Metadebates On Science 1999

METADEBATES ON SCIENCE
EINSTEIN MEETS MAGRITTE: An Interdisciplinary

Reflection on Science, Nature, Art, Human Action and Society
Series Editor
Diederik Aerts, Center Leo Apostel, Vrije Universiteit Brussel,
Belgium
Volume 1 Einstein Meets Magritte: An Interdisciplinary Reflection

The White Book of 'Einstein Meets Magritte'
Edited by Diederik Aerts, Jan Broekaert and Ernest Mathijs
Volume 2 Science and Art

The Red Book of 'Einstein Meets Magritte'
Edited by Diederik Aerts, Ernest Mathijs and Bert Mosselmans
Volume 3 Science. Technology, and Social Change

The Orange Book of 'Einstein Meets Magritte'
Edited by Diederik Aerts, Serge Gutwirth, Sonja Smets and
Luk Van Langenhove
Volume4 World Views and the Problem o[Synthesis

The Yellow Book of 'Einstein Meets Magritte'
Edited by Diederik Aerts, Hubert Van Belle and Jan Vander Veken
Volume 5 A World in Transition: Humankind and Nature

The Green Book of 'Einstein Meets Magritte'
Edited by DiederikAerts, Jan Broekaert and Willy Weyns
Volume6 Metadebates on Science

The Blue Book of 'Einstein Meets Magritte'
Edited by Gustaaf C. Comelis, Sonja Smets, Jean Paul Van Bendegem
Volume 7 Quantum Structures and the Nature ofReality

The Indigo Book of 'Einstein Meets Magritte'
Edited by Diederik Aerts and Jaroslaw Pykacz
Volume 8 The Evolution of Complexity

The Violet Book of 'Einstein Meets Magritte'
Edited by Francis Heylighen, Johan Bollen and Alexander Riegler
VOLUME6
Metadebates on Science
The Blue Book of
"Einstein Meets Magritte"
Edited by
Gustaaf C. Camelis, Sonja Smets and Jean Paul Van Bendegem

Brussels Free University
SPRINGER-SCIENCE+BUSINESS MEDIA, B.V.

Library of congress Cataloging-in-Publication Data
ISBN 978-90-481-5242-1 ISBN 978-94-017-2245-2 (eBook)

DOI 10.1007/978-94-017-2245-2
Printed on acidjree paper

AII Rights Reserved
© 1999 Springer Science+Business Media Dordrecht
Originally published by Kluwer Academic Publishers and Vrije Universiteit Brussel in 1999
No part of the material protected by this copyright notice may be reproduced or
utilized in any form or by any means, electronic or mechanical,
inc\uding photocopying, recording or by any information storage and
retrieval system, without written permission from the copyright owner.
Table of contents
General Introduction vu
Diederik Aerts
Acknowledgments xv
Editorial Introduction: Metadebates on Science xvii

Gustaaf C. Cornelis
1. "Was Einstein a Philosopher?" Deduction vs. Induction, the End 1
of Certitude and Conventionalism
A.M. Adam
2. Einstein, Ethics, and Action 17
Don Fawkes
3. Cognitive Science: Two Cultures or One? 31
Marek W. Bielecki
4. Godel's Theorem and Strong AI: Is Reason Blind? 43
Burton Voorhees
5. What about the Modernistic Concept of Consciousness? 63
Suzan Langenberg
6. Mind, Nature and the Emerging Science of Change: 71
An Introduction to Metamorphology
James Wilk
7. On Cognitive and Social Dimensions of Science: Constructivism 89
& Nonlinear Dynamics
William E. Herfel
8. Necessity of a New Paradigm in Experimental Research Taking 101
into Account Space and Time
Carmen Capel-Boute and Andre Koeckelenbergh
9. Biomedical Applications of Magnetic Fluids as a Specific 119
Interdisciplinary Problem
V.S. Lazarev, S. Roath, D.A. Yunusova and
O.K. Safonenko
10. Cosmology and Proliferation 139
Gustaaf C. Cornelis
v
VI TABLE OF CONTENTS
11. Resonance as a Unifying Principle between the Observer and 149

the 0 bserved
Paul J. Lewi
12. Early Greek Thought and Perspectives for the Interpretation 163
of Quantum Mechanics: Preliminaries to an Ontological
Approach
Karin Verelst and Bob Coecke
13. Landscapes of Sibylline Strangeness: Complementarity, 197

Quantum Measurement and Classical Physics
Arkady Plotnitsky
14. Metalinguistic Views of Quantum Mechanics and Its 213

Formalizability
Lars Lofgren
15. Quantum Truth, Non-Separability and Reality 225
Enrico Giannetto
16. Soliton Waves vs. the Particle Paradigm: The Elementary 237
Nature of the Physical World
Geoffrey Hunter
17. "Which Side Spinoza Would Have Taken (between 243

Einstein and Bohr) If He Had Lived to See the [Scientific]
Development of Our Days": An Analysis of Human Represen-
tation of the Physical Reality
Dan Nesher
18. General Relativity and Quantum Theory - Ontological 267

Investigations
Steven Weinstein
19. From Quantal to Material Level 281

Emmanuel Hemmerlin
20. On the Origin of Probabilities in Quantum Mechanics: 291

Creative and Contextual Aspects
Diederik Aerts, Bob Coecke and Sonja Smets
Index 303
DIEDERIK AERTS
THE GENERAL INTRODUCTION OF

EINSTEIN MEETS MAGRITTE
The series of books 'Einstein meets Magritte' presented here originates

from an international interdisciplinary conference with the same title,
which took place in Brussels in spring 1995. On the eve of the third
millennium, we assembled scientists and artists to reflect together on
the deep nature of reality and the knowledge and skill humankind has
gathered in this field. We had decided to call this meeting 'Einstein meets
Magritte' because we believed that meaningful keys could be found at
the place where the two meet. It is the way of the world that has made
Einstein and Magritte into icons of our culture. The purpose of the
conference was to reflect and debate without fear on the most profound
and timeless questions.
On one of those evenings, when the talks and discussions were long and
exhausting and the press were doing all they could to get Albert Ein-
stein and Rene Magritte in front of the microphones and cameras, a few
of my most loyal aides and myself succeeded in getting them safely and
quietly to a taxi, which then carried us off into the Brussels night. We
got out at Manneken Pis, since that was on Einstein's list, and we con-
cealed ourselves among the many tourists who were coming and going,
expressing their wonder in every language under the sun at the famous
little statue. And one of us was taking pictures: Einstein and Magritte
leaning against the railings, with us beside them, and one more, arm in
arm, and then another in case the first was no good, when suddenly I
felt a heavy slap on my shoulder:
"How you doing, mate?"
It was Jacky and his inseparable girlfriends Nicole and Sylvie, and
everyone embraced everyone else. I introduced Albert and Rene, and in-
terest was immediately shown, and I had my heart in my mouth, because
Jacky was a painter, poet and urban philosopher. We walked together
through the alleys of Brussels in dismal Belgian rain, over cobblestones
that glistened in the street lamps.
When we had provided for the inner man with 'Rabbit in Beer' and
'Mussels with Fries', and finally a 'Dame Blanche' topped with warm
chocolate sauce as apotheosis, Jacky enticed us to his house in the Rue
Haute where we threw ourselves into deep, soft armchairs. Albert and
Rene were offered the best places and as always Jacky told the story
Vll
@ 1999 Kluwer Academic Publishers.

viii DIEDERIK AERTS
of his life and discussed his rightness, as he did repeatedly, with a con-
fidence and suppleness that distinguished him so sharply from modern
science. Albert listened enthralled and Rene was fascinated, and once
more my heart was in my mouth, but Nicole winked reassuringly, and
Sylvie brought us snacks on cushions of Brussels lace and sweet white
wine in tall, old-fashioned crystal glasses.
The topic of discussion for the evening turned out to be 'the doubts
of modern science'. In science there is not a single hypothesis for which
one cannot find two groups of hard-working scientists, one of which can
'prove' the hypothesis while the other can 'prove' its negation. And the
more fundamental and important the question is, the more clearly the
situation turns out like this.
"It's crazy," maintained Jacky, "In fact science states that one doesn't
know anything anymore."
"That's right," said Albert, "Truth is not a simple concept, and I
believe that the history of science makes it clear how often erroneous
hypotheses have been believed over the centuries."
"A good thing too," replied Rene, "Things can only happen as a result
of the movement brought about by that constant doubt."
Meanwhile Sylvie came to join us and handed round pictures of the
exhibitions of Jacky's paintings and poems. Jacky suddenly got very ex-
cited, as if something had inspired him, and he leapt up and vanished
into his studio. A few minutes later he returned with his palette and
brush poised. Before I could stop him he had started painting violently
right at the spot where Albert and Rene were sitting. A large, gossamer-
thin piece of Brussels lace gradually took shape and Albert and Rene
vanished. Fortunately, my young assistants, Jacky's girlfriends and my-
self got away with just a few vicious daubs of paint in the face.
The series of eight volumes introduced here are not just the results of
the conference, as would be the case with a record of the proceedings.
The authors were invited to write with the events at the conference in
the back of their mind, so that the books would form a second phase in
the process of thought set in motion at the conference. A second phase
more clearly crystallised than the self-organising forum that arose during
the conference, but one which focuses on the same timeless questions and
problems.
The whole ensemble was already streamlined at the conference into
a number of main topics named after the colours of the rainbow - red,
orange, yellow, green, blue, indigo and violet, as well as white, the syn-
thesis of all colours. This order was maintained and led to eight separate
books in the series.
EINSTEIN MEETS MAGRITTE ix
Volume 1: Einstein meets Magritte: an Interdisciplinary Reflection

The White Book of Einstein meets Magritte
The white book contains more fully developed versions of the contribu-
tions made by the keynote speakers at the conference. So this white book
covers various scientific topics. In his article, 'Basically, it's purely aca-
demic', John Ziman asks himself what 'basic research' really is in today's
world. In his contribution, 'The manifest image and the scientific image',
Bas Van Fraassen analyses the considerable differences between the the-
oretical scientific description of the world and the way it appears to us.
He argues that most formulations of this problem may themselves be
tendentious metaphysics, full of false contrasts, and that insistence on a
radical separation between science and what we have apart from science,
and on the impossibility of accommodating science without surrender,
may be a way of either idolising or demonising science rather than under-
standing it. In the 'Microdynamics of incommensurability: philosophy of
science meets science studies', Barbara Herrnstein-Smith examines the
bemusing but instructive logical, rhetorical and cognitive dynamics of
contemporary theoretical controversy about science. In his contribution
'Subjects, objects, data and values', Robert Pirsig proposes a radical
integration of science and value that does no harm to either. It is argued
that values can exist as a part of scientific data, but outside any subject
or object. This argument opens a door to a 'metaphysics of value' that
provides a fundamentally different but not unscientific way of under-
standing the world. Ilya Prigogine discusses in 'Einstein and Magritte:
a study of creativity', the global transformation of a classical science
which was based on certainties into a new science that takes possibilities
as its basic concepts. Constantin Piron demonstrates in his contribu-
tion 'Quanta and relativity: two failed revolutions' that none of the two
great revolutions in physics, quantum mechanics and relativity theory,
have actually been digested by the physics community. He claims that
the vast majority of physicists still cling to the idea of a non-existent void
full of little particles, in the spirit of Leibniz or Descartes. Rom Harre
reflects on the significance of the theory of relativity. In his article 'The
redundancy of spacetime: relativity from Cusa to Einstein', he defends
the hypothesis that relativity theory is best interpreted as a grammar
for coordinating narratives told by different observers. In his contribu-
tion 'The stuff the world is made of: physics and reality', Diederik Aerts
analyses the consequences of the recent advances in quantum mechanics,
theoretically as well as experimentally, for the nature of reality. He anal-
yses the deep conceptual paradoxes in the light of these recent data and
tries to picture a coherent model of the world. In his contribution 'Da-
X DIEDERIK AERTS
sein's brain: phenomenology meets cognitive science', Francisco Varela

puts forward the hypothesis that the relation between brain processes
and living human experience is the really hard problem of consciousness.
He argues that science needs to be complemented by a deep scientific
investigation of experience itself to move this major question beyond
the sterile oppositions of dualism and reductionism. In his contribution
'What creativity in art and science tell us about how the brain must
work' William Calvin defends the prospects for a mental Darwinism
that operates on the milliseconds to minutes time scale, forming novel
ideas and sentences never previously expressed. Adolf Griinbaum in his
article 'The hermeneutic versus the scientific conception of psychoanal-
ysis: an unsuccessful effort to chart a via media for the human sciences'
argues that the so called 'hermeneutic' reconstruction of psychoanalytic
theory and therapy proposed by Karl Jaspers, Paul Ricoeur and Jiirgen
Habermas fails both as a channel and as alleged prototype for the study
of human nature. In his article 'Immortality, biology and computers',
Zygmunt Bauman analyses the shift that postmodern society has pro-
voked regarding the concept of immortality. He points out that strategies
of collective and individual immortality have shifted from the modern
deconstruction of death to a postmodern deconstruction of immortality,
and points out that the possible consequences of this process need to be
taken into consideration. Brian Arthur, in his article 'The end of cer-
tainty in economics', points out that our economy is very non-classical,
meaning that it is based on essentially self-referential systems of beliefs
about future economic conditions. He argues that our economy is inher-
ently complex, subjective, ever-changing, and to an unavoidable degree
ill-defined.
Volume 2: Science and Art
The Red Book of Einstein meets M agritte
And then Magritte comes in. Many obvious differences exist between
science and art. But the Science and Art volume of this series addresses
not only these differences but also the possibilities of crossing several of
the gaps between science and art. Several contributions deal with socio-
logical and philosophical elaborations of the similarities and differences
between science and art, while others approach science from an artistic
point of view and art from a scientific point of view. The volume also
considers several approaches that attempt to go beyond the classical
dichotomy between the two activities. In a special section, attention is
paid to the particular role played by perception in both science and art
as a regulator of human understanding. Together, these contributions
strive for an intensive interaction between science and art, and to a con-
EINSTEIN MEETS MAGRITTE xi
sideration of them as converging rather than diverging. It is to be hoped

that both science and art will benefit from this attempt.
Volume 3: Science, Technology and Social Change
The Orange Book of Einstein meets M agritte
The major subject of the orange book is that society as a whole is chang-
ing, due to changes in technology, economy and the changing strategies
and discourses of social scientists. The collected articles in the orange
stream discuss a range of specific societal problems related to the sub-
ject of social change, the topics of the articles range from the scale of
for instance sociology of health and psychohistory to more specific social
problems like for instance anorexia nervosa, art academics and the infor-
mation superhighway. Although the authors approach different subject
matters from dissimilar perspectives and work with various methods, all
the papers are related to the theme of science, technology and social
change. In the orange book the reader will find a lot of arguments and
hints pertaining to questions like: To what exactly will this social change
lead in the 21st century? What kind of society lies ahead? She/he will
be confronted to a plethora of enriching conceptions of the relationships
between social sciences and social changes.
Volume 4: World Views and the Problem of Synthesis
The Yellow Book of Einstein meets Magritte
A rapidly evolving world is seen to entail ideological, social, political,
cultural and scientific fragmentation. Many cultures, subcultures and
cultural fragments state their views assertively, while science progresses
in increasingly narrowly defined areas of inquiry, widening not only the
chasm between specialists and the layman, but also preventing special-
ists from having an overall view of their discipline. What are the motive
forces behind this process of fragmentation, what are its effects? Are
they truly inhospitable to the idea of synthesis, or do they call out,
more urgently than ever before, for new forms of synthesis? What con-
ditions would have to be met by contemporary synthesis? These and
related questions will be addressed in the yellow book of Einstein meets
Magritte.
Volume 5: A World in Transition; Humankind and Nature
The Green Book of Einstein meets Magritte
'A World in Transition. Humankind and Nature' is appropriately enti-
tled after its aim for an intrinsic property of reality: change. Of major
concern, in this era of transformation, is the extensive and profound in-
teraction of humankind with nature. The global scaled, social and tech-
nological project of humankind definitely involves a myriad of changes of
Xll DIEDERIK AERTS
the ecosphere. This book develops, from the call for an interdisciplinary
synthesis and respect for plurality, acknowledging the evolving scientific
truth, the need for an integrated but inevitably provisional world view.
Contributors from different parts of the world focus on four modes of
change: i) Social change and the individual condition, ii) Complex evo-
lution and fundamental emergent transformations, iii) Ecological trans-
formation and responsibility inquiries, iv) The economic-ecological and
socio-technical equilibria. Primarily reflecting on the deep transforma-
tions of humankind and on the relationship between humans and nature
it addresses major points of contemporary concern.
Volume 6: Metadebates
The Blue Book of Einstein meets M agritte
This book provides a meta-disciplinary reflection on science, nature, hu-
man action and society. It pertains to a dialogue between scientists,
sociologists of science, historians and philosophers of science. It covers
several topics: (1) the relation between science and philosophy, (2) new
approaches to cognitive science, (3) reflections on classical thinking and
contemporary science, (4) empirical epistemology, (5) epistemology of
quantum mechanics. Indeed, quantum mechanics is a discipline which
deserves and receives special attention here, for it still is a fascinating
and intriguing discipline from a historiographical and philosophical point
of view. This book does not only contain articles on a general level, it
also provides new insights and bold, even provocative theories on the
meta-level. That way, the reader gets acquainted with 'science in the
making', sitting in the front row.
Volume 7: Quantum Structures and the Nature of Reality
The Indigo Book of Einstein meets M agritte
This book refers to the satellite symposium that was organised by the
International Quantum Structure Association (IQSA) at Einstein meets
Magritte. The IQSA is a society for the advancement and dissemination
of theories about structures based on quantum mechanics in their phys-
ical, mathematical, philosophical, applied and interdisciplinary aspects.
The book contains several contributions presenting different fields of re-
search in quantum structures. A great effort has been made to present
some of the more technical aspects of quantum structures for a wide
audience. Some parts of the articles are explanatory, sketching the his-
torical development of research into quantum structures, while other
parts make an effort to analyse the way the study of quantum structures
has contributed to an understanding of the nature of our reality.
EINSTEIN MEETS MAGRITTE Xlll
Volume 8: The Evolution of Complexity

The Indigo Book of Einstein meets Magritte
The violet book collects the contributions that consider theories of evo-
lution and self-organisation, on the one hand, and systems theory and
cybernetics, on the other hand. Both can add to the development of an
integrated world view. The basic idea is that evolution leads to the spon-
taneous emergence of systems of higher and higher complexity or "in-
telligence": from elementary particles, via atoms, molecules, living cells,
multicellular organisms, plants, and animals to human beings, culture
and society. This perspective makes it possible to unify knowledge from
presently separate disciplines: physics, chemistry, biology, psychology,
sociology, etc. The volume thus wishes to revive the transdisciplinary
tradition of general systems theory by integrating the recently devel-
oped insights of the "complex adaptive systems" approach, pioneered
among others by the Santa Fe Institute.
Even these books only signify a single phase in the ever-recurring process
of thought and creation regarding the basic questions on the reality that
surrounds us and our place in it.
Brussels,
July 17, 1998.
ACKNOWLEDGEMENT S
The editors wish to thank the Center Leo Apostel (Clea) at Brussels
Free University and ORGAMED for their persistence in the organiza-
tion of the Einstein meets M agritte conference, for taking the initiative
to publish its outcomes, and for providing the dynamic structure and
stimulating surroundings to work in.
Explicit thanks are due to Sami Amira, Jan Broekaert, and Didier
Durlinger for their excellent handling of the material to be published.
Meeting the final deadline would not have been possible without their
flexibility that went well beyond any obligations.
We would also like to express our thanks to everyone at Kluwer Aca-
demic Publishers and the VUB-Press for the practical realization of the
volume.
Finally, and most important, we would like to thank the authors for
their contributions, as well as for their patience and responsiveness dur-
ing the editing.
Brussels, 5 June 1998.
XV
GUSTAAF C. CORNELIS
METADEBATES ON SCIENCE
INTRODUCTION
Through his paintings, creating astonishing, even absurd objects and

situations alike, Rene Magritte evokes a poetic climate, a seemingly
irrational, at least inconsistent, though wonderful dreamworld. Albert
Einstein's universe, how difficult it is to grasp for the non-specialist, is
not as surrealistic as one expects it to be. Einstein's world is (hope-
fully) logically consistent, Magritte's world is not. So, what if Einstein
would have met Magritte? Did they have anything in common, enough
to exchange thoughts about? Would they have bothered to listen to each
other? Probably they would, since both of them were evidently clever
enough to be open-minded.
As scintillating as counterfactuals might be, they are really quite
uninteresting from a historiographical point of view. However, from a
logico-philosophical point of view, the possible situations implied by the
counterfactual 'Einstein meeting Magritte' are plentiful, almost infinite.
What would have happened? Unfortunately, the antagonists are dead.
Hence, we can not make the two meet and see what happens. Therefore,
we can not really know what the outcome would have been. Yet, we can
speculate. Better still, we can devise an intriguing experiment: a meeting
between scientists on one hand and artists on the other that could give
a push in the right direction.
In May 1995, it all came to pass. On the occasion of the twenty-fifth
anniversary of the Vrije Universiteit Brussel the international conference
Einstein meets Magritte took place. Scholars, scientists and artists met
each other under the best of conditions. 'Einstein' and 'Magritte' got
acquainted, eventually.
The conference commenced with six plenary sessions, by Ilya Pri-
gogine, Barbara Hernnstein Smith, John Ziman, Rom Harre, Francisco
J. Varela and William H. Calvin, before it turned to the consideration
of specific problems in science presented in the seven parallel sessions.
Subjects regarding science and art, university and society, worldviews,
nature of life and death, quantum structures and complexity theory were
tackled in three hundred talks. In eleven other plenary sessions with
Adolf Gri.inbaum, Robert Edwards, Chris Langton, Zygmunt Bauman,
Bas Van Fraassen, Linda Scheele, Robert Pirsig, Julian Jaynes, W. Brian
XVII
© 1999 Kluwer Academic Publishers.

xviii GUSTAAF C. CORNELIS
Arthur, Constantin Piron, and Heinz Von Foerster, the spectrum of

present-day scientific research was completed. Five workshops, several
exhibitions, performances, posters on all kinds of topics and satellite
symposia on quantum structures and cybernetics completed the forum.
A great many lecturers linked their respective specialism to the Ein-
stein-Magritte duality, while others made a straightforward presentation
of their last results and conclusions. Together, through their papers, they
offered a cross-section of contemporary science. This book provides a de-
scription of the so-called 'blue stream' in the parallel sessions held at the
conference. Other volumes in this series cover the remaining contribu-
tions.
Many reactions were submitted to us, so much that, to our regret, not
all those who contributed papers could be invited to participate in the
conference and only a few can see their contribution published here. This
volume binds together twenty papers out of a total of thirty presented at
the conference as metadebates. They are divided here over five sections.
One might expect the metadebates to be the most farfetched contribu-
tions to the conference. If Einstein would really have met Magritte, that
encounter would very probable have resulted in a metadebate: a mosaic
of ideas would have emanated from it. Nevertheless, as would have been
the case with the two adversaries, the ideas presented here are far from
preposterous. Evidently, it must not be thought that this book will yield
immediate, internationally acceptable answers to the problems which
are identified. Yet, they do provide new insights. These unpretentious
articles reflect a scientific way of thinking without exception.
One might expect also the whole to be heterogeneous, but as the reader
readily observes, the following overview reveals a surprising correlation
between the articles, and even a relation between the sections seems to
subsist.
EINSTEIN, PHILOSOPHER AND SCIENTIST
In his contribution to this volume, it is argued by A.M. Adam (Univer-

site de Geneve, Switzerland) that Einstein had, indeed, a pronounced
philosophy of science, without being a philosopher of science. Doing so,
Adam runs counter to Pais's ideas, reasoning that Einstein's Induktion
und Deduktion in der Physik (1919) was clearly a defense of hypothet-
ical deductivism, hence Einstein was for sure a true philosopher. For
that matter, Einstein's philosophy of science was elaborated, expanded,
developed, and systematized by Popper.
Don Fawkes (Fayetteville State University, U.S.A.) goes even a little
further in his "Einstein, ethics and action". He argues that relativity
METADEBATES ON SCIENCE xix
theory apparently cannot provide support for ethical relativism. Nev-

ertheless, Einstein did defend certain ethics. Taking these as starting
point, Fawkes makes a case for a new course between relativism and
absolutism.
COGNITIVE SCIENCE: AN EVALUATION
The second section groups four evaluations of cognitive science. As an

interdisciplinary study, cognitive science can be viewed as merely the
combination of 'two cultures', as Marek Bielecki (California State Uni-
versity, U.S.A.) puts it. Yet, it concerns here an independent discipline.
On the other hand, naturalizing the philosophical element reduces cog-
nitive science to an exclusively natural science. According to Bielecki,
a 'naturalized philosophy' is incapable of addressing the fundamental
problem in cognitive science: the problem of meaning. He proposes Luh-
mann's system theory as capable of tackling the problem of intentional-
ity, as well as the problem of methodological diversity.
Burton Voorhees (Athabasca University, Canada) assesses Godel's the-
orem in view of strong artificial intelligence. Godel's theorem implies
certain restrictions on AI. In his paper, Voorhees discusses these impli-
cations and makes subsequently some philosophical considerations.
Suzan Langenberg (Belgium) presents a very intriguing view on cog-
nitive science. Without question the modern concept of consciousness is
very obscure and, according to Langenberg, in a sense double-faced: it
gives us the confusing experience of being the center of the universe and
simultaneously of being alienated from the world. She uses the Einstein-
Magritte 'coalition' as a starting point to develop a synthesis of this
contradiction.
The fourth paper, written by James Wilk (Brunel University, U.K.),
concerns an introduction to metamorphology: the study of transforma-
tion in nature. Perception of changes is fundamental for any biological
complex. So is its response to the ever changing environment. Wilk points
out that subsequently difficulties occur on the theoretical level. He shows
that a metamorphological approach tackles these problems sufficiently,
although it will change our conception of our surroundings thoroughly.
The last paper of this section is William Herfel's (University of New-
castle, U.K.) contribution. In his approach, the concepts of nonlinear
dynamics are used to analyse the dichotomy between the cognitive and
social aspects of science. After a study of the different positions taken
by several distinguished researchers like Churchland, Giere, Hacking,
van Fraassen, Galison, Latour and Woolgar, Herfel concludes that non-
linear dynamics indeed provide a conceptual framework for approaching
XX GUSTAAF C. CORNELIS
the complex relation between cognitive and social factors as a nonlinear

interaction.
EMPIRICAL EPISTEMOLOGY
Different disciplines, namely physical chemistry, biomedical sciences and

cosmology serve as the object of epistemological research. Carmen Capel-
Boute and Andre Koeckelenbergh (Universite Libre de Bruxelles, Bel-
gium) show in the first contribution to this section that the requirement
of reproducibility does not hold for physical chemistry, acknowledged as
a genuine science.
Vladimir Lazarev and his colleagues Roath, Yunusova and Safonenko
(Walt Disney Memorial Cancer Institute, U.S.A.) concentrate on bio-
medical research. Although their text is not easily accessible to non-
specialists, the philosophical and historiographical conclusions are very
illuminating.
The third and last paper concerns contemporary cosmology. Gustaaf
Cornelis (Vrije Universiteit Brussel, Belgium) shows that in spite of the
existence of an appropriate breeding ground and a suitable methodolog-
ical tradition, there is no proliferation of theories whatsoever appearing
in the development of modern cosmological thinking. Indeed, cosmogo-
nical theories based on extragalactic magnetism were developed during
the eighties, but as the only real attempt to contest standard theory,
they are slowly disappearing now altogether.
REFLECTIONS ON CLASSICAL THINKING

AND CONTEMPORARY SCIENCE
Two papers take classical approaches into consideration, searching for

'new' ways to tackle distinct problems in contemporary science. Both
articles are highly speculative and interdisciplinary, hence difficult to
class.
Paul Lewi (Janssen Research Foundation, Belgium) argues that the
universal resonance phenomenon present in almost every aspect of reality
can be viewed as the link between observer and the observed. Karin
Verelst and Bob Coecke (Vrije Universiteit Brussel, Belgium) try to show
that an 'ontological' approach of quantum mechanics based on early
Greek thought can be very fruitful. This could shed new light on some
hard problems in quantum physics.
METADEBATES ON SCIENCE xxi
EPISTEMOLOGY OF QUANTUM MECHANICS
This volume concludes with a section uniquely devoted to the episte-

mology of quantum mechanics. All contributions tackle the problem of
complementarity. By considering the Einstein-Podolsky-Rosen paradox,
Bohr's response to it, Bohm's hidden variables theory and Bell's the-
orem, Arkady Plotnitsky (University of Pennsylvania, U.S.A.) explores
how the debate on the relation between classical and quantum mechan-
ics, has implications for classical conceptions. His paper discusses the
significance of the controversy for classical rather than only quantum
physics, and the relevancy for traditional and postmodern philosophy
(Deleuze, Serres, Derrida).
Lars Lofgren (University of Lund, Sweden) argues that as a theory of
measurability, i.e. of what is measurable and is not measurable, quantum
theory ought to contain decidable levels as well as less constructive ones.
As a corollary, Lofgren concludes that such levels might resolve a recent
challenge against Bohr's wave-particle complementarity by Ghose, Home
and Agarwal.
In his contribution, Enrico Giannetto (Universita di Pavia, Italy) anal-
yses the definition of logical truth in quantum physics. An 'empirical'
logic and category theory are taken as a point of departure. The logical
incompleteness of quantum mechanics and non-separability which im-
plies the 'death of atomism' are discussed both from the physical and
logico-epistemological points of view, also dealing with the problem of
quantum reality.
Geoffrey Hunter (York University, Canada) reassesses the approach
of de Broglie. This fitting article provides a good introductory to the
non-specialist: the problem at stake is described very transparently.
The title of the article by Dan Nesher (University of Haifa, Israel),
concerning the epistemological aspects of quantum physics, is a counter-
factual: "Which side would Spinoza have taken (between Einstein and
Bohr) if he had lived to see the [scientific] development of our days."
Needless to say that this is reminiscent of the conference's title. The
article elucidates the controversy about complementarity.
In his text on General Relativity Theory and Quantum Theory, Steven
Weinstein compares the respective ontologies which go with these two
approaches to the world. According to Weinstein, GRT's ontology is re-
lated to either a particle based or a field based view. QT's ontology
is essentially a field based ontology. Weinstein points out that several
problems emerge when one tries to reconcile the two theories from an
ontological point of view. Nevertheless, according to Weinstein, a recon-
cilement remains conceivable.
xxii GUSTAAF C. CORNELIS
Emmanuel Hemmerlin's contribution regards his proposal to reduce

the fundamental constants to even a smaller number (as far as the in-
tention of the concept 'physical constant' goes). The paper ends with
some old, but still intriguing philsophical questions, pertaining to holism,
reductionism, dualism.
In the final and last, but certainly not the least article of this book,
Diederik Aerts, Bob Coecke, and Sonja Smets (Vrije Universiteit Brus-
sel, Belgium) present some intriguing results found by the CLEA-group
concerning a possible explanation for the probabilities in quantum me-
chanics. These probablities can now be seen as due to a lack of knowledge
on the interaction between measurement apparatus and the system un-
ders study, contrary to the 'classical' view that the lack of knowledge
(intrinsic or not) has to do with the state of the system.
AN INTERDISCIPLINARY REFLECTION ON SCIENCE,

NATURE, HUMAN ACTION AND SOCIETY
This caption was printed on the announcements of the conference early in

1994. Pointless to say that it serves equally well to describe the contents
of this book. Maybe the authors justly took it as a point of departure,
maybe the organizing committee merely selected the 'right' papers. Any-
way, they all surely nurture the inquiring mind; as Einstein and Magritte
both did by their ever so distinct but evenly prodigious and inspiring
work.
Brussels, May 1998.
A.M. ADAM
"WAS EINSTEIN A PHILOSOPHER?" DEDUCTION VS.

INDUCTION, THE END OF CERTITUDE AND
CONVENTIONALISM
1. INTRODUCTION
Instead of a survey, a direct attention to the literature on the debate of

whether Einstein was a philosopher will be offered. The question, "Was
Einstein a philosopher?", may sound a strange one, since Einstein was
considered as a philosopher scientist in a collection of papers published in
the series of the Library of Living Philosophers [29]. Still, book reviews
written by authoritative philosophers, E. Nagel [21] and G. Bergman
[2] adopted H. Reichenbach's claim, who contribute to this book [29],
whose thesis was that, Einstein's philosophy is by implication only. An-
other contributor, who agreed with Reichenbach, was F. Frank, one of
Einstein's first important biographers. His student Holton rebelled gen-
tly and debated this assertion indirectly, in a series of papers, offering
his deliberations on Einstein's philosophy of science as well as asserting
that there was a transformation between an early Einstein and a later
Einstein: the early Einstein was an empiricist and a positivist whereas
the later Einstein was a rationalist and a realist [17]. An attempt to
explain why in general Einstein was not discussed as a philosopher until
the end of the 1960s is given by the historian of philosophy of science, S.
Jaki, in a well documented argument. Holton is outstanding, says Jaki,
because of his profound consideration of Einstein's philosophy. Jaki does
not mention that Holton omitted a discussion on the neglect. Jaki him-
self goes on to explain it, arguing that Einstein was not in line with main
stream philosophy:
In all these statements Einstein denounced positivism, endorsed a realist
metaphysics, and professed his belief in the objectivity of physical reality.
These and other statements were so many public and empathic indications
of his belief that there was something absolute beneath the relative. Yet,
one would look in vain for a substantive trace of those statements in the
books and articles written on relativity by Frank, Rudolf Carnap, Reichen-
bach, and Herbert Feigl, all members of the Vienna Circle, who in the 1950s
and 1960s captured, in the United States at least, the role of authoritative
spokesman on behalf of Einstein in particular and of science in general.
Their systematic silence on many relevant statements and facts was only
1
2 A.M. ADAM
part of the strategy pursued by them. Instead of strategy, the word cru-
sade would be more appropriate. Reichenbach himself warned that logical
positivism should be looked upon as a crusade and not as an abstract phi-
losophizing. Intellectual crusades have their inner logic to which logical pos-
itivists were not immune. Or, as Herbert Feigl admitted well over a decade
ago: 'Confession, it is said, is good for the soul. Undoubtedly we [logical
positivists] made up some facts of scientific history to suit our theories'[18].
Another source of the ignorance of Einstein's philosophy is due to a lack

of knowledge of physics: Einstein's physics not only overwhelm physicists
such as Michelson but it overwhelmed many philosophers as well. As
strange as it may sound, there are many philosophers who did not study
the abc's of relativity and its epistemic, metaphysical and methodologi-
cal consequences for modern philosophy, and most philosophers paid no
attention to Einstein's philosophy. The merit of Einstein's contributions
to philosophy was, however, loudly debated in the pages of The British
Journal for Philosophy of Science and Studies of History and Philosophy
of Science; the debate was led by Zahar and his opponent Feyerabend.
Meanwhile, in 1982, Abraham Pais, in his first and most comprehensive
biography of Einstein [22], a study which superseded most Einstein's
previous biographies, suggested (the truism) that Einstein's scientific
insights were both philosophical and scientific. Zahar's papers were pub-
lished in a form of a book only in 1989 [30]; here the heuristics of Ein-
stein's philosophical view were examined. In 1993, Michel Paty publishes
his book in which he offers a reconstruction of the significant epistemical
import of Einstein's philosophy to early 20th century philosophy of sci-
ence [25]. Pais receives the book. Yet, in a recent book, Pais admits his
difficulties to appraise Einstein's contributions to philosophy, and argues
that Einstein's contribution to philosophy was, if at all, marginal [23].
Pais does not respond to Paty. Paty's book is unique in the field, in that
he reconstructs the epistemic debates between the scientists. For exam-
ple, the debates between Mach and Poincare and Planck on the basic
questions of the epistemic status of scientific theory, scientific method-
ology, etc. However, Paty doesn't explain if Einstein's philosophy is so
significant, why it was kept silent. Admittedly, Einstein's writings on
philosophical issues are few; it is not the quantity which counts here
but the quality. My thesis is that, Einstein's epistemic, methodological,
and metaphysical contributions to philosophy do not only indicate that
he directly contributed to philosophy, but in evaluating these contribu-
tions, it can be shown whether they were significant to early 20th century
philosophy of science. Furthermore, his philosophy was in opposition to
some of his own colleagues and friends' views who opposed his general
relativity and cosmology (Born). This can be illustrated by the open
WAS EINSTEIN A PHILOSOPHER? 3
debates on the sources of knowledge (scientific theory) and on scientific

method between the rationalists and the empiricists, and on the status
of scientific theory between the conventionalists and the realists. In these
debates Einstein offered new positions.
2. A CRUDE SOCIOLOGY OF KNOWLEDGE
Did Einstein talk philosophy and criticised it with philosophers? Yes.

Einstein was exposed to philosophy from an early age. He was about
13 when he first read Kant's Critique of Pure Judgment. He studied the
philosophies of Hume, Mill and Poincare [1902]. Einstein was directed to
Mach by his friend Besso, and he corresponded with Mach and admitted
Mach's crucial influence on his works. Not only the reconsideration of
the concept of gravitation; in a philosophical view, the ideas of learning
by trial and error as well as the significance of the idea of thought ex-
periment: both are critical aspects of Einstein's epistemology: both ideas
are discussed in Mach [19]. Einstein's reading included both his contem-
porary philosophers and historians of science. Beside reading the works
of Mach and Poincare [24], he read and corresponded with Meyerson
and not only wrote a review [9] of Meyerson's philosophical work but
also exchanged philosophical comments and criticisms at public confer-
ences [8]. Cassirer's philosophical work [5] on relativity was closely scru-
tinised by Einstein. All this is worth mentioning since all Einstein's cor-
respondences with philosophers like Meyerson and Cassirer, with whose
philosophies he disagreed, still await translation and publication. Fur-
ther, Schlick's critical book review of Cassirer book [5], was later praised
by Einstein. Schlick and Einstein studied each other works (roughly be-
tween 1910 and 1920) and corresponded and exchanged philosophical
views. With his friends, Einstein discussed philosophical issues as early
as his period in the patent office in Bern, a period of time in which he
was to become a professor of physics, a period of time in which he pub-
lished his famous papers of special theory of relativity-his friend who
translated most of his works to French, Solovine, discusses this in the
Introduction to their correspondence [13]. Einstein read Russell, neither
do I know of any evidence for the first time nor the first item he read.
Another point is historiographical-Einstein's correspondences. They
throw light on his intellectual development, since with his friends Ein-
stein discussed issues belonging to all the aspects of his life and in partic-
ular of his intellectual interests. This is the place where some historians
may find information which can be of use in the process of constru-
ing different aspects of Einstein's work and life. Similar is the case as
regards Einstein's contribution to philosophy. Nevertheless, in his pub-
4 A.M. ADAM
lications to the layman and professionals alike, Einstein's philosophical

views were argued in succinct manner. Einsteinian scholars may indeed
use the archives to this end; still Einstein's own published work is suf-
ficient. It was in public where Einstein first offered his philosophy, in a
language which is intelligible to the layman and the professional alike.
Thus the neglect of Einstein's philosophy is not due to the fact that he
kept it hidden and communicated it discretely. Neither is it due to the
(false) view that Einstein's philosophical contributions do not merit his
being discussed as a philosopher. In order to give further reasons and to
answer the question, why Einstein's philosophy of science was omitted
from the agenda of philosophers, a study of the sociology of early 20th
century philosophers and scientists is required; a study which is beyond
the scope of the present essay.
3. RAT I 0 N ALI S M V S. EMPIRICISM
According to the article "Rationalism" in the 11th edition of the En-

cyclopaedia Britannica [1910], the rationalist asserts that the source of
knowledge is mental, the sensationalist (the empiricist), that it is em-
pirical. As regards the issue of scientific method: the rationalist says,
it is deductive; the empiricist says, it is inductive. Both, the rationalist
and empiricist presuppose that, in principle, if one uses the right method
one should arrive at true knowledge; this I shall call the certitude the-
sis. The debate between the empiricist and rationalist on the sources
of knowledge and methodology of knowledge is central to the philoso-
phies of Bacon, and to Descartes' responses to Bacon, and it is central
to Hume's attack on Cartesian epistemology, and it is central to Kant's
consideration of Hume's attack and his own response to Hume. One of
the leading debates, which is traced back to the philosophies of Bacon
and Descartes, was common amongst late 19th century scientists. Mach
and Planck debated the sources of knowledge, Mach defended empiricism
whereas Planck defended rationalism. Indeed it became an unsettled cen-
tral debate, and was put again on the agenda of philosophers in Russell's
classic [26].
4. INDUCT IV ISM V S. DEDUCT IV ISM
In his paper "Induktion und Deduktion in der Physik" published on 25

December 1919 amongst other articles by Planck et al. in the widely read
news magazine Berliner Tageblatt Einstein made a compelling argument
in a classical philosophical form, to which I have seen only three earlier
references in the literature. First reference is in Don Howard [15] in
which a translation of the fifth paragraph of that paper is offered (See
below more on this context). Second reference is in Eisenstaedt [14] by

passing he says that Einstein's philosophy predates Popper's (I discuss
this below at the end of my essay). Third reference is most important,
Michel Paty [25], the only study which discusses Einstein's philosophy
in the current literature, and does not make much of this paper; so, I
will not discuss Paty's book here and leave it for an essay review which
is in progress.
The readership of the Berliner Tageblatt was the educated person, and
by directing attention to the objective thing science, Einstein aimed to
heal the political situation in Germany. This was not only published in
unstable Germany with its insane political situation, perhaps the situ-
ation is best characterised by George Grosz's paintings of the period.
Most importantly it was published after Einstein's theory of relativity
was confirmed by experience, and the results of the famous British eclipse
expeditions of November 1919 were known to many. This confirmation,
Sir Karl Popper confesses, consequently turned his attention, like of so
many of his peers, to the study of relativity. Enough was said about this.
The non-educated, and the educated people alike, were full with enthu-
siasm even though they did not know what it meant, but that something
big happened in physics, that Einstein created a theory which signifi-
cantly changed Newton's conceptions, and thus changed our conceptions
of the world.
It is also worth while to discuss Einstein's philosophical paper if only
because he breaks with philosophical tradition and offers new answers to
disturbing old questions such as, what is, if any, scientific method? What
is the structure of scientific theory? What should be the presuppositions
of the researchers who produce scientific theory?
5. IS EM PI R I CIS M TEN A B L E? N 0 .
In the first paragraph of his paper Einstein presents the standard answer
of the empiricist to the question of scientific method, the development
of natural science can be looked at as if it follows the inductive method:
Singular facts are chosen and grouped in such a way that the law of
nature which connects them becomes evident. By grouping these laws
more general ones can be derived until a more or less homogeneous
system would have been created for this set of singular facts. Start-
ing from these generalisations, the retrospective mind could then,
inversely, arrive back at the singular facts by pure reasoning. [6]
This is an elucidation of the empiricist view. General theories are induced
from singular facts. The main assumption shared by the empiricists, ever
since Bacon, was (the universal judgment) that all researchers should be-
6 A.M. ADAM
gin the inquiries about nature without any prejudgment. This is where
Einstein begins his argument against the standard answer of the empiri-
cist to the question of scientific method. He is not repeating the Humean
argument that inductive inference is invalid (he does this elsewhere [10]),
but Einstein denies that the researcher begins his inquiries tabula rasa.
Einstein argues that in principle the scientist cannot select singular facts
unless he has prejudgments. Einstein's attack is made by a counter ex-
ample, a singular case which counters (refutes) the universal judgment.
Einstein's counter example is an example of Galilee's scientific work. In
Einstein's opinion, the scientist has a prejudgment when he begins his
inquiries of nature. Otherwise, how should we understand, the essential
part of scientific discovery, the selection of facts done by the scientist?
Galileo could never have discovered the law of free falling bodies, had
he not maintained the preconceived opinion that the circumstances
which we really encounter are complicated by the effects of air resis-
tance so that one has to focus on cases in which air resistance plays
as marginal a role as possible. [6]
6. CRITICAL RAT I 0 N ALI S M IS P R 0 P 0 SED
Einstein, then, asserts that in fact the scientist begins in almost a dia-
metrical way, to the inductive method. The source of scientific theory is
not the singular facts by themselves but general hypothesis based on the
scientist's mental intuition which evaluates ensemble of facts. Then, the
scientist constructs a hypothetical theory with general laws and their
logical consequences, particular statements, which can be tested against
experience. Note that this is a characterization of what is known to be
the "hypothetico deductive" structure of scientific theory. Einstein de-
scribes it as follows:
Intuitive comprehension of the essentials about the large complex of
facts leads the researcher to construct one or several hypothetical
fundamental laws. From the fundamental law (system of axioms) the
researcher draws as completely as possible its consequences by purely
deductive logical method. These consequences, which often can only
be derived from the fundamental law by extensive elaborations and
calculations, can then be compared with experience yielding a cri-
terion for the validity of the supposed fundamental law. Both the
fundamental law (axioms) and the consequences form what we call
a "theory" [6].
Einstein is rejecting the scientific outlook of the empiricist: first and
foremost Galilee's main discovery cannot be understood according to
that view (i.e., that scientific method is inductive), secondly, contrary
to this view, the scientist has prejudgment, otherwise how could we un-
derstand his selection of facts; finally, we can better understand the way
of scientific discovery by assuming that scientific theory is hypothetico
deductive. Einstein illustrates that indeed all branches of physics need
to be understood in that way:
Newton's theory of gravitation, thermodynamics, the kinetical theory
of gas, modern electrodynamics and so forth, all arose in this way
and their foundations have in principle a hypothetical character [6].
The question which is left open is: what is the scientific method, if not
inductive? Does Einstein say that the scientific method is deductive? Or
does he think that science has any particular method? Or the question
we should ask is not, what is the method of science, if any, but rather
how do we conjecture a scientific theory? That is, how do we arrive at
scientific knowledge? Einstein, in my opinion, does not advocate that
the scientific method is deductive. I doubt that he thinks that this is the
case because when he arrives to describe the scientific method he begins
to describe it in a metaphoric language:
But he [the researcher] doesn't arrive at his system of thought in a
methodical, inductive way, he rather, snuggles to the facts by intu-
itive choice among the imaginable axiomatic theories [6].
Thus, in Einstein opinion, the method of science is neither inductive nor
deductive. There is no scientific method in the traditional sense. In the
sense that both, the traditional rationalist and empiricist presuppose
the certitude thesis. If anything, a scientific method cannot license true
knowledge. Here Einstein attempted to put an end to the feeling of
certainty in science (end of certitude). Nevertheless, the scientist creates
a hypothetico deductive theory. What is the truth value of a scientific
theory? Can we know, in principle, whether a scientific theory is correct?
Every student of early 20th century philosophy of science knows that
these are all central issues to heated debates between philosophers. These
are the questions Einstein was considering.
But before we turn to Einstein's answers to these questions, it is worth-
while to dwell on their opposition as represented by Born in particu-
lar since Born's assertions are in straight contradiction with Einstein's.
Twenty years later, in his book on philosophy of nature, Born asserts
that relativity illustrates that the scientist begins from hard facts rather
than preconceived ideas [3, pp. 141-142]. Born is not referring to Ein-
stein's philosophical views on this question. Prima facie it is quite strange
that Born argues that the empiricist way to describe relativity is bet-
ter. Born followed Einstein's publications and activity ever since 1915
and closely in the 1919-1920s from Gottingen where he held a post until
8 A.M. ADAM
1933. This is clear to every reader of the correspondences between them

[4]. Born did not approve of the philosophical changes made by the later
Einstein. This change of philosophy of science was in the public domain
since 1919- 1920. Not to mention Einstein's rejection of empiricism in
his famous Herbert Spencer Lecture [10]. Einstein illustrated his counter
examples against empiricism and inductivism by discussing how deduc-
tively he arrived at Relativity. Born knows of these differences between
his philosophy and Einstein's, and still Born suggests that the scientists
employ inductive method without considering Einstein's opposing views;
he says it in many places in his philosophical study of science. I further
would like to focus on Born's discussion of the issue of probability since
this issue was the kernel of scientific debates between Einstein and his
colleagues the physicists. I will not dwell here, however, on its scientific
or on its philosophical aspects. I consider Born's philosophical views not
only as the opposing views to Einstein's but also to show the relevance of
a discussion on Einstein's philosophy which is omitted even by his close
friend. Born then subscribes to the view that the inductive method is
the method of science. It is important to say that Born does not bother
to argue for his position; his convictions are so strong that any argument
seems, in reading his treatment of the issue of scientific method, to be
superfluous [3, p. 47].
7. THE END OF CERTITUDE
As regards the question of whether a scientific theory is erroneous, this

can be detected only if there is a logical error in its deduction. A scientific
theory needs to be evaluated first on its logical grounds, the question is
whether a scientific theory is consistent. The second test of a scientific
theory is, whether a fact is in accord with its consequences, that is, if
the theory is not in accord with experience, then the theory is false. In
Einstein's words:
A theory can thus be recognised as erroneous if there is a logical error
in its deductions, or as incorrect if a fact is not in agreement with its
consequences [6].
As regards the question whether a scientific theory can be proved to be
true, Einstein asserts in a long sentence that it cannot.
But the truth of a theory can never be proven. For one never knows
that even in the future no experience will be encountered which con-
tradicts its consequences; and still other systems of thought are al-
ways conceivable which are capable of joining together the same given
facts [6].
In this point Einstein's argument is that a scientific theory in principle

cannot be proved to be true, for he argues, we do not know if any future
experience will contradict its predictions.
This is where Einstein gives la coup de grace to the thesis of certitude
by directing our attention to the hypothetico deductive structure of a
scientific theory. A theory, in principle, cannot be proven to be true, be-
cause future experience may counter its consequences. This is the kernel
of the philosophical view which came to be known as critical rationalism.
What is said so far is the least one can say on the context of Einstein's
philosophical paper, and its contributions to philosophy.
8. EINSTEIN'S ALLEGED C 0 NV EN T I 0 N ALI S M
Consider what is to be done in the case when two scientific theories "are
capable of joining together the same given facts". This is where Einstein
cannot give a criterion for choice between the two theories. In this case,
the choice of the scientist then depends on the scientist's intuition, says
Einstein, where he attempts to explain why bright scientists hold contra-
dictory theories. Einstein, however, did not hold contradictory theories:
If two theories are available, both of which are compatible with the
given factual material, then there is no other criterion for preferring
the one or the other then the intuitive view of the researcher. Thus we
may understand how sharpwitted researches, who have command of
theories and facts, can still be passionate supporters of contradictory
theories. [6]
Interestingly enough the last three quotes from Einstein's paper serve a
respectable Einsteinian scholar, Howard, to argue that (I shall number
the sentences for the sake of convenience):
(1) What is significant [says Howard] is Einstein's claim that a the-
ory can never be proven true, not only because new discoveries might
overturn it, but also because alternative theories are "always conceiv-
able" which can account for the same evidence. (2) Einstein is saying
that in addition to the old Humean problem of inductive uncertainty,
we also have to acknowledge the radical underdetermination of the-
ory by evidence. (3) But this latter is just Schlick's point that "under
certain conditions several theories may be true at once, in that they
provide a different but in each case perfectly univocal designation of
the facts." [15, pp. 620-621]
Howard's statement (1) is true to the text, but his statement (2) is not.
If you recall, Einstein's attack on the inductive method is on the as-
sumption that the scientist begins tabula rasa in his inquiries of nature.
Einstein did not attack the invalidity of inductive inference, which is
10 A.M. ADAM
known to be the "old Humean problem of inductive uncertainty". Fur-

ther, and most crucial, Einstein does not say what (as Howard seems to
think) Schlick says. Einstein does not state Schlick's possible conditions
under which "theories may be true at once". Recall that Einstein asserts
that a theory can never be proven to be true and that when he says any-
thing about two theories which agree with the same set of facts, he does
not say that they are both true. Einstein rather says that the choice of
the scientist in this case depends on his own intuition. It is very bizarre
that Howard is off the mark because from Einstein [6] the translation of
the fifth paragraph only appears in Howard's paper.
Worse, Howard alleges that Einstein believes in conventionalism:
Einstein and Schlick both believed then that experience does not
uniquely determine a correct theory, from which it follows that one's
choice of a theory has the logical status of convention [15, p. 621].
I will not say a thing about Schlick's beliefs. As regards Einstein's, he
offers arguments and assertions. He does not use the verb "to believe"
at all in his paper of 1919. So I do not know what is the source of
Howard's observation. Further, Einstein in fact does not assert positive
criteria, that is, neither does he assert how we do recognise a theory free
of errors (a logically consistent), nor a correct theory (a theory which
agrees with experience). Einstein, rather, argues that: "A theory can thus
be recognised as erroneous if there is a logical error in its deductions, or
as incorrect if a fact is not in agreement with its consequences." Thus
Einstein offers negative criteria. From this does not follow Howard's
claim "that one's choice of a theory has the logical status of convention".
I should elaborate here. In a (misleading) footnote in his [16], Howard
references Einstein [7] as a supporting source to the (false) claim that
Einstein subscribes to conventionalism. According to Howard, Einstein
"concludes : 'In my opinion, Poincare is correct sub specia aeterni, in this
conception' " [16, p. 380]. There is no contradiction here. Einstein seems
to hold that conventionalism in mathematics is right. But Einstein says
no to conventionalism in science. How was Howard led astray? It begins
with his presentation of Einstein's argument which is not true to the
text since Einstein does not "conclude" with this remark. The truth is
that in his [7], Einstein after suggesting that Schlick's descriptions of the
axioms of mathematics as "implicit definitions" is "apt", then Einstein
was suggesting "that mathematics as such cannot predicate anything
about perceptual objects or real objects" [7, p. 30]. Then Einstein turns
to describe his philosophy of science in contrast with Poincare's. Einstein
goes on to describe how geometry can do this job, i.e., to predict, with
the addition of the hypothesis: "Solid bodies are related, with respect to
their possible dispositions, as are bodies in Euclidean geometry of three

dimensions" [7, p. 32]. Then Einstein goes on to assert that geometry
with this added hypothesis should be distinguished from "purely ax-
iomatic geometry" and he baptised his version by the name: "practical
geometry" . The latter is characterised in the following: "The question
whether the practical geometry of the universe is Euclidean or not has
a clear meaning, and its answer can be furnished by experience". Then
Einstein says that he attaches special importance to his view of geometry
since otherwise he would not be able to arrive at relativity. Thereafter
he suggests that if we do not agree to his (Einstein's) understanding of
practical geometry, that is if we deny the connection between geometry
and reality, then we arrive at Poincare's conventionalism. This is the
pretext to Einstein's statement "Poincare is correct sub specia aeterni,
in this conception" [7, p. 35]. Yet, Einstein continues with his consider-
ation and assessment of Poincare's views, and he says "but" [7, p. 36],
and after it, he objects to Poincare's view by arguing that we are yet to
posses certain knowledge of theoretical principles [of scientific theories];
and Einstein then tackles a further objection to his realistic interpreta-
tion of physics, i.e., relativity, and then he extends his claim to other
geometries [7, pp. 37-38]. Einstein therein suggests that experience in-
deed serves the scientist in his assessment of his given theory, and hence
the choice of the axioms of scientific theory is not a matter of conven-
tion. Einstein only then arrives at a conclusion before illustrating it by
ample examples. His conclusion begins with these words: "Therefore the
question whether the universe is spatially finite or not seems to me a
decidedly pregnant question in the sense of practical geometry. I do not
even consider it impossible that this question will be answered before
long by astronomy" [7, p. 41]. I elaborated on this point elsewhere [1]
like Paty who argues that Einstein was not a conventionalist with a sim-
ilar argument [25]. Einstein repeats the same in his replies to criticism
[12]. I will turn back to Einstein's paper of 1919.
Einstein, in this paper [6], offers to the general public, for the first
time, to the best of my knowledge, his philosophical outlook. And since
in this Einstein's philosophy differs greatly from his earlier philosophical
outlook, I will not follow Howard's discussion which goes to an earlier
phase of the correspondences between Schlick and Einstein. I am reluc-
tant to do this because of another reason which is external to this issue.
Howard observes that Einstein dissents from the neo-Kantians philos-
ophy. But he does omit mentioning that Einstein corresponded with
Meyerson and Cassirer. Howard also omits mentioning that Cassirer's
philosophical manuscript on the philosophy of relativity was scrutinised
by Einstein. And that Cassirer and Einstein corresponded. This is impor-
12 A.M. ADAM
tant because Howard mentions that in Schlick's review of Cassirer's book

in which Cassirer attempted to reconcile relativity and neo-Kantianism,
this attempt was rejected by Schlick as untenable. Further, Einstein
praised Schlick's review. Question: Why did Einstein work on a philo-
sophical work, make corrections and comments, but not tell Cassirer that
Cassirer's views are untenable? Did Einstein need to be enlightened by
Schlick? This is a question at which Howard does not arrive? Why? Be-
cause, to begin with, Howard omits mentioning the exchanges between
Einstein and Cassirer.
The last issue I shall mention here in passing is that Einstein in fact
discussed the method used by the researcher earlier in his address to
Max Planck of 1918, and Einstein later continued to discuss the issues
of his published paper in his address to the Prussian Academy of Sci-
ences in which he comments directly on Schlick's conventionalist view
that the axioms of mathematics can be considered as "implicit defini-
tions." This is in the domain of philosophy of mathematics not in the
domain of philosophy of science. This, Howard omits to mention as well.
Einstein, as is indicated in the above, argues in this paper [7] against the
conventionalism of Poincare in science and finds it appropriate in math-
ematics. In this Einstein elaborates on the view presented in his paper
[6] with ample examples in the literature. The third philosophical paper
in which Einstein repeats these views with further elaboration is in his
famous Herbert Spencer Lecture in 1933 [10]. Einstein's other published
philosophical reflections are scattered all over his writings, for example,
in some of his contributions to Russell's Festschrift [28] and in several
introductions he contributed to books on history and philosophy of sci-
ence. He also mingled his philosophy with some popular and scholarly
presentations of his physics.
9. C 0 N C L US I 0 N
I primarily aimed in my paper to indicate that Einstein has a philosophy

and that his philosophy is a breakthrough in the debate between the ra-
tionalists and the empiricists, and that he considers the conventionalist
view and rejects it. His philosophy of science is critical rationalist and is
in opposition to the empiricist philosophy of Born who advocates quan-
tum physics. I also made here an attempt to show that Pais's scepticism
regarding Einstein's contributions to philosophy is untenable; and that
Einstein's contributions were significant.
I should perhaps mention that Jean Eisenstaedt asserts in a sweeping
claim that as regards the main points of Popper's philosophical pro-
gram: (whatever it means) it is predated by a philosophical argument
by Einstein:
It is certainly not by chance that Einstein stated the main points of
the Popperian program many years before the publication of Logik
der Forschung [1935], as early as 1919: 'But the truth of a theory will
never be proved. Because one can never know if, in the future, some
experience will become known that would contradict its conclusions
(Einstein 1919)'[14].
It is regrettable that Eisenstaedt neither pauses to elaborate once he
made an explicit strong historical comment, i.e., Einstein's philosoph-
ical view predates Popper's; nor does he say how significant Popper's
philosophy was, if at all. Nor does he say, If Einstein contributed so
much to an acknowledged philosopher's philosophy of science, how come
Einstein's philosophy of science is rarely discussed or why he himself does
not discuss it but reference to by passing. I shall discuss Eisenstaedt's
important observation in another essay.
AFFILIATION
The author is currently at the Edelstein Centre for Philosophy of Science,

Hebrew University of Jerusalem, Jerusalem, Israel.
Earlier versions of this paper were read on 2 February 1995 in Dort-
mund University: I am thankful to Professors, J. Stolz and A. Hirshfeld,
for comments and the latter also corrected some errors in the translation
of [6]. On 1 June 1995 it was presented in a short version to the Congress:
Einstein meets Magritte. On 14 July it was presented in Geneva Univer-
sity. On 21 November 1995 it was presented in Tel Aviv University. I am
indebted to extensive criticisms given by my friend Mr. Douglas Gardner
of York University and for comments given by Professors Max Jammer
and Menachem Fish and Yemima Ben Menachem. The translation of [6]
was done by my friends Bernd Fleischman, Derek Lutterbeck and my-
self. I am the sole responsible for the views and mistakes presented in
this essay. While writing this essay I was appointed as a Research Asso-
ciate in Geneva University and benefited from the Swiss Israeli Cultural
Exchange Grant.
14 A.M. ADAM
REFERENCES
[1] Adam, A.M., "Einstein, Michelson, and the Crucial Experiment

Revisited", Methodology and Science, 25, 1992, pp. 117-228.
[2] Bergman, G., "Book review" of [29] in: The Philosophical Review,
LX, 1951, pp. 268-274.
[3] Born, M., Natural Philosophy of Cause and Chance, Oxford, The
Clarendon Press, 1949.
[4] Born, M., Albert Einstein/Max Born Briefwechsel1916-1955, 1969.
[5] Cassirer, E., Zur Einstein'schen Relativitiitstheorie, 1921.
[6] Einstein, A., "Induktion und Deduktion in der Physik", Berliner
Tageblatt, (Suppl. 4),1, 25 December 1919.
[7] Einstein, A., Geometry And Experience Sidelights On Relativity
London, Methuen & Co., 1921.
[8] Einstein, A., Conseil de Physique Brussels Bulletin de la Societe
franr;aise de philosophie, 22, 1922, p. 112.
[9] Einstein, A., "A propos de La deduction relativiste, de M. Emile
Meyerson", in: Revue philosophique de la France et de l'etranger, 105,
1928, pp. 161-166.
[10] Einstein, A., "The Herbert Spencer Lecture" which was titled "On
the Method of Theoretical Physics" Oxford, The Clarendon Press, 1933.
[11] Einstein, A., "Remarks on Bertrand Russell's Theory of Knowl-
edge", in [28].
[12] Einstein, A., "Replies To Criticism", in [29].
[13] Einstein, A., Lettres A Maurice Solovine, Paris, Gauthier-Villars
1956.
[14] Eisenstaedt, J ., "Low Water Mark In General Relativity", in:
Howard, D. and Stachel, J. (eds.), Einstein and the History of Gen-
eral Relativity, Einstein Studies, 1, Boston, Birkhauser, 1989, pp. 277-
292.
[15] Howard, D., "Realism and Rationalism in Einstein's Philosophy
of Science", Philosophia Naturalis, 1984, pp. 617-629.
[16] Howard, D., "Einstein and Duhem", Synthese, 1990, pp. 363-384.
[17] Holton, G., Thematic Origins of Scientific Thought, 1973.
[18] Jaki, S.L., "The absolute beneath the relative: reflections on Ein-
stein's theories", in: Ryan, D.P. (ed.), Einstein and the Humanities,
New York, Greenwood Press, 1987, pp. 5-18.
[19] Mach, E., Erkenntnis und Irrtum. Skizzen zur Psychologie der
Forschung, 1905.
[20] Meyerson, E., The Relativistic Deduction, 1925.
[21] Nagel, E., a book review: "Einstein's Philosophy of Science", on
[29] in: The Kenyon Review, XII, 1950, pp. 521-531.
[22] Pais, A., Subtle is the Lord: The science and the life of Albert
Einstein, Oxford, UP, 1982.
[23] Pais, A., Einstein Lived Here, Oxford, UP, 1994.
[24] Poincare, H., Science and Hypothesis, 1902.
[25] Paty, M., Einstein philosophe: La physique comme pratique philoso-
phique, Paris, Presses Universitaires de France, 1993.
[26] Russell, B., The Problems of Philosophy, Oxford, UP, 1912.
[27] Russell, B., The Scientific Outlook, London, Allen, 1931.
[28] Schilpp, P.A., The Philosophy of Bertrand Russell, Evanston, Illi-
nois, 1944.
[29] Schilpp, P.A., Albert Einstein Philosopher Scientist, 1951.
[30] Zahar, E.G., Einstein Revolution, New York, Open Court, 1989.
DON FAWKES
EINSTEIN, ETHICS, AND ACTION
1. QUESTIONS OF ETHICS ARE RELATIVE,

AREN'T THEY?
On the 26th of February 1979, Time magazine [5] ran an advertisement

describing its coverage of the Einstein Centennial. With a magnificent
picture of Einstein's thoughtful, sensitive face covering most of the page
the lead read, "Everything is relative. In the cool, beautiful language
of mathematics, Einstein demonstrated that we live in a world of rela-
tive values." (p. 53) This claim must have come as quite a surprise to
those who have read Einstein's published works on ethics, because it
is contrary to them. (See, e.g., [1]) "Everything is relative" is a misun-
derstanding that infects popular accounts of Einstein's work in physics
no less than accounts of his views on morality. The Special and Gen-
eral Theories of Relativity produce objective laws of nature, confirmed
true for any reference frame. Reference frames (or "bodies"), of course,
can move in relation to each other, and when they do move, this affects
measurements of distance and time from one body to another; hence,
the name "Relativity." But the laws are quite objective, and confirmed
by experience. Einstein's view of morality is rather similar:
It is the privilege of man's moral genius, impersonated by inspired
individuals, to advance ethical axioms which are so comprehensive
and so well founded that men will accept them as grounded in the
vast mass of their individual emotional experiences. Ethical axioms
are found and tested not very differently from the axioms of science.
Truth is what stands the test of experience. [1, p. 115]
We will return to this general approach below. But before doing so we
need to build a little groundwork to show that both moral relativism and
moral absolutism are intellectually bankrupt. Interestingly, quite similar
arguments show both of these somewhat popular doctrines to be faulty.
We begin with the most popular form of absolutism, that which is based
on commands from authority.
2. COMMANDS
We will deal with the view that morality is based on authority in its
most persistent form: That God is the authority. If we can show the
17
18 DON FAWKES
flaws in this position, it is unlikely that "lesser" authorities will have

much chance of standing as the foundation of morality. We will not
discuss nor question the existence of a supreme being or supernatural
entities; consideration of those topics is properly left to metaphysics and
theology. Rather, we simply ask, "If there is a deity, what bearing can
this have on morality?"
There seem to be three possible relationships between a deity or deities
and morality (For ease of expression we will employ the singular term
"deity" from here forward; however, what follows applies equally, with
very minor modifications, to a multiplicity of deities or supernatural
beings). The three possible relationships are as follows:
DET: The deity wills right acts because these acts are right. (Divine
Expert Theory)
DCT: Right acts are right because the deity wills them. (Divine Com-
mand Theory)
NDRT: The deity wills nothing concerning people's actions. (No Divine
Relation Theory)
NDRT is consistent with deism, the opinion held by many, perhaps
most, of the "founding fathers" of the United States. Deism holds that
although a deity began the universe at a time long past, the deity has
no dealings with it now, and certainly no concern for human beings;
whatever it is that "began" the universe is the deity, the Creator. In
other words, there is no relationship between the deity and morality.
This, of course, is possible but we will have little further to say about
it, since our primary interest is to discuss the possible bearing of a deity
on morality.
This leaves us to consider Divine Expert Theory and Divine Command
Theory. If there is a deity, does the deity will that we do right acts
because these acts are right, or are these acts right simply because the
deity wills that we do them? What kinds of reasons could be given to
answer this question?-We can approach this question in two ways: (a)
We can give reasons to believe that one of the views is true; and, (b) we
can reason by finding grounds for believing that one of them is false.
An Argument in Favor of DET and Against DCT
The most obvious consideration in favor of DET is that it is compatible

with the concept that most believers have of the deity. Most believers
hold that the deity is the most good (most loving), most knowing, and
most powerful being. Many believers hold that the deity is somehow
infinite in these ways. It seems obvious enough that a most good and
EINSTEIN, ETHICS, AND ACTION 19
most knowing deity would know the right acts from the wrong ones, and
would will that we do the right ones (DET). But it does not seem to be
likely that a deity like this would command a set of acts saying that these
acts are "right" just because the deity says so (DCT); in fact, believers
often refer to a supernatural being who acts in this way by a name other
than "deity". Thus, it is compatible with this notion of the deity that
the deity would make decisions on the basis of goodness and knowledge;
but Divine Command Theory is incompatible with this concept of the
deity.
Ways to State Divine Command Theory and its Consequences
Before we proceed further, we should state some of the alternative ways

in which Divine Command Theory can be expressed, and consider what
are its consequences. As it is expressed above, DCT is a "metaphysical"
claim; it is a claim about reality, about the way things are. Now this
metaphysical claim also can be expressed "linguistically"; expressed as
a claim about the meaning of words. In linguistic form, DCT could be
rendered by some such words as: "The word 'right' simply means 'what
the deity commands.' "; or, "The entire meaning of 'Act X is right' is
'the deity commands that we do X.' " There may be some who would
wish to make a great fuss over this technicality, but our primary purpose
in bringing it up is to make sure that we are able to recognize DCT if it
is raised. We will continue to discuss DCT in both forms, and everything
that we say about it will apply (with perhaps very minor modification)
to either form. As for the consequences of DCT, they are the following:
(a) If DCT were true then no acts would be right unless a deity existed;
if there is no deity, then there is no morality. And, (b) no matter what
the deity commanded us to do, it would be right. For instance, if the
deity commanded that we slowly torture-to-death a two-year-old child,
or commit genocide, then these acts would be right just because the deity
commanded them.
Arguments Against Divine Command Theory
We can now turn to the problems with DCT. There are four major
arguments of this kind, and we can label them the problems of meaning,
autonomy, variety, and interpretation.
a. Meaning. If right actions are held to be what the deity commands,
then the result is a curious emptiness in the meaning of the term "right".
For now it has no meaning apart from whatever is commanded. Perhaps
"right" in this setting is a way to remind us that the deity has power;
20 DON FAWKES
"right" is just a word that refers to a command backed-up by a threat.

We may be reminded here of the politician who says he or she favors
"family values" without bothering to tell us that by "family values" he
or she means the "values" that he or she happens to like. In short, we
cannot identify the meaning of a claim merely by finding its source, or
who likes or commands it. And the difficulty actually is worse for DCT
than it is in the politician's case. For we can at least ask the politician
to clarify what he or she means by "family values." But in the case of
Divine Command Theory, the meaning of "right" is supposedly as clear
as it can get! DCT simply makes the use of the word "right" a verbal
flourish with no meaning at all; it amounts to the command, "Do what
the deity says."
b. Autonomy. Closely related to the problem of meaning is the problem
of autonomy. The autonomous person is one who is (1) free to choose, and
(2) who "knows the options." Hence, autonomy can be eliminated either
by deprivation of freedom or by ignorance. (Being autonomous usually
is a matter of degree. So we can describe a person as being more or
less autonomous in a given situation; or, as being more or less increased
or reduced in autonomy by circumstances or choice.) Autonomy seems
to be necessary for praise or blame. For instance, we would not usually
praise or blame a person for an act performed under coercion or out of
ignorance. There are, of course, exceptions to this, but it is important to
note that these exceptions involve showing that an appropriate degree
of autonomy (freedom and knowledge) is present. But perhaps the most
serious practical consequence of believing Divine Command Theory to
be true is that the moral agent who does so simply surrenders moral
autonomy. For, such a moral agent has given up the possibility that he
or she can choose or "reason-out" standards of right and wrong: The
freedom to choose is at least limited by the command-backed-up-threat;
but more importantly, there is nothing to reason about (the problem of
meaning). The agent can only try to find out (at his or her own peril)
what the commands are.
c. Variety. This leads to a third problem with DCT. It is obvious that
there are many groups in the world today, and throughout history, who
have claimed to have the true commands of the deity. It is notorious
that these groups differ widely over the nature of the deity as well as the
content of the commands. Thus, we are faced with a variety of possible
deities. Which shall we obey? Suppose that someone reasoned as follows:
The Bible (or Koran or Bhagavad-Gita, etc.) is the true source of the
deity's commands to people. I know that the Bible is the true source
because it is the word of the deity revealed through the prophets. I
know that these prophets are the true prophets because it says so in
the Bible.
This is a case of circular reasoning (begging the question). Wherever
we enter the circle we find that the very thing for which we need evidence
is simply assumed. This leads to the following considerations: When we
make a judgement concerning which possible set of commands we ought
to follow we are, in effect, making a moral judgement. But if Divine
Command Theory is true, no moral judgement can be made without
first having the deity's commands. Divine Expert Theory does not lead
to this situation, but Divine Command Theory does. And this is the case
no matter whether our "judgement" is based on reason, or sentiment, or
"upbringing," or emotion, or intuition, or ... and so on. Now we should
note that such a judgement is in effect a moral one because, once made,
it determines the rest of our "moral duties." Some may wish to say that
the judgement itself is not a moral one. But even granting this, if we
must somehow decide which claims to divine revelation are genuine and
which are false, then reason, or intuition, etc. is the final authority in
the decision. This is contrary to the notion that a particular revelation
provides knowledge that is absolutely certain just because it comes from
the deity. (We should not need here to take one more turn around a
circular reasoning track by introducing a notion of "divine guidance" or
"divine inspiration," etc.)
d. Interpretation. Even if we could solve the variety problem for DCT,
we would be faced with interpreting the commands. Some of the most
troublesome moral issues have to do with interpreting alleged commands.
We can find, for example, (in the U.S.A.) members of the Christian
Knights of the Ku Klux Klan and members of the Southern Christian
Leadership Conference marching, carrying Bibles, and each claiming to
have the true interpretation of the deity's commands. (But these inter-
pretations arc contrary to each other.) Even people who are members of
the same sect and who have a shared tradition often disagree concern-
ing how to interpret the set of divine commands that they believe are
genuine. Now this is also a difficulty for those who accept Divine Expert
Theory, but at least they can try to reason things out or consult their in-
tuitions, etc. But if we adopt Divine Command Theory, then how can we
interpret the commands? To interpret requires making moral decisions.
But to do that requires further divine commands, and so on for as long
as we need to interpret ... Here again we are reminded, as we were when
we considered the problem of meaning, that according to DCT "things
are already as clear as they can get."- At least they are as clear as
commands can make them.
22 DON FAWKES
Consequences of Divine Expert Theory
Having given argument that supports DET on the assumption that there
is a deity, and having given argument against DCT (again assuming that
there is a deity), we need to consider the consequences of DET, "The
deity wills right acts because these acts are right." Let us return for
a moment to the example of torturing to death a two-year-old child.
Following DET, a most good, most knowing deity would know that such
an act is cruel, and that cruelty is wrong, and therefore, would not
command it, but rather would forbid it. This means that the deity would
recognize the wrongness of cruelty, and this shows that cruelty would
be wrong whether or not a deity existed. Hence, contrary to Divine
Command Theory, the first consequence of Divine Expert Theory is (a)
that the existence of a deity is not necessary for morality. The existence
of a deity is compatible in this regard with morality, but not necessary
for there to be right acts. A second consequence of D ET is (b) that the
deity would not will that we do any wrong acts. This is so not because
what-the-deity-says-goes (DCT), but rather because a most good and
most knowing being would will that we do right acts because they are
right.
Further Consequences, Complications, and Conclusions
Now we will consider three further topics: (1) General objections to the
view that morality must be based on divine command; (2) the claim
that the goodness of the deity guarantees that DET is true and that
a version of DCT is true; and (3) some general ways to broaden our
understanding, of issues related to autonomy, religion, and morality.
1. General objections to the view that morality must be based on divine
command. We can take the view that morality must be based on
divine command as the position that "If there is no deity, then there
is no morality." We saw above that this would be a consequence of
DCT. But now we will examine this notion, "No deity implies no
morality" (NDNM), on its own. The first thing to say about NDNM
is to point-out that it seems difficult to imagine how we could believe
that NDNM is true without committing ourselves to believing that
DCT is true. The second thing to ask is, "What grounds are there
to believe NDNM?" Some people may have believed NDNM because
they believed DCT, but we have shown that there is overwhelming
evidence against DCT. So why would anyone believe NDNM? The
third thing to say about NDNM raises a difficult issue which affects
a wide variety of topics related to autonomy, religion, and morality;
it is this: If there is a most good, most knowing, and most powerful
deity who created the universe, then why is there so much evil (for
example, undeserved suffering) in the universe? Now we cannot even
begin to discuss this complex topic here, but one caution we can offer
is to be careful of the fallacy of begging the question when thinking
about this issue. Finally, we can point-out that there are versions
of the problems of meaning, autonomy, variety, and interpretation
which also apply to NDNM.
2. The claim that the goodness of the deity guarantees that DET is true
and that a version of DCT is true. A person who believes that the
deity's goodness guarantees that DET is true might also add that,
therefore, there is a kind of "round-about" sense in which DCT is
also true. For it might be said, that "right acts are right because
the deity wills them," and this is so because the deity's goodness
guarantees that the deity can only will that people do right acts.
This appears to be a complex position and the following complex
remarks address it: Suppose it is said that the deity's goodness
guarantees that the deity would not (for instance) command that
people be cruel. Now we can ask, "Is the deity good because the
deity's character meets a standard that is independent of what the
deity wills, or is the fact that the deity wills to have a certain kind
of character the standard of goodness itself?"-If you think that we
now face the same question (in a new form) with which we began
our inquiry into commands, you are quite right! (And perhaps a
little tired!)-For consider that the question can be answered in
one of two ways:
a) The deity is good because the deity is, for example, benevolent,
merciful and just. But then are benevolence, mercy and justice
genuine standards of goodness? If so, then they are standards
which are independent of the deity's will. And thus, these are
genuine standards even if there is no deity. So, we are led back
to DET and its consequences.
b) On the other hand, if benevolence, mercy and justice are gen-
uine standards of goodness just because the deity wills that
they be part of the deity's character, then if the deity were
to will malevolence, ruthlessness, and injustice be part of the
deity's character, then these would be the genuine standards of
goodness. Thus we are led back to DCT and its consequences,
and its problems. Thus, we can see that nothing is added to our
discussion above by the claim that the goodness of the deity
guarantees that DET is true and that a version of DCT is true;
24 DON FAWKES
although, it is complicated (and perhaps somewhat tiring) to

show this. This does not detract from the point we made ear-
lier that if there is a most good deity then that is compatible
with DET.
3. Some further general points. Easy answers are not available to ques-
tions about autonomy, religion, and morality. Now we will address
a few ways in which our understanding can expand. The first thing
we need to note is that the arguments given above are compatible
with the possibility that (Y) 'the set-of-all-right-acts' is identical
to (Z) 'the set-of-all-acts-commanded-by-the-deity'. Such an iden-
tity of sets is sometimes called extensional equivalence, and this just
means that every member of the first set is also a member of the
second set, and that every member of the second set is also a mem-
ber of the first set. We have not argued that set (Y) is extensionally
equivalent to set (Z), but only that this is possible, and compatible
with DET. Second, we should remind ourselves that we have not
addressed the question of the existence of a deity, nor the question
concerning the existence of so much evil in the world. We only asked
"If there is a deity, what bearing can this have on morality?" Third,
we should also remind ourselves that we only defined deism, and
did not discuss the grounds which might be given for or against it
or similar views. Fourth, we should keep in mind that many of the
beliefs we have concerning autonomy, religion, and morality were
acquired at an age and in circumstances which were not likely to fos-
ter careful or reflective thinking. We should recognize that these are
not the best conditions in which to acquire knowledge (knowledge
which can importantly affect our lives); and thus, we should resist
the natural inclination against reflectively reevaluating these beliefs.
Fifth, we should note that the term "religion" could be misleading,
for we have only addressed those religions which are theistic (i.e.,
those that assert that there is a deity). But there are non-theistic
and/ or agnostic religions (such as Theravada Buddhism, U nitari-
anism, Taoism, and Confucianism) which we have not addressed at
all. Sixth, we have largely ignored a considerable body of analysis
among those who have studied the anthropology, sociology, psychol-
ogy, and mythology of religions. This body of analysis holds that
the relation between religion and morality is entirely different from
that which we have examined: Instead, it is said, that religion is
a social activity that gives dramatic and mythical expression to a
culture's "basic values"; that religion involves stories or parables
that express a culture's "world view," and "attitude toward life and
death," and "sense of mystery or transcendence" as well as lists of
"do's-and-don'ts"; that these stories and activities show the mean-
ing and social importance of histories, of "sacred texts" and "holy
places"; and, that religion may embody and call for the celebra-
tion of a culture's art and literature. Finally, we have given little
attention to understanding the concepts of "right" and "wrong",
and "good" and "evil." We have been concerned, primarily with
removing an impediment to our consideration of those issues. That
impediment is contained in DCT and NDNM. We can now get on
with the business of discussing ethics.
3. WHY ETHICAL RELATIVISM IS FALSE AND

WHAT IS TRUE ABOUT IT
Since Einstein's work can be represented only falsely as promoting rela-

tivism in ethics, what has motivated this popular view? It may be that
thinking people, who sensed the weaknesses of Divine Command The-
ory, were impressed with the findings of social scientists. These findings
revealed widely variant codes of morality among different cultures, soci-
eties, and even family groups. So, the conclusion that morality is relative
to one's family, or culture, or society seemed appropriate. But there is
a fatal flaw to this line of thought: We may as well conclude that geo-
graphy is relative to one's family, or culture, or society from the premise
that some families, cultures, or societies believe that the earth is flat,
while others believe that the earth is round.
However, there are several senses of "relativism" in ethics and these
warrant review concerning (a) whether they are true, and (b) whether
they are relevant to the possibility of objective ethical truth, confirmed
by experience. These senses of relativism are as follows:
(ri) What one person, family, culture, or society believes to be right
often differs from what other persons, families, cultures, or societies
believe to be right.
(rii) What is right in one situation or circumstance differs from what is
right in another situation or circumstance.
(riii) What is right is determined by one's family, culture, or society.
(riv) The meaning of terms like "right," and "good," etc. can only be
determined by reference to one's family, culture, or society.
By now there is an extensive body of evidence to show that (ri) is true.
But, as we already have observed, this is simply irrelevant to the ques-
tion of whether there are objective, principles of morality: The fact that
people, or families, or cultures, or societies disagree about something
26 DON FAWKES
does not show that there is no objective truth about it. But further-
more, concluding that "morality is relative" on such grounds assumes
that all moral propositions are on the same level of specificity; and, that
assumption is simply false. For, consider the following hypothetical ex-
ample: Suppose that we encounter two cultures with very different moral
rules concerning human reproduction. In culture A the rules encourage
the increase of population as rapidly as possible; in culture B reproduc-
tion is strongly discouraged. But suppose we then observe that in B the
ratio of population to resources is very high; while in A this ratio is very
low. We then further observe that the members of A and B agree com-
pletely concerning the preservation and enjoyment of people's lives. The
apparent moral disagreement now can be understood as a difference in
circumstances combined with a complete agreement about ethics. Having
said this much, what about (rii)?-Well, suppose that we demonstrate
(see section 4 below) that objective reasons show that lying is wrong
and truth-telling is right. Does this mean that it would be wrong to lie
in order to save the life of a Jew being pursued by Nazis?-Countless
examples like this (at every level of specificity) show that there are no
absolute nor universal moral rules. So, (rii) is not only true, but also a
significant part of morality. Now consider (riii). A little reflection will
make plain that (riii) is just another version of the view that morality
is founded on authority; (riii) replaces the deity's commands with those
of society, etc. So, we need not consider (riii) any further. As for (riv),
it is just the "linguistic" equivalent of the "metaphysical" (riii); so once
again, our work is done. (There are several more difficulties for (riii)
and (riv), and a concise exposition can be found in [2, pp. 160~172]. To
sum up then, our conclusions are that (ri) is true but irrelevant to the
question of whether ethics can be objective; that (rii) is both true and
relevant to morality; and that both (riii) and (riv) are false.
4. EINSTEIN, MAGRITTE, AXIOMS, AND PRACTICE
Writing of Magritte, Suzi Gablik says,

People have always looked for symbolic meanings in Magritte's pic-
tures, and in some cases managed to find them. Nothing caused him
greater displeasure. [3, p. 11]
She then quotes Magritte (in personal conversation) as follows:
To equate my painting with symbolism, conscious or unconscious,
is to ignore its true nature ... People are quite willing to use objects
without looking for any symbolic intention in them, but when they
look at paintings, they can't find any use for them. So they hunt
around for a meaning to get themselves out of the quandary, and
because they don't understand what they are supposed to think when
they confront the painting ... They want something to lean on, so they
can be comfortable. They want something secure to hang on to, so
they can save themselves from the void. People who look for symbolic
meanings fail to grasp the inherent poetry and mystery of the image.
No doubt they sense this mystery, but they wish to get rid of it. They
are afraid. By asking 'what does this mean?' they express a wish that
everything be understandable. But if one does not reject the mystery,
one has quite a different response. One asks other things. [3, p. 11]
So I will not ask that anyone find a meaning in the images I am about
to suggest, but only recommend that one contemplate the images along
with the prose to follow. Perhaps the images can play as a heuristic. It is
not necessary to reject the "mystery of the image" in order to see such
a heuristic role. One can bring a meaning of one's own to an image, and
appreciate the mystery too (I suppose Magritte brings some of his own
meaning to his works by giving titles to them). Consider then: Evening
Falls, 1964 [3, fig. 62].-I suggest, "Ordinary morals remain mostly only
with dogmatic structures shattered." And consider: The Two Mysteries,
1966 [3, fig. 110].-I suggest, "Description falls short of image, but has
its roles; it exists too. Image falls short of object ... Object falls short
of description ... Think of practice as object here." These are specula-
tions that I bring to the paintings, expressed allegorically for heuristic
purposes.
We can return from speculation to Einstein for a concrete example of
the use of "principles" or "axioms" in the understanding of moral rules .
. . . we do not feel at all that it is meaningless to ask such questions as:
"Why should we not lie?" We feel that such questions are meaningful
because in all discussions of this kind some ethical premises are tacitly
taken for granted. We then feel satisfied when we succeed in tracing
back the ethical directive in question to these basic premises. In the
case of lying, this might perhaps be done in some way such as this:
Lying destroys confidence in the statements of other people. Without
such confidence, social cooperation is made impossible or at least
difficult. Such cooperation, however, is essential in order to make
human life possible and tolerable. This means that the rule "Thou
shalt not lie" has been traced back to the demands: "Human life
shall be preserved" and "Pain and sorrow shall be lessened as much
as possible."
But what is the origin of such ethical axioms? Are they arbitrary?
Are they based on mere authority? Do they stem from experiences
of men and are they conditioned indirectly by such experiences?
28 DON FAWKES
For pure logic all axioms are arbitrary, including the axioms of ethics.
But they are by no means arbitrary from a psychological and genetic
point of view. They are derived from our inborn tendencies to avoid
pain and annihilation, and from the accumulated emotional reaction
of individuals to the behavior of their neighbors. [1, p. 115]
So, the rules need not be all on the same level of specificity. More gen-
eral rules may be called "principles", and these can serve the purpose of
helping to keep the more specific rules consistent with the point of moral-
ity. They also can help us to apply the more specific rules to concrete
cases, and help to eliminate apparent conflicts. But the entire structure
is based in experience.
Think of morality as a game, a social game. It is partly constituted of
rules in varying degrees of specificity. Applications of the rules depend
on circumstances. Rules apply in some situations and not in others, and
there is always space for originality. But the morality game has not only
rules and circumstances, but also a point. [6, p. 150] That point is to
promote the common good.
Without apprehending the point of morality we can get only rules that
are mere commands of authority or whims of culture. Just as morality
has nothing to do with the cynicism of ethical relativism, so also does it
have nothing to do with pious moralizing backed by commands. Moral-
ity has to do, instead, with careful recognition of natural facts, and with
recognizing and attempting to deal with real problems, like poverty in
the midst of plenty; and yet, morality docs not posit any utopia. Moral-
ity, instead, charges us to work, to strive, and to enjoy our efforts in the
knowledge that, although we may achieve much, those who may follow
us will find more to do. The choice is ours, and it is the same for each
of us in every age, every culture, and every generation: We can delude
ourselves in escapist cynicism, or pious certainty, or comforting confor-
mity; or, we can face ourselves and the natural world, and get on with
a moral life. It is in the (at least tacit) context of this understanding of
morality that interpersonal relationships, and intercultural relationships
are possible, beyond force competition. It often can be of help, when
we are unsure of a judgement, to recall the point of the game. A little
reflection should show how far removed is this understanding of morality
(though it is only a sketch) from that of either moral relativism or moral
absolutism. But what they say is likely to be what we shall get, if we
cannot make clear the point of the game.
There is a saying that "Too great a sense of identity makes a man
feel he can do no wrong; and, too little does the same." The human
history of religious and ideological violence shows that identification with
either the appeal to absolutism or to relativism produces people who
think they can do no wrong. In fact, it is surprising how often these

dogmas combine in curious ways to produce people who really believe
that they can do no wrong, and that they can do nothing else. But this
result, perhaps, should not be surprising: The person who really believes
that values are relative, and/or that values come from the command of
authority has no way of dealing with moral disagreement other than
a "flight or fight" response. For, with either dogma, there is nothing
to discuss: One cannot reason with commands but only follow or not
follow them; one cannot reason with "relative values" but only note their
differences. Hence, one can only terminate interaction by "flight," or
"resolve" moral disagreement by force; and so, the characteristic pattern
of interpersonal and intercultural interaction becomes force competition.
Those who fall victim to dogma, once or twice, all too often also fall
victim to the demands of that primitive reptilian-brain, the "R-complex"
[4, p. 55 ff.] within each of us, the seat of aggression. When the cerebral
cortex is crippled by dogma, the primitive brain takes over. Yet our
growing destructive abilities and the growing need for environmental
cooperation make it more obvious each day that it is in our individual
and collective interest to get beyond force competition. But if morals
are believed to be "relative" or believed to be "founded on commands,"
then no ethical context is available for interpersonal or intercultural
interaction beyond force competition. It is here that ethical theory is
highly relevant to ethical practice. If we cannot find a reasoned ethical
context for interaction, we are practically doomed. So, a solution to this
ancient, philosophers' puzzle has become a practical need no less than an
intellectual challenge; and yet, if we are lucky, practice may be moving
ahead of theory. To this both Einstein and Magritte would surely raise
a cup. Truth is what stands the test of experience.
AFFILIATION
Don Fawkes
Fayetteville State University
Fayetteville, North Carolina
U.S.A.
REFERENCES
[1] Einstein, A., "The Laws of Science and the Laws of Ethics", 1950, as
found in: Einstein, A., Out of My Later Years, Secaucus, The Citadel
Press, NJ, (Rev. ed.), 1956.
[2] Feldman, F., Introductory Ethics, Englewood Cliffs, Prentice Hall,
NJ, 1978.
30 DON FAWKES
[3] Gablik, S., Magritte, Little Brown and Company, Boston 1976.
[4] Sagan, C., The Dragons of Eden, Random House, New York, 1977.
[5] Time Magazine, 26 February 1979.
[6] Wittgenstein, L., Philosophical Investigations, Macmillan, New
York, 1958.
MAREK W. BIELECKI
COGNITIVE SCIENCE: TWO CULTURES OR ONE?
1. THE NATURE AND FUNDAMENTAL ASSUMPTIONS

OF COGNITIVE SCIENCE
Cognitive science, a mixture of psychology, computer science, neuro-

physiology, linguistics and philosophy has been hailed by many as the
first genuine science of cognitive phenomena. Its composition signals
that the discipline employs methodologies of the 'two cultures,' of the
natural sciences represented most clearly by neurophysiology and of the
humanities represented by philosophy. Cognitive science, thus, becomes
an interesting new area for the Erkliiren- Verstehen controversy. There
are voices which, citing this methodological eclecticism, deny cognitive
science the status of a separate discipline. Others disagree arguing that
a 'naturalized' philosophy is not much different from psychology or even
neuroscience. Many researchers believe that there is a central idea that
unifies all the different approaches belonging to cognitive science. It is
the idea represented by the model of algorithmic processing of informa-
tion with all its related concepts and techniques. Thus the main goal of
cognitive science in this view is to understand and explain the nature
of perception, memory, learning, linguistic abilities, etc. by representing
them in terms of their underlying computational processes.
1.1. Philosophical Roots of Cognitive Science
The dominant paradigm is not without serious difficulties, however, as

it embraces one of the philosophical perspectives in the perennial con-
troversy over what it means to be human. This perspective is part of a
rich philosophical tradition going back to the ideas of Plato, Descartes,
Hobbes, and Leibniz, which contributed to the rise of modern science. It
is already in Plato where we find the view that reasoning is a rule driven
process and that reality has essentially the mathematic-logical charac-
ter. Descartes can be attributed with making the claim that is central
to cognitive science, that all of our knowledge is given by means of men-
tal representations that reside in the mind and somehow correspond to
elements of the external world. Thinking, according to Descartes, is the
process of manipulating these representations according to certain, pri-
marily logical rules. In Hobbes we find the idea that 'reason is nothing
31
32 MAREK W. BIELECKI
but reckoning', i.e., a kind of calculation performed on general symbols

stored in the mind. Finally, from Leibniz comes the idea of a universal
symbolic language whose simplest elements represent basic features of
the world. All of these ideas lie at the very heart of modern science.
The more recent origins of cognitive science can be found in the work
of Frege, Russell and Whitehead on the foundations of logic. Among
the immediate precursors of cognitive science are also the philosophical
assumptions of logical positivism and even, according to H. Dreyfus [1],
Husserl's phenomenology, which can be viewed as the culmination and
closure of the Cartesian tradition in philosophy.
1.2. Assumptions of the Standard Model of Information Processing
The stimulating role of the model of mind as an algorithmic machine

manipulating symbolic representations has lead to establishing cognitive
science as an independent domain of research. Newell & Simon [2] have
tried to capture the key features of this model in the idea of a "physical
symbol system," which they define as a "machine that produces through
time an evolving collection of symbol structures" and which, they claim,
"has the necessary and sufficient means for generating intelligent action."
The basic assumptions of this model can be summarized as follows:
i) information exists in all processes and structures, such as the DNA,
where regularities are present, and information processing consists
in modification of such regularities;
ii) the information processing that takes place in such systems as the
brain-mind or the computer is algorithmic in nature, i.e., it involves,
as in logical systems, the formal process of manipulation of unin-
terpreted symbols in terms of unequivocal rules that refer only to
geometric shapes of symbols and not to any of their possible mean-
ings;
iii) the sequences of symbols and procedures for their manipulation are
capable of representing aspects of the external reality, which enables
the system in which the processing takes place to generate behavior
that is correlated with these aspects;
iv) the sequences of symbols are capable of representing other such
sequences, which leads to the emergence of dynamic memory and
recursive processes that are crucial for intelligent behavior; and fi-
nally
v) information processing can and should be studied without reference
to the physical structure of the system that does the processing.
Other frequently adopted assumptions that are claimed to char-
acterize this model include the thesis about intentionality of such
COGNITIVE SCIENCE: TWO CULTURES OR ONE? 33
processes and about their "functional modularity," i.e., their de-

composability into functionally simple components, and a related
meta-thesis about the "functionalist" method of explanation that is
said to be the only proper for such systems. These assumptions are
even more controversial than the basic five and belong to the area
of study that can be called "the foundations of cognitive science."
This area is almost exclusively the domain of philosophy.

Since its inception the computational model of information processing
has played an important role in linguistics, psychology, neuroscience, and
philosophy. In linguistics it has influenced Chomsky's idea of Universal
Grammar and more recently lead to the emergence of computational
linguistics. In computer science it has lead to the emergence of artifi-
cial intelligence (AI) as an independent subdiscipline. In psychology it
has paved the way to the rise of cognitive psychology after a period
dominated by behaviorism. In neuroscience it has influenced the models
explaining how the brain processes sensory stimuli. Finally, in philos-
ophy, according to some, it has helped to find solutions to seemingly
intractable problems arising in connection with attempts to explain the
nature of thought processes. One of such celebrated problems is the so
called "ghost in the machine" problem. D. Dennett calls it "Burne's
problem" and offers his own, he believes more satisfactory solution than
that of Humc [4].
Some philosophers consider the attempts to apply the computational
model of the mind to tackling philosophical problems completely mis-
taken and not even worth refutation. One of few exceptions is H. Dreyfus
[5]. His criticism, however, does not extend to the scientific method as
such, as he accepts the view that by studying the brain we may get a
better understanding of the nature of mind. Similar views are expressed
by .J. Searle [6], who additionally thinks that we need to modify basic
scientific and philosophical conceptions about the nature of causality. A
more deeply sceptical position is adopted by T. Winograd [7], who in-
spired by the ideas of a biologist and system theorist, H. Maturana paints
a bleak picture of the traditional research effort in cognitive science.
Regardless, however, of whether one is optimistic or not about the
prospects of current research trends in cognitive science, its influence on
philosophy cannot be denied. Part of this influence stems from provid-
ing philosophy, especially philosophy and methodology of science with
new models, e.g., models of intertheory relations and models of cross-
disciplinary endeavors. More importantly, however, because philosophy
is itself a participant in this endeavor, the philosophy's own identity,
the way it poses and tries to answer "philosophical" questions has been
influenced. P.M. Churchland writes for example, 'Very shortly it will

no longer be possible to do major work in the philosophy of science
without drawing on themes from the philosophy of mind and from the
related disciplines of computational neuroscience, cognitive psychology,
and connectionist AI,' [8, p. xiv]. It is thus worthwhile to take a closer
look at the place and role of philosophy in the scientific endeavor known
as cognitive science.
2. PH I L 0 S 0 PH I CAL AND MET H 0 D 0 L 0 G I CAL

PROBLEMS OF COGNITIVE SCIENCE
The domain of the foundations of cognitive science is frequently char-

acterized by simply listing the problems that fall in its scope. These
problems can be divided into three broad categories: methodological,
ontological and epistemological. The central methodological problems
include: the nature of the research method proper to cognitive science,
its method of explanation and the scope of its studies, the nature of its
fundamental concepts and theories, and the relationship between various
concepts and theories of the disciplines that constitute cognitive science.
The fundamental ontological issues include: the nature of entities and
constructs (such as the physical symbol system, an information process-
ing system and symbolic representation) and processes (e.g., interpreta-
tion and algorithmic processing) that cognitive science presupposes and
their relationship to commonsense concepts of folk psychology and to
physical or biological structures and processes. Finally, the fundamental
epistemological issues include: the nature of knowledge, the methods of
processing and applying knowledge by intelligent systems, the mecha-
nisms of memory, learning and concept formation.
As it almost always happens in philosophy, all these methodological,
ontological and epistemological issues are interdependent. For example,
the central problem that lies at the heart of cognitive science, the prob-
lem of the origin and nature of intentionality cuts across all three cate-
gories.
2.1. Methodological Issues
One of the main methodological issues in cognitive science is the problem

of its proper method of explanation. After the failure of behaviorism and
mentalism the only remaining candidate is functionalism. According to
functionalism, cognitive states are just functional states of sufficiently
complex systems. Each functional state of the cognitive system is defined
by reference to the system's specific mode of processing inputs into out
puts. In contrast to behaviorism, in functionalism cognitive states are

not just equated with the system's observable behavior, but instead they
are identified by the causal relations that they bear to each other and
to inputs and outputs. On the other hand, in contrast to the materialist
'identity theory', functionalism allows different physical systems, e.g., a
brain and a computer, to be in the same functional states.
The explanation of the cognitive system's, such as a brain or a com-
puter, ability to exhibit intelligent behavior invokes the idea of "func-
tional modularity," i.e., the idea of a hierarchy of processes (Dennett's
'intentional systems') beginning with the highest level processes which
correspond to directly observable intelligent behavior, through the 'semi-
intelligent', constituent subprocesses, down to 'unintelligent' subsubpro-
cesses residing at the lowest level. As a result of such functional decom-
position Ryle's ghost is supposed to 'get exorcized' from the machine.
This model of explanation encounters several difficulties. One of them
concerns the specific method that is used to represent information in
cognitive systems. The model most frequently adopted, according to
which information is represented as sentences, meets the criticism that
it cannot account for the observed high speed of processing complex
information. P.M. Churchland suggests a replacement of the outdated
Deductive-Nomological model of explanation with his alternative, con-
nectionist PA ('Prototype Activation') model which rests on the idea of
activations in neural networks [8].
The version of functionalism that assumes that the processes that
transform functional states are algorithmic is called 'computationalism'.
Intelligence and thinking on this view are a product of algorithmic, most
likely very complex processes. According to the famous Church-Turing
thesis any arbitrary algorithmic process can be simulated by a universal,
symbol manipulating system, such as the Turing machine. Because the
computers embody the idea of the Turing machine, they are claimed to
be able to simulate thought processes taking place in the minds, whether
human or alien. This perspective generates a number of philosophical
questions, such as whether a perfect simulation is the same as duplica-
tion and how to even tell one from the other. This problem involves the
methodological status of the Turing test which was considered by its au-
thor to be the only meaningful way to approach the question of machine
intelligence. These features of computationalism have become a source
of arguments aimed at showing that functionalism is untenable. One of
the most celebrated such arguments is known as the Searle "Chinese
Room". Searle's argument is directed not just against functionalism but
against the whole idea of the Turing test. It consists in providing an
example of a system that satisfies the conditions of the Turing test but
cannot be called intelligent or even understanding of what it is doing.

This argument is meant to support Searle's main philosophical thesis
that semantics (meaning) cannot be obtained from syntax alone, i.e.,
by means of syntactically specified manipulation of formal symbols [9].
Arguments for and against Searle's example play an important role in
debates in the foundations of cognitive science. In such debates there
is a frequent reflection of the realism-instrumentalism controversy con-
cerning the status of theoretical concepts and abstract models, and a
controversy about the ontological status of functional states and of the
commonsense concepts of folk psychology.
2.2. Ontological Issues
At the roots of almost all methodological problems related to cognitive

science, including Searle's thesis about the relationship between syntax
and semantics, reside ontological and epistemological issues, such as the
nature of language and its role as a representation tool and the origins
of intentionality of information processes.
J. Searle, for instance, in spite of his opposition to the idea of reduction
of semantics to syntax, maintains that thinking is a strictly biological
phenomenon no different than, say, digestion. Thus, in order for a system,
whether natural or artificial, to have mental states it is necessary that it
possesses 'causal powers' at least equivalent to those of the human brain.
The running of a program, no matter how complicated, on even the most
powerful computer can never produce the necessary causal powers.
The attempts made so far to solve the mystery of intentionality by
invoking the causal attributes of material substances have proven un-
successful. Even more sophisticated attempts have failed, such as those
of D. Armstrong [10] to reduce intentionality of thought to intentionality
of action, and this in turn to behavior of systems capable of homeostasis.
The problem of intentionality remains thus the central unsolved philo-
sophical problem in the foundations of cognitive science as well as in the
philosophy of mind and language. It clearly indicates the need for a new
approach that would be both acceptable to 'non-naturalized' philoso-
phers and useful to cognitive scientists as a set of guiding principles in
working out models of cognitive systems.
Among other ontological problems there are: the problem of ontolog-
ical status of commonsense concepts such as belief, doubt, desire, etc.
and their relation to the cognitive system's functional states and algo-
rithmic processes, the problem of 'qualia', i.e., the problem of content
of perceptions, and the problem of 'propositional attitudes', i.e., of cog-
nitive states such as belief, doubt, hope, desire, etc. whose content is
represented by assertions. All these problems are related to the problem

of intentionality. They also have an epistemological dimension as they
involve the issue of the nature and representation of knowledge.
2.3. Epistemological Issues
In philosophy the central epistemological problem is the problem of the

nature, structure, origins, and growth of knowledge and justification
of beliefs. In cognitive science the central epistemological issue is the
problem of the representation of different kinds of knowledge and the
problem of the mechanism of learning. Traditionally philosophers have
distinguished two fundamental kinds of knowledge, 'declarative knowl-
edge' which is the 'knowledge that ... ', e.g., that the computer is an
algorithmic machine, and 'procedural knowledge' which is the 'knowl-
edge how ... ', e.g., how to use a computer. Additionally, some cognitive
scientists found it useful to introduce the notion of 'tacit knowledge',
i.e., unconscious declarative knowledge.
G. Ryle [11] claimed that all types of intelligent actions such as prob-
lem solving or language use involve procedural rather than declarative
knowledge and argued therefore that the former must be more funda-
mental than the latter, i.e., that the latter must presuppose the former.
Because the assumption that intelligent action involves only declarative
knowledge would lead immediately to an infinite regress of explanations
of intelligence in terms of intelligent ability to use declarative knowledge.
A standard answer to Ryle's argument provided in cognitive science in-
vokes the idea of tacit knowledge and the idea of a hierarchy of infor-
mation processes which at the most fundamental level are wired into
the structure of the cognitive system and thus executed automatically
without the necessity of making intelligent or conscious decision.
These ideas, including the idea of tacit knowledge, do not help to solve
the problems of knowledge acquisition (or learning) and of belief fixation.
The latter problem is also known as the frame problem. According to J.
Fodor [12] the frame problem is the problem of finding a nonarbitrary
strategy for restricting evidence in the course of rational belief fixation.
A related problem is the problem of learning, especially of genuinely
novel concepts. If learning, Fodor maintains, entails either establishing
or failing to establish belief in a preexisting concept or hypothesis then
that new concept must already be present prior to the beginning of the
learning process. Thus learning, he argues, cannot create new represen-
tations, but only varying combinations of preexisting basic representa-
tions. Therefore concepts must be innate and represented in an inner
language of thought. The language of thought hypothesis is, however,
hard pressed when it comes to explaining perception and the processing

of sensory data.
The problem of finding the most efficient representations for different
kinds of knowledge, especially sensory information, and for their retrieval
methods has lead in recent years to the emergence of a new paradigm in
cognitive science. This paradigm was initially called 'parallel distributed
processing' ('PDP') and is currently known as 'connectionism'. The coex-
istence in cognitive science of two competing paradigms, connectionist
and symbolic AI leads naturally to questions about philosophical as-
sumptions adopted within each approach and about the perspective of
future research envisioned in each of them.
3. C 0 N N E C T I 0 N I S M V S . S Y M B 0 L I C A I
The standard (paradigmatic) example of information processing adopted

in the traditional (symbolic processing) paradigm is problem solving,
where the 'problem space' can be represented symbolically and where
the search for a solution consists in searching for a path in the problem
space leading from an initial state to a final, solution state. The ability to
use natural language, and other cognitive phenomena, are represented in
this paradigm as problem solving tasks. In the case of linguistic behavior
the rules for generating such behavior are simply the rules of grammar,
and the basic units of information to which the rules are applied are
words or sentences. In this model, it is assumed that the rules are rep-
resented in the form of a program executed in the brain or computer
each time the language generates or processes linguistic information. It
is also frequently assumed that most of the knowledge is stored in the
mind in the form of sentences and that such knowledge can be manip-
ulated algorithmically. Many cognitive scientists believe that in general
thinking has the same linguistic character.
On the other hand, the proponents of connectionism argue that the
fact that people are frequently unaware of the rules of grammar they
obey proves that contrary to what the classical approach assumes these
rules must be tacit and thus the information processed by such rules
cannot have a symbolic nature. Additionally, they point to cognitive
phenomena such as perception which pose a real challenge to the clas-
sical approach. They postulate therefore a replacement of the classical
model that represents thinking as an algorithmic process of symbolic
information processing with the model which draws its inspiration from
neurophysiology and adopts the view that knowledge is stored in the
brain-mind as activation levels of groups of neurons and that information
processing has a 'subsymbolic' character. The seminal work in which the
ideas of connectionism were first fully articulated was published in 1986

[13], but the precursor ideas can already be found in the work of such
neurophysiologists as, Rashevsky, Rosenblatt, Konarski, and Grossberg.
According to some philosophers the connectionist approach offers the
most adequate explanatory model of cognitive phenomena, the best foun-
dation for attempts to 'naturalize' epistemology, and the best approach
to a number of purely philosophical problems [8]. However, anyone who
appreciates the depth and seriousness of the problem of intentionality
has to approach such claims, especially the last one, with great skep-
ticism. My view is that the conceptual difficulties encountered by both
paradigms, in particular their inability to explain the nature of 'unsuper-
vised learning' and concept formation, the key processes which separate
natural cognitive systems such as humans from artificial ones, make it
unlikely that these paradigms will be capable of continuing to provide
useful explanatory principles in cognitive science. The shortcoming of
the classical paradigm, of symbolic information processing is that it re-
duces cognitive systems to their formal structure, and thus it tends to
view problems in philosophy and other cognitive science disciplines as
logic problems. On the other hand, the connectionist paradigm reduces
cognitive systems to neural 'black boxes' whose internal workings are
mostly irrelevant, except for the final distribution of node activations.
This approach relegates philosophy entirely from the picture and views
cognitive science as a subdiscipline of neuroscience.
4. TOWARDS NEW FOUNDATIONS

FOR COGNITIVE SCIENCE
Besides the two leading paradigms there is one other approach worth
to be mentioned here. It is the 'emergent computation' approach (14],
which adopts the view that an emergent pattern of behavior can be inter-
preted as processing information. Its modeling tools of choice are cellular
automata, genetic algorithms, and adaptive neural nets. This approach
comes closer than the classical ones to capturing the dynamic, adaptive
nature of natural cognitive systems. This robustness comes at a price,
however. It is not clear at all how the emergent transient patterns can
be harnessed to perform specific, repeatable tasks of information pro-
cessing. The main stumbling block in this and in all other approaches
adopted in cognitive science and other scientific disciplines is the in-
correct notion of information they presuppose. This is a result of their
inability to tackle the problem of intentionality (meaning).
There exists at present only one theoretical perspective that is robust
enough conceptually to provide a new general framework for cognitive
science. It is the conceptual framework of systems theory, especially in

its version developed by N. Luhmann which is rooted in the theory of
autopoiesis and phenomenology. Luhmann maintains that the identity of
a particular knowledge group or research field consists in 'its particular
way of solving insolvable problems, ... ', p. 81 [15]. For the 'hard' sciences
such core problem is 'the complexity of complexity,' for the 'soft' sciences
(the humanities) it is 'the meaning of meaning.' There are two types of
complexity, one based on operation and the other on observation, each
a result of 'enforced selectivity.' At the same time, complexity enforces
selection, because the system has to choose only some among many
possible relations it could possibly establish. By 'operation' he means any
process involved in the system's reproduction, by 'observation'-any act
of distinguishing for the creation of information. Meaning is a property of
dynamic systems only, and it can be described in the first approximation
as the link between the actual and the possible. Because the actuality is
unstable, it requires enforced selectivity of one of many possible futures,
i.e., it requires the focusing of attention. Meaning is thus nothing but
a 'way to experience and handle enforced selectivity' or in other words
meaning is a 'new and powerful form of coping with complexity under the
unavoidable condition of enforced selectivity.' Consequently, the problem
of complexity of complexity and the problem of meaning of meaning are
articulated as two aspects of a more fundamental problem of enforced
selectivity. This helps to bridge the methodological gap separating the
two cultures. It also implies a new approach to cognitivion, where the
cognitive system can be described as an autopoietic system capable of
self-observation.
Drawing inspiration from Luhmann's ideas, we can also suggest a new
classification of disciplines and their corresponding objects of study.
One can distinguish two main categories of such objects: autopoietic
and non-autopoietic systems. The autopoietic systems are systems that
are self-organizing and self-referential. There are three distinct classes
of autopoietic systems: living systems (cells, brains, organisms), psychic
(conscious) systems, and social systems (interactions, organizations, and
societies). Among non-autopoietic systems one can distinguish systems
that are self-organizing but not self-referential, such as some biological,
physical, or formal systems (e.g., animals, cellular automata, and adap-
tive connectionist systems). Finally, there is a class of systems that are
neither self-organizing nor self-referential, where most of the Artificial
Intelligence and physical systems belong. Whether artificial systems can
ever become autopoietic is an open question.
D. Dennett [4] is undoubtedly on the right track with his apt anal-
ogy he offers to justify why AI researchers should study philosophy.
They should do so, unless they are content to reinvent the wheel every
few days and when AI reinvents a wheel, it is typically square, or at
best hexagonal. His justification as to why philosophers should study
AI is unconvincing, however. Philosophers' wheels, he says, are perfect
circles that require in principle no lubrication and can go in at least
two directions at once. If by this he means that philosophers' wheels
can go in either direction, then he is perhaps right but there is nothing
wrong with it. Philosophers should certainly keep in mind the lesson
learned from the failure of behaviorism or logical positivism, and from
the insurmountable difficulties encountered by the symbolic AI and, I be-
lieve, also connectionismconnectionism. The lesson is precisely opposite
to what Dennett implies. Dennett, the Churchlands, and other cognitive
scientists are trying to come up with solutions to philosophical problems
by invoking scientific models, for example to solve the problem of in-
tentionality by adopting the connectionist model of cognition. A more
meaningful approach involves, I believe, a move in the opposite direction,
where shortcomings of scientific models are exposed by pointing out the
unsophisticated and simplistic nature of philosophical assumptions un-
derlying such models. This is particularly true of cognitive science where
philosophy can and should play a unique and special role.
AFFILIATION
Marek W. Bielecki
Philosophy Department
California State University
Hayward, CA 94542
USA
Department of Philosophy and Sociology

University of Warsaw
Warsaw
Poland
REFERENCES
[1] Dreyfus, H., Husserl Intentionality and Cognitive Science, MIT

Press, Cambridge MA, 1982.
[2] Newell, A. & Simon, H., "Computer science as empirical enquiry",
Communications of the Association for Computing Machinery, 19, 1976,
pp. 113-126.
[3] Garfield, J., "Introduction", in: Garfield. J. (ed.), Foundations of
Cognitive Science: the Essential Readings, Paragon House, N.Y., 1990.
[4] Dennett, D., "Artificial Intelligence as Philosophy and as Psychol-
ogy", in: Garfield, J. (ed.), Foundations of Cognitive Science: the Es-
sential Readings, Paragon House, N.Y., 1990.
[5] Dreyfus, H., What computers can't do: A critique of artificial reason,
(2nd ed.), Harper and Row, N.Y., 1979.
[6] Searle, J., The Rediscovery of the Mind, MIT Press, Cambridge MA,
1992.
[7] Winograd, T. & Flores, F., Understanding Computers and Cog-
nition, Addison- Wesley, Reading MA, 1987.
[8] Churchland, P.M., A Neurocomputational Perspective. The Struc-
ture of Mind and the Structure of Science, MIT Press, Cambridge MA,
1989.
[9] Searle, J ., "Is the Brain's Mind a Computer Program?" in: Scientific
American, 262, no. 1, 1990.
[10] Armstrong, D., A Materialist Theory of the Mind, Routledge &
Kegan Paul, London, 1968.
[11] Ryle, G., The Concept of Mind, Hutchinson and Co., Ltd., London,
1949.
[12] Fodor, J., The Modularity of Mind., MIT Press, Cambridge MA,
1983.
[13] Rumelhart, D. E. & McClelland J. L. (eds.), Parallel distributed
processing, Vol. 1 & 2. MIT Press, Cambridge MA, 1986.
[14] Forrest, S. (ed.), Emergent Computation: Self-Organizing, Collec-
tive, and Cooperative Phenomena in Natural and Artificial Networks.,
MIT Press, Cambridge MA, 1991.
[15] Luhmann, N., Essays on Self-Reference, Columbia UP, N.Y., 1990.
BURTON VOORHEES
GODEL'S THEOREM AND STRONG AI: IS REASON BLIND?
1. INTRODUCTION
In an episode of the television series Star Trek an ambitious star fleet

researcher wants to requisition the android Mr. Data as an experimen-
tal subject. Data objected to this on the basis that one highly probable
result of the experiments would be the cessation of his existence. The
researcher claimed that Data was only a machine, hence star fleet prop-
erty, and had no say in the matter. A military court was convened. Data
was defended by Jean-Luc Picard, captain of the starship Enterprise on
which Data served as science officer. Picard's defense was that although
it was not clear that Data was a self-conscious entity, it would set a dan-
gerous precedent with regard to the rights of future self-aware constructs
to declare that he was property.
The existence of self-aware robots, androids, and computers is a cliche
in science fiction. In science fiction the author is free of the need to
provide detail as to how such devices might actually come about, and
can avoid the metaphysical question of their possibility.
In contrast, this possibility is a point of lively debate in cognitive sci-
ence. All too often, however, the two sides of this debate seem to be
speaking past, rather than to, each other. The terms of the debate cen-
ter on what has come to be called "strong AI". This is the thesis that the
necessary and sufficient conditions for a self-conscious computer are the
construction of a big enough machine capable of running sufficiently so-
phisticated computational algorithms. Or, equivalently, that there exists
a formal system having mind as a model. Opponents of strong AI argue
that human beings are qualitatively more than programmed machines
and no machine, even in principle, could ever be aware in the way that
humans are aware.
The actual possibility of computer consciousness has become a matter
of both philosophical and practical concern. It is possible, for exam-
ple, that if the strong AI program is successful machines may end up
replacing humans, relegating humanity to the evolutionary dust bin, a
possibility viewed with both excitement [1], and concern 1 . In the Star
1
An article in The Spectator (Oct. 8, 1994), for example, suggests this will happen
within the next 30 to 50 years-unless something is done.
43
44 BURTON VOORHEES
Trek episode I referred to, this question is gently raised by Mr. Data.
He points out that Lieutenant LaForge, another member of the Enter-
prise crew, has artificial eyes which provide him with vision which is
far superior to ordinary human vision. Why, Data asks, don't the rest
of the humans have their eyes replaced by similar cybernetic implants?
Picard's response, that this is inconceivable, begs the question. Support-
ers of strong AI should sec no problem with such replacement. Indeed,
the far more radical idea of achieving immortality by downloading mind
onto mainframe has been seriously suggested [2].
As that suggestion indicates, the question of whether or not machines
can think is intimately connected to a number of other questions: What
is thinking? What is consciousness? What is self-consciousness? These
questions have been central to philosophy from its very beginnings, but
the invention of digital computers has brought them down from the
high mountains of philosophical speculation to the cultivated plains of
practical application. The number of angels that dance on the head of a
pin may not be relevant, but the number of transistors on a silicon chip
certainly is!
2. FORMAL SYSTEMS
One direction of attack on strong AI is based on Godel's incompleteness

theorems. These, it is claimed, demonstrate that there are capacities
of the human mind that can never be matched by a computer [3]. This
challenge fails, however, since it does not establish that the mind actually
possesses the claimed trans-computational capacities [e.g., 4,5]. For this
reason, it is sometimes asserted that the incompleteness theorems are
irrelevant to the validity of strong AI, and have no place in the debate [6].
The purpose of this paper is to show that although the Godel theorems
do not challenge the validity of strong AI, they have major implications
for its consequences.
In doing this, I will follow the line opened by Penrose [7,8,9] in his
attempted refutation of strong AI on the basis of a reductio ad absur-
dum in which the Godel theorems play the major role. Unlike Penrose,
however, I will not claim that conclusions reached on the basis of these
theorems and strong AI are absurd. Rather, the intent is to explore
possible consequences.
Computers, as designed and programmed today, are algorithmic en-
gines. The machine hardware is designed to support universal compu-
tation, and the software consists of instruction sets for implementation
of various computational algorithms. The strong AI thesis is that this
is sufficient to reproduce thinking and self-consciousness. Some prelimi-
GODEL'S THEOREM AND STRONG AI 45
nary definitions are necessary to make this idea precise, and to formulate
the Godelian argument against it:
1. A Formal System consists of a set A of abstract symbols together
with a set R of rules for operating on strings of these symbols to
form new strings. The set A is called the alphabet of the system.
The set R is divided into rules of gmmmarwhich specify constraints
which a string of symbols must satisfy to be admissible, and rules
of inference which specify the way in which admissible strings can
be operated upon in order to form new strings. The grammar and
rules of inference must be compatible in the sense that it cannot be
possible to transform a non-admissible sequence into an admissible
sequence by application of any rules of inference.
2. Any admissible string which is specified a priori as a starting point
for application of the rules of inference is called an axiom. The
system is finitely determined if the rules and axioms can be specified
by a finite procedure. A finite sequence of admissible strings such
that each string is obtained from its predecessor by application of
one of the rules of inference is called a proof sequence (or proof),
and the final string in a proof is called a theorem of the system.
Two simple examples of formal systems will illustrate this definition:
1. The game of chess: The symbols are the game pieces; the admissible
strings are the ways in which the pieces can be arranged on a chess-
board; the grammar specifies the set of legal chess positions; there
is one axiom, the initial position at the start of the game (or, in a
chess problem, the specified initial position). The rules of inference
are just the rules governing the legal movement of pieces. Thus the
"theorems" of chess are all positions which can be reached from the
starting position by a legal sequence of moves. (Note, however, that
there are only three ultimately significant theorems: white mate,
black mate, and draw. The game consists of a struggle between
players to produce a proof sequence culminating in a particular
conclusion.)
2. Define a formal system A with alphabet {0, 1} and the strings as
all finite sequences of O's and 1 's, taken in any order. Admissible
strings are those which begin with a block of 0 or more O's followed
by a block of 0 or more 1 's. That is, admissible strings have the
form 0 ... 01. .. 1. There are two axioms, the strings 0 and 1, and
a single rule of inference: if S and S' are admissible strings then
( S, S') ----+ S" where S" is the string obtained by dropping the leading
O's from S' and adjoining the 1's of S' to the right of S. For example,
(0011, 0111) ----7 0011111.
46 BURTON VOORHEES
Formal systems, of themselves, are purely syntactic. That is, they

are meaningless without an interpretation, which provides semantics in
terms of a model of the system. The second example illustrates the purely
formal nature of a formal system in a way that the chess example does
not because in that example the system was already interpreted in terms
of a model. That is, the symbols of the alphabet are interpreted as pieces
in the game of chess. In the second example, no interpretation is given.
The symbols were taken as 0 and 1, but then could have been called A
and B, or even Joe and Sally. In other words, they are purely abstract,
and the rules of inference give nothing but the syntax of the system
language.
One interpretation which could be given to the system of the second
example is to associate to every string S the number of 1's contained
in that string. For example, 011 = [2] and 00111 = 0111 = 111 = [3].
With this interpretation, the rule of inference becomes [n]+[m] = [n+m].
The system with this interpretation is modeled by addition of the non-
negative integers.
The distinction between model and system gives the distinction be-
tween truth and proof
Truth is a semantic property of statements in the model. Provability is
a syntactic property of strings in the formal system. Thus (00111, 011111)
----+ 0011111111 is a provable theorem of the system A, having the in-
terpretation 3+5=8, a true statement in the arithmetic model of the
system.
With this distinction, it becomes important to determine whether or
not every true statement in a model has a formal proof. In his 1929 doc-
toral dissertation, Kurt Godel established a completeness theorem which
demonstrates the equivalence of the semantic and syntactic versions of
entailment for first order logic. In general, however, this equivalence does
not obtain. Given a formal system F, and a model M of this system, let
T(F) = T1 , T2 , ... be a list of all of the theorems which are provable in
F, and let P be a well posed proposition in M. Then:
1. F is complete if F is finitely determined, and there is some Tn E
T(F) such that the interpretation of Tn in M is either P or "' P.
Informally, F is complete if every true statement in M has a proof
in F.
2. F is consistent if for every P it is true that there do not exist Tr
and Ts E T(F) such that the interpretation of Tr in M is P and
the interpretation of Ts in M is rv P. That is, contradictions are
not allowed. One cannot prove that a proposition is both true and
false.
Algorithmic computer programs are formal systems. The symbols are
1 and 0, or their electronic equivalents, and the grammar allows all 0,1
sequences (up to some machine dependent maximum size). The axioms
are the input strings, and the rules of inference are the allowed logical
operations on bit strings. A program is a set of algorithms, which speci-
fies a finite sequence of logical operations whose implementation results
in the generation of a proof sequence terminating with the result to be
computed. The program itself can be coded as a sequence of O's and
1 's so the program plus initial data can be taken as input. This means
that the same program can be implemented on many different comput-
ers. That is, the same formal system can be realized in many different
models. (This is what is behind the computational view of physics, in
which the laws of physics are taken as specifying the grammar and rules
of inference of a formal system which is being implemented in matter.
The observed state of physical reality is then taken as the current state
of an ongoing computational process.)
On the basis of this "universal" nature of algorithmic computation the
supporters of strong AI argue that it doesn't matter whether a program
is running on a digital machine or in a human brain-all thought pro-
ceeds via a sequence of "brain states" and the possible states, together
with their rules of transformation, constitute only a particular model
of a formal system. Reproduce this system in a computer model and
presto-cogito ergo sum. The Godelian argument against this is based
on the fact that for formal systems which include ordinary arithmetic,
there are true statements in models of these systems which are not the-
orems of the systems themselves.
3. THE ARGUMENT FROM GODEL
In 1931 the mathematical logician Kurt Godel proved the following the-
orems [10]:
Let F be a formal system such that: a) F is finitely describable,
and b) F is strong enough to contain a sub-system with ordinary
arithmetic as a model. Then:
1. F cannot be both consistent and complete.
2. F is consistent if and only ifF cannot prove its own consistency.
To prove these theorems, Godel developed a way to code logical propo-
sitions as statements in arithmetic. He then showed that there was an
arithmetic statement which corresponded to the proposition G(F) =
"System F Cannot Prove This Statement." This led to the following
possibilities:
48 BURTON VOORHEES
1. F proves G(F): Then G(F) is a provable theorem ofF, but G(F)

asserts that G (F) is not a provable theorem of F. Hence F proves
both G(F) and cv G(F). This contradiction shows that F is incon-
sistent.
2. F proves cv G(F): Then cv G(F) is a provable theorem ofF, but
the negation of G(F) is a theorem stating that F can prove G(F).
Again there is a contradiction and F is inconsistent.
3. F can prove neither G(F) nor cv G(F): Then F may be consistent,
but it is incomplete since there is a well posed proposition in the
arithmetic model ofF which has no proof within F. Further, G(F)
is obviously true in this model since it asserts that F cannot prove
G(F), and this is true. The conclusion is that if F is consistent
then it is necessarily incomplete. Godel's second result follows as a
corollary to this.
Although the possibility of using Godel's theorems (or equivalently, the

undecidability of the halting problem) to construct an argument against
machine self-consciousness was raised and dismissed by Turing [11], the
first development of the argument was published by Lucas [3]. Lucas
argued that Godel's theorem implied that there was always something
which a mind could do which a machine could not-see that the Godel
sentence of the formal system defining that machine was true. Therefore,
minds could not be machines. This was disputed in a number of responses
[4-6,12-14], the most telling counter-arguments being given by Lewis [4]
and Benacerraf [5]. Lewis reformulated his understanding of the Lucas
argument as follows:
1. Let A be a formalization of the language of arithmetic, with sen-

tences in A being taken as true if they are true in the standard
model of A.
2. For any given machine M, define a function con(M) from the ma-
chine table of M to the sentences of A such that: a) If S is some
subset of the sentences of A which is the potential output of a ma-
chine M then con (M) is true if and only if S is consistent; b) If
the set S contains only true sentences then con( M) is true; c) If S
includes the Peano axioms for arithmetic then con(M) is provable
from S if and only if S is inconsistent.
3. Call a sentence s of A a consistency sentence for S if and only if
there is a machine M with set of possible outputs S such that s =
con(M).
4. Define a rule of inference R: "If s is a consistency sentence for S,
infer s from S."
5. Define the "Lucas arithmetic" L by taking p as a sentence in L if
and only if p belongs to every superset of the Peano axioms which
is closed under the ordinary rules of logical inference and the rule
R.
With this construction, Lewis claims that so long as the Peano axioms
are accepted as consistent, the Lucas arithmetic L cannot be generated as
the output of any Turing machine. If it could, it would have a consistency
sentence s which by construction was contained in L. This would imply
that L, and hence the Peano axioms were inconsistent.
Lewis now argues that although Lucas may believe that all of the
theorems of L are true, he cannot show that all of the sentences of L
are contained in his own potential output, but that is what is required
if his anti-mechanist argument is to succeed. Benacerraf [5] uses his own
reconstruction of Lucas' argument to claim that all that is shown is that
if Lucas is an instantiation of a Turing machine, he cannot know this
to be the case: "It seems to be consistent . .. that I am indeed a Turing
machine, but one with such a complex machine table (program) that I
cannot ascer-tain what it is." (Benacerraf's emphasis)
Lucas [15] has responded to these criticisms by saying that both Lewis
and Benacerraf have misunderstood his real point. He is not required to
best all possible Turing machines. Rather, his argument is a dialectic. If
somebody claims that his mind is a particular machine he can produce a
true sentence which that machine cannot prove. Further, he can do this,
no matter what machine it is claimed that he is, hence he can be no
machine. This argument evades the issue, however, since it is not clear
that Lucas actually possesses the capacities which he claims.
The most extensive presentation of the argument from Godel has been
given by Penrose [7-9]. As with Lucas, the basic form of this argument
is that there are things which the mind can do, which computers, even
in principle, cannot. Therefore, computation is not sufficient to produce
minds. Penrose gives the basic reasoning for this as follows:
IfF is a formal system then there is a Godel statement G(F) which
can be stated in the system F, but which cannot be proved or disproved
within F. Yet we, on the basis of the meaning of the symbols involved in
the statement, can know that it must be true.
Given F, the simplest Godel proposition is G(F) = "F cannot prove
this sentence." Through a complicated numbering scheme G(F) is trans-
lated into an arithmetic proposition involving numbers, and other ele-
ments of the alphabet of F. Penrose writes this as
50 BURTON VOORHEES
If F is consistent it can prove neither this proposition, nor its negation.

But we know that the proposition is true because we know that F cannot
prove the stated proposition, and we are able to make an interpretation
of the symbols. That is, we assign them meaning. ""' 3, for example,
means "there does not exist." Without that, it would not be possible to
know if the proposition were true or false. Penrose draws the conclusion
that:
" ... the concept of mathematical truth cannot be encapsulated in
any formalistic scheme. Mathematical truth is something that goes
beyond mere formalism."
But we are able to have insights into the world of mathematical truth
which are unavailable to formal systems. Therefore, we cannot be op-
erating on the basis of formalism alone. Formal computation may be
necessary for self-consciousness, but it is not sufficient.
Penrose realizes, however, that the Lucas argument fails to topple
the strong AI thesis so he attempts to strengthen it with a reductio ad
absurdum, and this is his real contribution: he has shifted the debate from
the truth or falsity of strong AI to the consequences which must follow
if strong AI is true.
Suppose that mathematical judgment is nothing other than the result
of algorithmic computation. If this is so, Penrose argues, then the perfect
communicability of mathematical truth implies that all mathematicians
must be using the same algorithm. This, however, implies that the formal
system which this algorithm defines must be forever beyond the reach
of mathematicians, who will not even be able to know if it is consistent.
To Penrose this idea is an absurdity:
"The whole point of our mathematical heritage and training is that
we do not bow down to the authority of some obscure rules that we
can never hope to understand. We must see-or at least in princi-
ple see-that each step in an argument can be reduced to something
simple and obvious. Mathematical truth is not a horrendously com-
plicated dogma whose validity is beyond our comprehension. It is
something built up from simple and obvious ingredients-and when
we understand them, their truth is clear and agreed by all."
The question at this point is why Penrose's argument should be taken
as anything other than a statement of personal belief? The strongest
response to the Penrose argument is to challenge human cognition: How
do we know that human minds are not subject to the same limits as
machines? This is the same response that Lewis and Benacerraf gave to
Lucas. We are Thring machines, but don't notice this because minds are
so complex in their programming that we cannot reach their limits 2 . In

fact, suppose that we know something which a formal system F cannot
prove. Very well, take a bigger system, F' which can prove everything
that F does, and this additional thing as well. What results is an infinite
regress. The strong AI supporter will say that the human mind must
reach a point in this sequence at which it can go no further, and that is
the formal system which models the mind, and is sufficient to produce
self-consciousness.
Another significant criticism of the Penrose argument has been made
by Sloman [16]. His criticism is twofold: first, that Penrose has not pro-
vided a clear definition of consciousness, so it does not make sense to
discuss the possibility, or lack thereof, in computers; second, Penrose fails
to distinguish between the ideas of proof and truth-when he claims that
he can "see" the truth of the Godel sentence of a formal system he is
making a statement which depends on a particular model. Since the
system can neither prove nor disprove that statement, it will be true in
some models of the system and false in others, and Penrose cannot really
"see" that it is true-his "seeing" is model dependent.
Interestingly enough, Godel anticipated all of this in his 1951 Gibbs
lecture [17], where he pointed out that if mathematics is taken as a
body of propositions, true in an absolute sense, then the implication
of his incompleteness theorems is that the propositions of mathematics
cannot be exhausted by mechanical enumeration. His second theorem
points this out with particular clarity:
"For it makes it impossible that someone should set up a certain
well-defined system of axioms and rules and consistently make the
following assertion about it: All of these axioms and rules I perceive
(with mathematical certitude) to be correct, and moreover, I believe
that they contain all of mathematics."
This follows since the assertion that the axioms and rules are correct is
an assertion of consistency and the system, if consistent, cannot contain
the proof of its own consistency, hence cannot contain all of mathematics.
Godel then made a distinction between 'objective' and 'subjective'
mathematics, defined respectively as the body of all true mathemati-
cal propositions, and the body of all humanly provable mathematical
propositions. The incompleteness theorems demonstrate that no com-
plete formal system can exist for objective mathematics, but do not
exclude the possibility that there might be some such system yielding
all of the axioms and theorems of subjective mathematics.
2
For a survey of the Penrose argument and responses to it, readers are referred to
[8]
52 BURTON VOORHEES
No human could ever determine the consistency of such a system, but

if it existed, then from a mathematical point of view the human mind is
equivalent to a Turing machine and the strong AI thesis is correct. Fur-
ther, there would be simple propositions of ordinary arithmetic whose
truth or falsehood could not be established by any humanly intelligi-
ble proof. Godel termed such propositions absolutely undecidable, and
concluded that either subjective mathematics was inexhaustible, in the
sense that its evident axioms cannot be generated by any finite system, or
there exist absolutely undecidable propositions in ordinary arithmetic 3 .
Godel took no position on which of these possibilities was correct, al-
though he did suggest that if there was a formal system which generated
all of subjective mathematics then it could not be one constructed by
human mathematicians, but could only be produced by some sort of
evolutionary process, a point discussed most recently by Casti [18], who
suggests this as a possible outcome of work on artificial life.
Penrose, on the other hand, finds the possibility of absolutely unde-
cidable propositions in ordinary arithmetic to be contrary to the entire
idea of mathematics. How, he asks, can a proof which can be followed
simple step by simple step, be incomprehensible?
4. INTUITION
In the final analysis, Penrose's ultimate appeal is to mathematical intu-

ition. It is this that allows the human mathematicians to "see" the truth
of theorems in a way that computers cannot. Further, since intuition
is an aspect of consciousness, computers (or at least algorithmic based
computers) cannot be conscious.
To approach the deeper issue underlying the preceding discussion, con-
sider the relation between formal systems and their models. Ordinary
arithmetic, for example, is a model involving concepts such as number,
addition, and multiplication. Concepts are semantic, and semantics is not
found in the purely syntactic formal system. But mathematical intuition,
at least in experience, involves concepts. It is inherently semantic.
The direct connection between the semantic nature of mathematical
intuition and consciousness is found in the idea of a reflection principle,
which asserts that even in certain cases in which a formal proof is not
available, the truth of a mathematical statement may be seen by "reflec-
tion" on the meaning of the terms; i.e., by calling into consciousness a
basis of background knowledge and mathematical experience, and in this
3
It is not necessary for subjective mathematics to coincide with objective mathe-
matics in order that it be inexhaustible, it is only necessary that there does not exist
a finite formal system which generates all of its axioms.
context reflecting on the statements in question. This is the idea which
Penrose uses to claim that he can see the truth of Godelian statements.
"I believe ... that our consciousness is a crucial ingredient in our com-
prehension of mathematical truth. We must 'see' the truth of a math-
ematical argument to be convinced of its validity. This 'seeing' is the
very essence of consciousness. When we convince ourselves of the va-
lidity of Godel's theorem we not only 'see' it, but by so doing we
reveal the very non algorithmic nature of the 'seeing' process itself."
Unfortunately, the rather vague connections between intuition and con-
sciousness, and the still problematic nature of consciousness, leave this
line beset with uncertainties. Penrose offers his own view in saying that
" ... consciousness represents some kind of contact with the timeless Pla-
tonic world of mathematical concepts."
Many other scientists have also expressed a belief in the non-algo-
rithmic nature of intuition [19]: "It was his intuition which was pre-
eminently extraordinary. So happy in his conjectures that he seemed
to know more than he could have possibly any hope of proving. The
proofs were ... dressed up afterwards; they were not the instruments of
discovery." (John Maynard Keynes, on Newton); "There are no logical
paths to [laws of nature], only intuition resting on sympathetic under-
standing of experience ... " (Albert Einstein) "There is no such thing
as a logical method of having new ideas, or a logical reconstruction of
this process ... Every discovery contains an 'irrational element' or a cre-
ative intuition." (Sir Karl Popper); "[Mathematics] is a series of great
intuitions carefully sifted, refined, and organized by the logic men are
willing and able to apply at any time." (Morris Kline). Other examples
of mathematical intuition have been described by Hadamard [20] in a
classic study. The characteristic factor is always some sort of "seeing" in
which an entire problem or idea is revealed at once. But, whether this
"seeing" is based on a computational process or not remains an open
question.
Mathematical intuition has a long history in philosophy. In Plato's
Meno, Socrates demonstrates that a slave, ignorant of mathematics, is
yet able to see the truth of a mathematical proof. In the Republic, Plato
suggests that mathematics provides a bridge between the ordinary sen-
sory world and the ideal world of purely abstract forms such as Truth,
Justice, and Beauty. He bases this idea on the fact that mathematicians
use sensory objects (e.g., written notation, geometric diagrams) in order
to express ideas which are purely abstract.
In this regard, Plato also criticizes mathematicians for focusing too
much attention on the sensory shadows rather than on the reality which
54 BURTON VOORHEES
they represent. In making this criticism, he might well have had the
formalist mathematician David Hilbert in mind. In an often quoted pas-
sage [21], Hilbert asserts that the objects of mathematical intuition are
" ... the concrete signs themselves, whose shape, according to the con-
ception we have adopted, is immediately clear and recognizable.".
For Plato, intuition was the faculty giving access to the abstract world
of Forms. He held that this was the true reality, of which the ordinary
sensory world was but a shadow.
Aristotle dissented from this view, holding that the universals (i.e.,
the Forms) had their existence only as they were instantiated in matter.
Nevertheless, the universals were real, and were the true objects of intu-
itive knowledge once the mind had abstracted them from the experience
of a certain amount of concrete exemplars. Further, the universals are
the principles behind the manifestations of matter so that knowledge of
them is the fundamental knowledge. Once the mind has grasped the uni-
versal, or essential nature of a thing, all other knowledge of that thing
follows immediately by logical deduction.
"Only intuition yields a clearer knowledge of reality than the knowl-
edge resulting from science and proof. The intuitive mind is the prin-
ciple of science, for it knows the principles which form the starting-
point of science.
" ... intuition is true in a higher sense than science because it provides
science with principles." [22]
Writing 2100 years later, Kant distinguished sensory and internal in-
tuitions and found both to be closely related to consciousness. In Kant's
view, sensory intuition leads to the apprehensions which the mind struc-
tures into the objects of sensory perception, while mathematical objects
are derived from the mind's intuition of certain necessary aspects of its
own functioning. Kant also believed that mind could never gain knowl-
edge of reality as it is, apart from the necessary categories of its mental
construction, a view in which he differed strongly from Plato and Aris-
totle.
A possible clue to the nature of intuition might also be found in educa-
tional research which indicates a relation between the ability to grasp and
manipulate abstract concepts, and the capacity for "distancing," that is,
mentally detaching from an immediate context in order to "view" it from
a more objective position [23].
5. S C H 0 0 L S 0 F MATHEMATICAL T H 0 UGH T
If a machine could have real insight into mathematical questions then it

could presumably develop other kinds of semantic awareness as well. But
how could a machine have intuition? It cannot be the result of a sim-

ple enumeration of possibilities-there are too many, and no mechanism
is known for determination of significance. By reflection on a particu-
lar problem a human mathematician may realize that a theorem must
be proved, and that the proof will require a new definition or axiom.
Further, this reflection indicates the general nature of the required def-
inition or axiom. Strong AI supporters are challenged to show how this
process could be formalized. How could a machine ever distinguish the
meaningful from the trivial?
In considering how a machine might develop intuition, however, the
question of how it is possible to have an experience of an abstract object
which has no sensory qualities must be addressed. This question was
finessed by Godel [24], who claimed that the question of the reality of
the objects of mathematical intuition is the same as that of the reality of
objects of the perceptual world. Perceptual objects are constructed in the
mind from data supplied by sensory intuitions, and mathematical objects
are constructed in the same way from data supplied by mathematical
intuitions.
There are different views in mathematics, however, regarding the real-
ity of mathematical objects. The three major schools of thought within
mathematics are:
Formalism: Mathematical objects have no real meaning or existence.
Mathematics is the manipulation of abstract symbols according to
formal rules. It is content free and new mathematical results are
human inventions.
Platonism: Mathematical objects have real existence whether any hu-
man is aware of them or not. Mathematicians gain access to the
world of mathematical objects through mathematical intuition, and
new mathematical results are discoveries.
Intuitionism: Only those mathematical objects exist which have been
constructed and can be viewed as intuitively present. Mathematical
results are human constructions4 .
6. AN 0 NT 0 L 0 G I C A L ASS U M P T I 0 N
Talking about the existence, or lack thereof, of mathematical objects,

raises the question of ontology. What is the nature of that which really
exists, and how can it be characterized? It is at this level of discussion
that the radical nature of the strong AI thesis can be seen.
4
Some intuitionists, exhibiting a certain hubris, would say create rather than con-
struct.
56 BURTON VOORHEES
The pre-Socratic philosopher Parmenides made two very famous state-

ments, usually translated as "It is the same to think and to be," and
"What cannot be thought cannot be." A more accurate rendering of
these is: "It is the same to be conscious and to exist," and "What cannot
be an object of consciousness, cannot exist. 5 "
Parmenides' statements can be interpreted as an ontological and epis-
temological assumption: "It is the same to exist and to be a possible
object of consciousness." Since the time of the Greeks, this has been the
fundamental assumption of science: what exists can be known 6 . Restrict-
ing consideration to mathematics, this has different implications when
considered from the viewpoints of the different mathematical schools.
Formalists deny the real existence of abstract mathematical objects.
Thus, in accepting the existential criterion of possibility as an object of
consciousness, they would deny that abstract mathematical objects are
possible objects of consciousness. This may seem to be contradictory.
After all, I draw a circle and am certainly conscious of it, how then can
it be claimed that a circle is not a possible object of consciousness? The
reply is that that circle right there is certainly real, but the abstract idea
of a perfect circle is not. No one will ever see a perfect circle, or have any
real consciousness of such a thing. Thus, since it cannot be thought, it
has no existence except as a set of symbols which may be manipulated
in certain ways.
A Platonist, accepting the ontological assumption would take it as
meaning that all mathematical objects, no matter how complicated, are
possible objects of consciousness. This raises the question of infinity.
That is, there are mathematical objects such as certain numbers, or the
Mandelbrot set, which have infinite complexity. The question is, how
could these ever be objects of consciousness?
It seems to me that the Platonist has two options. The first is to claim
that it is sufficient that the axioms defining an infinite set are possible
objects of consciousness, and that any suitably delimited finite portion
of these sets may be an object of consciousness. This is Penrose's view.
The second response is more mystical: to claim that the experience of
human consciousness gives the concept of infinite, hence unrestricted
5
I am grateful to Gary Miller for this remark.
6
Although it is sometimes said that all of philosophy is just a footnote to the
Greeks, there is too often the tendency to fall into Vico's"error of scholars" and
assume that Greek thought was primitive in comparison to that of today. A more
charitable, and fruitful view is to assume that these ancient thinkers were highly
intelligent people, wrestling in a serious way with very difficult questions, hindered
by the lack of a technical vocabulary of sufficient sophistication to accurately convey
the deeper subtleties of their thought.
consciousness, which therefore exists and can have infinitely complex

objects. This second response is just the ontological argument for the
existence of God.
An intuitionist accepts as real all explicitly constructed finite repre-
sentations of mathematical objects, but denies reality to infinite and
non-constructed objects. When an intuitionist claims to have proved a
theorem, it is a claim that a certain mental construction has been carried
out. An example of intuitionist thinking is given in the following state-
ment made by Rolf Landauer in a discussion of the limits to knowledge
[25]:
"I do not just question whether the real numbers, or the integers,
characterize the physical world. I assert that the real numbers, or
unlimited sets of integers, have no available physical representation,
and therefore cannot be specified, manipulated, used, etc. They do
not 'exist' just because they have been named and can be invoked in
a closed formal axiom system.
"I maintain an open mind about the existence of a largest number. I
suspect that God was not that digital in orientation, and that there
is no sharp cutoff, but rather a statistical sort of blur that sets in as
numbers get larger."
When combined with the Parmenidean assumption, the strong AI the-
sis creates problems. If mind is based on a formal system, then there are
perfectly well defined mathematical objects which cannot be thought-
they remain forever incomprehensible. We are forced to the conclusion
that these finite, apparently well defined mathematical objects do not,
in fact, exist. This conclusion would be especially bitter for a Platon-
ist such as Penrose, but it creates difficulties for formalists as well. The
formalist is faced with the idea that certain well defined sequences of
symbols, written down following allowed inference rules, will be forever
incomprehensible. The intuitionist would seem to get off scot-free, but at
the cost of having to deny much of modern mathematics, and radically
change our view of the physical world.
One possible out is to deny the Parmenidean ontology. Doing this,
however, requires acceptance of the idea that there are things which
exist, but which cannot, even in principle, be known. If the unknowable
can causally influence us, then we are at the mercy of unknown and
unknowable forces, to which we can only respond with some form of
New Age irrationalism.
It was the Greeks who first introduced the idea of a rational macrocosm
which could be rationally known by the human microcosm. This idea of a
rationally comprehensible cosmos is the basis of science, and lies behind
58 BURTON VOORHEES
the statement, attributed to Leucippus, that the purpose of science is

" ... to free man from fear." Giving this up returns us to the position of
warming ourselves around the fire, surrounded by the infinite darkness
of the unknowable. Or, in Dante's imagery, human reason produces but
a small region of soft light in the darkness of hell.
Penrose is correct in saying that we would be surrendering a 2500 year
tradition of reason as a source of knowledge. And that, it seems to me,
is the real crux of his opposition to strong AI.
The situation may not seem so bad as all this since we can, on the
strong AI thesis, construct self-conscious computers which exceed our
own limited capacities. Even so, however, we-or to be more precise-our
computer successors, remain stuck in the position of Zeno's Achilles, who
will never catch the plodding tortoise. Any computer we construct must
necessarily remain finite, and the size of the universe sets a maximum
size on the possible size of computers (if we accept the computational
hypothesis of physics, then the universe itself is the largest computer).
Thus it would seem that on the strong AI hypothesis there is a funda-
mental limit to knowability. We must admit that even in principle, we
cannot know the nature of things.
7. EXTENSION BEYOND MATHEMATICS
As indicated at the beginning of this paper, the strong AI question goes

to the heart of some of the major perennial questions of philosophy, and
so has implications extending far beyond mathematical logic.
Lucas argues that his progression is a dialectic. Given any machine, he
can prove that he is not that machine. But in so proving, he shows how
to formalize a more powerful machine. In mathematics, what is initially
seen in intuition will later be captured in a net of symbols, equations,
and words. Are we then Achilles, forever pursuing the tortoise of the
unknown? Or are we perhaps the tortoise, with formal reason close at
our heels?
Suppose that we start with ourselves. Each of us, I assume, believes
that he or she exists on the basis of first person introspective experience.
We experience a self with a personal history and an apparent continuity
of identity. That is, the sufficient reason for our assumption of personal
existence is self-consciousness. Because we can "think" our self, we as-
sume that it must exist.
Speaking in extremes, a Platonist would assert that this self must have
existed prior to its becoming an object of consciousness, and indeed,
that all conceivable "selves" must exist, whether or not they are ever
realized as an actual person. From this perspective a Platonist might not
dispute the possibility of a self-conscious computer, but would question
the nature of the conditions which would produce this result. For the
Platonist mathematical insight, and by extension self-consciousness, is
beyond the implementation of a formal program.
The extreme formalist would say that there is no abstract "person"
there-all that is present is a formal program which is modeling it-
self within its environment. An extreme intuitionist, on the other hand,
would hold that this self only existed to the extent that it could be
known in intuition, making existence a concomitant of self-knowledge.
With this background, let's return to the argument from Godel's the-
orems, and the infinite regress which results. With respect to our in-
trospective awareness of self, this sequence can be seen as a dialectic
process of self-becoming. The formalist, and the strong AI advocate,
would describe this in terms of a program with the capacity for learn-
ing and self-modification. This program would be so constructed as to
generate a sub-program modeling its vehicle of manifestation within the
environment which this vehicle inhabits. Then it would model itself as
a program modeling its vehicle of manifestation, and etc.
In this view, "self-consciousness" is nothing but the capacity of the
most highly developed model in this sequence to manipulate symbols
having an interpretation as a "self" sub-program. The resulting pro-
cess of continuing self-invention is driven, in human terms, by internal
psychological factors (e.g., a desire to resolve apparent contradictions,
hunger) and halted by others (e.g., comfort, satiation). The intuitionists
would claim that the process of self-construction (creation) would always
terminate at some finite stage, and remain agnostic with regard to the
nature of the intuition by which this constructed self was known.
For the Platonist, on the other hand, the infinite limit of this se-
quence of partially realized selves exists a priori and the movement of
self-discovery, or self-remembering is a result of a natural attraction to-
wards this ideal limit. True "self-consciousness" is now beyond the formal
capacity for self-modeling-it also involves the capacity to meaningfully
interpret these models from the intuited viewpoint of the final ideal self.
In the Platonic dialogues the approach to this ideal is through the So-
cratic dialectic, which strips away false identifications 7 . From the Aris-
totelian position, on the other hand, the ideal self can exist only as it is
embodied in matter, and is to be abstracted from empirical observations
of the existing self. In this view, the ideal Platonic self is "remembered"
7
This is very similar to the "nyte, nyte" exercise in esoteric Buddhism, in which
a person is instructed to say "nyte, nyte" to all thoughts and perceptions: "I am not
that, I am not that."
60 BURTON VOORHEES
through a process of self-observation. A computer could only become

self-conscious if it, too, had the capacity to carry out this process.
In the formalists bottom-up view a self is constructed from below, each
level generating the next higher one out of itself and external input. This
process either stops due to internal and/or external factors, or is endless.
In the Platonist top-down view a self is gradually "remembered" as one
moves toward the ideal self8 .
So, will a computer ever think? Since I started with some science
fiction, I will end with some as well, taken from a discussion of the
conditions necessary for a thinking computer in the book Time Enough
for Love by Robert A. Heinlein [26]. Minerva, a former computer, now
embodied as a woman:
" ... speaking from experience, all theories are empty. How a com-
puter becomes self-aware remains as much a mystery, even to com-
puters, as the age-old mystery of flesh-and-blood self-awareness. It
just is. But so far as I've heard ... self-awareness never arises in a
computer designed only for deductive logic and mathematical calcu-
lations, no matter how big it is. But if it is designed for inductive
logic, able to assess data, draw hypotheses therefrom, test them, re-
construct them to fit new data, make random comparisons of the re-
sults, and change those reconstructions-exercise judgment the way
a flesh-and-blood does, then self-awareness may occur. But I don't
know why and no computer knows. It just does." (p. 371)
Lazarus Long responds:
"... all machinery is animistic-'humanistic,' I want to say, but that
term has been preempted. Any machine is a concept of a human
designer; it reflects the human brain, be it wheelbarrow or giant
computer. So there is nothing mysterious in a machine designed by a
human showing human self-awareness; the mystery lies in awareness
itself, wherever it's found." (p. 372)
"... those big computers with awareness potential but without self-
awareness are that way because nobody bothered to love the poor
things. That's all. Babies or big computers- they become aware
through being given lots of personal attention. 'Love' as it's usually
called .... With babies you breast-feed them and nibble their toes and
talk to them and blow in their bellybuttons and make them laugh.
Computers don't have bellybuttons, but attention works just as well
on them." (pp. 373~374)
8
In a lighter vein, one might say that the formalist attempts to bootstrap con-
sciousness while the Platonist looks for a sky hook.
ACKNOWLEDGMENTS
This paper derives in part from a survey paper presented at the Work-
shop on Logic held in July, 1992 at the University of Victoria. The
present paper is based on work carried out in January, 1995 during a
visit to the Santa Fe Institute, whose hospitality is gratefully acknowl-
edged. Early versions of the paper were presented as seminars at the
Santa Fe Institute, January 24, 1995 and at Athabasca University, April
27, 1995. A shorter version has been published in Complexity under the
title "Gi:idel's Theorem and the Possibility of Thinking Machines".
Supported by operating grant OGP 0024871 from the National Science
and Engineering Research Council of Canada.
AFFILIATION
Faculty of Science
Athabasca University
Athabasca, Canada TOG 2RO
burt@cs. athabascau. ca
REFERENCES
[1] Minsky, M., "Will robots inherit the Earth?", Scientific American,
271, No. 4, 1994, pp. 108-113.
[2] Moravec, H., Mind Child, Cambridge, MA, Harvard Univ., 1988.
[3] Lucas, J.B., "Minds, machines, and Godel", Philosophy, 36, 1961,
pp. 112-127.
[4] Lewis, D., "Lucas against mechanism", Philosophy, 44, 1969, p. 231.
[5] Benacerraf, P., "God, the devil, and Gi:idel", The Monist, 51, 1967,
pp. 9-32.
[6] Good, I.J., "Gi:idel's theorem is a red herring", British Journal for
the Philosophy of Science, 19, 1969, pp. 359-373.
[7] Penrose, R., The Emperor's New Mind, NY, Oxford Univ., 1989.
[8] Penrose, R., "Precis on the emperor's new mind: concerning com-
puters, minds, and the laws of physics", Behavioral and Brain Sciences,
13, 1990, pp. 643-705.
[9] Penrose, R., Shadows of the Mind, NY, Oxford U. Press, 1994.
[10] Godel, K., "On formally undecidable propositions of Principia
mathematica and related systems: I", In: Solomon Feferman (ed.),
Collected Works, Vol. I, NY, Oxford University Press, 1986.
62 BURTON VOORHEES
[11] Turing, A., "Computing machinery and intelligence", MIND, 59,

1950, pp. 433-460.
[12] Coder, D., "Godel's theorem and mechanism", Philosophy, 44,
1969, pp. 234-237.
[13] Bowie, G.L., "Lucas' number is finally up", Journal of Philosoph-
ical Logic, 11, 1982, pp. 279-285.
[14] Lewis, D., "Lucas against mechanism: II", Canadian Journal of
Philosophy, IX, No. 3, 1979, pp. 373-376.
[15] Lucas, J.R., "Mechanism: a rejoinder", Philosophy, 45, 1970, p.149.
[16] Sloman, A., The emperors real mind, preprint.
[17] Godel, K., "Some basic theorems on the foundations of mathemat-
ics and their implications", Unpublished Gibbs Lecture. To appear in:
Collected Works, Vol. III.
[18] Casti, J.L., Complexification, NY, Harper Collins, 1994, pp. 166ff.
[19] Goldberg, P., The Intuitive Edge, Los Angeles, J.P. Tarcher, 1983.
[20] Hadamard, J., An Essay on the Psychology of Invention in the
Mathematical Field, Princeton, Princeton University Press, 1945.
[21] Hilbert, D., "On the infinite", (Stefan Bauer-Mengelberg, trans.)
in: Van Heijenoort, J. (ed.), From Frege to Godel: A source Book in
Mathematical Logic, 1819-1931, Cambridge, Harvard University Press,
1967, p. 376.
[22] Kal, V., On Intuition and Discursive Reasoning in Aristotle, Lei-
den, E.J. Brill, 1988, p. 47.
[23] Ellsworth, T. & Sindt, V., "Helping "Aha" to happen: the contri-
bution of Irving Sigil" ,in Educational Leadership, 51, 5, 1994, pp. 40-49.
[24] Godel, K., "What is Cantor's continuum problem?", The American
Mathematical Monthly, 54, 1947, pp. 515-525.
[25] Casti, J.L., & Traub, J.F. (eds.), On Limits, Santa Fe Institute
preprint 94-10-056.
[26] Heinlein, R.A., Time Enough for Love, NY, Putnam, 1973, p. 371.
SUZAN LANGENBERG
WHAT ABOUT THE MODERNISTIC CONCEPT OF

CONSCIOUSNESS?
The Meeting of Einstein and Magritte: a Metaphor
In this essay I will use the development of a schizophrenic illness as a way

of showing that schizoid people, apart from suffering from a possible neu-
robiological defect or genetic abnormality, make the paradoxical charac-
ter of the modernistic concept of consciousness visible. vVe can compare
the modernistic rupture between mind and body with the rupture be-
tween the cognitivist and subjectivistic experience of the schizophrenic.
This experience, expressed in language, behaviour, art and/ or illness,
not only seen as caused by a neurological defect or environmental fac-
tors, can be understood as an extreme manifestation of the modernistic
concept of consciousness ...
In the schizophrenic, hyperrefiexivity meets with deep subjectivistic
feelings in an everlasting conflicting game without any prospect on the
integration of these human opposites. In this light we can ask 'Does
Einstein really meet Magritte, or is the skill of the first discriminating
the art of the other?' Can they complement each other or does this
meeting result in an overflowing of areas in which the specificity of each
sphere fades away into obscurity? Can we call it the great fear of our
modernity, that sometime all the precise, careful and distinct areas of
our sciences will be intermixed, setting us back into the Middle Ages? Is
the modernist human being filled with fear, in the context of a coming
new, vague, uncertain century, in the context of a world in transition,
experiencing the disappearance of les grands recits (Lyotard) because of
the possible discovery that our concept of truth is based on a mistake?
Is this the great difficulty which undermines the philosophical discourse
nowadays, that when we really want to understand the interdisciplinary
construction of our society, when we effectively want to have a grip on
the complexity of our social behaviour, we have to meet the different
other every day as a participant and not as a distant observer, without
fear, without projection, without the disturbing and troubling thought
that such confrontations will be followed by a transgression of our moral,
cognitive, aesthetic and expressive borders, when we are in contact with
that otherness? The fear of losing our singularity, our individuality, what-
ever it may be, has fragmented our life-world up into categories based
63
64 SUZAN LANGENBERG
upon dichotomical judgements. Einstein fights, loves, hates, likes, meets

Magritte in the schizophrenic.
What to do when someone is not capable of sublimating his confusion

concerning his own Being in the World into an ingenious piece of art,
like for instance the surrealist painter Giorgio de Chirico? This very
schizoid person-it has been proved in his diaries-painted in his early
work (at the very beginning of the twentieth century) aspects of the
Stimmung of a schizophrenic life-world: the sense of meaninglessness and
meaningfulness, the significance and insignificance, described as anti-
epiphany, an experience in which the familiar has turned strange and the
unfamiliar familiar. He wrote: There are more enigmas in the shadow of
a man who walks in the sun than in all the religions of the past, present
and future. [5, p. 45]
But, in the end, what is so important about the concept of conscious-
ness nowadays? We know that human beings have something, an instru-
ment, by which they think, they make words, they express themselves,
they conceptualize their perceptions, by which they can remember, by
which they have a memory, and we call that instrument our conscious-
ness. Consciousness is the anchor of our institutionalised world. The fact
that we have filtered that concept out of the chaos of nature has opened
the way to cognitivity and truth. But there are still some questions:
How do we think? How do we know? When do we know? What do we
know when we say we know something about something?
A young schizophrenic writes in his diary:
What is the intellect? Probably the force which asks the question
'What is the intellect?' If the intellect asked itself what itself is, can
it answer itself? [5, p. 333]
Many modernist philosophers like Nietzsche, Heidegger, Foucault have
questioned the apparent unquestionability of that concept of conscious-
ness.
CONTRADICTION
The modernistic concept of consciousness forms the foundation of our

institutions. The way we treat people in the medical health sector, the
methods we use for research in human sciences as well as in natural sci-
ences, the way in which we maintain law and order, the way in which we
judge, the way in which we judge our feelings, our behaviour and the be-
haviour of the other. .. all these ways can be reduced to and explained by
a so-called concept of consciousness. The simple principle of Descartes: I
think therefore I am (cogito ergo sum) was the very beginning of modern
THE M 0 D ERN IS TIC C 0 N C E P T 0 F C 0 N SCI 0 U S NESS 65
thought, of modern science. With the conceptualisation of thinking as a

constitutional element of being, we can make a distinction between mind
and body, as Descartes had done. But with that distinction we have a
kind of clarity about an area that is untouchable, that is invisible, that
probably exists on a border, that probably exists in a confrontation, that
probably exists in a dialogue.
We have made classification systems, we have made categories, we
have made rules to be able to live together, we have made political and
economic systems in which we had the possibility to live. For a long
time we have had the illusion that we could make, govern and control
the world. For a long time this was our reality, it was our conviction, it
was our truth, it was the safety of the modern human being. The concept
of consciousness made it possible.
However, most of us have had, or still regularly have the experience
or the impression that, when we really say what we think, others will
think we are mad. So, many of the things we say are censured by our
reflections and know-how of the culture we live in, are dependent on
tradition, the context, the situation we are in. The Cartesian modernist
concept of the mind, consciousness seen as radically different from the
body in which it exists in a mysterious way, a sort of mind in a machine,
was the first successful attempt of science to exist besides the domination
of the church for centuries. Consciousness offers an immediate approach
to the inner ideas which represent the world in a certain manner. (I
cannot have knowledge about things outside myself unless by means of
reflections on the ideas inside me. I will refer here to the concept of
indirect presentation of Kant). These two aspects of Cartesianism are
integrated in our usual way of understanding that has become evident
and inevitable. Implications of this Cartesianism for the concept of the
self can be reduced to two aspects: liberation and reflexivity.
Complete realisation of the essential being (as consciousness) asks for
a release from the body with its 'rooted' passions. That will be the only
way to realise rational self-control. Cogito ergo sum refers to the certainty
of the knowing power of the inner mind; the inevitable participation of
the mind in every conscious social action.
Kant rather emphasizes the importance of liberation and reflexivity
as a result of a working consciousness. It is not only a criterion of cer-
tainty, but it is also a structure too that constitutes and creates the
world of our experience with its categories time, space, causality and
substantiality. Kant makes an absolute distinction between the world of
possible human experience (phenomenon) and the present existence of
Being (noumenon). Kant seems to have shown, that it was the world
that had to arrange itself in accordance with the conditions of human
66 SUZAN LANGENBERG
consciousness rather than the reverse. Only what the human mind was
capable of conceiving, could exist. Being was shown to depend on Know-
ing. The primacy of consciousness as the constituting center was bound
up with three notions: the freedom of consciousness, its self-creating na-
ture, and its status as a complete and lucid source of truth. Beside the
almighty power of the church it was necessary to create an even more
powerful concept of knowing based upon cognitivist truth, a product of
the mind as a human source, instead of knowing based upon the truth
of an almighty god as a product of faith.
In the modern era however, consciousness as a seemingly self-aware
foundation and transparent medium of representation, is also found to
be surrounded by and imbued with a kind of obscurity. Consciousness
discovers at its borders an element of darkness, an apparently inert den-
sity in which it is embedded, an unthought which it contains entirely,
yet in which it is also caught. This contradicts the other notion of con-
sciousness that all of reality exposes itself before the mind and at the
same time is constituted by it. This gives a sort of tunnel-vision solip-
sism: an experience in which consciousness is felt to be both an ultimate
constituting source and a limiting and mysterious channel-omniscient
and omnipotent, but only within a sphere, the boundaries of which are
acutely felt from within.
REFLECTION
Reflection, the distinguishing of the objects we see, of our feelings, our

desires, the fact that it is possible to control ourselves by using the ratio,
the fact we can give names to such powers, is an implicit result of the
modernistic concept of consciousness. It creates the possibility for us to
live together, in a more or less harmonious way. Reflection according to
Jacques Derrida means: the act by which the ego, after having stripped
away its natural immediacy and returned into itself, becomes conscious of
its subjectivity in relation to counterposited objectivity, and distinguishes
itself from it. [5, pp. 345-347] This meaning of reflection conflicts with
the modernist concept of consciousness because of the different primacy
that is given to human action.
When we can define rationality as reflexivity, as the capacity toques-
tion what had hitherto been treated as given; to bring to reflection what
had only been used; to think about our thinking, we can interpret ex-
cesses like schizophrenic thinking and behaviour in all their different
expressions, their incongruity, as products of a kind of hyperrationality:
forms of irrationality generated by rationality itself.
THE MODERNISTIC CONCEPT OF CONSCIOUSNESS 67
INTENTIONALITY
Many schizophrenics tend to lose their sense of active and integrated

intentionality. Instead of serving as a kind of anchoring center, the self
may be dispersed outward where it fragments into parts that float among
the things of the world; even the most intimate thoughts and inclinations
may appear to emanate from some external source or mysterious foreign
soul, as if they were the workings of another psyche.
This existence accounted for me great fear. In the unreal scene, in
the murky quiet of my perception, suddenly 'the thing' sprang up.
The stone jar, decorated with blue flowers, was there facing me, de-
fying me with its presence, with its existence. To conquer my fear I
looked away. My eyes met a chair, then a table; they were alive, too,
asserting their presence. I attempted to escape their hold by calling
out their names. I said, 'chair, jug, table, it is a chair. ' But the word
echoed hollowly, deprived of all meaning; it had left the object, was
divorced from it, so much so that on one hand it was a living, mock-
ing thing, on the other, a name, robbed of sense, an envelope emptied
of content. Nor was I able to bring the two together, but stood rooted
there before them, filled with fear and impotence. (. .. ) And they, the
doctors, too, thought I saw these things as humans whom I heard
speak. But it was not that. Their life consisted uniquely in the fact
that they were there, in their existence itself. [5, p. 49]
But we also know that the self can come to seem pre-eminent and all-
powerful. When someone gets fixed on the self, he or she transgresses the
borders of common accepted conceptualisations. For instance a banana
is accepted under the category 'fruit'. When someone describes a banana
as belonging to the category of yellow and bent objects, he or she isn't
wrong, but is behaving very deviantly concerning conceptualisation of
things.
Schizophrenics complain about the inadequacy of language, about the
insufficiency of words: But what is a train? It's a word. The word has
nothing to do with a solid thing like a train. [5, p. 51]
Children have these sorts of thoughts. But do adults?
THE PUBLIC AND THE PRIVATE
One of the most fundamental aspects of modernity is the intense focus on

the self (both as an object as well as a subject of experience) and on the
value and power of the individual. We cannot find these features in more
primitive and integrated societies. What is the relationship between the
self and the world, between the inner world of experience, of feeling, of
68 SUZAN LANGENBERG
inner perception, of inner speech and the outer world, with the different
other, with the unusual, with the unfamiliar?
Descartes was the first philosopher who explicitly formulated a concept

of the instrument, our consciousness, that realises this relationship in
terms of self perception. But this concept of self perception can lead to an
extreme interpretation of the self; the self fixated on its own inwardness.
In this case the meditation will not stop.
Descartes meditated within a certain context: religious power ended,
the development of different sciences began. Within that specific frame-
work his meditation made sense because of the principle of true thinking,
of objectivity, made it possible to escape the continuing domination of
the church onto the moral and scientific development of human nature.
In the social interaction of human beings this context is no longer
current. It is our history that again and again must be reinterpreted by
each individual in each event. Which context will stop our thinking, will
stop our intense focus on the self and make the passage from inner to
outer world possible? The concept of self perception is a possible rupture
of the inner from each possible outer.
KAFKA
In Kafka's Description of a struggle we can find the entire progression

of a schizophrenic illness; from self-consciousness and a very critical way
of perception through a thorough self-alienation and solipsism, and on
to the dissolution of both self and the world. The story exemplifies the
impossibility to think for the other, to know what the other is feeling.
The narrator meets situations completely alienated from his own feel-
ings, from his own primary intentions, in an invented story of the other.
The logic of his own thinking projected onto the other. His own self-
experience gets involved with the imagined and projected experience of
the other, which results in an entire dissolution and derealisation. What
starts as intelligible-logically based interpretations of the situation he
is in, escalates into a chaotic, derealised, solipsistic and depersonalised
obscure self-perception.
Experiences of alienation from the world and from oneself, followed by
moments of brief-lived solipsistic triumph and a more pervasive ontolog-
ical collapse.
And I hope to learn from you how things really are, why it is that
around me things sink away like fallen snow, whereas for other peo-
ple even a little liqueur glass stands on the table steady as a statue ...
Why, after all, should I feel ashamed of not walking upright and tak-
THE M 0 D ERN IS T l C C 0 N C E P T 0 F C 0 N S C I 0 U S NESS 69
ing noTmal steps, of not tapping the pavement with my stick ... ? Am
I not mtheT entitled to complain bitter-ly at having to skip along the
houses like a shadow without a cleaT outline, sometimes disappearing
in the panes of the shop windows? [3, p. 825]
Kafka, in this story, is an excellent exponent of the modernist, and also

postmodernist spirit. Extreme forms of dehumanisation and derealisa-
tion as a result of hyperreflexivity, as a result of an extreme form of
dealing with social interaction proceeding from an extreme focus on the
disfunctioning self, leads the reader to a confrontation with a madman,
the narrator. But ... is he mad? From the start this is an ordinary story of
daily triviality. But the platitudes, the banalities develop into an extreme
reality ... reasonings that resemble ours, projections that are common
for modernist human beings, feelings as an extreme result of an absence
of dialogue. The felt incapability to check or to talk about this imagined
reality with the other is an extreme exponent of what schizophrenics feel
when they 'make' a psychosis or hallucination as an ultimate way out of
the human prison: locked up in the unfitting mask in which rationality
and subjectivity are intertwined in an eternal fight, a modernist fight
against a modernist invention.
Maybe it was that small, veTy still pause between day and night, when
ouT head, without ouT expecting it, falls foTwaTd and in which eveTy-
thing, without ouT noticing it, stands still because we don't look at
it, and then disappeaTs, while we stay behind alone with a bent body.
Then we look aTound 'US but see nothing anymoTe; we don't feel any
resistance of the air eitheT, but inward we gmsp ouTselves onto the
memoTy that at a certain distance there must be houses with roofs
and, fortunately, with angular chimneys, by which the darkness in
the hmtses will flow thmugh the attics to all the different rooms. [3,
p. 845]
In my view, this passage exemplifies an extreme longing for a 'normal'

experience of the world instead of the inner chaos or the outer threat.
But, the normal. .. does it exist?
This story exemplifies both sides of the empirico-transcendental dou-
blet of modern thought: most of the time we find ourselves occupying
a transcendental position, in fact an almost solipsistic one identifying
ourselves with a narrator who experiences himself as the center of the
universe. Yet we also experience the self as alienated and passive, the
narrator is continually discovering his thoughts and feelings outside him-
self.
70 SUZAN LANGENBERG
CONCLUSION
How can we know that what we experience of the world is not merely
a hallucination? We have met a number of artists who have their own
interpretation/hallucination of the world outside them. They have found
a universal accepted modus for sublimating their vision upon life- and
system-world. There are a greater number of human beings who cannot
cope in such a manner with their existential questions. They cannot
make people listen to their critical attitude towards family, society or
'dehumanised' daily living.
Leo Beyers gives a possible opening to other ways of conceptualisa-
tion when human beings want to cope with the complexity of human
behaviour. The will to speak ('zeggingswil'), present in all of us, gives
us, as a human condition, as a way of being, a concept to make cooper-
ation thinkable between cognitive, moral and emotional claims in social
events. The will to speak happens and, only if noticed, brings humans
into the inter-esse, the being between. Inter-esse is a tenuous state that
throws humans into conflicting qualities of life, while being in the event,
without classifying the event. Conflict is understood as a driving force
in bringing together simultaneously attention, intention, cognition, and
feeling again and again and again and again ...
AFFILIATION
Suzan Langenberg
Academie Leo Beyers for Art and Life Sciences
Kortenaken
Belgium
REFERENCES
[1] Beyers, L., Conflict en inter-esse, VUB-PRESS 1.s.m. Stichting

CINCOOP, 1994.
[2] Foucault, M., Folie et deraison; histoire de la folie a l'age classique,
Gallimard, Paris, 1961.
[3] Kafka, F., Verzameld werk, Querida, Amsterdam, 1983.
[4] Lyotard, J-F., Le Differend, Les Editions de Minuit, 1983.
[5] Sass, L. A., Madness and Modernism - Insanity in the Light of
Modern Art, Literature and Thought, Basic Books, 1992.
JAMES WILK
MIND, NATURE AND THE EMERGING SCIENCE OF

CHANGE: AN INTRODUCTION TO METAMORPHOLOGY
The universe was put into equations the way the history of France
was put into madrigals.
~Maeterlinck
INTRODUCTION
Over these few short pages, I should like to offer you a brief introduc-
tion, or rather, a lengthy invitation, to an emerging field of scientific
study. Rather than attempt, impossibly, to convey the richness of the
tapestry, and its significance, I shall confine myself instead to drawing
your attention to just a few of the threads running through it, and say
a little about why I think these threads worth pursuing.
A NEW SCIENCE OF CHANGE?
The word "metamorphology," not a pretty word, comes from the Greek
"metamorphosis," the equivalent of the Latin "transformation." The Ox-
ford English Dictionary reminds us that a century ago it was applied to
the study of "change of form" solely in the context of post-embryonic
ontogenesis. Today it refers more broadly to the study of transformation
in general~the emerging science of change.
This admission should at once arouse our deepest suspicions! After
all, which branch of science does not study change? Does physics, and
indeed each of the special sciences, not deal with change of one kind
or another? When one has said everything the various sciences have to
say about specific classes of changes, what more is there to say on the
subject of change? Surely, one might argue, if there are any fundamental
principles underlying change in general, these would turn out to be none
other than the laws discovered by physics!
Now on the basis of prevailing views on change we should be fully
warranted in drawing such a conclusion. And yet, I have come to believe,
to do so would be a grave error. If nothing else, I hope in this brief
discussion to indicate just a little of why I think so.
71
72 JAMES WILK
INVARIANCE, CONTEXT, AND SCIENTIFIC INQUIRY
Now you might say that, as scientists, whatever our particular discipline,
we study certain observed invariances of interest to us. In fact, each
branch of science concerns itself with a certain class of invariances. We
study the contexts within which certain invariances occur, and we seek,
empirically, to define the limits of those contexts~that is, to identify
where the invariances do not occur. We aim finally, when all goes well,
to account for those context-bound invariances in terms of a smaller
number of others which are considered more fundamental, and which in
turn account for a wide range of still other invariances too, some perhaps
as yet undiscovered.
So we may study mirages, for example, and come eventually to delin-
eate precisely the physical contexts in which mirages can uniquely be
observed, and do the same, say, for rainbows, and account for each of
these context-bound invariances in terms of, at most, a few of the laws
(the more fundamental invariances) of optics and atmospheric physics.
Or again, we may study certain invariances in the degree of darkening
of the skin on exposure to sunlight, and come to account for these, in
the first instance, say, with reference to the quantity of melanin, and so
on. But whilst the appearance of mirages on a hot country road, or of a
rainbow after a shower, or of a tan after a holiday may all be considered
to be changes, what we are normally interested in, as scientists, is to
account for observed invariances as they happen to be exhibited in these
specific changes.
The changes themselves may provide us with observational fodder,
they may offer us the very clues we need to penetrate more deeply into
the nature of things, but they provide only one starting point of our
inquiry and perhaps some of the signposts along the way. Change is not,
in such cases, the subject of our investigation itself. How could it be, or
rather, why should it ever be?
CHANGE IS INESCAPABLE
Well, certain changes can indeed be, and often are, the very subject of
a scientific investigation, but normally only tacitly. So let me state the
obvious.
Once we turn our attention from physics to the biological sciences,
let alone the human sciences, we note that change, as a phenomenon,
has long been as indispensable in scientific explanation as it has been
ubiquitous. For any biological assembly, the perception of and differential
response to changes in its environment has long been taken tacitly to
A N I NT R 0 D U C T I 0 N T 0 MET A M 0 R P H 0 L 0 G Y 73
be fundamental. We need look no further than such common or garden

examples as birds flying south for the winter.
CHANGE IS INADMISSIBLE
But now let me point something out that may be less obvious.
To rely on the notion of change in nature, in our explanations, or to
refer to it at all, is 'Officially' an illicit move. That is, within nature as
Officially conceived over the past three centuries, there is no room for
the concept of change. For change is a difference over time, and, logically
speaking, any difference in a given particular can only be a descriptive
difference. For some particular p to be regarded as different at time T2
to the way it was at some earlier time T1 , it must be identifiably the
same p, which is now different in some (descriptive) respect.
What is more, we can readily prove the following from first principles:
To state the matter somewhat informally, for an assembly a to perceive or
respond to a change in some particular p, a must at the very least detect
a descriptive difference in some pattern or invariance of p--a descriptive
difference in pattern, moreover, which is of significance from a's point of
view.
DESCRIPTION, POINT OF VIEW, AND SIGNIFICANCE

IN SCIENTIFIC EXPLANATION
But whoa! Wait a minute. How did description and point of view and
significance (meaning, no less!) suddenly appear amongst the would-be
fundamentals of the universe? How did these three interlopers manage
to gatecrash the explanatory party? If it is indeed the case that, in
logic, any attempt to invoke the concept of change is, however tacitly, to
presuppose the concepts of description and point of view and significance,
can we permit scientific explanations to rely so heavily, as they therefore
currently do rely implicitly, on such concepts as these? Can we permit a
key role for description and point of view and meaning (Heaven forfend!)
in the fundamental mechanics of the physical universe?
The short answer goes something like this: We might as well, and for
two reasons. First of all, it is unlikely that we could get by in scientific
explanation without making use of the notion of change, at least outside
of basic physics (and even there, for reasons we cannot go into here, it is
unlikely that we could get by without it for long). And second of all, the
concepts of description and point of view and significance turn out to
be analysable as more austere and theoretically parsimonious concepts
than we might at first be inclined to fear. In fact, within the theo-
retical edifice of metamorphology, these concepts are logically derived
74 JAMES WILK
[25], fairly straightforwardly, from a small stock of fundamental notions

from amongst the basic concepts of predication and the basic concepts of
measurement. Arguably, the only novel concept we need to import is the
concept of imparity, or unequal weighting [26]. These "exotic" sounding
concepts-description, point of view, significance-are therefore derived
from basic notions which are already presupposed by virtually all scien-
tific inquiry of the past 300 years, and thus would appear, on analysis,
to give no special cause for concern.
If we cannot do without a concept (sc. change) which presupposes
the concepts of description and point of view, and if the latter can be
analysed into concepts already presupposed by the very enterprise of em-
pirical science, why not give the gatecrashers a dance card? Or, to change
the metaphor, why not, as we in fact go on to do in metamorphology,
let these poachers turn gamekeepers?
There are plenty of reasons, frankly, against, at least reasons aplenty
for the faint-hearted. For even to admit description and point of view ex-
plicitly into the mechanics of the physical universe, however ineluctably,
has some far-reaching consequences for our conception of the physical
universe and of scientific explanation.
But before I turn to consider some of the consequences, let me pick
up another couple of threads of which I will need you to be aware.
FROM EVENTS TO STATES-OF-AFFAIRS;

CONTINUITY, PATTERN, AND FLUX
In metamorphology, we avoid the understandable but muddling ten-

dency in science to reify "events" -a tendency many have long criti-
qued-by turning our attention from events to states-of-affairs described
in terms of pattern or invariance. We seek to account for states-of-affairs
obtaining at certain points in time, and eschew all talk of events.
Conceptually, this is closely related to a second further thread I must
tease out for you. Namely, in metamorphology it is persistence and not
flux which is the focus of our inquiry. This may sound odd for an engage-
ment concerned with the study of change! But first of all, we distinguish
flux sharply from change. To oversimplify rather drastically our own
account of the matter, flux refers merely to variance-the temporal dif-
ferentiae in nature which make the attribution of change possible at all.
Flux may be random, chaotic, or it may be simply patterned, and it
is characterised as much by continuity as by change-where the latter
refers to an attribution of a descriptive difference in some defined in
variance. The tides-here is example of patterned flux-may fluctuate
continuously, but as the tide tables show, they can hardly be said, in
AN INTRODUCTION TO METAMORPHOLOGY 75
this strict sense, to change. A 'sudden' solar eclipse from our (layman's,
frightened natives') point of view is only continuity from an astronomer's
point of view. If a solar eclipse just failed to occur when expected by the
astronomers-well now, that would be a turn-up for the books!
As to the question of where change comes into our investigations, I
shall return to this in a few moments. For now, let us note that we seek
in this area of work to account for pattern, that is, the persistence of
descriptive invariance in states of affairs.
A "PHOTOG RAP HI C NEGATIVE" VIEW OF NATURE
Moreover, we adopt, if you will, a "photographic negative" view of na-

ture. On this figure-ground reversal of the Official view of the universe,
we hold that, all things being equal, we are entitled to expect continuous,
random flux everywhere; the persistence of any particular order or pat-
tern in any region of this universe is thus viewed as highly improbable,
and needs accounting for. From this perspective, persistence presupposes
mechanism, and it is our job, as scientists, to elucidate the mechanism
that accounts for the persistence of any descriptive invariance-pattern.
If there were ceaseless, unconstrained, random flux, there would be,
in any assembly we might be observing, an infinity of possibilities that
might now be realised just here. But only a small subset of the logical
possibilities are actual. The behaviour of the assembly is thus said to
exhibit constraint.
THE SEARCH FOR CONSTRAINTS
Our scientific inquiry is aimed at revealing the nature of these con-

straints. These constraints will include certain universal, empirically
established invariances we normally refer to as "laws of nature" or of
physics (e.g., that light cannot travel faster than c).
But our inquiry need not stop there, that is, we need not confine our-
selves to citing only such universal invariances as happen to apply. For
in specifying more fully the set of constraints-on-variance that preclude
all states of affairs other than the one we are seeking to account for,
the theoretical edifice of metamorphology provides us with a rich set
of methods for additionally specifying, with the highest degree of sci-
entific rigour, those far more numerous sources of constraint which are
not universal at all but quite concrete, local, and idiosyncratic to the
situation.
76 JAMES WILK
NEGATIVE EXPLANATION
Our scientific explanation is not complete until we can satisfy a ques-

tioner's quite specific "why this rather than that." To do so we shall find
it essential to specify constraints which may be idiosyncratic to this as-
sembly, in this context. And so, in metamorphological explanation, the
concept of flux-and-constraint replaces the much more epistemically per-
missive concept of cause- and-effect. A state-of-affairs is now only to be
regarded as having been explained scientifically once it can be demon-
strated to be the only state-of-affairs not currently precluded [3]. As you
can begin to see, perhaps with some horror, our explanations are always
explanations in terms of unities.
At the risk of damaging my credibility further still, perhaps now to
a point beyond all repair, I must confess to you that the concepts with
which we are able to operate rigorously, in metamorphology, include
purpose, intention, and design, which again can be defined in terms of
imparity and rigorously derived theoretically [25] from some pretty basic
logical and physical concepts, once we allow description an explicit and
legitimate place in our armoury of fundamentals.
THE NATURE OF CHANGE
Change is a difference over time-a descriptive difference. And so change

is always change-from-a-point-of-view.
In metamorphology, we analyse situations in terms of an agent assem-
bly (that which does) and a patient assembly (that to which something
is done), and we look at an agent's observation/attribution of change
in the context of the agent's intervention into the free fall of the pa-
tient assembly, as seen from the agent's point of view, i.e., with respect
to a set of aspects heeded. The agent's intervention to create a pre-
defined change in a state-of- affairs-from one description of the patient
to another-is viewed as a design problem. The agent's behaviour at any
point in time is seen as the only behaviour currently possible given the
design constraints within which the agent is operating. The patient as-
sembly, whether it is basically inert or whether it can also be analysed in
agent terms, similarly is viewed as being in the only state not currently
precluded.
FREE FALL AND INTERVENTION
\Ve can define a "descriptive space" as a set of (mutually orthogonal)

descriptive dimensions (A set of dimensions is mutually orthogonal if
for any given particular, its value on any of these dimensions cannot
A N I NT R 0 D U C T I 0 N T 0 M ETA M 0 R P H 0 L 0 G Y 77
be deduced from its value on any or all of the others). The description
of a particular along all of the dimensions constituting the descriptive
space can be given by a vector (n-tuple) specifying the coordinates of the
particular's "representative point" [1] within the space (For shorthand,
rather than speaking of 'the location of a particular's representative
point' within the space, we shall refer to 'the particular's location' in
the space). The concept of a descriptive space is developed much more
rigorously in [25]. As a particular's description along these dimensions
changes over time, the particular will come to occupy different loca-
tions within the space, and we can track the trajectory of the particular
through the descriptive space over time.
For some agent assembly to intervene in the free fall of some patient
assembly is for the agent to act upon it such that the patient's trajectory
through a defined descriptive space will be different to the trajectory it
would have followed had the agent not so acted. But it is only meaningful
to consider the "free fall" of an assembly with respect to the possibility
of some agent's intervention. Free fall is thus defined as an assembly's
trajectory through some descriptive space over time~the description
of some assembly as it would unfold over time~if the agent did not
intervene at all. Free fall is thus always relative to a point of view of~a
set of descriptive aspects heeded by~an agent. The agent intervenes in
order that the patient will be descriptively different over time~from the
agent's point of view~to what it would have been in the absence of his
intervention.
THE NATURE OF BEHAVIOUR
Some descriptive pattern, some invariance is to be replaced by another

which is preferred by the agent. In metamorphology, whether in the hu-
man realm or in the wider biological realm, the agent's purposes and
point of view are paramount, and any valid scientific explanation must
take account of them, not seek (per impossibile) to explain them away.
Behaviour is intervention, the purposeful achieving of a selected descrip-
tive state-of- affairs in preference to the alternatives. (Cf. [27] on be-
haviour as the control of perception.)
But what is sauce for the goose is sauce for the gander. The patient's
response to the agent's intervention will itself be an intervention on the
patient's part. The patient must be construed as an agent. And it is
always with reference to the first agent's knowledge of the nature of
the-patient-assembly-as-agent that the intervention is (ideally) to be de-
signed, and (properly) understood. We can only understand the 'inter-
action' between two assemblies by understanding what each is "trying
78 JAMES WILK
to do" (There need be nothing 'anthropomorphic' about this notion, but

of course man, we should argue, is best understood anthropomorphically
anyway) [15]. Behaviour [27] is thus understood in metamorphology, as
in cybernetics, as the behaving entity's varying of all kinds of things
which do not matter to it, in order to maintain or secure the things
(descriptive states-of-affairs) which do matter to it.
What is more, our analytical methodology here ensures that we can
never forget that the patient assembly into whose free fall an agent
assembly intervenes is never part of some common universal system of
nature-not part of the territory, as it were-but is part of the-territory-
as-mapped-by-the-agent. An agent intervenes in a context-as- under-
stood. Our universe is thus seen not as a grand unitary system but as a
collection of Leibnizian windowless monads, each mirroring the activity
of the other monads from its own unique, subjective point of view.
SCIENTIFIC OBJECTIVITY
This perspective thus brings with it a distinct view of what objectivity

must needs involve in the scientific engagement.
The subjectivity of the object of study is the asymptote of objective
understanding.
The sine qua non of the scientific engagement, in our view, is the elu-
cidation of mechanism. In science we aim to achieve greater and greater
fit between the distinctions in terms of which we seek to understand
the mechanism of the assembly, on the one hand, and the distinctions
from which the assembly is itself actually operating on the other hand
(Metamorphology, by the way (but not incidentally!), is rooted firmly
in a philosophy of science which is uncompromisingly foundationalist,
essentialist, and realist, and is committed unreservedly to a correspon-
dence theory of truth, and in all of these respects it is thus at direct
odds with the whole 'postmodern' tradition).
THE OFFICIAL VIEW
According to the Official view, which has held sway for over three and a
half centuries, the universe comprises a vast, hugely complex and richly
interconnected causal system of events, governed by a relatively small
number of universal causal laws, some yet to be discovered. This Official
universe is hierarchically arrayed in a sequence of interlocking stratified
layers (or nested shells), with the phenomena of subatomic physics at the
lowest level (or at the core), and with layers of 'higher-level' phenomena
successively arrayed upwards from this base (or outwards from this core).
AN IN T R 0 D U C T I 0 N T 0 MET AM 0 R P H 0 L 0 G Y 79
The laws, known and unknown, governing the behaviour of the vari-
ous classes of interacting elements together determine the course of ev-
erything that transpires in the universe. On the Official view, if there
is any room for slippage between these universal laws and individual
goings-on, it can only be the result of a kind of leakage-some residual
indeterminism in the universe due either to quantum uncertainty or to
the predictive difficulties of non-linear equations [22].
THE END OF THE OFFICIAL UNIVERSE
There was once, and for many there still is, a heady and enticing ratio-
nalist dream of nature as such a universal, richly interconnected system
unfolding from a relative handful of fundamental laws. I have come to
believe, however, that this dream is just that: a dream. In its place,
metamorphology posits an uncompromisingly empiricist alternative.
In metamorphology, we replace the Official triad of notions of object-
and-forces, cause-and-effect, and conformity to universal laws or regular-
ities, with an alternative triad of notions: flux-and-constraint, purpose-
and-design, and adjustment to locally prevailing conditions. And we re-
place the Official picture of the universe as consisting of a hierarchically
ordered system of interacting, homogeneous classes of events with an
alternative picture of the universe as consisting of a myriad of idiosyn-
cratic, autonomous phenomena, unfolding and responding to the vagaries
of local circumstances which are always unique and unrepeatable.
The success of Official, rationalist science at uncovering general laws
with wide applicability shows not that all phenomena in the universe are
to be completely described by such regularities. Rather, it only shows
that the severely limited rationalist methodology, however redoubtably
successful it has been within its own terms of reference, is suited only
to discovering such universal regularities and is eo ipso oblivious to the
vast majority of distinct phenomena that occur in the universe and which
equally merit rigorous empirical, scientific study [28].
THE REALITY OF DIVERSITY
On our view, the rich diversity of phenomena, everywhere apparent, is

real and of fundamental explanatory importance. I mean this in two
ways.
First, it is no use pretending that we can ever arrive at universal laws
governing all phenomena everywhere, laws which, taken in aggregate,
explain everything whatsoever. The diversity of phenomena, even the
diversity of phenomena which you are now in a position to observe from
wherever you may be sitting or standing or reclining as you read this,
80 JAMES WILK
is not merely something to be explained away, but part of the essential

mechanics of the universe. There are, to be sure, regularities to be found
governing whole classes of phenomena. Indeed, it would be surprising if
all the pieces of my china dinner service did not display quite a number
of common invariances between them, or if one could not make remarks
about the mechanics of the Krebs Cycle as it occurred in cells in Rich-
mond, Surrey which did not differ from the mechanics of the Krebs Cycle
as it occurred in cells in Richmond, Virginia or in Kuala Lumpur for that
matter. But the narrow yet breathtakingly fertile investigatory paradigm
of Official science-comprising a tiny subset, roughly a twelfth or less, of
the set of detailed empirical methodologies scoped out by Francis Bacon
in his Novum Organum of 1620 which set out the basis of modern sci-
ence-this circumscribed Official set of methods, was suited only to the
discovery of such universal regularities and, alas, whatever happened to
be unique and idiosyncratic was simply filtered out of account.
Likewise, as Eddington has pointed out [6], it was hardly surprising
if Official physics has revealed to us increasingly a world consisting en-
tirely of classes of inert, atomistic entities "bound to one another by
mathematical equations forming a deterministic scheme," for in physics,
traditionally, mathematics has been the model of exact inference, and
wherever the mathematical chain could not be completed the physicist
was forced to admit that knowledge of the kind he makes his stock-
in-trade was not yet for the having [6, p. 270]; I have critiqued the
pretensions of Official physics elsewhere [22], but the important point
here is that once we find ourselves forced, as scientists, to take seri-
ously the description of assemblies singled out by biological-let alone
human-agents with a point of view (a set of descriptive aspects being
heeded by the agent), then the diversity of phenomena can be seen to be
real and efficacious and now must enter fundamentally into our scientific
explanations, that is, into the explanantia as well as into the explananda.
To summarise our line of argument on this point, which I have been
able to do little more than allude to in this introduction: If we cannot
do without the concept of change, then we cannot do without interven-
tion, without purpose and description and point of view, in our scientific
explanations of states-of-affairs. And the purposes of agents, especially
given the infinitely redescribable nature of reality, are as diverse as you
please. Diversity will not go away.
This, then, is my first point about the reality of diversity and its fun-
damental explanatory importance. But there is a more critical point still,
namely, second, that we must not expect that all explanations will be of
the same form, as Aristotle, Ryle, Toulmin and others have also been at
pains to point out [29]. Explanations of why this rather than that-the
AN INTRODUCTION TO MET AMORPHO LOGY 81
level of rigour we demand in any metamorphological account, in which

we show that no other state-of-affairs was currently possible-will them-
selves be of the most diverse form. Explanations of the behaviour of a
violinist performing in an orchestral concert, or of a gardener pruning
the roses in a particular manner, or of a philologist performing certain
analytical operations on a text, if the explanations of their behaviour are
to be genuinely objective and scientific, will need to rely principally not
on the knowledge of the behavioural scientist but on the knowledge of the
musicologist, the horticulturalist, and the philologist respectively. There
is no one form of knowledge which holds the copyright on irreducible
scientific truth [24]. The form of an ultimately acceptable explanation
will depend on what the questioner is seeking specifically to know.
But what is particularly exciting in all this for the natural scientist is
that whereas the Official methods gave us access only to regularities as
a subject for scientific investigation, the metamorphological approach
opens up the rich diversity of the unique, the unrepeatable, the id-
iosyncratic to rigorous scientific investigation and understanding. If the
mechanics of the universe are infinitely diverse and richly intertwined
with description-from-a-point-of-view, there is a vast, rich field in which
empirical science can make its contribution to the world by revealing
and elucidating those mechanisms. My own empirical work, in meta-
morphology, has largely focussed on the exploration of the idiosyncratic
but rigorously demonstrable, ineluctable mechanics of one-off situations,
whether in nature or in the world of affairs [21].
THE BROADENING OF PHYSICS
The laws of physics are at a very high level of abstraction and ap-
ply equally to phenomena that are widely dispersed in space and time.
However, the laws of physics tell us very little about what goes on in
the universe, and for all but a relatively small number of very simple
phenomena they explain-and, even ultimately, can explain--very little
indeed. This is not to imply the least discredit to physics (where indeed
would we be without it and its dazzling record of achievements?), any
more than it is a discredit to the aeronautics industry that their prod-
ucts are quite useless for getting about town. Universal laws, such as
those discovered by physicists, are handy and, when aptly abstracted
and accurately delineated, quite effective for getting us to the sorts of
theoretical and practical destinations to which such laws are useful for
taking us. But there are other sorts of places, more local destinations,
where we shall need to go, and there we must rely on more local forms
of-equally scientific-transport.
82 JAMES WILK
This dissident, metamorphological view of the universe gives more

scope to empirical science than the rationalist, 'systemic' universe ever
allowed. It permits us rigorously to investigate empirically an infinitely
greater range of phenomena in the universe. What is more, where the
'same' phenomenon can be predicted or retrodicted on both types of
explanatory account (that is, from within an Official, causal framework
or from within a metamorphological framework of flux-and-constraint),
the phenomenon can virtually always be predicted or retrodicted with
far greater precision or specificity within our dissident, 'negative' epis-
temology, as we have amply shown [25].
THE ORDINARY UNIVERSE:

REDISCOVERING THE REALITY OF EVERYDAY LIFE
Most excitingly, in my view, metamorphology permits us to deal rigor-

ously with the rich, concrete, idiosyncratic details of real life, to treat
diversity seriously as an essential feature of the universe, and to accord
fundamental explanatory significance to those very features of everyday
life which the man or woman in the street knows to be of importance.
From this point of view, the rules of bridge or cricket, and the kinds
of information contained in Hoyle or Wisden, may turn out to be ev-
ery bit as fundamental as the laws of physics in accounting scientifically
for what transpires at a given juncture in space-time, if what transpires
there is, say, a card game or a cricket match.
THE NATURE OF MIND
Which leads us to something I have not yet called by name, without

which my short paper would be like a synopsis of Shakespeare's Hamlet
omitting to mention the Prince of Denmark. I am referring here, of
course, to mind. Mind in human life, as some of my own work has argued
[25], appears from this point of view as an abstraction referring to the
realm of finding our way about in the world according to our designs,
the realm of intervening in contexts according to our understanding of
those contexts. It is the realm of intervention in contexts-as-understood.
MIND AND BODY
What then, on this view, is the relationship of mind and body? Quite
simply, on this view, there is no relationship between mind and body.
The alleged mind-body 'relationship' is like the relationship between
the separation of powers in the American Constitution, on the one
hand, and the arrangement of partition walls in the architecture of the
AN IN T R 0 D U C T I 0 N T 0 MET AM 0 R P H 0 L 0 G Y 83
White House, on the other hand. One has nothing to do with the other.
'The mind' and 'the body' are both, equally, abstractions, but they are
abstractions- abridgements, 'arrests in description' [8]-along very dif-
ferent lines. To the extent that we are only concerned with the handful
of aspects of some assembly corresponding to the tiny handful of as-
pects studied by physics (aspects concerned with, or derivable from, the
descriptive dimensions of distance, time, mass, etc.), to that extent we
abstract the physical aspects. To the extent that we are concerned to
understand the mechanisms of behaviour by placing the behaviour in
the realm of conduct ("the complex background in which it is enacted"
[4]) to make it intelligible, to that extent we abstract the mental.
There is an important sense in which 'the mind' is no more closely
related to the brain than it is to any other organ of the body [25]. And
for my own part I can confidently assert, after more than two decades
of investigation as a neurophysiologist, cybernetician, cognitive scientist
and philosopher of mind, that the study of the brain has so far thrown
no light (but much darkness) on our understanding of the mind, nor
is it ever likely to do so. The whole business, on my reading of things,
involves a philosophical error of the first water [25].
Now 'traditionally', cognitive science has started from the false premiss
that we have nature (equated spuriously with the so-called 'physical
world') pretty well understood, at least in broad outline. In other words,
we've got nature taped. The anomaly remaining to be accommodated
within the physical scheme, so the story goes, is mind, and cognitive
science seeks accordingly to understand the nature of mind and its place
within nature as thus, Officially, understood. The approach adopted by
my colleagues and myself, however, has been to start from the other
end and proceed in the opposite direction. We aim, as D.J. Stewart has
aptly remarked [14], not to reduce psychology to physics but to bring
physics up to the level of psychology. And so, in stark contrast to the
prevailing view in cognitive science, we have taken the view that we have
not yet even begun to understand the mechanics of nature-blinded, as
we have been, by the physicists whom we have permitted, in E.A. Burtt's
phrase, to make a metaphysic out of a method. But if we understand
nature aright, along the incomparably richer lines of metamorphology,
mind is no longer an anomaly at all.
THE MIND OF NATURE
For mind, more generally, from this point of view, can be seen to refer
to those aspects of phenomena in the universe which inherently involve
description, significance and point of view, where a defined pattern has
84 JAMES WILK
its source in imparity, and so is the result of a design, and no less so even
where, as in Darwinian evolutionary theory, there is no designer [30].
Design or purpose-what Cudworth (1628) [5] called "plastic natures"-
can be shown, in metamorphology, to be as essential to understanding
the mechanics of the majority of observable phenomena in the universe
as are the concepts of mass/energy, length, and time.
The framework of metamorphology thus not only provides a philo-
sophical and natural-scientific basis for taking purpose as central to our
understanding of human conduct. For more broadly, it provides a the-
oretically parsimonious basis for including purpose and meaning within
nature, yielding a new and readily tractable understanding of the rela-
tionship between the mental and the physical.
A GENERAL THEORY OF INTERVENTION
What has so far interested some people the most, however, about the
field of metamorphology-particularly in government and industry-has
been, predictably, the emerging practical applications of this work. The
core of the detailed theoretical structure of metamorphology, which we
have hardly even alluded to, is a General Theory of Intervention which
provides an account of what it is for anyone or anything to act and
what it is for anyone or anything to be acted upon. This theory turns on
the concepts of description, point of view, agent, patient, free fall, and
intervention.
TRANSFORMATION UNLEASHED:
THE POWER OF THE NUDGE
Practically speaking, this has yielded, in turn, a method of agent/patient

analysis for systematically identifying the smallest intervention in a given
situation that will be sufficient to trigger an all-or-none transformation
or 'flip' to some pre-defined, desired state-of-affairs [31].
I have sometimes put the matter facetiously by saying that if chaos
theory shows us how a butterfly flapping its wings in Peking can alter
the weather over New England, this methodology provides a means of
identifying which butterfly has to flap its wings for how long just when
on which street corner in Peking to guarantee a beautiful Labor Day
weekend on Cape Cod. More seriously put, the method, closely related
to the method of negative explanation, permits the apparent complexity
of any situation to be filtered to reveal the smallest intervention by an
agent assembly that will result in the desired transformation in a patient
assembly. By identifying the set of constraints that make the present be-
haviour of the patient assembly the only behaviour currently possible, we
AN IN T R 0 D U C T I 0 N T 0 M ETA M 0 R P H 0 L 0 G Y 85
can pinpoint which constraints, if lifted, will yield the transformation de-
sired. This theoretical capacity for "minimalist intervention" has already
been applied successfully to creating significant real-world transforma-
tions by means of very small sets of precisely pinpointed actions. Some
of the practical implications have already been tested impressively by a
number of sponsoring government agencies and multinational corpora-
tions, in applications ranging from the resolution of intractable problems
to the overnight transformation of corporate culture.
TRANSFORMING HOW NATURE IS SAID
It is heartening, to say the least, to find that what has seemed to me and
my colleagues to be such a rich field of study, such a powerful explana-
tory framework, and such a profound and fertile epistemological shift,
should turn out, so early in the game, to have such profound empiri-
cal and pragmatic consequences. These consequences, of course, sound
startling only from within the narrow stable of the Official View. If we
can include within Nature a cluster of fundamental concepts tradition-
ally excluded, we can come to adopt an epistemology possessing better
fit to the way Nature herself in fact would appear to operate~namely,
locally, concretely, and in terms of unities, within a universe in which
description and meaning play a critical determining role.
The founder of modern scientific methodology, Francis Bacon, said
that nature can be commanded only by obeying her. If we can achieve
this better fit, this greater obedience of our explanations and interven-
tions to the actual dictates of Nature, then we can expect far greater
(and more ecologically robust) command of Nature than the Official
universe has so far accommodated.
AFFILIATION
Dr. James Wilk

Interchange Associates
Toronto
Ontario M6R 1M1
Canada
REFERENCES
[1] Ashby, W.R., An Introduction to Cybernetics, Methuen & Co, Lon-

don, 1956.
[2] Austin, J .L., Philosophical Papers, Clarendon Press, Oxford, 1961.
86 JAMES WILK
[3] Bateson, G., "Cybernetic Explanation", American Behavioral Sci-

entist, X, 8, April 1967, pp. 29-32. Reprinted in Bateson, G., Steps to
an Ecology of Mind, Paladin, London, 1973.
[4] Bedford, E., "Emotion" Proceedings of the Aristotelian Society,
LVII, 1956-57.
[5] Cudworth, R., The True Intellectual System of the Universe, 1628.
Reprinted by Thoemmes Press, Bristol, 1994.
[6] Eddington, A.S., The Nature of the Physical World, Cambridge
University Press, Cambridge, 1928.
[7] Gosling, J.C.B., Plato (in the Arguments of the Philosophers series),
Routledge & Kegan Paul, London, 1973.
[8] Oakeshott, M., Experience and Its Modes, Cambridge University
Press, Cambridge, 1933.
[9] Powers, W.T., Behaviour: The Control of Perception, Wildwood
House, London, 1974.
[10] Powers, W.T., Living Control Systems I, Gravel Switch, The Con-
trol Systems Group, Kentucky, 1989.
[11] Powers, W.T., Living Control Systems II, Gravel Switch, The
Control Systems Group, Kentucky, 1992.
[12] Ryle, G., The Concept of Mind, Hutchinson's University Library,
London, 1949.
[13] Ryle, G., Dilemmas, Cambridge University Press, Cambridge, 1954.
[14] Stewart, D.J., "A Ternary Domanial Structure as a Basis for Cy-
bernetics and Its Place in Knowledge", Kybernetes, Gordon & Breach
Scientific Publishers, 1989.
[15] Taylor, R., Action and Purpose, Englewood Cliffs, N.J., Prentice-
Hall, 1966.
[16] Toulmin, S., The Uses of Argument, Cambridge University Press,
Cambridge, 1958.
[17] Wilk, J ., "Culture and Epistemology", The International Journal
of Systems Research and Information Science, October 1989.
[18] Wilk, J., "Some Common Myths About Change" Invited Address,
British Academy of Management Annual Conference, Proceedings, De-
cember 1991.
[19] Wilk, J., "The Art of the Nudge: Minimalist Intervention and the
Science of Change" Invited Address to the Strategic Planning Society,
Privately Circulated, June 1993.
[20] Wilk, J., The Nature of Change, in press, 1994-1997.
AN I NT R 0 D U C T I 0 N T 0 M ETA M 0 R P H 0 L 0 G Y 87
[21] Wilk, J., Principia Metamorphologica: Novum Organum, Brunei

University, London, 1995.
[22] Wilk, J ., "The Metaphysics of the Physical World" invited address
to the Annual Conference of the Cybernetics Society (UK), September
1995, Proceedings, in press, 1995.
[23] Wilk, J., The Power of the Nudge: Pinpointing Effective Action in
the Face of Apparent Complexity, in press, 1997.
[24] Winch, P., The Idea of a Social Science and Its Relation to Phi-
losophy, Routledge & Kegan Paul, London, 1958.
[25] ibid. [21]' [20].
[26] ibid. [14], [21], [20].
[27] ibid. [9], [10], [11].
[28] ibid. [21], [22], [20].
[29] ibid. [12], [13], [16], [24], [2], [7], [21], [22], [20].
[30] ibid. [14], [21], [20].
[31] ibid. [17], [18], [19], [21], [23].
WILLIAM E. HERFEL
ON COGNITIVE AND SOCIAL DIMENSIONS OF SCIENCE:

CONSTRUCTIVISM & NONLINEAR DYNAMICS
INTRODUCTION
Juxtaposing the work of Einstein and Magritte has made for a wonder-
fully diverse and stimulating conference. As Michel Foucault has pointed
out, many of Magritte's works have entertained themes about the open-
endedness, indeterminateness and complexity of the relations between
language, world and imagination. His paintings reveal a world whereby
simple objects from our experience are combined in novel ways, providing
an elegant way to construct strange new phenomena from the familiar.
Einstein, on the other hand, urged us to accept a theory in which it is
possible to explain all novelty within a deterministic framework. The the-
ories of relativity show how it is possible to derive a complex world from
the geometrical structure of space-time alone. Many things could be said
about how we react when asked to think about the relationships between
these world-views. And indeed, many imaginative responses have been
forthcoming in the course of these proceedings. Unfortunately, I possess
neither the time nor the space to say anything more that is immediately
about the relationship between the work of these two great artists.
Nevertheless, my work meshes quite well with several themes addressed
at this conference. Following are a few words about how this paper fits
with others presented here. I am attempting a synthesis between the
work of social constructivists (particularly Bruno Latour) and that of
complex adaptive systems theorists (the likes of Ilya Prigogine, Heinz
von Foerster, Brian Arthur and Christopher Langton, to name a few
representatives present at the conference). Despite the surface tensions
between these diverse vantage points, I think that at a deeper level these
ideas can be married quite happily. To see how, one must first see the
task of philosophy of science from my perspective. In my view, the job
of philosophy of science is to construct models of the scientific process.
( C.A. Hooker and I have recently shown to what extent traditional phi-
losophy science succeeds in this regard in a paper presented to the 1995
International LMPS Congress.) Since scientific research is a social pro-
cess, this work becomes fundamentally one of social dynamics. Social
constructivists have been busily amassing data toward this end.
89
© 1999 Kluwer Academic P1tblishers.
90 WILLIAM E. HERFEL
My larger project is to provide a framework within which this data can

be assimilated, interpreted in such a way that it sheds light on traditional
questions in the philosophy of science, as well as the methodology for
understanding its social significance. Specifically, my project has three
mms:
1. To use the concepts of nonlinear dynamics to articulate clearly the
results of the social constructivists.
2. To undermine the formalist notion of rationality, so that it is pos-
sible to view science as simultaneously cognitive and social.
3. To apply critiques of neo-Darwinism, from the perspective of com-
plex dynamical self-organising systems theory, to work in evolu-
tionary epistemology (like that discussed so eloquently by Barbara
Herrnstein-Smi th).
At a later time I hope that the picture of science I am in the process of
constructing can be applied to the kind of science policy issues addressed
at this conference by John Ziman. In this context, however, my remarks
must be brief. I will focus on one issue from aim 1., only touching on
related issues from the other two aims. I will apply the concepts of
nonlinear dynamics to analyse the social constructivist discussion of the
dichotomy between the rational and the social. Even with such a narrow
focus, only a sketch of the argument can be presented here. These points
will be elaborated upon at a later date.
WHY CONSTRUCTIVIST PHILOSOPHY OF SCIENCE

NEEDS SOCIAL DYNAMICS
Recently, constructivist accounts of science have been proposed by phi-

losophers (e.g., Churchland and Hooker [3]; Giere [7]; Hacking [9]; van
Fraassen [23]), historians (e.g., Galison [5]) and sociologists (e.g., Latour
and Woolgar [13], [14]), sharing the view that scientific facts (and in
some cases entities) are human constructs. Constructivist accounts con-
trast with those provided by formalists who sought to explain science in
terms of application of a priori, formally and eternally valid methodology
(say e.g., the work of Hempel and Popper).l Formalists trivialise social
factors, for the details of scientific dynamics don't matter, so long as
ultimately the only social aspect that counts is the actualisation of the
application of the proven methodology to the subject matter at hand.
In fact, the only other way that political, economic or social influences
can enter into this account are as impediments to the process. Thus
the traditional account entails that science must be insulated from such
1
The literature is vast ([7] [9] [11] for good summaries).
CONSTRUCTIVISM & NON LINEAR DYNAMICS 91
"outside" interference. This is the source of the traditional dichotomy

between rational scientific aspects and nonrational social aspects in sci-
ence.
Constructivists, by and large, recognise that far from rendering opaque
the world which could be made otherwise transparent by unbiased ap-
plication of scientific method, macrosocial inputs (e.g., capital) provide
vital material resources enabling the scientific enterprise to operate. Ac-
cepting this does not require one to be a social determinist. I do not sub-
scribe to the view that science can be reduced to sociology any more than
it can be reduced to observation and formal logic, taking very seriously
the material constraints on the fabrication of facts. Certainly imagina-
tion and creativity play a role, but scientists don't simply "dream up"
scientific facts, any more than an architect can "dream up" a building.
An architect who repeatedly produces designs that cannot be realised in
structures won't remain an architect very long.
Researchers in fields from meteorology to biology to economics have
been constructing models of complex systems. A unified language of non-
linear dynamics is starting to emerge. Philosophers, historians and soci-
ologists must also construct models in order to understand the process
of scientific activity. I contend that complex systems dynamics provides
tools appropriate for this task. The application of these tools to the
study of science will lead to a better understanding of cognition within
a social context, moving the discussion beyond the crude dichotomy of
the rational and social.
ON THE DICHOTOMY OF THE RATIONAL

VERSUS THE SOCIAL
In Laboratory life, Latour and Woolgar adopt as a methodological

principle the thesis that the social and cognitive are not dichotomous.
They discuss how traditional approaches to the study of science embrace
a dichotomy between the "social" and the "scientific" or "cognitive" con-
tent of the activity of research, a strategy deployed by the scientific prac-
titioners themselves in order to challenge the results of their competitors
[14]. Depending on the account one follows, the cognitive involves either
"intellectual" or "technical" factors or both. On this conventional view,
normal science is dominated by the "scientific", and it is only in cases
of "extreme secrecy, fraud or other relatively infrequent occasions" [14]
that extraneous social factors intrude into the practice of discovering
knowledge. To this they respond:
Our intention is to understand how this distinction features in the
activities of scientists, rather than to demonstrate that emphasis on
one or the other side of the duality is more appropriate for our un-
derstanding of science. [14]
Among the methodological reasons for rejecting the dichotomy Latour
and Woolgar cite Barnes' "symmetry thesis": social factors are no more
or less relevant for explaining the successes of science than explaining
its failures [14].
Applying the tools of nonlinear dynamics permits a finer-grained ac-
count than is currently available in the social studies of science literature.
Considered from the perspective of nonlinear dynamics, the dichotomy
becomes a strategy employed in the modeling of science as a complex
system. Imagine a model of science in a state space whereby dynami-
cal constraints pick out a collection of trajectories through the space,
a specific trajectory selected by particular initial conditions. To con-
struct such a model we must choose features of the system which will
appear as dimensions represented by orthogonal axes. For example, we
could choose for our dimensions: available methodology, economic re-
sources, career interests and technological expertise. One reading, from
our present perspective, renders the dichotomy between the rational and
the social conceptual: in the terms of systems dynamics, this amounts to
stating that the social and the cognitive dimensions can be realistically
modeled on orthogonal state space axes.
It is compatible with Latour and Woolgar's account that the dynamics
of science is not (conceptually) decomposable along such lines, thereby
social and cognitive factors could not lie on orthogonal axes. But let us
suppose that they can. Even then the dichotomy could be maintained
dynamically only if the space in which science's dynamics could be cap-
tured has a linear topography. In such a (linear) space all vectors relevant
to the dynamics can be decomposed into the sum of orthogonal vectors.
Thus translated, even without the assumption that the cognitive and
social are conceptually inseparable, we can still retain a Latourian ob-
jection. For if the dynamics of science is nonlinear then any model in
a linear space, in which the dynamics is decomposed into the additive
play of independent forces, is bound to mislead. Hence even without La-
tour's radical assumption, which amounts to a very strong form of social
holism, we are still left with quite a powerful conclusion: rational and
social factors in science are dynamically interdependent making social
elements intrinsic to scientific activity.
NONDECOMPOSABILITY IN NONLINEAR DYNAMICS
We can better understand nondecomposability in nonlinear dynamics by

looking at one of its canonical examples, the early result of Lorenz [15].
C 0 N S T R U C T IV IS M & N 0 N LINE A R D Y N AM I C S 93
The paper documented the first systematic study of chaotic behaviour

generated by a small system of nonlinear deterministic equations. By
now this example has been discussed at length in the literature [2], [8],
[21] so my discussion will be condensed.
The Lorenz system consists of three ordinary differential equations in
x, y and z:
dx
-lOx+ lOy
dt
dy
rx- y- xz
dt
dz 8
--z +xy
dt 3
with r representing the control parameter. While no general analytic
solution of the equations is known outside the stable range (r > 24.74),
the behaviour of the system is observed by programming a computer to
iteratively generate a map of the trajectory of the system. 2 The result
is the now familiar Lorenz at tractor, the first strange attractor to be
observed with the aid of a digital computer. The attractor is strange
because, although the behaviour it describes is highly structured, it is
aperiodic. There are three aspects of the system's complexity:
1. While the behaviour is mathematically determined, it is aperiodic,
or non-repetitive. The trajectory neither fills the space nor visits
any point more than once.
2. Like many other chaotic trajectories, it exhibits a fractal structure.
Such structure is exemplified by self-similarity across scales. That
is, any magnified portion of the trajectory reveals a fine-grained
geometrical structure similar to the unmagnified trajectory.
3. The system is sensitively dependent on initial conditions. Simply
put, nearby trajectories diverge exponentially. Two time series gen-
erated by starting the system with initial conditions infinitesimally
close to one another diverge rapidly, and are eventually, after rela-
tively few iterations, no more alike than two trajectories generated
by initial conditions chosen at random. Thus errors made in the
measurement of the initial conditions of a system exhibiting such
sensitive dependence will grow exponentially with time, making pre-
dictive accuracy quickly evaporate as temporal span of prediction
is lengthened.
2
The discretization of the differential equations is an artefact of modeling them
on a digital machine. Such a method has been shown to be reliable in a large range
of situations. Nevertheless, the study of chaos has given rise to questions concerning
the generality of traditional numerical techniques [22].
These features combine to produce a dynamics that is at some scales

highly structured, while at others apparently random. Most important
for our present purposes is the question: What is the source of the nov-
elty? The complexity and unpredictability of the system is directly at-
tributable to the nonlinearly interactive terms, xz and xy, in the equa-
tions of motion [21]. Consider a system of linearly interacting equations
by ignoring the xz and xy terms. 3 Then z = z(O)exp( -(8/3)t), evolv-
ing independently of x and y. A similar result occurs in the x and y
dimensions. In the linear case the dynamics can be decomposed into
independent elements, one for each dimension. Thus, we can follow the
dynamics in any particular dimension without worrying about what goes
on in the others. In a recent paper Dyke [4] describes linear spaces as:
... the horne of systems of independent "forces" that can be separated
out in such a way that their individual contributions to the trajectory
of the system can be examined by themselves, small increments of
"force" produce small changes in trajectory, and we can say clearly
what the trajectory would have been had a given individual "force"
not been operating.
No such decomposition is possible in the nonlinear case. We cannot
follow the dynamics of each dimension separately. What happens in y
and z depends nonlinearly on what happens in x. Further, the complexity
and unpredictability of the Lorenz system does not occur in the linear
case where the interactive terms are not present. Thus the novel features
of the system come from the nonlinear interactions between variables in
separate dimensions of the system.
IS SCIENCE IN ACTION
A NONLINEAR DYNAMICAL SYSTEM?
I am not in a position to prove that science is a nonlinear dynamical sys-

tem. First we must address the issue of whether we should think about
science as a dynamical system at all. Certainly science has a history,
and it is not outlandish to attempt to understand that history by con-
structing models. To be dynamic, such a model would describe science's
components in terms of their states, and change in the states would be
described in terms of the "forces" that play among them. At this junc-
ture I think the only alternative to such an approach is to marvel at the
miracle of scientific prowess. I do not pretend to have a tidy set of equa-
tions upon which to base a dynamical model of science. Nevertheless, we
3
This could be done simply by adding a new parameter, a, to the equations as a
coefficient to both the xz and xy terms, and considering the case for a = 0.
C 0 N S T R U C T IV IS M & N 0 N LIN EAR D Y N AM I C S 95
are beginning to amass a list of features that nonlinear systems possess,

and under what conditions they possess them.
Finding that science possesses features typical of nonlinear systems
would support the dynamical approach I advocate, moving us beyond a
simplistic discussion of a "dichotomy" between rational and social factors
in science. Consider the following evidence:
1. As Kuhn [12] has argued, science does not linearly collect facts. Sci-
ence develops both by the elaboration of paradigms as well as by
the discontinuous replacement of old paradigms with new. Kuhn's
account has weaknesses, e.g., it does not accommodate the fact
that traditions of instrumentation and experimentation are partially
autonomous from theory, making scientific decision-making appear
more arbitrary than it actually is [5]. Again we are confronted with
a crude dichotomy: this time between rationality and relativism.
But as a first approximation, The structure of scientific re-
volutions, portrayed dynamically, is an account of the nonlinear
evolution through revolution of science.
2. Science has an irreversible character, while linear dynamics is gen-
erally reversible. Just as it is highly unlikely that the shattered
pieces of broken glass will jump back onto a table and reassemble,
entrenched results are much more difficult (although not impossi-
ble) to contest than are the tenuous ones at the cutting edge of
research [13]. For example, improvements in instrumentation (of-
tentimes provided cross-disciplinarily) allow for better experimen-
tal evidence. The use of x-ray diffraction to uncover the structure
of DNA is a familiar example. Some entrenched results are over-
turned, but by and large science (even in its revolutionary periods)
builds on past successes, making previously accepted results ever
more costly to reject.
3. As opposed to linear systems, the majority of the behaviour of sci-
entific communities does not appear periodic. The same patterns of
activity do not repeat as scientists successively explore their subject
matter as experiment, instrumentation and theory evolve. Seem-
ingly trivial periodic behaviour may play a role. But each new re-
sult, each new technique, leads to novel behaviour. At the level of
science as a whole we observe the bifurcation of disciplines into sub-
disciplines (at an ever increasing rate). Prigogine has pointed out
that this is a generic characteristic of the behaviour of nonlinear
far-from-equilibrium systems [16].
4. As Popper has observed, if we could reliably predict the future re-
sults of science, there would be little point to doing science. Science's
future may be impossible to predict for exactly the feature we know
makes some nonlinear systems impossible to predict: namely, small

discoveries, communications, technical improvements, etc. combine
in complex ways to produce large-scale consequences whereas were
initial circumstances slightly different the results may have turned
out largely different. Imagine, for example, the result if any one of
the contingencies Watson [24] describes had kept him, then an in-
significant post-doc, out of the Cavendish Laboratory. At the time,
Watson not ending up in Cambridge would have been a minor event
in the grand scheme of the history of theoretical biology, not even
rating a mention in the history books. Unbeknownst to those in-
volved in the decisions, however, the result could have been large
indeed.
Combining all of the above, the dynamics of the scientific process resem-
bles aperiodic speciation in biological evolution more than it resembles
the periodic swing of a simple pendulum. If I am correct, and science's
dynamics is nonlinear, then its dynamics cannot be decomposed into
the operation of independent factors whose effects can be "summed up"
in order to construct a model of its behaviour. Here Latour's insistence
that the cognitive dimensions of science cannot be properly understood
independently of its social dimensions gains new force: it could well be
that the success of science depends sensitively on social and cultural as
well as cognitive factors. The future success of science may rely on pre-
serving a delicate dynamic reliant on the complex nonlinear interaction
between all factors that influence research.
NONLINEAR SCIENCE
One might argue that order, not chaos, is what we want to promote
in science. And the surest way to guarantee order in a system is to
constrain it to linearity. But the robust chemical oscillators found in
organisms are nonlinear, as are a wide variety of other biological phe-
nomena [8]. There is no simple correspondence between orderly periodic-
ity and health. While chaotic cardiac fibrillation leads quickly to death,
aperiodicity is a ubiquitous feature of other healthy biological rhythms.
No general rule governs whether organic oscillations are normally either
chaotic or quasi-periodic (we rarely find strict periodicity). Instead it ap-
pears that life depends on a wide range of context dependent dynamical
patterns.
It is particularly noteworthy that aperiodicity occurs throughout neu-
ral processes. The olfactory systems of higher animals rely on chaotic
searching mechanisms in order to rapidly respond to smells [20]. Active
desynchronization through chaotic fluctuation is essential for healthy
C 0 N S T R U C T IV ISM & N 0 N LINE A R D Y N AM I C S 97
nervous system function in Homo Sapiens-the breakdown of which may

be the cause of Parkinson's tremors [6]. Aperiodicity may be crucial
for cognition. For instance, the dynamics of the central human cogni-
tive apparatus (otherwise known as the BRAIN) is quite complex, with
mounting evidence that chaos plays a crucial role [17].
Nonlinear dynamics, including the possibility for controlled chaos, may
be essential for maintaining robust flexibility in complex systems adapt-
ing in changing environments [20]. We are beginning to understand how
to accomplish such flexible control. Shinbrot et al. [19] review the results
(both numerical and experimental) of their experience with controlling
chaotic systems. They employ a local controller that can gently but con-
tinually nudge a chaotic system onto a chosen unstable periodic orbit. 4
Such orbits, away from which the smallest perturbation drives the sys-
tem, exist, densely packed, within the chaotic regime. Such techniques
can be used to steer complex systems from within without necessarily
having detailed knowledge of the global dynamics of the system.
COGNITION IN SELF-ORGANISING SCIENCE
Consider techno-science to be a complex social system, an evolving insti-

tutionalised structure, for obtaining information about, and construct-
ing devices for, the manipulation and prediction of the environment. One
aim of philosophy of science should be to understand this institution in
order to improve and coordinate it. The structure is complex and deli-
cate. The attractors, as well as the basin boundaries, may well be fractal
whereby small events can have large consequences. But we can recog-
nise the coherence: we can identify such complex patterns. Nonlinear
dynamics lends precision to our pretheoretic recognition.
If Latour is correct, and cognition is distributed throughout the sci-
entific process, from decisions made at the laboratory bench to political
decisions of rival research groups to economic decisions of funding agen-
cies, then an account of scientific cognition will emerge only from an
accurate study of this multi-dimensional dynamic. As Galison [5] has
argued, while acknowledging the role of "sociological influences" cuts-off
the rationalists' caricature of science as the conjunction of formal logic
and observation, placing science in a constructive social context need
not render science or scientists unreasonable. In fact, within the context
of social dynamics, scientific decision making becomes less, not more,
arbitrary.
4
See also Hooker, Penfold and Evans [10] for a similar local controller, announced
as exemplifying a new control paradigm, capable of steering nonlinear (including
chaotic) systems.
In the past, treating science in its complexity, accounting for even a

small subset of the multitude of factors involved, was practically impos-
sible. Accounts attempting to capture the complexity of science before
the advent of nonlinear dynamics, even if realistic, were bound to ap-
pear fluffy and imprecise. Only recently has science provided effective
tools for modeling complex systems. Nonlinear dynamics can sharpen
our analyses of science as a process. One aspect of scientific research
that has been ignored by traditional philosophic analyses is the posi-
tive feedbacks occurring between the market, technology and knowledge.
Structures emerging from these feedback cycles have recently been ex-
amined by Stanford economists (Arthur [1]; Rosenberg [18]). I advocate
a research program in science studies that can apply these insights in or-
der to further our understanding of how science works, so that it can be
gainfully employed in addressing the pressing global difficulties we face
as we march headlong into the twenty-first century. Through a careful
study of the dynamical structure of science's attractors we have a means
of beginning to deal with its complexity. We require such understanding
in order to intelligently coordinate such a potent self-organising institu-
tion.
AFFILIATION
William E. H erfel
Department of Philosophy
The University of Newcastle
New South Wales
Australia
P L WEH @cc. newcastle. edu. au
ACKNOWLEDGMENTS
I would like to thank John Atkins, Wayne Christensen and Barry Hodges
for free exchange of ideas and moral support. Special thanks go to Cliff
Hooker for his careful reading and critique of the manuscript and to
Chuck Dyke for critical comments and provision of his manuscript pre-
publication. Thanks to Kelly Boyle for proofreading and recommenda-
tions for stylistic improvements. This research was made possible in part
by support from ARC Grant AC9031991.
C 0 N S T R U C T IV IS M & N 0 N LINE A R D Y N AM I C S 99
REFERENCES
[1] Arthur, W.B., Increasing returns and path dependence in the econ-
omy, University of Michigan Press, Ann Arbor, 1994.
[2] Berge, P. et al., Order within chaos, Thckerman (tr.), Wiley, New
York, 1984.
[3] Churchland, P.M. and Hooker, C.A. (eds.), Images of science,
University Press, Chicago, 1985.
[4] Dyke, C., "Addition accretion accumulation acceleration alteration:
technology and nonlinearity.", (forthcoming).
[5] Galison, P., How experiments end, University Press, Chicago, 1987.
[6] Garfinkel, A., "A mathematics for physiology", American journal
of physiology, 245, 14, 1983, R455-466.
[7] Giere, R., Explaining science, University Press, Chicago,1988.
[8] Glass, L. and Mackey, M., From clocks to chaos, University Press,
Princeton, 1988.
[9] Hacking, 1., Representing and intervening, University Press, Cam-
bridge, 1983.
[10] Hooker, C.A., Penfold, H.B., and Evans, R.J., "Cognition
under a new control paradigm", Topoi, 11, 1992, pp. 71-88.
[11] Kourany, J. (ed.), Scientific knowledge, Wadsworth, Belmont, 1987.
[12] Kuhn, T., Structure of scientific revolutions, University Press,
Chicago, 1962.
[13] Latour, B., Science in action, Cambridge, Harvard, MA, 1987.
[14] Latour, B. and Woolgar, S., Laboratory life, University Press,
Princeton, 1986.
[15] Lorenz, E.N., "Deterministic nonperiodic flow", Journal of the
atmospheric sciences, 20, 1963, pp. 282-93.
[16] Peacocke, A.R., An introduction to the physical chemistry of bio-
logical organization, Clarendon, Oxford, 1983
[17] Rapp, P. et al., "Experimental studies of chaotic neural behav-
ior: cellular activity and electroencephalographic signals", in: Othmer,
H.G. (ed.), Nonlinear oscillations in biology and chemistry, Springer-
Verlag, Berlin, 1986.
[18] Rosenberg, N., Exploring the black box: technology, economics,
and history, University Press, Cambridge, 1994.
[19] Shinbrot, T. et al., "Using small perturbations to control chaos",
in Nature, 363, 1993, pp. 411-417.
[20] Skarda, C. and Freeman, W., "How brains make chaos in order
to make sense of the world", Behavioral and brain sciences, 10, 1987,
pp. 161-195.
[21] Stewart, I., Does God play dice? The mathematics of chaos, Basil
Blackwell, Oxford, 1989.
[22] Stewart, I., "Warning-handle with care", Nature, 355, 1992,
pp. 16-17.
[23] Van Fraassen, B.C., The scientific image, Clarendon, Oxford,
1980.
[24j Watson, J.D., The double helix, Weidenfield and Nicholson, Lon-
don, 1968.
C. CAPEL-BOUTE AND A. KOECKELENBERGH
NECESSITY OF A NEW PARADIGM IN EXPERIMENTAL

RESEARCH TAKING INTO ACCOUNT SPACE AND TIME
I have found that the scientific community may remain very conser-
vative and reluctant to put in question its fundamental criteria to char-
acterize the scientific method in experimental research: that is repro-
ducibility wherever and whenever the experiments are made in so-called
"identical conditions" ...
Happily, when I met Prof. H.S. Frank, the internationally known phy-
sicochemist author of a model of the structure of water, (in the Mellon
Institute in Pittsburgh (USA) in 1966), he introduced me to a biologist
and an economist Prof. Edward Dewey, (founder of the Foundation for
the Study of cycles), and they all, considering my views, recommended
me to read the book of Thomas Kuhn: "The structure of scientific rev-
olutions". I am grateful to them for warning me that "scientists fail to
reject paradigms when faced with anomalies or counter instances. They
could not do so and still remain scientists".
I knew thus that "inability to tolerate crisis" could drive "to desert
science", but also that "to reject one paradigm without simultaneously
substituting another is to reject science itself " ... "The act reflects not
on the paradigm but on the man". Pioneers in cosmo-biology like Tchi-
jewsky (1895-1964) in the ex-Soviet Union, or Piccardi (1895-1972) in
Italy had suffered until their death, and still afterwards, the criticism of
their colleagues, specially those of the same discipline.
Furthermore, a large interdisciplinary approach was more than unusual
in 1958 when we had held the 1st International Symposium on "Solar-
terrestrial Relationships in Physical-chemistry and Biology" at the Royal
Observatory of Belgium, under chairmanship of its director, Prof. Paul
Bourgeois [17, 3, 2].
It was specially intended to discuss an original methodology and its re-
sults, collected in different places, called the "Piccardi-chemical tests",
which had led him to present already in 1954 at an interdisciplinary
meeting in the Max Planck Institiit fiir Biophysik in Frankfurt, his "so-
lar hypothesis". It brought a coherent interpretation of evident abnor-
mal annual and long-term variations, appearing from mean values of
hundreds of daily experiments, carried out in so-called "identical condi-
tions", when ordered in function of time.
Forty years ago the scientific community had to face the choice either
101
102 C . CAP E L- B 0 UTE AN D A . K 0 E C K ELEN BERGH
to reject all the peculiarities and abnormal facts already observed inde-
pendently by many scientists in their different fields of research or to
accept a new paradigm that could interpret them; as stated by T. Kuhn
"paradigm changes do cause scientists to see the world of their research
engagement differently". Let me illustrate this with his most familiar
demonstration showing "what were ducks in the Scientist's world before
the revolution are rabbits afterwards".
Reproducibility in space and time, is based on the dogma that we
could operate in our laboratories in "identical conditions".
This meant not only a correct control of the inner parameters of the
systems observed, but also of the environmental factors already known
as able to act on a particular process, like light, temperature, pressure
and humidity, in a phytotron for example. But effects of low-energy
electric, magnetic or electromagnetic fields from our natural environment
were mostly ignored. Air pollution was known well before Lavoisier and
Priestley could identify the normal composition of pure air and our vital
need for oxygen ... because everybody could smell it! But radioactive
or electromagnetic pollution required not only specific instrumentation
to be detected, but also knowledge of their existence in our natural
environment and possible vital effects in spite of their low energy level.
It was when studying the effects and ways of action of such physical
factors, as they were used in industry for physical treatment of hard wa-
ters to prevent scaling in boilers, that abnormal facts were noticed quite
independently by G. Piccardi and myself, with different physical factors
and processes studied in carefully controlled laboratory conditions [15,
16, 18, 20].
1. the effects from such physical treatments of water could result from
an action on water itself persisting for days or hours.
2. laboratory test-reactions to characterize any effect of treated (ac-
tivated) water against untreated water (control in otherwise simi-
lar conditions) showed differences that could vary considerably and
even change from positive to negative effects in the course of time.
These could no more be considered as usual "experimental errors".
To test his hypothesis that the disturbing factors were of the same
nature, and could also act uncontrolled on water itself, Piccardi chose a
simple hydrolysis precipitation reaction (Bi Ch + H2 0 = Bi OCI + 2
H Cl) that could be carried out simultaneously with pre-treated or un-
treated distilled water, to observe in suitable conditions the variability
of the effect of a controlled physical treatment of the water. The effects
observed were on the germination, growth and subsequent sedimentation
of the resulting precipitate. The anomalous variability observed on the
NEW PARADIGM IN EXPERIMENTAL RESEARCH 103
sedimentation rates, even in the most rigorously standardized control

conditions led to characterize the effect only by the visible positive or
negative sign of the difference in the level reached by the precipitates
for each couple of reactions, carried out simultaneously at room tem-
perature. We can't enter here again in the discussion of the conditions
of the reactions, chosen to be rigorous but simple enough to enable a
single operator to carry out simultaneously ten couple of reactions to
be compared, and to repeat the test in function of time, in statistically
sufficient number of experiments.
The numerical result of series of Piccardi chemical tests is thus ex-
pressed as the frequency of a positive result (accelerated sedimentation)
as apparent effect of the controlled factor differentiating the two reac-
tions compared.
This numerical "variability index" may be plotted in function of time
as percentage for the daily series of experiments, or calculated as weekly,
monthly or annual means of these daily percentages (for technical details
and full discussion of all experimental conditions see [17, 3]).
The same test could be carried out differentiating the conditions of the
environment, either in a normal laboratory (test F), inside a Faraday-
cage (test D) or only with horizontal metal screens (test D'). It could
also be used to characterize in the same way the variability observed
for the screen effect in reactions with normal water (Test P N) or with
activated water (test P A) [5, 6].
Mean values of 50%, from a statistically sufficient number of experi-
ments, could mean just hazard or no effect for the factor differentiating
the couple of experiments repeatedly compared. With this simple orig-
inal methodology, Piccardi could reach, patiently, major conclusions in
favour of his hypothesis:
- after one month (600 experiments in March '51) it was clear that
significant but opposite results and patterns of variability were obtained
in his laboratory, either outside (F = fuori = 75%) or inside (D = dentro
= 30%) the Faraday-cage, just by screening the experiment for electrical
parameters in the environment.(Fig. 1) [14]
- after one year (more than 7000 experiments) there was an unusual
shape of annual pattern, with opposite trends for F and D - tests, par-
ticularly marked in March.
- after 4 years of experiments (more than 30000 in Florence) there was
a decaying trend of both tests F and D, that could be related with the
minimum of solar activity.
- positive correlations could also be established between the fluctuations
observed in Brussels and Florence, although mean values of the F tests
could be different, appearing as a latitude effect.
104 C. CAPEL-BOUTE AND A. KOECKELENBERGH
% T ~F
70
60
50
40
30
20
I 9 51 1952 1953 I 9 54 I 9 55
Figure 1. Monthly means of% T (accelerated sedimentation) observed on 20 daily

experiments in Piccardi Chemical tests-in normal laboratory conditions in the labo-
ratory (F-test)-inside a Faraday-cage shielded with a 0.1 mm copper sheet (D-test).
The "solar hypothesis" required first to reconsider any possible effects

of solar activity in the biosphere acting, directly or indirectly, through
geophysical factors, as electric, magnetic or electromagnetic disturbances
in the environment, in spite of their low-level of energy.
Piccardi could also explain with it, the abnormal annual variation,
taking into consideration the linear movement of the Sun in the Galaxy.
In this movement, the velocity of the Earth is more than double in the
equinox of March, when the Earth moves nearly in the direction of its
equatorial plane, than in September when it moves in a direction near
to its North South axes.(Fig. 2) [14]
With this new view, Piccardi could predict a dissymmetry in envi-
ronmental conditions in the northern and southern hemispheres of the
Earth, and a latitude effect.
To sustain this solar hypothesis, it was necessary to organize long-term
observations in different places, with a rigorously standardized method-
NEW PAR AD I G M IN EX PER I MENTAL RES EAR C H 105
Apex Apex
I ji
Figure 2. The Piccardi "Solar hypothesis" (A) Trajectory of the Earth in the Galaxy
(B) Resulting speed of the Earth in the spring and autumn (C) equinox
ology, carrying out simultaneously 10 couples of differential precipitation

reactions. It was done with Piccardi's synchronous mixer from 70° lat.
N (Tromsoe-Norway) to 70° lat. S. (Base Roi Baudouin Antarctic) and
in many places inbetween, with results of the tests confirming already in
1958, the dissymetry between Northern and Southern hemispheres.(Fig.
3,4,5) [17, 18]
The longest series of tests in Florence (1951 to 1972) and in Brussels
(1956 to 1978) showed that the environmental conditions responsible for
such fluctuating phenomena could change considerably from one solar-
cycle to the next one; that is, with the magnetic polarity of the couple
of spots developing on the sun in the 22 years-Hale cycle [7, 8, 19].
Dr. Koeckelenbergh, as an expert in solar physics, will point next at
the actual knowledge in astrophysics and geophysics that keep sustain-
ing Piccardi's "solar hypothesis". The so called "Solar Hypothesis" is
founded on the free motion of our Sun with his planetary system through
the Galaxy. We do not speak about the general galactic rotation. In our
Galaxy, at the solar system location there is a gravitational field of 10- 10
106 C. CAP EL-B 0 UTE AND A. KOEC KELENB ERGH
50
N
40-
30-
I I I I I I I I I I I I I
J MM J S N J
Figure 3. Mean annual variation of D'-test (under a horizontal copper sheet) in North-
ern and Southern hemispheres in 1957- 1960.
to 10- 11 Newton and a magnetic field of 3 nTesla (or gamma= 3 x10- 5

Gauss). G. Piccardi (1954) considers the magnetic field vector for his
hypothesis and A.A. Shpitalnaya (1979) concludes a possible influence
of the gravitational field on the solar activity. Nevertheless the measured
fields are of the order of the sensitivity of our best magnetometers.
NEW PAR AD I G M IN EX PER I MENTAL RESEARCH 107
0 -
50-
.1610
X
/
JO -f 1sso
R
I
0 100 200
Figure 4. Relation of the results of the D-tests in Florence with solar activity from
solar rotations 1610 to 1700. Ordinate: mean values of 20 daily experiments for 13
successive 27 days rotation of the sun ( =20 x 27 x 13 = 7000 observations for each
point) Abscissa: same mean values for Wolf's number R characterizing the importance
of solar spots.
The proper motion of the sun by reference to its local system of stars
is characterized by a speed vector of 19.4 km/s towards the direction of
the Apex which is situated in the constellation of Hercules. The plane of
the planetary solar system (the plane of the ecliptic) is tilted from 52.3°
on the Apex Axis. The orbital velocity vector of the Earth has a mean
amplitude of 29.9 kmjs and is normal to the radius vector Sun-Earth. It
turns in this plane with the annual motion of the Earth. In the plane of
the ecliptic the vernal equinox is turned approximately from 270° of the
plane who contains the poles, the Apex and the direction of the center
of the Galaxy. Thus, around the month of march, the vectorial addition
of the two vectors-reported on the Earth center-presents a maximum
in March and a deep minimum in September. We note that the northern
108 C. C APEL-B 0 UTE AND A. KOECKELENB ERGH
f
I'
,r~ '
.fJJ.', ' ' N
,;
~
'
'
''
' '
BRL'XELLES
FLOR E!'.:CE
Q.. - - - - - ----1-------- LlBREVll.LE
Q --------- l.EOPOLOVIl.l.E
'· FORT DAUPHIN
.. I<ERGUE l.E N
I
s
Figure 5. Dissymmetry of results of Piccardi F-tests in Northern and Southern hemi-
spheres in September 1958.
hemisphere is mainly front of the Apex during the whole year.(Fig.7)

The Earth magnetosphere is in interaction with the solar wind and
the heliosphere. Before 1958 our knowledge on the heliosphere and the
magnetosphere was very poor. Piccardi was very well inspired with the
formulation of his solar-galactic hypothesis in 1954, to explain both the
abnormal systematic annual variation in the sequence of his D-test (un-
der shield), which appeared as not sinusoidal with a singularity during
March that he related to the variation of the relative velocity of the
60-
so-
I r
5 0 +5
Figure 6. Effect of solar flares on mean values ofF and D-tests in Florence, significant
on keyday (0) only for test F, and not significant for 5 days before or afterwards.
Earth referred to the galactic center-and the dependence of the lat-

itude of the observing station, as it was known for the flux of cosmic
particles in the Earth magnetic field.
Er,;\~~-h c
Cn,·tcr
Figure 7. Combination of the motions of the Solar system towards the Apex and of
the Earth around the Sun. The largest arrow is the resulting vector of the proper
motion of the Solar system towards the Apex and of the projection on the same axis
of the tangential speed of the Earth in its orbital motion.
Now the solar wind appears as a radially continuous flux of an ionized

plasma of protons, electrons and 15% of helium nucleus (alpha particles).
That flux is spatially modulated by the slight general dipolar magnetic
field of the sun (of the order of + 5 Gs at the solar pole) and highly
perturbed by the local solar active regions. At the Earth orbit the total
residual field is measured around a few tenths of nT. The neutral plane
of the solar field is approximately its rotational equatorial plane. But
the perturbations by the solar activity create a neutral surface better
figured by the "ballerine skirt model" .(Fig. 8) The terrestrial magneto-
sphere, resulting from the interaction of the Earth magnetic field and the
charged particles of the solar wind, has theoretically a toroidal structure.
The solar wind creates a dissymmetry. In front of the sun the solar wind
faces the field lines of the Earth magnetic field acting as a "magneto-
spheric shield". In the opposite direction the structure is extended as a
"magnetospheric tail". Near the magnetic poles the field-lines form two
magnetic horns where the charged particles, originating also from the
solar wind as from cosmical galactic central regions and other sources,
SOLi\R BALLERINE SI<IKT
Figure 8. The "solar ballerine skirt". An approximate schema of the structure of the
magnetic field of the sun in-between the opposite N and S magnetic hemispheres
(neglecting the effect of the rotation of the sun)
travel towards the ground, generating auroral manifestations. In periods

of high activity of the Sun these may extend down to 50° lat.
The general solar field reverses its polarity during the solar activity
minimum. Thus the period of the magnetic solar cycle is twice the pe-
riod of the undecennal cycle: 22 years. The dynamical flux of charged
particles from the solar wind is deeply perturbed by the reversing of
the general field; the nature of the interaction is periodically reversed
for each hemisphere and the situation at the Earth orbit is very confuse
during a few months around the minimum of the cycle. During its revo-
lution around the Sun, the Earth and its magnetosphere, travel through
the wavelings of the so-called "ballerine skirt", changing of magnetic
sector and environment. At the boundaries of the sectors there is an
important discontinuity.
Piccardi had significantly correlated his F-test (without shield) with
solar flares.(Fig.6) Vladimirsky has indicated the association of the F-
test (without shield) and the sector discontinuities [23, 24].
The hypothesis of Piccardi is corroborated by these analyses as well
for the abnormal annual variation as for long term fluctuations. Now it
seems impossible to ignore the fundamental importance of cosmic and
heliophysical variables in the understanding of all the environmental
fluctuating phenomena [25].
From a methodological viewpoint some major conclusions, must first
be mentioned:
112 C. C APEL-B 0 UTE AND A. KOECKELENB ERG H
- in experimental research, time must be taken into account as a co-

ordinate.
- the relation with solar activity is not linear and presents a kind of
saturation when the number of Wolf (characterizing sunspots) is above
100 as shown by Becker in 1958 [2].
- screen effects and numerous experiments in altitude or underground
sustain the hypothesis also of effects of cosmic rays (C.R) with a negative
correlation with the sunspot cycle (maximum of C.R. for minimum of
sunspots known as Forbush effect)
- simple correlations with one environmental parameter must not be
expected.
- it is impossible to screen for all physical factors in the whole E. M.
spectra, from E.L.F. radio waves to C.R., that have been shown could
have significant effects in aqueous medium [9] and fluctuate uncontrolled
in the near and far environment [26].
- nevertheless, adequate specific experimental conditions and appro-
priate statistical methods of time-series analysis permit to guide the
research for environmental factors responsible for fluctuating phenom-
ena.
Without entering into the discussion of the appropriate statistical
methods used by F. De Meyer at the KM.I. (Koninklijk Meteorolo-
gisch Instituut-Belgium) [11], I would just like to quote his conclusions
concerning the fundamental questions which led to start such different
long-term series of experiments repeated in usually so-called "identical
conditions"
- first, "the time-variability" of the observed effects cannot be con-
sidered as being purely random... because the statistical information
content changes in time which implies nonstationarity of all time-series
considered" .
- "the sample means of either treatment effect (F and D') versus screen
effects (PN and PA) in Brussels are almost certainly unequal, and the
sample means of all three tests in Florence (F, D and P) are definitively
different. For all chemical tests mean values and standard deviations
are significantly different for successive subdivisions in the observing
interval.
- second, significant annual and long-term variations are present in
all data sets, but no common long-term pattern is observed ... no un-
ambiguous results point to a single or dominant cause". This is clearly
shown by spectral analysis of the data of monthly means chemical tests
in Brussels and Florence compared with some external factors usually
considered, as temperature, pressure.
As another example I would like to quote the remarkable work of

a microbiologist, Dr. Fiero Faraone (1992). He has observed daily the
frequency of incidence of CSD (Differentiated Sector Colonies) of het-
erogeneous bacteria collected from the outside air from 1970 to 1982,
and the frequency of CSD of staphylococcus aureus in laboratory cul-
tures from 1984 to 1991, and in 1992. The frequency of CSD expressed
in percentage of the total colonies counted, showed also mean values
with such an abnormal annual pattern with its minimum in March; but
a long-term variation was clearly in opposition to the sunspot numbers
of the undecennal solar activity. The CSD phenomenon is said to be
explained by microbiologists in terms of mutations.
The hypothesis of effects of Cosmic Rays (varying in opposition with
sunspot numbers) is sustained by series of experiments carried out in a
gallery under 1400 m of rocks, with clearly lower results than at sea level
[27, 28].
The studies of the "Piccardi-effect" in function of time, latitude, alti-
tude, screening conditions and with different controlled low-energy phys-
ical factors, used for treatment of water, have clearly contributed to
answer the question of the origin and nature of the disturbing factors
involved, including as well non-ionizing as ionizing radiation, in our near
and far environment.
Piccardi could already claim that with this "new conceptual paradigm,
we could solve the problems that have led the old one to a crisis" to use
the words of Kuhn. But still following Kuhn, "given a paradigm interpre-
tation of data is central to the enterprise that explores it". This is not an
easy task with the complex interrelated environmental factors, possibly
acting in the biosphere, as well as with their direct or indirect effects
on such complex systems as are even water itself, a fortiori, unstable
physico-chemical processes, and, moreover, living organisms.
When I met Piccardi for the first time in 1950, we were already both
convinced by our different experiments, carried out quite independently,
on different physico-chemical processes, that the low-energy factors stud-
ied acted on water itself.
The action could be of primary radiolysis of water (as for the indirect
action of ionizing radiation on biological systems) or just, (as Piccardi
suggested already in 1959) on the unstable and peculiar complex struc-
ture of water, to explain the effects of low-energy E.M. fields, specially in
the low frequency range. He developed this last hypothesis in his mono-
graph " The chemical basis of medical climatology", published in USA
in 1962 and translated in Russian in 1967 [18].
It can also explain the strange hysteresis effects, more or less retained
by water for hours of even several days after a physical treatment, as
114 C. CAPEL-BOUTE AND A. KO ECKELENBERG H
when water has been submitted to known structural changes by heating

or freezing.
Already in 1968, at our 2nd Brussels Symposium (unpublished) Pic-
cardi could conclude from extended research on fluctuating phenomena
that "many chemical systems and all biological systems, that are com-
plex heterogeneous systems far from equilibrium in aqueous medium,
shared not only the property of being from the thermodynamic point
of view open systems, exchanging matter and energy with the environ-
ment. But from the kinetic point of view they shared also the property
to react to external signals, even of very low-energy level, with results
fluctuating in the course of time."
Nevertheless, at last in 1972 (the year of Piccardi's death) B. GalOr,
as a physicist, in a discussion on the origin of irreversibility in Nature,
questions the concept and even the possibility of existence of a thermo-
dynamical isolated system.
A review of facts [9] shows that "water appears to be a most sensitive

medium to many physical factors radiated in our environment, from
the atmosphere and from space, from which it cannot be completely
isolated.With a low-energy input, as in any information process, they
can trigger important structural changes in water itself'. Water may
thus receive, transmit and even memorize or amplify, effects of the ever
changing physical conditions originating in the near and far environment.
All the research on the "Piccardi-effect" brought not only the foundation
of "chrono-chemistry", but an important contribution to the thesis of
Frank Brown Jr. confirming the action of external low energy physical
factors in chrono-biology.
It could even suggest that water could be an essential link to couple
cosmologic and electromagnetic asymmetries from the Universe with the
irreversible biologic evolution of Life on Earth.
Not only a new vision of the world is required from scientists to ac-
cept such concepts but new tools and structures are urgently needed in
fundamental and applied research as stated in the resolution presented
by our International Committee for Research and Study of Environmen-
tal Factors at the lOth Intern. Biometeor. Congress in Tokyo in 1984.
It stated: "as scientific workers from all disciplines concerned with the
problems of the complex relationships of living organisms with their near
and far environment, we stress for the urgent need to create interdisci-
plinary institutes for biosystemic research and long-term observations in
this field" .
AFFILIATION
C. Capel-Boute and A. Koeckelenbergh

Universite Libre de Bruxelles
Brussels, Belgium
REFERENCES
[1] Allen, C.W., Astrophysical Quantities, edit. Athlone Press, London,

1955.
[2] Becker, U., "Sonnenaktivitat und chemische test." In: Symposium
International sur les relations entre Phenomenes solaires et terrestres
en chimie physique et en biologie, Presses Academiques Europeennes,
Bruxelles, 1960, pp. 113-116.
[3] Capel-Boute, C., "Observations sur les tests chimiques de Pic-
cardi effectues a Bruxelles de 1950 a 1958." In: Symposium International
sur les Relations entre Phenomenes solaires et terrestres en chimie-
physique et en biologie, Presses academiques Europeennes, Bruxelles,
1960, pp. 51-109.
[4] Capel-Boute, C., "L'oeuvre scientifique de G. Piccardi", Medicina
Termale e Climatologia, 22, 1974, pp. 69-74.
[5] Capel-Boute, C., "New energy forms and the quality of life", Biosci.
Commun., 2, 1976, pp. 350-361.
[6] Capel-Boute, C., "Les effets d'ecran et de facteurs physiques am-
biants en chimie physique, dans les milieux aqueux", In: Les facteurs
d'Ambiance et les Climats interieurs, Cahiers de medecine du travail,
Bruxelles, XIV-2, 1977, pp. 213-225.
[7] Capel-Boute, C., "Assessment of origin and Effects from Natu-
ral Fluctuations and Man-Made Alterations of our Physical Environ-
ment", Biometeorology, 7, Part 2, Suppl. to Int. J. Biometeor., 24, 1980,
pp. 173-182.
[8] Capel-Boute, C., "Fluctuating phenomena in Physical Chemistry
and biology", Cycles, 34, 2, 1983, pp. 49-53.
[9] Capel-Boute, C., "Water as receptor of environmental informa-
tion", Biometeorology, 10, Part 2, Suppl. to Int. J. Biometeor., 29, 1985,
pp. 71-88.
[10] Delcourt J.J., Magnetisme terrestre, edit. Masson, Paris, 1990.
[11] De Meyer, F. and Capel-Boute, C., "Statistical analysis of
Piccardi chemical tests", Int. J. Biometeor., 31-4, 1987, pp. 301-322.
[12] GalOr, B., "The crisis about the origin of irreversibility and time
anisotropy", Science, 176,4030, 1972, pp. 11-17.
[13] Kuhn, T.S., The structure of scientific revolutions, The University
Chicago press, (Ist. Phoenix ed. 1964).
[14] Piccardi, G., "L'influence des phenomEmes terrestres, solaires et
cosmiques sur les reactions physico-chimiques et biologiques", Ciel et
Terre, 5-6, 1956, pp. 227-244.
[15] Piccardi, G. and Cini, R., "The action of an electromagnetic field
of 10 KHz frequency on the chemical tests: the problem of the influx of
"atmospherics"", Geofisica e meteorologia, 4, 1956, p. 25.
[16] Piccardi, G., "On the structure of water and the influx of low-
frequency electromagnetic fields", La Ricerca Scientifica, 29, 1959,
p. 1252.
[17] Piccardi, G., "Expose introductif- les tests chimiques", In: Sym-
posium International sur les Relations entre Phenomenes solaires et
terrestres en chimie-physique et biologie, Presses Academiques Europe-
ennes, Bruxelles, 1960, pp. 9-49 and pp. 121-130.
[18] Piccardi, G., The chemical Basis of Medical Climatology, Charles
C. Thomas, Springfield, 1962.
[19] Piccardi, G. and Capel-Boute, C., "The 22-year solar cycle and
chemical tests", J. Interdiscip. Cycle Res., 3, 1972, pp. 413-417.
[20] Pourbaix, M. and Rorive-Boute, C., " Procedes electrochim-
iques anti-incrustrants", In: Proceedings 2nd meeting of CITCE, Tam-
burini, Milano, 1951, pp. 240-241.
[21] Schoffeniels, E., "Generalizing information theory to an open sys-
tem", Biometeorology, 10, part 2. Suppl. to Int. J. Biometeor., 29, 1985,
pp. 41-56.
[22] Schnol, S.E. and Udaltsova, N.V., "On space-geo-physical corre-
lations in biological and physical chemistry processes", (3rd Intern. Sym-
posium in Pushchino- 26/9 -1/10/1993) CIFA-news, 15, 1994, pp. 2-6;
CIFA-news, 16, p. 6.
[23] Vladimirsky, B., Problemy kosmicheskoi biologii, vol. 5, Leningrad
Nauks, 1989, pp. 210-221.
[24] Vladimirsky, B., "Sector structure of interplanetary magnetic field
and F-tests of Piccardi", CIFA-news, 14, Febr. 1994 pp. 3-4.
[25] Vladimirsky, B. and Bruns, A.V., "Heliogeophysical distur-
bances influence upon the results of the measurements of gravitation
constant", CIFA-news, 19, May-June 1995, p. 2.
[26] Wedler, E., "Radiation in our environment from the atmosphere

and from space", Biometeorology, 10, part 2. Suppl. to Int. J. Biome-
teor., 29, 1985, pp. 59-70.
[27] "Geo-cosmic relations; the earth and its macro-environment", Fmc.
of Intern. Congress on Ceo-cosmic Relations, Amsterdam, 19-22 April
1989, Pudoc, Wageningen, 1990.
[28] "Exact, natural and human Sciences in the presence of uncontrolled
environmental factors", Fmc. of Intern. Symposium held in Brussels,
3 oct. 1992, Ed. Internat. Committee for Research and Study of Envi-
ronmental Factors (ICEF or CIFA) ULB (C.P. 196) 50 av. F.D. - 1050
Bruxelles, 1993.
See also for historical background, more recent developments and other
international interdisciplinary meetings the quarterly bulletins of CIFA
(ICEF) "CIFA-news" (Brussels) or "NOVOSTI-CIFA" ed. by Dr. Na-
talia Udaltsova, Secretary General of CIFA-Institute of theoretical and
experimental Biophysics of the Russian Acad. of Sciences, Pushchino
(Moscow Region) 142 292 Russia.
[29] Manzelli P., Masini G., Costa M., I segreti dell'acqua. L'opera
scientifica di G. Ficcardi, Ed. Di Renzo, Roma, 1994 (with full list of
publications by G. Piccardi established by M. Costa in the department
of Chemistry of the University of Florence, 9, Via Gino Capponi, 50121
Firenze.
LAZAREV, ROATH, YUNUSOVA & SAFONENKO
BIOMEDICAL APPLICATIONS OF MAGNETIC FLUIDS AS A

SPECIFIC INTERDISCIPLINARY PROBLEM
1. MAGNETIC FLUID: A SUBSTANCE
The best way to introduce the notion of magnetic fluids in terms of

readability and interest towards other than one's own discipline (as the
Editor requests) is to start with quotations from the keynote address by
Dr. R.E. Rosensweig, whose name is associated with the foundation of
the magnetic fluids research branch, to the Seventh International Con-
ference on Magnetic Fluids (1995, January):
"It is a fact that only solids that are strongly magnetic are found in nature, not
fluids. The strongly magnetic fluids( ... ) are colloids, most often produced by chemical
synthesis. The fluid consists of tiny particles of a magnetic solid stably dispersed in
a liquid carrier. ( ... ) The particles of magnetite are prevented from sticking to each
other by a molecular layer of soap or detergent that coats the surface. In this respect
the stabilization of a magnetic fluid has much in common with the technologies of
inks, paints and detergents, all of which rely on a molecular surface layer to suspend
tiny particles in one liquid to another" [1, p. 3].
"The particles in [magnetic fluids] are minute, only about one-hundred atoms across,
and each particle is like a little compass needle suspended in the liquid ( ... ). So when
a magnetic field is imposed, the little particles rotate into alignment with the field
and the fluid becomes magnetic overall. Unlike a compass needle each particle jiggles
around a lot due to the very strong Brownian motion. The motion is essential as it
keeps the particles suspended in the liquid instead of settling to the bottom" [1, p. 4].
"It is a curious fact that so far the availability of raw magnetite, soap and solvent
is concerned, magnetic fluids could have been synthesized hundreds of years ago but
in fact they have been in existence for only about thirty years" [1, p. 3].
2. MAGNETIC FLUIDS: AN INTERDISCIPLINARY

RESEARCH BRANCH: SPECULATIONS
The first inventors of magnetic fluids (MF) or, at least, MF precursors

were chemists [1, p. 3]. However, only in 1992-95 in the United States
special sessions devoted to MF "have appeared spontaneously in the
meetings of the .. . American Chemical Society ... , American Physical
Society ... and Materials Research Society... " [1, p. 2]. There is "a rich
specific heritage that underlines the topic of magnetic fluids, representing
as it does a unique confluence of the disciplines of electromagnetism and
119
120 LAZAREV, ROATH, YUNUS OVA & SAFONENKO
hydrodynamics concerning a complex fluid of interest to colloid science.

New combinations of old, unrelated subjects often excite the interest of
scientists from the separate disciplines and beyond" [1, p. 5].
The research of magnetic fluids displayed previously unknown phe-
nomena in the fields of hydrodynamics and crystallization [1, p. 7]. Var-
ious technological applications of MF are known in different fields of
science and technology (e.g., applications in bearings, seals, dampers,
devices for heat and mass transfer; in techniques of sink/float separa-
tion of minerals, in techniques of separation of gold from alluvial de-
posits, and grinding of ceramics [1, pp. 6-7]; the unique properties of MF
made it possible to use MF-based techniques also in computer devices,
aerospace and petroleum technologies [2] etc.). Moreover, just from the
formal viewpoint MF research may be considered as originally located
at the junction of physics, chemistry, mechanics and material sciences;
nowadays, as it follows from some of the above- listed applications, it
may be considered also as a part of engineering.
However, to note the interdisciplinary character of the research branch
in this way is just to observe and proclaim, but not to document. We may
document, of course, certain selected patents, articles, R&D reports etc.,
but what is the guarantee that our examples are representative and
unprejudiced? A good aid to support or disprove such observations as
those listed above is scientometrics.
3. MET H 0 D 0 L 0 G I CAL A P P R 0 A C H AND MET H 0 D S
Scientometrics is that branch of the Science of Science dealing with quan-

titative assessment of science indicators (like the number of researchers,
amounts of funding to support research activity etc., but mostly the
quantity of the scientific documents (articles, reports, books, patents
etc.) as the latter is a normal immediate formally assessable result of re-
search activity. Scientometrics does not deal so much with the content of
the indices under assessment. Our idea is that the "obviousness" of the
current conclusion of Chapter 2 ought to be supported by documented
data, obtained by "mere" scientometric calculations that-though often
looking superficial-are never open for prejudice. It goes without saying
that scientometric study is also -on a larger scale-used in the present
work for investigating the characteristics of biomedical applications of
magnetic fluids as an interdisciplinary problem. At the same time, anal-
ysis of the content of relevant publications is simultaneously used for the
same purpose. So, if we reformulate one of the basic questions put by
the present Conference as "Do we have to oppose the understanding of
science 'within object' (e.g., peer assessments, expert evaluation, prog-
APPLICATIONS OF MAGNETIC FLUIDS 121
noses made by the specialists of a domain under consideration) against a

form of the understanding of science 'beyond object' (e.g., scientometric
evaluation from outside a domain under consideration)?", there would
not be any need to oppose them to each other, both approaches being
applied together.
In scientometrics the level of interdisciplinarity is being assessed, e.g.,
by means of the disciplinary structure of the main research fulfilled by
employees of co-authors, by data on participation of researchers in "non-
profile" conferences, by disciplinary structure of cited references, etc.
[3]. Although we have some data on the subject structure of the main
research fulfilled by the employees of a representative number of MF
researchers and producers, the present study is based mostly on the
analysis of scientific documents and references cited in them.
A usual problem of such studies is an enormously large number of
documents to be involved to warrant their representative quantity. Since
not all kinds (species) of documents are equally significant, it is a usual
practice to include in such studies only selected kinds of them (e.g.,
papers published in the leading specialized journals or written by the
most cited authors from a domain). For the MF research branch that
hasn't any specialized journal, totally devoted to this subject (as well
as the branch itself has not as yet a nomination corresponding to the
existing scientific tradition and is being called in the same manner as
its subject), the best formal way to select the documents to be under
study is to use publications from the Proceedings of the International
Conferences on Magnetic Fluids (ICMF) [4]. Another reason for this
choice is that many MF researchers are involved in other research, too,
and a search for the documents according the authors names brings a
lot of publications not referring to MF at all [4].
Generally, proceedings of the international scientific conferences seem
to be rather a reliable source of documents under study: the so-called
"self-regulating mechanisms of science" are supposed to enable the selec-
tion of the international scientific events of potentially the most promi-
nent papers devoted to the most prospective trends of a research branch
development, while scientific conferences are usually "the first occasion of
communicating scientific findings" [5, p. 237]. In the above-cited keynote
address by Dr. Rosensweig the assessment of the increase of the domain
members' quantity was fulfilled by the count of the attendees of the
ICMF [1, p. 2].
In order to estimate the level of interdisciplinarity of MF research in
the periods, related to the terms of the Fifth and Sixth International
Conferences on MF (ICMF-5 and ICMF-6) being held (1989 and 1992),
122 LAZAREV, ROATH, YUNUSOVA & SAFONENKO
we studied the disciplinary structure of the references to journal articles,

cited in the corresponding Proceedings. (Citations to journal articles are
normally of about 70% of all the citations from any representative sam-
pling; and our case was regulated by the same rule: 74.24% of such
citations are in the ICMF-5 and 74.20% in the ICMF-6 Proceedings.)
Specialization of the cited journals was identified with the aid of the
special headings of the subject categories of the Science Citation Index
(which is one of the most usual ways of doing it [6], first applied to the
MF research branch study in [7]) and, if a cited journal was not reflected
there, its specialization was determined either according to the headings
of the Ulrich's International Periodical Directory or to common sense.
Such journals, non-identified with the aid of the Science Citation Index,
received 6.51% out of all the citations to journal articles in the ICMF-6
Proceedings, while for the ICMF-5 this portion is even smaller. It is im-
portant to note that, according to the Science Citation Index headings,
some journals are related to more than one discipline; in these cases the
journal citations were calculated so many times, as many disciplines a
journal was related to.
This method was applied to the articles in both Proceedings, specially
devoted to biomedical applications of MF. However, since the latter is
the main subject of the present research, we also analyzed the content, as
well as subject and disciplinary structure of the publications on biomed-
ical applications of MF themselves-but not in the format of the articles
in the Proceedings, however, but in a format of abstracts in the cor-
responding abstracts compendiums. The latter was chosen as a more
representative sampling for studying the characteristics of productivity:
not all the abstract authors are asked for submission of a full paper
to the Proceedings. On the other hand, references in the abstracts are
small in number, optional and occasional; therefore not representative
for scientometric research.
4. M A G N E T I C F L U I D S : A N I N T E R D I S C I P L I N A R Y
RESEARCH BRANCH:
DOCUMENTED SCIENTOMETRIC DATA
The disciplinary structure of the mentioned references in the whole Pro-

ceedings is presented in Table 1. The number of disciplines, presented by
the cited journals, has increased from 30 (which is, by itself, an unusu-
ally large number of disciplines, cited by such a compact and relatively
"young" domain) up to 44 (the increase is 46.90%). It means that, first,
MF research is really a highly interdisciplinary one. Further, such an
increase is normally interpreted as evidence of the increased use of the
knowledge generated in disciplines other than the research branch under

study [6 etc.], while an increased use of such "interdisciplinary informa-
tion" is believed to be a very strong determinant of the scientific value
of results being obtained in the research branch under study [8 etc.].
In any case, the expansion of the thematic scope of the cited scientific
literature is believed to be an apparent sign of the progress of scientific
development (this viewpoint is reviewed in [9]). Third, the main re-
flected disciplines and disciplinary groups documented by scientometric
investigation are really physics, chemistry, mechanics, material sciences
and engineering. Fourth, we are pleased to notice that such disciplines,
mentioned in Chapter 2, as crystallography, instruments and instrumen-
tation, aerospace engineering and technology, computer applications and
cybernetics as well as metallurgy and mining, are featured in Table 1;
that means that an ordinary scientometrician with the least knowledge
of what MF research is, can document the links of this branch with the
disciplines, discussed in Chapter 2.
Moreover, the list of Table 1 is more complete than the one discussed in
Chapter 2. Certainly, we are aware about most of the liaisons, mentioned
in Table 1, but did not mention them in Chapter 2 because our memory
failed ... Scientometrics did not. The two approaches of characterizing a
research branch- understanding of science 'within object' and a form
of the understanding of science 'beyond object' really ought to be applied
together.
Fifth-and now we are approaching at last to the heart of the matter-
there is an enormous increase not only of the number of citations to bi-
ological and medical journals, but also-of their disciplinary structures.
5. B I 0 M ED I C A L A P P L I C AT I 0 N S 0 F M A G NETIC FLUID S
A SUBSTANTIAL DIVISION OF
THE RESEARCH BRANCH
Table 2 plots the data on the relative increase or decrease of the portion
of citations to the journals in various disciplinary groups in the ICMF-
6 Proceedings as compared with the Proceedings of the ICMF-5. By
relative increase or decrease we mean the alteration not in number of ci-
tations but in their portion expressed in per cent. Besides 357.45 per cent
increase of citations to biological and medical journals, 65.90% of all the
disciplines cited in the ICMF-6 Proceedings are biological and medical
disciplines, while the part of medical and biological publications in the
Proceedings is but 12.87%. In the ICMF-6 Proceedings citations to bio-
logical and medical journals have the 4th rank after physics, mechanics,
engineering and there are more citations to medical and biological jour-
124 LA ZAREV, ROATH, YUN US OVA & S AF ON EN KO
nals than to that of material sciences and chemistry. Why?-We think

that the reason for such strongly increased information links of MF re-
search with biology and medicine may be the increase of development of
applications of MF-based techniques in this area. The supposed increase
of development of a certain research direction means its important role
in the structure of a research branch.
TABLE I
The disciplinary structure of the bibliographic citations to the journals in the

Proceedings of the ICMF-5 and ICMF-6
The disciplines presented The quantity of bibliographic

by the cited journals citations to the items repre-
Disciplinary groups senting these disciplines in
the Proceedings of
ICMF-5 ICMF-6
a % T a % Tl
Mechanics 122 26.49 1 83 14.09 2
("Physics" group including:} 182 29.49 190 32.26
Physics 81 17.57 2 107 18.17 1
Physics, Applied 64 13.88 3 47 7.98 4
Physics, Fluids and Plasmas 13 2.82 7 1 0.17 34
Physics, Condensed Matter 12 2.60 9 13 2.21 11
Physics, Atomic, Molecular
and Chemical 11 2.38 10 18 3.06 8
Physics, Mathematical 1 0.22 23 4 0.70 20
( "Mater·ial Sciences"
group including:} 57 12.36 74 12.56
Material Sciences 57 12.36 4 73 12.56 3
Material Sciences, Ceramics 1 0.17 34
("Chemistry" group including:} 40 8.68 45 7.64
Chemistry, Physical 30 6.51 6 35 5.94 6
Chemistry 4 0.87 13 7 1.19 14
Chemistry, Applied 3 0.65 14
Chemistry, Inorganic and
Nuclear 1 0.22 23 2 0.34 26
Chemistry, Analytical 1 0.17 43
1
a is an absolute figure of the magnitude of the index; % is its percent expression;
r is the rank of the index magnitude. Blank means the absence of citations to the
disciplinary journals in one of the studied Proceedings. The same symbols are used
in the rest of the table
AP PLICA TIO NS 0 F MAGNETIC FLUIDS 125
["Engineering" group including:} 20 4.35 79 13.41

Engineering, Mechanical 13 2.82 7 35 5.94 6
Engineering, Chemical 5 1.05 11 7 1.19 14
Engineering 1 0.22 23
Engineering, Electrical and
Electronic 1 0.22 23 41 6.96 5
("Biology and Medical Sciences"
group including:} 13 2.82 76 12.90
Biophysics 5 1.05 11 10 1.70 13
Cardiovascular System 2 0.43 17 6 1.02 16
Hematology 2 0.43 17 4 0.70 20
Immunology 2 0.43 17 15 2.55 9
Gastroenterology 1 0.22 23
Medicine, General
and Internal 1 0.22 23
Biochemistry and Molecular
Biology 2 0.34 26
Radiology and Nuclear
Medicine 14 2.38 10
Cytology and Histology 5 0.85 18
Oncology 4 0.70 20
Medicine, Research and
Experimental 4 0.70 20
Biology 3 0.51 24
Pharmacology and Pharmacy 2 0.34 26
Biology, Miscellaneous 2 0.34 26
Microbiology 1 0.17 34
Pediatrics 1 0.17 34
Physiology 1 0.17 34
Surgery 1 0.17 34
Mnltidisciplinary jonrnals 14 3.04 6 12 2.04 12
Acoustics 3 0.65 14
Crystallography 3 0.65 14 3 0.51 24
Mathematics 2 0.43 17 2 0.34 26
Mathematics, Applied 1 0.17 34
Instruments and
Instrumentation 2 0.43 17
Aerospace Engineering and
Technology 1 0.22 23 2 0.34 26
Optics 1 0.22 23 6 1.02 16
Technology: Comprehensive
126 LA ZAREV, ROATH, YUN USOV A & SAFONENKO
Work 2 1 0.22 23
Polymer Sciences 5 0.85 18
Energy and Fuels 2 0.34 26
Geosciences 2 0.34 26
Computer Applications and
Cybernetics 1 0.17 34
Metallurgy and Mining 1 0.17 34
Meteorology and Atmosphere
Sciences 1 0.17 34
Photographic Technology 1 0.17 34
I Total dzsczplznary cztatwns 461 589
TABLE II
The disciplinary distribution of the bibliographic citations to the journals in the
Proceedings of the ICMF-5 and ICMF-6 (only highly cited disciplines and disci-
plinary groups)
The discipline or dis- The quantity of bibliographic Relative
ciplinary group citations to the items repre- increase
senting these disciplines in (+)/decrease (-)
the Proceeding::; of in the I CMF-6
ICMF-5 ICMF-6 Proceedings
a % r a % r %
"Physics" group 189 39.48 1 190 32.26 1 - 18.29
Mechanics discipline 122 26.46 2 83 14.09 2 - 46.75
"Material Sciences"
group 57 12.36 3 74 12.56 5 + 1.62
"Chemistry" group 40 8.68 4 45 7.64 6 - 11.98
"Engineering" group 20 4.83 5 79 13.41 3 +177.63
"Biology and Medical
Sciences" group 13 2.82 6 76 12.90 4 +357.45
6. B I 0 MEDIC A L A P P L I C AT I 0 N S 0 F MAGNET I C FLU IDS

AN INTERDISCIPLINARY RESEARCH PROBLEM:
DOCUMENTED SCIENTOMETRIC DATA
The conclusion formulated in the title of the present chapter seems to be

obvious and is apparently derivable from the above Table 1. However,
2
The only discipline name that was taken from the Ulrich's Periodical International
Directory since a directly corresponding name was not found in the Science Citation
Index
A P PLICA T I 0 N S 0 F MAGNETIC FLUIDS 127
TABLE III
The disciplinary structure of the bibliographic citations to the journals in
the articles on biomedical applications published in the ICMF-6 Proceedings
The disciplines presented The quantity of bibliographic
by the cited journals citations to the items repre-
senting these disciplines in
the Proceedings of the ICMF-6
a % r
Biochemistry and Molecular Biology 14 14.14 1
Immunology 14 14.14 1
Biophysics 10 10.10 3
Material Sciences 9 9.09 4
Chemistry 5 5.05 5
Hematology 4 4.04 6
Oncology 4 4.04 6
Physic, Applied 4 4.04 6
Radiology and Nuclear Medicine 4 4.04 6
Cytology and Histology 4 4.04 6
Medicine, General and Internal 3 3.03 11
Multidisciplinary journals 3 3.03 11
Chemistry, Inorganic and Nuclear 2 2.02 13
Chemistry, Physical 2 2.02 13
Engineering, Electrical and Electronic 2 2.02 13
Biology 1 1.01 16
Biology, Miscellaneous 1 1.01 16
Cardiovascular System 1 1.01 16
Chemistry, Analytical 1 1.01 16
Chemistry, Physical 1 1.01 16
Crystallography 1 1.01 16
Engineering, Chemical 1 1.01 16
Medicine, Research and Experimental 1 1.01 16
Microbiology 1 1.01 16
Pediatrics 1 1.01 16
Pharmacology and Pharmacy 1 1.01 16
Physics 1 1.01 16
Physics, Condensed Matter 1 1.01 16
Physiology 1 1.01 16
Surgery 1 1.01 16
Table 1 plots the disciplinary structure of references in all the Proceed-

ings articles, not in the ones specially devoted to biomedical applications.
If we pass to the latter, we should see that 29 disciplines are presented by
the periodicals cited in the papers on biomedical MF applications pub-
lished in the ICMF-6 Proceedings (Table 3). The increase of the number
of disciplines cited in the articles on biomedical MF applications pub-
lished in the ICMF -6 Proceedings as compared with the ones published
in the Proceedings of the ICMF-5 is 222.22% (from 9 to 29 disciplines),
128 LAZARE V, R 0 AT H, Y UN US 0 VA & SA F 0 N EN K 0
the number of cited biomedical disciplines increased from 6 to 17, and

the number of the non-medical ones-from 3 to 12 disciplines. Bearing
in mind that the expansion of the thematic scope of the cited scientific
literature is believed to be a sign of the progress of scientific develop-
ment, we dare to suppose that biomedical applications of magnetic fluids
is not only a highly interdisciplinary research problem characterized by
the recent growth of interdisciplinary level, but also a recently progress-
ing research sub-field. Cf. with the analogous data for the disciplinary
structure of citations in the Proceedings on the Second and the Third
Symposia on Bone Marrow Purging and Processing (having been held
in the same years) which is another "young" research branch that has a
substantial intersection zone with the one under study (MF based tech-
niques are used for bone marrow purging): 32 vs. 24 cited disciplines,
i.e., an obvious decrease. Other indirect evidences of the recent progress
of the research problem are featured in Chapter 7; the structure of cited
disciplines is to be compared with applicability of some selected research
works to various medical and biological objectives in the following chap-
ters.
7. BI 0 MEDICAL AP PLICA TIO NS 0 F MAGNETIC FLUIDS

A PROGRESSING AREA OF RESEARCH?
(FURTHER SCIENTOMETRIC DATA)
The title hypothesis is also supported by the productivity data of the

abstracts from ICMF-5 and ICMF-6 Conferences: the abstracts com-
pendium of the ICMF-5 contains 11 works on MF biomedical applica-
tions (7.98% of the total number of abstracts), in the ICMF-6 com-
pendium there are 32 abstracts (11.43%). The absolute increase is 190.-
91%, the relative increase is 43.23%. Numbers of full papers published
in the Proceedings are correspondingly 5 for the ICMF-5 (7.35% of the
total amount of publications) and 13 (12.87%) for the ICMF-6, the abso-
lute increase is 160.00%, relative increase being 75.10%; i.e., more than
the relative increase of the number of all the papers published in the
Proceedings of the ICMF-6 (48.53%). The portions of the biomedical
papers from the biomedical abstracts are correspondingly 45.45% for
the ICMF-5 and 40.62% for the ICMF-6, while the full papers/abstract
part for the whole Proceedings are 49.27 and 36.07%. It means that
MF biomedical applications papers' authors succeeded more in terms of
publication productivity in the ICMF-6 Proceedings than the domain
in general. Also, there is a substantial increase in the numbers of insti-
tutions and countries whose representatives happily published their ab-
stracts and full papers: thus, if the representatives of only 5 countries and
AP PLICA TIO NS OF MAGNETIC FLUIDS 129
13 institutions published their abstracts in the ICMF-5 compendium, in

the compendium of the ICMF-6 the representatives from 11 countries
and 34 institutions published their works; the corresponding data for
the Proceedings being 3 countries I 7 institutions for the ICMF-5 and
8 countries I 20 institutions for the ICMF-6. The above data seems to
support our hypothesis as reflecting the increase of productivity of the
subdomain in all the possible terms and the appearance of newly in-
volved institutions (or at least of institutions whose researchers become
interested in participating in the top MF international forum).
8. S U B J E C T S T R U C T U R E A N D D I S C I P L I N A R Y
STRUCTURE OF THE RESEARCH
OF BIOMEDICAL APPLICATIONS
The obtained data on the disciplinary structure of references in the pa-

pers devoted to MF biomedical applications support the opinion that the
research problem is both a highly interdisciplinary and progressive one
(the latter being supported also by the data of the previous Chapter). It
is also of interest to obtain a formali:6ed notion of the thematic structure
of the research themselves as reflected in the content of the ICMF-5 and
ICMF-6 abstracts (to support or disprove our opinion). So as not to miss
any possible trend, we read thoroughly all the corresponding abstracts,
trying to classify first the subjects of their research.
The results of this attempt are presented in Table 4, the latter demon-
strating that three main subjects still exist, viz. cell separation and drug
transportation systems, and MF-based imaging techniques. As for two of
them, viz. cell separation and imaging techniques, there is an apparent
tendency to an increase, whereas for the use of magnetic fluids as drug
carriers the increase is a minor one. Little attention seems to be paid to
the studies of the interactions of MF with biological structures in gen-
eral. A visible expansion of the S'ubject structure of the studies is obvious.
This also looks to be a sign of a research progress.
Another way of presenting the thematic structure of the research is to
determine disciplines under which framework research may be classified
(as it was done with the cited references). For the content of the abstracts
such an attempt was also made (Table 5) since it is of an interest to
compare the disciplinary structure of published and cited works. The
most prominent finding plotted in Table 5 is the increase of research in
such vital areas as oncology and immunology.
130 LAZARE V, R 0 AT H, Y UN US 0 VA & SA F 0 N EN K 0
9. RETURNING T 0 SPECULATIONS:
POPULAR REVIEW OF SOME RECENT WORKS ON
MAGNETIC FLUIDS BASED
DRUGS TRANSPORTATION SYSTEMS
Drugs transportation is being undertaken by means of an applied mag-

netic field to a MF drug carrier. Though the subdivision of the research
is both important and promising, there is not such a great absolute in-
crease in the number of abstracts ( +40.00% ), while in relative terms the
decrease is -58.25%.
The mentioned decrease does not seem to be occasional since "such
drug transporting system must have definite, very often controversial
( ... ) characteristics:
- Particles in magnetic fluid compositions stabilized by biocompatible

substances must have appropriate shape and size ( ... ) in order to
pass through capillary systems of organs and tissues without posing
the threat of vessel embolisation.
- Compositions containing both [magnetic] fluid and a chemothera-

peutic agent must have sufficiently high magnetization [to be cap-
tured in a target site at the physiological velocities of blood flow].
- Magnetically guided carrier systems must be able to transport the

necessary amount of pharmacologically active compounds, and there
must be an effective mechanism for the release of drug from the
carrier at the target site.
All components of the drug carrier complex must be non-toxic, fi-

nally biodegradable and removable from the reticuloendothelial sys-
tem" [10, p. 335].
Besides these problems, it seems that just not very much attention is
paid to the problem of the studies of interactions of MF with "biological
structures" (i.e., "blood components, membranes of cells and subcellu-
lar structures" [11, p. 266]). However, "investigations in this field are
being done in oncology ( ... ) , in the treatment of diseases of heart and
blood vessels ( ... ) and hollow organs" [10, p. 335]. There are reports of
successful experimental treatment of some malignancies with the aid of
water-based MF drug carriers [12] and of a number of malignancies with
the aid of ferromagnetic microspheres that immobilize antitumor drugs
and consist of microcrystalline ferric oxide and a polymeric matrix [13],
TABLE IV
The subject structure of the abstracts on biomedical applications published in the
ICMF-5 and ICMF-6 Proceedings
Subject Number of the abstracts in the abstract
(purpose or compendium
sphere of the ICMF-5 of the ICMF-6
application
or an objective) original reviews total original rev1ews total
cell separation" 2 0 2 7 0 7
drug carriers 5 0 5 6 1 7
image techniques 1 0 1 6 1 7
biomedical research
in general 1 0 1
alignment of biome-
dical assemblies 0 1 1
experimental medi-
cine and oncology
in general 0 1 1
development of
immunological
compositions 1 0 1
fabrication of
carbon 1 0
hyperthermia of
cancer 1 0 1
immobilization of
cytochrome-c-oxidase 1 0 1
obturation of
digestive tracts
external fistulas 1 0 1
obturation of
hollow organ
fistulas 1 0 1
smear examination 1 0 1
studies of bio-
transformation of
magnetic powders 1 0 1
studies of MF
interactions with
biological struc-
tures 1 0 1
antifiammatory
effects 1 0 1
treatment of
plants 1 0 1 1 0 1
" Italicized are the titles of the mam general subjects
132 LA Z ARE V , R 0 AT H , Y UN US 0 VA & S A F 0 N EN K 0
TABLE V
The disciplinary structure of the abstracts on biomedical applications pub-
lished in the ICMF-5 and ICMF-6 Proceedings
Scientific discipline Number of the abstracts in the abstract
compendium
of the ICMF-5 of the ICMF-6
a % a %
Biophysics 2 15.38
Hematology 2 15.38 3 8.57
Medicine, General and Internal 2 15.38 3 8.57
Biology 1 7.69
Oncology 1 7.69 4 11.34
Cardiovascular System 1 7.69 1 2.86
Cytology 1 7.69
Genetics 1 7.69 1 2.86
Immunology 1 7.69 8 22.86
Surgery 1 7.69 2 5.71
on the development of a ferromagnetic antitumor dextran-ferrite [13]

and on the MF obturation of hollow organ fistulas [14], on the technical
possibility to use MF as the carrier for drug transportation into an oc-
cluded vessel [15]. Hence, such scientometrically defined disciplines such
as oncology, surgery and cardiovascular systems are again included in
the list~but this time by means of the analysis of the context of the
relevant publications.
10. AGAIN REVISITING SPECULATIONS:

POPULAR REVIEW OF SOME RECENT WORKS ON
MAGNETIC FLUIDS BASED
CELL SEPARATION SYSTEMS
MF-based and related techniques are good for cell separation systems
used, e.g., for human lymphocyte subset removal, tumor cell removal,
selection of self-replicating bone marrow cells, hybridoma selection, sep-
aration of endothelial cells [16].
The most acute challenge in the context of the listed applications
is bone marrow transplantation which is an aid in the management of
leukemia, cancer, and lethal radiation exposure. Human lymphocyte sub-
set removal and the selection of self-replicating bone marrow cells are
among the possible problems to be resolved during the procedure of bone
marrow transplantation:
Removal of human lymphocyte subsets using a suitable magnetic bead
labelled with an antibody to the targeted lymphocyte and being per-
formed by methods of magnetic filtration, is still a good biological com-
promise in allogenic bone marrow transplantation (BMT) where graft

versus host disease is to be controlled in this way and yet the graft ver-
sus leukemia effect to be retained. As for autologous BMT, where there
may be a residuum of, e.g., leukemic cells or neuroblastoma, or, e.g.,
breast tumor, in the sample collected, purging of marrow samples is ac-
cepted as part of the current approach to tumor treatment. For many
tumors there are specific antibodies and, using similar techniques to
those for removing lymphocytes subsets, these unwanted and presum-
ably life threatening residual malignant cells can be removed, e.g., by
ferromagnetic beads or simple filtration systems although refinement of
this technology is still awaited.
Separation technology using several magnetic based approaches in this
field is directed towards the purification of stem cells from blood or bone
marrow. Currently it is possible to regenerate human bone marrow using
stem cells only, when they are obtained in this way. Techniques include
the use of monosized composite polymer particles (magnetic beads) and
ferrocolloids [16].
The promising feature of the mentioned and unmentioned applications
is the existence of some commercial systems developed for cell sepa-
ration. E.g., in [17] experiments are described on "selective removal of
breast cancer cells from artificial mixtures containing fresh, healthy bone
marrow cells and breast cancer cell line-derived cells" [17, p. 488] with
the help of binding the lectin soybean agglutinin, mixed with polystyrene
magnetic beads "Dynabeads".
The authors of [18] produced colloidal magnetite-binding erythrocytes
considered for a cell processing scheme as being selective in the pro-
cess of binding. Also, a technique is being developed for immobilization
of polysaccharide-mediated colloidal particles onto human platelets en-
abling the potential use of the latter as a blood-component-based mag-
netic label for different cell fractionation schemes [19].
The brief outlook to the recent conference papers gives an impression
that the publication increase in this area (250.00% absolute and 20.35%
relative increase) is not just occasional and reflects the real state of
affairs regarding the reviewed direction of the research. An enthusiasm of
researchers, which is a good prerequisite for a success, is also obvious. It
should be emphasized that such disciplines as immunology, surgery and
radiation and nuclear medicine are again in the list of related disciplines
as the result of the above analysis.
134 LA Z ARE V, R 0 AT H , Y UN US 0 VA & SA F 0 N EN K 0
11. FURTHER SPECULATIONS: POPULAR REVIEW OF

SOME MAGNETIC FLUIDS BASED IMAGE TECHNIQUES
As one of us reported, "an area where magnetic and paramagnetic fluids

have been developed and established is that of the enhancement of mag-
netic resonance imaging. Briefly, both signal enhancement and improved
tissue contrast can be achieved by using paramagnetic, superparamag-
netic or ferro magnetic particles. Clearly, the needs for such a system
include:
1. non-toxic particles,
2. particles giving maximum enhancement,
3. delivery of the particles to the appropriate site.
Specific receptors and linkages of the enhancing particles to those recep-
tors by, e.g., antibodies or naturally targeting entities need more devel-
opment. However, imaging of experimental myocardial infarction ( ... )
and the gastrointestinal tract ( ... ) have shown that the contrast manip-
ulation may be highly effective" [16, p. 530]. Imaging for experimental
inflammation should be mentioned in the same way [20].
Second, Roentgen contrast diagnostics should be mentioned. A recent
example is an antitumor roentgenocontrast magnetically controlled fer-
romagnetic fluids described by Brusentsov et al [21] that contain 20-60
weight percent of dextran-ferrite, Dactomycin, and 40-80 weight percent
of verografin or conrey.
The increase of the number of abstracts devoted to image techniques
(+600.00% absolute and +140.70% relative) could hardly, however, be
taken into account seriously, because only one paper was presented at
the ICMF-5. A number of biological and medical disciplines where MF-
based image techniques are applicable, is, in any case, very impressive
(cardiovascular systems, internal medicine, oncology and a lot of others
that are not exemplified here). The research seems to be progressing,
too.
12. STILL M 0 R E SPEC U LA T I 0 N S :

SOME OTHER APPLICATIONS
The description of the studies of the rest-mostly specific than general-

of the applications is contained in some 1/3 of the ICMF- 5 and some
36% of the ICMF-6 abstracts. There are no small groups as we see, but
the distribution of the subjects-one abstract per two conferences except
the "treatment of plants" (one abstract per one conference )-does not
allow us to make any distinctly generalized grouping. Neither does it
inspire us to produce a review of the rest of the applications.
A P PLICA T I 0 N S 0 F M A G NET I C F L U I D S 135
At the same time, a new leading sphere of MF biological applications,

viz. agriculture, demonstrated itself at the ICMF-7 in January, 1995.
However: the Seventh Conference on scientometric analysis is beyond
the objectives of the present study.
13. CONCLUSIONS
Magnetic fluids is a relatively "young" research branch that is termed

in the same manner as its subject. It has no specialized journal totally
devoted to the subject, neither an officially registered scientific society
to unite the domain. In spite of these signs of an "insufficient maturity"
this research branch looks very much mature in terms of interdisciplinar-
ity -both as consuming the information from other disciplines and as
providing various fields of science and technology with the products of its
own R&D.
Biomedical applications of MF is a highly interdisciplinary problem.
The most prominent features are: the growths of the number of disci-
plines being linked with it; a great number of the medical and biological
disciplines, where the future practical applications of the R&D results
are very promising; a sufficient increase of the links with non-medical dis-
ciplines; a stable character of the main spheres of applications (founded
on the extension of the particular applications); practical applicability of
MF-based techniques in cell separation procedures; growth of the num-
ber of researchers being inspired by the targets of this research problem.
The present work is based on the combination of the two approaches:
"within object" (i.e., the analysis of the content of the problem by analyt-
ically reviewing the relevant literature) and "beyond object" (a sciento-
metric-i.e., formal and quantitative-evaluation of science indicators).
The study was fulfilled by the team of two scientometricians and two
experts in magnetic fluids (a physician researcher and a chemist). Such
an approach seems to be promising for bridging the gaps between the
formal way of assessment of a scientific discipline development (that may
be superficial and biased, but never prejudiced) and the assessment of
the latter by the representatives of the same discipline (peer evaluation
that is never superficial, seldom biased, but often prejudiced).
AFFILIATION
V.S. Lazarev
Department of Scientific Information and Patentology
Belarusian State Polytechnic Academy
Minsk, Belarus
Stuart Roath
Cancer Research Division, Walt Disney Memorial Cancer Institute
Altamonte Springs, FL, USA
O.K. Safonenko
Laboratory for Thermomechanics of Magnetic Fluids
Belarusian Polytechnic Academy
Minsk, Belarus
D.A. Yunusm~a
Research Institute of Hematology and Blood Transfusion
Minsk, Belarus
REFERENCES
[1] Rosensweig, R.E., Scientific, technological and cultural aspects of

magnetic fluids: Keynote address to the Seventh International Confer-
ence on Magnetic Fluids, Bhavnagar, Jan 9-14, 1995, Department of
Physics, Bhavnagar University, Bhavnagar, s.a.
[2] Raj, K., Moskowitz, B., Lanphear, S., "Advances in ferrofluid
technology", in: Abstracts of the Seventh International Conference on
lt1agnetic Fluids, Bhavnagar, India, 9-4 Jan., 1995, Bhavnagar, s.a.,
p. 18.
[3] Kozlyakovskiy, P.A., Research of the specialists needs in inter-
disciplinary information, Nauchnyie i tekhnicheskiye biblioteki SSSR, N
6, 22-27, 1977 (in Russian), Komlev, A.D., Inter-branch interaction
and its information-and-bibliographic support, PhD Thesis, Leningrad,
1981 (in Russian), Neeley-Jr, J.D., "The management and social sci-
ence literature: an interdisciplinary cross-citation analysis", Journal of
the American Society for Information Science, 32, 1981, pp. 217-223,
Porter, A.L., Chubin, D.E., "An indicator of cross-disciplinary re-
search", Scientometrics, 8, 1985, pp. 161-176. Ajiferuke, 1., Burell,
Q., Tague, J ., "Collaborative coefficient: a single measure of the degree
of collaboration in research", Scientometrics, 14, 1988, pp. 421-433.
[4] Lazarev, V.S., "When Science Citation Index is of no use ... ", in:
Tyutyunnik, V .M. (ed.), Information Science and Science of Science:
Proceedings of the 3rd International (5th Tambov) Scientific Conference,
Tarnbov, 28-29 November, 1994, Tambov, INIC Publishers, 1994, pp. 43-
44.
[5] Martens, B., Saretski, T., "Conferences and courses on biotechnol-
ogy. Describing scientific communication by exploratory method", Sci-

entometrics, 27, 1993, pp. 237-260.
[6] Lazarev, V.S.," Studies of documentary information flows by means
of citation analysis", Voprosy bibliografovedeniya i bibliotekovedeniya, 7,
1986, pp. 114-125, 1986 (in Russian).
[7] Lazarev, V.S., "Bibliometric studies of the proceedings of the Fifth
International Conference on Magnetic Fluids. Structure of bibliographic
references as an indicator of knowledge foundation of the field of science",
Magnitnaya Gidrodinamika, N 1, 1993, pp. 109-116 (in Russian).
[8] Lazarcv, V.S., "Bibliographic references as the aid of assessment of
scientific value of cited papers", in: Eve-Marie Lacroix, Local Orga-
nizing Committee of the 7th International Congress on Medical
Librarianship, ed., Health information for global village: Proceedings of
the 'lth International Congress on Medical Librarianship, Washington,
D.C., May 10-12, 1995, Washington, 1995, pp. 276-277.
[9] Karas, G.A., Lazarev, V.S., "The use of some scientometric meth-
ods for rapid assessment of scientific activity of scientific research institu-
tions", Vestnik Akademiyi meditsinkikh nauk SSSR, N 4, 1991, pp. 61-65
(in Russian).
[10] Ruuge, E.K., Rusetski, A.N., "Magnetic fluids as drug carriers:
Targeted transport of drugs by a magnetic field", Journal of Magnetism
and Magnetic Materials, 122 (1-3: Proceedings of the 6th International
Conference on Magnetic Fluids), 1993, pp. 335-339.
[11] Ahalaya, M.G., Kakiashvili, M.S., Volter, E.P., "Investiga-
tion of the mechanisms of magnetic fluids interactions with biological
structures", in: Fifth International Conference on Magnetic Fluids. Riga,
18-22 September, 1989: Abstracts, Salaspils, 1989, pp. 266-267.
[12] Brusentsov, N.A., Lukashevich, M.V., Lykov, V.V., Knorin,
E.A., Pechennikov, A.V., Chechernikov, V.I., Martynova, L.F.,
"The magnetic fluid for medicine application", in:Fifth International
Conference on Magnetic Fluids. Riga, 18-22 September, 1989: Abstracts,
Salaspils, 1989, pp. 258-259.
[13] Syrkin, A.B., Brusentsov, N.A., "Ferri- and ferromagnetics in
experimental oncology and medicine", in: Sixth International Conference
on Magnetic Fluids. July 20-24, 1992: Programme & Abstracts, S.l., s.a.,
pp. 480-481.
[14] Rusakova, N.N., Nikolaenkov, Y.V., Nechaeva, A.B., "Mod-
eling of magnetic fluid effect on the magnetofluid obturation of hollow
organ fistulae", in: Sixth International Conference on Magnetic Fluids.
July 20-24, 1992: Programme & Abstracts, S.l., s.a., pp. 492-493.
[15] Rusetski, A.N., Ruuge, E.K., "Magnetic fluids as possible drug

carrier for thrombosis treatment", in:Fifth International Conference on
Magnetic Fluids. Riga, 18-22 September, 1989: Abstracts, Salaspils, 1989,
pp. 256-257.
[16] Roath, S., "Biological and biomedical aspects of magnetic fluid
technology", Journal of Magnetism and Magnetic Materials, 122, (1-3:
Proceedings of the 6th International Conference on Magnetic Fluids),
1993, pp. 329-334.
[17] Morecki, S., Slavin, S., "Combination of magnetic and chemo-
cytotoxic cancer cell depletion for autologous bone marrow transplanta-
tion", Israel Journal of Medical Sciences, 24, 1988, pp. 488-493.
[18] Plavins, J ., Lauva, M., "Study of colloidal magnetic-binding
erythrocytes: Prospects for cell separation", Journal of Magnetism and
Magnetic Materials, 122, (1-3: Proceedings of the 6th International Con-
ference on Magnetic Fluids), 1993, pp. 349-353.
[19] Plavins, J ., Harris, J ., Roath, S., "Electron microscopy and
magnetic study of colloidal magnetic bound human platelets", in: Ab-
stracts of the Seventh International Conference on Magnetic Fluids,
Bhavnagar, India, 9-14 Jan., 1995, Bhavnagar, s.a., pp. 247- 248.
[20] Brady, T.J., "MR receptor and antibody imaging using monocrys-
talline iron oxide nanoparticles", in: Sixth International Conference on
Magnetic Fluids. July 20-24, 1992: Programme fj Abstracts, S.l., s.a.,
p. 180.
[21] Brusentsov, N.A., Jurchenko, N.Y., Jarovskaya, N.P., Zon-
tov, S.V., Syrkin, A.B., "Antitumor roentgenocontrast magnetically
controlled fluids", in: Abstracts of the Seventh International Conference
on Magnetic Fluids, Bhavnagar, India, 9-14 Jan., 1995, Bhavnagar, s.a.,
pp. 239-240.
GUSTAAF C. CORNELIS
COSMOLOGY AND PROLIFERATION
INTRODUCTION
Proliferation
This talk concerns the diffusion of hypothetical knowledge. Hence, the

concept of proliferation is defined as uncontrolled growth: too many dif-
ferent theories emerge. The other interpretation, namely as sociological
dispersion or circulation of knowledge through different types of media,
is not of any relevance to us now. The following picture shows what the
concept 'proliferation' stands for.
different theories
,----,,---------------~
time
The originally thick line represents a standard theory, which becomes

gradually more and more contestable, because alternatives (thin lines)
emerge almost continuous. There's more going on here than just a Kuh-
nian crisis: quantitatively and qualitatively as well, the situation can be
hardly overseen, due to the-at first sight-completely irrational way in
which alternatives show up and are elaborated.
Non-Specialist Literature
In the non-specialist literature, in general, cosmology is regarded as

an utmost speculative science. If this opinion is not shared in certain
books-mainly the case, when the author is a cosmologist him- or her-
self-the publication itself makes the reader's head spin: he or she will
139
140 GUSTAAF C. CORNELIS
probably deem cosmology as pure speculation. Anyway, many among

the cosmologists themselves view their professional activity as highly
speculative.
This appreciation could result in the particular view that a prolifer-
ation of theories should occur in the development of 20th century cos-
mology. However, in this paper it will be shown that this is not the
case.
Plan
First, the early stages of modern or twentieth century cosmology will be

dealt with. Second, the problem of the effective geometry of the universe
is examined. Third, the current views are discussed and finally, the future
developments are mapped.
EARLY STAGES
General relativity theory implied a group of dynamical models for the

universe. Initially (around 1916) Einstein thought the universe as static,
so he ignored all evolutionary models by modifying the equations.
In 1921 Alexander Friedman rediscovered the different possibilities
for models in Einstein's work. George Lemaitre did the same thing a
couple of years later (in 1926, to be precise). He went one step further
by devising a plausible cosmogony. The development is shown in the
graph on the next page.
Although Vesto Slipher found the galaxies receding in 1914 (he pub-
lished his discovery in 1915), Einstein was seemingly not acquainted with
these findings.
In 1923 Edwin Hubble began his observations of the galaxies and in
1928 he was able to formulate his famous law. By that time Lemaitre had
advanced his cosmogonic hypothesis. According to his view, the universe
emerged out of a region as small as the solar system, the primeval atom.
During the next decade a system was developed to categorize all
the different models, which were viewed as quite possible at that time.
They were called the Friedman-Lemaitre-Robertson-Walker-models, or
FLRW-models for short.
A subgroup turned up to be in close agreement with Hubble's observa-
tions of spiral galaxies and the cosmologists saw these models as highly
possible descriptions for the universe.
COSMOLOGY AND PROLIFERATION 141
1914
._s_li.;.p-rhe_r""""'!"'_..,. .......................X·························~
*
1915
1916
Einstein
1Einstein I
1917 I
~ e Sitter
1918
,..... .p.-.jmodel
1919
f
1920
1921 Fried:m.an
1922
1923 '
r----'------,·4·······;
Hubble:
1924 dis tan.c es
1925 '
1926 Lemaitre
1927
1928
~-· · · · · · · ·' Lemaitre:
expansion
Hubble: ~ ................ !
.....
1929
1930
Hubble's law
Hubble: ..
geometry
Lemaitre:
1931 primae vel
atom
The fat arrows depict real influences, while the faint ones stand for supposed or
secondary influences (when they did not happen, they are crossed). The vertical
dotted line distinguishes practical astronomy from theoretical cosmology on the
right.
Einstein revised willingly his so-called blunder of 1916: the fact that
he had renounced the dynamical models in the first place.
For a while, the physicists did agree upon the hypothesis that the
universe originated out of a physical singularity-precisely the hypoth-
esis of Lemaitre. However, they did not so regarding the future of the
universe: three possibilities remained. So, at this early stage in the de-
velopment of modern cosmology, although an almost infinite amount
of models was (mathematically) provided, roughly three groups among
the cosmologists existed, corresponding to the three physically feasible

models.
time
Ku111Hilr..<
The universe came into existence at time to; we are now at t 1 . These models
accorded respectively with an open, closed or flat universe. The tangent line
stands for the constant acceleration of the galaxies, which is seen as implausible.
The extension of the line until it intersects the time-line, results in the so-called
Rubble-time.
EFFECTIVE GEOMETRY
The question which one of the models to choose is equivalent to the ques-
tion which overall geometry the universe has. In effect, it was Hubble's
original idea to determine the curvature of the universe by observation,
by determining the spreading of the galaxies. However, merely count-
ing them won't suffice, simply because a lot of the existing galaxies are
barely if not visible.
Through the seventies and eighties several other astronomical tech-
niques were developed to measure the curvature. The principle they all
had in common was the fact that there was an equivalence between
the curvature and the so-called deceleration-parameter qo. This quan-
tity measures how fast the universe expansion diminishes. One way to
determine the deceleration is to determine the total mass of the universe
and to check whether it matches the critical mass, the theoretical mass
necessary to decelerate the expansion-rate, that much as to bring the
expansion to a standstill.
Two positions were and are still taken by cosmologists and astronomers
alike. The first one is to postulate that the universe is flat (so the ratio
mass j critical mass equals unity) and to search for enough mass to keep
the universe gravitationally in control. The second one is just to measure
the mass without any preconception.
The first group had to tackle the problem that only 1 percent of
the needed mass was visible. Another 9 percent could only be detected
indirectly-the so-called dark matter (for example, by dynamical mod-
els of galaxies and velocity-distributions in clusters of galaxies). Still 90
percent was missing, aptly named the missing matter. There was no way
to account for this matter, except by stating that it simply had to ex-
ist, because the universe had to be flat-there were doubtful indications
that the geometry was almost Euclidean, ranging from mere metaphys-
ical considerations to misinterpretations of physical theories. Anyway,
some cosmologists went out for some exotic particles, just enough to
constitute the missing matter. First they thought of massive neutrinos,
then they tried out axions. Needless to say that cosmologists and parti-
cle physicists were best friends during the eighties. The zoo of particles
found a reason for existence in cosmology.
The different particles implied specific cosmogonical assumptions. In
a certain sense, different and altogether new cosmological theories emer-
ged. But they disappeared almost as fast as they were developed.
The search for the missing matter had its inevitable repercussion on
the dark matter models. More than a dozen were tried out, but seemed
to be in conflict with observational data or the standard big bang cos-
mogony, which had proven to be quite reliable. Less than half a dozen
remain, so there is still no consensus on the geometry of the universe.
Cleverly, standard theory was recently defined (by Jim Peebles) in
such a way that this question could be left open. This fact ultimately
confirms that a proliferation of cosmological theories failed and still fails
to occur.
CURRENT VIEWS
Alternatives to Standard Theory
Were there never any genuine alternatives to standard theory? In the

late forties, the idea of a 'steady state universe' was developed. Maybe
the universe did not have a singular origin. Matter was continually being
created, but the expansion kept the overall mass-density constant. This
proposal could be reconciled with relativity theory and the observational
data.
In 1965 the standard big bang theory became confirmed by the 2. 7 K
background radiation, which could solely be accounted for by the hy-
pothesis of a hot origin. The steady state theory became refuted, but
the underlying principle is still being defended by a rather small, but no

less illustrious group of cosmologists (Hoyle and Narlikar).
However, in the same year another theory that did not accept the
unique singularity hypothesis was proposed. It was the 'oscillating uni-
verse model' of Robert Dicke. According to Dicke, the universe periodi-
cally expanded and contracted, a phase which was followed by an overall
explosion and another period of expansion .
..............................,
,.. ... ,...
,... '•,,1 ,.•'
I
II
"' ... I1111II "'
time
Everything said about the foregoing stages was uncertain on physical

grounds, and it was not clear whether an infinite or a finite series did
precede-although a finite series would suffice and an infinite amount
stated severe thermodynamical problems.
The oscillating universe model faded away since it was not falsifiable,
but the fundamentals (cyclism) are nowadays defended by some philoso-
phers, theologians, and space-time theoreticians.
Some of those theoreticians (namely, Brans and Dicke) developed a
whole new relativity-theory, postulating the gravitational constant time-
dependent. They did so, because there was an inconsistency in the time-
distance scaling of the universe, based on Hubble's law and Einsteinian
relativity. However, the proposal had no general impact, since the in-
consistency could be wiped out within standard theory thinking.
During the eighties, cosmogonical theories based on extragalactic mag-
netism were developed. In the so-called plasma universe, the galaxies
came into being by a magnetic radiation-matter interference. Recently
it has been shown that a magnetic origin for galaxies is unlikely, although
the interstellar magnetic influence on stars and planets is indisputably
there. Some scientists have tried to match their theory of extragalactic
magnetism with the 'empirical proof' of standard theory, that is, the
existence of the 2.7 K cosmic background radiation. They succeeded.
They could also explain some structural features of the galaxy in An-
dromeda, although a straightforward dynamical approach does the same
thing. Since there are no means to confirm or falsify the existence of in-
tergalactic magnetism, the group of adherents stays rather small and the
theories tend to disappear.
This application of magnetism-in science and pseudo-science taken

together, it is a frequently occurring 'deus-ex-machina'-was the only
real attempt to contest standard theory in recent times.
What about inflation-theory? The inflation-theories can hardly be seen

as alternatives at all to standard theory. And if one argues to do so, it
should be mentioned that only a few inflationary models were (consecu-
tively) developed. This diagram shows the elaboration of the inflation-
concept in cosmology.
The inflationary concept can give an account for the vastness of the
universe, although this anomaly was not the initial problem to the the-
ories. They are undoubtedly very speculative, but approximately half
of the cosmologists regard them as quite reasonable and are willing to
accept them.
FUTURE DEVELOPMENTS
It is interesting to see that many cosmologists believe in a convergence

of their theories to one major theory of everything. No consensus exists
on the content of that utopian theory, but they all agree that there has
to be some or other global theory. Some call this theory the blueprint of
the universe or God's thoughts or the philosopher's stone. Relevant to
us is the fact that the cosmologists through their thinking are limiting
the quantity of theories.
The end of cosmology is not to be expected in the near future, although

some physicists think otherwise. Anyway, a lot among them are trying to
stabilize and to ground the standard theory on a solid base. Obviously,
ad hoc hypotheses are devised in this process. This does not mean that
one should speak of a proliferation of theories, since ( 1) no new 'big'
theories seems to emerge and (2) there are no reasons at all to revise the
standard theory thoroughly.
1967 horizon-problem
1979 ( flatness-pro olem

_l
monopole-pro olem
~
I
tll.rlnel.ing
I
J
inflation: Guth's theory
inflation: Linde's theory

"bubbles
__re_ro,g_eM_r_·ry____~)
( _____M
1982
t
inflation: MW inflationtheory
--1 Heisenberg
uantum-in!1ation
Pee "bles-Zel'dovich
1983 inflation: chaotic inflation
CONCLUSION
During its early development, the standard theory of cosmology grouped

three alternatives regarding the geometry of the universe. No prolifera-
tion occurred, notwithstanding the fact that an infinite amount of possi-
ble models for the universe were implied. For sure, cosmology was (and
still is, for that matter) very speculative, but it did not get off the rails.
For the recent developed theories, the same holds. If the cosmologists
are right and a theory of everything will be found, no proliferation of
scientific theories will occur in the future.
Is there no triviality in all this? If one considers astrology, it is readily
seen that one has to deal there with an enormous amount of different
views on the same topic-as many as there are astrologers. Western as-
trologers respectively take completely dissimilar systems (even Chinese
astrology) as their starting point, but then vary idiosyncratically on that
basis. One can hardly speak of a consensus among the astrologers. Maybe
the concept of proliferation can shed new light on the demarcation-
problem.
Though, if one considers quantum mechanics, it is also seen that there
are at least between twenty and thirty different more or less accepted
interpretations going. In fact, every quantum physicist has his own meta-
physically induced view on the topic.
And if one considers particle physics, more than five different theories
are in competition with each other (ranging from the so-called standard
theory-and its derivatives-to string theory). They seem to be all in
mutual contradiction.
SST
FUM c haoti£ :infhtio II.

Co n.tiiLuo 1JS :infht ion.
What about cosmology? This last diagram shows clearly that, al-
though several different cosmological theories undoubtedly existed (and
some still exist), the overall picture remains quite surveyable. SST stands
for: Steady State Theory, OUM for: Oscillating Universe Model, PUM
for: Plasma Universe Model-the magnetic alternative. The lines are
partly dotted, meaning that these theories are slowly disappearing out
of the picture. You have the inflation theory development, not in con-
flict with Standard Big Bang Cosmology, but not generally accepted as
a part of it
We can conclude that, in spite of the existence of an appropriate breed-
ing ground and a suitable methodological tradition, there is no prolifer-
ation whatsoever appearing in the development of modern cosmological
thinking.
AFFILIATION
Dr. Gustaaf C. Camelis

Senior Research Assistant, F.S.R. Flanders, Belgium
Center for Logic and Philosophy of Science
Vrije Universiteit Brussel
REFERENCES AND FURTHER READING
[1] Davies, P.C.W. (ed.), The New Physics. Cambridge, Cambridge

University Press, 1989.
[2] Hawking, S.W. and Israel, W. (eds.), General Relativity: An Ein-
stein Centenary Survey. Cambridge, Cambridge University Press, 1979.
[3] Gribbin, J ., In Search of the Big Bang. Quantum Physics and Cos-
mology, Londen, Bantam Press, 1986.
[4] Kuhn, T.S., "The Structure of Scientific Revolutions", International
Encyclopedia of Unified Science, Chicago, University of Chicago Press,
2, 2, 1962.
[5] Peebles, P.J.E., Schramm, D.N., Turner, E.L. and Kron,
R.G., "The case for the relativistic hot Big Bang cosmology", Nature,
352, 1991, pp. 769-776.
[6] Overbye, D., Lonely Hearts of the Cosmos. New York, Harper &
Row, 1990.
[7] Weinberg, S., The First Three Minutes. A Modem View of the
Origin of the Universe. New York, Basic Books, 1977.
PAUL J. LEWI
RESONANCE AS A UNIFYING PRINCIPLE BETWEEN THE

OBSERVER AND THE OBSERVED
1. RESONANCE
Resonance is most often associated with vibrating mechanical and elec-

trical systems. Pendulums, strings, springs, membranes, sound boxes,
etc. are illustrations of simple mechanical resonating systems which we
encounter in daily life. The behavior of each of these can also be repro-
duced, although less visibly, by wiring suitable electrical and electronic
components together. More complex resonating systems are found on
all scales of physical dimension from the microworld to the macrocos-
mos, ranging from elementary particles, molecular assemblies, bridges,
planetary systems to galaxies.
In crude terms, one could state that resonance occurs whenever the
output of a system is returned to it as input, as in a classical feedback
loop. Each of the mechanical devices which we mentioned before, when
set in motion, will vibrate at a given frequency called the eigenfrequency
of the system. (The prefix 'eigen' is derived from the corresponding Ger-
man adjective, which means proper, dedicated, specific etc., but which
is used here in the sense of principal, fundamental, canonical, etc.) In
what follows we distinguish between eigenvibration and eigenvariation
to indicate that the system supports sustained oscillations or not. The
distinction is not essential, as both are considered here as the result of
resonance in the system.
The concept of a feedback loop can be properly depicted by means
of the alchemist symbol and Jungian archetype of the Ouroburos (Fig.
1), the snake that bites its own tail, representing at the same time the
'materia prima' and eternity [1]. On the one hand, the static perception
of the symbol produces a snake that holds its tail between its teeth.
A dynamic vision of the symbol, on the other hand, is that of a snake
that once bit its tail, causing a sensation of pain, causing it to bite its
tail again, causing a sensation of pain again, and so on ad infinitum. Of
course, there is no way of telling which of the two visions, the static or
the dynamic one, is closest to reality. Possibly, both play a role in our
understanding of the world. The story goes that the Ouroburos symbol
led Kekule in 1866 to conceive a static six-carbon ring structure of ben-
149
150 PAUL J. LEWI
Figure 1. Ouroburos or archetypal symbol representing primeval matter and eternity.

In our context, the symbol represents the occurrence of resonance in a system.
zene which is composed of three single and three double bonds [2]. Al-
ternatively, the benzene molecule is also regarded as resonating between
two hexagonal configurations with different electronic distributions.
The phenomenon of resonance is so ubiquitous that most of the time
we are not aware of it. But whenever we look closer at what makes the
world tick, vibrating and interacting systems abound, often in surprising
disguise. Our approach may be likened to that of the Pythagoreans who,
once they discovered the link between sound vibrations and the physical
dimensions of the objects that produced them, attempted to extend the
principle to the whole of the universe. Kepler's harmony of the celestial
spheres is a typical testimony thereof. In this paper we will draw atten-
tion to instances of resonance in a field where it is hardly expected to
occur, namely in the statistical analysis of data.
2. D AT A AND IN F 0 R MAT I 0 N
Data analysis is a branch of statistics which attempts to extract informa-

tion from tabulated data. For example, one may consider the chemical
elements and compile their macroscopic chemical and physical properties
in the form of a table. These properties include atomic weight, specific
heat, density, boiling point, acidity, etc. We refer to the rectangular ar-
rangement of the numbers by objects and by properties as the data.
Information refers to relationships and structure in the data. For exam-
ple, Mendeleev, through a long process that has not been (and perhaps
never will be) formalized, found his famous periodic arrangement of the
elements which has become a paradigm in chemistry ever since. While
the generation and compilation of data is often a highly codified scien-
tific activity, the extraction of information from data draws from both
science and art.
RESONANCE AS A UNIFYING PRINCIPLE 151
3. FACTORS
In data analysis one may discern two basic approaches. Cluster anal-
ysis, on the one hand, aims at producing taxonomies or classifications
which show the similarities of things according to recorded measure-
ments. It also may show the phylogenetic descent of the objects from
ancestors that are inferred from the data. This approach may be called
Aristotelian, as it accepts the complexity of the world as it is. It is exem-
plified by the work of Linnaeus and has produced the hierarchical divi-
sion of the animal kingdom into classes, orders, families, genera, species,
etc. Factor analysis, on the other hand, attempts to find hidden factors
which may explain the variation in the data. The underlying assumption
here is that the number of factors is less than the number of observed
properties. This approach may be termed Platonic, as it attributes real-
ity to the pure factors from which our imperfect observations emanate.
One is reminded here of Plato's tale of the cave in which projections are
made on a wall by forms that are invisible to the observers [3].
Factor analysis originated around 1930 in the context of the study
of intelligence by means of psychometric measurements by Spearman,
Holzinger, Thurstone and others [4]. These early factorists considered the
multitude of psychometric measurements as mixtures of a few primary
traits which included general intelligence (the famous IQ), abstraction,
language, music, visualisation, etc.
In order to explain the concept of factors, Thurstone reminded us
that any of the millions of colors that we can distinguish is reducible
to a mixture of only three primary colors (e.g., red, blue and green).
These can be regarded as the factors of color vision [5]. Their meaning
derives from the three types of cone cells in the human retina, which
possess specific photoreceptors with maximal absorption for red, blue
or green light, respectively. Note that absorption of photons at specific
wavelengths by molecules is intrinsically a resonance phenomenon.
When Mendeleev in 1869 arranged the chemical elements by groups
and periods, he identified atomic weight and chemical valence as the two
factors which explained most of the chemical and physical properties that
were known in his days. His approach allowed to correct some errors in
the data and successfully predicted the existence and the properties of
Gallium, Germanium and Scandium [6]. Much later, the physical mean-
ing of the two factors became apparent when they were identified with
the first two quantum numbers [7]. Note that the latter are solutions of
a mathematical equation that describes the resonance properties of the
electron.
152 PAUL J. LEW!
4. EIGENVALUE DECOMPOSITION
Jacobi, around 1846 derived the periodicities of the planetary system

from a table of physical coefficients [8]. The latter appeared in the system
of equations that described the motions of the planets around the sun
according to Newton's laws. The method was named eigenvalue decom-
position as it decomposes a table of coefficients into a set of eigenvectors
and their associated eigenvalues. Note that eigenvalues are the equiv-
alents of eigenfrequencies of oscillating systems and that eigenvectors
describe how each component of the system is affected by the different
eigenvibrations that the system allows. The number of eigenvectors and
associated eigenvalues of a system depends on the number of indepen-
dent modes of vibration. Strings, membranes and sound boxes possess
respectively one, two and three modes of vibration. Color vision and the
periodic arrangement of the chemical elements, as we have seen, have
two and three modes of variation, respectively.
The name eigenvalue decomposition is usually reserved for mechan-
ical, electronic and quantum mechanical systems for which theoretical
models have already been developed. The term factor analysis is often
used in empirical studies of psychology, sociology, biology, economy, etc.
that generate large amounts of data and for which adequate models still
have to be developed. Both contexts share the same objective, namely
to describe a system in terms of its principal modes of vibration or vari-
ation. Our bias here is toward the factor analytic approach which starts
from empirical data rather than from a mathematical model. Before we
can proceed, however, we must provide a geometrical interpretation of
factors.
5. DATA SPACES
Let's assume a table which describes the first 36 chemical elements (from
Hydrogen to Krypton) in terms of two properties, say atomic weight and
boiling point [9]. Such a table is called bivariate because it involves two
properties, each of which constitutes a different point of view. Conven-
tionally, for practical and aesthetic reasons, we associate the rows of
the table with the chemical elements and we relate the columns to the
properties. In theory, however, the roles of rows and columns are in-
terchangeable. A familiar geometrical interpretation of the table is the
Cartesian diagram in which atomic weight and boiling point are associ-
ated with the horizontal and vertical coordinate axes, and in which each
element is represented as a point in the plane of the diagram. Although
the geometrization of mathematics and physics is usually attributed to
Descartes, its origin has been traced back to the 14th century. Indeed,
RES 0 NAN C E AS A UNIFY IN G P R INC I P L E 153
the transition from geographical longitude and latitude to more abstract

physical coordinates is attributed to the nominalist philosopher Nicole
Oresme [10]. The bivariate diagram of atomic weight and boiling point
reveals the periodic structure of the chemical elements as observed ear-
lier by Mendeleev, although this is the result of a most fortunate choice
of the two points of view. We will indicate below that factor analysis may
lead to a choice of properties which is most likely to produce meaningful
information from the data.
If we now look at the chemical elements as a two-dimensional system
of points in the plane diagram of atomic weight and boiling point, we
may determine its two principal axes of inertia. In the case of an elliptic
cluster of points, the principal axes coincide with the major and minor
axes of symmetry of the system. By construction, they are mutually per-
pendicular and each in turn accounts for a maximum of the variation
in the data. At the beginning of the century, Karl Pearson described
the principal axes as the lines of closest fit to the data [11]. Hence,
they provide the most parsimonious description of the information in
the data. In fact, the principal axes are the factors or eigenvectors of
the data table. The variations in the data accounted for by the princi-
pal axes represent the associated eigenvalues. We refer to the latter as
eigenvariations. Hence, the importance of an eigenvector or factor can
be determined from its associated eigenvariation. This provides the basis
for a geometrical interpretation of factors in terms of principal axes and
the variations accounted for by them.
In the foregoing, we have described the rows of a table as points in
a space spanned by the columns, which is referred to as column-space.
Mutatis mutandis, since the roles of rows and columns are interchange-
able, we may also describe the columns of the same table as points in a
space spanned by the rows, which is then called row-space. The two data
spaces are conjugate, which means that a change in one, automatically
brings about a corresponding change in the other. We also obtain two
sets of principal axes, one in column-space and the other in row-space.
These principal axes are conjugate two-by-two. Each pair expresses one
and the same eigenvector or factor of the data and is associated with the
same eigenvariation. In what follows we make no distinction between the
terms eigenvectors, principal axes or factors, which are used as synonyms
here.
6. FACTOR PLOTS
The notion of conjugate data spaces may seem trivial in the case of a
bivariate table such as described above. Let's therefore consider the mul-
154 PAUL J. LEWI
tivariate case, in which the 36 elements are described by, say, 8 chemical
and physicochemical properties. The first two eigenvectors of column-
space may serve to construct a bivariate diagram in which the two co-
ordinate axes represent the two factors, and in which each point defines
one of the 8 properties. Likewise, the first two eigenvectors of row-space
build another bivariate diagram in which the two coordinate axes rep-
resent the same two factors as above, and in which each point refers
to one of the 36 elements. We refer to these diagrams as factor plots.
Note that the factor plot for elements and the factor plot for properties
are conjugate. (They can be superimposed by making the origin and
the coordinate axes coincident.) One factor plot reproduces Mendeleev's
arrangement of the chemical elements by groups and periods [12]. The
other factor plot reveals the pattern of intercorrelations between the
properties. In particular, the latter shows that atomic weight and boil-
ing point alone explain nearly as much of the variation in the data as
all 8 macroscopic properties together. It is not surprising therefore, that
the bivariate diagram of atomic weight and boiling point is capable of
revealing the periodic structure of the elements. It is well known, and
has been pointed out before, that the factors identified by Mendeleev
are associated with the first two quantum numbers. These, in turn, are
solutions of an eigenvalue problem which describes the resonant wave
characteristics of the electron. Since the macroscopic properties of the
elements can be related to the resonant states of the electrons in the outer
shells of the atoms, it is not surprising that factor analysis is capable
of unveiling the underlying quantum-mechanical structure. A surprising
fact, however, is that mathematical algorithms of factor analysis can be
likened to processes that occur in the resonating systems from which the
data themselves have been generated [13]. This will be discussed in more
detail below.
7. ALGORITHMS
Resonance, as we have remarked above, occurs when we set a system

in motion and then feed the output back as input. This concept can be
applied to a table of data in order to extract its eigenvectors.
Let's consider a data table X. An arbitrary vector u, with as many
elements as there are rows in X, is multiplied with the row-side of X.
The result is a vector v, with as many elements as there are columns in
X. Next, we multiply v with the column-side of X, which produces
an updated vector u. At this point we have completed one cycle of
the algorithm, which can be symbolized by means of the flow diagram
of Fig. 2. Note that a vector represents an axis in space and that an
RES 0 NANCE AS A UNIFYING P R INC I P L E 155
...
\ '
' \
''
f. ~1
2 \\ \
\
~\ -------
' I
'~
Figure 2. Cycle of the NIPALS algorithmic for the extraction of the dominant row-
eigenvector u and the conjugated column-eigenvector v from a data table X. The
diagram illustrates the analogy with a resonant system in which the output is fed
back as input.
eigenvector of a table represents a privileged axis, namely the principal

axis of the corresponding system of points. Successive repetitions of the
cycle will produce the conjugate vectors u and v, the directions of which
eventually converge to those of the dominant eigenvector of row-space
and column-space, respectively. Each time when a cycle is completed,
the numerical values of u and v are multiplied by the eigenvariation
that is associated with the dominant eigenvector. (In order to prevent
that the numbers in u and v either increase or decrease indefinitely, it is
mandatory to renormalize u or v after the completion of a cycle.) After
a few cycles, the vectors u and v converge to the dominant eigenvector
of row- and column-space, within the precision of common calculators.
The mathematical formalism of the eigenvector algorithm is given in
appendix A.
The first step of the algorithm can be regarded as the initial excita-
tion of the system, after which the output is fed back as input. The cycle
is iterated until both the input and the output become stationary and
a resonant state has set in. One easily recognizes the archetypal sym-
bol of the Ouroburos (Fig. 1) in the flow diagram of the algorithm for
eigenvector extraction (Fig. 2).
156 PAUL J. LEWI
Figure 3. Coupled Ouroburos symbol. In our context, the symbol represents the oc-
currence of resonance in two coupled systems.
After extraction of the dominant eigenvector, one can deflate the data
table X for the variation that is accounted for by that eigenvector (u or
v). If the iterative procedure, which we described above, is now applied
again to the residual table, we obtain the next dominant eigenvector.
This can be repeated until the table is exhausted, which occurs when
all the eigenvectors have been extracted and when the residual values
in the table become zero. The mathematical definition of deflation is
presented in the appendix. Technically, the algorithm for extraction of
eigenvectors from a table of data, which we described above, is called
NIPALS (which is short for non-linear iterative partial least squares)
[14]. It is also related to the so-called powering method which is used for
the extraction of eigenvectors from a table of intercorrelations between
properties [15].
Frequently, a situation occurs when the same set of objects is described
by properties that were determined in different contexts. We have al-
ready discussed the example of the table which describes 36 chemical
elements by 8 macroscopic properties, such as atomic weight, specific
heat, boiling point, etc. Let's now consider an additional table in which
the same 36 elements are described by 6 quantum-mechanical proper-
ties, such as atomic volume, Pauling electronegativity, ionization energy,
etc. [9]. An interesting problem now is to find out whether the quantum
mechanical properties can be predicted with a reasonable degree of accu-
racy from the macroscopic ones. This is a classical problem of regression.
It can also be seen as the problem of finding common resonance modes
between the two coupled systems that are described by the two tables.
The Ouroburos archetype can be helpful again if we consider two snakes
which bite each other's tail (Fig. 3). This symbol exemplifies a system
of two coupled oscillators, which under certain conditions exhibit a com-
mon mode of eigenvibration.
To fix our ideas we refer to the two tables by means of the symbols
Xl and X2. (We assume that corresponding rows in the two tables refer
X1 \.11 V2. X2
,. ..-' ---- .....

', ...
" ''
~/,, ------
___ /,t:r~r -
.,
___,.~-,.
T I ,'
V1
Figure 4. Cycle of the PLS algorithm for the extraction of the dominant common row-
eigenvectors ul, u2 and the conjugated column-eigenvectors vl, v2 from two coupled
data tables Xl, X2. It is assumed that corresponding rows in Xl and X2 refer to the
same objects. The diagram illustrates the analogy with two coupled resonant systems
in which the outputs are fed back as inputs.
to the same objects.) The cycle is initiated by multiplying an arbitrary

vector ul with X2, which produces v2, which, when multiplied with
X2, produces u2. The latter is multiplied with Xl and produces vl,
which, when multiplied with Xl, yields an updated ul. At this point
we have completed one cycle of the algorithm, which is symbolized in
the flow diagram of Fig. 4. After iteration of the cycle, one finds that
the values of ul, vl and u2, v2 become stationary and that a resonant
state has been obtained. At that stage, the conjugate vectors ul and vl
have converged to the first common dominant eigenvector in the raw-
and column-space of Xl, respectively. Likewise, the conjugate vectors
u2 and v2 have converged to the same common dominant eigenvector in
the row- and column-space of X2, respectively. As it was the case with
the single table, one can deflate the tables Xl and X2 for the variation
that is accounted for by previously extracted eigenvectors (for example,
ul or vl). The procedure, defined above, can be applied to the residual
tables until all common eigenvectors have been extracted. At that point,
all the values in the residual table Xl become zero. The mathematical
details are defined in appendix B. Technically, this common eigenvector
algorithm is referred to as PLS (which is short for partial least squares)
[16].
158 PAUL J. LEWI
The common eigenvector algorithm, as described above, is useful when

laboratory tests are designed in order to predict unknown real life situa-
tions, when instrument readings have to be calibrated for gauging spec-
imens of unknown constitution, etc. Given the numbers in an additional
row of the first table, one can derive the numbers in the corresponding
row of the second table from the computed eigenvectors.
Data tables which describe the same set of objects can be coupled to-
gether in various ways and their common eigenvectors can be determined
in much the same way as above. Each of these eigenvectors represents a
common mode of resonance of the coupled system. Each of these circuits
can also be represented by an extended form of the Ouroburos, one snake
biting one or more tails of the others and each being directly or indi-
rectly connected to all the others. The importance of resonances, both
in natural systems and in the analysis of data that are obtained from
them, has led to a view of the world which is briefly explained below.
8. WORLD VIEW
We view the world as being composed of resonating systems, possibly

connected with one another to form larger systems. The latter, in turn,
may form resonating supersystems, possibly connected with one another,
and so on. These systems may resonate in the strict sense of oscillatory
behavior as in mechanical, electric and quantum-mechanical systems or,
more generally, in the sense of specific interactions as occur between
drugs and proteins, substrates and enzymes, antibodies and antigens,
between persons with positive or negative affinities, etc.
We acquire data about systems or supersystems in the world either
directly, by observation through the senses, or indirectly, by means of
measurements with instruments. So far, we have only discussed the in-
strumental way of data acquisition, since this is the one that is best
understood today (although the same arguments may be applicable to
sensorial data). Most of the data that are acquired (perhaps as much as
eighty percent) are explicitly dependent on time or frequency. In that
case, fundamental cycles and rate constants of the system that produced
the data can be determined by means of Fourier and Laplace transform
methods. This applies to spectroscopic data, climatological, economic
and demographic time series, biological rhythms, etc. We have focused
our analysis, however, on the more general type of data where objects
have been described in terms of their (more or less) immutable proper-
ties. We have shown that, in this case, factor analysis allows to determine
the fundamental modes of resonance. The latter can be represented visu-
ally in the form of factor plots which show similarities between objects,
intercorrelations between properties and specific interactions between

objects and properties. In a sense, factor analysis attempts to reconstruct
the world system. The reconstruction is imperfect, however, because of
the limited sensitivity and accuracy of the measuring instruments, be-
cause of noise and interferences that inevitably creep in, and also because
of fundamental constraints (such as the Heisenberg uncertainty relation
between position and velocity).
In summary, we state that, on the one hand, instruments respond
to resonances in the world systems and produce data in which these
resonances are encoded numerically. Data analysis, on the other hand,
tries to decode the data and attempts to reconstruct the world systems
in terms of resonances that are derived from the data. Because of reasons
cited above, the reconstruction can never be perfect, but its limitations
can be understood. This world view, based on coding and decoding of
resonances, is depicted schematically in Fig. 5.
It is tempting to extend the resonance paradigm beyond the field of
theoretical and experimental sciences. One may also consider, for exam-
ple, specific interactions between one individual and another, between
individuals of society and between different parts of society. In partic-
ular, consciousness may be related to the coding and decoding of reso-
nances within composite neural systems. Hence, we conclude that reso-
nance phenomena are common to the observer and to the world that is
observed, even when the observer is himself the object of observation.
Resonance thus forms a link between the objective and the subjective,
between science and art, as well as between rational analysis and poetic
vision.
9. APPENDIX
A. A cycle of the NIPALS algorithm can be formulated by means of two

steps, which are also indicated on the flow diagram of Fig. 2.
Step1: v = xr. u and v ~ vj(vT. v) 112
Step2 : u = X · v
where X represents the data table, and where the vectors u and v even-
tually converge to the dominant eigenvector of X in row- and column-
space, respectively. The dot symbol indicates the rows-times-columns
product for matrices, and the superscript T indicates transposition of
rows and columns. (Vectors are regarded here as single-column matri-
ces.) The normalization of v in step 1 of the cycle is required in order
to prevent u and v from becoming infinitely large or small.
After completion of a few cycles, the elements of u and v usually
converge with sufficient accuracy to their theoretical values.
160 PAUL J. LEWI
World Recons~roded
S!jstero World S!jsrem
...
...
... ... ...
Figure 5. World view involving resonances in world systems which are encoded in data
by means of instruments. The world systems are reconstituted, in part, by decoding
the data in terms of the eigenvectors that can be extracted from them.
The next dominant eigenvector is then obtained after deflation of X

for the variation accounted for by u. This produces a residual data table
X:
RES ON ANC E AS A UNIFYING PRINCIPLE 161
where I represents the unit matrix of row-space. The algorithm can be
repeated until all eigenvectors of X have been extracted, at which point
all elements of X become zero.
B. A cycle of the PLS algorithm can be defined in four steps, which are
also indicated on the flow diagram of Fig. 4.
Step! : v2 = X2T · ul and v2----- v2/(v2T · v2) 112
Stcp2 : u2 = X2 · v2
Stcp3: vl=X1T·u2 and vl--vl/(v1T·v1) 112
Step4 : ul = Xl · vl
where Xl and X2 represent two coupled data tables with corresponding
rows, and where the vectors ul, u2 and vl, v2 eventually converge
to the dominant common eigenvectors of Xl and X2 in the common
row-space and the two column-spaces, respectively. The normalization
of vl and v2 is required in order to prevent ul, u2 and vl, v2 from
becoming infinitely large or small. After completion of a few cycles, the
elements of ul, u2 and vl, v2 usually converge with sufficient accuracy
to their theoretical values. The next dominant common eigenvector is
then obtained after deflation of Xl and X2 for the variation accounted
for by ul. This produces the residual data tables Xl and X2:
Xl----- (I- ul · ulT julT · ul) · Xl

X2----- (I- ul· ulT julT · ul) · X2
where I represents the unit matrix of the common row-space. The algo-
rithm can be repeated until all eigenvectors of Xl and X2 have been
extracted, at which point all elements of Xl become zero.
AFFILIATION
Pa'ul J. Lewi
Center for Molecular Design
Janssen Research Foundation
Beerse, Belgium
REFERENCES
[1] von Franz, M.-L., Alchemy: an introduction to the symbolism and

the psychology, (Studies in Jungian psychology by Jungian analysts,
nr 5), Inner City Books, Toronto, 1980.
[2] Stoerig, H.J ., Kleine Weltgeschichte der Wissenschaft, Vol II, W.
Kohlhammer Verlag, Stuttgart, 1965.
162 PAUL J. LEWI
[3] Plato, Republic, (428-347 BC); Oldroyd D., The Arch of Knowl-
edge. An introductory study of the history and methodology of science,
Methuen, New York, 1986.
[4] Cattel, R.B., Factor analysis, An introduction and manual for the
psychologist and social scientist, Harper and Brothers, New York, 1952.
[5] Thurstone, L.L., Multiple-factor analysis. A development and ex-
pansion of the vectors of mind, University of Chicago Press, Chicago,
Ill., 1947.
[6] Kolodkine, P., Dmitri Mendeleiev et la loi periodique, Ed. Seghers,
Paris, 1963.
[7] Hund F., Linienspektren und periodisches System der Elemente,
Berlin, 1927; D'Abro, A., The rise of the new physics, Vol 2, Dover
Publ., New York, 1951.
[8] Jacobi, C.G.J., "Ueber ein leichtes Verfahren die in der Theorie der
Sekularstoerungen vorkommender Gleichungen numerisch aufzuloesen",
J. reine und angew. Math., 30, 1846, pp. 51-95.
[9] Table of periodic properties of the elements, Sargent-Welch Scientific
Cy., Skokie, Ill., 1968.
[10] Borchert, E., Die Lehre von der Bewegung bei Niclaus Oresme.
Beitraege zur Geschichte der Philosophie und Theologie des Mittelalters,
Aschendorffscher Verlagsbuchhandlung, Muenster, 1934.
[11 J Pearson, K., "On lines and planes of closest fit to systems of points
in space", Phil. Mag., 2 (6th series), 1901, pp. 559-572.
[12] Lewi, P.J., Multivariate data analysis in industrial practice, Re-
search Studies Press (J. Wiley), Chichester, Eng., 1982.
[13] Lewi, P.J., "Pattern recognition. A chemometric point of view",
Chemometrics Intell. Lab. Syst., 28, 1995, pp. 23-33.
[14] Wold, H., "Soft modelling by latent variables: the non-linear it-
erative partial least squares (NIPALS) algorithm", In: Goni, J. (ed.),
Perspectives in probability and statistics, Academic Press, London, 1975,
pp. 117-142.
[15] Hotelling, H., "Analysis of a complex of statistical variables into
principal components", J. Educ. Psychol., 24, 1993, pp. 417-441.
[16] Wold, S., Albano, C., Dunn III, W.J., Esbensen, K., Hell-
berg, S., Johansson, E. and Sjostrom, M., "Pattern recognition:
finding and using patterns in multivariate data", In: Martens, H. and
Russwurm, H. Jr. (eds.), Food research and data analysis, Applied
Science Publ., London, 1983, p. 147.
KARIN VERELST AND BOB COECKE
EARLY GREEK THOUGHT AND PERSPECTIVES FOR THE

INTERPRETATION OF QUANTUM MECHANICS:
PRELIMINARIES TO AN ONTOLOGICAL APPROACH
1. INTRODUCTION
At the origin of our approach lay two encounters between Greek thought
and Quantum Mechanics, one of them deliberately conceived by its au-
thor, the other being a meeting between new QM-concepts and one of
the oldest problems of human thinking about the world. The first en-
counter has been presented in a short lecture by C. Piron [74, p. 169] in
which he attempted to develop a realistic QM-interpretation based on
two concepts fundamental to Aristotelian metaphysics, viz. potentiality
and actuality. The second one is the doctoral dissertation of D. Aerts [3].
It both by content and title dealt with the problem of the One and the
Many, the central theme Plato inherited from fifth century philosopher
Parmenides of Elea. In what follows we will attempt to make clear that
these encounters are not of a purely coincidential nature. We intend to
develop in this paper an analysis and re-evaluation of these old questions
and their solutions. Our contention is that this analysis might amount
into new perspectives on the interpretation of QM, since the enigmas
and paradoxes of early Greek thought, and the solutions presented to
them in the "classical period" are, more than we realize, bound to mark
our way of looking at and reasoning about the world 1 . The thought in-
struments developed by Plato and Aristotle, in order to solve the riddles
following out of so called "pre-Socratic" thought, which are epitomized
in the, according to both classical thinkers apparent, contradiction be-
tween the "world views" of Heraclitus of Ephesus and Parmenides of
Elea, are in use up to the present, be it in slightly modified forms 2 . Our
1
We find support for the relevance of this position in Schrodinger's fine little book
dedicated to the subject [87, p. 3ff, p. 159].
2
"Two great warring traditions regarding consistency originated in the days of the
Presocratics at the very dawn of philosophy. The one, going back to Heraclitus, insists
that the world is not a consistent system and that, accordingly, coherent knowledge of
it cannot be attained by man. (... ) The second tradition, going back to Parmenides,
holds that the world is a consistent system and that knowledge of it must correspond-
ingly be coherent as well, so that all contradictions must be eschewed." [83]
163
164 KARIN VERELST AND BOB COECKE
position will be that this classical contradiction has slipped through the
ages unimpaired, but in different forms, such as to make it hardly recog-
nisable in our present epistemological and ontological concepts, both
in philosophy and in science. A revelation of this implicit presence by
reconstructing the outlines of its historical pathway then becomes the
necessary first step towards an approach for the tackling of problems
it eventually causes in to-day's science. The argument will lead us to
the conclusion that the paradoxes appearing in QM represent such a
problem. A sketch of some possible strategies will complete this attempt
at clarification. Methodologically therefore the arguments in this paper
will be based both on scientific and philosophical grounds; even more,
it can be considered our aim to show that, when it comes to a proper
understanding of the significance and implications of basic "scientific"
findings, both are inextricably intertwined, since science, seemingly so
different, follows a path of deep conceptions laid down much earlier in
the development of human consciousness. We are tempted to see the ori-
gins of the QM-paradoxes as consequences of the ontological "choices" of
Plato and Aristotle. Their effort concerned the stabilisation of the world
of constant change, thus saving the possibility of certain knowledge in or-
der to escape the contradictions between stable and unstable, knowable
and unknowable that appear on the level of what happens in reality, as
expressed mainly by Heraclitus and Parmenides. This led to the concep-
tion of logic as a standardising rule for thinking and, much later, exper-
iment as a standardising rule for experiencing [34, IV: 99, p. 381]. Both
originated out of needs felt in the context of the macroworld, and reach
now their limitations in the study of the microworld. At the moment
when the interactions between things become as important as the things
themselves, the separating intervention in reality, first conceptualised by
Plato and Aristotle, seems to reveal itself as an illusion. For this reason
in this paper no position will be taken in the debate between rationalists
and empiricists. From our point of view these philosophical stances come
down to the same on the level of the ontological question: all imply the
metaphysical world-structure put forward by Plato and Aristotle as a
solution for the ontological paradoxes raised by pre-Socratic thought.
2. THE HISTORICAL CONTEXT
Pre-Socratic philosophy [70, Introduction] is the general name given to a

rather differentiated group of Greek thinkers, living from the 7th to the
5th century BC. The two main currents that are of concern to us here
are the Heraclitean and the Eleatic school of thought. The Heracliteans
claimed to be the followers of Heraclitus of Ephesus, of whose work
GREEK THOUGHT AND QUANTUM MECHANICS 165
only some fragments survive. Plato, followed by Aristotle, assigned to

him the flux-theory, the doctrine of permanent motion and unstability
in the world. The consequences of this doctrine are, as both Plato and
Aristotle stressed repeatedly, the impossibility to develop stable, certain
knowledge about the world, for an object, changing each instant, does
not allow for even to be named with certainty, let alone to be "known",
i.e., assigned fixed, objective characteristics. A fragment that might have
inspired Plato ([45], Cratylus, 66, 402(a); Theaetetus, 72, 160 (d, e))
for this interpretation is the famous riverfragmenf3 in which Heraclitus
states that because of the ever ongoing flow of the waters, it is impossible
to step into the same river twice. He stresses the ever ongoing change
or motion (both being aspects of the same process) that characterises
this world and its phenomena. This is so because Being, over a lapse
of time, has no stability. Everything that is at this moment changes
at the same time, therefore is not. This coming together of Being and
non-Being at one instant is known as the principle of coincidence
of opposites 4 . It is crucial to see that this principle is connected to
the possibilty of motion, for being in motion implies to be and to be
not at the same time at a certain place on a certain moment. It further
implies the unity of the world in the sense that there are no separated
objects 5 , its ontology is dynamic. However, this does not mean that Her-
aclitus meant to say that there is absolutely no stability in things. The
Heraclitean concept of harmony 6 , closely connected to the principle of
coincidence of opposites, rather points in the direction of a permanent
and momentaneous re-instantiation of things in the world in a web of
totally interconnected events. Reality is One. This does not imply the
"unreality" of the things we experience, it states our absolute intercon-
nectedness with the world we experience. Tradition transmitted as its
counterpart the Eleatic school, named after its inspirator Parmenides
of Elea. Of him more extant textfragments are known, although there
3
7to'taJ.lOl<; '!Ot<; auto'i<; EJ.l~atVOJ.lEV 't£ Kat OUK EJ.l~atVOJ.lEV, EtJ.lEV't£ Kat OUK EtJ.lEVln die-
selben Fli.isse steigen wir und steigen wir nicht, wir sind und wir sind nicht. [33, DK
49a, p. 161]
4
The formula stems from Nicolaus Cusanus, a thinker on the edge of the Middle
Ages and early Renaissance. But compare, e.g., Cusanus's "circular theology" with
the meaning of the "Ouroburos", the circular snake of ancient alchemy. For Cusanus,
see [34, p. 248ff]; for the Ouroburos, see [59, pp. 367-368].
5
ouK EJ.lOU, una tou Myou aKm)cravta<; OJ.lOAoyEtv cro<p6v Ecrnv EV navm dvat Raben sie
nicht mich, sondern den Sinn vernommen, so ist es weise, dem Sinne gemaB zu sagen,
alles sei eins. [33, DK 50, p. 161]
6
On the (ir)relevance of the flux-theory for Heraclitus's thought and possible in-
terpretations of his concept of apJ.lovia, see [56, introduction]; also [57, p. 193].
remains doubt about their true meaning. This perception of tradition

started with Plato and Aristotle, who saw him as the opponent of Hera-
clitus, teaching the non-existence of motion and change in reality, reality
being absolutely One, and being absolutely Being7 . This brings to mind
the paradoxes of Zeno of Elea [57, p. 263ff], according to tradition a dis-
ciple of Parmenides [45], Parmenides, pp. 204-206, 128(c-d)). The most
famous of them is the paradox of motion, a less famous one deals with the
possibility of there being separate things in the world: the paradox of the
One and the Many. Zeno's until now essentially unresolved paradoxes8
were intended to show that motion and separated things cannot exist,
precisely because their existence demands the coming together of being
and non-being at the same time on a certain place at a certain moment in
a world whose fundamental feature it is to exist. Non-Being is a contra-
diction in itself [93, p. 193]. The atomists' reaction to the Eleatic enigma
[57, pp. 407-408] and to our experience of change was the introduction
of non-Being in the form of the "void" . The void is nothing else than
non-Being, in which Being, in the form of a plurality of Eleatic Ones,
moves about 9 . They can stick together but not mingle with each other,
because of their Eleatic nature 10 . Change thus reduces to motion, but
the origin of this motion, and the paradoxical acceptance of the being of
non-Being, remain riddles unsolved. Because-from the Eleatic point of
view-in the real world everything is, eternally and indivisibly, it is im-
7
to EOV EC!'tt Being is. (actually it says: The Being is). [84, p. 45].
8
Dijksterhuis remarks sharply: "( ... ) we don't know how gravity manages to give
velocity to the body, and very often the given explanation comes down to pretending
to understand on the microlevel what should be explained on the macrolevel. But we
know since Zeno of Elea, that here lurks an essential difficulty of the conceptualisation
of motion ... " [34, p. 203, our translation]. See also the comments on modern attempts
to explain Newton's laws of gravitation by means of different kinds of bizarre particles,
the latest one being the theory of the graviton, by C. Piron in the text of a conference
given November 8, 1996 for CLEA in Brussels [78, p. 2].
9
to KEv6v "void" fulfills the role of non-being. That you can grant for the possibility
of motion by the introduction of the void was discovered by Melissus, a follower of
Parmenides, and of whom Leucippus, founder of atomism, was a disciple according
to tradition [57, pp. 397-398]. Furthermore ihoJ.tos, i.e., without 'tOJ.l~, without cuts,
unseparated in itself, undivided. It is important to see that this originally meant that
there are no "parts" in this Being between which there can be "parts" of non-Being,
the non-Being having a different character than the all-pervading, homogeneous void
or empty space of later times.
10
KtvOUV'tClt 'tE cmvex&s at a'tOJ.lOt 'tOV ai&va (... ) ~ 'tE 'tOU KEVOU <pucns ~ lhopisoucra bca-
(j'tllV aut~v touto 1tapamceuaset (... ) Les atomes se meuvent continument durant l'eter-
nite ( ... ) Car Ia nature du vide, qui separe chaque atome en lui-meme, produit cet
effet ( ... ). [29, in the 'lettre a Herodote', pp. 102-103]
possible to speak of something that is not in a way that makes sense. But,
as said before, the "contradiction" seen by classical philosophy between
Heraclitus and Parmenides is not necessarily a correct understanding of
the earlier "philosophies". One could as well infer that Heraclitus and
Parmenides do articulate the same world-experience, the former as the
experience of reality over a lapse of time, the latter as the experience
of the absolute reality of this moment (to understand better what this
means, try to deny by yourself you are experiencing yourself as existing
at this moment) 11 . This has nothing to do with the intellectual question
what it means to exist, or whether our existence is "real" or not. These
questions concern things "as such", objects, and their identity in past
and future. But this type of interpretation-which is the interpretation
of classical philosophy and of science, and which entails a representation
of reality outside of its actual and momentaneous experience doesn't
make sense, because for Heraclitus no things "as such" do exist, and for
Parmenides there is no motion, which implies that there is no time. It
is our conviction that, rather than revealing the contradiction between
the "thought systems" of the two pre-Socratic "philosophers", Plato's
interpretation reveals the difference between their world-experience and
what we think to be ours, constructed on the rational base laid down by
classical philosophy. The non-existence of metaphysical world views in
the pre-Socratic period is then due to a different kind of awareness of
one's being-in-the-world that characterized the transition from mythical
awareness to rational self-consciousness [54, pp. 67ff]. The hallmark of
this awareness is transparence for the stream of events that constitutes
the world [18, pp. 81ff]; there is no such thing as the separation be-
tween subject and object 12 . This separation precisely coincides with the
coming-to-be of rational self-consciousness. In philosophy, this change
will be codified in the metaphysical systems of Plato and Aristotle. The
process found its completion only in the early modern period, when ra-
tionalised self-consciousness developed the scientific way of observation
and explanation of the world. The problem of change (and motion as
a special case of change) and the problem of the existence of separable
objects in the world appear as two sides of the same coin. It is at the
origin of the principle of contradiction, formulated by both Plato
and Aristotle.
The contradicting conclusions deriving from pre-Socratic philosophy
11
Medieval thought knew this experiencing of experience as the nunc stans, the
"standing now". [10, p. 210]
12
"it is still the primary function of the noos to be in direct touch with ultimate
reality." [46, p. 52]
were of a major concern to Plato and Aristotle, because they challenged

the existence of truth and certainty about the world and therefore about
the actions of human beings in it. This uncertainty had given rise to a
philosophical discipline, Sophism, that simply denied any relation be-
tween reality and what we say about it ( [45], Theaetetus, 42, 152( d,e)).
Its subjectivism stems from a radical empiricism, which holds that things
are for me as I perceive them. But since reality as we perceive it is al-
ways in a process of permanent change this implies, as Plato points out
in the Theaetetus 13 , also the non-existence of stable, individual things
in the world. But then again this means that Sophism is nothing else but
an instance of the ontology of permanent change, already formulated by
earlier thinkers like Heraclitus. This is what, according to Plato, follows
for knowledge out of the ontology of universal change, and "everything
is equally true" reveals itself as the epistemological formulation
of the coincidence of opposites, which we met before as the base
of the universal change theory. On this soil the principle of contradic-
tion rests, because if you allow contradiction, you will be allowed to
say whatever: ex falso quodlibet ([45], Theaetetus, 150, 182(e)-183(a)).
But how to conceal this prohibition with our experience of permanent
change in the world? This will only be possible by stabilising human
world-experience in a world-picture, strong enough to survive the para-
doxical present into the past and the future 14 . Strong foundations must
be laid to grant the possibility to experience entities as objects outside
of the stream of events, and therefore to speak about them in an "ob-
jective" way. That is where, from our point of view, the true origin of
philosophy and in a later stadium science are to be situated. The main
problem for Plato and Aristotle can be described therefore as the con-
struction of a world-picture that would 1) grant the ultimate stability of
things, necessary as a solid base for certain knowledge, and at the same
time 2) allow for non-being, necessary for the existence of the change,
motion, manifold etc., that we experience by our senses. The common
feature of both their solutions was the division of the world in two
separated, though connected, layers: an unchangeable, motionless
'Parmenidean' or 'Eleatic' one which grants certainty about both ob-
jects and names, and a second 'Heraclitean' one, changeable and mov-
ing, which allows change and motion in the world as presented to our
13
Plato explicitly refers to Protagoras's "man is the measure of all things, existing
and non-existing". ([45], Theaetetus, 160(d,e))
14
The necessary condition that made possible this construction of stabilising world-
pictures or "world views", was the earlier coming-to-be of the "inner mind-space", in
which the non-present could be re-presented as present [54, p. 54ff].
senses. This feature of a two-layered world (a "world behind the world"),

is what makes their world views metaphysical ([10], One/ Thinking,
p. 23). In respect to this, we consider the differences between the two as
rather superficial. The ways they chose to achieve this are nevertheless
very different. They will be treated in more detail below.
3. THE SOLUTIONS 0 F PLATO AND ARISTOTLE
Plato's system displays an explicit two-world structure: the eternal

world of Forms or Ideas, and the world of changeable phenomena. The
relation between the two consists in the participation of the phenomena
in the Forms, the Forms granting them ultimate stability and knowabil-
ity. It is the hierarchy in the level of Being (in the level of reality) of
the Forms and the participation of things in the Forms (the participa-
tion theory) that allows for the possibility of an object to have contrary
properties without creating a contradiction between them that would
arise out of the simultaneous presence of Being and non-Being 15 . The
reasoning runs as follows: if there be a Form "Being" and another Form
"motion", then it is clear that "Being" must have a deeper, broader level
of reality, because everything that moves exists, but not everything that
exists moves. Something can be "non-moving" in the sense that it is
because it participates in the Form "Being", and is not in the sense that
it participates in the Form "Rest" and therefore doesn't move. Because
of the different level of reality of the Forms "Being" and "Rest", this will
not lead to the ontological impossibility of something not being on the
level of Being itself. This constitutes the ontological part of his system.
On the level of knowledge this is reflected in the structure of language.
Plato is the first to discriminate between the predicative and existential
use of the verb 'to be' 16 . This is possible because classes of concepts do
"mingle" the same way as classes of Forms ([45], Sophist, 400, 253(b,
c); id., 401, 253(d, e)). On the epistemological level, this discrimination
plays exactly the same role as the existence hierarchy of the Forms does
ontologically. It allows for the possibility to speak about a thing having
contrary properties without ending up in contradiction. Plato thus is the
first to formulate the principle of contradiction 17 . However, since his
epistemology remains fully embedded in his ontology, the principle fol-
15
Plato's elaboration for the Forms Being, Motion and Rest in ([45], Sophist, 388,
250(b,c)). A more general ontological formulation can be found, e.g., ([45], id., 392,
251(d); 413, 256(d, e)).
16
"Platonists who doubt that they are spectators of Being must settle for the knowl-
edge that they are investigators of the verb "to be". [72, p. 223]
17
Already in the Phaedo: ([45], Phaedo, 348, lOl(d, e)).
lows as a property of the way we can speak about the world directly out
of the participation theory. The principle of contradiction is the formu-
lation on an epistemological level of the participation theory 18 .
Plato's epistemology contains, because of this principle, and therefore
because of the existence of the Forms, an implicit logical structure [20].
The fundamental (and unproven) axiom on which his system rests is the
existence of the Forms proper. The stability of things and their knowa-
bility is granted by an essence (the Forms) existing before and apart
from them. The problem of motion gets its solution by means of the
degrees of reality that exist between the Forms mutually. In this way
he gives a foundation to the stability of the world and to its knowability,
without excluding properties like change and motion out of it. Thus he
escapes the pre-Socratic enigma.
Aristotle solves the problem of stability and knowledge-of stabil-
ity as a necessary condition for knowledge-in an at first glance to-
tally different manner. But he starts from the same premiss: that the
world, experienced as external, should be knowable as such, knowable
objectively-i.e., in its quality as a collection of objects- by a subject.
As said before, this separation of the world and the knower causes the
falling apart of the world in the stable, knowable, Eleatic noumenon, and
the unstable Heraclitean phenomenon, thus yielding a representation, a
metaphysical "world view" problematic vis-a-vis the changing reality
open to experience. The reasons for the Stagiryte's rejection of Plato's
system are the difficulties that are raised-from his point of view-by
the Form-ontology. The difficulties are threefold: the Third Man, the
unlimited number of Forms, and motion ([90], I, 990b(15)- 991a(8);
there also footnote c, and 991a(9ff)). Aristotle's objections reduce to one
major theme: the rejection of the Forms as stabilising essentials existing
separatedly from the things they instantiate. How then does he guaran-
tee the existence of stable things that can undergo change and motion
without allowing for the coincidence of opposites? And how does he
save the possibility to speak about them as being and non-being with-
out falling into contradiction? With him, and contrary to Plato, the
Eleatic and Heraclitean layers coincide in one world. A thing (object of
experience) is an essential form, a substance which realizes itself in
an undifferentiated material receptacle 19 , which can be seen as a sub-
stratum for existence, not as existence itself. Matter in itself has neither
individuality, nor quality. A thing consists of Form and matter at the
same time. Things therefore are not reflections of idealised Forms in
18
Epistemological formulation: [45], Sophist, 414, 257(a, b, c); id. 418, 258(b,c).
19
The unoKetJ.lEvOv. See [90, VII, 1028b(84)-1029a(34)]
a separated world, but instantiations of Form-termed substance 20 -in

matter. This instantiation or realisation is seen as a process in a course
of functional development that leads to a certain endpoint, and in this
sense as the "goal" to be achieved. How, then, does he explain and jus-
tify the nature of things in their coming-to-be and being? This he does
by introducing the theory of the four causes. These causes are not
to be interpreted in our strict causal sense; they represent the reasons
that make that something is what it is [90], I, 983a(24-34)]. They are not
causes but becauses21 . The material cause is the undifferentiated sub-
stratum for existence, in which the essential nature of things will find its
expression. The formal cause is exactly this essence, the thing's sub-
stantial nature. But because this evolution-this motion towards-takes
place in and through the material substrate, it is a process that can
never be completed. The efficient cause represents the influences from
the outside world that cause the process of motion towards realisation
of its true nature. The final cause is the endpoint of this realisation,
the completion of the transformation from potentiality to actu-
ality. Why does Aristotle separate potential from realised being? This
is the core of his metaphysical system, because the transformation from
potential to actual is his way of understanding motion. An entity that
has realised its substance doesn't change or move anymore. But this will
never be the case for a particular thing in the world, for a thing can
only be in motion in reference to something else. And since an endless
regression of causes of motion would be absurd, he postulates the exis-
tence of the truly actualised and therefore motionless Form, the First
and motionless Mover, God. An even more important point is that the
impossibility of the total actualisation of the things in the world follows
out of their being Form and matter at the same time: the entity is never
completely its actualised self, since matter remains indifferent towards
its evolution. The only truly actualised thing-true Form-is God. Here
we find the ontological ground for the contradiction principle:
an entity cannot be in the potential and in the actual state in reference
to one single other thing [90, IV, 1009a(24-39)]. This allows Aristotle to
unite the Eleatic and Heraclitean 'worlds' of his metaphysical system.
20
It will be noted that "substance" in the Aristotelian sense has nothing to do with
the material connotation that seems evident in our use of that concept. Aristotelian
reality on the level of matter is a continuum; it is the Form or substance that separates
Being from non-Being.
21
atno<;, cause, bears connotations different from the modern concept of causality.
Causal expressions in both the Platonic and Aristotelian sense would include (apart
from the 'strictly causal' ones): Why is this statue so heavy? Because it is made of
bronze. Why is he taking after-dinner walks? Because of his health. See [92, p. 134].
We bring to mind that his system was constructed this way to obtain
this result, to grant the possibility to deal with the apparent paradoxical
nature of 'real' reality: stability and motion should both be accounted
for. The world, thus stabilised ontologically, can now be made accessible
to thinking. How does he construct a framework for knowledge such that
a relation to this stabilised, but divided reality can be achieved? Here
we enter the vast area of Aristotelian logic. The principle of contra-
diction, based on the separation of being and non-being in the world,
can now be established as the basic axiom for correct thinking [90, IV,
1005b(8-34)]. Although Aristotle states explicitly its unprovability [90],
IV, 1005b(35)-1006a(16)], its introduction is justified in the framework
of his metaphysics, where the danger that it would cause the emergence
of a static, Eleatic world-picture, incompatible with our experience, had
been neutralised. The three fundamental principles of classical (Aris-
totelian) logic: the existence of objects of knowledge, the principle
of contradiction and the principle of identity, all correspond to
a fundamental aspect of his ontology. This is exemplified in the three
possible usages of the verb "to be": existential, predicative, and iden-
tical. The Aristotelian syllogism always starts with the affirmation of
existence: something is22 . The principle of contradiction then concerns
the way one can speak (predicate) validly about this existing object,
i.e., about the truth and falsehood of its having properties, not about
its being in existence. The principle of identity states that the entity
is identical to itself at any moment (a=a), thus granting the stability
necessary to name (identify) it. It will be clear that the principle of con-
tradiction and the principle of identity are closely interconnected. In any
way, change and motion are intrinsically not provided for in this frame-
work; therefore the ontology underlying the logical system of knowledge
is essentially static, and requires the introduction of a First Mover with
a proper ontological status beyond the phenomena for whose change and
motion he must account.
These different positions regarding the stable essence of things will
cause the Fight of the Universals, the question whether the substances
precede (ante re) or coincide with (in re) the things they instantiate.
During the Middle Ages this debate will give rise to a third possible
position: nominalism. It holds that substances (Universals) do not exist
except for our mental activity. But then the debate had already shifted
in a purely epistemological direction, while at its origin were mainly
ontological questions. That these questions even in the epistemologi-
cal treatment of nominalism don't disappear, becomes clear when one
22
With Aristotle, negation always is a secondary step in the process of reasoning.
GREEK T H 0 U G H T AND Q U ANT U M ME CHAN I C S 173
considers the difficulties each nominalist theory has to grant soundly for
the possibility to use general concepts, an indispensible tool for scientific
theory [19, p. 85].
4. RELATED C 0 N C E P T I 0 N S IN EARLY M 0 DE R NIT Y
It is a commonplace that the transformation from medieval to modern

science coincides with the abolition of Aristotelian metaphysics as the
foundation of knowledge. Not abandoned until the twentieth century,
however, was Aristotelian logic as a base for reasoning. Our aim until
now can be summarised as showing that the main principle of syllo-
gistic logic, the principle of contradiction, contains itself an ontological
rule. The rule is that, contrary to our daily experiences, intellectual
processes be standardised to remove change (and as a special case: mo-
tion) out of the world to assure the possibility of naming and classifying
unambiguously entities as objects. But the change and motion we expe-
rience in the world do remain. In the Stagiryte's system, the possibility
for change and motion was granted for exclusively on the ontological
level (the transformation of potentiality into actuality). Furthermore,
although Aristotle separated the disciplines of the theory of Being from
the theory of Reasoning (i.e., ontology from epistemology), we showed
above that the latter's basic rules and categories are ontological princi-
ples as well. Dropping Aristotelian metaphysics, while at the same time
continuing to use Aristotelian logic as an empty "reasoning apparatus"
implies therefore loosing the possibility to account for change and mo-
tion in whatever description of the world that is based on it. The fact
that Aristotelian logic transformed during the twentieth century into
different formal, axiomatic logical systems used in to-day's philosophy
and science doesn't really matter, because the fundamental principle,
and therefore the fundamental ontology, remained the same [40, p. xix].
This "emptied" 23 logic actually contains an Eleatic ontology,
that allows only for static descriptions of the world. From this point of
view, the debate during the Renaissance between the proponents of Aris-
totelian natural physics and the re-emerging corpuscular theories, can
be seen as a debate on how Aristotelian logic as the base of reasoning,
23
This "emptiness" is different from the 'emptiness of twentieth century formal
logic'. But the latter can be seen as a natural consequence of the former, since it was
developed to deal with logical problems that arise out of the ontological nature of
the rules of logic, as we hope to make clear further in this article. This applies, e.g.,
to Russell's "theory of types" (and to all theories based on "meta-reasoning"), that
can be read as modern (i.e., "purely epistemological") theories of categories in the
Aristotelian sense.
given its inherent ontological nature, can be brought into agreement with
the changing world of our senses24 . This viewpoint also implies that the
metaphysical structure fundamental to the older philosophical systems
actually remains present in science, be it in a different and, in fact, less
clear way.
Our aim is not to describe all subtle differences between the alternative
conceptions of the "real world" advanced at the verge of modern science;
this has been done by others in a brilliant way; e.g., in [34] and [50]. Our
concern now is to see whether it is possible to bring at the surface the
essential characteristics of a common line of reasoning which would al-
low to place science back into the philosophical development sketched
above, and to check whether this clarification of its fundamental con-
cepts sheds new light on present-day questions related to the enigma of
Quantum Mechanics. Our argument will lead us to the conclusion that
this is the case indeed: the ontological role of the Eleatic and Heraclitean
layers in the metaphysical reconstruction of reality is played in science
by the increasingly absolute conceptions of space and time, instead of
but necessarily correlated to the development of modern conceptions
of the nature of "matter". This becomes manifest in the "desubstan-
tialisation" of space and in the increasing parallelism between "space"
and "time". We will briefly consider the role of "experimentation" as
an observational practice designed to apply the ontological rule present
in the scientific way of reasoning on our world-experience, by chang-
ing "perception" into "observation". The formalisation of the empirical
component of the cognitive, "objectified" world-experience made it pos-
sible to bring perception into agreement with the ontological structure
of logical reasoning, not the other way around. Therefore the "episte-
mological" revolution brought about by science can be described not
as the abandoning of metaphysics, but as the complete absorption of
the metaphysical structure into the procedures of its formalised "oper-
ational" components, cognitive and empirical, of this "objective" (i.e.,
objectified) world-experience. The obscured relation in science between
"act" and "perception" then allows for the conception of logical reason-
ing as a representation of the ontological structure of reality and for the
successful application of science to the natural world: reality is adapted
to the ontological structure of science, not vice versa25 . The certain base
24
This of course is by no means to say that the debate was considered this way by
those who where implied. A brilliant exposure of the backgrounds of the debate be-
tween Galilei and the Jesuit scientists can be found in [82]. On Newton's backgrounds,
see [35, p. 55ff]. A more general discussion in [89].
25
The Greeks, whose intuitions about the relation man-world seem to have been
for knowledge is thus granted for in the most absolute sense, the circle is
closed: ontology is not a "problem" anymore. In this context the prob-
lem of the validity of knowledge takes on a new shape. Epistemology,
the critical commentary of the process of science, becomes more and
more the philosophical discipline, and can be considered as a discipline
or theory of knowledge not separated from, but "without" metaphysics:
it nourishes itself ontologically on science. The apparently sole prob-
lematic point concerns the relation of the "subject" with the objective
world-the debate between empiricists and idealists-a "question" that
for evident reasons never can be "solved" inside its framework: the re-
lation between knowledge and the real world that is implicitly supposed
here is replaced in science by the relation between knowledge and a re-
construction of the real world via an "empirical" procedure containing an
ontological rule that shapes the relation between the human "observer"
and reality on a much deeper level. The separation between subject and
object has by now been completed. Let us now see how early modern
science solved the problem of "refilling" logic with an ontology that al-
lows for a world of change and motion, and therefore for the description
of the world of our sensual experience.
5. THE S 0 L UTI ON S OF EARLY MODERNITY
At this point in our analysis our problem can be summarised as follows:

In the course of a process governed mainly by religious and societal con-
flicts, early modern natural philosophy emancipated itself from its Aris-
often more clear then ours, were sharply aware of the nature of technical interfer-
ence with the course of reality. One of the original meanings of the word J.llJxav~,
tool, technical device, is 'trick, deceit'. Whether this justifies their distrustful atti-
tude towards it is of course a different matter. This ancient clarity also holds for more
fundamental concepts. A quotation from von Fritz is worthwile: "( ... ) the concepts
of the "obscure" Heraclitus are all perfectly clear and can be very exactly defined.
In contrast, the empiricist Sextus, whose arguments seem so clear and easy to many
readers, has no clearly identifiable concept of either logos ("reason") or nous ("mind")
at all. Nous with Sextus is either identified with logos or considered a manifestation
of it. Logos, where Sextus speaks in his own name, is most often "logical reasoning"
or the capacity of logical reasoning ( ... ).But where Sextus reports the views of other
philosophers, logos becomes just the alternative to ais thesis ("perception"), whatever
this alternative may be, and so loses all clearly identifiable meaning. Yet it is highly
illustrative of the change which the concept of nous had undergone between
Heraclitus and Sextus that Sextus, in trying to explain Heraclitus's concept, begins
by connecting it with a term the preponderant meaning of which is "reasoning" and
ends by almost identifying it with "sensual perception." Heraclitus's own concept
of nous, as we have seen, was clearly distinguished from both but somewhat
more nearly related to the latter than to the former." [46, pp. 42, 43] Our boldtype
and translation of Greek terms.
totelian metaphysical foundation. Although indispensible for its correct

understanding, these religious and societal influences do not concern us
here; we will confine ourselves to an investigation of how this emanci-
pation was achieved, and by which intellectual means the problems that
arose out of it were solved by its main executors. Therefore we will look
into the altenatives that were formulated for the Aristotelian conceptions
that came under attack first, viz. of matter and motion, and how the rec-
onciliation with the Eleatic ontology, present unimpaired at the core of
Aristotelian logic, was granted for in a way that remains commensurable
with the phenomena of change and motion in the natural world. The first
step was taken by Galilei in an attempt to explain at first the behaviour
of light, later on of all manifestations of matter that are accessible to the
senses [82, chapter I], by re-introducing Democritean atomism as theory
of matter and of our perception of it. This implies a quantitative explana-
tion of the qualitative changes that we experience in the macroworld on
the level of the basic structure of material bodies; change reduces to mo-
tion. Following the same philosophical line, he established the description
of the physical behaviour of phenomena observed in standardised con-
ditions by way of mathematical laws, allowing for future verification 26 .
It was indicated before that atomism requires the introduction of a con-
cept of non-Being or "void", the precursor of "empty space", and the
indestructibility and incorruptibility of the "a-tomoi" if it wants to re-
main logically consistent. Equipped this way it would provide the perfect
metaphysical stuffing for logic's hidden ontology, if there didn't remain
some major deficiencies to deal with, viz. the total lack of understand-
ing of the origin of the motion that it makes logically possible, or the
question how the invisible atomoi constitute the sensible things in the
world. Nevertheless, in the context of our argument, Galilei's atomism
is of uttermost importance. Not only because it was the true reason for
his condemnation by the Church [82], but because the re-introduction of
corpuscular explanations for the nature and properties of material bod-
ies and the enigma of change and motion was a possible alternative to
the Cartesian ontology of the identity of matter and space. This
identity followed from Descartes's contention that extension and position
are the essential attributes of matter [80, pp. 46~50]. For Descartes space
is substantial: it is responsible for the fact that something is what it is,
exactly because of its position in it. He holds as well that the essential
26
Galilei's famous dual method, composed of Metoda risolutivo (analytical method
based on experimental data) and metoda compositivo (synthetical method, generalises
the principles found by the former and proves by prediction and verification that they
hold for the phenomena under study). [34, p. 259]
GREEK T H 0 UGH T AND QUANTUM ME CHAN I C S 177
characteristic of material bodies is their extension, and that is why in

his system matter and space ultimately coincide. In this way, Descartes
remained consistent with the ontological restrictions of the Stagiryte's
logical framework. But the impossibility to show that there can be things
with these primary qualities alone, makes his system untenable 27 . The
attempt to develop the possibilities of atomism to a sound philosoph-
ical base for the newly emerging science of nature was undertaken by
Pierre Gassendi 28 . His primal concern was a nominalist critique of
the concepts of both Aristotle and Descartes, a critique which can be
accomplished successfully only by replacing their logico-ontological cat-
egories by "physical" ones. Aristotle's space was to be rejected because
it was conceived as an accidens29 to substantial form, expressed in the
famous formula that it is "the number of place". Gassendi was con-
vinced that atomism provided a tool strong enough to overthrow once
and for all the Peripatetic doctrine of substances and qualities, while
avoiding Cartesian "absurdities" [16, p. 176]. Hence he radically turned
upside down the ontological hierarchy grounded by Aristotle but was-as
Descartes [80, p. 48]-carefully aware of the necessity to remain within
the ontological constraints of its logical formulation 30 . That is why he
27
"A good reason for this is the fact that we cannot measure primary qualities at
all unless we can perceive secondary qualities. ( ... ) This is the point of Berkeley's
argument that a material thing as conceived by Locke is an impossibility", the Lockean
"primary qualities" of matter being nothing else than the specification of Descartes'
concept of extension. [80, p. 49]
28
The importance of whom for the development of key-ideas in early modern science
is until to-day heavily underestimated. It doesn't surprise us much, however, that
Schri:idinger's finetuned philosophical intuitions recognised this already decades ago.
See [87, p. 75], also [50, pp. 34, 92-94]; [16, p. 174].
29
In the Peripatetic sense, a quality that cannot exist apart from the substance to
which it belongs. We would say a "property", not from a chunk of matter, but from
an essential Form.
30
Gassendi's work presents us with one of the rare instances that reveals, both by
content and structure, explicitly the metaphysical nature of the basic categories of
modern natural science: "C'est !'idee qui nous parait ressortir de Ia presentation de
Ia premiere partie de Ia Physique dans le Syntagma, le De Rebus Naturae Universe.
Cette 'Physique' fait immediatement suite a Ia 'Logique' par quoi commence l'ouvrage
( ... ). II ne s'y trouve pas en effet de 'Metaphysique', et Gassendi s'en explique des le
debut: ce n'est pas que Ia metaphysique soit sans objet, ou inaccesible, c'est qu'il n'y
a pas de distinction entre physique et metaphysique ( ... ). Ce sont done bien
des categories physiques qui prennent ici Ia place de !'ontologie aristotelicienne, en
meme temps qu'elles re<;oivent un contenu oppose a celles de Ia physique d'Aristote.
L'atomisme sera la realisation adequate d'un tel projet, mais !'on a vu que
celui-ci apparait ( ... ) a partir de Ia critique des 'formes substantielles', apporter une
nouvelle conception du "mouvement nature!", ressusciter 'l'espace des Anciens' contre
did not content himself with the reformulation of the "void" of classical
atomism, as it leaves the fundamental question of the origin of motion
unanswered. The introduction of an atomistic explanation for the nature
of material bodies and their properties required a different conception
of space as well, for abolishing the categories "substance" and "quality"
causes the downfall of the ontology of "potentiality" and "actuality",
and therewith of the possibility to grant for change and motion. Space,
infinite, divisible at infinity and indifferent to its material content [16,
p. 179] is the seat of the stable individuality of things. Space con-
stitutes the true Eleatic layer in the metaphysical set-up of
natural science: things take position in it while it remains immobile
and identical to itself. Space proceeds the material existence of things
and is a necessary condition for it, not vice versa; it fulfills the same role
as the substances with Aristotle. Atomism, Eleatic by nature, provides
a non-qualitative explanation of the composition and motion of mate-
rial entities in it. Gassendi's real coup, however, is the establishment of
the rigorous parallelism between "absolute space" and "absolute time".
The conclusions arrived at concerning the ontological status of space will
now be applied to time. Time is not the "number of motion" as with
Aristotle. It is motion that depends on time, not vice versa. Time con-
stitutes the true Heraclitean layer in the metaphysical set-up
of natural science. Gassendi thus reaches the completion of his quest
for sound metaphysical foundations for the new natural science: change
is motion, while space is the modus of existence of things permanent,
exactly the same way as time is the modus of existence of things suc-
cessive31. Here we find another fundamental reason for his rejection of
the Cartesian substantiality of space: nobody would defend the substan-
tiality of time. The reality of space and time is extra-substantial and
extra-accidential. They have dimensions and/or properties, but, being
incorporeal, no functions nor qualities [16, pp. 177-181]. Therefore the
dimensions of space can coincide with the dimensions of matter without
causing any interference. All this clears the road for the theoretical de-
le 'lieu Aristotelicien', retablir le 'vide' dans Ia Nature, proposer une nouvelle notion
du "Temps" etc." [16, pp. 172-173]. Our boldtype.
31
"II faut en effet, comme il le dira plus loin dans le Syntagma a propos du mou-
vement, faire une distinction radicale entre le mode d'existence des 'choses
permanentes' et celui des 'choses successives', distinction a laquelle cor-
respondent respectivement l'espace et le temps. (... ) Espace et temps sont
infinis, l'un selon les dimensions, !'autre selon Ia succession (... ). Espace et temps
ont des 'parties inepuisables', d'ou Ia contingence de Ia situation du monde hie et
nunc. Espace et temps sont enfin inalterables et invariables quel qu'en soit le contenu:
... l'espace reste identique et immobile, comme le temps s'ecoule toujours
de meme maniere." [16, p.179ff], our boldtype.
G R E E K T H 0 UGH T AND QUANTUM ME C HAN I C S 179
velopment of the mathematical description of the behaviour of material

objects in space and time in terms of their positions and velocities, as
had been initiated already by Galilei. The fundamental paradox however,
although neutralised again, did not completely disappear. The indivis-
ibility of the atom and the infinite divisibility of space somehow break
the strict mutual correspondence between their proper dimensionalities,
a necessary condition for their ontological complementarity. The infinite
divisibility of space and time 32 implies (following Gassendi) continuity.
And continuity, that key-concept to classical mechanics, was the well
of which the problems that later led to QM sprang33 . For Newton,
Gassendi's picture provided the perfect ontology to solve the problems
inherent in the mechanical description of nature 34 . However, Newton re-
alised very well that the force he introduced 35 to explain motion needed
itself an explanation [28, p. 40]. Already when writing the Principia he
searched for the explanation of the phenomenon the effects of which he
adequately described therein. For this explanation he turned, as so often,
primarily to the Ancients. Not without disappointment he notes that we
do not know of any solution provided by them [35, p. 57]. After various
attempts to develop a sound world view on the base of an "aether" that
would allow his "Force" to be transmitted mechanically [35, pp. 57-58],
he was forced to admit its impossibility, because the effects of such an
aether were not observed on planetary motion. Reluctantly 3 tl but con-
sequently, he introduces "a most subtile Spirit pervading every-body",
accounting for both inertial force as the cause of uniform straight-line
32
[69, p. 3] Although we totally agree that "extension" and "divisibility" are at
the origin of the present-day QM-problems, we think it established and not without
relevance for eventual remedies that these notions are themselves not primitive in this
respect, because they follow from past attempts to come to terms with the paradox.
33
"Our helplessness vis-a-vis the continuum, reflected in the present difficulties of
quantum theory, is not a late arrival, it stood godmother to the birth of science" [87].
34
He uses almost the same wording as Gassendi: "I. Tempus Absolutum, verurn,
& mathematicum, in se & natura sua, sine relatione ad externum quodvis,
aequalibiter ftuit, alioque nomine dicitur Duratio ( ... ) II. Spatium Absolutum,
natura sua sine relatione ad externum quodvis, semper manet similare &
immobile( ... )" in [71, p. 6] (Scholiurn to the Definitions). Our boldtypc. Reference
to this also in [78, p. 2]. "Tempus Relativum" and "Spatium Relativum" are the tools
that allow us to measure and describe motion in concreto, thence they are easily
confused with 'real'-i.e., absolute-space and time.
35
"What he proposed was an addition to the ontology of Nature." Westfall, R.S.,
cited in [28, p. 51]. Also [50, p. 101].
36
In a letter to Bentley, 25 Feb. 1693, Cited in [28, footnote 42, p. 52]
motion and gravitation as actio in distans 37 . Force thus turns out to

be a by necessity immaterial instantiation of space. Not only is
force a physical entity with a proper ontological status [32, p. 229], but
to bridge the abyss between material and immaterial, between Being and
non-Being, this status has to go beyond the categories of existence of the
phenomena it should grant for [63, p. 125]. It is crucial to realise that
Newton needed the co-incidence of God and space to account for gravity,
in exactly the same way Aristotle needed the First Mover to account for
motion 38 . Post-Newtonian classical physics, however, after being purified
by eighteenth century enlightement of all such "superfluous" hypothe-
ses, disguised by equating force and acceleration the "causal paradox"
that is the consequence of its metaphysical circularity [32, p. 228]. The
persistent problem is the impossibility to prove the existence of an ab-
solute "frame of reference" because the immaterial parameters of space
and time escape, at least in principle, experimentally controllable obser-
vation. It is exactly this ontological status beyond that makes "similar
and immobile" space to the only warrant for the repeatability of ex-
perimental observations [50, p. 84], which is another way to highlight
its Eleatic character, and its enigma as well. The essentially Leibnizian
approach of what was called later "analytical mechanics" seemed to de-
liver an acceptable-because mechanical-alternative, while allowing at
the same time for the treatment of similar classes of physical problems
as "Newtonian" or "vectorial" mechanics [58, p. xxi]. The principle of
least action presents a description of mechanical systems by minimizing
a quantity that measures the "action" of the system under consideration
as if it were a single particle moving in a plenum, the particles of this
plenum remaining mutually separated, but being always in direct contact
with each other. The mathematical "configuration space"' represents
space as if force were the 'real' manifestation of its n-dimensional geom-
37
In the second edition of the Principia, which dates 1713: "Adjicere jam liceret
nonulla de Spiritu quodam subtillissimo corpora crassa pervadente, & in
iisdem latente; cujus vi & actionibus particulae corporum ad minimas distantias
se mutua attrahunt, & continguae factae cohaerent; ( ... ) Sed haec paucis exponi non
possunt; neque adest sufficiens copia Experimentorum, quibus leges actionum hujus
Spiritus accurate determinari & montrari debent. [71, pp. 173-174] Our boldtype.
38
"( ... ) it is possible to see Newton's ideas as the "fruition of a long tradition" ex-
tending from Aristotle through Newton, a tradition in which Aristotle's finite plenum
was slowly and by painful steps converted into the void, infinite, three-dimensional
framework of the physical world required by classical physics. Newton's God-filled
space was the penultimate development in the process by which concepts of space were
developed by attributing to space properties derived from the Deity; after Newton's
time, the properties remained with the space while the Deity disappeared
from consideration." [35, p. 60] Our boldtype.
GREEK THOUGHT AND Q U ANT"C M MECHANICS 181
etry [58, p. 13]. But this at first glance purely formal approach hides the
Cartesian-Leibnizian ontology of the unreality of space, with Descartes
because space coincides completely with matter as pure extension; with
Leibniz because space is nothing but the relations between the objects
'in' it. That the basic quantity describing with Leibniz the dynamical
behaviour of the system, the vis viva or living force 39 , is represented by
a scalar is clearly of more than merely formal importance, it is a conse-
quence of his ontological position. This annihilation of non- Being~the
vacuum, empty space~from physics opens the possibility to treat space
as a merely "relativistic" phenomenon, a position fully adopted later by
Einstein in the context of his Special Relativity 40 . The problem of the
initial origin of forces or motions remains equally unsolved. Even worse,
analytical mechanics lacks a 'natural' way to provide for the stabilis-
ing frame of reference for experimental observation as is present in the
explicit ontology of the Newtonian treatment. Appearances are saved
by stating that the universe presents the same aspect from every point
(apart from "local irregularities") [50], but in the ontological setting of
the logical framework~i.e., the framework of the separation between
Being and non-Being, between system and environment, between cause
and effect~this can be upheld only by an act of Divine creation, as is
the case with Leibniz, but of course not with his followers in the later
school of analytical mechanics. Saying that the notion of absolute space
is redundant because the natural laws are invariant under coordinate
transformations [48] is turning upside down the chain of justifications.
As should be clear, the whole metaphysical set-up of natural science is
such that they should be so. That the role of "causal black box" played by
space is now fulfilled by the "environment" 41 , doesn't change anything
to the fundamental circularity of the reasoning.
39
Almost identical with our kinetic energy. Together with the "work of the force"
or potential energy the two fundamental scalar quantities on which the study of
equilibrium and motion of analytical mechanics rests [58, p. xxi].
40
"The introduction of a "luminiferous ether" will prove to be superfluous inasmuch
as the view here to be developed will not require an "absolutely stationary space"
provided with special properties, nor assign a velocity-vector to a point of the empty
space in which electromagnetic processes take place." [37, p. 38].
41
"What is system, for instance, is described by phases or states; environment is
not, and cannot, be represented in such terms. Rather, environment is the seat
of "external" forces ( ... ) This apparently necessary and innocent partition of the
world into system and environment, ( ... ) has the most profound consequences for
the notion of causality. For according to it, the notion of causality becomes bound
irrevocably to what happens in system alone, ( ... ) the state-transition sequence.
( ... )what happens in environment has thus been put beyond the reach of
causality. Environment has become acausal" [85, p. 19]. Our boldtype.
6. THE RE- EM ERG EN CE 0 F THE PARADOX:

LATE MODERNITY
The depth of the problem has been brought again to the fore by the
results obtained both theoretically and experimentally by QM. For not
only does the theory incorporate states for quantum entities that imply
non-local behaviour [8], but such effects have by now been established
incontrovertibly in various experimental settings, thus shortcutting for
once and for good eventual attempts to explain them away statistically42 .
Furthermore it was proved to be impossible to describe soundly two or
more separated systems within the present formalism [3]. As is suffi-
ciently known, the state of a physical system in QM is described within
mathematical Hilbert space. For every physical entity the collection of
all its properties constitutes its state in Hilbert space. Exactly in the
same way as the state of a classical entity is represented in phase space,
the Hilbert space contains all possible states the system could have. In
this sense, both phase space and the Hilbert space represent the envi-
ronment in which the dynamical transition of one state to another of the
entity will take place. To avoid that the individual state be blurred in the
statistical ensemble proper to the 'orthodox' interpretation, C. Piron in-
troduces a new notion of physical state based on the concept of property
as related to experimental projects with well-defined certain results [75,
p. 398]. These results are called yes-results, while any other outcome is
considered to be a no-result. To this end, a test that could eventually be
performed is associated to a property of the system under consideration,
42
C. Piron in [76]: "... the orthodox QM (the 'new testament' as Pauli named it)
with its credo and its principles is dead and definitively dead." A lot of acrobatic
attempts were made to rescue the modernist picture, by extending the quantum pic-
ture with an underlying modernist kind of world (for example D. Bohm [17]), or by
rejecting it (for example A. Einstein [36] and J. Bell [13]). But at the same time,
every advance on an experimental level (in particular A. Aspect [11] and H. Rauch
[81]) confirmed that these attempts were bound to fail. The general community of
physicists reacted to this by a kind of ontological ignorance attitude which evolved
towards a pragmatic quasi botanic empiricism, or by some new age like mystification
tendencies (see for example 'The Tao of Physics' by F. Capra). Alternatively, a not
unimportant group of people saw the inadequate arsenal of mathematical (and in par-
ticular logical) tools as the origin of all these problems (especially J. von Neumann &
G. Birkhoff [15], J. Jauch [52], C. Piron [73], H. Neumann & G. Ludwig [60], [61] and
D. Foulis & C. Randall [42]). This gave rise to an enormous development of mathe-
matical attributes and structures, of which quantum logic is the most known among
philosophers of science (see for example [14] and in particular [65]). Unfortunately,
the mathematical expertise required made the development and study of quantum
logic an essentially mathematical occupation such that the conceptual development
stagnated (a confirmation of this fact can be found in [66] and [77]).
G REEK T H 0 U G H T AN D QUANT U M ME CHAN I C S 183
in such a way that, once you are certain in advance to obtain the desired
result, you can assign to the system an element of reality [7 4, p. 170]
as conceived by Einstein [38, p. 137]. Introducing Aristotle's dynamical
terminology, a property is called "actual" when the result of the test
is certain; when uncertain it is said to be "potential" [7 4, p. 171]. A
property conceived this way is always potential or actual. It is, however,
possible that the property tested by an experimental project, and the
one tested by the experimental project obtained by exchanging the "yes"
and the "no" results of the former, are both potential. Whenever this sit-
uation appears we are dealing with quantum-like entities, contrary to
classical ones where this can never be the case. But it is not a priori ob-
vious how to define a particular entity and thus how to assign properties
to it. All comes down to account for the set of tests which matches the
collection of its properties and thus the entity itself. This presupposes
the possibility to separate the phenomenon under consideration from the
rest of the universe [31]. But, although it is stated explicitly that the cer-
tainty of the experimental project is an objective feature of the system
without reference to our knowledge or beliefs [69, p. 5], we are supposed
to know in advance what to do with the system in order to get this cer-
tain result. Indeed, the system should be prepared in a precise way that
is related to our existing knowledge of its properties. This presents no
problem when we are dealing with phenomena that are accessible to our
daily experience, like with the breaking of a piece of chalk [76, p. 208].
The necessary assumptions about its properties are then given by our
daily experience. This enables us to drop the preparation procedure that
is essential to 'true' experimental observation. Experimental observation
is indeed more than the "contemplation of the astronomer" [69, p. 8], it
is an intervention that prepares the system in such a way that we obtain
an entity-separated from its environment-together with a set of prop-
erties determined by the possible yes-no experiments. Again, it is the
preparation of the system that moulds the 'real thing' so as to fit into a
definite metaphysical set-up, primarily by separating "the system" out
of its "environment". Whatever be the structure of the set of possible
answers, the modelisation of the set of possible yes-no experiments im-
poses on the set of possible properties the mathematical structure of a
complete lattice [53, p. 844]. This means, as indicated before, that the
"real thing" is forced to fit into the scheme of an intrinsically
Eleatic ontology. The preparation procedure is a 'black box'
generating that ontology. This does not imply that it is senseless
to perform experimental projects. Natural laws are verified by experi-
mental observation, and experiments are performed in reality, so there
must be some kind of agreement with what happens in reality. But it is
precisely the kind of relationship that exists between reality and experi-
mental observation that should be rendered more clear; this implies that
the consequences of the fact that each experimental project is an inter-
vention in reality should be itself subject of investigation. By making ex-
plicit the structures underlying the sets of possible yes-no questions and
of possible answers, the Geneva School approach realises an important
progress in this respect, though until now the ontological aspect of the
problem has been left untouched. Embarking on an analysis of this as-
pect could nevertheless shed some light not only on the relation between
the "entity" and the "real-world phenomenon", but also on the status of
our conceptions of space and time, which form, as has been argued, the
ontological foundation for the metaphysical set-up present in scientific
theory. This provides the background against which Piron's attempts to
get hold of a "realistic" conception of space are to be seen. The inevitable
re-appearance of ontological circularity to which also these attempts are
subject can be clarified on a more theoreticallevel 43 . The tacit assump-
tion underlying the position that a thing coincides with a collection of
properties is that "being" in the sense of "existence" is but a property
amongst the others. We discussed before at length why this assumption
is untenable in the context of an-implicit or explicit-Eleatic ontology
as the one present in the cognitive and empirical procedures of science.
The notions "object" or "entity" necessitate the introduction of Being
in some form, ontologically beyond the object's predicable properties, so
as to grant for its required stability and separability, as well as for an
origin for the dynamics governing the processes of change and motion
to which it is exposed. In this ontological setting to be is not a prop-
43
A specific discussion of this problem with respect to Piron's approach of space
and time will be the subject of a future publication. Equally problematic but outside
our scope would be an analysis of the concept of 'field', arising out of General Rel-
ativity. Be it sufficient to refer to Schrodinger, who explains: "at any rate the very
foundation of the theory, viz. the basic principle of equivalence of acceleration
and a gravitational field, clearly means that there is no room for any kind of
'force' to produce acceleration save gravitation, which, however, is not to
be regarded as a force but resides on the geometry of space-time" [86, p. 1).
This 'making real' of the geometry of space as the 'force' behind dynamical processes
in four-dimensional space-time reveals more clearly the underlying ontology and links
General Relativity firmly to the 'Leibnizian' tradition within physics. But this con-
tinuum, in which gravity manifests itself, is itself the source of a host of ontological
problems. Regarding this, Piron cites in his lecture Einstein saying "that according to
general relativity space is endowed with physical qualities and in this sense
an ether exists ( ... ) but this ether must not be thought of as endowed with
the properties of ponderable media ( ... ) nor may any concept of motion
be applied to it" [79, pp. 1~2]. Our boldtype. By trying to explain his own concept,
Einstein reveals how close Newton and Leibniz actually stand.
GREEK T H 0 U G H T AND Q U ANT U M MECHANIC S 185
erty. It follows that existence is not a predicate [80, p. 36] and existence
itself is not within reach of experimental preparation, nor observation
[50, p. 84]. Regarding the paradoxical results attained by QM, we then
are led to the conclusion that only the development of an explicit and
properly ontological foundation for basic categories as 'entity', 'motion'
as well as for the intervention represented by their experimental obser-
vation remains open as a fruitful strategy. A sketch of a possible further
elaboration of the results attained thus far by the Geneva School with
regard to the foregoing analysis is presented in the last section of this
paper. Alternatively, leaving behind the principle of contradiction as con-
stituting ontology at the core of scientific procedure both empirical and
cognitive, could be taken into consideration. Interesting steps to come
to terms with problems related to this approach have been made in the
recent past. From attempts to develop logical systems that escape the
rigorous consistency demanded by the principle of contradiction sprang
relevant considerations, but these remain trapped within the framework
of the ontological core common to all logical systems44 . Another ap-
proach recognises the ontological nature of the problem, but limits itself
to the construction of ontologies for "possible worlds" that fit "a" logic
in which the principle of contradiction has been "neutralised" somehow
[83]. Since all logics still do partake in the same fundamental principle,
these "possible worlds" necessarily remain Eleatic. Our position would
be that the origin of the problem resides within the ontological nature
of reality itself This viewpoint might clarify the incurable presence of
paradoxes in logical systems of all kinds. Paradoxes then would appear
because of an ontological incompatibility between logic and "real" real-
ity. Therefore our approach would be to take paradox itself as the starting
point for the construction of an ontological framework. We see the work
of Spencer Brown [88], and, more recently, Kaufman 45 as supportive
in this respect. That this strategy might open relevant perspectives for
QM is indicated by the recent discovery of Aerts et al [7]. The principle
underlying such ontology has been adequately formulated in the past.
It is the already mentioned principle of coincidence of opposites, to be
found in the work of fifteenth century philosopher Nicolaus Cusanus. It
provides an excellent example to show that abandoning the principle of
contradiction implies the loss of neither the capacity to reason soundly,
nor the possibility to use mathematics [30, viz. Capitula XII to XVII]. It
does imply, however, the necessity to abandon on a conscious level the ar-
44
"Every logician in the end divides propositions into those which are acceptable
and those which are not." [40, introduction]
45
Oral communication. See also [55]
tificial separation between "subject" and "object", the distance between

things being a mere aspect of their instantiation. As an indication for
its operational viability we suspect the possibility to consider circularity
and self-reflexivity as tools rather than as problems. Another interesting
possibility might be the development of a sound conception of "physical
space", homogeneous from all points, because circumference and centre
coincide in all points, and without the need to introduce a force-like
"First Mover", since distance and interaction reveal themselves as the
same thing. This again might add to opening up interesting perspectives
on the "non-locality" of elementary physical phenomena manifest at the
core of QM.
7. CONTEMPORARY PHYSICS AS PRODUCTS OF

EARLY GREEK THOUGHT:
POST-MODERNITY AND BEYOND?
Let us now concentrate on one of the words in the title which might
seem rather innocent and obvious, namely 'products'. In a first reading
of this title one could think that it might easily be replaced by 're-
sult' or 'consequence'. We nevertheless attach a very definite meaning to
this word 'product' which is definitely not covered by 'result' nor 'con-
sequence'. In contemporary mathematics, which deals with the study
of families of well-defined formal structures (or, in a more advanced
language, 'categories of mathematical objects' 46 ), products of two struc-
tures that belong to the same family (i.e., two mathematical objects that
belong to the same category) are usually defined as a third structure in
this same family which has the two given structures as 'faithful' sub-
structures (in categorical language this means that there exist structure
preserving maps, called 'morphisms', from the two given objects into
the product object). In fact, this means that the product is larger than
the given structures, but essentially not different. This way of defining
products is an obvious consequence of the specific 'scientific method' in
the context of which mathematics is used: one only wants to deal with
objects that are well defined. Once one has defined such a well-defined
collection of objects, one sticks with it and studies it. Even more, an
object has only a meaning as an element of this well-defined collection.
As a consequence, if one wants to introduce operations on the objects in
46
For an outline of the mathematical theory of categories introduced by Eilenberg
and Mac Lane [39] we refer to [12], [62], [64]. In fact, some approaches and authors on
category theory see a possibility of introducing 'aspects of undefinedness' within this
formalism, and consider this aspect as the main argument to apply category theory
as a foundation for mathematics.
this collection in order to study and characterize it, one remains faithful
to this collection, i.e., one does not leave it 47 . This situation in mathe-
matics had a major influence on the way how people tend to describe
compound systems in physics: for a given description (i.e., structural
characterization) of individual physical entities one is still attached to
the idea that the compound system should be described within the same
family of structural characterizations. Within the context of Newtonian
mechanics, this does not pose a problem. Within the context of QM,
however, one is able to produce a description procedure for the joint
system, but looses completely the mathematical consistency48 required
for a 'good theory'. Unfortunately, by yielding the conclusion that the
'Hilbert space structure', the structure in which QM-entities are de-
scribed, is not 'the good universal structure' for their description, these
results have only fostered the search for 'bigger' and 'bigger' structures
in which one could hope to find 'consistent products' for the description
of the compound system. Never the a priori idea of universality has been
put into question. According to our previous sections, we think that this
specific attachment to universality of a description points the finger at
the real problem at the source of this malaise concerning the descrip-
tion of compound entities in physics. The presence of a second entity in
a compound system definitely changes the context of the first one, and
as such, a description that incorporates a contextual ingredient should
take this presence into account in an explicit way 49 . Unfortunately, for
the QM formalism, this is not the case 50 based on ideas developed by
Aerts, Gisin and Piron in [4], [5] and [47], and yielding the framework
of [5] and [21], shows that in the traditional QM description, aspects
like the order in which one performs consecutive measurements on dif-
ferent individual pseudo-entities within the compound system are even
not a part of the context, but an ingredient of the formal representa-
tion of the entity itself.. Solving this problem comes down to starting
from correct collections of primitive notions in order to reconstruct the
47
We have to remark that quite recently, since the development of the mathematical
theory of categories, one became very interested in 'relating' the different well-defined
collections of mathematical objects. However, one is still attached to the same kind
of scientific method, but now on a meta-level.
48
This inconsistency is encountered as well on a purely structural level, in the sense
that there exist two unequivalent procedures for obtaining a so called tensor product
[2], both when one starts from an operational or empiricist point of view [3], [77], [91].
49
For the reasons stated before, this could considerably enhance the coherence of
the theory.
so
An explicit construction in which one of the present authors tries to take the
context into account for a situation of 'the many' (see [22], [27]
proper formalization for the concept under consideration, in this case

the concept of compoundness51 . With the goal to aim to an as reduced
as possible reformulation of the formalism that arises when one starts
from as few as possible primitive notions, D.J. Moore started a catego-
rization program52 for the foundations of physics (see [67], [68],
[69]). Within this scheme, other primitive notions have been incorpo-
rated (see [9], [23], [24], [25], [27]) yielding a categorical description of
the concept of compoundness (see [27]). Without falling into the trap of
empirical fragmentation 53 , it seems to be possible for this concept to skip
the universality principle which has always been the a priori for theoret-
ical physics. One could wonder whether other physical concepts could
be treated in the same way? A conclusive question? When dealing with
'questions' we are confronted again with the problem of definedness. Old
and endless debates on the yes or no necessity of explicit representations
of 'states' of reality could definitely be omitted by allowing aspects of
undefinedness within the explicitation of a concept such as states and
experimental projects; quoting Moore on the latter in [69, p. 3]: "I shall
take a particular physical system to be some part of the ostensively exter-
nal phenomenal world, supposed separated from its surroundings in the
sense that its interactions with the environment can either be ignored
or effectively modeled in a simple way. The restriction to a reasonably
circumscribed aspect of a phenomenon is crucial: physics is, first and
foremost, the study of detailed models of specific situations ... I shall
take a definite experimental project relative to a physical system to be a
real experimental procedure where we have defined in advance what would
be the positive response should we perform the experiment ... Of course
the specification of a precisely delimited positive response is also an ide-
alization: in general terms, all that we require from an assignment rule is
that it be sufficiently clear in each case whether or not one should assign
the response 'yes', this within subjectively reasonable limits with respect
to borderline cases"; where he motivates the necessity of consecutive and
complementary 'idealizations' by: "... the problem of linguistic demarca-
tion in the face of vagueness or contextuality." As it is very reasonable to
consider human creation as an ability to 'invent' new objects outside an
a priori given class {xj¢(x)} with a characterizing predicate form¢, the
collection of possible designable experimental projects cannot be a class
51
A more detailed discussion on this matter can be found in [69].
52
This program, when consistently developed in the total framework of the theory,
could amount in an explicit Aristotelian ontological set-up, exactly in the sense Piron
[74] intended.
53
Which occurs as well on a theoretical as on an operational level.
GREEK T H 0 U G H T AND Q U ANT U M MECHANICS 189
and in particular, cannot be a set. To quote Moore (private communica-

tion, 27-09-97): "... this boils down to a kind of object-subject dichotomy:
states and properties are about the exterior world and definite experimen-
tal projects are about the internal world. Building physical theories is a
fallible attempt to relate one to the other ... " Still, any operational the-
ory attempts to remain unaware, by formal ignorance, of these possible
creative acts consisting of designing definite experimental projects. This
ignorance could be motivated by the ever ruling set theoretical basis for
mathematics. In order to understand the problems that arise, one could
consider entities like my past or the entity's past, also studied in [25].
It is obviously impossible to define a predicate ¢ that characterizes all
my possible pasts that I can have on a next instance of time, since this
would require Laplace's supreme intelligence: having all knowledge on
possible events that might be imposed by the context between now and
the considered next instance of time. Contrary to the existence of a set
of states as a criterion for the reality of an entity, the existence of prop-
erties as reference to the elements of reality in the Geneva approach [3],
[67], [69] can be the starting point for a more flexible setup: the entity
'my past' for example can be described by consecutive creations of prop-
erties, themselves not being part of a predefined set. This idea has been
generalized and taken as a starting point of a formal scheme referred to
as the Induction 54 formalism, which takes in an a priori (and not only
in an explanatory) way 'creation' into account, as well 'human' creation
(choice/invention) as 'material' creation (mutual Induction as interac-
tion), and of which a complete technical development can be found in
[9], [25], [26], [27]. This approach succeeds in avoiding the aspects of
isolation in the definition of an entity: the 'particular physical system'
fuzziness to which Moore refers is explicitly not present. As such, this
strategy aims at the elaboration of the results of the Geneva school in
the direction of a formalism that is explicitly compatible with the results
of the foregoing ontological analysis. Formally as well as conceptually,
such a theory of creation, by constituting the description of the emerging
properties of the entities as possible parts of compound systems, treats
"interaction with the context of an entity as a part of a measurement
process [4], [5], [21], [23]" and "entanglement of individual entities [22],
[27]" on the same level, i.e., in terms of mutual Induction of properties
[9], [25], [27]. We finally want to remark that the idea of going beyond de-
finedness, going beyond set-large collections of states, seems to be most
conveniently expressed through the use of so-called quasi-categories [1,
54
Here, 'induction' does not refer to any of its usual philosophical significances, but
rather refers to physical theories like electro-magnetism.
p. 31], a generalization of the above mentioned mathematical categories,

of which the undeniable necessity thus emerges from both physical and
metaphysical considerations. To quote Hersh in [49, p. 22-23]: "We can
try to describe mathematics, not as our inherited prejudices imagine it
to be, but as our actual experience tells us it is . ... What are the main
properties of mathematical knowledge, as known to all of us from daily
experience? ( 1) Mathematical objects are invented or created by humans.
(2) They are created, not arbitrarily, but arise from activity with already
existing mathematical objects, and from needs of science and daily life"
8. ACKNOWLEDGMENTS
We thank Sonja Smets, Hubert Dethier, Wilfried Van Rengen and Rudolf
De Smet for discussing the content of this paper. We thank David Moore
for the indication of problems and references connected to the subject
of this paper. We thank Didier Durlinger for technical help while imple-
menting the ancient Greek font. Bob Coecke is Post-Doctoral Researcher
at Flanders' Fund for Scientific Research.
AFFILIATION
Karin Verelst
Center Leo Apostel
Brussels, Belgium
Bob Goecke
Center Leo Apostel-FUND
Brussels, Belgium
REFERENCES
[1] Adamek, J., Herrlich, H. and Strecker, G.E., Abstract and

Concrete Categories, J. Wiley & Sons, 1990.
[2] Aerts, D. and Daubechies, I., Helv. Phys. Acta, 51, 1978, p. 661.
[3] Aerts, D., The One and the Many, PhD-thesis, Free University of
Brussels, 1981; Found. Phys., 12, 1982, p. 1131.
[4] Aerts, D., J. Math. Phys., 27, 1986, p. 202.
[5] Aerts, D., Found. Phys., 24,1994, p. 1227.
[6] Aerts, D., Coecke, B. and Smets, S., "On the Origin of Probabili-
ties in Quantum Mechanics: Creative and Contextual Aspects", in: Cor-
nelis, G. and Van Bendegem, J.P. (eds.), Einstein Meets Magritte:

Metadebates, Kluwer Academic Publishers, 1998.
[7] Aerts, D., Broekaert, J., Smets, S., "A Quantum Mechanical
Description of the Liar Paradox", submitted for the proceedings of the
20th Int. conf. on Philosophy, Boston, 1998.
[8] Aerts, D., "The Entity in Modern Physics: the Creation-Discovery-
View of Reality", in: Peruzzi, E. (ed.), Identity and Individuality of
Physical Objects, Springer Verlag, Berlin, 1998.
[9] Amira, H., Coecke, B. and Stubbe, I., "How Quantales Emerge
by Introducing Induced State Transition in Operational Theories", Helv.
Phys. Acta, 1998, to appear.
[10] Arendt, H., The life of the mind. One / Thinking, one-volume
edition, Harcourt Brace Jovanovich Publishers, London, 1978.
[11] Aspect, A. et al., Phys. Rev. Lett., 49, 1982, p. 1804.
[12] Barr, M. and Wells, C., Toposes, Triples and Theories, Springer,
1995.
[13] Bell, J., Physics, l, 1965, p. 195.
[14] Beltrametti, E.G. and van Fraassen, B.C. (eds.), Current Is-
sues in Quantum Logic, Plenum, 1981.
[15] Birkhoff, G. and von Neumann, J., Ann. Math., 37, 1936,
p. 823.
[16] Bloch, O.R., La philosophie de Gassendi. Nominalisme, Materia-
lisme et Metaphysique, Martinus Nijhoff, LaHaye, 1971.
[17] Bohm, D., Phys. Rev., 85, 1952, p. 166.
[18] Campbell, J ., The Masks of God, volume I, Primitive Mythology,
Arkana, Penguin Books, N.Y., 1991.
[19] Carnap, R., "Empiricism, Semantics, and Ontology", in: Boyd,
R., Gasper, P. and Trout, J. D. (eds.), The Philosophy of Science,
MIT press, Massachusetts, 1992.
[20] Castaneda, H.N., "Plato's theory of relations", in: Bunge, M.
(ed.), Exact Philosophy, D. Reidel Publishing Comp., Dordrecht, 1973.
[21] Coecke, B., Hidden Measurement Systems, Doctoral Dissertation,
Free University of Brussels, 1995; Helv. Phys. Acta, 70, 1997, p. 442;
Helv. Phys. Acta, 70, 1997, p. 462.
[22] Coecke, B., "A Representation for Compound Quantum Systems
as Individual Entities: Hard Acts of Creation and Hidden Correlations"
and "A Representation for Spin-S Quantum Entities as 2S Individual
Spin-1/2 Quantum Entities", Found. Phys., to appear, 1998.
192 KARIN VEREL ST AND B 0 B C OECKE
[23] Coecke, B. and Valckenborgh, F., Int. J. Theor. Phys., 37,

1998, p. 311.
[24] Coecke, B. and Moore, D.J., "Decompositions of Probability
Measures on Complete Ortholattices in Join Preserving Maps", Preprint,
FUND-DWIS, Free University of Brussels, 1998.
[25] Coecke, B. and Stubbe, I., "On a Creation-Based Extension of
Operational Theories", Internal Report VUB.FUND.1998.3, 1998.
[26] Coecke, B. and Stubbe, I., "Quantum Mechanics as a Theory of
Induction", Preprint, FUND-DWIS, Free University of Brussels, 1998.
[27] Coecke, B., "Compoundness as Diagrams in Quasi-Categories",
notes of a lecture given at IQSA-Atlanta, August 1997; Coecke, B. and
Stubbe, I., "On a Duality of Quantales emerging from an Operational
Resolution", Int. J. Theor. Phys., to appear, 1998; Coecke, B. and
Stubbe, I., "The Category of Mutual Inductions for Entities Described
by Pseudo States", In preparation, 1998.
[28] Cohen, B.I., "Newton's third law and universal gravity", in:
Scheurer, P.B. and Debrock, G. (eds.), Newton's scientific and
philosophical legacy, Kluwer Academic Publishers, Dordrecht, 1988.
[29] Conche, M., Epicure. Lettres et maximes, texte etabli par Marcel
Conche, PUF, Paris (1987).
[30] De Cusa, N., "De Docta Ignorantia", in: Hoffmann, E. and
Klibansky, R. (eds.), Lipsiae, In Aedibus Felicis Meiner, 1932.
[31] Daniel, W., "Bohr, Einstein and Realism", Dialectica, 43, 1989.
[32] Dellian, E., "Inertia, the innate force of matter", in: Scheurer,
P. B. and Debrock, G. (eds.), Newton's scientific and philosophical
legacy, Kluwer academic publishers, Dordrecht, 1988.
[33] Diels, H. and Kranz, W., Fragmente der Vorsokratiker, erster
Band, Weidmann, Dublin, Zurich, 1971.
[34] Dijksterhuis, E.J ., De mechanisering van het wereldbeeld, 7th ed.,
Meulenhof, Amsterdam, 1950-1996.
[35] Dobbs, B.J.T., "Newton's alchemy and his 'active principle' of
gravitation", in: Scheurer, P. B. and Debrock, G. (eds.), New-
ton's scientific and philosophical legacy, Kluwer academic publishers,
Dordrecht, 1988.
[36] Einstein, A., Podolski, E. and Rosen, N., Phys. Rev., 47, 1935,
p. 777.
[37] Einstein, A., "On the electrodynamics of moving bodies", in: The
principle of relativity. A collection of original papers on the special and
general theory of relativity. Notes by A. Sommerfeld., Dover publicatons,

New York, 1952.
[38] Einstein, A., "Can quantum-mechanical description of physical
reality be considered complete?", in: Wheeler, J.A., and Wojciech,
H.Z. (eds.), Quantum Theory and Measurement, Princeton University
Press, Princeton, 1983.
[39] Eilenberg, S., and Mac Lane, S., Trans. Am. Math. Soc., 58,
1945, p. 231.
[40] Epstein, R.L., The semantic foundations of Logic, Vol. I, Kluwer
academic publishers, Dordrecht, 1990.
[41] d'Espagnat, B., Conceptual Foundations of Quantum Mechanics,
W.A. Benjamin, 1976.
[42] Foulis, D.J ., and Randall, C.H., "Manuals, Morphisms and
Quantum Mechanics", in: Marlow, H. (ed.), Mathematical Foundations
of Quantum Theory, Academic Press, 1978.
[43] Foulis, D.J., Piron, C. and Randall, C.H., Found. Phys., 13,
1983, p. 813.
[44] Foulis, D.J., and Randall, C.H., Found. phys., 13, 1983, p. 843.
[45] Fowler, R.N., Plato in twelve volumes, Loeb Classical Library,
Heinemann, London, 1967.
[46] Von Fritz, K., "Nous, noein, and their derivatives in pre-Socratic
philosophy", in: Mourelatos, A.P.D. (ed.), The Pre-Socratics, a collec-
tion of critical essays, Princeton University Press, Princeton, 1974~1993.
[47] Gisin, N., and Piron, C., Lett. Math. Phys., 5, 1981, p. 379.
[48] Harre, R., "The redundancy of Spacetime: Relativity from Cusa
to Einstein", in: Aerts, D. (ed.), Einstein Meets Magritte: White Book,
Kluwer Academic Publishers, 1998.
[49] Hersh, R., in: Tymoczko, T. (ed.), New Directions in the Philos-
ophy of Mathematics, Birkhiiuser, 1986, p. 9.
[50] Jammer, M., Concepts of Space. The history of theories of space
in physics, 3rd. ed., Dover Publications, N.Y., 1954~1993.
[51] Jauch, J.M., and Piron, C., Helv. Phys. Acta, 36, 1963, p. 827.
[52] Jauch, J.M., Foundations of Quantum Mechanics, Addison-Wesley,
1968.
[53] Jauch, J.M., and Piron, C., "On the structure of quantal propo-
sition systems", Helv. Phys. Acta, 42, 1969.
[54] Jaynes, J., The origin of consciousness in the breakdown of the
bicameral mind, Houghton Mifflin Company, Boston, 1976.
[55] Kaufman, L., "Virtual Logic", talk delivered at the Einstein Meets
Magritte Conference, June 1995.
[56] Kirk, G.S., Heraclitus. The Cosmic Fragments, Cambridge Uni-
versity Press, Cambridge, 1975.
[57] Kirk, G.S., Raven, J.E., and Schofield, M., The Presocratic
Philosophers. A critical history with a selection of texts, Cambridge Uni-
versity Press, Cambridge, 1983.
[58] Lanczos, C., The variational principles of mechanics, fourth ed.,
Dover publications, New York, 1970.
[59] Luck, G., Arcana Mundi. Magic and the Occult in the Greek and
Roman Worlds. A collection of Ancient Texts, The Johns Hopkins Uni-
versity Press, Baltimore and London, 1985.
[60] Ludwig, G., "An Axiomatic Basis of Quantum Mechanics", in:
Neumann, H. (ed.), Interpretations and Foundations of Quantum the-
ory, B.I.-Wissenschaftsverlag, 1981.
[61] Ludwig, G., and Neumann, H., "Connections between Differ-
ent Approaches to the Foundations of Quantum Mechanics", in: Neu-
mann, H. (ed.), Interpretations and Foundations of Quantum theory,
B.I.-Wissenschaftsverlag, 1981.
[62] Mac Lane, S., Categories for the working Mathematician, Springer,
1971.
[63] McGuire, J.E., and Rattansi, P.M., "Newton and the 'Pipes of
Pan"', Notes and records of the Royal Society of London, 21, 1966.
[64] McLarty, C., Elementary Categories, Elementary Toposes, Oxford
Sci. Publ., 1995.
[65] Mittelstaedt, P., "Classification of Different Areas of Work Affer-
ent to Quantum Logic", in: Beltrametti, E.G. and van Fraassen,
B.C. (eds.), Current Issues in Quantum Logic, Plenum, 1981.
[66] Moore, D.J., Helv. Phys. Acta, 66, 1993, p. 471.
[67] Moore, D.J., Helv. Phys. Acta, 68, 1995, p. 658.
[68] Moore, D.J., Int. J. Theor. Phys., 36, 1997, p. 2211.
[69] Moore, D.J., "On State Spaces and Property Lattices", British J.
Phil. Sc., to appear, 1998.
[70] Mourelatos, A.P.D., The Pre-Socratics, a collection of critical
essays, Princeton University Press, Princeton, 1974-1993.
[71] Newton, I., Isaaci Newtoni, Opera quae extant omnia, Commen-
tariis illustrabat Samuel Horsley, Landini: exc. Joannes Nichols, 1889.
[72] Owen, G.E.L., "Plato on Non-Being", in: Vlastos, G. (ed.), Plato:

a collection of Critical Essays, vol I, Doubleday, 1971.
[73] Piron, C., Foundations of Quantum Physics, W.A. Benjamin, 1976.
[74] Piron, C., "Le n~alisme en physique quantique: une approche selon
Aristote", in: The concept of physical reality. Proceedings of a conference
organized by the Interdisciplinary Research Group, Univ. of Athens, 1983.
[75] Piron, C., Erkenntnis, 16, 1981, p. 398.
[76] Piron, C., "Quantum Mechanics Fifty Years After", in: Lahti, P.,
and Mittelstaedt, P. (eds.), Symposium on the Foundations of Modern
physics, World scientific, 1985.
[77] Piron, C., Mecanique Quantique Bases et Applications, Presses
Polytechniques et Universitaires Romandes, Lausanne, 1990.
[78] Piron, C., "Un nouveau regard sur le monde physique", preprint,
Universite de Geneve, 1996.
[79] Piron, C., "Quantum Mechanics and Relativity: two failed rev-
olutions", in: Aerts, D., (ed.), Einstein Meets Magritte: White Book,
Kluwer Academic Publishers, 1998.
[80] Quinton, A., The Nature of Things, Routledge & Kegan Paul,
London and Boston, 1973.
[81] Rauch, H., Helv. Phys. Acta, 61, 1988, p. 589.
[82] Redondi, P., Galilei, eretico, Einaudi, Turino, 1983.
[83] Rescher, N., and Brandom, R., The logic of inconsistency. A
study in Non-Standard Possible- World Semantics and Ontology, Basil
Blackwell, Oxford, 1980.
[84] Riezler, K., Parmenides, Text, Ubersetzung, Einfuhrung und In-
terpretation, Vittorio Klostermann, Frankfurt a.M., 1970.
[85] Rosen, R., "The Roles of Necessity in Biology", in: Newton to
Aristotle. Toward a Theory of Models for Living Systems, Birkhiiuser,
Boston, Basel and Berlin, 1989.
[86] Schrodinger, E., Space- Time Structure, Cambridge University
Press, Cambridge, 1950-1985.
[87] Schrodinger, E., Nature and the Greeks and Science and Human-
ism, Canto, Cambridge University Press, 1954-1996.
[88] Spencer Brown, G., Laws of Form, George Allen and Unwin, ltd.,
London, 1969.
[89] Toulmin, S., Cosmopolis, the hidden agenda of modernity, Univer-
sity of Chicago Press, Chicago, 1984.
196 KARIN VERELST AND BOB GOECKE
[90] Tredennick, H., Aristotle. Metaphysics, books I-IX, Loeb XVII,

Loeb Classical Library, Harvard University Press, Cambridge, Massa-
chusetts, 1996.
[91] van Fraassen, B.C., Quantum Mechanics, Clarendon Press, Ox-
ford, 1991.
[92] Vlastos, G., "Reasons and causes in the Phaedo", in: Vlastos, G.
(ed. ), Plato: a collection of Critical Essays, vol I, Doubleday, 1971.
[93] Vlastos, G., Studies in Greek Philosophy, volume I: The Preso-
cratics, Princeton University Press, Princeton, 1993.
[94] von Neumann, J., The Mathematical Foundations of Quantum
Mechanics, Princeton University Press, 1955.
ARKADY PLOTNITSKY
LANDSCAPES OF SIBYLLINE STRANGENESS:

COMPLEMENTARITY, QUANTUM MEASUREMENT AND
CLASSICAL PHYSICS
Quantum mechanics has divided our description of the physical world

into two distinct counterparts-classical at the macrolevel and quantum
at the microlevel-and by so doing initiated a debate that has contin-
ued ever since. On the one side of this debate one finds arguments for
the necessity of replacing (and attempts to replace) quantum mechanics
with classical or classical-like theories-from Einstein's initial skepticism
and the Einstein-Podolsky-Rosen (EPR) argument to hidden variables
and beyond. On the other side, one encounters attempts to account for
classical and quantum physics as both necessary for our understanding
of the physical world. Niels Bohr's interpretation of quantum mechanics
as complementarity follows this second approach, with radical implica-
tions for classical and quantum physics, and, as I shall suggest here, for
their relationships. I shall also suggest that, contrary to a common view,
Bohr's interpretation reveals several affinities and interactions between
both sides of the debate at issue.
Complementarity originates in Bohr's argument that quantum physics
entails what he called complementary descriptions-descriptions that
are mutually exclusive (and, thus, not applicable simultaneously at any
point) but equally necessary for a comprehensive overall account of quan-
tum phenomena. This argument leads to difficulties in developing a re-
alist interpretation of quantum physics since at any given point at least
two mutually exclusive "pictures" are possible and there appears to be no
underlying (classically) complete configuration that such pictures would
partially represent. By a realist interpretation I refer to a possibility of
representing quantum "reality" by ascertaining a one-to-one correspon-
dence (in practice or in principle) between this reality and a physical
description, or even only a partial, incomplete representation of an in-
dependent physical reality, conceived on the model of classical physics.
According to Bohr, while possible or even necessary in classical physics,
the concept of physical reality as existing by itself and in itself indepen-
dently of observation and measurement-and thus, in a certain sense,
of representation-becomes equally inapplicable in quantum physics.
These formulations need not deny-and Bohr never did-a certain ma-
terial "efficacity" (to oppose this term to classical causality) of quantum
197
198 ARKADY PLOTNITSKY
phenomena. This efficacity, however, can neither be ascribed the sta-

tus of independent physical reality nor be seen as a "wholeness" that
is only partially represented by complementarity descriptions. Quantum
physics, Bohr argues, does give us a sufficient amount of data in order to
build a comprehensive theoretical description, while disallowing classical
ontological or epistemological claims-either objectivist or, importantly,
subjectivist or positivist, which have been (in my view, mistakenly) as-
cribed to Bohr [1].
Bohr develops complementarity into a comprehensive framework, who-
se aim is to offer an interpretation of quantum mechanics fully consistent
both with the experimental data of quantum physics and (which is not
always sufficiently emphasized) with relativity. As such, complementar-
ity also entails both an irreducible division (in terms of their description)
and an irreducible interaction between quantum micro-objects, described
by means of quantum-mechanical formalism, and a measuring macro-
apparatus, described the interaction by means of the formalism of clas-
sical physics. Both, thus, are capable of quantum interaction, through
which and only through which complementary configurations themselves
emerge in view of mutually exclusive experimental settings that one can
arrange. Bohr leaves open the question of the physics of transition from
the quantum (micro)world to the classical (macro)world, which remains
an outstanding problem of modern physics and the subject of continuing
research.
Complementarity is now primarily associated with wave-particle com-
plementarity, which suspends unequivocal identification of two classes
of phenomena-particle-like, such as electrons; and wave-like, such as
light or other forms of radiation-found in classical physics. In quan-
tum physics such unequivocal identification is no longer possible, since
depending on how we arrange our instruments we will observe either a
particle-like or a wave-like picture (although never both together), and
we must consider both in order to give a comprehensive description.
There are several reasons for the prominence of this particular comple-
mentarity, although complementarities connected to spin have become
equally important in the wake of Bohm's reformulation of the EPR argu-
ment, hidden variables, and Bell's theorem, if usually without invoking
the idiom of complementarity itself. For Bohr himself the most crucial
specific complementarity is that of coordination and causality. It dislo-
cates the causal spatia-temporal description that defines classical sys-
tems and that allows one to know with certainty the state of a system
at any point, once it is known at a given point. According to Bohr:
LANDSCAPES 0 F SIBYLLINE STRANGENESS 199
"The very nature of the quantum theory ... forces us to regard the
space-time coordination and the claim of causality, the union of which
characterizes the classical theories, as complementary but exclusive
features of the description, symbolizing the idealization of observa-
tion and definition respectively" [2].
The complementarity of coordination and causality is, thus, linked

to another complementarity, conceived in more general philosophical
terms-the complementarity of (the idealizations of) observation and
definition. The latter is given less attention in Bohr's work subsequent
to the so-called Como lecture, which introduces complementarity and
which I cite here. In the lecture itself, however, it plays a central role, and
Bohr shows how other complementarities can be seen as its consequences
or correlatives. That includes wave-particle complementarity, or those of
position and momentum, or time and energy, most closely tied to Heisen-
berg's uncertainty relations. The latter express the irreducible limits on
the possibility and, in fact, preclude (this is crucial) simultaneous ex-
act measurement of both variables within each pair. According to Bohr,
more radically, uncertainty relations preclude one even from considering
such variables as simultaneously applicable or unambiguously definable
at any point by virtue of always mutually exclusive experimental ar-
rangements, which would allow one to measure and define each variable
without ambiguity. Therefore, Bohr argues, one cannot assign classical-
like states to quantum systems. Uncertainty relations (along other com-
plementarities) can be connected to wave-particle complementarity, in
particular via Max Born's interpretation of Schrodinger's wave-function
as "mapping" the probability-distribution defining the statistical predic-
tion of quantum mechanics.
Here, I would like to explore implications of Bohr's framework for the
relationships between classical and quantum physics. The main conclu-
sion is that, in Bohr's scheme, a violation of quantum mechanics by as-
signing physical states conceived, on the classical model, as independent
of observation and measurement ("independent physical reality") would
imply a violation of classical description for the measuring instruments
involved -unless, in view of the EPR argument and Bell's theorem, the
Lorentz invariance is violated and propagation of signals with infinite
speed is allowed. In Bohr's framework classical macrophysics and quan-
tum microphysics are intertwined in a peculiar way: an assumption of
classical-like description at the microlevel would imply a breakdown of
classical description at the macrolevel (at least, in the case of measur-
ing instruments), while non-classical microdescription is consistent with
classical macrodescription [3].
It is, I think, appropriate to see the ascription independent physical

states as a violation of quantum mechanics in Bohr's or, more broadly,
in the so-called orthodox interpretation, even though the resulting in-
terpretations may retain its data, as do hidden variables theories, such
as Bohm's, which of course employ a formalism expressly different from
that of quantum mechanics [4]. The question of an independent physical
reality is especially significant here, perhaps more significant than the
question of causality, the second key issue dividing classical and quantum
models. The possibility or, conversely, impossibility of simultaneously as-
cribing definite values to both conjugate variables and thus classical-like
states to the system are responsible for different rules of calculating the
probabilities of the outcomes of key experiments in quantum measure-
ment. This quantitative difference arises from the qualitative difference
defined by the possibility (as in the case of classical physics) or con-
versely impossibility (as in the case of quantum mechanics) to ascribe
definite values to the variables involved regardless of whether they are
actually measured or not [5]. The inequalities of Bell's theorem arise in
view of similar considerations and, in particular, the (im)possibility of
an independent existence of physical variables or states, more so than on
the (im)possibility of hidden variables as such, since the latter may also
be conceived interactively, that is, without assuming states independent
of observation or measurement.
It is not that we can simply abandon-how can we?-classical physics
and classical, or classical-like, dimensions of quantum physics, for ex-
ample, in view of the correspondence principle, the classical description
of measuring instruments, or our dependence on the complementary us-
age of classical concepts. The present argument is for the richness of
both classical-like and quantum-like approaches (and there are many
in each category) to physics or philosophy, and of the interactions be-
tween them-instead of dividing them, dogmatically or antagonistically,
which unfortunately has not been uncommon in the history of the philo-
sophical debate concerning quantum mechanics. Moreover, it is far from
always clear which- classical-like or quantum-like-approach is either
more effective or even more revolutionary. Thus, Einstein's position con-
cerning quantum physics has been argued to be, at least in some of its
aspects, more revolutionary than Bohr's [6]. From (and before) Einstein
and Bohr to the most recent work in physics and philosophy, we have
a wealth of productive approaches on both sides, which may well be
the most crucial lesson of the history of quantum physics. We might
need more radical interpretations still-either more radically realist or
more radically anti-realist, or both. Einstein and Bohr meet and do not
meet (or complementarily both) on many roads, where they may also
LANDSCAPES 0 F SIBYLLINE STRANGENESS 201
encounter Magritte, or Picasso and Duchamp.

The overall configuration just outlined is philosophically generaliz-
able, which also allows one to describe in these terms several contempo-
rary philosophical frameworks and their relations to different scientific
or mathematical models-such as Gilles Deleuze and Riemannian geom-
etry, Michel Serres and chaos theory (in juxtaposition to quantum me-
chanics), or Derrida and Godelian incompleteness and undecidability-
as well as their relations to different artistic models, be it Cubism,
Duchamp and Magritte, or Joyce and Pound. This configuration allows
one to rethink the relationships between classical and nonclassical ap-
proaches, which can be defined in other fields on the model of classical
and quantum theories in physics, and different epistemologies they en-
tail. I cannot consider these issues here (beyond a few brief indications)
and shall remain within the limits of Bohr's framework.
Heisenberg's uncertainty relations express mathematically the limits
on the possibility of simultaneous exact measurement-and, according
to Bohr, ascribability and even unambiguous definition-of complemen-
tary variables. In the formalism of quantum mechanics, variables (or
observables, as they are also called) are represented by what are called
operators in a Hilbert space-p (momentum), q (coordinate), and so
forth-to which standard mathematical operations, such as addition and
multiplication, can be assigned. Details of the mathematics of Hilbert
spaces need not be considered here. Most crucial is that a peculiar (for
physics) feature of this formalism is that the multiplication of such oper-
ators in general does not commute, specifically when conjugate variables
are involved-pq does not equal qp. It became quickly apparent in the
wake of Heisenberg's introduction of his (matrix) quantum mechanics
that it is this feature that reflects most fundamentally the difference
between classical and quantum physics. The difference between the two
products pq and qp allows one to establish certain quantitative limits,
which are represented by Heisenberg's uncertainty relations. According
to Bohr, this noncommutability represents not only the fact that conju-
gate variables, such as momentum or position, cannot be simultaneously
measured with unlimited precision, but that they cannot even be fully
determined and unambiguously defined simultaneously, as they can be
in the case of classical physics. Hence, Bohr argues, the formalism of
quantum mechanics entails a physics that dislocates the realist inter-
pretation entailed, or at least permitted, by classical physics and the
classical-say, Hamiltonian-formalism that describes it.
As a result, in Bohr's interpretation, quantum-mechanical formalism
cannot be seen as representing any (independently existing) physical re-
ality, but must instead be seen as reflecting the impossibility of such a

representation and, indeed, of a reality conceived on the model of classi-
cal physics. Instead Bohr sees quantum-mechanical formalism as repre-
senting the interaction between quantum (micro) objects and measuring
instruments, described in terms of classical physics at the macrolevel.
Bohr connects both formalisms, the classical physics of measuring in-
struments and the quantum physics of the objects under observation or
measurement, or what we infer as such "objects" by means of observa-
tion and measurement, using the classical physics of measuring instru-
ments. In contrast to classical physics, this interaction is, Bohr argues,
irreducible: it cannot be disregarded or compensated for so as to enable
one to return to a classical-like theory. At the microlevel measuring in-
struments are, of course, capable of quantum interactions with quantum
objects. A full determination of one complementary variable, however,
requires an experimental arrangement which makes it impossible to do
the same for the other. Quantum- mechanical formalism and uncertainty
relations reflect this mutual exclusivity. As Bohr writes:
The limited commutability of the symbols by which such variables
are represented in the quantal formalism corresponds to the mutual
exclusion of the experimental arrangements required for their unam-
biguous definition. In this context, we are of course not concerned
with a restriction as to the accuracy of measurement, but with a
limitation of the well-defined application of space time concepts and
dynamic conservation laws, entailed by the necessary distinction be-
tween measuring instruments and [quantum] objects. (PWNB 3:5;
emphasis added) [7].
Both our descriptions of our instruments and the pictures they produce
are classical, and we can have no other instruments and can "see" no
other pictures. Thanks to the efficacity emerging within the interactive
(classical at the macro level, quantum at the microlevel) configuration
of quantum measurement, we infer quantum objects from the "sum" of
these pictures and by correlating some of them. These pictures, how-
ever, never quite "sum up," at least not in the way they do in classical
physics. There is no underlying complete classical-like picture from which
these complementary classical "parts" are derived, which impossibility
is the essence of quantum mechanics and/ as complementarity. There is
no "whole" that is the sum of these "parts" -or even that is more, or
less, than the sum of its parts. This radical strangeness is perhaps what
troubled Einstein most, even though he was no stranger to strange-
Sibylline- landscapes, and created a few of his own [8].
LANDSC AP ES 0 F SI BYLLINE ST R A NG E NE SS 203
It is crucial that quantum mechanics limits only the simultaneous mea-

surement, applicability and definability of both conjugate variables and
only within the same experimental arrangement. Either variable by it-
self can be measured, at least in principle, with full precision and, thus,
be well-defined at any point. The latter fact may suggest that quantum-
mechanical description is incomplete and that one can develop a more
complete classical-like account of the data at issue. According to Bohr,
however, instead of the incompleteness of quantum theory, uncertainty
relations signal the inapplicability to quantum physics of a classical, and
especially realist, picture-either strictly causal or classically statistical
(i.e., implying an underlying non-statistical configuration incompletely
represented by a statistical account). This does not mean that a single
variable can be attributed an existence independent of an experimental
arrangement through which it is defined as a variable. Such an attribu-
tion, in contrast to definition through observation or measurement, is
just as impossible for a single variable as for both variables. The latter
nuance figures significantly in Bohr's reply to the EPR argument, which
played a key role in sharpening his view [9].
According to Bohr, the EPR argument failed to demonstrate the in-
completeness of quantum mechanics, which was the main stated aim
of the argument. According to both parties, however, it demonstrated
something that may be more important-namely, the profound connec-
tions, first, between quantum mechanics and relativity and, second, be-
tween quantum and classical physics. The EPR argument concerns a sit-
uation in which-in an apparent conflict with quantum mechanics-both
the position and momentum of a particle appear ascribed definitively,
without uncertainty, in view of the following considerations. Although
(due to uncertainty relations) in quantum mechanics one can never si-
multaneously measure both the momentum and position of any particle
with full precision, quantum mechanical formalism does allow one to
calculate without uncertainty the combined momentum of and the rel-
ative distance between two particles, since corresponding operators do
commute. If these particles are then allowed to interact, the possibility
of measuring either the position or momentum of one particle at any
given point (which is allowed by quantum mechanics) and the law of
momentum conservation allow one to predict with certainty (but not
simultaneously) either the position or the momentum of the other parti-
cle, and thus ascribe both to the other particle without interfering with
it. Since, EPR claim, in quantum mechanics such ascriptions are impos-
sible, their argument shows the incompleteness of quantum mechanics.
Bohr argued that the application of the EPR assumptions to quan-

tum mechanics needed qualifications, beginning with what quantum me-
chanics allows as concerns predictions of the values of both conjugate
variables, such as position and momentum. Quantum mechanics claims
only that, due to uncertainty relations, the exact simultaneous mea-
surement of both within the same experimental arrangement is never
possible. In the EPR situation, too, such simultaneous measurements,
or even simultaneous predictions, are impossible. It is the impossibil-
ity of simultaneous measurement and only this impossibility- rather
than impossibility of predicting either complementary variables on the
basis of other data and calculations-that defines quantum mechanics.
Predictions with certainty of either the momentum or the position of a
given particle are possible under certain conditions, such as those speci-
fied by the EPR argument; exact measurement of both simultaneously is
never possible under any conditions, for example, if one wants to simul-
taneously verify such predictions. EPR do introduce the question of the
simultaneous measurement and prediction of conjugate variables at the
end of their article and stress, correctly, that this is a crucial point, since
both separability (independence of physical processes and measurement
in spatially separated system) and locality (the impossibility of instan-
taneous propagation of signals and, hence, the Lorentz invariance and
relativity) are at stake [10]. The significance of this point transcends the
difference of the two positions, and remains a crucial issue in the debates
concerning quantum physics, especially in the wake of Bell's theorem,
which brought the questions at issue into a new and sharper focus and
gave them deeper physical and philosophical significance [11].
Bohr argued that the interactions between quantum objects and mea-
suring instruments were not sufficiently taken into account by EPR,
while a rigorous analysis of these interactions disallows their key as-
sumptions and exposes "an essential ambiguity" in the EPR argument.
He also argued that no physical disturbance and thus no physical action
at a distance and hence no violation of relativity takes place in the EPR
situation if treated by means of quantum mechanics, the point that Ein-
stein partly contested later in terms of separability [12]. In view of Bell's
theorem, such a violation would in fact occur if one were to assign an
underlying classical-like picture (either in terms of standard or hidden
variables) to quantum data-and, in particular, if one sees these data as
reflecting reality existing independently of observation or measurement.
The EPR argument and, subsequently, Bell's theorem were crucial for es-
tablishing that point. For Bohr, the data of quantum-mechanics disallows
the possibility of ascribing a classical-like picture of states-states inde-
pendent of observation and measurement-to quantum systems. Bohr
LANDSC AP E S 0 F SI BY LLINE ST R AN GE NE SS 205
concludes that "the apparent contradiction [argued by EPR] in fact dis-

closes only an essential inadequacy of the customary viewpoint of natural
philosophy for a rational account of physical phenomena of the type with
which we are concerned in quantum mechanics ... [T]he interaction be-
tween object and measuring agencies ... entails ... the necessity of a final
renunciation of the classical ideal of causality and a radical revision of
our attitude towards the problem of physical reality" ( QTM, 145-46).
As I shall suggest here, however one pursues such a revision (closer to
Bohr or to Einstein), the latter issue may even be more crucial to EPR
(and especially to Einstein himself) than the question of causality [13].
The EPR argument was a major contribution to the debate concern-
ing quantum theory, and it was given a further significance when Bohm
introduced his hidden variables interpretation of quantum-mechanical
data in 1952, and, then, in the wake of Bell's theorem, in 1964. Hid-
den variables theories seek to supplement quantum-mechanical statisti-
cal data by additional variables affecting physical states, unobtainable
by means of- "hidden" from-quantum mechanics, although operative
within a nonstatistical, causal configuration and possibly eventually ex-
perimentally obtainable or inferable. Quantum mechanics, then, would
be subsumed by a classically complete and causal framework. Such a
framework may or may not involve the concept of independent physical
reality, depending on how one locates hidden variables-in such an inde-
pendent reality itself, apparatus, or more interactively. As will be seen,
one cannot, in principle, exclude the possibility of a hidden variables
theory which is, in a certain sense, causal, but does not entail classical-
like states independent of observation or measurement. In contrast to
Bohm's, such a theory, if it exists, may even be local (i.e., Lorentz in-
variant).
Bell's theorem introduced a dramatic twist into the hidden variables
question by stating that Bohm's theory was no exception and that any
hidden variables theory compatible with the statistical predictions of
quantum mechanics would have to be, in Bell's phrase, as "grossly non-
local" as Bohm's theory. Bell's theorem makes it impossible (or just
about) to maintain simultaneously the data of quantum mechanics and
relativity within a classical-like theory. This, once again, may be ascer-
tained at least for hidden variables theories presupposing an independent
physical reality (conceivably described by means of hidden rather than
standard variables, which would arise only through a measuring process)
[14]. Insofar as quantum-mechanical and relativity data remain valid,
however, Bell's theorem places severe constraints upon the possibility
of classical-like theories accounting for quantum data-the data that
violates local causality and disallows an independent physical reality.

The quantum-mechanical situation has significant implications for the
relationships between classical and quantum physics. Bohr's interpreta-
tion fundamentally depends on the description of measuring instruments
in terms of classical physics. By the same token, quantum data become
connected to the classical (such as Hamiltonian) formalism describing
measuring instruments used to obtain them, since in Bohr's interpreta-
tion every aspect of quantum-mechanical formalism, in particular uncer-
tainty relations, are related to the classical physics of the instruments-
such as classical conjugate variables describing or coordinated with the
instruments, energy and momentum conservation laws, and so forth.
Bohr establishes a rigorous relation between the classical, such as
Hamiltonian, formalism of the apparatus and the quantum-mechanical
formalism of the quantum system, and connects uncertainty relations
b..pb..q "" h both to the commutation relations for the corresponding con-
jugate variables of the particles as operators in a Hilbert space obeying
the commutation rule
h
pzqj - qjpz = 21ri 8z1
and to the classical Hamiltonian conjugate variables of the apparatus,
defined by well-known equations
This scheme is then argued to be correlative to the mutually exclu-

sive character of experimental arrangements, through which (and only
through which) one can unambiguously establish and define either one
or another complementary variable, but never both together.
It appears to follow, given Bohr's logic, that if one assigns classical-like
states to quantum objects and, by implication, a realist interpretation
to the data described in a nonrealist way by quantum mechanics, one
is no longer able to maintain a classical description of the measuring
instruments. Here, I can only indicate the main logic of this argument
[15].
Bohr's analysis is predicated on the classical features of measuring
instruments and their interactions with quantum objects, which are re-
flected in the formalism of quantum mechanics. Bohr derives the impos-
sibility of ascribing classical-like states to a quantum system from these
features-and from the experimental data at issue in quantum mechan-
ics. Therefore, insofar as the standard quantum-mechanical data is valid
LANDSCA P E S 0 F SIBY LLI NE ST R A NG E NE SS 207
and (in view of Bell's theorem in the case of hidden variables) insofar as
the Lorentz invariance demanded by relativity is maintained, the possi-
bility of such states would imply a breakdown of the classical description
of measuring instruments. For the classical describability of measuring
instruments becomes the only vulnerable link in Bohr's logical chains.
One needs to consider both standard and hidden variables because one
must examine all available situations in which classical-like states may,
in principle, be ascribed to quantum data, and establish that, if one
does so, the classical describability of measuring instruments would be
violated in each of these situations. Bell's theorem eliminates the case
of hidden variables, since it implies that any hidden variables frame-
work consistent with the predictions of the standard quantum theory
would lead to violation of the Lorentz invariance. The EPR situation
as treated by means of non-relativistic quantum mechanics entails no
violation of relativity, in contrast to treating it by means of Bohm's
theory or, it appears, any theory that presupposes independent states.
Establishing fully rigorous connections between quantum mechanics and
relativity entails quantum electrodynamics. Thus theory leads to further
epistemological complexities, which cannot be considered here.
In view of some recent arguments, we may not be able, even by using
Bell's theorem, unconditionally to exclude the possibility of local (i.e.,
Lorentz invariant) and locally causal hidden variables theories of a non-
realist type-that is, theories that are causal but that (in contrast to
Bohm' s theory/ ies) would disallow an ascription to a quant urn system
of classical-like states independent of observation. The formulation of
Bell's theorem needs to be amended accordingly. Hidden variables in
this case would have to be seen as "located" in the interaction between
quantum objects and measuring instruments, an interaction that would
itself be seen as irreducible, just as in Bohr's interpretation. Hence, the
possibility of such theories would not undermine the present argument.
Such theories would offer the intriguing possibility of a picture of the
physical world that is both causal and local but that does not presuppose
an independent physical reality at the microlevel [16].
In Bohr's interpretation, classical macrophysics and quantum micro-
physics are intertwined in an asymmetrical way: attempts to introduce
a more classical or classical-like description at the microlevel appear to
break down classical description at the macrolevel, while non-classical
quantum microdescription is fully consistent and sustains classical ma-
cro-description, at least in the situation of quantum measurement. In
this scheme quantum (micro) physics, relativity, and classical (macro)
physics are irreducibly interlinked, and both modes of description, classi-
cal and quantum-mechanical, are necessary. One cannot perhaps replace

one of them-say, quantum theory (including quantum field theory) with
a classical-like framework-without affecting or undermining others at
some level.
Bohr's view remains open as to the mechanism of transition from the
quantum (micro)world to the classical (macro)world or of their coexis-
tence and joint functioning, for example, in the situation of quantum
measurement. The problem, as I said, remains outstanding and is re-
flected in the number of so-called paradoxes, such as that of Schrodinger's
cat (although such paradoxes may be best seen as manifesting this prob-
lem rather than logical paradoxes). It is difficult to predict along which
lines its solution will proceed [17].
According to Bohr, rather than physical (micro)reality, quantum me-
chanics represents something else-namely, the structure of the interac-
tion between quantum objects, which disallow a realist interpretation,
and classical measuring instruments, whose physics allows for such an
interpretation. It connects the possibility of realism and classical concep-
tuality at the macrolevel and their suspension at the microlevel. In more
philosophical terms (which can be extrapolated to the relationship be-
tween classical and non-classical theories elsewhere and which will help
dispel some major misunderstandings of their relationships), in Bohr's
interpretation quantum theory represents the relationship between that
which is representable in a classical, and specifically causal and real-
ist, sense and that which is unrepresentable by any means, classical or
non- classical, or through any concept (including the concept of the ab-
solutely unrepresentable) but to which one can relate-obliquely, but
rigorously-only by means of a non-classical and specifically counter-
realist theory. There emerge deep connections between philosophy and
physics, both its mathematical formalism and its experimental technol-
ogy, in turn, according to Bohr, irreducibly joined in quantum physics,
or rather within the juncture of classical and quantum physics as com-
plementarity.
The richness and complexity of this juncture and multiple possibil-
ities it offers to physics and philosophy may well be more significant
than any particular interpretation of it. As I have argued here, we
might need to proceed by connecting (naturally, with critical discrimina-
tion) rather than by dogmatically separating different interpretations-
complementary, hidden variables, macrorealist, or still other. Current
theories remain manifestly incomplete. Special relativity and general rel-
ativity, quantum electrodynamics and other quantum field theories, and
so forth describe, in classical and quantum terms, various macro and
L A N D S C A P E S 0 F S I B Y L L I N E S T R A N G EN E S S 209
micro aspects of the physical world. For the moment we can at best
only correlate or conjoin some of them and try to maintain their consis-
tency with each other (and, of course, with experimental data) within
sufficiently workable limits. Whether physics can ever be fully brought
together remains an open question; and, as I have stressed here, the
problem of transition from quantum microphysics to classical macro-
physics remains outstanding. It is conceivable that future theories (or
new data) will transform both classical and quantum theories and their
relations, perhaps by means of a more homogeneous single theory. It
is also conceivable that future developments will preserve the joint sig-
nificance of classical and quantum physical and philosophical theories,
which has defined the century of physics that began with Planck's dis-
covery of quantum physics in 1900 and that is now at its end-or will
produce landscapes of as yet unencountered or even as yet inconceivable
strangeness.
AFFILIATION
Arkady Plotnitsky
Department of English
Purdue University
West Lafayette, IN 47907-1356, USA
aplotnit@acpub. duke. edu
REFERENCES
[1] Among recent studies considering Bohr's philosophical positions are

Faye, J ., Niels Bohr: His Heritage and Legacy. An Anti-Realist View of
Quantum Mechanics. Kluwer, Dordrecht, 1991; Folse, H.J., The Phi-
losophy of Niels Bohr: The Framework of Complementar·ity. North Hol-
land, Amsterdam, 1985, and Folse, H.J., "Bohr on Bell", in: Cush-
ing, J.T. and McMullan, E. (eds.), Philosophical Consequences of
Quantum Theory: Reflections on Bell's Theorem, University of Notre
Dame Press, Notre Dame, Ind., 1989; and Hanner, J., The Description
of Nature: Niels Bohr and Philosophy of Quantum Physics, Clarendon,
Oxford, 1987. I have considered Bohr's (anti)epistemology in detail in
Plotnitsky, A., Complementarity: Anti-Epistemology After Bohr and
Derrida, Duke University Press, Durham, NC., 1994, and Plotnitsky,
A., "Complementarity, Idealization, and the Limits of Classical Concep-
tions of Reality", SAQ, 94, 2, 1995, pp. 527-570.
[2] The Philosophical Writings of Niels Bohr (hereafter PWNB), 3 vols.
Ox Bow Press, Woodbridge, Conn., 1, 8, 1987.
[3] I am indebted to discussions with Philip J. Siemens and Anthony J.

Leggett's work (see [5]).
[4] There are several versions of Bohm's theory. The most recent one is in
Bohm, David and Hiley, B.J ., The Undivided Universe, Routledge,
London, 1993. All of them entail a violation of the Lorentz invariance
(even though this violation cannot be observed in practice).
[5] See Leggett, A.J., "Experimental Approaches to the Quantum
Measurement Paradox", Foundations of Physics, 18, 9, 1988, and "On
the Nature of Research in Condensed-State Physics", Foundations of
Physics, 22, 2, 1992.
[6] Fine, A., The Shaky Game: Einstein, Realism and the Quantum
Theory, University of Chicago Press, Chicago, 24, 1986.
[7] See also "The Causality Problem in Atomic Physics", New Theories
of Physics, Institute of Intellectual Cooperation, Paris, 1939.
[8] The expression "the Sibylline strangeness of the landscapes" is bor-
rowed from Wheeler, who applied it to Einstein's general relativity in
Wheeler, J.A., "Beyond the Black Whole", in: Woolf, H., (ed.),
Some Strangeness in Proportion, Addison-Wesley, Reading, Mass., 1980,
p. 341.
[9] Both articles were published under the same title, "Can Quantum-
Mechanical Description of Physical Reality Be Considered Complete?"
Physical Review, 47, 1935, pp. 777~780., and in: Physical Review, 48,
1935, pp. 696~702. and republished in: Wheeler, J.A., and Zurek,
W.H. (eds.), Quantum Theory and Measurement (hereafter QTM),
Princeton University Press, Princeton, NJ., 1983.
[10] Separability is equally crucial to Einstein, and it can perhaps be
maintained (along with locality) under the conditions of quantum me-
chanics, although many recent authors argue that quantum mechanics
(as a local theory) must be seen as violating Einstein-like separabil-
ity requirements in view of Bell's theorem. Cf., most contributions to
Philosophical Consequences of Quantum Theory: Reflections on Bell's
Theorem, Cushing, J.T., and McMullan, E. (eds.), opus. cit. [1], ex-
cept for Arthur Fine's article, "Do Correlations Need to be Explained?",
which maintains a more subtle position in this respect. I surmise that
Bohr would argue the contrary and such an argument can be derived
from his reply to EPR, in part because the principle of separability usu-
ally (although not always) entails the concept of independent reality.
The question of separability does not affect my main argument, in so-
far as locality is maintained in either interpretation. Throughout this
article I mean by relativity the Lorentz invariance and the impossibil-
LANDSCAPES OF SIBYLLINE STRANGENESS 211
ity of instantaneous propagation of physical signals (there may be some

ambiguity in interpreting Einstein's special relativity as well).
[11] The literature on the issue is immense. See Cushing, J. T. and
McMullan, E. (eds.), opus. cit. [1], for a representative philosophically
oriented selection. A number of books and collections and many articles
have appeared since.
[12] Quantum mechanics may be, and has been, seen as non-local theory
in various ways, for example, in the sense of violating the separability
principle, or, as it is sometimes called, quantum "entanglement," and so
forth. This non-locality remains consistent with relativity.
[13] Pauli argues a similar point in his letter to Max Born (The Born-
Einstein Letters, Walker, New York, 1971).
[14] See, for example, Bohm and Hiley, The Undivided Universe,
pp. 106-116. Their theory is, again, not Lorentz invariant.
[15] The full analysis is an article, now in preparation, "Quantum Theory
and the Classical Physics of Measurement in Niels Bohr".
[16] The point appears to emerge or to be implied in a number of recent
treatments. See, in particular, Laszlo E. Szabo, "Is Quantum Mechan-
ics Compatible with a Deterministic Universe? Two Interpretations of
Quantum Probabilities", Fo'undations of Physics Letters, June, 1995. As
I have argued, the probabilities at issue and the outcome of Bell's in-
equalities depend on ascribing definite values to the variables involved
at all times, irrespective of whether or not they are actually measured
rather than on determinism or causality, which are conceivable within a
more interactive scheme. Szabo's argument is in accord with and indeed
depends on this point. Szabo's article leaves open the question of what
such a theory would look like. His point is that the possibility of such a
theory cannot be excluded in advance, even given Bell's theorem.
[17] On some of these issues, see Anthony J. Leggett's work on macrore-
alisnl cited earlier [5], and, along different lines, recent work by Diederik
Aerts and coworkers. See Aerts, D., "Quantum Structures, Separated
Physical Entities and Probability", Foundations of Physics, 24, 9, 1994,
and references there.
LARS LOFGREN
METALINGUISTIC VIEWS OF QUANTUM MECHANICS

AND ITS FORMALIZABILITY
1. METALINGUISTIC VIEWS
When we talk of a language, we need in general a meta-language. The

reason is that no language can be completely introspective (cf. the lin-
guistic complementarity to be outlined below; cf. also [6], [8]).
Also for quantum mechanics, we need in general a meta-language for
understanding quantum mechanical measurability-which is a linguistic
phenomenon [6].
In general, questions of formalizability, like questions of completeness,
require meta- object- distinctions. Related are metamathematical type
distinctions, or levels of inference (in provability or realizability).
Let us remind of a metamathematical concept of a formal theory in
a formal language. A formal theory has axioms and rules of inference
which are decidable as such, meaning that there is an algorithm which
decides of strings of symbols whether they form an axiom or consti-
tute an inference according to a rule of inference. Any sentence which
makes sense with respect to the intended interpretations in the lan-
guage, must be formalizable as a well-formed formula (abbreviated wff),
where well-formedness is decidable. This does not mean that the truth,
or provability, of a wff need be decidable, only that the property of being
well-formed, i.e., of being interpretable in the language, is decidable. In
a many-sorted language, where the sorts refer to types, the sort (type)
of a well-formed formula is decidable. The sorts (types) to be considered
here, refer to kinds of ascertainment.
In quantum mechanics there are two kinds of ascertainment at play.
Namely, to ascertain by quantum mechanical measurement, and to as-
certain by proof in the quantum mechanical theory. By a basic measure-
ment sentence we refer to a sentence which can be directly ascertained
by a quantum mechanical measurement (with no further higher level
inferences for the ascertainment).
Completeness Question: Can the basic measurement sentences be
formalized as wff's in some quantum measurement language?
The question may be regarded a formal correlate to the question of
what a quantum physical quantity is, or how to demarcate such quan-
213
214 LARS LOFGREN
tities from other. The completeness question immediately leads to the

necessity of "levelling" quantum theory into types of inference.
For suppose that T is a sound quantum theory (everything provable
in T is true). If A is a verified basic measurement statement, and T
contains the inference (theorem) A=? B, i.e.
1-r (A=? B),
then, although B of course must be true, it does not in general follow
that B is also a measurement sentence. What are the inferences (theo-
rems) of T which preserve measurability?
Here the quantum mechanical concept of measurability turns out to
be confronted against (metamathematical concepts of) T-inferribility
(involving types of constructivity of the T-inferences). Linguistic realism
[6], rather than physical realism, comes into focus.
That such a confrontation is "real", and not just an "academic hair-
splitter", is seen from a recent challenge against the Bohr "wave-particle
complementarity" by Ghose, Home, and Agarwal [3] in terms of their
interesting double-prism experiment. As we are about to see, the chal-
lenge presupposes that no distinction is made between measurability and
inferribility. A rather simple separation between measurability and in-
ferribility, in terms of degrees of constructivity, will yield a constructivist
understanding of the wave-particle complementarity, whereby it is not
affected by the experiment.
By way of further examples, our understanding of the two processes
of ascertainment within linguistic realism is helpful for a critical under-
standing of an hypothesis of a programmable experimentability which
is suggested by Deutsch in connection with his proposals for "quantum
computation" (cf. [8]).
In our exposition we are using the concept of language in a quite
general sense, summarised as follows.
The Cornplernentaristic Conception of Language: In its com-
plementaristic understanding, the phenomenon of language is a whole
of description and interpretation processes, yet a whole which has no
such parts expressible within itself. This constitutes a paradigm for
complementarity, the linguistic complementarity. Any other known
form of complementarity, from proposals from Bergson to Bohr, have
been found [5] reducible to the linguistic complementarity, and the
reductions themselves do provide an understanding of the comple-
mentarities. There are various related ways of looking at the linguis-
tic complementarity:
(i) as descriptional incompleteness: in no language, its interpreta-
tion process can be completely described in the language itself;
METALINGUISTIC VIEWS OF QM 215
(ii) as a tension between describability and interpretability within
a language;
(iii) as degrees of partiality of self-reference (introspection) within
a language: complete self-reference within a language is impos-
sible;
(iv) as a principle of "nondetachability of language".
2. ARE THERE WELL-FORMED FORMULAS

WHICH COMPREHEND SETS, PROGRAMS,
OR MEASUREMENTS;
A COMPARATIVE UNDERSTANDING
In several familiar languages we meet the question whether it is possi-

ble to comprehend the essence of certain holistic concepts in terms of
decidable syntactic criteria of well-formedness on its describing formu-
las. We will make a comparative review of such questions in set theory
language, in programming language, and in quantum mechanical mea-
surement language. In the latter case, the question is whether decidable
syntactic criteria on the well-formed formulas for the basic measurement
sentences can be given, such that their interpretations will coincide with
the idea of quantum mechanical measurements.
Decisive for all three cases is the linguistic complementarity [5], [7]. We
find that for a full comprehension, an ultimate linguistic relativization
is unavoidable, which indicates shortcomings of pure syntactic criteria
of well-formedness.
In Set Languages, with the usual set-notation, S = { x : Px}, we
may look at the predicate Px as a description of the set S, or of the
set S as an interpretation of Px, in a set language L. If L is too rich,
allowing well-formed formulas like x ~ x, we know from the Russell
paradox that not all well-formed formulas can be interpreted as sets
(do not have extensions which are sets). The question is whether we
can give syntactic criteria on well-formedness for the formulas of L such
that the language becomes a proper set language. That is, where every
well-formed formula (set-formula) can be interpreted as a set, and where
every conceivable set can be described.
We meet such attempts at well-formedness in various proposals for
comprehension axioms.
Axiom of Typed Comprehension (used in Russell's theory of types; cf. [10]).
All variables are here typed, such that, if x E y, and x is of type n (an
integer), then y is of type n + 1. A well-formed predicate must here respect
this type condition. Thus, none of the predicates x E x, its negation x ~
216 LARS LOFGREN
x, or x E y & y E x, is well-formed. Any well-formed predicate Px is

comprehensible as a set S = { x: Px }; if x is of type n, then S is of type
n+ 1.
Axiom of Stratified Comprehension (used by Quine [14]). Any predicate
Px, which is well-formed in a stratified sense, is comprehensible as a set
S = {x: Px}. Here the variables are not really typed, but the requirement
of stratification on Px means, essentially, that in any subformula x E y of
Px, it is possible to assign integers to the variables such that the integer
for y is 1 greater than the integer for x. For an individual, however, and
only for individuals, we have x = { x} (which is impossible in the theory of
types).
Axiom of Relative Comprehension. For any predicate Px which is well-
formed in a set language without any type or stratification conditions, and
with any already established set y, there exists a set S that contains just
those elements x of y for which Px holds true, namely S = {x: Px & x E y}.
If y is not a set, neither is in general S.
The first two axioms try to secure set interpretability by restricting

the predicates, as objects for interpretation, by syntactic criteria of well-
formedness. Both are successful in the sense that consistent set theories
are obtained. But are they complete? Do the set theories describe all
sets (all objects which are naturally conceivable as sets)? Newer con-
structs, by Scott [15] and others, demonstrate sets with self-membership
(forbidden in Russell's typed set theory) much wider than that allowed
by Quine's stratified comprehension.
The axiom of relative comprehension, on the other hand, refers not
only to a pure descriptive well-formed part, Px (which is the only part
in the first two axioms), but also to a semantic part, namely an already
realized set y. That is of course a failure with respect to the goal of
a descriptive set theory. But it is a way out of the descriptive incom-
pleteness, with a taste of complementarity (cf. the "nondetachability of
language", or the need for both descriptions and interpretations, or the
necessity to relati vize the general notion of set to language). For further
comments, we refer to [5].
Next, consider a programming language L, where descriptions are
programs for a "universal" Turing machine and interpretations are the
corresponding computational behaviours (computation of partial recur-
sive functions). Only well-formed programs (formulas) are accepted by
the Turing machine to make it run.
We know from the linguistic complementarity that there can be no de-
scription (program) in L which completely describes its interpretations.
Let us try to make the interpretations more precise by moving from
the partial recursive functions to the total recursive functions. These,
unlike the partial recursive functions, are understandable according to a
classical function concept, which makes the objects clearly interpretable.
Then we cannot any longer describe within the language, in terms of
syntactic well-formedness conditions on the programs, which programs
will be interpretable (as total recursive functions).
This illustration of the tension aspect of the linguistic complementar-
ity for a programming language shows a limitation of the possibility of
imposing syntactic criteria on well-formedness of programs in order to
describe all and only the well understood (total) recursive functions.
Finally, a Quantum Mechanical Measurement Language pro-
vides a situation which is similar to that of the two previous cases. With
measurement a kind of interpretation [4], [9], quantum theory in the
form of the Schrodinger equation together with the projection postulate
forms a description (in a quantum theory, T, in a quantum mechanical
measurement language, L), attempting to describe (by the projection
postulate) its interpretation (cf. [6]). By the linguistic complementarity
we can hope for at most a partial success.
In this comparative perspective, it is doubtful whether we can provide
decidable syntactic criteria ofwell-formedness for the basic measurement
sentences such that these will have interpretations that coincide with the
quantum mechanical concept of measurement.
As we are about to see next, a similar conclusion is obtained by starting
out directly from the quantum theoretical concept of observable.
3. MEASURABILITY AND IN FE RRIB IL ITY;

A TIE IN TERMS OF LEVELS OF CONSTRUCTIVITY
We are confronted with two modes of ascertainment, by physical mea-

surement, and by linguistic syntactic inference. In general, the two modes
are kept apart by Cartesian or Heisenberg cuts, arguable in terms of
problems of complete self-reference. Forms of partial self-reference are,
however, legitimate. Which may also be expressed in terms of realiz-
able degrees of self-reference in a language, degrees of introspection in a
language, etc.
We will look into the possibility of performing, not another cut, but a
tie, let be loose, between physical measurability and linguistic syntactic
inferribility. Namely, in asking if they can have in common a lowest level
of constructivity (realizability) in a hierarchy of such levels.
In von Neumann's formulation of quantum mechanics [11] the observ-
ables correspond to self-adjoint operators acting on a Hilbert state space.
If A is a self-adjoint operator corresponding to some observable, then its
spectral values are interpreted as the possible values which one may
218 LARS LOFGREN
obtain in a measurement of this observable.

The characterization, or construction, of observables in terms of oper-
ators is obviously fundamental for the generation of basic measurement.
Primas [12], pp 62-3, in referring to pioneer quantum mechanics, explains
further how to construct new observables from old.
If A is a self-adjoint operator, then there exists a unique spectral resolution
E on the spectrum n of A such that
A= in w E(dw) .
... If A is a self-adjoint operator corresponding to some observable, then its

spectral values are interpreted as the possible values which one may obtain
in an ideal measurement of this observable .
.. . A real Borel function F of an observable A represents a new observable
F(A) which can be measured by the very same apparatus used for A by
replacing the scale of of its meter by a new one in which every number w is
replaced by F(w). In terms of von Neumann's spectral theorem, this means
that the spectral resolution of A
A= in w E(dw),
implies the spectral resolution of F(A)
F(A) =in F(w) E(dw).
We notice here a first trace of a merger between two ideas. On the

one hand, the idea of a mathematical construction of new physical (self-
adjoint) operators F(A) from old A, whereby new measurement values
F(w) result from old w.
This F-construction is, at least in von Neumann's original formulation
[11], without any restriction on F to be realizable in some constructivist
perspective. von Neumann writes, page 248:
"If the operator R corresponds to the quantity R, then the operator F(R)
corresponds to the quantity F(R) [F(.\) an arbitrary real function]."
And, on the other hand, we have the idea of an instrument con-
struction, of how to construct a new measuring instrument from an old,
where realizability conditions are obviously present. In order for one ex-
perimenter to effectively communicate to another how a meter scale is
to be obtained, he must resort only to constructivist processes.
There are no ties between the two ideas in von Neumann's quantum
theory with "F(.>..) an arbitrary real function". Most real functions are
not even computable.
We seem to have a real problem here. How is the quantum theory T,
its rules for well-formedness for the basic measurement sentences and its
rules of inference, to be formulated that we in T can decide which infer-
ences from measurement statements are again measurement sentences.
First of all, we have to impose on quantum theory the condition that
the F's be computable. Otherwise, we could think of quantum mechanics
as an effective phenomenon being able to answer noncomputable prob-
lems.
But such a computability restriction on the F's is not enough. It would
allow for quantum theory arbitrary complex inferences, only that they
are recursive (computable )-which every rule of inference, for any formal
theory, is anyway.
In order to give to quantum theory an intended meaning of measure-
ment theory, if not of a full measurability theory, it is necessary to equip
it with levels, distinguishing fundamental measurement inferences from
higher level (less constructive) inferences which indeed do occur in quan-
tum theory. Examples of such higher level inferences are theorems about
the noncomputability of the domain-problem for the quantum mechan-
ical operators [2]. The theorem that tunnelling (in the double prism ex-
periment; see the following section) is an "exclusive wave-phenomenon",
is also on a level higher than that of a basic measurement statement (cf.
[7]). The "superselection rules" may be taken to indicate a need to go
above first level rules for basic measurement statements. True, that in
Primas' algebraic theory (cf. [12, 13]), superselection rules are describ-
able. But not in a complete sense since there is reason to believe that
they are not decidable.
Quantum theory does indeed contain very complex inferences and, as
we will examplify in the next section, it may even be of physical interest
not to treat all its inferences on a par but to try to distinguish between
them in terms of levels of constructivity with physical relevance as well
as linguistic.
A development of levels which are both logical, like syntactic con-
straints on well-formedness, and also physical, like quantum physical
measurement constraints, is not likely to appear in some absolute way.
Such a solution would seem to imply a physical theory of our linguis-
tic cerebral processes (beyond mere measurements). Rather, it points
toward a necessary linguistic relativization with language in its comple-
mentaristic conception. The general philosophy of linguistic models (cf.
[6]) for quantum theory is a step in this direction.
In particular cases, the simpler idea of a quantum theory with only
two levels (basic measurement statements, and inferences which are not
basic measurement statements) may be quite helpful even without some
220 LARS LOFGREN
precise demarcation of the levels. The challenge from the double-prism

experiment of the Bohr wave-particle complementarity, may be taken as
an example.
4. LEVELS OF CONSTRUCTIVITY ENFORCED BY THE

DOUBLE-PRISM EXPERIMENT
Recently a "double-prism experiment" has been proposed by Chose,

Home, and Agarwal [3] as a challenge to the Bohr wave-particle comple-
mentarity.
In the experiment a "beam-splitter" in the form of a double-prism
is used. Since quanta are supposed to be indivisible, experiments to
split them are expected to exhibit revealing properties. The choice of a
double-prism as beam-splitter, instead of say a semitransparent mirror,
is interesting in a further sense. Namely, that it is then possible to infer
a simultaneous wave and particle nature of a single photon state of light
under investigation. This is in [1] argued to contradict the wave-particle
complementarity.
Coincidence
Anti
coincidence
Photon detection
L-T"""-1 counter
----•1
Single
photon
pulse
Beam Photon detection
spliuer counter
Figure 1. A single experimental arrangement to display both classical wave and

particle-like propagation of single photon states of light. After Chose, Home, and
Agarwal [1].
As illustrated, the double-prism prepares for a reflection path and a

tunnelling path. A source is used which emits a single photon state of
light. The prism gap is chosen such that if transmission along the tun-
nelling path occurs, which is indicated by a click in a photon detector
counter Dt in that path, then the transmitted phenomenon must have
wave-nature (not preventing a simultaneous particle nature). In the re-
flection path there is another photon detector counter Dr. Repeated
runs indicate strict anticoincidence (no coincidence) between the two
counters, supporting the hypothesis that the behaviour of the emitted
entities is particle like. Obviously, the experiment supports further hy-
potheses about a simultaneous wave and particle nature of the emitted
single photon states of light.
The inference of a wave-and-particle nature of the photons is suggested
(cf. [1]) as a falsification of the Bohr wave-particle complementarity.
However, as we have argued in [7], the inference of a wave-and-particle
nature is on a level which is above that of strict measurability. The wave-
nature of the entity which is transmitted along the tunnelling path is
never directly measured.
Therefore, the result of the experiment does not challenge Bohr's wave-
particle complementarity in its constructivist understanding preventing
a simultaneous direct measurement of wave-like and particle-like prop-
erties.
Our argument in [7] is based on the injection of linguistic information
levels for inferences in quantum measurement theory. These levels can
also be referred to complexity classes of realizing automata. Thereby
the concept of automaton will occur as the (loose) tie between physical
measurability and linguistic inferribility.
This is how we think of the double-prism experiment as highly inter~
esting. It raises the quest of a levelled approach to quantum mechanics as
a theory of measurement. Not with some arbitrary introduction of lev-
els. But with a hierarchy where, on a lowest level, physical constructivity
in terms of measurability will coincide with linguistic constructivity in
terms of metamathematical realizability.
5. C 0 N C L U S I 0 N S
Our central quest for wff's for basic measurement sentences may be
looked at as a modern realization of Bohr's plea for using natural lan-
guage with parts of classical physics for describing measuring instru-
ments and experimental findings-in the hope of reaching an unambigu-
ous communication of experimental results.
Our conclusions are that this seemingly simple quest for wff's for basic
measurement sentences in fact is too complex to allow a positive solution
in terms of formalism. What is needed is a complementaristic resolution
222 LARS LOFGREN
taking also experimentalism into account or, equivalently, a shift from

logics to language in its complementaristic conception.
As a modern experimentalist version of the quest, we want to mention
Deutsch's hypothesis of a programmable experimentability (sometimes
referred to as universal quantum "computation"; see [8]). Although an
interesting approach, there seems to be lacking an attachment to the
quest for wff's for basic measurement sentences (recall the complemen-
taristic nature of the problem).
We have contrasted the aim for communicable experimental ascer-
tainment by measurability, against von Neumann's early formulation of
quantum measurement theory-which turns to be the formal side of the
coin neglecting the constructivist (experimentalist) side.
The exposed partial tie between measurability and inferribility in terms
of a common level of constructivity, may be looked at as a continuation
of a historical development of connections, in terms of various informa-
tion concepts, between certain physical and cognitive quantities [16], [3],
[17].
AFFILIATION
Lars Lofgren
Department of Information Theory
University of Lund, Box 118, S~221 00 Lund, Sweden
lofgren@dit.lth. se
REFERENCES
[1] Chose P, HomeD, and Agarwal G , "An Experiment to Throw

More Light on Light", Physics Letters A, val 153, 1991, pp. 403-406.
[2] Hellman, G., "Constructive Mathematics and Quantum Mechanics:
Unbounded Operators and the Spectral Theorem", Jour-nal of Philosoph-
ical Logic, 22, 1993, pp. 221-248.
[3] Landauer, R., "Information is Physical", Physics Today, 44:5, 1991,
pp. 23-29.
[4] Lofgren, L., "Complementarity in Systems Thinking." In: Pmceed-
ings: of the European Congress on System Science, part I. Lausanne:
AFCET, 1989, pp. 293-316.
[5] Lofgren, L., "Complementarity in Language : toward a general un-
derstanding." In: Carvallo, M. (ed.), Nature, Cognition, and System II:
complementarity and beyond, Theory and Decision Library, Dordrecht-
London-Boston, Kluwer, 1992, pp. 113-153.
[6] Lofgren, L., "Linguistic Realism and Issues in Quantum Philoso-

phy." In: Laurikainen, K., and Montonen, C. (eds.), Symposia on
the Foundations of Modern Physics: the Copenhagen Interpretation and
Wolfgang Pauli, Singapore-New Jersey-London-HongKong, World Sci-
entific, 1993, pp. 297-318.
[7] Lofgren, L., "General Complementarity and the Double-Prism Ex-
periment." in: Laurikainen, K., Montonen, C., Sunnarborg, K.
(eds.), Symposium on the Foundations of Modern Physics 199..4; 70 Years
of Matter Waves, Paris: Editions Frontieres, 1994, pp. 155-166.
[8] Lofgren, L., "Nonseparability of Inferribility and Measurability in
Quantum Mechanics as a Systema Magnum." in: Trappl, R. (ed.), Cy-
bernetics and Systems '98, vol 1, Austrian Society for Cybernetic Stud-
ies, Vienna, 1998, pp. 113-118.
[9] Margenau, H., "The Philosophical Legacy of Contemporary Quan-
tum Theory." In: Colodny, R. (ed.), Mind and Cosmos, University of
Pittsburgh Press, 1966, pp. 330-356.
[10] Mendelson, E., Introduction to Mathematical Logic, third edition.
Monterey, CA, Wadsworth and Brooks, 1987.
[11] von Neumann, J., Mathematische Grundlagen der Quantenme-
chanik, Berlin, Springer, 1932.
[12] Primas, H., Chemistry, Quantum Mechanics and Reductionism.
Berlin, Springer, 1983.
[13] Primas, H., "Endo- and Exo-Theories of Matter." in: Atman-
spacher H. and Dalenoort G. (eds.), Inside Versus Outside, Berlin,
Springer. 1994, pp. 163-193.
[14] Quine, W., "New Foundations for Mathematical Logic." Amer.
Math. Monthly, 44, 1937, pp. 70-80.
[15] Scott, D., "A Different Kind of Model for Set Theory." Mimeogra-
phed (unpublished) paper read at the 1960 International Congress on
Logic, Methodology and Philosophy of Science, Stanford, 1960.
[16] Szilard, L., "Uber die Entropieverminderung in einen thermody-
namischen System bei Eingriffen intelligenter Wesen." Z. Physik, 53,
1929, pp. 840-856.
[17] Zurek, W., "Thermodynamic Cost of Computation, Algorithmic
Complexity and the Information Metric." Nature, 341, 1989, pp. 119-
124.
ENRICO GIANNETTO
QUANTUM TRUTH, NON-SEPARABILITY AND REALITY
1. INTRODUCTION
Since the beginning of the XXth century the appearance of quantum

physics has marked an epochal change in physics or at least in our in-
sight of it. An epochal change, however, which for many reasons is not
yet completely recognised. On one side, already at the level of its formal
(axiomatic) systematisation a rationalisation process has been operated
hiding the general as well as the particular problems. The general prob-
lems are related to the breakdown of every conceptual (waves or par-
ticles, etc.) schema and of every representation of the world (quantum
"truth" and "reality" are undetermined). Among the "particular" prob-
lems we can recall the questions of casuality and discontinuity, the needs
of a new mathematics and a new logic, the "violations" of general princi-
ples like the ones regarding the uniformity of nature or the conservation
of energy, etc. [1].
On the other side, physicists' community has been reducing quantum
physics just to a new, mere calculus and experimentation techniques:
techniques able to secure us with a wider, technical and practical, domin-
ion over nature even inside its fine structure. The microphysical domain
becomes a new source and fund for new instrumental uses of nature at
disposal of the human will to power. This was pointed out in Heideg-
ger's analysis, but in my opinion it holds only at the level of a sociological
description of the practices of physics [2].
Moreover, some physicists do not conceive quantum theory as a "uni-
versal theory" but, by cutting out the semantical differences, they con-
sider it as a theory with a very limited domain of validity in respect of
the wider domain of classical physics; or they also try to delegitimate
quantum theory just reducing it to a wider classical theoretical frame-
work.
Such a "psychological" inertia to recognise the epochal change involved
in quantum physics, in my opinion, is breaking down as long as it is show-
ing itself as devoid of any ground. Indeed, quantum physics allows us
to have a new insight also in classical and relativistic physics. Further-
more, the present "revolution" related to the so-called chaos physics is
showing us that the specific features of classical and relativistic physics
as long as believed to be in opposition to the ones of quantum physics
225
226 ENRICO GIANNETTO
are only illusory. Predictability, completeness, and indeed determinism

too, break down also for classical and relativistic physics: they reveal
themselves as theoretically and experimentally non-effective. The link of
chaos physics with quantum physics does not lay only on the ground of
the conception of quantum physics as a historical example of an indeter-
ministic physics, but is also historically realised within the works of Max
Born since fifties. Indeed, starting from the dialogue with Einstein on
quantum physics, Born has showed the unpredictability, indeterminism
and incompleteness within classical and relativistic physics: he showed
this just for the most simple systems, one-degree-of-freedom systems, and
already for a linear error propagation from initial conditions [3]. The ex-
istence of this link of chaos physics with quantum physics is the reason
I have said that quantum physics leads to an epochal change in physics
or at least in our insight of it.
This epochal change, in my opinion, involves the relations between
physis and logos themselves and leads to the breakdown of the logic, of
the epistemology and of the ontology of (classical) physics. And we have
to remember that classical physics has to be conceived as the actual
realisation of a process of rationalisation of the world, which is proper
own to the history of western metaphysics, and of the correlated process
of the technical dominion over nature by mankind [4].
From this point of view, I would like to deal with the so-called problem
of non-separability as it rises within quantum physics. I will briefly dis-
cuss the Einstein-Podolsky-Rosen paradox from a theoretical and episte-
mological perspective. Indeed, in my opinion, a principle of separability
of the world has been the first hidden presupposition at the ground of all
the physics since its origins. A principle which is the theoretical counter-
part of the political and technical divide et impera at the ground of the
whole rationalisation process of the world and of the technical dominion
over nature.
2. E P R: N 0 N- S EPA R ABILITY AND UN DEC IDA B I LIT Y
The question of non-separability within quantum mechanics, related to

the analysis of the Einstein-Podolsky-Rosen paradox, has been discussed
from several points of view. In my opinion, a technical and conceptual,
right solution to EPR was given on one hand-at least partially-in a
paper by J. Rothstein and, on the other hand, an independent, deep
interpretation of EPR pointing out non-separability was furnished by H.
Primas [5].
Here, it is possible to remember very briefly that Rothstein has ana-
lysed an EPR version which involves the separate measurements of com-
QUANTUM TRUTH & REALITY 227
plementary observables on both partial subsystems derived from the

"splitting" of a unique global system: the macroscopically registered re-
sults are revealed by a set of macroscopic "observers" as mutually non-
interfering and Lorentz-equivalent. These observers are within equiva-
lence classes, that is all the members of the same class agree on the
proper quantum descriptions of the results. However, members belong-
ing to different classes give mutually inconsistent, contradictory descrip-
tions, except the case in which wave functions present in their descrip-
tions are replaced by mixtures, that is density matrices. If one states that
quantum descriptions by wave functions were complete, they become mu-
tually contradictory. Vice versa, if the quantum descriptions are given
by statistical mixtures and hence they are incomplete, they become non-
contradictory.
The quantum mechanical description of individual subsystems 81 and
82 of a previous composed system 812 can be given only by terms of
mixtures as objective states of 81X 82 (non-pure states) just because the
measure process which allows us to obtain information is an irreversible
one. Rothstein has stated that in some way this argument allows to
deduce the second law of thermodynamics, through relativity, as a nec-
essary condition for the operational consistency of quantum mechanics.
There is, as noted by Rothstein, the breakdown of the myth of the
theoretical completeness and non-contradiction features of quantum me-
chanics, which has been constituting the rationalisation of Bohr's com-
plementarity and Heisenberg's indeterminacy relations. The presumted
completeness of quantum mechanics, which was used as an epistemolog-
ical obstacle in relation to the trials of formulating other theories (like
Einstein's unified field or hidden variables' ones) more complete than
quantum mechanics can no longer be accepted. However, even if such
incompleteness has to be partially understood as Einstein said, that is as
a constitutive inability of quantum mechanics to characterise completely
the presumed physical "reality", its deep meaning has to be related to
a Godel-like incompleteness [6].
Indeed, since Godel's theorems were published in 1931 (if not since
Lowenheim-Skolem appeared pointing out the fundamental limits of any
formal theory with a finite or enumerable number of axioms), it should
be clear that theoretical "completeness" is a myth. These theorems, as
well known, show the syntactical and semantical incompleteness (oth-
erwise the theory would be contradictory), the presence of undecidable
enunciates and the impossibility of proofs of validity, consistency and
non-contradiction, the expressive inadequacy, the non-characterizability
and non-unicity of models of any formal ("finite") theory [7].
In the EPR case, we have to deal with the undecidability between the
two mutually contradictory enunciates which correspond to the different

descriptions given by the two observers which could perform two different
measurements. Thus, from an epistemological point of view, the quantum
mechanics incompleteness does no more seem as an unpleasant feature
to be avoided, but a structural, logical characteristic of any theory: here,
incompleteness is not mathematically or logically a priori given, but it
follows from the logic of experimental, physical possible operations on
individual subsystems of a global system, which involve quantum truth-
gaps.
However, what does imply the need to use density-matrices to describe
all the subsystems of the world? Indeed, the concept of state for any in-
dividual subsystem loses meaning, breaks down. That is, non-separability
of the world is the physical counterpart, correspondent feature of the for-
mal use of density-matrices (non-separability is implied also in relation
to "spacelike" distances which do not allow any exchange of physical
interactions). Quantum "reality" is non-separable.
Coming now to Primas' analysis, it could be said that the world is a
whole which is not constituted by (separable) parts. From such a perspec-
tive, quantum physics radically differs from the reductionism and me-
chanicism involved as the paradigmatic background of classical physics,
by which complex phenomena were treated in terms of few fundamental,
elementary objects and of their interactions. Indeed, as Primas pointed
out, within the used quantum field theories and also grand unification
theories physicists go on to represent physical processes in a very superfi-
cial way, just ignoring such non-separability feature of quantum physics.
Nuclei are conceived as constituted by protons and neutrons, atoms as
constituted by nuclei and electrons, and so on. However, we are able
only to break from the outside atoms in protons, neutrons and elec-
trons, and these particles as free particles behave in a very elementary
way under the action of the proper kinematical group of transformations.
Atomism is definitely dead, and we cannot perform a tensorial-product
decomposition of physical "reality". The fact that we can disintegrate
nuclei, atoms, molecules in elementary systems does not imply that nu-
clei, atoms, molecules are constituted by elementary sub-systems: it is
not possible to attribute any kind of individual existence to such sub-
systems. For example, a molecule can be described only as an entangled,
non-separable system of protons, neutrons and electrons together [8].
3. N 0 N- S EPA R ABILITY AND THE

QUANTUM RELATIVISTIC FRAMEWORK
At the quantum-mechanical level, indeed, non-separability implies the

unobservability, non-measurability of physical variables related to indi-
vidual subsystems of the world [9]. If we consider the quantum-relativis-
tic-field theoretical framework, we have to recognise the unobservability,
non-measurability of fields related to individual particles which represent
the new dynamical variables of the theory.
This implies that regarding quantum mechanics we can use the wave-
function language only for the universe as a whole, whereas, regarding
quantum relativistic field theory, we should use field language only for
the universe as a whole. Otherwise, we have to use density-matrices for
individual particles. That is, it would become necessary to consider a
quantum-relativistic-field theory of the universe as a whole: indeed, such
a theory has independently begun to be analysed for another purpose in
the framework of a quantum cosmology as a so-called third quantization
theory [10].
Furthermore, another point has to be stressed regarding quantum rel-
ativistic field theory. Indeed, in quantum chromodynamics (QCD) and
then also in quantum electrodynamics (QED) there appeared correlation
phenomena which cannot be eliminated also over spacelike distances:
these phenomena have been explained just introducing the concept of
a dynamical confinement of fields and of correspondent confined phases
of the theory. We say that there is a breakdown of the cluster prop-
erty (for so-recalled unphysical fields) and of asymptotic completeness.
Indeed, we could say that these phenomena involve a sort of dynamical
non-separability. In my opinion, the analogy is very deep and fundamen-
tal and we can give a similar treatment: on one side, it suggests us the
possibility to conceive interaction phenomena on different scales as quan-
tum non-separability phenomena by operating a sort of kinematization
of interactions. On the other side, it suggests a new insight on quantum
non-separability as a sort of confinement (separability would involve also
an infinite indeterminacy on energy) and quantum incompleteness as a
sort of asymptotic incompleteness [11].
Indeed, a deep analysis, at an axiomatic level, on the dynamical con-
finement phenomenon has pointed out that individual separate fields are
non-local just because they are related to uneliminable specific interac-
tion correlations, and they cannot be considered as observables, measur-
able quantities: they are not physical, because they imply the breakdown
of the Lorentz-Poincare symmetry. That is, the non-phy~i::::al na""•1re of
such individual, separate, (non-local by interaction) fields, is needed from
the point of view of the preservation of the Lorentz-Poincare fundamen-

tal symmetry of the theory [12]. Separate fields are non-physical, ficti-
tious, mathematical constructions: they physically correspond to a whole
complex, non-elementary, non-separable "reality".
Thus, the relativistic analysis of EPR given by Rothstein must be
modified: the individual subsystems of the world are non-physical, just
because their definition imply a breakdown of Lorentz-Poincare symme-
try; they are confined within the universe. At a quantum-relativistic level,
the reality of particles loses meaning: only the universe is physical; non-
separability is the non-particle-like structure of the universe, the non-
divisibility by parts of the world. Is, however, a quantum-relativistic-field
theory (as well as a wave-function theory) of the universe possible? The
answer has to be negative because the universe as a whole is not observ-
able, it is not measurable as long as there exists nothing outside: any
observer is always within it, and the world as a whole is completely inde-
terminate. The only viable description is by density-matrices, related to
the various chosen observers, which show the impossibility of a separate
description of the observer and of the remaining part of the universe.
As David Finkelstein said, a quantum universe does not exist, but only
a universal quantum, that is, we have to deal with an irreducible multi-
plicity of non-separable physical processes which can never be described
as a uni-verse or a totality [13].
4. C 0 N C L US I 0 N S
First of all, we have to reconsider some questions. Is it true that in

classical and relativistic physics atomism, reductionism and mechanicism
are actually realised? That is, within classical and relativistic physics
does a principle of separability actually hold? No, it does not. Both
Mach and Poincare have respectively shown that classical and relativistic
dynamics imply non-separability [14]. Indeed, Mach has shown that the
origin of inertia has to be related to the presence of all the masses in the
universe; the universe is mechanically non-separable:
d2 /dt 2 ('E-imiri/'E-imi) = 0
Poincare, following this kind of argumentation has shown that the rela-
tivity principle for the universe is a non-separability principle and is like
a "gauge-invariance" principle whose transformations reduce to identity
when one considers only physical, invariant, relational- "universal" vari-
ables.
Moreover, Whitehead has given us a completely relational formulation
of special relativity and has further stressed that the relativity principle
is nothing else than a principle of universal relatedness, that is a prin-
ciple of non-separability of the world [15]. Recently, the logician Martin
Davis has shown that the indeterminacy principle is nothing else than
a relativity principle, and vice versa I have shown that the relativity
principle is nothing else than an indeterminacy principle [16]. Indeed,
non-separability was already a feature of Leibniz' dynamics, related to
the actual meaning of the so-called pre-established harmony [17]. From
this point of view, quantum and relativistic physics can be understood in
terms of radical processes of de-construction of the meta-physical New-
tonian interpretation of the language of mechanics and so also in terms
of radical modifications of Newtonian classical physics by introduction
of Leibnizian relationist developments. Thus, there is a breakdown of
the last illusions of classical and relativistic mechanics, as features of
the process of rational and technical dominion over the world, related to
the western meta-physics.
Indeed, since its Greek origins, at least after Anaximander, physi-
cal theoretization, as related to a philosophical physics or to a scien-
tific physics after Galilei, became meta-physics and nothing else than
a research of the etPXT/, that is of material or formal, onto-logical or
logico-epistemological grounds, from which starting to reconstruct and
to explain the variety of physical phenomena.
The indeterminacy of microphysical, elementary and simple entities
as material grounds of physis does not only imply the breakdown of any
kind of reductionism, but also the breakdown of such a project of ra-
tionalisation and dominion over nature. Reductionism presupposes the
possibility of a determination of the microphysical entities and there-
fore is nothing else than a particular specification of determinism. How-
ever, non-separability as a further indeterminacy of the physical pro-
cesses implies the non-physicality itself of elementary, simple, local and
individual entities through which it would be possible to (linearly) re-
construct the world. There is not only the impossibility of a material
ground for nature, but also of any formal one, that is of any onto-logical
or logico-epistemological ground. Indeed, the unobjectivability of any
first-quantization physical property regarding individual physical enti-
ties as logical subjects, related to unobservability (for non-separability),
implies the impossibility of any formal ground like the one exemplified by
energy and the breakdown of any subject-predicate logic. The unobjec-
tivability of any second quantization individual entities as unobservable
and non-physical (related to non-separability) implies the breakdown of
any logical reconstruction of the world in terms of the couple subject-
object (for non-separability) and so of any epistemological ground. The
unobjectivability of the third quantization world as a totality-universe
implies also the breakdown of any holistic ontology of the world being
and of any representation of the world: that is the breakdown of any

reduction of the world to representation or image. From a logical point
of view, hence, there is no possibility to define the objects themselves of
a physical universe of discourse [18].
The world shows itself as non-separable and indeterminate, like Anax-
imander's TO cnr£Lpov: indeed, the separability of the world by parts
or particles was the first hidden presupposition of any numerical-quan-
titative determination of physical processes, the first presupposition of
any measurement, of any >.o-yoc.; as ratio (for the first time introduced
by Pythago-rean mathematical physics), of the determinacy principle
which lays down the non-contradiction principle, as well as of any prin-
cipium individuationis of both objectivity or subjectivity. Thus, by the
recognition of a quantum radical non-separability of the world, there is
the breakdown of the presuppositions themselves of any physical, philo-
sophical or scientific, representation-theoretization and therefore of the
presuppositions of the quantum physics itself.
Hence, at variance with Heidegger's analysis which was embedding
quantum physics within a technical-scientific paradigm of dominion over
nature which would not be possible to overcome as long as it has its roots
in the life world ( Lebenswelt) of the human will to power; at variance with
the seeming sociological features of the "quantum establishment" within
the physicists' community; at variance with the apparent new forms of
the technical dominion over the world allowed only by a pmgmatical and
statistical use of quantum physics; quantum physics, from indeterminacy
to non-separability, has given actually rise to an epochal change, pointing
out the lack of any foundation ground of the process of the rational and
technical dominion over nature by which the human species has been
constituting its peculiarity, and showing the unavoidable limits which
nature presents in respect to its reduction to a resource fund (Bestand) of
exploitation at disposition of the human will to power, and in respect to
the constitution of its absolute mastery by men ( Gestell) [19]. We have no
more to comprehend the world but just to be com-prehended within the
world: non-separability implies a sort of physical hermeneutics, related
to a physical, non-ontological Dasein (indeterminate, non-separable part
of the world).
Would it be possible that such a consciousness, required more and
more from our world transformed by technics itself as realized by such
a science, overcomes and transforms the life world itself which has given
rise to it? Or should we wait for the complete breakdown of such his-
torically and genetically determined life world under the heaviness of its
proper own contradictions? Could be enough for us in such a waiting
and hope the consciousness of the physical irreducibility of nature (of
the illusory character of life world dominion over nature), whereas na-
ture is even subjected to any seeming-even if non-effective -violent
dominion of the life world at the interspecific, intra-specific, ecological,
biological, social and political levels? In any case, physics, beyond any
ethical indication, shows us a physis of which, conscious or not, we are
parts belonging to that secret love which is the non-separability of the
world.
AFFILIATION
Enrico Giannetto
Dipartimento di Fisica "A. Volta"
Universita di Pavia
Pavia
It alia
REFERENCES
[1] For these issues, see for example: Giannetto, E., "La logica quan-
tistica tra fondamenti della matematica e della fisica" in: Bartocci,
U. & Wesley, J.P., Blumberg (eds.), Foundations of Mathematics
€3 Physics, 1990, pp. 107-127; Giannetto, E., "Toward a Quantum
Epistemology" in: Dalla Chiara, M. L. & Galavotti, M. C. (eds.),
Atti del Convegno S.I.L.F.S. Temi e Prospettive della Logica e della
Filosofia della Scienza, CLUEB, Bologna, 1988, pp. 121-124; Gian-
netto, E., "L'epistemologia quantistica come metafora antifondazionis-
tica" in: Petruccioli, S. (ed.), Immagini Linguaggi Concetti, Theoria,
Roma, 1991, pp. 301-322; Giannetto, E., "On Truth: A Physical In-
quiry" in: Cellucci, C. & Dalla Chiara, M., Atti del Congresso 'Nuovi
problemi della logica e della filosofia della scienza', CLUEB, Bologna,
1991, pp. 221-228; Giannetto, E., "Il crollo del concetto di spazio-
tempo negli sviluppi della fisica quantistica: l'impossibilita di una ri-
costruzione razionale nomologica del mondo" in: Boniolo, G. (ed.), As-
petti epistemologici della spazio e del tempo, Borla, Roma, 1987, pp. 169-
224.
[2] Heidegger, M., "Die Frage nach der Technik" in: Vortrage und
Aufsatze, Neske, Pflillingen, 1954; Heidegger, M., "Wissenschaft und
Besinnung" in: Vortrage und ... , op. cit.; Giannetto, E., "Heidegger
and the Question of Physics", in: Kiss, 0. & Ropolyi, L. (eds.), Pro-
ceedings of the "Conference on Science and Hermeneutics", Veszprem
1993, Reidel, Dordrecht (in press).
[3] Giannetto, E., "Note sulla complessita: Max Borne la nascita della
nuova fisica del caos", in: Cellucci, C. and Dimaid, M.C. (eds.), Atti
del Congresso, ETS, Pisa, 1994, pp. 317-330.
[4] See [1], [2], [3].
[5] Rothstein, J ., "Physics of Selective Systems: Computation and Bi-
ology", International Journal of Theoretical Physics, 21, 1982, pp. 327-
350; Primas, H., in: Miller, A.l. (ed.), Sixty-Two Years of Uncer-
tainty, Plenum Press, New York, 1990, p. 233; Primas, H., Chem-
istry, Quantum Mechanics and Reductionism, Springer Verlag, Berlin,
1983; Primas, H., Time-Asymmetric Phenomena in Biology: Comple-
mentary Exophysical Descriptions Arising from Deterministic Quantum
Endophysics, preprint, LFC-Zurich, 1988; for a general overview, see
also: D'Espagnat, B., Conceptual Foundations of Quantum Mechan-
ics, Benjamin, Menlo Park, 1971; Bohm, Hiley, The Undivided Uni-
verse, Routledge, London, 1993; Kitchener, R. (ed.), The World View
of Contemporary Physics, State University of New York Press, Albany,
1988.
[6] Giannetto, E., "The Epistemological and Physical Importance of
Godel's Theorems" in: Walkowski, Z.W. (ed.),First International Sym-
posium on Godel's Theorems, World Scientific, Singapore, 1993, pp. 136-
147.
[7] See references given in the paper quoted in [6].
[8] See papers quoted in [5].
[9] See also the demonstration of the impossibility to attribute a wave
function to a single (one-particle) well-defined physical system in quan-
tum mechanics, given in the following paper: Preparata, G., What is
Quantum Physics? Back to the QFT of Planck, Einstein and Nernst, lec-
ture given at the IX Winter School on Hadron Physics, Folgaria (Italy)
6-13 February 1994, MI-TH 94/3 preprint. However, in my opinion, it
is not enough to realize a many-particle theory to deal with the non-
separability problem.
[10] Caderni, Martellini, "Third Quantization Formalism for Hamil-
tonian Cosmologies", International Journal of Theoretical Physics, 23,
1984, p. 223.
[11] Giannetto, E., Teoria quanto-relativistica delle fasi macros co piche
della materia condensata: la transizione fiuido-solido, PhD thesis, Uni-
versity of Messina, Messina, 1992; Giannetto, G., "Towards a Quan-
tum-Relativistic Understanding of the Phases of Matter: The Fluid Solid
Transition", Physics Essays, 6, 1993, pp. 98-109.
[12] See [11] and Kugo, Ojima, Supplement to Progress in Theoretical
Physics, 66, 1979, p. 1.
[13] Finkelstein, D., "The Universal Quantum", in: Kitchener, R.

(ed.), The World View ... , op. cit., pp. 75~89.
[14] Mach, E., Die Mechanik in ihrer Entwickelung historisch-kritisch
dargestellt, Brockhaus, Leipzig, 1883; Poincare, H., "L'espace et le
temps", Scientia, 12, 1912, pp. 159~171; Giannetto, E., "Henri Poin-
care and the rise of special relativity", Hadronic Journal, Supplement,
10, 1995, pp. 365~433; Giannetto, E., Lectures on Relativity, mimeo-
graphed paper, University of Pavia, Pavia, 1993.
[15] Giannetto, E., "Mach's Principle and Whitehead's Relational For-
mulation of Special Relativity" conference delivered at the International
Congress on 'Mach's Principle', Tubingen, July 1993 (in press); White-
head, A.N., An Enquiry on the Principles of Natural Knowledge, Cam-
bridge University Press, Cambridge, 1919; Whitehead, A.N., The Con-
cept of Nature, University Press, Cambridge, 1920; Whitehead, A.N.,
The Principle of Relativity with applications to Physical Science, Cam-
bridge University Press, Cambridge, 1922.
[16] Giannetto, E., "On Relativity Theories and Leibniz" conference
delivered at the International Conference 'Albert Einstein', Ulm, March
1992 (in press); Giannetto, E., Heidegger and ... , op. cit.; Giannetto,
E., Note sulla ... , op. cit.; Davis, M., "A Relativity Principle in Quan-
tum Mechanics", International Journal of Theoretical Physics, 16, 1977,
p. 867.
[17] Leibniz, G.W., "Leibnizens mathematische Schriften" Gerhardt,
C.G., (ed.), Halle, 1850~63; Cassirer, E., Leibniz' System in seinen
wissenschaftlichen Grundlagen, Elwert, Marburg 1902 & Wissenschaftli-
che Buchgesellschaft, Darmstadt, 1962; Cassirer, E., "Erkenntnisprob-
lem in der Philosophie und Wissenschaft der neuren Zeit" Berlin, 1911 ~
1920; Giannetto, E., Lectures on Leibniz, mimeographed paper, Uni-
versity of Messina, Messina, 1988; Giannetto, E., Lectures on the His-
tory of Energy Concept, mimeographed paper, University of Messina,
Messina, 1991.
[18] See references given in [1].
[19] See references given in [2].
GEOFFREY HUNTER
SOLITON WAVES VS. THE PARTICLE PARADIGM: THE

ELEMENTARY NATURE OF THE PHYSICAL WORLD
1. PARTICLES AND WAVES
Whether the physical world is composed of indivisible atoms, or alter-

natively is an infinitely divisible continuum, is a long standing question
in natural philosophy. The proposition that the world is composed of
atoms (the smallest particles) is a natural extension of the physics of
macroscopic solid bodies to a microscopic domain. Solid bodies are the
tangible constituents of the macroscopic world, whereas commonly ob-
served waves (e.g., in water) occur as the concerted, oscillational motion
within a physical medium apparently consisting of material particles in
the liquid or gaseous state rather than as a solid. Thus waves have been
regarded as a phenomenon in physical systems rather than the funda-
mental basis of the physical world. Specifically Newton postulated that
light is composed of a stream of particles, and after the discovery of the
wave properties of light (diffraction and interference) it was thought to
be a wave motion within some underlying medium (the "ether").
The question of whether the elementary nature of the physical world
is composed of particles or of waves, became a paradox in the early part
of this century (1905-30) through the discoveries that:
- light exhibits both wave properties (diffraction and interference)
and particle properties (Compton and photo-electric effects),
- electrons exhibit the wave properties of diffraction and interference
as well as behaving as particles having mass and electric charge.
Orthodox ("Copenhagen") quantum theory accepts the paradox with-
out attempting to explain it. That both light and electrons behave as
particles in some experiments and as waves in others is accepted as an
inexplicable phenomenon of quantum mechanics. This orthodox philos-
ophy asserts that it is not possible to know the nature of the photon or
the electron, and that it is the act of measurement that causes the wave
or the particle to manifest itself.
The general thesis of this article is that both light and electrons are
localized ("soliton") waves; their wave properties are derived from their
intrinsic wave nature, and their particle properties are derived from the
237
238 GEOFFREY HUNTER
localized nature of the wave. Thus the paradox of the wave-particle du-
ality is, in principle resolved because the electron or photon is a localized
wave whose centre behaves like a particle.
2. THE NON-ELEMENTARY NATURE OF PARTICLES
The concept of a Particle is that it is point-like, its size possibly being fi-
nite but small (on some scale). For the particle concept to be elementary
(i.e., without internal structure that must be described in terms of more
elementary concepts), the "particle" would have to exist at a point, or
it would have to be a rigid body-probably a rigid sphere. If the ele-
mentary particle existed at a point, and if it is something physical (as
distinct from mathematical), then the elementary physical object would
have to have an infinite energy density, which is physically untenable;
there is no physics at a mathematical point. The other possibility of
the particle being a rigid body would require the transfer of momentum
across the particle's finite size during a collision to imply either infinite
forces within it (which is also physically untenable), or alternatively an
infinite speed of transmission of the impact across its finite size, which
is inconsistent with the principle that the velocity of light is the upper
limit for the transmission of all material interactions. These arguments
are from Lucas and Hodgson [1]. Hence the concept of Particle is not
Elementary: the nature and structure of particles (such as electrons, pro-
tons, etc) must be described in terms other than "particle" -probably
as waves-the other principal paradigm of theoretical physics. The con-
cept of elementarity is of something that is conceptually selfsufficient;
thus an elementary finite-sized particle would have to be a rigid body,
otherwise we would need some more elementary property to describe its
internal elasticity.
3. WAVES ARE ELEMENTARY
A wave always has a finite extent-the energy of the wave is distributed

throughout its volume; the energy density is finite everywhere. When a
wave collides with another physical object it deforms: the impact spreads
across the extent of the wave at a finite speed (often equal to, and always
no larger than the velocity of light). The forces involved are finite. When
a wave collides with a "solid" object its direction of propagation will be
bent (i.e., diffraction). Since the angle of bending (around an edge or
through a slit) will depend upon the impact parameter of the wave with
the edge, this bending can also lead to "interference"; i.e., preferential
bending (higher intensity) in certain directions. The classical concept of
ELEMENTARY NATURE 0 F PHYSICAL W 0 R L D 239
"interference" (ala Huygens principle) is probably artifactual as in the

classical, Maxwellian theory developed by Richard Prosser [2].
4. THE DE BROGLIE HYPOTHESIS
Einstein (1905) quantified the idea that mass is simply a form of energy
in his famous equation:
(1)
Scientists have tended to regard particles (especially the electron and
proton) as "elementary" constituents of nature, the mass of the particle
being one of its most distinctive properties. The idea that a particle is a
localized wave was introduced by de Broglie [3]. By combining Einstein's
equation (1) with Planck's equation:
E = hv (2)
relating the energy, E of a quantum "oscillator" to its frequency v, he
postulated that an elementary particle such as an electron is some kind
of localized (soliton) wave whose frequency of oscillation is related to its
mass by:
(3)
That is what we regard as the (inertial) mass of the particle is, ac-
cording to de Broglie's proposal, simply a manifestation of a localized
oscillation of a field (most likely the electromagnetic field). From this
standpoint mass is not an elementary property of a particle, but rather
a property derived from the localized motion of the (electromagnetic)
field.
For an electron, the computed frequency of the wave is v = 1.2356 x
1020 sec- 1 (Hertz), corresponding to a wavelength of>.= cjv = 2.4263 x
10- 12 metre.
5. THE PART I C L E WAVE AND THE 0 B SERVER
De Broglie postulated that the time-dependence of this wave motion in

the rest-frame of the particle would have the form:
7/J = exp(i21fvt) = exp(iwt) (w = 21rv) (4)
and he then performed a Lorentz transformation on this time-factor from
the electron's rest frame to the frame of a observer moving at velocity v
(along the x axis) relative to the electron by:
t = 1(t' + x'vjc 2 ) (5)
240 GEOFFREY HUNTER
where 1 = 1IJ1- (vlc)2 and the primed variables are the coordinates
(x', t') in the observer's frame of reference. Thus the wave (4) becomes
in the observer's frame (x', t'):
·l/J = exp(i2nvlx'vlc2 ) x exp(i2nv1t') (6)

which becomes (using equation (3) to eliminate the frequency v in the
space factor),
1/J = exp(i2nlx'mvlh) x exp(i2nv1t') = (7)

exp( i2nx' I)..) x exp( i21rv1t')
where)..= hl(!mv) is the de Broglie wavelength of the electron moving

with velocity v and dynamic (relativistic) mass 1m. The space factor
in equation (7), exp( i2nx 1 I>-.) is simply Schrodinger's wavefunction for a
free particle (electron). This result is very remarkable; from the postulate
that the electron is (internally) some kind of wave motion, one can de-
rive (using nothing more than a Lorentz transformation) the Schrodinger
wavefunction for the particle as seen by an "observer" such as a diffrac-
tion grating. It seems to have the potential to relate Schrodinger's wave
mechanics to Special Relativity and the theory of classical (electromag-
netic) waves. Such a unification would be a big step forward in theoretical
physics. Note that the time factors of (4) and (7) are (for low velocities,
1 ~ 1) virtually identical.
Schrodinger's wave is a kind of "beat frequency" (much lower fre-
quency and longer wavelength than de Broglie's internal electron wave)
produced by the relative motion of the particle and the observer.
This idea has the potential to resolve the long-standing problem of
the wave-particle duality paradox; how can the electron sometimes be-
have like a particle and at other times as a (Schrodinger) wave ? The
answer is that the particle is always a (de Broglie) wave localized around
its centre of energy (mass) that we tend to regard as a point particle.
Its Schrodinger wave is a manifestation of its internal wave motion as
experienced by a relatively moving observer (or apparatus).
6. PH 0 T 0 N S AS S 0 LIT 0 N WAVES
The Photon has been modeled as a soliton wave [4]. The model is a
nonplane wave solution of Maxwell's equations that has the correct spin-
angular momentum of the photon (±n), the two values corresponding
to right and left circularly polarized photons. Although the mathemat-
ical form of the electric and magnetic fields is obtained by solution of
the linear Maxwell equations, they are constrained by the relativistic
ELEMENTARY NATURE 0 F PH Y S I CAL W 0 R L D 241
principle of causality to lie within a circular ellipsoid whose long axis is

the axis of propagation. The length of the ellipsoid is one wavelength
and its cross-sectional circumference is also one wavelength; hence it is a
prolate ellipsoid whose diameter is approximately one-third (accurately
1/7T) of its length. This causally constrained soliton model of the Photon
predicts some of the experimental properties of monochromatic light:
- the optimal resolving power of a monochromatic microscope is sta-
ted in textbooks of elementary physics to be "a little less than a
third of a wavelength"; 1/7T is "a little less than a third",
- the minimum slit-width (or orifice diameter) required for trans-
mission of circularly polarized light is A.j1r; this effective photon
diameter was accurately confirmed by our own experiments with
microwaves [4] within the experimental error of a half a percent,
- an antenna in the form of a helix whose diameter is equal to its
pitch divided by 7T, produces a narrow beam of circularly polarized
radiation whose wavelength is equal to the pitch of the helix [4],
- the threshold intensity for the production of multiphoton-absorption
in focussed laser beams is predicted by this photon diameter of A.j1r;
the threshold beam intensity corresponds to adjacent photons in the
beam touching each other. We have called this intensity the photon's
intrinsic intensity, because it is the average light intensity within the
photon's cross-sectional circle of diameter A.j1r perpendicular to its
direction of propagation.
Its value is given by:
(8)
For example at 523 nm (middle of the visible spectrum) this intrinsic

intensity is 1 megawatt per square centimetre. At beam intensities higher
than the intrinsic intensity photons necessarily overlap; i.e., there are
two or more photons in the same place at the same time, which is the
required condition for multiphoton absorption to occur [4].
This prediction of the threshold beam intensity for multi photon absorp-
tion to occur correctly predicts the requisite experimental intensities
(typically produced in focussed laser beams). The predicted threshold
intensity has also been confirmed by multi-photon absorption experi-
ments with radiation from a CW C02 laser (>.. = 10.5 micrometres, Ip
= 6 watts per square centimetre); thus the expression (8) scales with
wavelength (as A. 4 ) correctly.
242 GEOFFREY HUNTER
7. THE W A YE-P ARTICLE DUALITY PARADOX
Recognition that the concept of particle is not elementary, and that

the elementary concept of wave can be used to construct "elementary"
particles (photons, electrons, etc.) that have both particle-like (locality
and momentum) and wave-like (diffraction and interference) properties,
may be the way to resolve the Wave-Particle Duality Paradox [5].
The concept of particle must be abandoned as an elementary ingredi-
ent of physical theory (as distinct from a pragmatically useful, approx-
imate concept). Our further understanding of the nature of the "parti-
cles" of physics is likely to involve soliton waves.
This conclusion suggests that the currently-in-vogue paradigm in the-
oretical physics (in which interactions between particles are represented
by exchange of other particles) should be abandoned in order to describe
the physical world in terms of truly elementary concepts.
AFFILIATION
G. Hunter
Department of Chemistry
York University, Canada
REFERENCES
[1] Lucas, J.R., and Hodgson, P.E., Spacetime and Electromag-

netism, Clarendon Press, Oxford, 1990.
[2] Prosser, R.D., "The Interpretation of Diffraction and Interference
in terms of Energy Flow", International Journal of Theoretical Physics,
Vol. 15, 1976, pp. 169-180.
[3] Theses presentees par Louis de Broglie, premier these "Recherches
sur La Theorie des Quanta", Masson et Companie, Editeurs, Libraires
de l' Academie de Medecine, Paris, 1924.
A reconsideration and extension (in English) has been given by Mioara
Mugur-Schachter ("Quantum Mechanics and Relativity: Attempt as a
New Start", Found. of Physics Letters, Vol. 2, 3, 1989, pp. 261-286.
A translation of Chapter 1 of de Broglie's thesis entitled The Relation-
ship Between the Quantum and Relativity was published (American
Journal of Physics, Vol. 40, Sept. 1972, pp. 1315-1319).
[4] Hunter, G., and Robert L.P., "Wadlinger Photons and Neutrinos
as Electromagnetic Solitons", Physics Essays, Vol. 2, 1989, pp. 158-172.
[5] Diner, S., Fargue, D., Lochak, G., and Selleri, F. (eds.), The
Wave-Particle Dualism, D. Reidel, Dordrecht, 1984.
DAN NESHER
"WHICH SIDE SPINOZA WOULD HAVE TAKEN (BETWEEN

EINSTEIN AND BOHR) IF HE HAD LIVED TO SEE THE
(SCIENTIFIC) DEVELOPMENT OF OUR DAYS": AN
ANALYSIS OF HUMAN REPRESENTATION OF THE
PHYSICAL REALITY
1. INTRODUCTION: THE QUEST TO UNDERSTAND

SCIENTIFIC REPRESENTATION OF REALITY
Bohr tells us that in his discussion with Einstein at Princeton in 1937,

regarding the understanding of modern science, they did not get beyond
a humorous debate on "which side Spinoza would have taken if he had
lived to see the development of our days" [7, 237]. It is well known that
an important element in Einstein-Bohr controversy about modern sci-
ence was its philosophical-epistemological interpretation (e.g., [7]; [14,
pp. 683-4]). Hence, the issue of this article is the epistemological prob-
lems concerning the interpretation of science of our day in the framework
of evolutionary epistemology. Elsewhere I interpreted Spinoza's theory
of knowledge as evolutionary epistemology [30], [32] and I consider his
philosophy of nature (as distinct from "natural philosophy") as the best
basis for philosophy of science. The quest here is to understand science as
one of the human cognitive modes of representing reality. I will analyze
the epistemological controversy between Einstein and Bohr in regard
to the interpretation of scientific theories and their capacity to repre-
sent reality, and compare their philosophical positions with Spinoza's. I
believe that through the philosophical discussion of the epistemological
problem of scientific representation of reality we can advance to a better
understanding of our scientific theories. This is expressed forcefully by
Einstein:
The reciprocal relationship of epistemology and science is of notewor-
thy kind. They are dependent upon each other. Epistemology without
contact with science becomes an empty scheme. Science without epis-
temology is-insofar as it is thinkable at all-primitive and muddled
([14, pp. 683-4]; cf., [8, p. 1]).
We may understand this philosophical controversy between Einstein and
Bohr as similar to our present-day controversy between Metaphysical Re-
alism and Phenomenalism (or Internal Realism), respectively (cf., [42,
243
244 DAN NESHER
Ch. 12]; [38, pp. 49-54]; [40, Ch. l]; [11]; [10]). I will attempt to ana-
lyze this problem in the framework of Spinoza's naturalism (as distinct
from physicalism) and Peirce's logic of cognition (semiotics). I believe
that Spinoza and Peirce reject both, Metaphysical Realism and Internal
Realism, and take a third, pragmatist, perspective, which I call Repre-
sentational Realism (cf., [29], [30]). There are various ways to explicate
realism (and therefore anti-realism). I would like to give a wider defini-
tion for realism than Dummett's, and to suggest that it has two tenets
that only together explicate it: (1) the ontological tenet, that there is
reality which exists independently of its cognitive representation; and (2)
the epistemological tenet, that this reality is represented by our cognitive
minds: we know it. Dummett defines realism as accepting (1), but instead
of (2) has a stronger condition: "that reality renders each statement in
the class determinably true or false, again independently of whether we
know, or are even able to discover, its truth-value" ([11, p. 55]; cf., [38,
p. 49ff.]). The question is whether, under my explication of realism, the
"metaphysical realist" and the "internal realist" are realists at all. The
metaphysical realist accepts naively the existence of external reality (1)
but since the truth of a statement is determined by reality independently
of whether we know it or not, it is not clear whether humans can know
that there is such external reality (2). The internal realist, by reject-
ing the possibility of knowing reality outside of our cognitive states (1 ),
must also be skeptical, like Hume, or categorical, like Kant, about the
existence of this mysterious transcendental reality (2).
The question of realism is whether we humans can know that there
is reality external to our cognitive experience, and if so, what is its na-
ture. This is Peirce's distinction in between the negative and the positive
knowledge of reality. The first is grasped in the way we learn that there
is something different from us; the second in the way we learn that there
is something which corresponds to our cognition. The negative knowl-
edge of reality evolves already in the child's awareness of the real by
being in error or in ignorance, but we are aware of it in every surpris-
ing fact frustrating our expectation [36, 5.233-234, 5.311]. In this case
we are aware that by knowing that what we expected is an erroneous
subjective cognition, we also know that there is something independent
refuting our ideas, and "to this we give the name of the rear' [37, vol.3,
8]. Popper, with his celebrated theory of refutation actually accepted
only the negative concept of reality, while in his later writings he tried
in vain to show how we should leap from frustrated hypotheses into the
eternal truths.
According to Spinoza and Peirce we can achieve representational
knowledge of external reality only through our causal interaction with
WHICH SIDE SPIN 0 Z A W 0 U L D HAVE TAKEN ? 245
such reality. Our positive description of real objects is based on our

"natural instinct for truth", namely, reacting positively to these objects
in the perceptual process (cf., [44, 17#31; [36, 7.220]; [31], [32, #7]).
Peirce meant to understand reality as independent from human cog-
nition representing it, since without being forced from outside, how can
investigators progress toward one and the same conclusion? The question
for Peirce is whether we can cognize or describe positively "some exter-
nal permanency", "something upon which our thinking has no effect"
[36, 5.384], which is independent of "the thinking part of the mind" [36,
7.338]. The questions are, in what sense is there an "independent real-
ity" of the human thought representing it (cf., [36, 5.553, 5.564]), and in
what sense does "reality" depend on (connect to) our thought or other
cognitions by which it is represented (cf., [36, 1.578, 7.336-345, 5.405-
410]; on the controversy between Einstein and Bohr about the definition
ofreality cf. [17, 7.2]).
2. THE EPISTEMOLOGICAL CONTROVERSY BETWEEN

EINSTEIN AND BOHR ABOUT PHYSICAL REALITY
AND ITS SCIENTIFIC REPRESENTATION
The above question of representation of external physical reality became

more problematic owing to the discovery of the atomic ("quantum")
reality revealing new behavior of its objects, deviating from the behavior
of objects of classical physics (cf., [8, III, p. 60], [8, II, pp. 71-73], [7,
pp. 222-3, 233-34], [5, pp. 237-8]; [4, p. 176]). Hence, we should be
careful in using the concept object for the entities of the atomic domain
in order not to picture to ourselves the classical concept of object with
its classical properties (cf., [8, I, p. 53]; [13, p. 71]). This new situation
calls for either a new theoretical explanation with non classical concepts
referring to non-classical properties of this revealed atomic reality or a
new philosophical epistemology, or both (cf., [8, I, p. 92ff.], [14, pp. 683-
4]; [22, p. 128ff.]).
Nature seems to be somehow a continuous whole and the classical
distinction among discrete objects and between the subject observer and
object observed become obscure and problematic (cf., [8, I, pp. 53-57,
62-68], [8, II, pp. 91-3]).
The philosophical controversy concerning the understanding of the
wholeness of the atomic ("quantum") world, is presented by two seem-
ingly unsatisfactory opposing positions: Einstein's "classical [atomic] re-
alism" and Bohr's phenomenalism (camp. [43, pp. 158-160). This is in-
deed a very crude characterization of both epistemological positions but
it is essentially correct. Especially in regard to Bohr's philosophy and
246 DAN NESHER
epistemology of science there are many different interpretations, and

the question is whether we can ascribe to Bohr a third position, a sort
of realism which is neither classical realism nor anti-realism (e.g., [20,
pp. 21-27]; [23, Chs. 5, 8]; [28, pp. 195-199!-245], [18, Ch. VIII]).
Assuming these two philosophical positions, we face a hard dilemma:
either to accept the classical separation of objects, even for the atomic
domain, in order to maintain the external reality and its representation
independent of the observer and his measuring instruments, or rather,
to retain non-separability of the wholeness of experimental phenomena
and renounce reality independent of its representation by the cognitive
mind. These two philosophical attitudes fed the controversy between
Einstein and Bohr for many years and their analysis is important for the
clarification of the epistemological problem of scientific representation of
reality (cf., [14, pp. 11-13, 672-674]; [5, pp. 145-6], [8, II, p. 2], [8, III,
pp. 1-7]).
Bohr in his article "Discussion with Einstein on Epistemological Prob-
lems in Atomic Physics" [7], about their controversy regarding the un-
derstanding of modern science, writes:
Not least through a new discussion with Einstein in Princeton in
1937, where we did not get beyond a humorous contest concerning
which side Spinoza would have taken if he had lived to see the de-
velopment of our days, I was strongly reminded of the importance
of utmost caution in all questions of terminology and dialectics [7,
pp. 236-237].
From my understanding of Spinoza's philosophy, he would not have
taken either side. This is because both Einstein and Bohr are right
and wrong in respect of Spinoza's conception of reality and its cog-
nitive representation. Spinoza theorized Nature as a continuous whole
and suggested that the mental and physical domains have such a spe-
cific connection that despite the wholeness of Nature there is no modal
causal relation between cognitive minds and physical systems but only
a real union of mind and body such that the represented physical reality
remains independent of its cognitive representation [30], [32].
Therefore, in respect to Spinoza's philosophy, Einstein is right in
claiming that the represented reality is independent of its representa-
tion by the cognitive mind [14, p. 81]; but he is wrong in his conception
of reality absolutely independent of its measurement that nevertheless
can be represented completely by scientific theory [16]. Einstein's posi-
tion is expressed clearly in some programmatic paragraphs of his famous
paper written with Podolsky and Rosen (EPR):
WHICH SIDE SPIN 0 Z A W 0 U L D HAVE TAKEN ? 24 7
"In a complete theory there is an element corresponding to each

element of reality. A sufficient condition for the reality of a physical
quantity is the possibility of predicting it with certainty, without
disturbing the system. [16, p. 138]
If, without any way disturbing a system, we can predict with certainty
(i.e., with probability equal to unity) the value of a physical quantity,
then there exists an element of physical reality corresponding to this
physical quantity. [16, p. 138]
While we have thus shown that the wave function does not provide
complete description of the physical reality, we left open the question
of whether or not such a description exists. We believe, however, that
such a theory is possible" [16, p. 141]; cf., [14, pp. 666-72, 681-84],
[15, pp. 276-7]; comp. [21, p. 66]; cf., [19, pp. 266-271]).
These formulations, especially the (originally) emphasized "this crite-
rion ... of recognizing a physical reality", and similar others (e.g., [13],
[14], [15], remind us of Peirce's nominal definition of reality where the
ideal scientific theory corresponds to reality and represents it completely
(5.407f.). With charity we could interpret these conceptions of complete
theory and reality as nominal (ideal) definitions, but then, if only ideal
theory is complete and true; and since there are non-ideal theories the
controversy with Bohr would be vacuous because not only quantum the-
ory but every scientific theory is incomplete (cf., [5, p. 148], [7, pp. 230-
235]; [14, pp. 83, 666-674]; [43, p. 153]; Born in [15, p. 277]). Nevertheless
it seems reasonable to understand that what Einstein meant by arguing
for the incompleteness of the quantum theory is that it is indeterministic
and unrealistic in respect to its representational function of "individual
systems"; according to this interpretation Einstein believed that quan-
tum theory violates the basic characters that are required from physical
theory: determinism and realism, and that even in atomic physics this
kind of completeness of theories can be achieved (cf., [14, 13, pp. 672-3],
[15, pp. 276-7]; [20, pp. 143ff., 222]; [19, pp. 259(E), 264ff.]).
Bohr, also in respect of Spinoza, is right and wrong; he is right "that
no sharp separation can be made between an independent behavior of
the objects and their interaction with the measuring instruments which
define the reference frame" [7, p. 224], namely, in the physical domain;
but he is wrong in that by looking for complete certainty in science and
knowing that we have no complete control over the relation between the
measuring instrument and the measured system, he decided to avoid
speaking about the measured system existing independently of our cog-
nitive representation; the result was that he restricted "physical reality"
to our phenomenal description of the "whole experimental arrangement"
and thus endorsed a direct interaction between mind and objects (e.g.,
248 DAN NESHER
[7, pp. 233-4], [5, p. 146ff.], [8, II, pp. 72-73]). Bohr's phenomenalism
(and instrumentalism) is expressed clearly in the following:
In our description of nature the purpose is not to disclose the real
essence of the phenomena but only to track down, so far as it is pos-
sible, relations between the manifold aspects of our experience ([8, I,
p. 18]; cf., [8, I, p. 4]).
On the line of objective description [i.e., the unambiguous commu-
nication-e.g. [8, III, p. 3], it is indeed more appropriate to use the
word phenomenon to refer only to observations obtained under cir-
cumstances whose description includes an account of the whole ex-
perimental arrangement. In such terminology, the observational prob-
lem in quantum physics is deprived of any special intricacy ... [8, II,
p. 73]; cf., [6, p. 24]; comp., [17, p. 58]).
Strictly speaking, the mathematical formalism of quantum mechanics
and electrodynamics merely offers rules of calculation for the deduc-
tion of expectations about observations obtained under well defined
conditions specified by classical physical concepts [i.e., common-sense
communicable physical concepts-[8, II, p. 73] ([8, III, p. 60]; cf., [8,
II, pp. 3, 68, 71], [7, pp. 222-3, 233-34, 237-8]).
Bohr's philosophical position was such that while he took the effect of
the measuring instrument on the measured object (system), and even of
the human body on the former, as parts of the experimental perceptual-
observational process, he despaired of the possibility of describing "un-
ambiguously" their interactions in the physical realm (cf., [8, II, p. 39],
[8, III, p. 3]; [23, Ch. 3.5]). In order to avoid ambiguous (non-objective)
language in the description of this uncontrolled relation, he invented his
principle of "complementarity". This enabled him to apply the math-
ematical formalism of quantum mechanics to different descriptions of
separated experimental arrangements without relating the different re-
sults to an independent reality of a particle; hence "an unambiguous
meaning can be attributed to such an expression as 'physical reality' "
([5, pp. 145, 151], cf., [8, II, pp. 1-2], [8, III, pp. 2-7]; [22, p. 179]).
According to Bohr, "the interaction between the measuring instru-
ments and the objects forms an integral part of the phenomena" [8,
III, p. 4]. The phenomena are the "experimental arrangements", and
the "evidence about atomic objects obtained by different experimental
arrangements exhibits a novel kind of complementary relationship" [8,
III, p. 4]. Since in quantum physics evidence of different experimental
arrangements "appears contradictory when combination into a single
picture is attempted" Bohr concluded that even if there is such reality
that determines causally our phenomenal experience we cannot know it
WHICH SIDE SPIN 0 Z A W 0 U L D HAVE TAKEN? 249
unambiguously (objectively); hence our "purpose is not to disclose the

real essence of the phenomena" [8, I, p. 18]; yet, such contradicting evi-
dence can be explained away with the notion of complementarity which
"exhausts all conceivable knowledge about the object" [8, III, p. 4]. But
what is this object? Is it experientially "in" the phenomena or transcen-
dentally "behind" them?
Let us analyze Bohr's phenomenalism and his problem with the con-
ception of the physical reality of science:
The epistemological problem under discussion may be characterized
briefly as follows: For describing our mental activity, we require, on
one hand, an objectively given content to be placed in opposition to
a perceiving subject, while, on the other hand, as is already implied
in such assertion, no sharp separation between object and subject
can be maintained, since the perceiving subject also belongs to our
mental content [8, I, p. 96].
This discussion is about the "objectivity of phenomena" of the quan-
tum domain, which is different from the classical perceived material ob-
jects. Bohr put this epistemological problem in phenomenological terms
which echo Kant's distinction between the cognitive states ( Vorstellun-
gen) and their content or sum (Inbegriff), i.e., the appearance, which is
viewed as their object (cf., [24, B236/ Al91, A105]). Bohr follows Kant's
phenomenological holism, in claiming that there is no sharp separation
between subject and object, since both the object and the perceiving
subject belong to our mental content. In Weizsacker's words we find a
telling interpretation of Bohr's Kantian position that the objects are the
"sum" or the "content" of the phenomena themselves:
The fact that classical physics breaks down on the quantum level
means that we cannot describe atoms as "little things". This does
not seem to be very far from Mach's view that we should not in-
vent "things" behind the phenomena. But Bohr differs from Mach
in maintaining that "phenomena" are always "phenomena involving
things", because otherwise the phenomena would not admit of the
objectification without which there can be no science of them. For
Bohr, the true role of things is that they are not "behind" but "in"
the phenomena [47, p. 185].
This Kantian epistemological and ontological position, which was prob-
ably imparted to Bohr by Hofding, his philosophical teacher and friend,
preceded the formulation of quantum theory and helped Bohr in its
epistemological explanation (cf., [18, Chs. III, VIII]). It is interesting to
see whether phenomenology can enable Bohr to show the completeness
of quantum theory by explaining away complementarily the different
250 DAN NESHER
contradictory results of different atomic experiments in respect of "the

same atomic object" (cf., [20, p. 156ff.]; [23, Ch. 3.6]). In explicating his
concept of "phenomenon" and his phenomenological epistemology Bohr
writes:
It is certainly more in accordance with the structure and the interpre-
tation of the quantum mechanical symbolism, as well as with elemen-
tary epistemological principles, to reserve the word "phenomenon"
for the comprehension of the effects observed under given experi-
mental conditions. These conditions which include the account of
the properties and manipulation of all measuring instruments essen-
tially concerned, constitute in fact the only basis for the definition of
the concepts by which the phenomenon is described ([6, p. 24]; [20,
pp. 157-8]; cf., [8, II, p. 73]).
The question is whether Bohr's atomic object is the cognitive content of
different experiments or if he also had to invoke external reality beyond
our "objectively" given mental content. As to the latter, is this atomic
object a Kantian transcendental object that somehow determines the
experimental phenomena but is outside our scientific knowledge? Or is
it the pragmaticist real external object represented by such knowledge?
(comp., [20, p. 156]; [28, 10.10]).
In an interesting analysis Faye shows the development of Bohr's expli-
cation of his concept "phenomenon" in the course of his philosophical dis-
cussion with Einstein, and especially in the context of the EPR thought
experiment. Thus before 1935 Bohr's definition of "phenomenon" was
the atomic object to which different effects of the different experimen-
tal arrangements are ascribed as its different property aspects, and thus
assuming phenomenon as the same object independent of the experimen-
tal measuring arrangement. Later he redefined "phenomenon" such that
the special effects of different experimental arrangements are "different
types of quantum phenomena" ([6, p. 22], [8, II, pp. 64, 71-74], [8, III,
pp. 2-7]; cf., [18, pp. 191-95]). Foles interprets Bohr's position as a kind
of realism which combines scientific phenomenalism with metaphysical
realism:
Thus Bohr makes it obvious that such classical terms as "position"
and "momentum" are "deprived of all meaning" apart from the con-
text of their application to describe particular observation interac-
tions of phenomenal objects as they appear in specific observational
interaction ... He does not assert that the very notion of such an
independent reality is itself without meaning or that there is no need
to refer to such atomic object in the description of observation as
interaction ([20, p. 156]; my emphasis).
WHICH SIDE SPINOZA WOULD HAVE TAKEN? 251
It is clear, therefore, that the phenomenal objects of different contrasting

observations cannot be "the same object" to which all these different
described properties are related. (comp. [20, pp. 164-5]). The outcome
must be to ascribe the contrasting properties to non phenomenal physical
object and this results in the same paradox of using ambiguous language
to describe this object, which because of this cannot be the objective
scientific language of science.
It is interesting to see that Bohr, while trying by his complementar-
ity to avoid the phenomenal paradox of contrasting phenomena related
to different experimental arrangements (e.g., particles or waves), had to
introducing a new paradox between complementary "contrasting phe-
nomena" and the underlying referred atomic object which functions like
the ideal "single picture" of the "atomic system" or like "a consistent
picture of the object under investigation" of classical physics but with-
out its consistency (cf., [8, III, p. 4ff]). The phenomenalist way out of
this paradox, namely, avoiding the ascription of contrasting properties
to "the same atomic object", is to declare this physical object indescrib-
able. Thus Bohr's general epistemological lesson for the entire scientific
enterprise must restrict scientific knowledge to phenomenal objects only,
and maintain that the external reality, though probably causing our ex-
perience nevertheless cannot be the object of science or, in Kant's terms,
cannot be unknown. This reality is at most an "abstract" metaphysical
presupposition, namely, a transcendental object. But if contrasting prop-
erties can be ascribed to such an object how do we know that it is "the
same atomic object"? (cf., [20, p. 244ff.]). In this context it is interesting
to see how Folse defends Bohr's position as realist, while admitting it
to be basically a Kantian position which combines "empirical realism"
with the "transcendental object" (cf., [24, A370]). This is actually done
by accepting, as Kant did, Burne's position that
It is both impossible and unnecessary to expect that natural science
justifies its realistic outlook by empirically demonstrating that there
is a reality existing independently of experienced phenomena [20,
pp. 242-3].
Avoiding this Humean-Kantian position and taking the Spinozist-Peir-
cean perspective we can see that representing external realities is exactly
what science does by proving or justifying its theoretical claims about
them (cf., [32], [33], [34, IV]). However, owing to the problem of explain-
ing epistemologically the quantum theory Einstein and Schrodinger ar-
gued about the intrinsic incompleteness of this theory. Bonner explicates
Bohr's phenomenology and its difficulty with a comprehensive descrip-
tion of physical reality:
252 DAN NESHER
Bohr's notion of objectivity differs from the classical account, how-

ever, in that he stresses that our descriptions of nature are not de-
scriptions of independently existing realities, but descriptions of our
encounter with such realities ... Quantum physicists, restricted to
describing experiments in terms of the whole apparatus-system in-
teraction, find themselves in the same position as people trying to
describe in detail the totality of their mental activity, "since the per-
ceiving subject also belongs to our mental content" ([8, I, p. 96], [23,
pp. 146-147]; cf., [20, pp. 154-167]).
Thus for Bohr the "physical reality" is the "atomic observed phenomenal
object". Moreover, this observed atomic object itself is at best only the
phenomenal features of the measuring instrument which are interpreted
as the measured atomic object. The different conceptions of physical
reality of Bohr, Einstein and Spinoza can be shown in the following
scheme of the experimental arrangement: (1) The Experimental Situa-
tion- Observation-Measurement and Represented "Physical Reality" for
Einstein, Bohr, and Spinoza:
Person Observing
Measuring ~ ~ Phenomena Mind

Measured Instrument .: .: (Bohr's Reality) .. ... ... .. .. .. .. .... .
Object
Human Body
(Einstein's Objective Reality
II
Hence Bohr is essentially phenomenalist claiming that scientific knowl-
edge is only the "descriptions of our encounter with such [independently
existing] realities". Accordingly, the description of human phenomeno-
logical experience is the only reality available to humans and this is
exactly what Kant dubbed "empirical realism" (cf., [24, A370]; comp.
[34, II]; [23, Ch. 5.2]). It is interesting that in the above scheme Ein-
stein's concept ion of physical reality is separated from Bohr's physical
reality, namely, what is outside or external to our experience. According
to Einstein we represent physical reality through our sensual encounter
with the Measuring Instrument but can describe this reality as being
independent and undisturbed by being measured. For Spinoza (if he
had lived to see the development of modern science) there is a contin-
uous interaction between Human Body, Measuring Instrument and the
Measured Object and thus the Measured Object cannot be really inde-
pendent and undisturbed, and yet, through this phenomenal encounter
W HI C H SIDE S PIN 0 Z A W 0 U L D HAVE TAKEN ? 253
we can learn and represent relatively truly the relatively independent

measured atomic object.
3. CRITICISM 0 F ' PH EN 0 MEN A LIST ' AND

'CLASSICAL REALIST' EPISTEMOLOGIES
In this context it is interesting to read Peirce's criticism of the phe-

nomenalist-positivist position of Comte, Poincare, and Pearson, while
suggesting the pragmaticist alternative about scientific theories, verifi-
cation and reality:
An explanatory hypothesis, that is to say, a conception which does
not limit its purpose to enabling the mind to grasp into one a variety
of facts, but which seeks to connect those facts with our general con-
ception of the universe, ought, in one sense to be verifiable; that is to
say, it ought to be little more than a ligament of numberless possible
predictions concerning future experience ... But Comte's own notion
of a verifiable hypothesis was that it must not suppose anything that
you are not able directly to observe. From such a rule it would
be fair to infer that he would permit Mr. Schliemann to suppose he
was going to find arms and utensils at Hissarlik [according to his
hypothesis about the city of Troy], but would forbid him to suppose
that they were either made or used by any human being, since no
such being could ever be detected by direct perception ... Comte,
Poincare, and Karl Pearson take what they consider to be the first
impression of sense, ... and they separate these from all the intellec-
tual part of our knowledge, and arbitrarily call the first real and the
second fictions. (Peirce, 5.597 [1903]; bold letters added).
In the same line Schrodinger argues against the doctrine of contemporary
quantum mechanics that holds that "models with determining parts that
uniquely determine each other, as do the classical ones, cannot do justice
to nature", and therefore, requires a rejection of realism [43, p. 153].
Schrodinger writes:
Reality resists imitation through a model. So one lets go naive realism
and leans directly on the indubitable proposition that actually (for
the physicist) after all is said and done there is only observation,
measurement. Then all our physical thinking thenceforth has as a
sole object the results of measurements, which can in principle be
carried out, for we must now explicitly not relate our thinking any
longer to any other kind of reality or to a model. [43, p. 157].
Thus the framework of complementarity as "a ligament" of contrasting
phenomena became a license for the phenomenological understanding of
physical reality: "a radical revision of our attitude ["the customary view
254 DAN NESHER
point of natural philosophy"] as regards physical reality" ([5, pp. 151,

149]; cf., [8, I, p. 93ff.], [5, pp. 145-6, 149-151], [8, II, v-vi, Ch. II], [8,
III, pp. 1-6!-7], [7, pp. 209-211]). Schrodinger put it in a witty remark:
"Bohr wants to complement away [wegkomplementieren} all difficulties"
(Schrodinger, in [35, p. 425]; comp. [16, p. 139]; the alternative (2); [14,
p. 674]; [43, pp. 153-155!-160]; [9, pp. 20-28]; [17, Ch. 3]; [4, Ch. 2]). Bohm
and Hiley have the same kind of criticism of Bohr's phenomenology:
Bohr would never allow the type of language that admitted the in-
dependent existence of any kind of quantum object which could be
said to be in a certain state. That is to say, he would not regard it
as meaningful to talk about, for example, a particle existing between
quantum measurements even if the same results were obtained for a
given observable in the sequence of such measurements. Rather, as
we have seen, he considered the experimental arrangement and the
content (meaning) of the result to be a single unanalysable whole
([4, p. 18, also 23]; cf., [7, pp. 232-4]; comp. [24, Bxxv-xxvi]; cf., [1,
p. 51]; [45, p. 63]).
In the same line Espagnat criticizes this phenomenalist-operationalist
approach which in order to maintain certain and complete physical
knowledge restricts physical reality to operational statements and their
perceived results.
Indeed, strictly speaking, nothing is really certain to us except our
operations. If science is required to be certain, then it can make
only operational statements. However, operational assertions have no
meaning except in reference to the community of the operators ... If
I demand that science be certain, then the notion of any scientific
object whatsoever reduces completely and is totally exhausted by the
notion of a given set of operations that we perform and of the results
that our mind can perceive ([17, pp. 130-131], cf., [17, pp. 17-19];
[43, p. 153]).
It is interesting to see that from Bohr's (and Heisenberg's) phenomenalist
interpretation of quantum mechanics, which does not distinguish clearly
between human knowledge and its object, some physicists inferred the
effect of the former upon the latter (cf., [21, p. 83]). In this theoretical
situation Stapp went even further to conclude an "idealike" physical
world, which is similar to the Leibnizian world of "Monades" (cf., [45,
Ch. 11]).
I think that part of Einstein's and Bohr's philosophical mistakes, from
a Spinozist-pragmaticist point of view, is that both were somehow influ-
enced by Kantian philosophy yet, they embraced different aspects of it
(cf., [13, pp. 61-62], [14, pp. 672-674]; [27, p. 249]; [17, p. 63]; [18, xix,
pp. 172, 197-211]; [25]). Einstein believed in the reality of things in them-
selves completely independent of human activity but that nevertheless
can (miraculously) be known (cf., [18, xix, 180]); and Bohr viewed phys-
ical reality as identical with the experiential phenomena, namely, depen-
dent on human cognition (cf., [8, I, pp. 93, 103]; [18, p. 211]; [46, p. 59];
[4, pp. 16-19]; [45, pp. 65, 117-119]; [41, p. 51])]. Thus, Einstein and
Bohr were involved with "the so-called 'deep truths"', in Bohr's terms,
namely, "statements in which the opposite also contains deep truth" [7,
p. 240]. For example, when Einstein speaks of "incomplete description"
of physical reality by quantum theory he means the reality of things in
themselves, while Bohr by "complete description" means the formalism
of quantum theory as applied to the observed phenomena only (cf., [5,
p. 145ff., pp. 149-150], [7, pp. 222-224, 237-8]; [43, pp. 153-4, 157-160];
[14, p. 681ff.]). On the incompleteness of Quantum Theory from there-
alist (not the naive realist) perspective and its apparent completeness
from the phenomenological standpoint, Schri:idinger writes:
The rejection of realism also imposes obligations. From the stand-
point of the classical model the momentary statement content of the
\If-function is far from complete; it comprises only about 50 per cent
of a complete description ["the other half then remains completely
indeterminate" - p. 132]. From the new standpoint it must be com-
plete for reasons already touched upon at the end of sect. 6 [p. 157].
It must be impossible to add on to it additional correct statements,
without otherwise changing it; else one would not have the right to
call meaningless all questions extending beyond it [as Bohr does] ([43,
p. 159]; comp. [14, p. 668ff.]).
Thus, returning to Einstein's and Bohr's philosophical positions in their
1935 controversy about scientific theories, their completeness, and the
nature of physical reality, it seems to me that they are both some-
how wrong. First, they are wrong because they hold uncritically a basic
"philosophical instinct" to believe in the completeness of scientific the-
ories; but from Spinoza's and Peirce's pragmaticist point of view there
cannot be a complete (certain) description of external reality since hu-
man knowledge of it is limited and fallible. Moreover, Einstein's "cri-
terion of reality" is too strong since there is no completely independent
"undisturbed" reality, and Bohr's "conception of reality" is too restricted
(devised to ensure complete knowledge) and thus his "phenomena" can-
not describe external physical reality (cf., [43, pp. 153, 155, 157ff.]; [17,
p. 65ff.]). However, I suggest that when a pragmaticist epistemology is
adapted "we are nearing the goal where logical order to a large extent
allows us to avoid deep truth", where contradicting positions are both
256 DAN NESHER
true and false (Bohr's wording, [7, p. 240]). It is true that the represen-
tation of the atomic domain depends upon the cognitive mind but we
should not identify, as Kant and Bohr do, the phenomenal-representation
with the represented reality (cf., [33]). In order to differentiate these two
components, representation and physical reality, we should analyze the
observational-measurement situation.
In a theory of representation a distinction should be made between
the role of the perceiver-observer in measurement and the relation be-
tween the measuring instrument and the measured object (system). The
measuring instrument is what the experimentalist perceives and the mea-
sured system is what she observes through the perception of the former.
As Schrodinger expresses it:
The systematic arranged interaction of two systems (measured object
and measuring instrument) is called measurement on the first system,
if a directly-sensible variable feature of the second (pointer position)
is always produced within certain error limits when the process is
immediately repeated (on the same object, which in the meantime
must not be exposed to any additional influences) [43, p. 158].
We can measure an atomic system only mediately when perceiving a
middle-size measuring instrument interacting with this system. If we can-
not perceive the measuring instrument we cannot experimentally mea-
sure and thus cannot observe an atomic System even where an interac-
tion between them occurs (comp. [7, p. 209]). However, we should be
careful not to confound perception with measuring operation and un-
derstand that even perception and measuring operation are two compo-
nents of observation. Hence, the measuring instrument can operate and
affect the measured system without being perceived; and also we can
perceive the measuring instrument when the latter is not in operation
and therefore does not affect the system to be measured. In perception
(as a part of the observational procedure) the relation of the mind to
the measuring instrument is not mediated as such. This, of course, leads
to a serious question as to whether "Schrodinger's cat", as an instru-
ment measuring the decay of the radioactive substance, can be in an
indeterminate superposition if it is always to be perceived. It seems that
Schrodinger's intention in his famous thought experiment is to show that
according to the doctrine of contemporary quantum theory "an indeter-
minacy originally restricted to the atomic domain becomes transferred
into macroscopic indeterminacy, which can then be resolved by direct
observation"; but then if there is an indeterminacy in the measuring in-
strument itself (the cat), there cannot be any observation of the atomic
system. Hence, if this is the result of the quantum theory, it "prevents
WHICH SIDE SPIN OZA WOULD HAVE TAKEN? 257
us from accepting as valid a "blurred model" [quantum theory] for rep-

resenting reality" [43, p. 157]. Here Schrodinger distinguishes between
the incomplete quantum theory which he metaphorizes as "a blurred or
poorly focused photograph" , and the complicated quantum reality itself
which he metaphorically describes as "clouds and fog patches" (cf., [43,
pp. 157, 155] [translated by [26, p. 197]]). Lockwood mistakenly ascribes
Schrodinger's second metaphor for a quantum "complete theory", i.e.,
"a picture of clouds and fog patches", to the incomplete quantum the-
ory of wave functions which according Schrodinger is only "a blurred or
poorly focused photograph" (cf., [43, pp. 153, 157-8] and [26, p. 197]).
Therefore, the macroscopic cat as a measuring instrument and its health
condition as the pointer, cannot be indeterminate, uncertain, but always
"within certain error limits" [43, p. 158].
But serious misgivings arise if one notices that the uncertainty af-
fects macroscopically tangible and visible things, for which the term
"blurring" seems simply wrong. [43, p. 156].
Here as elsewhere Schrodinger uses "blurring" for an incomplete theory,
and not for a cloudy and foggy reality as we have already seen above; but
the "macroscopically tangible and visible things" are neither "blurring"
nor "cloudy and foggy" things. Thus the apparent paradox is that if
the measuring instrument is an indeterminate quantum system then its
"pointer" (e.g., the eat's health condition) cannot possess a well-defined
position; but if its well-defined position is a necessary condition for exper-
imental evaluation of quantum theory, then the quantum theory, which
predicts a not well-defined position of the pointer, is impossible; but
then the pointer can have a well-defined position and quantum theory
is possible, and so on. The way out of this paradox is to show that the
Measuring Instrument operates at the "quasi-classical" level and has
"quasi-locality" parts such that the prediction of their position by quan-
tum theory are "always produced within certain error limits" (cf., [43,
pp. 156-158]; [8, III, p. 5]; [14, pp. 670ff.]). This is actually the argument
of Bohm and Hiley.
The immediate experience in this world is that which is described by
what is called common sense ... Within the domain of such experi-
ence it may be said that this [the overall atomic-e.g., [4, pp. 161,
178] world is manifest . ... it is what can be held in the hand, the eye,
and, of course, scientific instruments. Its [the quasi-classic manifesta-
tions] basic characteristic is that it contains certain relatively stable
structures that make the holding possible. These structures must not
only be relatively stable, but also essentially local. ... Without such
a [sub-]world we would not be able to make sense of our observations
258 DAN NESHER
of matter ... [4, p. 176]

The distinction between the quasi-classical sub-world and the atomic
(quantum) overall-world is the ontological counterpart of the epistemol-
ogy distinction I made between perception and observation in the mea-
surement procedure. In perception there is an immediate, direct, and
familiar relation between a person and a middle-size ("quasi-classical")
measuring instrument, i.e., "that which is described by what is called
common sense", while in observation the representation of the observed
system is theoretical, mediated by the measuring instrument (cf., [17,
11.1]; [34, III, IV[8]]). Bohm and Hiley's enterprise here is to avoid a
seemingly unabridgeable dichotomy between the common-sense classical
world and the scientific atomic world and suggest to including the quasi-
classical world as a limited case of the atomic world, thus permitting an
explanation of perception and communication [4, Ch. 8].
The reception of a small number of quanta gives only the vaguest
sense of optical stimulation. Meaningful perception requires a large
number of quanta and therefore, along the lines we have already
explained, this will imply an essentially classical behavior. Rather
we are simply calling attention to the observed fact that meaningful
sense perception and communication has to go through the [quasi]
classical level in which the effects of this wave function can be essen-
tially left out of account ... the overall quantum "world" can manifest
itself in the more limited [quasi-]classical "sub-world" [4, p. 178].
Thus for observation to be materialized, the measuring instrument must
be a "quasi classical" object which can be perceived; in this case the
interaction between the human body and the perceived measuring in-
strument is also a "quasi-classical" relation, namely, its components are
"quasi-separated" and "quasi-localized"; this leaves the measuring in-
strument essentially independent of the human body of the perceiving
person and, therefore, observation is possible. (cf., [4, 8.6]). But even
taking into account the effect of the perceiver-observer as a physical
body upon the measuring instrument and through it upon the measured
system (due to physical non-separability), it is still not in respect of the
representational function of "the thinking part of the mind".
4. T H E RELATIVE IN DE P END EN C E 0 F
REALITY AND COGNITIVE MIND
The properties of the atomic system after the measurement are not al-
together independent of the measurement interference but, according
to my analysis, are independent of their cognitive representation owing
to this measurement (cf., [33, IV]). However, regardless of how much
the properties of human body are entangled with the properties of the
perceived measuring instrument the epistemological distinction must be
between the representing Mind on one side, and the measuring instru-
ment on the other. But the same argument is even stronger for the
representation of an atomic observed system, which physically cannot be
distinguished completely from the measuring instrument and the body of
the observer; they constitute together an extended continuous natural
system whose components are only modally (relatively) separated (the
essential non-separability of the physical domain). Therefore, any possi-
ble division among all three of them is only relative in respect of their
modes of being. Their modal interaction causes modifications, new real
magnitudes or even new particles with their specific properties, since the
idealization of the atomic systems cannot be like the classical idealized
objects (separated and localized) (cf., [7, pp. 201-2], [8, II, p. 71], [8,
III, p. 2]; [4, Ch. 8 & pp. 386-390]). And yet, the measuring process is
determined, like any natural process according to the laws of nature and
cannot "disturb" or "interrupt" these laws. (cf., Spinoza, Ethics IIIPr).
Schrodinger states rightly that,
... in the realism point of view observation is a natural process like
any other and cannot per sebring about an interruption of the orderly
flow of natural events [43, p. 158]; cf., [43, p. 160#10]).
This is the case since observation qua measurement is itself a natural
process, yet, as such it really changes the flow of some specific events, but
according the laws of nature and without interrupting them. Now, some
interpreters of quantum theory suggest that the observation-measuring
operation due to the intervention of consciousness "creates" new realities
that were not in the interacting modes of being before. That is to say,
consciousness brings about an interruption of the orderly flow of natural
events (cf., [9, pp. 20-26]; camp. [1, p. 54ff.]; [4, Ch. 2[pp. 16-19, 218]]).
But if the physical observation-measuring process is separated from the
mental process of perception, and if our minds have no causal effect upon
the represented physical objects, then why should this atomic process be
understood differently from other natural physical processes? It seems
to me that the problem with Bohr's approach to quantum theory is
that he continued to use the classical concepts of particles and waves
and thus had to adapt the Kantian phenomenology as his philosophical
epistemology of science in order to describe unambiguously the resulting
experimental phenomena. This was expressed clearly by Weizsacker:
Niels Bohr is the only physicist in our time who-as far as I know,
without having been influenced by Kant-proceeded from fundamen-
tal insight similar to Kant ... Only in this framework will physicists
260 DAN NESHER
be able to do justice to Bohr's doctrine of the indispensability of

classical concepts ([47, p. 342ff., 345]; cf. [8, I, pp. 8, 17, 53, 77, 94],
[8, II, pp. 26, 39, 72, 88], [8, III, pp. 3, 12, 24]; comp. [3, p. 245ff.]).
The Spinozist alternative considers nature with its natural processes as
a non-homogeneous continuum with different degrees and intensities of
bulks of density and fields of thinness in which the interactions of pro-
cesses change the structural patterns of these processes. At the quantum
level of this continuum the interaction of such processes exposes either
their bulks which appear as particles or its thinness which appear as
waves because we still do not have better concepts than these classical
ones. This is the relative ("modal") separation of the modes or processes
of nature but not a real or absolute separation. It is similar to the "undi-
vided universe" of Bohm and Hiley (e.g., [4, p. 352, Ch. 15.12]) but not
to Bohr's "wholeness" which is restricted to the phenomenal experience
alone (e.g., [8, II, pp. 71-2], [8, III, pp. 2, 4]; cf., [28, p. 91]). Therefore, in
natural processes there is a sort of "creation" of new, but only relatively
new, realities from more basic structures "potentially" contained in the
interaction of more comprehensive physical systems (cf., [4, p. 18ff., 3.2,
p. 218]); this creation is actually a relative separation of the elements
from the system containing them potentially, which occurs according the
laws of nature governing all natural processes. The specific structures of
the individual things (containing more basic and embedded in less ba-
sic ones) are Spinoza's "essences" of things (systems) (cf., Ethics IID2,
Al" D-L 7DS). It seems that this Spinozist "picture" of physical reality
calls for new and non-classical concepts of physical processes and it might
be that with this conceptual revolution the philosophical interpretation
would be realistic but without requiring either Bohr's phenomenological
detachment from unobserved reality or Einstein's absolutely separated
and undisturbed external physical reality (cf., [20, Ch. 7]). However, in
its phenomenological interpretation quantum theory does not venture
to explain these changes and creations of new processes since it is for-
bidden to go beyond the descriptions of the results of measurements to
represent the real systems involved in the experimentation (cf., [4, 2.2]).
The cognitive perception-observation of the operation of the physical
process of measurement adds nothing physically to it but only represent
its results (cf., [43, p. 160]; [4, Chs. 6, 8]). Thus, the cognitive mind does
not change ("collapse") the wave states of the measured system through
the perception of the measuring instrument. This is the case since the
perceiver-observer represents perceptually the measuring instrument af-
ter the measurement process has terminated and she theoretically repre-
sents the measured system (object) only mediately, when the perceptual
process completed (cf., [4, 6.1]).
WHICH SIDE SPINOZA WOULD HAVE TAKEN? 261
5. REALITY AND REPRESENTATIONAL REALISM
We need three components to explain human representation of reality:

A. The relative dependency between the cognitive mind and the reality
represented which is necessary for the natural process of represen-
tation.
B. The relative independence of reality from the cognitive mind in order
to explain that reality can force itself on human cognition.
C. The relative independence of the mind from the external physical
reality to enable us to explain its errors (being only relatively true)
in representation. To explain the relation between minds and the
external reality, we should have recourse to what I call "representa-
tional realism" (cf., [29], [30], [31], [32, #3, 7]).
The following explanation of the representational situation and the indi-
rect representation of reality is based on Spinoza's and Peirce's concep-
tions of nature, its physical and psychical components, and their laws.
(2) The Representational Situation of Mind and Reality:
NATURE
Person Perceiving
Mental Causality in Mind
Real Object Cognitive Mind
,Sign------- ~E-----~0.----~L
I'
l ''
'' Human
,
Indirect Representation~ Union
' Mind
of the Physical Object ' "'"' .. "'"'"'"' "'"'"'""
"'T"'• -------------- ......................... -- ........ ---- of
AI'" ''
' Mind-Body
''
'
Physical Object----------..,.. {OR+ bp} f--4--------Human Body
Physical Causal Interaction of Bodies
In the representational situation the actual connection between the cau-

sal process of perceptual sign interpretation, Sign- OI --+ E --+D--+L
(E, D, L are respectively, Peirce's Emotional, Dynamical and Logical
262 DAN NESHER
interpretants), and the causal interaction between the Human Body

and the Physical (Real) Object, is the ontological union U of the
perceiver's mind and body. This whole structure (given in bold signs
and lines) makes possible the indirect Representation R~ of the external
Object by the representing Signs. The interaction between the two bodies
modifies them, and the modification in the human body {oR+ bp} is the
combined effect of both the Real (Physical) Object OR, and the Person
(human) Body Bp)· The entire situation of the indirect representation
mediated by the mind-body union can be schematized as follows:
(3) The Indirect Representation: R~ (Sign, Object):
R 1 ((Sign----+ E ---+ D---+ L) + [U(Sign, {OR+ bp} )] +
Physical Object)
(comp. [32, pp. 172-4], [33], [12]).
Following is the explanation of the relation between the mind and
the real object represented in respect of the three required components
above:
1. The two interacting objects, the physical object and the human
body, are relatively (modally) independent of each other since they
are not absolute parts of one another. (These are the relations
among all modally separated sub-systems of nature, the "modes" in
Spinoza's philosophy; cf. EthicsiiP13S-Postulates.)
2. For perception there must be a physical interaction of the relatively
independent human body (with its brain [and eyes]) and physical
object.
3. The modification {OR+ bp} occurs in the human body as well as
in the Physical Object, but differently (because they are different
objects), and the modification in the human body is united with the
mental modification in the mind. Yet, the human mind's perceiving
process is connected with the real (physical) object only through
this union with "its own" body and the causal interaction of these
two objects.
4. Therefore, the mental modification, the percept and its interpreta-
tion in signs, are dependent on the reality of the interacting physical
objects because the latter are relatively independent of the mental
processes.
5. Since the representation of the external physical object is based on
this combined modification {OR+ bp}, it will forever be only an in-
direct representation, and therefore, it will always be an incomplete
representation (comp. [4, Ch. 15.13]).
WHICH S IDE S PIN 0 Z A W 0 U L D HAVE TAKEN ? 263
6. Human cognitive representation of physical objects is based on the

modification of the human body {oR+ bp} by the external object,
and on the correlate Mental modification: the percept. Yet the later
cognitive interpretation of the percept proceeds in abstraction and
generalization, where this process is relatively autonomous of the
percept and evolves according to the background knowledge and
the laws of the mind to represent the external object.
7. If reality were dependent on human cognition, then human minds
would enjoy "their inward freedom which determines their experi-
ential cognition" [36, 2.138] and could not be in conflict with their
cognitive modes.
8. Since reality is relatively independent of the cognitive mind, and the
mind is relatively independent of reality, the relative falsity and the
relative truth of human cognitive representations is explainable.
From the representational realist point of view we understand that the
objects represented by our minds are not created by them while being
represented, and in this sense there is an independent reality. Of course,
if we take our representations as the only reality, then it is dependent on
our cognition, yet we must emphasize the distinction between our repre-
sentational knowledge being dependent on our minds and the represented
reality which is independent of from the representational operation.
AFFILIATION
Dan Nesher
University of Haifa
Israel
REFERENCES
[1] Bell, J., "John Bell: A Discussion of the Mysteries of Quantum

Physics". In: Davies, P.C.W. & Brown, J.R. (eds.), The Ghost in
the Atom, Cambridge University Press, 1986.
[2] Beller, M., "The Birth of Bohr's Complementarity: the Context and
the Dialogues", Studies in History and Philosophy of Science, Vol. 23,
No. 1, 1992, pp. 147-180.
[3] Beller, M., "Einstein and Bohr's Rhetoric of Complementarity".
In: Beller, M. Cohen, R.S. & Renn, J. (eds.), Einstein in Context,
Cambridge University Press, 1993.
[4] Bohm, D. & Hiley, B. J ., The Undivided Universe: An Ontological
Interpretation of Quantum Theory, Routledge, 1993.
264 DAN NESHER
[5] Bohr, N., "Can Quantum Mechanical Description of Physical Real-

ity be Considered Complete?" Physical Review, 48, 1935, pp. 696-702.
Reprinted in: Weeler, J. A. & Zurek, W.H. (eds.), Quantum Theory
and Measurement, Princeton University Press, 1983, pp. 145-151.
[6] Bohr, N., "The Causality Problem in Atomic Physics," in: New
Theories in Physics, Paris, International Institute of Intellectual Collab-
oration, 1939, pp. 11-30.
[7] Bohr, N., "Discussing with Einstein on Epistemological Problems in
Atomic Physics". In: Schilpp, P. A. (ed.), Albert Einstein: Philosopher-
Scientist, The Library of Living Philosophers, Inc., Evanston, Illinois,
1949.
[8] Bohr, N., The Philosophical Writings of Niels Bohr: Vols. I-III
{1934-1958-1963), OxBow Press, 1987.
[9] Davies P.C.W., & Brown, J.R., The Ghost in the Atom, Cam-
bridge University Press, 1986.
[10] Devitt, M., Realism and Truth, Princeton University Press, 1984.
[11] Dummett, M., "Realism", Synthese, 52, 1982, pp. 55-112.
[12] Einstein, A., "Maxwell's Influence on the Development of the Con-
ception of Physical Reality". In: James Clerk Maxwell: A Commemora-
tion Volume, Cambridge University Press, 1931, pp. 66-73.
[13] Einstein, A., "Physics and Reality". Journal of the Franklin In-
stitute, 221, 1936, pp. 349-82. Reprinted in: Einstein, Out of My Later
Life, 1956, A Citadel Press Book, 1991, pp. 59-97.
[14] Einstein, A., "Autobiographical Notes." In: Schilpp, P. A. (ed.),
Albert Einstein: Philosopher-Scientist, The Library of Living Philoso-
phers, Inc., Evanston, Illinois, 1949.
[15] Einstein, A., "Letters to Max Born and Max Born-s Comments
(1950)". In: French, A. P. (ed.), Einstein: A Centenary Volume, Har-
vard University Press, 1979, pp. 275-7.
[16] Einstein, A., Podolsky, B. & Rosen, N., "Can Quantum-
Mechanical Description of Physical Reality Be Considered Complete?"
Physical Review, 47, 1935, pp. 777-80. Reprinted in: Weeler, J.A. &
Zurek, W.H. (eds.), Quantum Theory and Measurement, Princeton
University Press, 1983, pp. 138-141.
[17] Espagnat, B. d', In Search of Reality, Springer-Verlag, 1983.
[18] Faye, J., Niels Bohr: His Heritage and Legacy-An Anti-Realist
View of Quantum Mechanics, Kluwer Academic Publishers, 1991.
WHICH SIDE SPIN 0 Z A W 0 U L D HAVE TAKEN? 265
[19] Fine, A., "Einstein's Interpretation of the Quantum Theory." In:

Beller, M., Cohen, R.S. & Renn, J. (eds.), Einstein in Context,
Cambridge University Press, 1993, pp. 257-273.
[20] Folse, H., The Philosophy of Niels Bohr, North Holland, 1985.
[21] Heisenberg, W., "The Physical Content of Quantum Kinematics
and Mechanics (1927)". In: Wheeler, J.A. & Zurek, W.H. (eds.),
Quantum Theory and Measurement, Princeton University Press, 1983,
pp. 62-86.
[22] Heisenberg, W., Physics and Philosophy: The Revolution in Mod-
ern Science, Harper & Brothers Publishers, 1958.
[23] Honner, J., The Description of Nature: Niels Bohr and the Phi-
losophy of Quantum Physics, Clarendon Press, 1987.
[24] Kant, I., Critique of Pure Reason {1781-1787), Tr. by N.K. Smith,
Macmillan & Co. Ltd., 1929.
[25] Kaiser, D., "More Roots of Complementarity: Kantian Aspects
and Influences", Studies in History and Philosophy of Science, Vol. 23,
No. 1, 1992, pp. 213-239.
[26] Lockwood, M., Mind, Brain 8 Quantum: The Compound 'I',
Blackwell, 1990.
[27] Margenau, H., "Einstein's Conception of Reality". In: Schilpp,
P.A. (ed.), Albert Einstein: Philosopher-Scientist, The Library of Living
Philosophers, Inc., Evanston, Illinois, 1949.
[28] Murdoch, D., Niels Bohr's Philosophy of Physics, Cambridge Uni-
versity Press, 1987.
[29] Nesher, D., "A New Conception of Language: A Reconstruction of
Peirce's Philosophy of Language as a Theory of Cognition", in: The Tasks
of Contemporary Philosophy, Holder-PichlerTempsky, 1986, pp. 198-205.
[30] Nesher, D., "Remarks on the Consistency and Evolutionary Char-
acter of Spinoza's Philosophy". Iyyun, 36, April1987, pp. 83-123. (He-
brew).
[31] Nesher, D., "Truth and Meaning in Semiotic Process: A Peircean
Approach". Delivered at the Charles S. Peirce Sesquisentennial Interna-
tional Congress, Harvard University, September, 1989. (Manuscript)
[32] Nesher, D., "Spinoza's Theory of Truth". In: Hunter, G. (ed.),
Spinoza: The Endurin Questions, University of Toronto Press, 1994,
pp. 140-177.
[33] Nesher, D., "Pragmaticist Realism:The Third Philosophical Per-
spective as the 'intermediate point' Between 'Metaphysical Realism' and
'Internal Realism' ", 1997. (Manuscript)
266 DAN NESHER
[34] Nesher, D., "Peircean Realism: Truth as Meaning of Cognitive

Signs Representing External Reality", Transactions of the Charles S.
Peirce Society, Vol. XXXIII, 1, 1995.
[35] Pais, A., Niels Bohr's Times in Physics, Philosophy, and Polity,
Clarendon Press, Oxford, 1991.
[36] Peirce, C.S., Collected Papers Vols. 1-8, Harvard University Press,
1931-1958. (References to Volume, e.g., 5, and paragraphs, e.g., 119, i.e.,
5.119).
[37] Peirce, C.S., Writings of Charles S. Peirce, A Chronological Edi-
tion. Vol. 1-5, Indiana University Press, 1982.
[38] Putnam, H., Reason Truth and history, Cambridge University
Press, 1981.
[39] Putnam, H., Representation and Reality, MIT Press, 1988.
[40] Putnam, H., Realism With a Human Face (Edited and Introduced
by James Conant), Harvard University Press, 1990.
[41] Redhead, M., Incompleteness Nonlocality and Realism: Prolegom-
ena to the Philosophy of Quantum Mechanics, Clarendon Press, Oxford,
1987.
[42] Rorty, R., Consequences of Pragmatism, University of Minnesota
Press, 1982.
[43] Schrodinger, E., "The present situation in Quantum Mechanics."
Naturwissenschajten, 23, 1935. Trans. J.D. Trimmer, in: Weeler, J.A.
& Zurek, W.H. (eds.), Quantum Theory and Measurement, Princeton
University Press, 1983, pp. 152-167.
[44] Spinoza, B., The Collected Works of Spinoza, Vol. I. Edited and
Translated by E. Curley. Princeton University Press, 1985.
[45] Stapp, H.P., Mind, Matter, and Quantum Mechanics, Springer-
Verlag, 1993.
[46] Wheeler, J., "John Wheeler: A Discussion of the Mysteries of
Quantum Physics." In: Davis, P.C.W. & Brown, J.R. (eds.), The
Ghost in the Atom, Cambridge University Press 1986.
[47] Weizsacker, C.F. von, The Unity of Nature, Farrar Strauss Gi-
roux, 1980.
STEVEN WEINSTEIN
GENERAL RELATIVITY AND QUANTUM

THEORY-ONTOLOGICAL INVESTIGATIONS
1. INTRODUCTION
It is common to note that the two great theoretical frameworks of

twentieth-century physics, quantum theory and general relativity, are
prima facie incompatible. Within the physics community, the incompat-
ibility is largely discussed on a technical level. In this paper, I would
like to frame the problem more conceptually, focussing on the respective
ontologies of the two frameworks.
When I speak of ontology, I have in mind the idea that theories make
use of a formal (often mathematical) language for talking about the
world. The ontology of the theory then just consists of the objects and
properties to which the names and predicates of the language refer. Al-
though this ontology may vary according to different ways of regiment-
ing the language, there is frequently a common or standard formulation.
Thus the ontology of astronomy is roughly understood to be planets,
stars and galaxies (the objects), which have their various masses and
shapes ("internal" properties) and positions and velocities ("external"
properties).
For the most part, fundamental physical theories refer only implicitly
to the objects themselves, for it is the properties that are mathemati-
cally quantified. The language of Newton's mechanics and his theory of
gravitation enables one to refer to any object having a mass, a position,
and a velocity. Thus Newton's theory is as much about planets as it
is about apples. Maxwell's theory of electromagnetism refers to objects
that have the property not only of mass but charge as well. In that re-
spect it is not significantly different from Newton's theory. But it also
refers to something called the electromagnetic field. The values of the
field are properties, but they are properties not of Newtonian objects
but of spacetime itself.
Historically, Maxwell's theory may be the first theory in which space-
time plays the role of an "object", a bearer of properties. Previously,
space and time had simply been thought of as the background in which
objects lived. (Of course, one can anachronistically reconstruct, e.g.,
Newtonian theory from the modern perspective.) As we shall see, general
relativity expands on this development, while quantum theory retains a
267
268 STEVEN WEINSTEIN
largely passive view of spacetime while suggesting entirely new ways to

think about properties.
2. GENERAL RELATIVITY
General relativity is among other things a theory of gravitation, which

is to say it is a theory about the interactions of massive bodies. The
sources of gravitational attraction include not only the mass and energy
inherent in "bodies" but the energy inherent in various fields, as well.
Classical (non-quantum) general relativity is almost universally for-
mulated as a geometric theory, in which gravitation is manifested as
curvature of spacetime .I This geometric approach builds on Minkowski's
formulation of special relativity as a theory of "fiat" 4-dimensional space-
time, where spacetime is represented by a 4-dimensional differential man-
ifold with Minkowski (fiat) metric. General relativity extends this, using
the tools of pseudo-Riemannian geometry to talk about more general
spacetimes, ones with a Lorentz metric gab (of which the Minkowski
metric is a special case). The metric carries information about the dis-
tance between points on the manifold, and thus encodes the geometric
structure of the manifold. The relevant curvature measures are given by
the Riemann tensor Rabcd' and the related Ricci tensor Rab = Rcacb and
scalar curvature R =gab Rab· These curvature tensors are all made up of
derivatives of the metric; they describe curvature by describing how the
metric changes when one moves in various directions.
The curvature of spacetime is constrained by the stress-energy-momen-
tum distribution Tab, which in addition to encoding the mass-energy at
each spacetime point also encodes the flow of mass-energy. Einstein's
equation quantifies the way in which Tab constrains the metric gab:
(1)
Thus we have a sort of relativistic generalization of Newton's theory

of gravity, in which the spatio-temporal distribution of mass and energy
determines the gravitational field, here represented by curved spacetime.
The language of general relativity is the language of tensors on a man-
ifold, the language of differential geometry. Insofar as one regards the
points of the manifold as representing spacetime points, one must regard
the stress tensor Tab, the metric tensor gab, and the latter's associated
curvature tensors, as assignments of physical quantities (energy, mo-
mentum, spacetime curvature) to the various points. This language of
1
See (1], though, for an algebraic approach to general relativity.
GENERAL RELATIVITY AND QUANTUM THEORY 269
properties at points is suggestive of a field ontology, but the stress ten-

sor need not be constructed purely out of fields. Bulk matter will do as
well-one simply assigns a matter density and velocity to each point in
spacetime occupied by the massive object.
The implicit assumption involved in construing tensors as fields in
spacetime is that the points of the differential manifold are points of
spacetime. But stripped of the metric, the bare manifold has very little of
the structure one associates with spacetime. Among other things, there
is no causal structure, no notion of whether two points are space-like or
time-like related. This suggests that it is only in the presence of a metric
tensor that it makes sense to refer to points as points in spacetime. In
that spirit, we would say that the statement that a tensor field such as
Tab represents a distribution of properties in spacetime makes sense only
in the presence of a metric tensor.
Now, the idea that the stress tensor Tab represents a spacetime prop-
erty distribution only in the presence of a metric is not really problem-
atic, since the metric is already built into the stress tensor. 2 However,
what are we to say about the metric itself-does it represent a space-
time distribution of properties? This is a tricky question. On the one
hand, once we have a metric, we have spacetime points, in the sense
that the metric defines spatial and temporal distances between points.
These points have definite physical properties attributed to them-the
derivatives of the metric describe the gravitational field. On the other
hand, from a strictly mathematical perspective, the metric is an attribu-
tion of properties to points on the manifold, and the manifold does not
in itself represent spacetime. From the former perspective, the metric
is a field like any other tensor field, and one might think it should be
treated as such. From the latter perspective, the metric plays a rather
special role, and it is not clear that it should be regarded as a conven-
tional field, in the sense of a spacetime property distribution. Support
for the former view comes mainly from practical concerns-after all, one
can measure the gravitational field just as one can measure the electro-
magnetic field. 3 Support for the latter view comes primarily from formal
concerns, in particular the role the metric plays in the definition of other
fields and their stress tensors. We shall not settle this issue here! 4
2
For example, the stress tensor for a massless scalar field is given by Tab
\7 ac/JY'bc/J- ~gab(\7 cc/JY'c¢).
3
It must be said, however, that the strength of the gravitational field at a point
is not well-defined, unlike the strength of the electromagnetic field. This despite the
fact that the Riemann tensor Rabcd is well-defined at a point.
4
The interested reader will find a lucid discussion of this issue in [2], which contains
3. QUANTUM THEORY
There are various quantum theories. There is quantum mechanics, by

which I mean the quantum theory of non-relativistic particles. Then
there are the quantum field theories, which are theories of systems with
infinitely many "degrees of freedom": these include theories of the fun-
damental forces such as electroweak theory and QCD, as well as theories
of various many-body systems, such as superconducting matter, super-
fluids, plasmas, electrical conductors, etc.
Let us consider quantum mechanics first. In quantum mechanics, one
works with fixed numbers of objects, such as elementary particles, atoms,
or protons. A collection of these objects constitutes a "system." The
"state" of a system is technically a map from the configuration space
of the system to the complex numbers. For instance, two unconstrained
massive, spinless particles in three-dimensions have a configuration space
of JR 6 , and the map is given by a (normalized) "wave-function" that
assigns a complex-valued "amplitude" to each point in the configuration
space. The square of the amplitude, integrated over a region of the space,
gives the probability of finding the system in one of the configurations
in that region. For a single, spinless particle, the configuration space
is typically three-dimensional physical space, and the integral of the
squared amplitude evaluated over a given region of this space yields the
probability of finding the particle in that region.
Note that the complex nature of the amplitude enables the wave-
function to encode not only position information, but momentum in-
formation as well. The momentum information is extractable in var-
ious ways, among them taking the Fourier transform of the original
wave-function, and evaluating the squared amplitudes of the resulting
"momentum-space" wave-function. What is important for our purposes
is simply that a maximally specific description of the state of any quan-
tum system encodes probabilistic distributions of position and momen-
tum (among other properties), and that these distributions resist being
"squeezed" past a certain point. What this means is that, after a certain
point, a system may have its position specified more closely only at the
cost of specifying its momentum less closely. Thus the precise properties
of any quantum system are inherently indefinite.
Despite the uncertainty in the properties of atoms, electrons, or what
have you, one can still speak of localized, particle-like objects which
possess the properties, even though they only "possess" them proba-
bilistically. Thus the ontological situation in quantum mechanics is one
in which there are a definite number of objects at any given time, and
further references to the literature.
G ENE R A L R E LA T IV IT Y AN D Q U ANT U M THE 0 R Y 271
in which these objects persist over time. So far, so classical. What is un-
classical, again, is that the objects do not have definite properties at a
given time (of position, momentum, spin, etc.) and do not have definite
trajectories over time. Furthermore, if the systems have interacted, these
properties will be "entangled", so that, e.g., the position of one object
will in general be correlated with the position of another.
The ontological situation in quantum field theory is somewhat dif-
ferent. Indeed, the formal apparatus of the theory was developed to
accommodate physical situations in which the number of particles can
change. The general treatment is thus one of systems with "infinite de-
grees of freedom." The idea of infinite degrees of freedom is suggestive of
classical fields, and indeed there are many cases in which quantum fields
are analogous to classical fields, in that at each point in spacetime, one
has operators (technically, operator-valued distributions) that represent
various field properties (such as the electric field in the x direction), and
that allow one to extract the probability of observing various values for
those properties at the various points in space. The possibility of de-
scribing such quantum systems either in terms of an indefinite number
of particles, or in terms of probabilistic values of a field, is the essence
of wave-particle duality. 5
So we have a situation in which there is a sort of wave-particle duality.
But because the number of particles is indefinite, it would seem odd to
say that the ontology is particles. One of the founders of quantum field
theory, P.A.M. Dirac, shared this view:
If one can create particles, then the question of which are the fun-
damental constituents of matter ceases to have a definite meaning.
Previously, one could say that one only had to analyze a piece of
matter as far as possible, and get at the ultimate constituents in
that way. But if one can create particles by atomic interactions, then
one cannot give a definite definition for an elementary particle. [3,
p. 19]
In an important sense, a sort of field ontology appears to be forced on
us in relativistic quantum theory in a way that it is not in classical,
relativistic particle theory, for not only the properties of the particles
but their very number is indeterminate.
One might bite the bullet at this point and argue that, just as prop-
5
Note that the "wave-particle duality" often attributed to the non-relativistic
quantum theory of particles is a bit of a misnomer, since the wave-function for a system
with more than one particle cannot be thought of as a wave in three-dimensional
space. (In an n particle system the wave function typically exists in a space of 3n
dimensions.)
erties of objects can be indefinite, so can the objects themselves, and

therefore quantum field theory can be construed as being a theory about
particles. However, one runs into real trouble when considering quantum
fields in curved spacetime, which is to say in the presence of gravity. In
such situations, the vacuum ("no particle") state is observer-dependent,
and since the particle content of the theory turns out to depend on the
vacuum state, one has no preferred notion of particles. One observer's
vacuum turns out to be another observer's "bath" of particles. (See [4]
for further discussion.)
It would appear, then, that quantum field theory might be better said
to simply be about fields. The field representation of a massless scalar
field, without spin, involves an algebra of operators J;(x, t) and canon-
ically conjugate operators 7r(i, t). As the label indicates, the position
and time serve to parametrize these operators. As noted above, they
describe a property distribution about points in space at a given time,
and consequently invite the interpretation that they represent a quan-
tum version of a field. On this interpretation, the principle difference
between a quantum and a classical field is that the property distribution
is indeterminate and probabilistic.
There are two main problems with this interpretation of quantum
fields as probabilistic versions of classical fields. The first is that there is
no particular reason to privilege the parametrization in terms of x and
t. One can equally well Fourier-transform, and represent the quantum
field by an algebra of operators parametrized by the momentum p and
the time. This suggests that the field interpretation is at best optional.
The second problem is that any field with internal degrees of freedom
will have an operator algebra parametrized by those degrees of freedom
as well. So for instance a field whose degrees of freedom include position,
the z component of spin and the third component of isospin would have
an operator algebra parametrized as J;(x, O"z, h, t). 6 Now, this could be
interpreted as a spatia-temporal distribution of properties a la the clas-
sical field concept, though note that for every point in space at a given
time one will have four operators J; and four operators 7r (since spin and
isospin have two possible values each). But what is worse for the field
point of view is that it is perfectly plausible to construct a quantum field
theory without any reference to space at all, simply because the spatial
degrees of freedom play no privileged role in the formalism of the theory.
There are undoubtedly those who will argue that the idea of a quantum
6
This is a mild abuse of language. Technically, the degrees of freedom of the quan-
tum field are the various (classical) functions ¢(x, uz, h, t), not the properties such
as position, spin, and isospin, that parametrize them.
GENERAL R E LA T IV IT Y AND QUANTUM THE 0 R Y 273
field theory which is not in any sense a spacetime theory is absurd, the
argument being that because observations take place at points in space
at particular times, any theory must have spatial degrees of freedom in
order for it to yield predictions for observations. Though this argument
has some appeal, it is highly problematic, for it rather begs the question
of how to represent observations in quantum theory. In general, I would
suggest that we be wary of it, for one of the fundamental problems in
physics is the reconciliation of quantum theory and gravity, and it may
be just such prejudices regarding the nature of observation that obstruct
the way toward a reconciliation of the two theoretical frameworks.
4. GENERAL R E LA T IV IT Y AND QUANTUM THE 0 R Y
Let me briefly rehearse the conclusions regarding the ontologies associ-

ated with general relativity and with quantum theory. General relativity,
it will be recalled, accommodates both particles and fields as the material
constituents, though it describes both in a field-like way, ascribing prop-
erties to points of the spacetime manifold. Its classical nature consists in
the fact that all of these properties of the particles or fields are definite
properties-at any point in space at any time, the stress-energy tensor
takes on definite values. The primary ontological point of contention
in general relativity is whether one ought to characterize the spacetime
metric itself as some sort of field. If it is a field, then is it the same sort
of field as other fields? 7
Quantum theories, on the other hand, are characterized in part by the
indefiniteness of the properties they describe. In quantum mechanics,
the indefinite properties may be ascribed to underlying objects called
"particles," but a full relativistic treatment involves a progression to the
notion of quantum fields. Here a particle ontology is essentially unten-
able. Moreover, the internal (non-spatial) degrees of freedom which go
along with quantum field theories suggest that it may furthermore be in-
appropriate to characterize the theories as theories of fields, in the sense
of spacetime property distributions.
To begin to grasp the tension between general relativity and quantum
theory, consider again Einstein's equation Gab = 81rTab· The stress ten-
sor, on the right, invariably requires a specification of the properties of
matter such that the matter has well-defined values for various quanti-
ties at each point in spacetime, and this is simply not obtainable from
any quantum-theoretic treatment of matter. The four-momentum of a
particle, for example, is not well-defined, both because it is impossible
7
One argument against treating the gravitational field as "just another field" is
that it is difficult to provide a notion of local gravitational energy.
to completely localize a quantum-mechanical particle, and furthermore

because one must ultimately trade-off increased definiteness in position
(at a time) with decreased definiteness in 3-momentum (or vice-versa).
How might one go about incorporating quantized matter into general
relativity? One might try to "quantize" the stress tensor, the idea being
that just as the classical tensor Tab is essentially a function of the classical
field variables¢ and Jr, its replacement should be an operator Tab which
is a function of 1J and ir. This makes a fair amount of intuitive sense,
since if one thinks that the physical matter under consideration has a
quantum uncertainty associated with it, one would expect that its stress-
energy-momentum properties would reflect this.
Quantizing the stress tensor, however, presents numerous difficulties.
One, it immediately raises the heretofore postponed question of whether
or not to quantize the metric, for the stress tensor is a function not
only of the matter variables ¢ (reverting for simplicity to the covariant
formalism and thus effectively incorporating 1r), but of the metric 9ab as
well. I.e., do we want an equation of the form
(2)
or
Gab = 8Jr \ Tab(J}, 9ab)) 1/J (3)
or perhaps something else? 8

The first equation (2) couples quantized matter to quantized space-
time. This implies a modified ontology for general relativity, whereby
matter has indefinite properties, and whereby spacetime has indefinite
properties as well. Note that the notion of a quantized spacetime, a
quantized gravitational field, is suggestive of the more general notion of
a quantum field theory as a specification of certain families of proper-
ties which are not necessarily spatia-temporal. In treating the metric as
something to be quantized, as "just another field", we end up treating it
as the most general sort of quantum field, as a field which is not essen-
tially a spatia-temporal distribution of properties, since the quantization
of the metric deprives one of a well-defined spacetime background.
Despite the compelling ontological unity suggested by (2), in which
"everything" is a quantum field, the problems are legion. It perhaps
suffices to note that no one has yet succeeded in quantizing vacuum
8
Equation (2) is simply a heuristic. It is most closely analogous to the constraint
equations of canonical quantum gravity. See [5] for a clear introduction to this research
program.
GENERAL RELATIVITY AND QUANTUM THEORY 275
gravity, i.e., an equation of the form GabW = 0, much less couple it to

a quantized stress tensor, of which more below. 9 Part of the difficulty
involved in doing so is to select an appropriate representation of the
relevant operators. If one did have such a representation, then one could
try to give (2) a physical interpretation. On the face of it, if one were
told that the "system" was in some eigenstate of the stress tensor, then
this equation would appear to allow one to determine the probability of
observing various metrics given a certain stress-energy distribution. But
note that an eigenstate of the stress tensor cannot be a state in which the
stress-energy-momentum of the system has a definite classical value in
the usual sense because it is quite unclear what it even means to say that
the stress-energy-momentum is "definite" if the background metric is
quantized. Definite at every point in space at some time? What is meant
by "space" in the context of a quantized metric? In short, an equation
such as (2) is riddled with both technical and conceptual difficulties. 10
If something like it actually turns out to be mathematically viable, it
will no doubt require a major conceptual change to make physical sense
of it. There is every reason to think that a fully quantum theory of
gravity would be incompatible with any ontological structure, be it that
of general relativity or non-relativistic quantum theory, which makes
essential reference to a background spacetime. (See [11] for discussion.)
What about equation (3)? This is known as the semi-classical Einstein
equation; it couples quantized matter to classical spacetime. It is occa-
sionally used when doing quantum field theory in curved spacetime, as a
way of getting an approximate handle on the way in which the evolution
of a quantum field affects the structure of the spacetime in which the
field is embedded. However, its use in even this limited role is heavily
circumscribed [4], and virtually no one takes it seriously anymore as
a candidate for a fundamental theory [12]. One can see one aspect of
the difficulty as follows. Suppose we have a massive object which is in
a superposition of states, such that there is equal probability of find-
ing it on either side of some room, with vanishing probability anywhere
in the middle. Then according to (3), the gravitational field associated
with this object would be the field one would classically obtain were the
mass in the center of the room, since that is the expectation value for
the position of the object (and one expects in this situation that the
9
See [6] and [7] for excellent reviews of quantum gravity and discussions of the
:;orts of problem one encounters. See [8] and references therein for news of recent
progress in this area.
10
My quick dismissal notwithstanding, Rovelli (see [9], [10] ) has recently proposed
a way of dealing with some of the issues raised here.
expectation value of the stress tensor will be a simple function of the ex-
pectation value for the position). Now perform a position measurement.
One will find the object on one side of the room or the other. So at the
moment of measurement, when the object becomes localized, we must
say that the metric changes discontinuously. Besides the fact that this
seems intuitively unphysical, such a theory represents a major departure
from the framework of general relativity!
Now, this quick example by no means constitutes a refutation of the

semi-classical approach. For one thing, it could be that introducing mea-
surement and reduction of the state vector is not playing the game
fairly, that one must deal only with closed systems when considering the
coupling of quantized matter to classical spacetime. However, there are
equally serious difficulties for a closed system treatment. The conceptual
problems are less obvious simply because the quantum theory of closed
systems is problematic. But consider what happens if one adopts the
many-worlds/decoherent histories approach to closed systems (see [13]
for a review), in which the system continuously "branches". If one were
to invoke the semi-classical equation (3) in this context, one would have
a situation in which an observer would in theory be able to detect the
branching, because the (classical) state of the gravitational field would
be a function of the (quantum) matter distribution in all the branches,
not just her own. The gravitational field would not, in general, have any-
thing to do with the observed (within the branch) matter distribution.
This is of course in contradiction with what we observe, and so it seems
as if the closed system approach to semi-classical gravity fares no better
than the open system approach. In short, semi-classical gravity seems to
be a rather implausible halfway-house.
In the face of the apparent difficulties with putting together classical

general relativity and quantum theory in the ways sketched above, it
is notable that no one has yet come up with a formal argument that
the quantization of the sources of the gravitational field implies that the
field itself must be quantized, though such an argument does exist for the
quantization of the electromagnetic field [14], [15]. This is related to the
fact that in electromagnetism, the field strength couples to the charge of
the source, whereas in gravity, it couples to the mass. If one considers the
limit in which the mass goes to zero, the uncertainty principle predicts
an enormous indeterminacy in the position or momentum of the source
particle, and a correspondingly large indeterminacy in the value of the
field. However, for gravity, the field vanishes as the mass goes to zero
[16].
5. CON CL UDIN G REMARKS
The prima facie incompatibility of the general relativistic and quantum

theoretic ontologies does not necessarily indicate an insuperable incom-
patibility. The discussion in this paper is offered more in the spirit of
indicating that something has to give. In general relativity, we see some
room for flexibility in the question of whether or not we take the space-
time metric to be a field in its own right, and in quantum theory we see
the possibility, suggested by quantum theories with internal degrees of
freedom, of conceiving of a quantum field as something quite unlike a
classical, spatiotemporal field.
Of course, despite this flexibility, we still have to face the apparent
incompatibility between the definite physical properties which are the
province of general relativity, and the indefinite, probabilistic properties
of quantum theory. It is notable, however, that general relativity is a
typical classical theory in that the properties (of particles, of spacetime
points, etc.) are objectively possessed, whereas in quantum theory, prop-
erties are indeterminate and become determinate (but not necessarily
objective (see [17])) only on observation. An investigation of the notions
of observation and objectivity in general relativity and quantum theory
would be a useful complement to the ontological issues raised here.
ACKNOWLEDGMENTS
This paper is the write-up of a talk delivered at the IQSA symposium,

Einstein meets Magritte conference, Brussels, May, 1995. Thanks to the
audience in Brussels, and audiences at Cambridge University and Impe-
rial College, for helpful questions and comments. The bulk of the paper
was written during my stay at Imperial College, London, and I would
like to express my gratitude to the members of the Theoretical Physics
Group there for their hospitality. I would particularly like to thank Chris
Isham for many enjoyable and interesting discussions on conceptual and
technical problems in quantum theory and general relativity.
AFFILIATION
Steven Weinstein
Northwestern University
Evanston, Illinois
USA
REFERENCES
[1] Geroch, R., "Einstein algebras", In: Communications in Mathemat-

ical Physics, volume 25, 1972, pp. 271-275.
[2] Stachel, J., "The meaning of general covariance: The hole story",
In: Earman, J., Janis, A., Massey, G., and Rescher, N. (eds.),
Philosophical Problems of the Internal and External Worlds, Pittsburgh-
Konstanz Series in the Philosophy and History of Science, Pittsburgh,
University of Pittsburgh, 1994, pp. 129-160.
[3] Dirac, P.A.M., Directions in Physics, New York, Wiley, 1978.
[4] Wald, R.M., Quantum Field Theory in Curved Spacetime and Black
Hole Thermodynamics, Chicago, University of Chicago Press, 1994.
[5] Ashtekar, A., Lectures on Non-Perturbative Canonical Gravity, Sin-
gapore, World Scientific, 1991.
[6] Isham, C.J., "Canonical quantum gravity and the problem of time",
In: Integrable Systems, Quantum Groups, and Quantum Field Theories,
London, Kluwer Academic, 1993, pp. 157-288.
[7] Kuchar, K., "Time and interpretations of quantum gravity", In:
Proceedings of the 4th Canadian Conference on General Relativity and
Relativistic Astrophysics, Singapore, World Scientific, 1992, pp. 211-314.
[8] Ashtekar, A., "Polymer geometry at Planck scale and quantum
Einstein equations", In: General Relativity and Gravitation: Proceedings
of the 14th International Conference on General Relativity and Gravita-
tion, Singapore, World Scientific, 1997.
[9] Rovelli, C., "What is observable in classical and quantum gravity?",
In: Classical and Quantum Gravity, volume 8, 1991, pp. 297-316.
[10] Rovelli, C., "Quantum reference systems", In: Classical and Quan-
tum Gravity, volume 8, 1991, pp. 317-331.
[11] Isham, C.J., "Structural issues in quantum gravity", In: General
Relativity and Gravitation: Proceedings of the 14th International Confer-
ence on General Relativity and Gravitation, Singapore, World Scientific,
1997.
[12] Kibble, T.W.B., "Is a semi-classical theory of gravity viable?",
In: Isham, C.J., Penrose, R., and Sciama, D. (eds.), Quantum
Gravity 2: A Second Oxford Symposium, Oxford, Clarendon Press, 1981,
pp. 63-80.
[13] Halliwell, J.J., "A review of the decoherent histories approach to
quantum mechanics", In: Greenberger, D. (ed.), Fundamental Prob-
lems in Quantum Theory, New York, Annals of the New York Academy
of Sciences, 1995.
[14] Bohr, N. and Rosenfeld, L., "Zur Frage der Messbarkeit der
elektromagnetischen Feldgrossen", In: Kgl. Danek Vidensk. Selsk. Math.-
fys. Medd., volume 12, number 8, 1933.
[15] Bohr, N. and Rosenfeld, L., "On the question of the measurabil-
ity of electromagnetic field quantities (English translation)", In: Cohen,
R.S., and Stachel, J. (eds.), Selected Papers by Leon Rosenfeld, Dor-
drecht, Reidel, 1978, pp. 357-400.
[16] Rosenfeld, L., In: DeWitt, B.S. (ed.), Conference on the Role
of Gravitation in Physics, Proceedings, W.A.D.C. Technical Report 57-
216; ASTIA Document No. AD118180, Wright Air Development Center,
Wright-Patterson Air Force Base, Ohio, 1957.
[17] Rovelli, C., "On quantum mechanics", LANL physics archive, hep-
th/9403015.
EMMANUEL HEMMERLIN
FROM QUANTAL TO MATERIAL LEVEL
1. THE CONSTITUENT/CONSTITUTED RULE
The purpose of this communication is to present a possible new writing

of fundamental physics that has to do with the interpretation of Quan-
tum Mechanics. In fact, the first thing that has to be said is that, in that
matter, there is a rule that is systematically transgressed. It is an epis-
temological rule but at our level it is so trivial that, to my knowledge, it
has never been expressed as such by philosophers. That rule that I call
the 'constituent/constituted rule' just says that:
i/ the properties of the components are not those of the thing consti-
tuted (for instance we say that a brick is hard and a house comfortable,
but we do not say that a house is hard or that a brick is comfortable),
ii/ consequently, it would be highly confusing to give the same name
to the components as to the constituted thing.
Like I said, at our level it is trivial and in general we do not do such a
confusing thing. But in the particular case of physical entities we do. We
do so, because of a lack of words, and because have not been made aware
that the methods of physics would make us go so far in the nature of
being, so deep below the level of matter. Matter is the sensitive word: in
fact matter is made of atoms and atoms are themselves made of quantal
entities. Therefore matter is like a town, atoms are like houses, quantal
entities are bricks and timbers. So, we should not call them 'matter'.
They are not matter, they are components of components of matter
In fact we have been accustomed to doing the contrary because of the
old conception (cfr. Democritus) according to which the whole universe
is made of void and full, the full being named 'matter'. But it leads to
the planetary model of the atom, which, as we know, does not work.
So, we should consider it as obsolete and embrace a new conception in
agreement with the constituent/constituted rule according to which we
have to distinguish the material level, the atomic level, and finally the
quantallevel. And, as the properties of a brick are not the properties of
a town, properties of quantallevel are definitely not those of matter. To
understand them we have to forget all we know about matter: properties
of matter such as solidity, definite shape, deterministic behaviour are
constructed properties. They have nothing to do with the fundamental
281
282 EMMANUEL HEMMERLIN
level where, in accordance with the rule "not the same level, not the
same properties", there is a priori no solidity, no definite shape, no
deterministic behaviour.
What has to be understood by that is made clear by the quantal
formalism of the wave function. For instance, what has to be understood
by 'no definite shape' for the entity is, according to that formalism:
- no precise boundary (therefore, a fuzzy shape),
- deformation of that fuzzy shape when moving. (Nothing to do with
fuzzy sets.)
With that new concept of extended and deforming shape it is clear that
we have neither precise position, nor precise speed, and that is what is
behind the Heisenberg inequalities. (Note: if we listen to Heisenberg [1],
the origin of what was not yet laid down as a rule-and not applied
to the word matter-may be found in the work of Democritus himself.
As to J.M. Levy-Leblond [2], he maintains the word 'matter' at the
fundamental level, but suggests to see it as "without qualities".)
2. CONSEQUENCE AS TO THE WRITING OF

THE FORMALISM
From what we have said it follows that an electron, for instance, is not
matter. That has an immediate and fundamental consequence as, if an
electron is not matter, the mass is a quantity of matter. Therefore an
electron must not be characterized by a mass!
How to proceed in order not to have the mass at the fundamental level
is simple. Let us take a relativistic form of the Schrodinger equation, that
of Dirac for instance, and just consider its mathematical structure: it is
a one-constant structure. Therefore only one constant must be used,
and, as we know nothing about the nature of reality at quantal level,
this constant will be primarily unknown. So, let us name it ko, thus
obtaining:
(i'"f 11 811 - ko)w = o (1)

or, more explicitly:
8 ~
(i oxO + ia\l- f3ko)W = 0 (2)
Now, we write the conventional form, with mass mo of course, but also
with two other constants, n and c:
8 ~ 2
(in at+ inca\!- f3moc )w = 0 (3)
F R 0 M Q U ANTAL T 0 MATER I A L LEVEL 283
So, we have there three constants instead of one and, if one is enough,
why three? ... As we want to compare that conventional form with the
new expression, we multiply relation (2) by nc, thus obtaining:
8 ~
(in 0 t + incaV'- {3nck 0 )iJ! = o (4)
Then, by identification we obtain the way that ko is related to conven-
tional physics:
n(koc) = moc 2 (5)
Now, when deriving the Schrodinger equation, one has to set:
iJ! s '!1:. iJ!exp(i(koc)t) '!1:. iJ!exp(iwot) (6)

Thus, we see that koc is a pulsation (for non physicists let us say 'the
frequency of a vibration'), and therefore the final relationship is:
(7)
So, for non physicists, we have on the left side the expression of energy in
conventional Quantum Mechanics, and, on the right side, its expression
in conventional relativity. But, of course, we have done no physics at all.
We have just obtained the way that the relevant concept-the pulsation
wo-is related to the concept of mass, irrelevant at quantal level. And,
if we have obtained something 'already known', it is not really surpris-
ing (if not it would be wrong). But it was not (openly) postulated at
the beginning. In conclusion, ko is no longer an unknown constant: it is
the relativistic equivalent to the asymptotic pulsation of the wave func-
tion wo and therefore the pulsation (or frequency) replaces the mass at
quantal level. But if the mass is somehow 'something', the pulsation is a
characteristic of something, something that we do not know.
3. N OUMEN AL PHYSICS
Now, it is interesting to see what happens if-in accordance with rela-

tionship (5)-we replace in the conventional expressions mo by n":} We
cannot prove it in this short presentation-for that see references [3] and
[4]- but, just as in relationship (1), n disappears from every (testable)
equation, and c appears only under the form ct = x 0 , which means two
things:
i/ the mass is not the right characterization for a quantal entity (we
have seen why: an electron is not matter, it is a component of com-
n
ponent of matter), and therefore disappears when the mass dis-
appears;
ii/ in fundamental physics, the right variable is not t but x 0 (or ix 0 ),

which is already known although its full meaning is not clearly ac-
cepted.
But, as all variables are now space variables there is no more room in
'relativistic' physics for a constant, like c, that would be a speed: that is
why c disappears, as may be seen in relationship (1). And, of course, I
write 'relativistic' in inverted commas, because if there are no speeds,
they are neither relative nor absolute.
Thus, ordinary time is to be replaced by ix 0 , which is a length~or a
'length'~but not just a time expressed in unit of length, and, therefore,
ordinary time is not the right variable in fundamental physics. However
we live in ordinary time, and that is a mystery. But philosophers do
have their word in the fact that the world has to be fundamentally dif-
ferent from what we perceive, although intimately related to it. Thus,
in Kant's vocabulary the 'noumenon' is the thing as it is in itself, not as
we perceive it, and that is related to that theme of the Conference that
"the Map is not the Territory" (Korzybski about Magritte's painting of
the pipe). But it also says that what we perceive is not the territory, but
just a sort of map, useful only for practical purposes. As an example,
not known in the time of Kant, colors in our perception are just electro-
magnetic frequencies in the physical world: both are intimately related
but fundamentally different.
So, by reference to Kant I would say that fundamental physics~with
ix 0 and ko, and with a description of quantal entities by complex num-
bers, the only ones in the formalism that may account for interferences~
is 'noumenal' physics, and that ordinary physics, with ordinary time and
mass, is 'phenomenal' physics. Between the two, with ordinary time only
(and pulsation), we have 'nearly fundamental' physics. But, of course, by
'noumenal physics', I do not mean that we get all of the 'noumenon': we
get only its mathematical aspects. But, clearly, we have overstepped the
phenomenon level and that is what I really mean by 'noumenal'. (Accord-
ing to Kant we have no means of knowing anything about the noumenon,
and there we fall into disagreement. But, if nobody can know what the
noumenon is, and as we get something that is below the phenomenon
level, to know if we may say that it is noumenal in the fundamental sense
is an undecidable question.)
Coming back to the question of the 'map', according to Kant, space
and time, now we would say ordinary Euclidean 3-dimensional space
and ordinary time, are the 'a prioris' of our perception. This is well
known, although the distinction we make between ordinary space and
time and their physical versions generally seems not to be understood in
FR0 M QU ANTAL T 0 M ATE RIAL LEVEL 285
that way. But, what is less known is that according to Kant, matter-
and mechanical causality-are also 'a prioris'. They are 'a prioris' of
our understanding. Therefore, at this point I must pay tribute to the
astonishing precocity of Kant's analysis, itself rooted in the old Greek
questioning.
4. F R 0 M N0 U MEN A L T 0 0 R DIN A R Y PH Y S I C S
Now we have to come back from 'noumenal' physics to ordinary physics.

For that we have first to define a formal equivalent of ordinary time:
df xo
t=- (8)
c
Such a 'creation' is necessary to give us a framework for describing ex-
periments: thus we can say that c is a natural creation constant.
If we stop there we have 'nearly fundamental' physics. But, of course,
we want to get classical physics in its conventional form. For that we
must continue to proceed by definitions:
(9)
and:
df E
m=- (10)
c2
thus creating two very useful, but artificial, concepts. Just as, in order
to deal with thermics, we must set:
(11)
according to which the quantity of heat Q is no longer a reality of its

own. Similarly, energy and mass are not to be considered as realities of
their own. Therefore n,
like J, is an artificial creation constant. And,
by doing that, in passing from pulsation to mass through energy, we
pass from the quantallevel to the material level. (Note: the relationship
between definition and post been pointed out by H. Poincare. For the
question of 'creation' constants, see also J.M. Levy-Leblond [5].)
5. INTERPRETATION 0 F QUANTUM MECHANICS
We can now come to the interpretation of Quantum Mechanics: if we

have no more mass, we have obviously no more corpuscles, or, as we say,
particles. That is clear (even if the 'obviously' may be discussed), but
clearer still if we consider the Heisenberg inequalities. The first one is

written:
(12)
and, in accordance with what has been said above, we have to remove in
order to get back to the wave level. Asp= lik, this is very easy indeed.
We obtain:
(13)
which is, as you know if you are physicist or mathematician, something

not mysterious at all as it is just a mathematical theorem (on Fourier
transforms). Applied to free quantal entities it implies that the more
concentrated the fuzzy shape is, the more quickly it gets out of shape:
then nothing can be said about 'the' speed.
So, we have got a mathematical theorem, without li of course, and di-
rectly applicable at the wave level. It is only when you want to introduce
the corpuscle that you have to use li in order to create its momentum p
by
(14)
as a complement to equation (9) defining E. (Now we understand what is
behind that formal quantity, the momentum, as we understand what is
behind that other formal quantity, the energy, that we have got so accus-
tomed to, and we also understand the 'relativistic' link between them as
k is the spatial equivalent to w). And then, by introducing the corpuscle
and its momentum-that we call its 'quantity of movement'- therefore
coming back to the Heisenberg inequalities, we create the interpretation
problem!
But at the wave level, indeed, we have no problem. More exactly, we
have only one problem: wave collapse, or wave reduction. Of course I
should perhaps say 'wave function' instead of just 'wave', but what I
have in mind should already be clear. I also think we should say 'mys-
tery' instead of 'problem', because we must not have preconceived ideas
about noumenal physics. However, it seems to me that all the known in-
terpretations of Quantum Mechanics are attempts at not to purely and
simply accepting the reduction, or at not accepting it as something phys-
ical: the reason is that it is viewed, by those who may let think that
they want the description of the whole universe by physics be total, as
an 'ugly scar'. (The G.R.W. theory [6], [7] may also be examined in such
FROM QUA NT AL TO MATERIAL LEVEL 287
a context.) But cannot the reduction be seen as an indication that we

have to change our mind on the nature of quantal entities?
So, now is the moment, in order to advocate it, to note that the
wave reduction accounts for the said impact effects: in fact, inside quan-
tal formalism, there is no other way to precisely describe them, the
'widely' outspread wave function becoming a 'narrowly' outspread one,
for instance in the volume of an atom. And therefore the 'impossible'
wave/particle duality-nonprecisely located versus precisely located-is
not mandatory. (More precisely, we would say that there is no duality in
the formalism, and that the formalism accounts, or claims to accounts,
for every phenomenon.)
In other words, the line of thinking followed here compels us to face
the wave reduction mystery as something objectively happening and,
according to me, happening without intervention of human mind. (If I
am right such an objectivity of the reduction-occurring even if not
observed-does leave marks and therefore, according to me, it could be
tested by subsequent observations [8].) I would add that, as we have got
two mysteries, we may think that they are related (reduction making
ordinary time go ahead), and, for me, that is the real source of irre-
versibility and creation. I also think that the wave reduction accounts
for the mechanical friction and that is, of course, also related to irre-
versibility.
The next question is about reality: according to the wave nature of
the quantal entities, if we want something real to replace the mass at
quantallevel, I think that we are led to calling it 'ether'. But, of course,
it is not a material ether, and thus, a parallel may be made with what
Einstein calls "the ether of the theory of General Relativity" [9] (see also
C. Piron [10]). However, in that non-material 'quantal' ether nothing
material really moves: only properties of space are changing, so that we
can also say that it is an 'empty shape' that moves.
Moreover, it moves in accordance with the wave vector k written in the
'quantal' ether. Therefore, in that ether the absolute speed of the fuzzy
deforming shape, although not experimentally obtainable, is written ...
Accordingly, we are led to think that, among the infinity of possible in-
ertial frames, there is only one in which the supposed value of k is the
real one. (And we have also to note that, according to the relationship
of dispersion w = f(k) in 'covariant' Q.M., it appears in an understand-
able manner-and with no reference to any separate principle-that the
speed of a quantal entity may not be greater than c, suggesting that 'co-
variant' Q.M., just exhibiting the symmetry between x 0 and x, contains
the predictions of 'relativity' with no appeal to a separate principle. Of
course reciprocal is not true.)

According to these views, we would just say that to dictate relativistic
principles to the wave reduction appears to be no longer mandatory. (See
also J.S. Bell interrogated by P.C.W. Davies [11]. As to other possible
objections to the here suggested interpretation of Quantum Mechanics,
see [4].)
6. PH I L 0 S 0 PH I CAL 0 PEN IN G S
As to the philosophical openings, the main one is obviously that, when

matter no longer appears as something fundamental but as something
constituted, the old mind/matter problem has to be replaced by the
'mind/question mark' problem. Now, if the mind/matter problem was
considered as a solutionless problem, it was because of the difference of
nature between mind and matter, whereas now, we are confronted with
a question mark ... And, if we have to face the external obviousness that
the world, and our cells, obey to mechanical causality, we have also to
face the internal obviousness that we are alive, and conscious.
So, coming back to the constituent/constituted rule, we find it le-
gitimate to ask the following question: as the properties of something
constituted are not the properties of the components, is it not possible
that life comes into being in such a process from simple to complex?
But, on the contrary, we see that this path leads to matter, mechan-
ical causality and classical physics. Therefore, is it not also perfectly
legitimate to explore the reverse concept, and think that the emerging
property is not life, but death? ... In any case my conclusion is that it is
legitimate.
AFFILIATION
Emmanuel Hemmerlin
50 place du Quebec
M ontpellier, France
REFERENCES
[1] Heisenberg, W., Physics and philosophy, Harper & Brothers, New
York, 1958.
[2] Levy-Leblond, J.M., Une matiere sans qualites?, lecture at College
de France, 1990.
[3] Hemmerlin, E., On the use of the relationship hv = mc2 to trans-
form physical equations.
F R 0 M QUANT A L T 0 MATER I A L LEVEL 289
[4] Hemmerlin, E., On the concept of energy in Quantum Mechanics.

[5] Levy-Leblond, J.M., "On the Conceptual Nature of the Physical
Constants", Rivista del Nuovo Cimento, 7, 1977, pp. 187-214.
[6] Ghirardi, G.C., Rimini, A., Weber, T., "Unified dynamics for
microscopic and macroscopic systems", Physical Review, D34, 1986,
pp. 470-491.
[7] Ghirardi, G.C., Rimini, A., and Weber, T., "The Puzzling
Entanglement of Schrodinger's Wave Function", Foundations of Physics,
lB, 1988, pp. 1-27.
[8] Hemmer lin, E., Objectivity of the reduction in Quantum Mechanics:
towards an experimental answer?
[9] Einstein, A., "Ether and Theory of Relativity", lecture at the Uni-
versity of Leyde, 1920, in: Refiexions sur l'electrodynamique, l'ether, la
geometrie et la relativite, Gauthier-Villars, Paris, 1972.
[10] Piron, C., "Quanta and Relativity: Two Failed Revolutions." in:
Einstein meets Magritte, The White Book, Kluwer Academic Publishers,
New York, 1998.
[11] Davies, P.C.W., and Brown, J.R., The ghost in the atom, Cam-
bridge University Press, Cambridge, 1986.
D. AERTS, B. COECKE, S. SMETS
ON THE ORIGIN OF PROBABILITIES IN QUANTUM

MECHANICS: CREATIVE AND CONTEXTUAL ASPECTS
1. INTRODUCTION
At the beginning of this century, a lot of changes as well in the field of

politics, art and sciences have led to a change of paradigms and ways in
which people think and interact with '(the/their) world'. That climate
of change lasted the whole century, due to the inertia of old ideas and
the required time people needed to build 'new images' (or even, new
world views) that incorporate the new findings. Nonetheless, as soon as
old paradigms get overruled, new ones appear. This is something which
also happened after the 'invention' of quantum physics. The modernist
deterministic world guided by Laplace's prime intelligence had to make
place for one in which appear pmbabilities that have a mysterious status,
as we will explain now. The probabilities of classical statistical theories
[1, 3], e.g., statistical, mechanics, thermodynamics, classical probability
calculus, have never been considered to be an obscure subject, because
they can be explained as being due to a lack of knowledge about an even-
tual deterministic underlying reality. So, these classical probabilities are
only a mathematical formalization of the lack of knowledge about the
system under study. When quantum mechanics was born as an intrinsic
probabilistic theory, the question was raised rapidly of whether these
quantum probabilities [5, 10, 13, 17] can also be explained as due to a
lack of knowledge. The field of research investigating this problem was
referred to as the search for hidden variable theories, the hidden vari-
ables describing this so called deterministic underlying reality. During
the years many theorems (e.g., the famous no-go theorem of J. von Neu-
mann [4], or its elaborations [6, 7, 8].) have shown that hidden variable
theories for quantum mechanics are impossible, indicating that quantum
probabilities are of a fundamentally different nature than classical prob-
abilities and seemingly not due to a lack of knowledge. Some physicists
formulated very clearly their opinion: quantum mechanical probabilities
are ontologically present in reality itself. These ontological (or objec-
tive) probabilities destroyed the classical picture of the world in such a
way that the search for an image of what really happens in the 'physi-
cal world' had been abandoned, and still is so in many fields of micro-
physics. As such, a large quantity of the contemporary community of
291
292 D. AERTS, B. COECKE, S. SMETS
physicists consider 'real' physics as something definitely complementary

to anything to be understood as possibly eligible by 'realism'. Without
going into any debate on this, we will show that it is indeed possible
to find a picture of quantum entities where these 'strange' probabilities
are explained. In fact, even formally, we only have to introduce a specific
new concept within the theory of physics, that was not explicitly present
before, namely a model of aspects of creation.
More explicitly, when we consider a measurement on a physical en-
tity, these aspects of creation can be modeled as an interaction of the
physical entity with its measurement environment, in such a way that a
measurement on this entity provokes a change of state of the entity that
depends on the interaction with its measurement context. As it is explic-
itly shown in [15, 16], there are two aspects that determine structural
differences, in the sense that we obtain quantum-like probability struc-
tures, if the measurements needed to test the properties of the system
are such that:
1. The measurements are not just observations but provoke a real change
of the state of the system.
2. There exists a lack of knowledge on what precisely happens during the
measurement process.
The first aspect, the change of state, can be interpreted as an 'act of
creation' on the entity under study. It is indeed the external device that
provokes the change of state during the interaction with the entity. If
there is not such a change of state, we can consider the measurement
as a discovery. The second aspect, the presence of the lack of knowledge
on the precise act of creation which results from an interaction with the
measurement context, lies at the origin of the so called indeterministic
nature of quantum measurements and can be formalized as a lack of
knowledge on the precise measurement that is actually performed [15,
16]. We can formalize the foregoing ideas somewhat more concretely in
the following way:
1. With each real measurement e corresponds a collection of determin-
istic measurements e>,, called 'hidden measurements'.
2. When a measurement e is performed on an entity in a state p, then
one of the hidden measurements e>, takes place. The probability finds its
origin in the lack of knowledge about which one of the hidden measure-
ments effectively takes place.
In [15, 16, 22, 23, 24, 26] it has been shown that we are indeed able to
recover the probabilities that appear in quantum theory by considering
every measurement as a collection of so called 'hidden measurements'
PROBABILITIES IN QUANTUM MECHANICS 293
on which we introduce a lack of knowledge through a so called math-
ematical weight-representative for the relative frequency of occurrence
of the hidden measurements when the original measurement actually
takes place. These ideas have been further developed on a formal level
within an abstract mathematical setup called the hidden measurement
formalism for physical measurements [15, 16, 22, 23, 24, 26, 27]. For an
overview of other applications of the hidden measurement approach we
refer to [18, 19, 20, 21, 24, 25, 28, 29, 30, 31, 32, 33].
In this paper we proceed as follows. First we briefly describe a classical
physical entity, and we explain the meaning of probabilities within the
framework of classical physics. Secondly, we explain in which way quan-
tum entities and quantum probabilities differ from such classical ones.
In the following section we introduce aspects of creation, and we show
how they fit in our traditional picture. Finally we present a model within
macroscopic physical reality that generates exactly the quantum proba-
bilities, and in which appear only probabilities that correspond with a
lack of knowledge.
2. P R 0 B A B I LIT IE S IN CLASSICAL SYSTEMS
In this section we introduce the notion of probability, and how it occurs

in classical physics. In classical physics we describe classical systems. A
classical system can be conceived as an entity that represents a well-
defined part of reality, and which is separated from its environment. A
second important aspect that characterizes a classical physical entity is
that we are in principle able to attach definite values true or false to
all the physical quantities necessary and sufficient to give a full repre-
sentation of our physical system. We explain this in more detail. As a
starting point we consider the elements of physical reality, introduced
by Einstein, Podolsky and Rosen who said in their 1935-paper [2]: 'If,
without in any way disturbing a system, we can predict with certainty
the value of a physical quantity, then there exists an element of physical
reality corresponding to this physical quantity'. We now quote Piron [14]
to explain what we mean by an actual property1 : 'A property is actual
(or the system possesses an 'element of reality') if it is 'certain' that the
positive result would be obtained if the experimenter decided to per-
form the corresponding experiment', where it is pre-supposed that every
property that one can attach to a physical system is testable, i.e., there
exists an experiment that verifies it. Since the set of all possible physical
quantities which could describe our entity are completely characterized
1
More details on this specific notion of a property can be found in [9, 11, 34].
294 D. AERTS, B. GOECKE, S. SMETS
by the set of all properties of the entity (actual, or not, which is called
potential [9, 11, 34]), we are for every classical entity able to attach a
value true or false to every property respectively corresponding with the
property being actual or potential 2 • Of course, in certain situations we
might not know exactly the value of every property of the physical entity
(although they really exist). This means in fact that we have a lack of
knowledge on the properties of the entity. The mathematical tool deliv-
ered to deal with such a situation of a lack of knowledge on a classical
entity is a probability measure [1, 3]. In such a probability representa-
tion, if something is true to our knowledge it requires that we attach a
value 1 to it, and if it is false to our knowledge it requires a value 0. In
a lack of knowledge situation, we are dealing with values lying between
(and not equal to) 0 and 1, representative for a lack of certainty. The
most important example where we encounter such a lack of knowledge
situation is the one of a statistical ensemble, i.e., a collection of entities
to which we attach only global parameters (for example temperature,
density, velocity distributions etc ... ). Consider for example a quantity
of a liquid within a barrel: For every individual molecule in this liquid,
we know that it is located within the barrel, but we don't know where
exactly in the barrel; as a consequence, the properties related to the
individual molecules are not exactly known; nonetheless, we do know
which ones are very probable and which aren't.
3. P R 0 B A B I LIT IE S IN QUANTUM SYSTEM S
For more than sixty years now, scientists have been trying to really un-
derstand quantum mechanics. This long period is at least partly due to
the fact that we are very attached to the classical picture of nature. Ac-
cording to Einstein and many others-who believed in what we can call
the classical picture of nature -the quantum description which should
represent an entity is incomplete. This has been a point of discussion for
many years, and we have now come to the conclusion that we believe that
it is indeed possible to give a complete realistic description of a quantum
entity. But what was the cause for this so called incompleteness? The fact
is that when the formalism of quantum mechanics had been constructed
around 1927 to describe microscopic systems, the only mathematical ex-
pression to describe an entity was an expression about possibilities or
2
At this point, the present reader might get a feeling of conceptual overkill. There-
fore, we already mention that it will be here that the troubles start in the quantum
case: false will not be implied by potential. As such, this more evolved conceptual
picture will enable us to point at the differences between the classical and quantum
system in a scheme that hosts both of them [9, 11].
P R 0 B A B I LIT IE S IN QUANTUM M E CHAN I C S 295
tendencies. In most cases where we use the standard quantum formalism,

we can't give definite values to all physical quantities which are necessary
and sufficient to give a full representation of our physical system. The
probability assignments are the most the theory can say about a quan-
tum system we want to observe or to measure. In fact, one can prove that
for every possible preparation we could make when we want to check if
an entity has a well defined property, there are always other properties
about which we cannot be certain. Referring to the previous section, this
means that not all potential properties are false. When we perform an
experiment that tests them, both answers might occur, namely false and
true. In fact, only very few properties that are not actual will be false.
When we use the standard quantum formalism, one has to accept that
for these properties that are not true and not false, there exist proba-
bilities that do not describe just a lack of knowledge about the exact
situation of our entity, but that are a priori linked with the quantum
world. That's why one also calls them objective probabilities, contrary
to the classical probabilities that refer to an incomplete knowledge of
a given situation. These classical probabilities, which are the same for
every human being who wants to observe an entity, correspond in fact
to a lack of knowledge of the 'state' of the entity. 3 Until now, one has
not been able to give a real content to this picture of objective (onto-
logical) probabilities because they imply that the existence of an entity
can be asserted only with a certain level of probability. In this article
we want to explain another kind of view which doesn't need objective
probabilities. This view, which we have called the hidden measurement
approach, assumes that quantum probabilities are caused by a lack of
knowledge such as we are used to from the classical case. Here, quantum
probabilities aren't objective, but the kind of lack of knowledge differs
from the one in classical physics in the sense that a lack of knowledge on
the state of the system gives rise to a classical probability model, and a
lack of knowledge concerning the measuring process itself introduces a
non-classical model. The presence of the quantum probability is due to
the fact that before the measurement process starts, it is not determined
which of the hidden measurement processes shall take place. To explain
what we really mean here, we shall first have to take a closer look at the
aspects of creation which are inherent to measurements that change the
state.
3
We consider the quantum state to give us an as complete as possible representation
of the elements of reality of the entity under consideration. With the words of C. Piron
[14]: 'The state is nothing else than the collection of all actual properties of the given
system.'
4. ASP E C T S 0 F C REA T I 0 N
Within the hidden measurement approach of quantum mechanics we are

able to explain the appearances of quantum-like probabilities. Measuring
or observing a physical reality involves to some extent in the quantum
case, also the creation of the reality observed. What we mean is that
some of the elements of physical reality are being created in the course
of measurement. The act of creation that sometimes takes place dur-
ing a measurement is something we are familiar with from our ordinary
macroscopic world. The fact that it only now attracts our attention is
that in the quantum case, measurements that in ordinary reality are
just observations also seem to contain a creative element in the quan-
tum case. We will give an example, which has already been put forward
in [20], and which will explain the 'creation-discovery'-view a bit more.
Consider a survey to determine the opinion on nuclear energy of an arbi-
trary selected group persons. A few interviewers have to find out whether
one is 'for' or 'against' the use of nuclear energy. The aim is to obtain
the true opinion of the members of the survey, therefore the interview-
ers have been provided with full information about the problem. Before
an interviewer asks the question to find out if one is 'for' or 'against'
nuclear energy, a lot of the members of our survey group will already
have a strict opinion (for or against) on the matter, while other mem-
bers haven't thought the subject over in advance and so haven't made
their mind up yet. By asking the question the interviewer will, from
those members who had a strict opinion in advance, 'discover' if they
are for or against nuclear energy. The ones who didn't have an opinion in
advance, so the ones who are in doubt about the subject when the ques-
tion is being asked, will receive enough information so that also they can
give a 'for or against' answer. We see here that by this experiment per-
sons are pulled out a 'doubtful'-state into a 'for or against'-state. This
change of state is caused by the creation-aspect that happens during
the measurement or survey. How the change of state will evolve in the
'doubtful-person'-case, depends for a large part on the persuasive force
of the interviewer and on the information he offers. In other words : we
say that the change of state depends in this 'doubtful-person'-case on the
influence of the measurement context. When we are only dealing with a
discovery, there's almost no influence of the measurement context. Let's
now look at the probabilities. When an interviewer meets a member of
the survey group, he does not know his/her opinion and without explic-
itly asking the question, he can only make a probabilistic prediction. We
make the following assumption on the probabilities characteristic of the
interaction with the interviewer and the doubtful person: an interviewer
P R 0 B A B I LIT IE S IN Q U ANT U M ME CHAN I C S 297
meeting such a member of the survey has a 40 % chance of finding a

person who will be for nuclear energy and a 60 % chance of finding one
against (here, we attribute probabilities to the change of state of per-
sons). Suppose that after we have asked the question, of all our 1000
members we will find 400 for and 600 against. Possible initial distribu-
tions of the members of the survey which yield this result could be: (i)
in advance, 400 are for and 600 are against, (ii) in advance, alllOOO have
no definite opinion, (iii) n are for (we choose n between 0 and 400), ~n
are against, and 1000-~n have no opinion. We want to stress the fact
that the result of the survey depends both on the initial distribution of
the members of the survey and on the assumption we made on the in-
teraction of the interviewer with a doubtful person. We will now analyse
this in more detail. Although all above examples yield the same result,
this result emerged from situations with a different nature. In (i) there
was no act of creation by the interviewer since all members of the survey
had a definite answer in advance. Thus this situation corresponds to a
classical situation (compare to 'all properties are true or false' in section
2). In all the other situations the result is partially due to our proba-
bilistic assumption of the interviewer creating an answer 'for or against'
during his interaction with a member of the survey group. The proba-
bilities attributed to the change of state of a doubtful person, can not
be attributed to a lack of knowledge of a more complete specification
of the real state of the person, because a more complete specification
simply does not exist. The probabilities about the state in which the
other persons with a fixed opinion will end up, can indeed be attributed
to a lack of knowledge of a more complete specification of the real state
of the person. As we will point out in the next paragraph, the creation-
discovery picture gives a very natural explanation for the existence of
quantum probabilities. Quantum probabilities are attributed to the lack
of knowledge concerning the measurement process and not to the lack
of knowledge on the state of the system.
5. THE QUANTUM MACHINE
We'll go more into detail to justify our creation-discovery picture for

the quantum case. The way to do this is to talk about what we call the
quantum machine 4 . This machine provides a model that is applicable to
a spin-~ quantum entity. The 'lack of knowledge' on the measurement
process will be presented by the parameter ). E [0, 1]. The entity in
4
A variant of this model has first been introduced in [15, 16]. A similar modelization
is also applied in [12].
Fig. 1 is a point particle that moves on the surface of a sphere with

center 0 and radius 1. If the point has coordinates v, we denote the state
corresponding to this point as Pv· A measurement eu on the entity in a
state Pv is defined in the following way: Consider a straight line segment
with one of its endpoints in the point u of the sphere, and the other
one in the antipodal point -u. We'll denote this segment by [-u, u]. We
project v orthogonally on [-u, u] and obtain the point v'. This point
defines two segments [-u, v'] and [v', u] (see Fig.1).
Figure 1. Illustration of a measurement eu on our model for a spin-~ entity when the
initial state is Pv.
Consider now a random variable ,\ defined on the segment [-u, u], and
suppose that the relative frequency of appearance of the possible ,\ is
uniformly distributed on [-u, u]. If,\ E [ -u, v'], the point corresponding
to the state of the entity moves to u along [v', u] and we obtain a state
Pu· If ,\ E ]v', u], the point moves to -u along [v', -u] and we obtain
P-u· As a consequence, there are two possible outcome states for this
measurement eu: Pu and P-u· From a more mechanical point of view, the
segment [-u, u] can be seen as a uniform 'elastic', that can break in every
one of its points in [-u, u] with the same probability. Before we put the
machine on, there is no way to find out in which point the elastic will
break. In the case we push the start-button of the quantum machine, we
get the following image: the point particle falls from its original location
on the elastic such that its 'falling-direction' is orthogonal to the elastic,
and sticks to it. Then the elastic breaks and the particle is torn to one
of the two original endpoints of the elastic. If we decide to perform a
measurement in the u direction when we start with a particle located in
a point v on the unit sphere, it is easy to calculate probabilities for a
transition of the particle to u or -u:
1+cos,B ,8
P[v --+ u] = = cos 2 - = 1 - P[v --+ -u]
2 2
P R 0 B ABILITIES IN QUANTUM MECHANIC S 299
where (3 is the angle between the vectors u and v. These probabilities

are the same as one obtains in a Stern-Gerlach measurement on spin-~
quantum entities, i.e., this model system generates quantum probabili-
ties. For more details on this model system we refer to [15, 16, 18, 21,
22, 25, 29, 30, 32]. A similar model for a quantum entity described in a
three dimensional Hilbert space can be found in [28]. At every moment
it is possible to look at what happens within this machine. A classical
machine will be one where we install a rope which has the ability only
to break in one point, such that we get classical probabilities which are
in principle for a well defined initial location of the point particle only
1 or 0 (we omit the neglectable situation of the point-particle falling on
the breaking point, since this is an event with a so called zero measure).
Given the quantum machine, we are now able to visualize an emergence
of the quantum probabilities. As we mentioned before, the probability
appears due to a lack of knowledge about the measurement process.
In our quantum machine this lack of knowledge is exactly referring to
the point where the elastic breaks. We don't have any means to find out
where the elastic will break when we set our quantum machine in action.
The quantum probabilities might be looked at in this way. We mean that
our quantum probabilities are calculated in a way in which we take the
probability over different classical machines, put on action. Each of the
different classical machines considered has a rope of the same length as
the big quantum machine's elastic, but now the rope is only breakable
in one point. All these classical machines differ in the sense that their
ropes will break in another point if set in action. We need to consider
as many classical machines as the quantum-elastic has breaking points.
Now stated shortly, a quantum machine in action is the same as one of
all the considered classical machines in action. The only point is that we
don't know which classical machine with its specific breaking point-rope
will actually be put into action. This special classical machine is our hid-
den measurement. Although this machine may be hidden, it influences
the result of the measurement in such a way that it is responsible for
the creative aspect.
AFFILIATION
Diederik Aerts,
Senior Research Associate at the Fund for Scientific Research-Flanders,
Center Leo Apostel & FUND, Brussels Free University,
Belgium
300 D. AERTS, B. GOECKE, S. SMETS
Bob Goecke,
Post Doctoral Researcher at the Fund for Scientific Research-Flanders,
Center Leo Apostel f3 FUND, Brussels Free University,
Belgium
Sonja Smets,
Research Assistant at the Fund for Scientific Research-Flanders,
Center Leo Apostel f3 FUND, Brussels Free University,
Belgium
REFERENCES
[1] Kolmogorov, A.N., Foundations ofthe theory of probability, Trans.

N. Morrison, 1933.
[2] Einstein, A., Podolsky, B., and Rosen, N., "Can quantum-
mechanical description of physical reality be considered complete", Phys.
Rev., 47, 1935, p. 777.
[3] Halmos, P., Measure theory, Van Nostrand, 1950.
[4] von Neumann, J., The Mathematical foundations of quantum me-
chanics, Princeton University Press, Princeton, 1955.
[5] Mackey, G.W., Mathematical foundations of quantum mechanics,
Benjamin, 1963.
[6] Jauch, J., and Piron, C., "Can hidden variables be excluded in
quantum mechanics", Helv. Phys. Acta, 38, 1963, p. 827.
[7] Bell, J .S., "On the Einstein-Podolsky-Rosen paradox", Physics, 1,
1964, p. 403.
[8] Kochen, S., and Specker, E.P., "The problem of hidden variables
in quantum mechanics", J. Math. Mech., 17, 1967, p. 59.
[9] Piron, C., Foundations of quantum physics, W.A. Benjamin, 1976.
[10] Suppes, P. (ed.), Logic and probability in quantum mechanics, D.
Reidel Publishing Company, 1976.
[11] Aerts, D., The one and the many, Doctoral Dissertation, Free Uni-
versity of Brussels, 1981.
[12] Gisin, N., and Piron, C., "Collapse of the wave packet without
mixture", Lett. Math. Phys., 5, 1981, p. 379.
[13] Holevo, A.S., Probabilistic and statistical aspects of quantum the-
ory, North Holland, 1982.
[14] Piron, C., "Quantum mechanics fifty years after", in Lahti and
Mittelstaedt, P. (eds.), Symposium on the foundations of modem phy-
sics, World Scientific, 1985, p. 208.
P R 0 B A B I LIT IE S IN QUANTUM ME CHAN I C S 301
[15] Aerts, D., "A possible explanation for the probabilities of quantum
mechanics", J. Math. Phys., 27, 1986, p. 202.
[16] Aerts, D., "The origin of the non-classical character of the quan-
tum probability model", in: Blanquiere, A., Diner, S., and Lochak,
G. (eds.), Information, complexity, and control in quantum physics,
Springer-Verlag, Wien-New York, 1987, p. 77.
[17] Gudder, S.P., Quantum probability, Academic Press, 1988.
[18] Aerts, D., "A mechanistic classical laboratory situation violating
the Bell inequalities with 2J2, exactly 'in the same way' as its violations
by the EPR experiments", Helv. Phys. Acta, 64, 1991, p. 1.
[19] Aerts, D., "A macroscopical classical laboratory situation with
only macroscopical classical entities giving rise to a quantum mechani-
cal probability model", in: Accardi, L. (ed.), Quantum Probability and
related topics, volume VI, World Scientific Publishing, Singapore, 1991,
p. 75.
[20] Aerts, D., De Muze van het leven, quantummechanica en de aard
van de werkelijkheid, Pelckmans-Kok Agora, 1993.
[21] Aerts, D., "Quantum structures due to fluctuations of the mea-
surement situations", Int. J. Theor. Phys., 32, 1993, p. 2207.
[22] Aerts, D., "Quantum structures, separated physical entities and
probability", Found. Phys., 24, 1994, p. 1227.
[23] Coecke, B., "Generalization of the proof on the existence of hidden
measurements to experiments with an infinite set of outcomes", Found.
Phys. Lett., 8, 1995, p. 437.
[24] Coecke, B., Hidden measurement systems, Doctoral Dissertation,
Free University of Brussels, 1995.
[25] Aerts, D., and Aerts, S., "The hidden measurement approach to
quantum mechanics", in: Proceedings of the international conference on
the foundations of physics, Oviedo, Spain, Kluwer Academic, 1997.
[26] Coecke, B., "Classical representations for quantum-like systems
through an axiomatics for context dependence", Helv. Phys. Acta, 70,
1997, p. 442.
[27] Coecke, B., "A Classification of classical representations for quan-
tum-like systems", Helv. Phys. Acta, 70, 1997, p. 462.
[28] Aerts, D., Coecke, B., D'Hooghe, B., and Valckenborgh,
F., "A Mechanistic macroscopic physical entity with a three dimensional
Hilbert space description", Helv. Phys. Acta, 70, 1997, p. 793.
[29] Aerts, D., and Coecke, B., "The creation-discovery-view: to-
wards a possible explanation of quantum reality", in: Dalla Chiara,
M.L., Proceedings of the 1Oth international congress of logic, methodol-

ogy and philosophy of science, Kluwer Academic, 1998.
[30] Aerts, D., "The hidden measurement formalism: what can be ex-
plained and where paradoxes remain?", Int. J. Theor. Phys., 37, 1998.
[31] Coecke, B., and Valckenborgh, F., "Hidden measurements, au-
tomorphisms and decompositions in context dependent components",
Int. J. Theor. Phys., 37, 1998, p. 311.
[32] Aerts, D., "The entity and modern physics", in: Castellani, E.
(ed.), Identity and individuality of physical objects, Princeton University
Press, 1998.
[33] Coecke, B., "A representation for compound quantum systems
as individual entities: hard acts of creation and Hidden Correlations",
Found. Phys., 28, to appear (1998).
[34] Moore, D.J., "On state spaces and property lattices", Stud. Hist.
Phil. Mod. Phys., to appear (1998).
INDEX 303
AO')'O<;, 232 Kant ian position, 249

phenomenalism, 245, 248-250, 252,
20th century cosmology, 140 254
phenomenology, 251, 254, 260
act, 174 philosophical position, 248, 255
Anaximander, 231, 232 philosophy, 245
annual variation, 104, 106 realism, 197, 248-252, 255
aperiodic, 93 wave-particle complementarity, 198,
archetype 199, 214, 220, 221, 238
Jungian, 149 wholeness, 260
Aristotle, 54, 59, 81, 151, 163-173,
176-178, 180, 183, 188 category, 34, 40, 54, 63, 65, 67, 122,
logic, 172, 173, 176 140, 173, 177, 178, 180, 185-188,
metaphysics, 173, 176 190, 200, 244
Artificial Intelligence, 33, 40, 41 quasi-categories, 189
connectionist, 34, 38 cell separation, 129, 131-133, 135
strong, 43, 44, 47, 50-52, 55, 57- chaos,64,69,84,93,96,97,201, 225,
59 226
symbolic, 38, 41 charged particles, 110, 111
ascertainment, 213 chemical elements, 150-154, 156
astrology, 146 chemical test
Chinese, 146 Piccardi, 103, 104
astrophysics, 105 Church-Thring thesis, 35
atomic object, 251 cited scientific literature, 121-123, 127-
atomism, 80, 166, 176-178, 228, 230 129
authority, 17, 21, 26-29, 50 classical, 38, 39, 182, 183
autonomy, 19, 20, 22-24 atomism, 178
autopoiesis, 40 Einstein's [atomic] realism, 245, 247,
252, 255
Bell's theorem, 182, 198-200, 204, 205, logic, 172
207, 288 machine, 299
Beyers, Leo, 70 mechanics, 179
Bhagavad-Gita, 20 philosophy, 167
Bible, 20, 21 physics, 180, 197-203, 206-209, 221,
bifurcation, 95 225,226,228,230,231,245,248,
Big Bang theory, 143, 148 249, 251, 285, 288, 293-295
biomedical applications of magnetic incompleteness, 226
fluids, 120, 122, 123, 126-129, 135 probability, 291, 295, 299
Birkhoff, Garrett, 182 realism, 1
Bohm, David, 182, 198,200, 205, 207, cluster analysis, 151
254, 257, 258, 260 cognitive
Bohr, Niels, 197-208, 214, 220, 221, mind, 244, 246, 256, 258, 260, 261,
227, 243, 245-252, 254-256, 259, 263
260 psychology, 33, 34
304 INDEX
science, 31-34, 36-41, 43, 83 density, 66, 150, 227-230, 238, 260,
color vision, 151, 152 269, 294
common good, 28 Derrida, Jacques, 66, 201
common sense, 257 Descartes, Rene, 4, 31, 32, 64, 65, 68,
complementarity, 179, 197-202, 204, 152, 176-178, 181, 217
206, 208, 213, 214, 216, 220, 222, describability, 47, 207, 215, 219
227, 248 determinism, 80, 89, 91, 93, 226, 231,
linguistic, 214-217, 227, 248, 249, 247, 281, 282, 291, 292
251, 253, 254 Dirac, Paul Adrien Maurice, 271, 282
complex adaptive systems, 89 disciplinary structure of references,
compound systems in physics, 187- 121-123, 127-129
189 divine, 18-22, 25, 181
comprehension axiom, 215 doctrine, 17, 165, 177, 253, 256, 260
computationalism, 35 dogma, 27, 29, 50, 102, 200, 208
computer science, 31, 33 drug carrier, 129-131
Comte, Auguste, 253 Dummett, Michael, 244
concept of particle, 238, 242
connectionism, 38, 39, 41 earth magnetosphere, 108
continuous whole, 245, 246 eigenvalue decomposition, 152-154
continuum, 171, 179, 184, 237, 260 eigenvibration, 149, 152, 156
conventionalism, 3, 9-12, 91 Einstein, Albert, 1-13, 17, 25-27, 29,
corpuscular theories, 173, 176, 285, 53, 63, 64, 89, 140, 141, 144, 181-
286 184, 197, 200, 202, 204, 205, 226,
cosmic particles, 109 227, 239, 243, 245-247, 250-252,
cosmic rays, 112, 113 254, 255, 260, 268, 273, 275, 287,
creation, 59, 189, 260, 285, 287, 296, 293, 294
297 classical [atomic] realism, 245, 247,
act, 292, 296, 297 252, 255
aspects, 292, 293, 295, 296, 299 Einstein-Podolsky-Rosen (EPR), 197,
consciousness, 66 226
creation-discovery-view, 296, 297 Elea, 164-166,168,170-172,174,178,
Divine, 181 180, 185
human, 188, 189 layer, 178
information, 40 noumenon, 170
intuition, 53 ontology, 173, 176, 183, 184
Leibniz, 181 paradox of Zeno, 166
of properties, 189 Parmenides, 163, 165
Popper, 53 Parmenides' realism, 164
theory, 189 Zeno, 166
CSD (Differentiated Sector Colonies), elementarity, 238
113 empiricism, 1, 3-8, 12, 59, 69, 72, 74,
Cusanus, Nicolaus, 165, 185 75, 79-82, 85, 144, 152, 164, 168,
17 4, 175, 182, 184, 185, 187, 188,
data analysis, 150, 151, 159 251, 252
de Broglie, Prince Louis, 239, 240 energy, 84, 102, 104, 113, 114, 156,
de Chirico, Giorgio, 64 181, 199, 206, 225, 229, 231, 238-
INDEX 305
240, 268, 273-275, 283, 285, 286, induction, 189

296, 297 Forms or Ideas, 169
entanglement, 189 Foucault, Michel, 64, 89
environment, 29, 59, 63, 72, 97, 102- foundation(s)
105, 111-114, 181-183, 188, 293 basic categories, 185
measurement, 292 chrono-chemistry, 114
epistemology, 3, 4, 34, 36, 37, 39, 56, cognitive science, 33, 34, 36, 39
82,85, 164,168-170,172-175,198, epistemology, 39
201, 207, 226-228, 231, 243-246, General Relativity, 184
249-251, 253, 258, 259, 281 hypothetical, 7
evolutionary, 90, 243 knowledge, 173
pragmaticist, 255 logic, 32
EPR, 246 magnetic fluids research, 119
EPR argument, 203-205 mathematics, 186
ethics, 17, 25-29, 233 morality, 18
evolution, 84, 95, 96, 114, 140 natural science, 178
existence of objects, 172, 178 philosophy, 176
experiment, 43, 95, 97, 101-104, 107, physics, 188
112, 113, 130, 133, 134, 159, 164, Fourier, Joseph, 158, 270, 272, 286
17 4, 176, 180-185, 188, 189, 198- fractal structure, 93, 97
200, 202--206, 208, 209, 214, 218, functionalism, 33-36, 171
220-222, 225, 226, 228, 237, 241,
246-248, 250-252, 254, 256, 257,
Galilei, Ga!ileo, 174, 176, 179, 231
259, 260, 285, 287, 293, 295, 296
Gassendi, Pierre, 177-179
arrangement, 252
general relativity, 2, 140, 184, 208,
double-prism, 214, 219-221
267, 268, 273-277, 287
situation, 252
Geneva School approach, 184, 185,
standardasing rule, 164
189
thought, 3, 256
geophysics, 105
factor analysis, 151-154, 158, 159 God, 17, 57, 66, 145, 171, 180
factor plots, 153, 154, 158 Godel, Kurt, 44-49, 51-53, 55, 59,
far- from-equilibrium, 95 201, 227
fictions, 253 incompleteness theorem, 44, 51, 227
field, 269
earth magnetic, 109, 110 Heidegger, Martin, 64, 225, 232
Fight of the Universals, 172 Heisenberg, Werner, 159, 199, 201,
First Mover, 172 217, 227, 254, 282, 286
folk psychology, 34, 36 Heraclitus of Ephesus, 163-168, 170,
force, 179, 180 171, 174, 175, 178
formal system, 40, 43-52, 57 realism, 164, 167, 174
formal theory, 213, 219, 227 hermeneutics
formalism, 48, 50, 54-60, 90, 129, 150, physical, 232
155, 174, 182, 186-189, 198, 200- hidden variables, 197, 198, 200, 204,
203, 206, 208, 213, 221, 248, 255, 205, 207, 208, 227, 291
272, 274, 282, 284, 287, 291-295 Hiley, Basil J., 254, 257, 258, 260
306 INDEX
Hume, David, 3, 4, 6, 9, 10, 33, 244, Kafka, Franz, 68, 69

251 Kant, Immanuel, 3, 4, 54, 65, 244,
hyperrefiexivity, 63, 69 249-252, 256, 259, 284, 285
hysteresis transcendental object, 250
memory of water, 114 Kepler, Johannes, 150
Koran, 20
idealism, 175
image techniques, 131, 134 language,4,38,89, 151,186,213-215
ambiguous, 248, 251
incompleteness, 51, 247
arithmetic, 48
asymptotic, 229
categorical, 186
classical physics, 226
complementaristic conception, 214,
consistency, 48
219, 222
Godel, 44, 201, 227
differential geometry, 268
language, 214, 216
field, 229
quantum mechanics, 203, 208, 226-
formal, 213
229, 247, 251, 255, 257, 294
general relativity, 268
representation, 197, 262
language of thought, 37
indirect representation, 261, 262
Leibniz's universal symbolic lan-
inference, 6, 9, 45-49, 57, 80, 213, guage, 32
214, 217, 219, 221
linguistic realism, 214
inferribility, 214, 217, 221, 222 linguistic relativization, 219
inflation-theory, 145, 148 many-sorted, 213
instrumentalism mathematical, 267
realism-instrumentalism, 36 mathematics, 17
inter-esse, 70 mechanics, 231
interdisciplinary, 63, 101, 114, 119, meta-language, 213
120, 122, 123, 126, 128, 129, 135 natural, 38, 221
interpretation, 11, 19, 21, 23, 46, 50, nature, 36
56, 59, 68, 70, 101, 113, 139, 147, nondetachability, 216
152, 153, 167, 213-217, 226, 231, objective scientific, 251
243, 246, 249, 250, 272, 275 programming, 215-217
quantum mechanics, 163, 182, 197, quantum measurement, 213
198,205,208,254,281,285,286, quantum mechanical measurement,
288 215, 217
intuition, 6, 9, 10, 21, 52-55, 58, 59, schizophrenia, 63, 67
174, 177 self-reference, 217
intuitionism, 55, 57 set theory, 215, 216
invariance, 72-75, 77, 80 structure, 169
Lorentz, 199, 204, 207, 210 wave-function, 229
irreversible, 95, 114, 227, 287 Laplace, Pierre-Simon, 158, 189, 291
latitude effect, 104, 109
Jacobi, C.G.J., 152 Leibniz, Gottfried Wilhelm, 31, 32,
journal citations, 122 78, 180, 181, 184, 231, 254
Jung, Carl Gustav Divine creation, 181
archetype, 149 universal symbolic language, 32
INDEX 307
world, 254 mechanics

Leucippus, 166 classical, 179
level Mendeleev, 150, 151, 153, 154
constructivity, 217, 219, 220, 222 meta-language, 213
interdisciplinarity, 121, 128 metamorphology, 71, 73-79, 81-84
reality, 169 metaphysics, 18, 43, 143, 163, 169,
life, 27, 70, 82, 288 171-178, 180, 181, 183, 184
linguistic models for quantum the- realism, 1, 19, 243, 244, 250
ory, 219 Western, 226
linguistics, 19, 26, 31, 33, 38, 188, world view, 167, 170
213, 214 microbiology, 113, 125, 127
Linnaeus, 151 mind/matter problem, 288
Lockwood, M., 257 modal causal relation, 246
logic model, 31-33, 35, 36, 38, 39, 41, 43,
Aristotle, 172 46-48, 51, 52, 59, 80, 89, 91, 92,
classical, 172 94,96, 101,110,140-142,152,188,
logic as a standardising rule, 164 197, 199, 202, 219, 227, 241, 253,
logical positivism, 2, 32, 41 255, 257, 281, 292, 293, 295, 297,
logos, 175, 226 299
Lorenz, Edward N., 92-94 oscillating universe, 144, 147
Lyotard, Jean-Fran<;ois, 63 plasma universe, 144, 147
modern science, 31, 65, 80, 173-175,
Mach, Ernst, 2-4, 230, 249
177, 243, 246, 252
magnetic discontinuity, 111
modes of ascertainment, 217
magnetic fluids, 119
monads, 78
magnetic solar cycle, 111
many-worlds, 276 Moore, D.J., 188, 189
mathematics, 10, 12, 17, 51-53, 55- morality, 17-19, 22, 24-26, 28
58, 62, 80, 99, 100, 152, 185-187, motion of the sun, 107
189, 190, 201, 225
matter, 174, 281 neurophysiology, 31
dark, 143 neuroscience, 33
Maxwell, James Clerk, 239, 240, 267 Newton, 5, 7, 53, 106, 152, 166, 174,
measurability, 213, 214, 217, 219, 221, 179-181, 184, 187, 192, 194, 195,
222, 229, 230 231, 237, 267, 268
measurement, 17, 74, 93, 151, 158, Newton, 267
187, 197, 199-205, 207, 208, 213- Nietzsche, 64
215, 217-219, 221, 226, 228, 232, NIPALS, 156
237, 246, 253, 256, 258, 260, 276, nominalism, 153, 172, 173, 177, 247
292, 296, 298, 299 non-linear equations, 79
basic sentence, 213, 218, 219, 221, non-local behaviour, 182
222 non-separability, 226, 228, 230, 231,
context, 296 233
environment, 292 nondecomposability, 92
hidden, 292, 293, 295, 296, 299 Northern-Southern dissymmetry, 108
process, 189, 297, 299 noumenal physics, 283
308 INDEX
objectivity of phenomena, 249 classical, 167

ontology, 34, 36, 55-57, 68, 164, 165, philosophy of science, 1, 2, 4, 7, 8, 10,
168-177, 179-185, 188, 189, 198, 12, 13, 34, 78, 89, 97, 243
226, 231, 244, 249, 258, 262, 267, photons, 240
269-271, 273-275, 277, 291, 295 physical reality
optics, 72 elementary nature, 237
Oresme, Nicole, 153 physics, 2, 3, 5, 7, 11, 12, 17, 47,
Ouroburos, 149, 155, 156, 158, 165 58, 71-73, 75, 78, 80-83, 105, 120,
123, 147, 152, 173, 181, 184, 186-
paradigm, 31, 38, 39, 80, 95, 97, 101, 188, 197, 204
102, 113, 150, 159, 214, 228, 232, classical, 180, 197-203, 206-209,
238, 242, 291 221,225,226,228,230,231,245,
paradox, 163, 166,168,172,179, 182, 248, 249, 251, 285, 288, 293-
185, 208, 251, 257 295
apparent, 257 incompleteness, 226
causaL 180
phys~, 226, 231, 233
Einstein-Podolsky-Rosen, 226
Piccardi, G., 101-104, 108, 111, 113,
logical, 208
114
ontological, 164
Piron, Constantin, 163, 166, 182, 184,
quantum mechanics, 164, 185
187, 188, 287, 293, 295
Russell, 215
plasma universe, 144, 147
wave-particle, 237, 238, 240, 242,
Plato,31,53-55,58,59, 151,163-170
251
Zeno, 166 realism, 169
Parmenides of Elca, 56, 163-165, 167 plenum, 180
realism, 167, 174 PLS, 157
participation theory, 169 Poincare, Henri, 2, 3, 10, 12, 229, 230,
particles and waves, 237 253, 285
Pauling, Linus, 156 Popper, Karl, 5, 12, 90, 95, 244
Pearson, Karl, 153, 253 creative intuition, 53
Peirce, Charles Sanders, 244, 247, 253, positive feedback, 98
255, 261 positivism, 1, 2, 32, 41
Penrose, Roger, 44, 49, 50, 52, 53, precipitation reaction, 102
56-58 principal axes, 153
perception, 31, 36, 38, 54, 59, 64, 67, Principia, 179
68, 72, 149, 166, 174, 175, 253, principle
256, 258-260, 262, 284 coincidence of opposites, 165, 185
phenomena contradiction, 167, 169, 171-173,
fluctuating, 111, 112, 114 185
phenomenalism, 243, 245, 248-250, identity, 172
252, 254 probability, 8, 200, 247, 270, 271, 275,
phenomenology, 40 291-296, 298
philosophy, 1-3, 5, 7-9, 11, 12, 31- classical, 291, 295, 299
34, 36, 37, 39-41, 44, 53, 56, 58, Proceedings of the International Con-
163, 164, 167, 168, 175, 200, 208, ferences, 121
219, 237, 243, 246, 262 productivity data, 128
INDEX 309
proliferation, 139, 140, 143, 146, 147 Bohr, 197, 202, 203, 206, 246, 248,
proof, 45-48, 51-55, 144, 213, 227 250-252, 255
properties, 270 classical, 246
property, 40, 43, 46, 114, 120, 150- degrees of reality, 170
152, 154, 156, 158, 169, 170, 172, Einstein, 255
176, 178, 180, 182-184, 189, 213, Einstein's classical [atomic] real-
220, 221, 229, 231, 237-239, 241, ism, 1, 245, 247, 252, 255
245, 250, 251, 258, 267-274, 277, element of reality, 183, 189, 247,
281, 287, 288, 292, 293, 295 293, 295, 296
prophet, 20 empirical, 251, 252
psychology, 24, 31, 33, 83, 152 everyday life, 82
folk, 34, 36 external, 32
formalism, 55
quantum field theory, 208, 228, 270- Code!, 55
275 Heraclitus and Parmenides, 164,
quantum gravity, 27 4, 275 167, 174
quantum logic, 182 index of reality, 293
quantum machine, 297-299 internal, 243, 244
quantum mechanics, 147, 163, 174, interpretation of physics, 11
197-201, 203-206, 208, 213, 218, intuitionism, 55, 57
219, 221, 226-229, 237, 248, 253, Kafka, 69
270, 273, 283, 285, 286, 288, 291, Kant, 54, 252
294, 296 level of reality, 169
realism, 163 linguistic, 214
quantum probabilities, 291, 293, 295, metaphysical, 1, 19, 243, 244, 250
297, 299 naive, 253
quasi-categories, 189 objectivity of physical, 1
quasi-classic manifestations, 257 physical, 47, 214
quasi-classical, 257, 258 physical reality, 197, 200, 202, 205-
quasi-locality, 257 208,227,228,247-249,251-256,
quasi-localized, 258 260, 293, 296
Plato, 53, 55, 58, 59, 151
rationalism, 1, 3, 4, 6, 7, 9, 12, 79, quantum, 228
82, 90, 164 quantum mechanical interpretation,
real, 253 163, 197, 208
real union of mind and body, 246 quantum reality, 245
rcalims realism-instrumentalism, 36
Plato, 169 reality of mathematical objects, 55
realism, 3, 11, 31, 54, 55, 78, 149, 164, representational, 244, 261
165, 167-170, 175, 188, 204, 208, Schrodinger, 253, 257, 259
230, 244, 245, 252, 253, 255, 259, Sophism, 168
292 space and time, 178, 184
absolute, 167 Spinoza, 246, 260
anti-realism, 244, 246 relative (modal) separation, 260
Aristotle, 171, 172 relative dependency, 261
310 INDEX
relative independence of reality, 261 stabilising world-pictures, 168

relativism, 17, 25, 28, 29 stable, 170
relativity, 1-3, 5, 7, 8, 11, 12, 17, 89, steady state theory, 147
143, 144, 198, 203-205, 207, 227, steady state universe, 143
230, 231, 283, 287 strange attractor, 93
general, 2, 17, 140, 184, 208, 267, subject structure and disciplinary struc-
268, 273-277, 287 ture, 129
special, 3, 17, 181, 208, 230, 240, substance, 170
268 system, 270
representation, 174, 246
incomplete, 262 technical dominion, 232
reproducibility, 101, 102 theory of the four causes, 171
RESONANCE, 149 thinness, 260
resonance, 150, 151, 154, 158, 159 third quantization, 229
Thurstone, L.L., 151
scepticism, 12, 33 time, 174, 178
Schrodinger, Erwin, 163, 177, 184, 199, time-series analysis, 112
217, 251, 253, 255-257, 259, 282 truth, 46, 245
cat, 208, 256 two-layered world, 169
wavefunction, 240 two-world structure, 169
science, 167
scientific conferences, 121 uncertainty relations, 199, 201
scientific discovery, 7 unstable Heraclitean phenomenon, 170
scientific theories, 253 vacuum, 181
scientometrics, 120 variability index, 103
screen effects, 112 verification, 253
self-consciousness, 43, 44, 48, 50, 51, violence, 28
58-60, 68, 167 void, 166
self-knowledge, 59 von Neumann, John, 182, 217, 218,
self-observation, 40, 60 222, 291
self-organisation, 40, 90, 97, 98
self-remembering, 59 water, 102
sensitively dependent, 93 wave and particle nature, 221
solar activity, 107 wave reduction, 286
solar flares, 111 wave-particle duality, 242
solar hypothesis, 105 well-formed formula, 213
solar time, 111 wholeness, 246
solar wind, 108, 110 Bohr, 260
soliton waves, 237, 240 will to power, 225
Sophism, 168 world systems, 159
space, 174, 178 world view, 24, 158, 159, 163, 167,
special relativity, 3, 208, 230, 240, 168, 170, 179, 291
268
spectral analysis, 112 zeggingswil, 70
Spinoza, Baruch, 243, 244, 246, 247, Zeno of Elea, 166
252, 255, 260, 261 paradox, 166

Metadebates On Science 1999

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Metadebates On Science 1999

Uploaded by

Copyright:

Available Formats

METADEBATES ON SCIENCE

EINSTEIN MEETS MAGRITTE: An Interdisciplinary

Volume 1 Einstein Meets Magritte: An Interdisciplinary Reflection

Volume 2 Science and Art

Volume 3 Science. Technology, and Social Change

Volume4 World Views and the Problem o[Synthesis

Volume 5 A World in Transition: Humankind and Nature

Volume6 Metadebates on Science

Volume 7 Quantum Structures and the Nature ofReality

Volume 8 The Evolution of Complexity

Gustaaf C. Camelis, Sonja Smets and Jean Paul Van Bendegem

SPRINGER-SCIENCE+BUSINESS MEDIA, B.V.

ISBN 978-90-481-5242-1 ISBN 978-94-017-2245-2 (eBook)

Printed on acidjree paper

Editorial Introduction: Metadebates on Science xvii

11. Resonance as a Unifying Principle between the Observer and 149

13. Landscapes of Sibylline Strangeness: Complementarity, 197

14. Metalinguistic Views of Quantum Mechanics and Its 213

17. "Which Side Spinoza Would Have Taken (between 243

18. General Relativity and Quantum Theory - Ontological 267

19. From Quantal to Material Level 281

20. On the Origin of Probabilities in Quantum Mechanics: 291

THE GENERAL INTRODUCTION OF

The series of books 'Einstein meets Magritte' presented here originates

@ 1999 Kluwer Academic Publishers.

Volume 1: Einstein meets Magritte: an Interdisciplinary Reflection

sein's brain: phenomenology meets cognitive science', Francisco Varela

sideration of them as converging rather than diverging. It is to be hoped

Volume 8: The Evolution of Complexity

Brussels, 5 June 1998.

Through his paintings, creating astonishing, even absurd objects and

© 1999 Kluwer Academic Publishers.

Arthur, Constantin Piron, and Heinz Von Foerster, the spectrum of

EINSTEIN, PHILOSOPHER AND SCIENTIST

In his contribution to this volume, it is argued by A.M. Adam (Univer-

theory apparently cannot provide support for ethical relativism. Nev-

COGNITIVE SCIENCE: AN EVALUATION

The second section groups four evaluations of cognitive science. As an

the complex relation between cognitive and social factors as a nonlinear

Different disciplines, namely physical chemistry, biomedical sciences and

REFLECTIONS ON CLASSICAL THINKING

Two papers take classical approaches into consideration, searching for

EPISTEMOLOGY OF QUANTUM MECHANICS

This volume concludes with a section uniquely devoted to the episte-

Emmanuel Hemmerlin's contribution regards his proposal to reduce

AN INTERDISCIPLINARY REFLECTION ON SCIENCE,

This caption was printed on the announcements of the conference early in

"WAS EINSTEIN A PHILOSOPHER?" DEDUCTION VS.

Instead of a survey, a direct attention to the literature on the debate of

Another source of the ignorance of Einstein's philosophy is due to a lack

debates on the sources of knowledge (scientific theory) and on scientific

2. A CRUDE SOCIOLOGY OF KNOWLEDGE

Did Einstein talk philosophy and criticised it with philosophers? Yes.

lications to the layman and professionals alike, Einstein's philosophical

3. RAT I 0 N ALI S M V S. EMPIRICISM

According to the article "Rationalism" in the 11th edition of the En-

4. INDUCT IV ISM V S. DEDUCT IV ISM

In his paper "Induktion und Deduktion in der Physik" published on 25

below more on this context). Second reference is in Eisenstaedt [14] by

6. CRITICAL RAT I 0 N ALI S M IS P R 0 P 0 SED

1933. This is clear to every reader of the correspondences between them

7. THE END OF CERTITUDE

As regards the question of whether a scientific theory is erroneous, this