Professional Documents
Culture Documents

THE Knowable AND
THE Unknowable
Modern Science,
"Two Cultures"
Arkady Plotnitsky
Ann Arbor
A CIP catalog record for this book is available from the British Library.
Plotnitsky, Arkady.
Includes bibliographical references and index. Published in the United States of America by
ISBN 0-472-09797-0 (cloth : alk. paper) - ISBN 0-472-06797-4
501-dc21 2001005506
A CIP catalog record for this book is available from the British Library.
Plotnitsky, Arkady.
The knowable and the unknowable : modern science, nonclassical
thought, and the "two cultures" I Arkady Plotnitsky.
p. em.- (Studies in literature and science)
Includes bibliographical references and index.
ISBN 0-472-09797-0 (cloth: alk. paper)- ISBN 0-472-06797-4
(paper: alk. paper)
1. Quantum theory. 2. Science-Philosophy. 3. Literature and
science. I. Title. II. Series.
QC174.12 .P6 2001
501-dc21 2001005506
Und darum: Hoch die Physik! Und hoher noch
-NIETZSCHE
-BOHR

Contents
Acknowledgments ix
Preface xiii
Nonclassical Thought 29
Quasi-Mathematics 109
Conclusion 235
Notes 243
Bibliography 291
Index 301

Acknowledgments
First, I would like to thank many mathematicians and scientists for their
contribution to this project through their own work and thought and
mathematics and science in this study would not be possible without them,
and I hope that they (and other mathematicians and scientists) will forgive
me for what must be improved and refined, especially if they find something
that is erroneous here. I have tried hard to avoid such errors, but there are
(now St. Petersburg State University). They are among the greatest scientists
in the world in their respective fields of quantum theory and topology, and
I was extraordinarily fortunate to have studied with them. I also thank Gus-
taaf Cornelis, Kurt Gottfried, Barry Mazur, David Reed, Philip Siemens,
grateful to Tony Gonis and Patrice E. A. Turchi for inviting me to the first
I owe a very special debt of gratitude to Jacques Derrida for his work and
grateful to him for his support of this study itself, which, while it most
ence and Cultural Theory of Duke University had a special role in my work
on this project, which in part developed from several public lectures I was
invited to give as part of the 1996-97 series "Reconfiguring the Two Cul-
tures." The series, which took place amid the debates known as the "Science
tory, and sociology of science, further helped to shape the conception of this
ter, for the opportunity to be a part of this series and to work at the Center.
I am also grateful to her for many helpful discussions and support of this
Department at Duke, and to thank Fredric Jameson for his help and support
There are many others to whom this study is indebted. It would not be
including science majors, at Duke and Purdue. Teaching courses and semi-
ships and grants. I regret that essays on de Man and quantum epistemology;
and science and gender theory, which have also been supported by these fel-
lowships, could not be included in this study. They have appeared or will
appear elsewhere.
Acknowledgments * xi
I was a fellow at the Center for the Humanities at Oregon State Univer-
sity in the spring of 1994. I am grateful to the Center and to Oregon State
University for this support and to the members of the Center and other fel-
Peter Copek, who founded the Center and was its director until his death in
June 2001, when this book was in press, offered his generous support and
grateful to the Center and to Vanderbilt University for this special honor, as
I was the first William S. Vaughn Fellow appointed. I also thank the partic-
research on the subjects of this study, for which I am grateful to the Center
The research support from Purdue University in the summer of 2000, the
fall of 2000, and the summer of 2001, and the appointment as a University
Faculty Scholar provided major help at the final stages of this work. I am
grateful to Tom Adler, the Chair of English, and Margaret R. Rowe, the
Dean of the School of Liberal Arts, for their help and support. I am also
to Victor Raskin for many productive discussions and to Pat Sullivan for
This book appears as part of the series Studies in Literature and Science,
which is published in association with the Society for Literature and Science
(SLS). I thank the Society and its members for their role in shaping this
ambiance and a forum for some of the ideas of this study. I am grateful to
Katherine Hayles for her support of the project, which was instrumental for
and to LeAnn Fields, the executive editor at the Press, who supported the
xii * The Knowable and the Unknowable
project with exemplary attention and care at every stage. I thank Marcia
LaBrenz for editorial supervision of the book, Anne Taylor for copyediting,
and others at the Press who helped to bring the book to its published form.
fully acknowledged.
Relativity, and the Science Wars," and the accompanying exchange with
Florida, appeared in Postmodern Culture 7.2 (January 1997) and 8.2 (Jan-
edited by Jean-Michel Rabate (New York: Other Press, 2001), and a por-
Theory and Nonclassical Thought at the Fin de Siecle and the Philosophy of
son and Joseph Valente (Princeton, NJ: Princeton University Press, 2001).
Preface
thinking gives rise. This thinking and these theories radically redefine the
cal theories are seen as being beyond any knowledge or even conception,
while, at the same time, affecting what is knowable. Thus, according to Niels
the statement to which I continue to return throughout this study, "we are
tain limits (which nonclassical theories establish as well) that defines these
theories. By the same token, this impossibility also defines "the unknowable"
of my title as that which is placed by such theories beyond the limit of any
upon what can be known. Indeed, as will be seen, in these circumstances, the
does not deny that there are things that are, in practice or even in principle,
ever, classical theories are not concerned with the irreducibly unknowable
xiv * Preface
placed strictly outside their limits, rather than is seen, as it would be in non-
event, on the view adopted by the present study, the knowable and the clas-
sical are one and the same. By the same token, classical theories become a
and they have also contributed and often led to the emergence of nonclassi-
cal thinking historically. Indeed classical theories provide not only a path-
way to the unknowable but, by definition, the only such pathway. For how
could we otherwise know about the unknowable, or, more crucially, how
in this radical sense, rather than only imagine it, did the unknowable not
have manifest effects upon what we can know? These manifestations, how-
classical theories theorize both the knowable and the unknowable, found in
(from that of classical theories) relationship between the knowable and the
their place in intellectual history or culture as the radical nature of the non-
classical unknowable itself. Indeed both, this relationship and the nonclassi-
this study are those exemplified, in various ways and to various degrees, in
the works of Niels Bohr, Werner Heisenberg, Jacques Lacan, and Jacques
Derrida. This study devotes a chapter to each of them (a little less in the case
serves as the primary paradigm of nonclassical theory for this study. The
extraordinary in their own right, these ideas indicate the broad historical
mathematics and science, from at least the early nineteenth century to quan-
sometimes borrow these elements from other areas of human inquiry. Reci-
possible in my exposition of them; and I have tried to do the same for the
nonscientific subjects discussed here. While the book is not a primer on the
make the mathematical and scientific parts of the book available to nonspe-
included), makes the situation to which this project belongs (and that it
asymmetry persists, even though there may be more symmetry than is often
thought and even though there are, and have always been, arguably, begin-
ning at least with Plato, more than two cultures involved, or perhaps both
more than two and less than one. Partly real and partly imaginary, the
ity and this less than unity. The Greeks might have introduced this split
when they invented mathematics, arguably the first science in the full sense
itself apart from philosophy, poetry and the arts, politics, and to some
degree even language, although it could not be born or exist without them.
But then, neither this type of invention nor this type of divide could have a
occurred one single time, even leaving aside large-scale cultural entities (that
is, cultural multiplicities), for example, Babylonian, or, later, Arab mathe-
events. All of these-emergences of new sciences, the many (more than two
xvi * Preface
and less than one) cultures that give them birth, the two cultures and divides
to which they give rise, and so forth-occur all the time, sometimes without
establish once and for all (in many cases, even provisionally) what defines
each culture and what divides them. In the case of Snow's two cultures,
however, the divide persists. It is equally difficult to say whether the Snow
more than a dream. One of the persistent effects of the Snow divide is the
asymmetry, just invoked, of the ways in which we discuss the two cultures.
The nature of this asymmetry, or of the Snow divide itself, is outside the
places greater demands on a project like the one undertaken here and,
tation, however careful and rigorous one tries to be. Most of my arguments,
moreover, would apply whether or not one agrees with Bohr's interpreta-
very least viable and effective, even if not inevitable, however troubling or
While, in accordance with the outline just given, conceived more broadly,
the argument of this book is also a response to both long-standing and more
recent debates concerning the two cultures. The most recent stage of these
debates also involves what has become known as the "Science Wars," fol-
lowing the appearance of Paul R. Gross and Norman Levitt's book, Higher
Superstition: The Academic Left and Its Quarrels with Science (1994) and
Preface * xvii
Alan Sokal's hoax article published in the journal Social Text (1996). A
and later in England and the United States under the title Fashionable Non-
related publications have expanded these debates still further, both intellec-
intellectual scene. One of the aims of this study is to contribute to more pro-
hand, and in the humanities and social sciences, on the other, is, I would
The following outline is designed to help the reader navigate through this
ing its argument(s). For example, one might, after reading chapter 1 (which
ment in the first section, while the details of quantum mechanics and com-
(especially the second section of the latter) for the conceptual discussion of
the other hand, the actual sequence of the chapters is designed to make the
This chapter explains the key terms of the book and sketches the broad lin-
own key concepts and some of the key aspects of the century-long debate
patible with Einstein's relativity theory.) The EPR argument played a crucial
role in Bohr's thinking and indeed forced, but also enabled, him to refine his
thought believe.
tation and its radical nature to the maximal degree possible. The most rad-
ical aspects of his interpretation have rarely been given their due even in sci-
entific and philosophical literature. One of the main reasons for this neglect
is that the more subtle and sometimes minute nuances of Bohr's argument
on the subject.
mental to the general argument of this study concerning the nature and via-
Lacan's Quasi-Mathematics
which extends from imaginary and complex numbers, such as the square
Outline of the Chapters * xxi
ing that used in quantum mechanics (which, in fact, crucially depends on the
The context itself just invoked is, as is well known, explicit in Lacan's
deploys ideas borrowed from these and related areas of mathematics. While
mathematics posed during the Science Wars. Lacan's work has occupied a
special place in academic and intellectual debates for quite a while now,
logic and topology (the latter will be considered in some detail here in the
Lacan's work). This chapter, however, deals directly only with imaginary
primarily concerns, first, the way mathematics is used in Lacan and why it
things considered, Lacan is not as bad as some of his recent scientific critics
ics. Second, it concerns the philosophical (rather than more specifically psy-
tive from which this usage can be meaningfully considered. I would like to
This chapter examines the Science Wars and related debates concerning the
ideas and modern mathematics and science, including relativity, which was
rida's, when these texts engage or relate to mathematics and science and
concerning quantum mechanics, which may, with due qualification and cau-
tion, be seen as deconstructive. The chapter, thus, brings together the ques-
cal philosophy.
Chapter 5: Deconstructions
and science, most especially quantum physics, in some of the Science Wars
criticism, specifically in Gross and Levitt's and Sokal and Bricmont's books.
and the social sciences in the Science Wars books just mentioned is by now
mented upon fact that these books contain significant problems in their rep-
text of Lacan and complex numbers. As I argue there, Sokal and Bricmont
are almost worse on complex numbers than they are on Lacan, and, at
points, worse than Lacan is. I would further contend that an adequate treat-
Outline of the Chapters * xxiii
this study requires a rigorous and nuanced treatment both of this work itself
and of mathematics and science. The Science Wars books in question not
only, by and large, uniformly fail in the first task but often also, and, given
that the authors are scientists, less forgivably, fail in the second as well.
ter, the second section, closing this study, brings together Heisenberg's 1929
axis in the epistemology and, to some degree, even physics of quantum the-
Conclusion
the last day of his life), the conclusion offers a brief, codalike commentary
defined by the necessity of communicating those ideas that bring the cul-
tures involved-say, mathematics and science, on one side, and the human-
ities, on the other-to the limits of both what is known and unknown, or
in both cultures, the unknowable reaches arguably the farthest known lim-
its. This, however, may enable us to open more effective channels of com-
munication and even ethical relationships between our two cultures, or,
Abbreviations
ior, and the relationships between them. Indeed, these features define such
from some other objects, some of whose other properties, moreover, are dis-
regarded by the theory, in the way, for example, classical physics abstracts
studies. Thus, classical mechanics, the part of classical physics that deals
ory. It fully accounts, at least in principle, for its objects and their behavior
physical objects involved, say, planets moving around the sun, are disre-
garded by classical mechanics, which thus deals with idealized objects. The
equations of classical mechanics allow us to know the past state (or to oper-
ate under the assumption of such knowledge) and to predict the future state
cal physics, chaos theory, and electromagnetism (a wave, rather than parti-
same sense. While, thus, in general an idealization, within its proper limits
measurement.
Classical physics is, thus, by definition, realist and usually causal. Or,
2 * The Knowable and the Unknowable
again, it may be and usually is interpreted as such for most purposes of its
analysis and use; that is, one can combine theory (specifically mathematical
els, each comprising a set of idealized objects, whose causal behavior the
which relates to our ability to predict the outcome of such causal processes,
objects. Within its proper scope, classical physics offers both excellent
such) and from which it idealizes the objects of physical theories, and excel-
mechanics allows only for the latter, not for the former, and indeed rigor-
type with respect to the ultimate objects it considers, since it only describes
the effects of the interaction between these objects and measuring instru-
els need not be seen as describing, even approximately, the behavior, let
alone the ultimate nature, of actual physical objects (even though they can
be and often are seen as so doing), but only as serving to predict outcomes
els. The key aspects of the classical situation in physics just outlined can be
may itself be seen as derived from classical theories elsewhere and also, in
made by both Bohr and Heisenberg.1 Thus, the classical and the knowable
of my title are one and the same, denoting that which is available to knowl-
according to this view, what is knowable is classical, and only what is clas-
ducibly, in practice and (this defines the difference between classical and
with objects of classical theories. For example, it may not be, and in Bohr's
particles or waves, for example), which, however, does not imply that noth-
ing exists or everything stands still. The latter, naturally, is itself a classical
classical theories? For, in this understanding, only classical theories or, more
objects of nonclassical theories are not their objects insofar as one means by
the latter anything that can actually be described by such a theory. The
impact of such objects on what the theory can account for is crucial, how-
ever, and this impact cannot be described classically, which is what makes a
fact that the situation is subtler than just presented as concerns the ultimate
we can know-of the effects of the nonclassical upon the classical. Here and
throughout this study, I use the term "efficacity" in its dictionary sense of
power or agency producing effects but, in this case, without the possibility
such objects by any conceivable means, rather than only by means of this
theory itself. This inaccessibility does not refer to what can be further linked
itics, or culture, which a given theory would consider. Upon what happens
at these levels as such, or how it can possibly be seen by other theories, non-
classical theory does not make any claim, including the claim that it is inac-
such un-objects may not correspond in any way to the objects of a given
nonclassical theory, even though and because this theory conceives of its
indeed in any terms (such as "objects"). This latter view, too, must be seen
approximation (of the kind classical theories often pursue), but instead a
kind of irreducible rupture from whatever may physically exist. In all rigor,
speak in terms of the existence in space and time, or in terms of any specific
form of materiality we can conceive of, since they may not be applicable,
even in the sense of the remotest analogy. From this perspective, nonclassi-
is that between the knowable effects and their unknowable efficacity, for
which the ultimate objects of a given theory are responsible. The second is
that between these objects, qua objects of the theory, and that-those un-
and, indeed, one need not engage with it for most practical workings and
ever, at certain points this part of the nonclassical conceptual and epistemo-
There may, again, be a link between such un-objects and the objects of a
given nonclassical theory, and indeed what happens at these more remote
this type of link and this type of ultimate efficacity enter quantum theory
through the experimental data of quantum physics, for which certain un-
to be developed in the next chapter, this view is part of Bohr's "model" and
Strictly speaking one should put quotation marks around "nature," "ulti-
not claim the ultimate validity of this idealization for anything in nature,
beyond its role in the argument for the completeness (an exhaustive account
(and one finds such claims sometimes) that, at the ultimate level of the quan-
rather than only that we cannot see the objects of quantum mechanics in
As will be seen, Bohr argues that "there is no quantum world" in the pres-
ent sense. Or, at least, this statement and complementarity in general may
terms to be deployed in this context and often indeed subsume other denom-
tain this situation, even negatively, by any single term. Indeed, as will be
question.
The view just outlined, thus, accommodates the possibility that other the-
ories may define the un-objects of nonclassical theories differently and may
more classical and specifically realist lines. The ultimate viability of such
other hand, given that our physical theories are manifestly incomplete, mov-
6 * The Knowable and the Unknowable
most available quantum theories and even for relativity, or their extensions,
objects, such as elementary particles, which we now see as the ultimate con-
ity, the terms "particle" and "wave" cannot be applied to quantum objects
limit, can the terms "quantum" or "objects" be applied, or, ultimately, any
Einstein's so-called special relativity of 1905, which deals with the propaga-
tion of light in a vacuum. According to this theory, this speed (in a vacuum)
relative to it. It follows from the latter fact that in the case of light itself such
classical properties as time dilation cannot apply. Were it possible (it is not)
clock would stand still. Relativity may, however, be seen as a classical the-
mechanics was introduced. (As will be seen, the question whether such a
physics may be statistical, too, but not quite in so radical a way as quantum
An Introduction to Nonclassical Thought * 7
ment "God does not play dice" may well be true, but only in the sense that,
as Bohr observes on several occasions, it is not clear in what sense one can
may no more be applicable here than any other.) Eventually (in 1953, one
year before he died) Einstein came to accept this point, even though he
would still, apparently for this very reason, clearly prefer God's playing dice
rather than the games of quantum physics and its predictions, effective as
sical theories elsewhere), while it may be seen as merely auxiliary and ulti-
mately dispensable in classical physics. Indeed, one could also define non-
describes the behavior of its objects. Indeed, as I said, it is the latter view
realist in any sense hitherto available. Nor, in the view here adopted, are
Obviously, classical theories, too, involve things that are, at least at cer-
and, consequently it is what we can know and conceive of that are different
these unknowable objects or what they idealize and their impact upon what
qualifier here and throughout this study, where it features prominently (and
ries, since they serve as a pathway, indeed the only pathway, to establishing
the existence of and the connections to the unknowable. Or, more accu-
ries. The latter are able to use them, while leaving the ultimate nature of the
idealization. The models at stake in such theories and the way they con-
account for the effects in question. Indeed, "even" is not altogether appro-
model it considers. Nonclassically, one does not make even (now "even" is
more than any other claims, as regards more remote levels, such as that of
the ultimate constitution of nature qua nature (again, to the degree this term
argued to be consistent with available data and prediction that the theory
seen as part of, or indeed as, the ultimate "efficacity" of everything to which
a given nonclassical theory relates. For example, as I said, the data in ques-
linked to such "efficacities," while the efficacities themselves are seen or ide-
An Introduction to Nonclassical Thought * 9
making any ultimate claim of unknowability upon them. It follows that the
nonclassical unknowable.
While the ideas of Bohr and several other thinkers, such as Nietzsche and
we can speak of its effects."4 Reciprocally, "it would not be possible to seri-
ment, with which I began this study, "in quantum mechanics [as comple-
sitated by his nonclassical view, the sense to be explained in chapter 2.) This
and so forth-using them very much in the present sense. In particular, non-
"the quantum world") but only of the effects of the interaction between
the efficacity of these effects. This interaction is itself quantum (and thus
ments) and hence is unavailable to classical or, again, any treatment, even
though the effects of this interaction are available to classical physical and
model of what is knowable, of their ultimate efficacity or, at least, the ulti-
the basis of the phenomena that they consider, and explain its significance
for what we can know, and utilize the (manifest) effects of this interaction,
means and what is unknowable (the same parentheses apply) by any means,
classical or nonclassical.
that could in principle become known and, hence, classical in the present
able refers to something that cannot be known or conceived not only by any
means that is now available but also by any means that could ever be avail-
as Nietzsche, arguably the first thinker of the ultimate limits of both the
nonclassical and nihilism (but, importantly, not a nihilist himself), was first
theory, when he points out that his "argumentation does of course not
tools appropriate for its comprehension." Indeed, Bohr adds "it seems
likely that the introduction of still further abstractions into the formalism
will be required to account for the novel features revealed by the explo-
ration of atomic processes of very high energy" (PWNB 3:6). The history of
level of the effects of the unknowable is generated all the time. The
On this view, the question could only be whether or not the phenomena
enable any further access to the ultimate objects of a given theory and the
to some degree this change may be seen as in practice taking place in quan-
tum physics, whether the practitioners themselves like it (or even recognize
it) or not. It transpires, for example, when the results of experiments are
ical formalism that do not refer to space-time processes at the level of quan-
them, many others who also thought the epistemological cost exorbitant or
type, such as that of Einstein and Schrodinger, was far too complex and
in physics, only that its nonclassical character could be. Hence, Einstein and
Schrodinger did not think that they could abandon the search for what they
(which is in fact what greater completeness ultimately meant for them) epis-
inevitable or even valid account of the data it considers. Thus, most propo-
versions) belong to the latter group and, accordingly, see Bohm's theory as
Is then quantum mechanics itself, that is, the experimental data in ques-
tion in it and the mathematical formalism that accounts for it, uncircum-
interpretation? It may well be, and, while exercising caution, Bohr and oth-
ers were and many are inclined to think so. Here, however, I only argue for
tion may be challenged as well, just as, again, may be Bohr's claims (in
even a claim that such claims are ultimately impossible. We do not ulti-
than most, even if not all (although the latter possibility is not inconceivable
to the present author). My argument here, however, need not go that far. It
tions elsewhere, remain consistent (logically and with respect to the data
the current scene. This view also assumes that quantum mechanics itself
remains logically consistent and complete within its proper scope, that is, as
Bohr puts it, that it employs "a logically consistent mathematical formal-
ism" and that one cannot demonstrate that the consequences of this for-
(PWNB 2:57).
more classical alternatives, such as along the lines of Bohm's theories. While
it has attracted some public attention, the latter is a small minority view in
are is, again, a separate question, which I shall not, and for my purposes
need not, fully address here. I shall comment on some of these attempts later
tion, for example, in Michel Foucault's work, and relates to a wide and
argued, however, that the view of the "classical" here adopted is sufficiently
tion that would fully master the field(s) in question and to which all other
to sustain the assumption that this is possible or, again, ultimately possible.
But, as Derrida has stressed from the outset of his deconstructive project
even more effective once we understand the more complex dynamics under-
logic/intuition, and so forth, or more or less new, such as the classical and
example, are those used in quantum physics, say, between the wave and the
although the two types of epistemology thus designated are irreducibly dif-
ferent. They are not complementary, however, since they are not applicable
temology applies exclusively to "quantum objects" (to the degree the latter
author) for the following reason. It is true that its most radical forms may
beginning more or less with Nietzsche and then extending to the authors
actually rather limited). Indeed, it may be argued that nothing like quantum
mology, in its full measure, prior to the emergence of quantum theory, and,
then, as will be seen, some of its nonclassical effects are strangers still in
Kant or indeed to earlier critics of Sir Isaac Newton and (which is not quite
Kant began cautiously to delimit the realm within which this concept makes
sense (and to this day we are not done with this fixing of limits)" (emphasis
added).14 This is not that far from Bohr's agenda, at least from the causality
part of it ("the realm within which this concept makes sense" was delimited
by him as that of classical physics); and Bohr might well have been aware of
I shall return in chapter 2. One might also cite Ludwig Wittgenstein's state-
statement that might have been known to Bohr at some point in his lifelong
happen too, and what the law of causality is supposed to exclude will not let
causality and reality are at stake here and are mutually implicated. Bohr's,
phenomenon would indicate, and may indeed require a rereading from this
thought, at least of some of its key ingredients, can be traced as early as the
classical thinkers mentioned here, Bohr among them, credited the pre-
ever the degree of their viability) pose, first, the question of a more rigorous
16 * The Knowable and the Unknowable
cal, for example, how causal or realist, certain classical theories (wherever
they are found) really are or even can be. Or, again, more accurately, the
and Plato. At the same time, even as the tracing just indicated takes place, at
its radical limits, nonclassical thinking and, perhaps especially, some of its
community than they have ever been and are often resisted with increased
specifically during and in the wake of the Science Wars. It is this resistance
On the other hand, one can, and I here shall, use the term "radical" in the
of the word "radical." Indeed, as I shall argue, the question may well be
(it is, as I shall argue here against recent critics just mentioned, both), is yet
rigorous and radical enough for what is at stake at the philosophical limits
sequences of new physics are sometimes more radical than any "postmod-
ernism" can imagine, and they compelled Bohr to invoke repeatedly "the
phers in spite of their recognition of those aspects of new physics that (from
times as a result of their own work. The work of both Einstein and
Schrodinger, and more recently the work of John S. Bell (of Bell's theorem,
the latter case, and their views or even names are often appealed to as the
"The significance of physical science for philosophy does not merely lie in the
steady increase of our experience of inanimate matter, but above all in the
opportunity of testing the foundation and scope of some of our most elemen-
Galileo's earlier work demonstrates this point especially powerfully, but the
indicative here. This role of physics is in part why Nietzsche said: "Und
darum: Hoch die Physik! Und hoher noch das, was uns zu ihr zwingt,-unsre
Redlichkeit!" (And this is why: long live physics! And even more so that
Nietzsche's remark may be seen as part of his more general argument for
all our concepts, and forces us to develop new ones. Reciprocally, nonclas-
sical philosophy may help physics and, sometimes, even force physics' own
tested and advanced, and indeed created each other throughout their his-
tory, and, in the process, have productively shaped the ambient culture
ily leads to "right action," even though and because he also knew that even
the greatest thinkers, from Socrates and Plato on, and indeed on occasion
himself, are tempted to assume this. Or at least, they, and we, wish that such
18 * The Knowable and the Unknowable
were the case, for example, in the Science Wars, on both or, again, all sides
of it; and, naturally, it does not follow that we could not or should not, or
have no ground to, assess and criticize such action or knowledge. We can
but these are not the only grounds. Nor is it ever guaranteed that the conse-
although often they may be. Even if classical theories or classical epistemol-
ogy of ethical or political action were in fact always true, adopting them
indeed assume that we can even define either "right knowledge" or "right
action" with any certainty at any given moment, let alone once and for all?
The latter assumption, too, is something that Nietzsche and other nonclas-
sical thinkers mentioned here radically question and indeed deny. It, again,
of denying, "unless [one is] a fool," that many actions considered good or,
fact be good, or bad or evil, at any given moment. They may be so for the
most part but never always or once and for all, let alone for everybody. It is
rather a matter of rethinking why they are or may be so, or when and for
whom, and they may in fact be such for "other reasons than hitherto," and
by the Science Wars critics are often seen by the latter, as well as by many
bility and theoretical, scholarly, and intellectual norms and rigor. I would
could not deny differences between the work and attitudes of these thinkers
themselves and the fact that the reasons and effectiveness of nonclassical
many among them, however, specifically those discussed here, radicality is,
ity and discipline (in either sense) in their fields. In these cases one finds what
may even be called, strange as it may sound to their recent (and some ear-
enacted, first, after exhausting the possibilities it offers for a new configura-
An Introduction to Nonclassical Thought * 19
tion, which may in fact arise in part from within the old one. Second, it is
argues (rightly) as follows: "Modern theories did not arise from revolution-
ary ideas which have been, so to speak, introduced in the exact sciences
from without. On the contrary they have forced their way into research
physics-they arise out of its very nature. It is for this reason that the begin-
Even the point concerning the time of Copernicus might require further
However, this point does suggest (again, rightly) that there are other
ever field one considers. Thus, one finds more manifestly or (it may be
One can think, for example, of the cases of Nietzsche, Bataille, Deleuze, and
Lacan as different from those of Bohr, Heisenberg, and Derrida. These dif-
list, and indeed one finds arguments to that effect in their work. Thus, for
Nietzsche and (differently) Deleuze, one's sense of the "discipline" (in either
case. His strong sense of philosophical or even, in a certain sense, almost sci-
nature of his texts. In these cases one also confronts more complex discipli-
science. The latter are far from free from these complexities either. The spec-
here. The case of mathematics and science, or, again, specific cases, such as
plex than those involving the work of figures, such as those just mentioned,
20 * The Knowable and the Unknowable
elsewhere. Or at least this type of complexity may be kept at bay more eas-
ily in the disciplinary practice of mathematics and science rather than, say,
strongest possible case that the present analysis bypasses certain extrasci-
These aspects of the work of nonclassical theorists are entirely, and per-
haps inevitably, missed by most Science Wars critics. On the other hand, a
they are not always sufficiently familiar with mathematics and science not
(sometimes this is the case as well) but to test the epistemological limits of
aspects and limits, may in fact entail more radical philosophical concep-
tions. This knowledge, along with and reciprocal with his extreme episte-
haps what enables Bohr to go further than some of these thinkers do. This
is of course of no help to the Science Wars critics, who often have just as lit-
tle patience for Bohr and Heisenberg as they do for Lacan or Derrida. One
of their greatest problems is also this lack of patience, which is rarely helped
by anxiety, although this impatience has other sources, too. This, it is true,
may also be said about many of those on the other side, on other sides, of
the debates in question in the Science Wars and beyond. One must of course
be as critical as possible, but, again, rigorously and patiently so, for other-
wise criticism is rarely effective or even meaningful, even when we deal with
For one thing, we may, on the second, or third, look, discover that we were
that we were right. And then, the scientific critics in question themselves
infinite number of points, ... from which one gazes into what defies illumi-
nation [and sees] how logic coils up at these boundaries and finally bites it
own tail" (emphasis added).24 It may be, however, that it is in our encoun-
ters with nature that we test these limits most severely, as Bohr says; and this
is why Nietzsche invokes physics in this way, even though, while indebted
once said of what it shows or un-shows to us: "What could one have
this? "25 Accordingly, against the Science Wars critics and their anxieties and
fears, one might urge a maximal engagement with modern mathematics and
even at the risk of getting certain technical details wrong here and there.
utmost caution in this respect throughout this study. This maximal engage-
ment with mathematics and science may well be necessary if one is to reach
the limits of knowledge in these fields, and the threshold of the unknown,
and even the unknowable, which, it may be argued, defines all significant
in mathematics and science, that new knowledge, for example, that leading
of ever greater complexity. This process makes the life of knowledge more
The reasons and circumstances for the Science Wars and related criticism
and Heisenberg, ultimately became a target as well) are many, given the
diversity of the participants involved, on both or all sides, and the immense
take place. The targets, both figures and ideas, are many and diverse, and so
are the reasons for their being targeted. Some of them are obvious enough,
sometimes all too obvious, while others are quite complex. It may be, how-
not outright fear (although sometimes this fear nearly cries out, too), of
22 * The Knowable and the Unknowable
many such critics, mathematicians and scientists or others, is this: "the inex-
of Heisenberg's statement and in the broad sense Nietzsche gives to the sit-
ics and science to political practice. It may even become the source of better
matics and science. This practice, it is true, at the moment need not, by and
ories, such as and in particular quantum physics, in which one can be (and
most physicists are) quite successful without paying any tribute or even
attention to anything nonclassical. If there is one thing that the major target
of classical theories and ways of thinking (the degree of success of their chal-
lenges varies, of course). I leave aside here many other authors targeted by
the Science Wars critics, some of whom may well merit criticism (as some-
times do the authors just listed) but, by and large, may still be seen as missed
targets of the Science Wars critics. These critics are rarely, if ever, in a posi-
tion to properly discriminate between what is and what is not a proper tar-
sial strata of modern philosophical thinking, and its role in the humanities
is not unlike that of quantum theory in the discipline of physics. The works
of the figures to be specifically considered here are among the most contro-
work is controversial. But then so is, one might argue, the denomination
new concepts, indeed concepts that are forever new, thus defining it as, in
Nietzsche's famous phrase, always "the philosophy of the future." The term
"concept" itself must be, and here will be, used in the particular sense
Deleuze and Guattari give to it, rather than in any common sense of it,
itself, mathematics and science, on the one hand, and literature and art, on
the other.
different disciplines and the debates concerning these relationships over the
part given the nature of the fashion) risky term "interdisciplinarity" may be
may have different roles to play, rather than certain interactions between
such fields themselves for their own sake, be they more or less rigorous or
more or less loose. Such interactions and our views of them have sometimes
been too loose and superficial or, one might say, lacking in discipline (in
physics and, at the same time, stress the points of similarity it exhibits to our
conception of nature. We are not dealing here with more or less vague
analogies, but with an investigation of the conditions for the proper use of
making us familiar with the novel situation in physical science, but might on
the data in question in quantum physics are "objective" in the sense of being
and (again, keeping due differences in mind) of other theoretical fields. This
(rather than the appeal to objective reality at the quantum level) appears to
be the primary reason for Bohr's usage of this, in turn risky and often
(As will be seen, the term "irrational" has a more complex status in Bohr's
works.) One can of course hardly avoid hearing overtones of Kant's critical
philosophy in Bohr's invocation of "the conditions for the proper use of our
far more decisive here (however one reads or rereads Kant from this non-
classical perspective, as one perhaps ought to, as will be seen in chapter 3).
as such in chapter 2. Bohr says: "It is most important to realize that the
this kind, but with the inability of the classical frame of concepts to com-
causality (or even that of reality) and may be read as defining the essence of
it a bit to make my point more apparent: "In these theories we are presented
not with usual, if complex, intricacies of the classical kind, but with the
A very different frame of concepts may thus be required for such processes,
some of them classical and some new, nonclassical, as was indicated earlier
of theoretical practice.
recent works on the relationships between mathematics and science and the
caution) in this study. Indeed, Bohr, too, sometimes spoke more ambitiously,
briefly sketch an outline of this perspective. Many lines and points of inter-
atic ever since Max Planck's introduction of quantum theory in 1900, may
are fully defined thereby, as some have argued) to a more general intellectual
and cultural configuration that emerged around 1900 in the wake of, on the
other, certain trends in philosophy, literature, and the arts. Many of these
On the side of literature and the arts, they include (to give a limited list)
such literary icons as Franz Kafka, James Joyce, Ezra Pound, and Samuel
Beckett; Arnold Schonberg and serial music; and Pablo Picasso and Cubism
both in its own terms and in shaping the ideas of Heidegger and, then, the
however, other key links. Thus, while the work of Theodor Adorno, Walter
Benjamin, and Ludwig Wittgenstein is outside the scope of this book, the
work of Imre Lakatos, Thomas Kuhn, and Paul Feyerabend and their fol-
lowers in what has become known as science studies and related develop-
On the side of mathematics and science, key areas include relativity and
quantum physics and their extensions, and more recently chaos and com-
areas of mathematics and science that shape the intellectual landscape con-
There are many well-known connections between the authors and fields
connected to Cubism, while the latter was in turn affected by previous devel-
ideas, too, coincides (but is not coincidental) with the emergence of quantum
Joyce, Rainer Maria Rilke, Virginia Woolf, and many others has been well
its epistemology in the work of Bataille, Blanchot, Levinas, Derrida, and Paul
and irreducible, even if not always immediately manifest, which does not
respected and taken into account in exploring these connections or the work
of these figures in their respective fields. Accordingly, one also needs a care-
ferred from one field to another, whether one proceeds from the humanities
ities (or the social sciences). The present study adopts this view and tries to
implement it throughout.
Indeed, one rarely deals here even with unambivalent local determina-
tions, let alone with a single organizational synthesis, but instead with net-
fields and areas of inquiry that are involved, rather than a vertical subordi-
nation to or encompassing by any given field, bring the present view close to
tions, of which we can at least dream, as Bohr did. Accordingly, one can
never be certain what will or will not be brought together or when. Nor can
must use both), will work. This study's bet is that we might benefit by bring-
ing together a few more things than are together now, even though it
appears likely to be here to stay, at least for a while, even if the situation or
gave my "for now" qualification earlier). This is possible, given the manifest
mology either, even within the still relatively limited domain of its accep-
tance, as discussed earlier here. For, even leaving aside the persistence of
opinion are always in flux on all fronts and in all areas. One may detect a
ues. On the other hand, yet greater epistemological surprises may also be in
Nonclassical Thought
-MARCEL PROUST
efficacious
Einstein, Podolsky, and Rosen and which guides my argument in this chap-
ter. (Both articles were published under the same title, "Can Quantum-
Bohr invokes: "a final renunciation of the classical ideal of causality and a
stake. The main reason is that complementarity places the ultimate objects
ory and beyond all knowledge and conception. It is, as we have seen, this sit-
The first section serves as an introduction. It offers a brief outline of the key
junctures in the history of quantum theory and then introduces the key epis-
which no longer applies at the level of nature (quantum objects) but instead
ical and most extraordinary concepts. The fourth section considers uncer-
fifth section addresses the Einstein, Podolsky, and Rosen experiment and the
Locality has become the central issue in more recent debates concerning
quantum theory and, perhaps inevitably, has entered the current public dis-
fact "provides room for new physical laws [those of quantum mechanics],
the coexistence of which might at first sight appear irreconcilable with the
discovery of new physical laws-rather than being in conflict with the basic
ous character. Planck made his discovery in the course of his attempt to for-
mulate and then to interpret the radiation law for the so-called black body,
defined by the specific frequency of the radiation of the body and a univer-
which Planck himself termed "the quantum of action" and which turned
out to be one of the most fundamental constants of all physics. The indivis-
h and the frequency v, E = hvy.1 Planck's work and related developments rad-
ically transformed physics and our sense of the limits of our knowledge, sci-
entific and philosophical, and its claims upon nature and mind. The trans-
Quantum Mechanics, Complementarity, and Nonclassical Thought * 31
be in sight.2
continuity of radiation (assuming that Planck himself went even that far,
Planck's discovery, primarily due to Einstein. The idea met with consider-
Bohr at the time. Eventually, sometime around 1920, it won the physics
ticles. It also was around that time that the particles of light were named
"photons." The light acquired a more explicitly dual, wave and particle,
sical physics did not appear able to explain. More or less around the same
time, the more complex and pervasive nature of the wave-particle duality at
the quantum level became apparent as well. Both radiation, such as light,
wavelike according to the classical view, and particles (or what were classi-
cally seen as such), such as electrons, may manifest their existence, if not
themselves (this qualification has proved decisive for Bohr), in both wave-
objects has changed in the wake of quantum mechanics.4 At the same time,
it did not appear possible to ever observe both types of phenomena together.
On the one hand, this circumstance appeared to make the situation para-
Bohr, a way out of the paradox, almost a blessing: Since such incompatible
observational effects are always mutually exclusive and can never be simul-
ity was then used by Bohr to define the complementarity of certain features
conceptuality.
It took more than two decades to sort out the initial complexities by
means of quantum mechanics, also called at the time "the new quantum the-
itself more naturally to the nonclassical view and was indeed initially pre-
work of Max Born, Pascual Jordan, Paul Dirac, Wolfgang Pauli, and (pri-
and dealt with the motion of electrons at speeds significantly slower than
Dirac introduced his famous relativistic equation for the electron in 1928.
of the problems posed by Planck's discovery (which the old quantum theory
standard theory. The majority of even the most resilient critics, Einstein and
quantum theory. What bothered these critics and even some proponents
Quantum Mechanics, Complementarity, and Nonclassical Thought * 33
The old quantum theory dealt reasonably well (albeit not perfectly) with
statistical matters and was or, rather, seemed to have been analogous to
classical statistical physics. What was lacking was the mechanics describing
according to the modern view, which emerged during the first decades of
this century, courtesy primarily of Einstein. Both Bohr's 1913 theory of the
hydrogen atom (and much of his work that followed it prior to quantum
mechanics) and Einstein's 1916 work on the so-called induced and sponta-
neous emission did deal with individual quantum processes.6 While, how-
tum) jumps of electrons in the atom from one energy level to another, Bohr's
theory did not account for the mechanism of transition as classical physics
ity consideration into the individual quantum processes (rather than those
of quantum theory ever since and yet one more of Einstein's revolutionary
insights. These works, thus, further showed that the true mechanics was
like) mechanics was perhaps no longer possible and might even in principle
ately in the wake of Planck's discovery of his law, however, that, contrary
stein was among the first to show), this law was incompatible with a classi-
cal-like underlying picture. Indeed, for that reason, the way of statistical
was correct, even though part of his physics was wrong, an error that, as
The new quantum mechanics, which was expected to resolve these prob-
lems, was, however, nothing like classical physics. Certain aspects of the old
quantum theory, in particular, again, Bohr's 1913 theory of the atom and
most of Einstein's work, were harbingers of the new, and to many unap-
pealing, features that quantum mechanics was to retain and enhance rather
than eliminate, as some hoped it would. Skipping for the moment even
a manner analogous to classical physics. (As will be seen, it does make some
exact predictions, which, however, does not fundamentally change the epis-
ories would fail to do so), but it would not describe the behavior of physical
objects in the way classical physics would. Nor would it predict in the same
way either. Far from having eliminated chance from the theory, quantum
rather than only collective, behavior. Indeed, the collective behavior could
lawless. Quantum mechanics does not proceed in the way classical statisti-
cal physics does, from causal and deterministic individual behavior to sta-
that makes possible excellent statistical predictions but leaves little, if any
cal processes in space and time responsible for such predictions. We can
cal manner, but, in accordance with nonclassical epistemology, not the ulti-
observable phenomena (in the usual sense) can be properly related to the
non" in terms of the effects of the interactions between quantum objects and
dent on kinematics but are defined by the presence of mass of the bodies, as
kind) are involved in quantum mechanics, they are only those of certain
approach in the Como lecture is worth citing: "The new development [of
of motion by replacing from the very start the ordinary kinematical and
tory rhetoric aside, when Bohr says "fundamental" (or essential), he means
concept of new kinematics itself in his Como lecture, whose first section-
which also offers Bohr's own (much more elementary than Heisenberg's
Thus, the complexity of quantum phenomena opened the way to, even if
(the question is, again, under debate) not definitively entailed, nonclassical
that accounts for them, such as Bohr's complementarity and related inter-
tation." Not all of these reach the nonclassical limits in question in this
caution, given the differences between them and the thought of the different
figures involved. This includes those who are considered, and consider
themselves, close to Bohr. It is not always easy to ascertain the degree of this
Pauli, or (still differently) Dirac and John Von Neumann, since their episte-
mological views are not explicated to the same degree as Bohr's. I would
argue, however, that, once considered in all of its aspects, Bohr's interpreta-
tion (in the present reading or "interpretation") is unique and is the only
in Heisenberg's early work, as just indicated. One must also exercise caution
lecture in 1927 and developed by him through his life. In a way, it was his
life, as he in fact said: "It was, in a way, my life."10 In the process, however,
before it reached its ultimate and ultimately nonclassical version (by the late
Sometime between 1935 and 1938, in part under the impact of Einstein's
tal effects in question (PWNB 2:64). Instead, they would refer to experi-
mentally observed effects (the term persistently used in his writings hence-
objects. It would perhaps be more accurate, given the history of the term, to
and is one of the reasons for the complementary (mutually exclusive) char-
parts as a classical frame of reference, where we can, say, register the posi-
tion of a spot left by a "particle" (using this denomination and indeed this
classically account for the actual physical process that led to such changes in
the conditions of the measuring instruments, nor use such occurrences for
mechanics in 1925, had failed, as did those attempts (during the same
neither explain the sum total of these effects nor offer theoretical predictions
giving rise to one or the other type of phenomena (but never both types
type of phenomena (but never both together) appears (in either sense) as the
we can always arrange for such a setup and expect the appearance of the
corresponding type of phenomena (although not the same outcome for each
38 * The Knowable and the Unknowable
(in the way it can always be done, at least in principle, in classical physics).
Thus, we can observe either the wavelike effects or the particlelike effects of
the phenomena of both types at any point. In some respects the latter is true
are always particles, waves are always waves. A difference in the experi-
mental setup would not change the nature of either type of object. By con-
of "objects," while the latter are seen as, ultimately, unobservable as such,
because one cannot say that the complementary phenomenal effects in ques-
tion in any given set of experiments could pertain to the same object, even
though, technically, we can sometimes either repeat the same type of exper-
as different, unique each time, as concerns both the effect and the efficacity
under consideration.
only a human eye observing, say, the moon moving in the sky, famously
that each time in order to observe the moon (or what would be left of it) we
would need to shut something on the scale of Jupiter into it and study the
debris left. One must, however, also keep in mind the limited nature of this
analogy. While helpful, it can also be misleading, since, as will be seen later
in this chapter, the key epistemological features here discussed need not
Quantum Mechanics, Complementarity, and Nonclassical Thought * 39
out (PWNB 2:63-64, 73; 3:5). For the moment, as I said, we can neither
and both, efficacities and effects, reciprocally define each other. In this sense
Thus, while always mutually exclusive in quantum physics, the two types
(which has always been limited) for Bohr. The wavelike effects observed in
relating to this more complex situation rather than to any space-time wave
sidered later), to these effects and only to them, a view, again, anticipated by
refer to and indeed (this is sometimes forgotten) define its objects, such as
40 * The Knowable and the Unknowable
diately in the wake of quantum mechanics that one could never apply all of
the classical particle properties, or, conversely, all of the classical wave prop-
objects themselves. Indeed, one would not be able to apply all the necessary
ment even in this case, although considered in themselves, outside the con-
such properties would come into play in the classical manner in the classical
also follows that the situation just described pertains to the quantum inter-
action between quantum objects and the quantum aspects of the measuring
in Bohr's interpretation.15s
One can, thus, define the key variables involved in a perfectly classical
also why we can never simultaneously apply all the classical particle prop-
by those who use the concept. The quantum-mechanical situation just out-
account but can never apply simultaneously, since they entail mutually
made a virtue out of necessity and used this fact as a way to resolve these
This, it goes without saying, is not to say that quantum objects or, again,
more accurately, that which compels us to speak in such terms do not exist,
for example, when we are not there to interact with them by means of our
objects" of some sort and of their actions, such as those producing certain
To put it another way, the particular ("strange") architecture of the data or,
one might say in the spirit of quantum information theory, information that
one encounters in quantum physics compels one to infer the existence of cer-
tain (quantum) objects of nonclassical nature; or, again, these data, includ-
be interpreted in this way. In other words, the point here is that such attrib-
objects," this phrase should be read in this general sense of the existence of
something that is beyond the classical level of description and that (in the
and cannot know or conceive of what sort of objects quantum objects may
be, nor can we ultimately, in all rigor, think of them as "objects" (a rela-
physics. This, however, need not mean that nothing exists or that everything
stands still, as Parmenides was perhaps first to surmise, thus helping Plato to
"objective reality" (or the lack thereof) in quantum theory: "We often dis-
cussed his notions of objective reality. I recall that during one walk, Einstein
suddenly stopped, turned to me and asked whether I really believed that the
moon exists only when I look at it."17 Einstein's target is Bohr's argument
nothing "out there," for example, where we see the moon, when there is
nobody to look at it. (There are very few who would believe this, even
among the most committed antirealists.) Instead it tells us that it is far from
certain that, once there is nobody to look at it, it would be the moon or any-
thing we can possible think of, now or ever, say, a collection of elementary
particles. And then, which moon? We might do well to follow, to its perhaps
vow.... O, swear not by the moon, th' inconstant moon, / That monthly
changes in her circled orb, / Lest that thy love prove likewise variable"
(Romeo and Juliet, II.ii.109-11). The moon is never quite there, at least as
anything quite the same, even when we look at it, but it is also not quite up
a little by so doing. To cite Proust's comment: "the trees, the sun and the sky
would not be the same as what we see if they were apprehended by creatures
having eyes differently constituted from ours, or else endowed for that pur-
pose with organs other than eyes which would furnish equivalents of trees
and sky and sun, though not visual ones." Perhaps such "organs" (even if
this denomination could apply) would not furnish even that much, in any
event nothing equivalent. It is, now not surprisingly, at this juncture that
Proust invokes those "ideas that do not reveal [show] themselves but are
none the less efficacious [agissantes]."18 It is also clear that while he says
"ideas" he starts his meditation with the physical world and sees two paral-
lels, and indeed, interactive dynamics, material and phenomenal (in the
usual sense of appearance to the mind), at work. Both are always at work in
to Bohr's atom.
data in question, Bohr's nonclassical solution of the dilemma and, hence, his
EPR article, Bohr speaks of "the necessity of a final renunciation of the clas-
sical ideal of causality and a radical revision of our attitude towards the
over, the suspension of realism entails not only the impossibility of describ-
Quantum Mechanics, Complementarity, and Nonclassical Thought * 43
ing physical properties of quantum objects and processes but also the
other reality) one can hardly speak of causality. The point was well
accounting for quantum data), here in question, and both appear to have
could avoid it. "If a classical [physical] state does not exist, it can hardly
quotation marks become obligatory under these conditions, since the terms,
however, is again a crucial qualifier here since both concepts, that of reality
cal work and other human activities, or in everyday life. They are also nec-
essary for the functioning of nonclassical theories, since the latter must
work with the classical (knowable) effects and, hence, rely on classical ways
ultimately unknowable.
account for the overall configuration of the effects in question and, in the
such effects. This capacity ensures the status of quantum mechanics as, in
turn and classical physics. We need nonclassical theories for these purposes
rather than only in order to introduce the irreducibly unknowable that ulti-
mately defines them. Indeed, the latter conception itself emerges by virtue of
the rigor of theoretical knowledge, and even "provides room for new phys-
ical laws [here the laws of quantum mechanics], the coexistence of which
might at first sight appear irreconcilable with the basic principles of science"
(QTM, 146). This argument enables us to read the two statements, which
respectively, open and close Bohr's argument in his reply to EPR, together:
radical revision of our attitude towards the problem of physical reality ...
provides room for new physical laws, the coexistence of which might at first
sight appear irreconcilable with the basic principles of science" (QTM, 146,
accordance with the disciplinary rules of physics, above all, for the discov-
to Nonclassical Epistemology
theories involved. It was defined in terms of physics itself (that is, in terms
of the constitution of the data and the theories accounting for these data) as
the difference between classical, often also called Newtonian, physics and
the new physics, which emerged around 1900 and departed from classical
debate, extending to our time and sometimes joining the debate concerning
referred at the time, first, to this theory or, as it became known later, special
1915. The latter extended relativity theory to spaces whose geometry was
well. Second, the new physics referred to quantum theory, eventually lead-
The classical physical theories just mentioned, however, are also classical
in the epistemological sense of this study and hence are realist, as well as
causal (which, as I said, in general need not be the case classically). Or, more
accurately, they may be and (in contrast to quantum theory) commonly are
interpreted as such. At least this can be, and commonly is, done at the level
quantities abstracted from other properties that material bodies and their
behavior possess. One can see such idealized configurations of bodies and
Schrodinger, we may call these models "classical models" and the ideal itself
broader sense of this study. To cite Schrodinger in his "cut paradox" article:
In the second half of the previous [nineteenth] century there arose, from the
great progress in kinetic theory of gases and in the mechanical theory of heat,
an ideal of the exact description of nature that stands out as the reward of
It is worth interrupting the quotation for the moment to make the fol-
lowing observation. Schrodinger is well aware of the ideal nature of the clas-
the ultimate level of description in view of quantum theory (ten years old
and well accepted by the time), although Einstein's criticism renewed his
classical hopes (QTM, 153-54). However, as these sentences and his argu-
ment as a whole make clear, his appreciation of the ideal itself was as pro-
found as were his concerns for the possible loss of this ideal in view of quan-
Of natural objects, whose observed behavior one might treat, one sets up a
exactly, much more exactly than any experience, considering its limited
one side and two adjoining angles, as determining parts, also determine the
third angle, the other two sides, the three altitudes, the radius of the inscribed
circle, etc. Yet, the representation differs intrinsically from a geometric figure
in this important respect, that also in time as fourth dimension [this repre-
changes with time, that can assume different states; and if a state becomes
known in the necessary number of determining parts, then not only are all
other parts also given for the moment (as illustrated for the triangle above),
but likewise all parts, the complete state, for any given later time: just as the
part of the inner law of the concept that it should change in a given manner,
that it should continuously run through a given sequence of states, each one
of which it reaches at a fully determined time. That is its nature, that is the
Schrodinger also calls it, is both causal and realist (in view of the determi-
nation of all parts from a sufficient given set of parts). That is, the systems
theories and models. Some among such theories and models may be suc-
cessful (or successfully refined), while others may fail and may be aban-
doned. Newtonian mechanics is the primary and arguably still most suc-
often depend on it as well). Consider it, for example, as applied to the falling
bodies or (physically the same thing) the motion of planets around the sun.
We can idealize such objects and behavior, as massive material points (cor-
ization. In this case the "determining parts" would be such variables and
tion at any given point and the equations of classical mechanics (which
the fact that once more than two bodies are involved (say, the sun, the earth,
and the moon), the behavior of the system becomes, in general, chaotic,
causal, and, as I will explain presently, chaos theory is classical in the pres-
theory ultimately fails even on this scale and needs to be replaced by Ein-
stein's general relativity, since the latter is classical within these limits. The
situation becomes more complicated once very massive bodies, such as neu-
tron stars or black holes, enter the picture (even when one leaves aside the
sical physics fails both on very large scales and on very small ones when
tions, which include both models and interpretations, since the status of
relationships between this model and the experimental data or the behavior
of the natural objects in relation to the model. Thus, in the case of planetary
approximation, insofar as the realist and causal models describe and predict
the situation may be more complex since the relationships of such models to
the behavior of the natural objects involved are far less direct.
malism, which would relate, say, a given set of equations to the motion of
idealized physical objects. (This was to become a problem in the case of the
type of interpretation may precede and give rise to an idealized model rather
48 * The Knowable and the Unknowable
though they have helped mathematics to develop and refine some of its con-
by Galileo or (these are not quite the same) Newton and his successors,
especially Joseph Louis Lagrange and Sir Rowan Hamilton, who gave clas-
sical mechanics its modern form. Their work laid the mathematical and, in
the Hamiltonian type. As will be seen, however, the variables to which these
equations apply are mathematically very different from those used in classi-
cal physics.
relativity may be, and may ultimately need to be, seen as epistemologically
ily determinism (the predictive capacity of the theory) and causality (the
ing even in the case of the individual quantum processes, such as, say, an
electron's motion around the nucleus of an atom, and even at the level of
the article cited earlier, a classical-like model and a fulfillment of the classi-
cal ideal appeared all but impossible, which, in his view, indicated a neces-
the standard quantum theory lends itself and that are mathematically
equation) did not, and still do not, appear available to a classical and,
quantum theory, from Planck on, prior to and in the immediate wake of
tinue but, to this author, without much success at the very least at the level
Heisenberg and then Bohr made a virtue out of necessity by taking advan-
ics" dealing with the effects of quantum processes upon measuring instru-
objects, or, again, what we infer as such from the data in question in quan-
idealized in this model, since there is nothing the model says about them
feature of the model and the resulting theory, almost a betrayal of the task
as unidealizable by this model, while what they relate to in nature is not ide-
(as the nonclassical unknowable) as part of its model rather than attributing
this conception to whatever actually happens in nature at the level of its ulti-
other words, if one establishes the model just outlined. That is, comple-
the interaction in question) are seen as inaccessible and, hence, not subject
with the formalism of quantum theory, which predicts them, just as classi-
dealing with the properties and behavior of the objects themselves that it
sical macroscopic effects of the interactions between these objects and mea-
suring instruments upon the latter, rather than that of the ultimate objects,
mately, (given that even this is an idealization) even the concept of incon-
ceivability itself.
for this data (such as Bohmian mechanics); and so forth. While I do not
think, for the reasons to be explained later, that such challenges have so far
pretation of the relationships between the model and nature is built into the
nature, the character, of the model itself-a unique and elegant way of
One may also put this a little differently and perhaps more strongly.
other interpretations that hold the same view as to what quantum mechan-
ventable limits upon what we can, in principle, analyze or know or, again,
even conceive of, now or, again, ever. This is why Bohr said that, "in quan-
Bohr's sense], but with a recognition that such an analysis is, in principle,
excluded" (PWNB 2:62; Bohr's emphasis). On the basis of the present read-
mechanics, as just explained. Quantum mechanics may and has been inter-
preted otherwise. It is, again, another question of how effective such alter-
another question, too. I shall return to these questions later in this chapter.
On the point itself made by Bohr here complementarity, thus, goes fur-
opposed to, as in Bohr, strictly the effects of their interaction with measur-
effects in question, and hence the ultimate objects of theory, and, second,
correlatively, to the rules, such as Max Born's rule for using the wave func-
But, while extraordinarily effective, they remain ad hoc and exterior to the
theory; that is, they have neither mathematical nor physical justification
52 * The Knowable and the Unknowable
from within the theory itself, and (this is my point here) in Bohr's interpre-
Newton's), may not be concerned with such justifications either. This, how-
ever, is not quite the same as not being able to offer such a justification at all
from the first principles (which can be done reasonably well in classical
his 1913 theory of the atom, Bohr, too, only saw such a justification as
eventually able to offer such a justification (or at least what was accepted as
such) in the case of Galileo's physics, while in quantum mechanics this does
not appear to have been possible so far, even in interpretations short of non-
I shall now explain the main physical reasons, first, for the epistemologi-
cally classical nature, or, again, the possibility of the epistemologically clas-
sical interpretation of classical physics, and then the physical reasons for
as, ontologically, realist because it can be seen as fully describing all the
behavior. At least, again, such is the case for idealized systems or at the level
The resulting model can then also serve as a model for more complex and
state of the systems it considers (these systems may, again, be idealized and
determined (in the past) by and to determine (in the future) its states at all
know, again, at least in principle and in ideal cases (but also as a good
approximation in most actual cases) its state at any other point. Description
of this kind was, as Bohr observes, found to have a very wide scope, extend-
Quantum Mechanics, Complementarity, and Nonclassical Thought * 53
ing (with due nuances) to (statistical) kinetic theories of gases and electro-
wave theory of the latter, and, then, albeit, again, ultimately more ambiva-
sense. Thus, both classical statistical physics and, differently, chaos theory
(which is, in most of its forms, classical and is sometimes a direct extension
of Newtonian mechanics) are causal, or, again, at least they are assumed to
terms of idealizations and models as well, even if only because the quantum
aspects of the objects considered by such theories indicate the very different
nature of the ultimate reality, or the lack thereof, to which such models
apply.27 These theories, however, are not deterministic even in ideal cases in
view of the great structural complexity of the systems they consider. This
complexity blocks our ability to predict the behavior of such systems, either
exactly or at all, even though we can write equations that describe them and
ization. In general, it does not follow that deterministic theories are realist,
since the actual behavior of a system may not be mapped by our description
of it, even though we can make reasonably good predictions concerning that
behavior. Classical mechanics (or chaos theory) is, however, also realist
describe the behavior of its ultimate objects, such as molecules of a gas. It is,
or models that are approximate in this sense or further to theories that pre-
with, or, one might say, on the model of, classical models. Indeed, the latter
54 * The Knowable and the Unknowable
is our only source of such conceptions, for example, those responsible for
erties ("elements of reality," Einstein liked to call them) and behave in the
ingly, or for the assumption that somebody plays or does not play quantum
dice. Most realist theories and models in physics hitherto available may be
realist and classical theories are obviously one and the same and are treated
here offered) and yet treat the behavior of these systems as, at least in prin-
the classical ideal, as it was seen by both Einstein and, as was discussed,
From this viewpoint, insofar as it was not describing the physical space-
at the level of idealized models, quantum mechanics was almost not physics
for Einstein, at least not the way he practiced it with such extraordinary
none of the features of classical theories just sketched is, by definition, pos-
the only true discontinuity in quantum physics), that defines the difference
tum objects as something that is beyond any physical space-time (or indeed
any other) description; and second, between this idealization and whatever
that it was the first discontinuity that was especially crucial and vexing to
Einstein and Schrbdinger, and many of their followers, who would hardly
find the word idealization suitable here. Einstein spoke of "Jacob's pillow,"
gen, two main centers of new quantum theory, perhaps especially in view of
the almost mysterious (in either sense) predictive power of the theory. As I
shall explain, however, while this power may indeed be seen as mysterious,
insofar as we do not know and cannot even think in terms of any underly-
lined may and, in such cases as Einstein's and Schrbdinger's, did entail con-
will always remain naive in this sense. Perhaps no (mere) refinement of com-
mon experience, of everyday life, which gave rise to the conception of clas-
thinking. As I have indicated earlier, both Bohr and Heisenberg (rightly) see
Galileo on, this refinement reaches far beyond Aristotle, but not even nearly
far enough (which may in fact not be possible classically) for quantum
One might, however, also question why one should expect nature at
these levels to follow the classical ideal, even as an ideal or model, which is
something that emerges at a very different scale and in very different cir-
expect it to play or not play dice, Einstein's famous, if, as we have seen,
rowed from the experience of everyday life. It can only be carried over to
atomic phenomena when due regard is paid to the limitations placed on all
not the best way to handle the situation. In Bohr's post-EPR version, any
56 * The Knowable and the Unknowable
tions or phenomena.
ministic nor realist in any of the senses described earlier, most specifically in
processes. The latter may, at most, be said to be real or, again, may relate to
something that exists (even when we are not there to observe it), but this
in which it exists. This is how we are compelled to idealize these objects and
It is not only that the state of the system at a given point gives us no help
(PWNB 1:65-66).29 More radically, even this state itself cannot, at any
(q) and "momentum" (p). These limits are expressed by the famous formula
measurement (the same type of formula holds for time and energy). The
nature. This point is crucial to Bohr and part of his argument, cited earlier,
both causality and reality, "quantum mechanics provides room for new
physical laws, the coexistence of which might at first sight appear irrecon-
cilable with the basic principles of science." As could be expected, the two
bols whereby the product of the magnitudes involved depends on the order
of multiplication (i.e., PQ does not equal QP), even though this formalism
pears in the process as well, for the very simple reason that in the quantum-
mechanical and classical formalism. It can also be shown that in order for
that h equals zero, the symbolic quantities P and Q (usually still called
or schemes (as we have seen, there are several mathematically, although not
old quantum theory" but successfully used by him and others, especially,
of Complementarity
stein.32 Bohr's 1935 reply to EPR was arguably the most decisive work in
developing Bohr's ideas in the form considered here, although there is still
earlier evidence of this shift, around 1930, under the impact of his earlier
tarity" in 1958 may be the most definitive. To the present reader, however,
reinterprets, his earlier works, such as the Como lecture of 1927. The dif-
ferences between different post-EPR arguments are not negligible but are
not even nearly as significant as those between them and those prior to the
EPR arguments or, in any way, prior to Einstein's critique from 1927 on.
ogy and its nonclassical nature, although they are not always addressed or
carefully sorted out by Bohr's readers, who are not always helped by Bohr
either, since he tends to minimize these differences and to reread his earlier
in his 1935 reply to EPR, Bohr says: "I shall therefore be glad to use this
vious occasions" (QTM, 145). This is not altogether wrong, but he is not
quite right either. While certain key features, specifically the mutual exclu-
pear rather quickly from Bohr's writings, and for good reasons. This partic-
becomes quite clear. And there is none in Bohr's later writings. The comple-
tarity on the part of the physics community, although the latter problem
will speak of "a final renunciation of the classical ideal of causality" (QTM,
ideal of causality" (PWNB 2:41; emphasis added). The Como lecture also
sical physics" (PWNB 1:87), but, besides a significant nuance in the formu-
quite yet fully worked out so as to bring it to a nonclassical level.34 One can
even prior to the EPR argument, but clearly under the impact of Einstein's
6:361-70). This lecture may be seen as "a previous occasion" where "a gen-
reply to EPR, was indeed "indicated" by Bohr there, but not in any of his
Bohr's Como approach contained great benefits and ideas as well, begin-
ning with complementarity but far from ending there, and the Como lecture
tremendous. Perhaps most significant, even though not fully worked out, are
the irreducible role of measuring instruments and its radical (ultimately non-
free space as well as isolated material particles are abstractions, their proper-
ties on the quantum theory being definable and observable only through
1:57). This is just a few inches away from his ultimately nonclassical view,
themselves, which leads, in the lecture itself and especially later on, to the
his argument in the Como lecture. The latter presents its argument in terms
in the later works. By the time of "Discussion with Einstein," Bohr argues
for our orientation and adjustment to ordinary experience [on which classi-
cal physics is based]. It is in this respect that quantum theory has called for
"effects" (a term that itself became essential for Bohr), rather than quantum
objects themselves? To some degree, the preceding analysis explains the sec-
ond concept. The analysis to follow will explain the first and will link both
together. It is, again, the Como lecture that introduces the subject of and the
quantum theory, it seems that its essence . . . may be expressed in the so-
Quantum Mechanics, Complementarity, and Nonclassical Thought * 61
opening move in the development of one of Bohr's most radical and innov-
ative concepts. It took him twenty years and much further development of
action between quantum objects and measuring instruments upon the latter
ical experiment-which may be argued to contain all of the key physical and
shall discuss the double-slit experiment following Bohr or, in any event,
slit (A); at a sufficient distance from it a second diaphragm with two slits (B
and C), widely separated; and, finally, at a sufficient distance from the second
between the source and the two-slit diaphragm, which does not change the
essence of the situation.) A sufficient number (for a full effect it must be very
emitted from a source, is allowed to pass through both diaphragms and leave
their traces on the screen. Provisionally, I speak for the moment in terms of
Bohr's sense). Indeed, which is the main point here, the attempts to appeal to
tion ultimately lead to problems. Two setups are considered: in the first we
cannot know through which slit each particle passes; in the second we can,
Both slits are open, and no counters are installed (interference pattern)
Source
Source B
Counter
Figure 1.
counters, are made that would allow us to establish through which slit each
particle that hits the screen passes, a wavelike interference pattern will
what kind of setup can be actually realized), this pattern will emerge regard-
less of the distance between slits or the time interval between the emissions
of the particles, say, one hour, an eternity at the quantum scale of events.
Particles will "arrange" themselves in a pattern even though the next emis-
sion occurs after the preceding particle is destroyed after colliding with the
screen. This fact is, of course, remarkable, although routine by the stan-
More accurately, one should speak here of a pattern analogous to the traces
Quantum Mechanics, Complementarity, and Nonclassical Thought * 63
waves on the sand. In other words, if quantum objects were classical waves,
which, however, they are not in this interpretation, they would leave this
Bohr's and most standard interpretations (there are alternative views, espe-
character of the quantum world. This pattern will appear whether we deal
with what would be prior to the advent of quantum physics classically seen
said, on this view, quantum objects must be seen as entities other than either
particles or waves, or indeed anything, since the term and any notion of
could no longer apply to them. It is also worth keeping in mind that we see
on the screen only classically manifest traces of quantum objects. The latter
The outcome of this first setup may appear to imply that each particle
would spread, in a wavelike fashion, into a volume larger than the slit sep-
aration or would somehow divide into two and then relocalize or reunite so
between slits, too, can be very large relative to the "size" of the particles,
tion is more or less as follows. Although having both routes open always
mulates, any given particle passing through the slits should be seen as an
indivisible whole (i.e., the corresponding effects upon the measuring devices
confirm this-at the cost of losing the interference pattern, which circum-
arbitrarily long (light years), after particles passed the slits, while we can
If, however, there are counters or other devices that would allow us to
check through which slit particles pass (indeed even merely setting up the
through which slit particles pass. Thus, ironically (such ironies are charac-
of a pattern and, hence, with a higher rather than a lower degree of order, as
would be the case in, say, classical statistical physics. (Chaos theory is yet
something else.) The situation-that is, the fact that the knowledge or even
the possibility in principle of knowing through which slit the particles pass
mechanics appears to be irreducible. I shall return to this point later. For the
A, two intermediate states B and C, and then a final state E. (The latter is
analogous to the state of a "particle" at the point of its interaction with the
screen in the double-slit experiment.) First, we arrange to block the path via
state C but leave the path via state B open. (In this case we do not attempt
to install any additional devices to check directly whether the object has in
fact passed through state B.) In a large number (say, again, one million) of
trials we record the number of particles reaching state E. Then we repeat the
same number of runs of the experiment, this time blocking the path via B
and leaving the path via C open. Finally we repeat the experiment again
with the same number of runs, now with both paths open. In Leggett's
'via B' or 'via C.' " The probabilities of the outcomes of individual experi-
gence of the interference pattern when both slits are open in the double-slit
on its way to the screen. If we do, the above probability sum law (based on
themselves), would not be obeyed and the conflict with the interference pat-
2:46-47; QTM, 146-47). This is also why the ways of counting probabili-
One may also put it as follows. We must take into account the possibility of
a particle passing through both states B and C (and through both slits in the
double-slit experiments), when both are open to it, in calculating the prob-
in Bohr's interpretation, assume either that such an event in space and time
physically occurs for any single particle, anymore that we can assume that
one can walk into a building simultaneously through two doors, when these
In the light of this result, it is difficult to avoid the conclusion that each
and C. (The only obvious alternative would be to postulate that the ensem-
ble as a whole possesses properties in this respect that are not possessed by its
always find a definite result, i.e., each particular microsystem is found to have
longer see any interference between the two processes.) . . . (Clearly, we can
read off the result of the measurement only when it has been amplified to a
macroscopic [classical] level, e.g. in the form of a pointer position [of mea-
The first possibility corresponds to more familiar questions, such as, how
do particles know that both slits are open, or conversely that counters are
66 * The Knowable and the Unknowable
haps more general question, to be discussed later, of how order (such as that
found in the EPR experiment) arises from the absolute randomness of indi-
vidual quantum events. Indeed, in this case it may even happen without any
establishing such connections, since, as I said, the next particle can be emit-
ted long after the previous one has already been destroyed.
reason for this reaction is that, if one speaks in terms of particles themselves
the slits. Or, even more radically, individually, or (which would be even
cles appear somehow to "know" whether both slits are or are not open, or
whether there are or are not counting devices installed. In short, any attempt
tivity. The latter alternative was first proposed by Einstein and is legitimate
and, arguably, the most rational under this attribution. Or-this is the EPR
tion" and "momentum" (these variables of course define the very concept of
than to admit that, in this field of experience, we are [rather than with prop-
erties of quantum objects] dealing with individual phenomena and that our
objects and measuring instruments upon the latter (PWNB 2:51). The term
trivial application of the standard logical deduction under the nontrivial and
takes place and works only if Bohr's interpretation itself applies.) As will be
seen in the next section, a similar argument allows one to ascertain the local-
specifically denies that any departure from ordinary logic (for example, mul-
even when one speaks of single such attributes rather than, as would be
various uncertainty relations, and even at the time when the measurement
takes place. This stronger claim, or, again, the corresponding interpretation,
appears to have emerged in Bohr's arguments in the wake of the EPR arti-
cle, although, as I said, the Como lecture says nearly as much as well. It is
clear, however, that this is what Bohr has in mind, since in speaking of this
uncertainty relations. As he says in his reply to EPR, "In fact to measure the
position of one of the particles can mean nothing else than to establish a cor-
relation between its behavior and some instrument . . . which defines the
68 * The Knowable and the Unknowable
ior," not its position! This is why the term "phenomena" in the passage
ment makes no sense otherwise. The sentence also compels one to argue that
in Bohr's scheme this "behavior" itself, too, cannot be described but can
certain other parts, where the actual interaction with quantum objects takes
that Bohr speaks here in terms of particles. It is clear, however, that he does
so only provisionally and primarily in arguing with EPR, who use these
apply. What is referred to by particles is, in all rigor, only a class of properly
call interacting instruments. As Bohr says a bit earlier in his discussion of the
double-slit experiment:
that we are presented with a choice of either tracing the path of a particle or
cles hitting the screen], which allows us to escape from a paradoxical neces-
ments . . . and are just faced with the impossibility, in the analysis of quan-
behavior of atomic objects and their interaction with the measuring instru-
ments which serve to define the conditions under which the phenomena
We now see why Bohr needs his concept of phenomena as defined by the
mechanical context of its appearance, each mark on the screen in the dou-
ble-slit experiment would be perceived in the same way or as the same phe-
cal mechanics such conditions would of course be the same, assuming that
we deal with particles (in the case of waves these conditions would be the
same as well but there would be no pointlike traces, and different equations,
two different sets of conditions of the experiment. One of these sets of con-
to a small spot on the photographic plate, and the distribution of these spots
2:45-46; emphasis added). Each single spot, however, must be, again, seen
which the event occurs. Two different overall patterns, "interference" and
"no interference," pertain, thus, to two (very large) sets of different individ-
nomena is essential for Bohr's meaning and the consistency of his argumen-
objects themselves, are involved. Second, and perhaps most crucially, in our
case of classical physics. This is not an uncommon error (at least as Bohr's
on which point I shall comment in the next section in the context of the EPR
argument and counterfactual logic. The latter, however, disappear once this
lowed. Throughout his arguments with Einstein, Bohr stresses in such situ-
EPR type, "we must realize that . . . we are not dealing with a single
ity of a quantum phenomenon or effect in Bohr), and, in this sense, each phe-
Of course, not all such phenomena are "complementary" in the same sense
sense, rather than two aspects, actual or even possible, of the same entity
but must be regarded as complementary in the sense that only the totality of
the [observable] phenomena [produces the data that] exhausts the possible
Thus, on the one hand, quantum objects are (or, again, are idealized as)
ble from their interaction with measuring instruments and the effects this
its effects) and is outside any knowledge or conception, continuity and dis-
continuity among them (PWNB 2:39, 62). Indeed, as I have indicated at the
outset, one cannot speak here of a single efficacity of any two effects but
Quantum Mechanics, Complementarity, and Nonclassical Thought * 71
between quantum objects and measuring instruments. In this sense, there is,
effects and, it follows, manifesting itself only through these effects. These
tion of the quantum "atomicity" (in the original Greek sense of being indi-
visible any further) of matter, as against the view prompted by Planck's dis-
covery (thus the view still persists). By contrast, quantum objects themselves
than any other features, properties, images, and so forth, "wavelike" for
ments and quantum objects and to allow one to isolate the latter or their
and time. As Bohr writes, "the individuality of the typical quantum effects
finds its proper expression in the circumstance that any attempt of subdi-
the same nature, and, hence, it will always retain or reinstate complemen-
wholeness).
ingly, every event in question in quantum physics is also individual and sin-
gular in the same sense and is in itself not predictable or, more generally, not
event. This explains why Bohr uses the term in this context, in preference to
uality. Thus, along with the quantum atomicity as indivisibility, the quantum
from discontinuity at the level of quantum objects, where Planck "found" it,
nomena in Bohr's sense. Bohr directly argues that quantum physics reflects
early as 1932 (PWNB 2:6) and is made throughout his later writings, twice
a feature of atomicity in the laws of nature going far beyond the old doctrine
ing spirit the novel features of atomicity which pointed beyond the frame-
work of classical physics" (PWNB 2:33, 71; 3:2). Classical physics is based
Lucretius. This physics would describe the behavior of its objects as such
entities that can be rigorously described by quantum theory are certain indi-
suring instruments. From this perspective, the only "atoms" that can be rig-
are inseparable from quantum objects and make the latter irreducibly inac-
which one might best apply the term "Bohr's atom." This phrase is usually
was Bohr's first step on the long road to his nonclassical "atoms" just con-
sidered. Obviously, these "atoms" are not physically atomic, since they
stituents. They are atomic in the sense that no different form of access to
quantum objects is possible, regardless of how far we can divide and sub-
"(en)closed phenomena."
applied with caution, the analogy itself is not out of place, insofar as the
a black hole) found in the interior of a black hole, can never be seen. Black
microscopic black holes. This idea is in turn partly due to Hawking, who
appears, also of noncausality and arealism) into physics.42 This limit upon
how far our knowledge can penetrate into black holes in no way appears to
following Bohr, I have stressed from the outset, should we be alarmed by the
and Probability
The preceding analysis makes clear why, far from being restricted to the
wave and particle effects (which in a way virtually disappear into other
although in practice various arrangements are often convenient for the study
simply supplement each other and can be combined into a consistent picture
knowledge about the object. Far from restricting our efforts to put questions
characterizes the answers we can receive by such inquiry, whenever the inter-
action between the measuring instruments and the objects forms an integral
This view makes Bohr's very use of the term "complementarity" peculiar,
or derivable from a single entity of any kind, which gives a very peculiar
Quantum Mechanics, Complementarity, and Nonclassical Thought * 75
aspects of the situation seem inconceivable, this is because they in fact are, or
dependent).
tum" (p): AqAp h (h is, again, Planck's constant, and A is the precision of
nation of such variables, even each by itself, that becomes the next question.
physics, while other parts are subject to quantum behavior, which enables
Indeed, as we have seen, the situation is more complicated even in the case
between the behavior of atomic [i.e., quantum] objects and the interaction
which the phenomena appear" (PWNB 2:39; Bohr's emphasis). For, "under
2:40), ultimately even if one talks of a single such attribute under all condi-
tions between quantum objects and measuring instruments upon the latter,
of the quantum object involved. In this sense, it would not be quite accurate
146). The latter might have been the case for Bohr, who subsequently refined
most such formulations in any event. In this view, uncertainty relations have
as much, if not more, to do with reality (or the lack thereof) as with deter-
minism and causality (or the lack thereof) in quantum physics, and, as will
be seen presently, they have almost nothing to do with the accuracy of mea-
their interactions with quantum objects is of the same type as the data
pensate for (so as to avoid its significance for our theoretical description) the
define either one or the other variable of that type, say a change in the
tions between quantum objects and measuring instruments upon the latter
can, however, be described in the realist manner, since any single variable of
unity. (The latter fact led Einstein to think that something is amiss in quan-
tum theory, that it is perhaps incomplete. Not so, Bohr countered!) Such
variables can only now be seen as defining either the positional coordinates
of the point registered in some part of the measuring instruments or, con-
versely, a change in the momentum of another such part, under the impact
of its interaction with the object under investigation. They, even each by
tion laws" (this case is a bit more subtle) occurs at the level of measuring
This view gives a new and more radical meaning to Bohr's argument con-
tion and the measuring instruments which serve to define, in classical terms,
the conditions under which the phenomena appear" (PWNB 2:50). This
ena" (QTM, 150). For, while "in classical physics the distinction between
object and measuring agencies does not entail any difference in the charac-
does (QTM, 150; PWNB 2:50; 3:3). These and related statements by Bohr
do not mean (as it might appear) that, while parts of measuring instruments
78 * The Knowable and the Unknowable
are described by means of classical physics, the behavior (in space and
formalism. Bohr obviously says the former, but he clearly does not say,
here or elsewhere, and does not mean the latter. Instead, the difference
between the two descriptions is this: classical theory describes the classical
interpretation.
As Bohr points out, "it is true that the place within each measuring pro-
this interpretation rather than a problem, as it would be, and often has been,
space and time behavior of quantum objects themselves. At this point of his
argument and, according to Bohr, "perhaps more than any other feature of
spondence with classical mechanics" (QTM, 145, Note). For "by securing
its proper correspondence with the classical theory the [transformation] the-
mechanical description, connected with a change of the place where the dis-
(QTM, 150).
What makes this last point (which, thus, also conveys the deeper episte-
quantum objects are always on the other side of the "cut" and indeed may
object. The correlation is, according to Bohr, real, or rigorous, while the
identification can only be, in his words, "symbolic," insofar as, in view of
ously attribute such (or, again, any) properties to quantum objects or their
although they cannot be isolated from their efficacity by even conceiving, let
quantum objects and processes can never be isolated, either materially (from
the quantum level but only at the level of "measuring," or, we may say,
with quantum objects. About the latter we obtain only certain irreducibly
from the Como lecture on, Bohr denies that the quantum-mechanical for-
tion[s], we are dealing with the formalism which defies unambiguous expres-
(or Einstein) was, with the physical meaning of a theory as referring to the
speak of the physical reality in terms of such objects, of the physical mean-
ing (in space and time) of such mathematical objects and their properties?
Dirac, whose work may be seen as especially responsible for the view that
clearly denies that the key mathematical operations involved can have any
closer (which is not to say identical) to Bohr's than it may appear.46 In other
tion or particle motion. As will be seen in the next chapter, they may not
lier, the practice (more or less standard since Dirac's and then Von Neu-
appears to conform to this view more readily and may be seen as shaping
Bohr's view all along. His "matrices" and the formal (Hamiltonian) equa-
tions, retained from classical physics but now applied to these matrices, relate
the term "kinematics" may, again, be misleading here, since it suggests a rep-
matrices do not represent motion or, for that matter, anything physical at the
quantum level. They only enable statistical predictions concerning the out-
of the instruments used under the impact of their interactions with quantum
tation, as compared with both classical physics and with a possible classical
principle. This fact allows us both to explain and indeed to picture (visual-
ize) the process and predict the state of the system (it may consist of many
particles) at any point, or know it at any point in the past, once we know it
actual prediction but does not change the epistemology. (B) In the classical-
tum theory, such as Bohm's mechanics, which retains the uncertainty rela-
tions, we can, given the uncertainty relations, only measure either the
sured simultaneously with any given precision (within the capacity of our
tem under consideration and to track and predict the system's behavior at
any point. Both p and q can be seen pertaining to the system itself and as
Measurement:
pandq pandq
System S
the state of the system and tracking or predicting its evolution, but both
the state of the system at any given point and its evolution are presupposed
on the model of (A). The causal evolution of the system is defined either
Measurement:
System S
at any point and, hence, considering of the situation in terms of either (A)
tation marks around systems "S." The punctured lines indicate the inac-
Measurement:
"Black holes"
p . p.
Figure 2.
Quantum Mechanics, Complementarity, and Nonclassical Thought * 83
ics, however, while we can measure either the position or the momentum of
the effect defining each such phenomenon is both itself each time unique and
quantum objects themselves in the way classical mechanics does for classi-
cal objects. According to Bohr, from the outset of quantum physics in the
2:34). The reasons for this argumentation, from the perspective of Bohr's
full analysis would require technical considerations beyond the scope of this
study).
As we have seen, the exact (within the capacity of our instruments) pre-
ferent quantum objects (again, to the degree that we can speak of them as
through a slit, and hitting (or not hitting the screen) in the double-slit exper-
be the same even if we repeat the measurement of the same variable. For, to
begin with, a repetition of any given setup can never be strictly identical,
even in principle, as they are in classical physics, since the kind of control of,
subject to uncertainty relations. According to Bohr: "The very fact that rep-
least two different objects are always necessary for anything to be predicted
along the lines of classical physics (a fact that becomes crucial in the EPR
argument and Bohr's reply). In fact, and in part correlatively, a large num-
ber of repeated experiments are necessary, which makes the statistical char-
the concept of the individual physical event. The individuality of such events
offers us no laws that would enable us to predict with certainty the outcome
statistical physics, the laws of quantum mechanics also rigorously allow for
since, as I have already noted, one can no longer see the situation in terms
Quantum Mechanics, Complementarity, and Nonclassical Thought * 85
know why certain probability rules, say, those of Planck's law or Max
ities from the wave function, in fact apply, but they do. The fact that we can
these effects.
From this viewpoint, one can indeed hardly speak of anyone playing dice,
have seen, Einstein indeed was willing to accept after all, perhaps for this rea-
son. But that "[the Lord] should gamble according to definite rules, that
[was] beyond him," he said in 1953, one year before his death (QTM, 8), one
absolute lawlessness of each of them. That is, these rules manage to handle
this mysterious combination, which nature's hand deals us, if indeed we can
playing dice, although it seems to suit the situation rather better. In Bohr's
words: "It is most important to realize that the recourse to probability laws
with intricacies of this kind, but with the inability of the classical frame of
ing for the average behavior of a large number of atomic systems" (PWNB
or, in Bohr's interpretation, can possibly be, given the data in question in
quantum theory. For that matter, as we have seen here, it redefines every
analysis of, in the language of Bohr's reply to EPR, "this entirely new situa-
can such an analysis appeal to the ultimate efficacity of the effects in ques-
tion in, and accounted for (in terms of statistical prediction) by, quantum
of Bohr's statement that "in quantum mechanics, we are not dealing with an
Einstein said (in 1936) in the Schilpp volume, Albert Einstein: Philoso-
pher-Scientist, that "to believe this [i.e., that quantum theory offers an
without contradiction," and ultimately he thought that Bohr did most jus-
tarity eluded him. His reply also contains an exposition of his own views on
the subject. However, he saw the way quantum mechanics works "'as so
very contrary to [his] scientific instinct that [he could not] forgo the search
point, Einstein saw the situation in terms of the alternative for quantum the-
ory as that between being "complete and nonlocal" versus "local but
Bohr's argument, as he says, "translating [it] into his own way of putting
it," a translation that may in fact be precisely impossible, without losing the
terms of elements of reality.48 Thus, he was searching for a local and com-
plete conception of quantum physics. The latter, I argue here, becomes pos-
sible once one accepts Bohr's complementarity and its nonclassical episte-
way Einstein would envision (QTM, 150). From this perspective a counter-
aiming to show that in quantum mechanics, we are not dealing with an arbi-
tation as compatible with locality, which is, again, not altogether clear
such model nor such application may be possible and indeed did not appear
tion, even if not ultimately unavoidable, and his interpretation reflects the
possibility of this claim. Indeed, from this viewpoint, it is not even clear in
stein's position on this was complex), to "physical reality" has never been
Bohr did not think so, which made his interpretation at least sufficient, if
not inevitable.50 No local alternatives were on the horizon at the time, nor
form of order, pose perhaps most sharply one of the greatest enigmas of
nature at the level of its ultimate constituents and their effects upon what
appear to us as phenomena (in either sense). How does order arise from the
ment carries this question with it and, as will be seen presently, is indeed an
mechanics or of the quantum data itself, for one need not use quantum
which, I also argue, is local. The reasons for this and for my argument in this
theory) but are not simply, short of an interpretation, such as Bohr's, made
related results, any classical-like theory (such as Bohm's) based on the attri-
quence of the mathematical formalism. Bell's theorem tells us that any clas-
ble with the data of quantum mechanics, Bohr at the very least (which is
ory and a nonlocal but complete theory under the epistemologically classi-
The latter refined the EPR argument by more sharply focusing it on the pos-
ory. This alternative is suggested in the EPR article as well (albeit in a some-
paper (1935), cited earlier; David Bohm's reformulation of the EPR type of
experiments in terms of the so-called spin of particles (this does not change
theories; John S. Bell's theorem and his discontent with quantum mechan-
related questions, are all well-known developments and parts of the debate
in question.51
and minimal, criterion of physical reality: "If, without in any way disturb-
90 * The Knowable and the Unknowable
ing 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 phys-
not "interfering" with this system, since in his interpretation there is no clas-
that are then disturbed in the process.) This can indeed be done in quantum
mechanics for (this remains crucial) a single variable in certain cases, such as
ments on other systems (in the EPR argument, another particle), which have
investigation. For the sake of economy, I, again, speak for the moment of
this measurement, and hence without interfering with the object and with-
ment (PWNB 2:57). In fact, the particle and those parts of the measuring
the EPR situation, which involves two particles, rather than a particle and a
of the case, as it sometimes does for Bohr. In any event, predictions (limited
Quantum Mechanics, Complementarity, and Nonclassical Thought * 91
been in an interaction with the first particle but, at the time of measurement,
or between any measuring apparatus and the second particle. This circum-
stance led some, beginning with Einstein, to conclude that there are some
action at a distance." Bohr did not think that such connections are implied
I shall not consider here EPR's subtle argument and Bohr's equally subtle
and it perhaps cannot be done within the limits of a single treatment. The key
point is this. If one accepts that the EPR criterion of reality applies in quan-
more accurately and more crucially (this is in fact or in effect what EPR
physics rigorously disallows even the minimal form of realism entailed by the
EPR criterion. This is what Bohr argues. Or, again, at least he argues that one
poses here, since it makes the nonclassical view of quantum theory at least
It may be argued that Bohr's was at least the first such interpretation.
nologies, as considered here. Indeed, EPR argue that one can ascertain a
their paper, but is not developed in their article. This shift from the impos-
bility of ascribing even a single variable at any point to the second (or indeed
the first) particle of the EPR experiment is what changes the argument from
cal." While Bohr's argument in his reply is directed primarily against the
point clear in his reply, and he states it (at least) three times in "Discussion
with Einstein" (PWNB 2:39-40, 52, 61). This separation is impossible even
lier. Hence, such measurements would not involve this object itself at the
These are the circumstances that Bohr has in mind when he says that
is able to argue that "a criterion like that proposed by [EPR] contains ... an
are here concerned" (QTM, 146). This is the same ambiguity. For, in view
Ultimately we cannot do so even for a single such property, let alone both
complementary ones (as EPR, again, do in their argument), or, at the limit,
Bohr agrees that quantum mechanics allows for and enables such ("at-a-
(Bohr shuns this language) "states" exist, but this only means (a) that par-
need not follow, unless of course it is independently derived from the for-
malism itself, which, as I said, does not appear to be the case thus far.53 It
would follow that the efficacities of correlations are not available to quan-
tum-mechanical or any explanation. But then neither are most typical quan-
tum effects, many of which are the effects of correlations. In other words,
Bohr argues that a local but nonclassical (as opposed to nonlocal and clas-
sible. This possibility was missed by EPR's argument, although perhaps not
plementarity unacceptable.
It has been argued that the question of locality is irrelevant for Bohr and,
fact nonlocal.54 It would be difficult to agree with either view, especially the
mechanics and relativity is manifest in his reply to EPR and most of his sub-
sequent (and preceding) writings on the subject, including virtually all arti-
cles here cited. He states in a key (but usually ignored) note at the end of his
compatibility between the argument outlined in the present article [i.e., his
reply to EPR] and all exigencies of relativity theory" (QTM, 150n). Indeed,
took place in 1930, well before the EPR article, although, as Bohr observed,
(PWNB 2:57).
For Bohr, relativity, and hence locality, was a given, an "axiom" or "a
given, corresponding to the fact that on the macrolevel classical physics still
and the uncertainty relations. EPR's or, more properly, Einstein's challenge
was to prove that complementarity is, or may be refined to be, both com-
plete, within its scope, and consistent with relativity. Bohr's ultimate inter-
pretation allows him to see quantum theory and data as fully consistent
brings all three together (involving classical physics at the level of measuring
ble in physical (space-time) terms, since they must now be seen as classically
accessible than in the case of other quantum effects, many of which in fact
earlier.
These are the circumstances that enable Bohr to argue that "the apparent
this interaction "entails the necessity of the final renunciation of the classi-
cal ideal of causality and a radical revision of our attitude towards the prob-
lem of physical reality" (QTM, 145-46).55 Or, again, at least the corre-
one to effectively reply to the EPR argument. For this view makes it possi-
mechanics with relativity, the desideratum equally defining his and Ein-
available? It may or may not be; the question would requre a separate
Quantum Mechanics, Complementarity, and Nonclassical Thought * 95
invade some among the proposed alternatives. Are more radically different
only consistent with it? It is difficult to say, and it is the latter, quite enough
argue primarily for the latter, although he did not appear to think (rightly)
that there were other fully developed such interpretations at the time of his
writing (QTM, 145). This view of Bohr's argument is not inconsistent with
his other claims, such as that "in quantum mechanics, we are not dealing
enabling one to view quantum mechanics as a fully consistent and, within its
degree. Nor does it follow that it is nonlocal, since, on the one hand, the the-
ory itself is consistent with relativity. On the other hand, all available argu-
does not make (which still appears to be the case), beginning with and per-
which has been a major part of most arguments concerning quantum non-
all) counterfactual arguments, that is, arguments based on what could, but
we can only argue from within one among the possible complementary
arrangement at hand, say, when both slits are open and no counters are
96 * The Knowable and the Unknowable
installed, even when we only have a single spot on the screen. Our exami-
one that would allow us to determine through which slit particles pass. We
what could, but actually did not, happen to the quantum objects in the alter-
of any EPR type of experiment, and hence two actual physical situations,
are always required in order to confirm the EPR alternative predictions. Just
which they would be effects. Hence, EPR's and Einstein's other nonlocality
mechanical data. (As I have indicated, Bell's theorem itself does not prove
knew, and as both Arthur Fine and David Mermin have emphasized.56
refer to the latter. Or, again, they rely on the presupposition that both com-
ics is local.57 Einstein to some degree realized this point, although he con-
sidered this way out of the dilemma "too cheap": "The interpretation of the
measurement carried out in one part of space determines the kind of expec-
tation for a measurement carried out later in another part of space (coupling
Quantum Mechanics, Complementarity, and Nonclassical Thought * 97
parts of systems far apart in space)." On the other hand, as Mermin ele-
gantly shows, if one uses counterfactual reasoning, one cannot establish that
reasoning.58
pointed out that the applicability of classical physics at this level is difficult
main point at the moment is that Bohr's interpretation allows him to han-
ity, as Einstein did later, rather than only incompleteness. Once relativ-
tive. His argument is that this interpretation is possible, which also entails
that Bohr's reply to EPR addresses the EPR argument as such, that is, as an
sometimes done.
forth), with the local measuring system associated with the first particle.
That, however, is quite different from and indeed incompatible with any
suring system of the same type may be introduced for the second particle
performed), correlated with the one associated with the first particle, once
this second system is being in turn interfered with, for example, in order to
measure the value of a given variable for the second particle. Such correla-
tions are involved in Bell's theorem and its refinements and the discussions
(which are particularly striking versions of the EPR type of experiments and
refers to on this occasion, even if what he says is consistent with these "cor-
making in this form.62 Nor, of course, could Bohr have had in mind more
but his argument can, I think, be adjusted to these cases as well. My point is
that quantum entanglement and locality are both maintained and can be
indeed we cannot know, or even conceive of it, since any further analysis
able and, hence, the inaccessibility of "the quantum world" itself (to any
word) allow one to rigorously maintain both the consistency and locality of
tum theory appears to, and in Bohr's interpretation does, prohibit in prin-
ciple what it does not deliver in practice. But it delivers plenty of what can
tion in its technical sense, can be obtained and processed, for example,
this impossibility often enables knowledge and conceptions that would not
EPR and his nonclassical epistemology of quantum mechanics for his views
of the basic principles of science and, hence, for the disciplinary practice of
the argument of [EPR] does not justify their conclusion that quantum
which provides room for new physical laws [i.e., the laws of quantum
with the basic principles of science [but is ultimately not]. It is just this
Bohr, intriguingly, omits "reality" here. This omission may have been
sense, that is, an independent physical reality, defined by postulating the exis-
tence, on the classical model, of physical properties of objects (or, again, con-
ments within which quantum effects, marks left in our measuring devices,
manifest themselves. That, again, does not mean that a certain level of the
constitution of matter does not exist in some form, but only that the attribu-
ticle, may not be possible at that level. Nor, as we have seen, would it follow
(as some contend) that this suspension of the independently attributable par-
ticle identities, such as those of two "particles" in the EPR situation, in fact
entails nonlocality. Two quantum entities (for the lack of a better word)
the EPR case certainly "no question of a mechanical [i.e., physical] distur-
bance of [one] system under investigation" by our interference with the other
argument is, accordingly, not logically wrong. His assumption may well be
objects and measuring instruments, while indeed incompatible with the clas-
theory and what this theory actually unambiguously accounts for and how
it accounts for what it accounts for. The essential ambiguity of the EPR cri-
teria arises precisely from their failure to perceive that a complete and local,
tle and revealing of new aspects and "mysteries" of the quantum it may be.
In particular, it is, again, the failure to see that "under these circumstances
attributes, or even seeing such objects as particles. Or, at the very least, they
between the [quantum] objects and the measuring instruments in the field of
quantum theory." As such it also "provide[s] room for new physical laws
[of quantum mechanics], the coexistence of which might at first sight appear
Bohr's argument is, thus, as follows: were it not for the irreducibility of
(b) there would be no room for the laws of quantum mechanics as phys-
one wants to keep the locality requirement and, hence, compatibility with
any classical theory does for its data, which is what EPR tried, ultimately
sees as basic to science, both in general and insofar as such principles can be
applied in the case of quantum physics. Bohr argues as follows. Rather than
are the logical consistency of a given theory, its correspondence with the
hand (at least, again, under the conditions of locality), the EPR criterion of
102 * The Knowable and the Unknowable
reality). Clearly, one needs to (re)consider what the basic principles of sci-
ence are. This is a crucial point: physics itself, not philosophy, requires this
ural philosophy in our day and age carries weight only if its every detail can
siders nature at the quantum level, the level of its ultimate constituents. The
accordingly.
in accord with the defining aspects of the project and practice of classical
refers in this context (PWNB 3:1). It is true that Einstein and many others
the basic principles of science qua science may, at a certain point, come into
may be with classical physics. What are the basic principles of science,
according to Bohr's view? What would define the science and the discipline
(in either sense) of physics as, to use Galileo's locution, a (modern) "mathe-
matical science of nature"? There are, as I can see it, more or less four such
recalled, itself played a major role in the recent cultural debates, such as, but
again far from exclusively, the Science Wars.66 It may be shown that, in
best. The works involved are not without their own problems, sometimes as
among these principles, but perhaps also the most crucial ones.
(1) The mathematical character of modern physics. By this I mean the fol-
lowing, and I think, with both Galileo and Bohr, that this is what modern
cally representing the ultimate nature or structure of this world.67 The lat-
ferences in this respect between Galileo's and Newton's project and philos-
mentarity but does not eliminate significant differences between them, for
is almost tempted to say that, while his models are classical, his view of how
they relate to nature is almost, even if not quite, nonclassical. In any event,
claim that nature possesses a structure that can ultimately (at least by God)
(2) The principle of consistency. These theories (i.e., arguments and inter-
physics, must exceed their mathematical aspects, must offer logically con-
There are important further nuances to this principle as well, which I must
the practical limits of the functioning of science, for the (sufficiently) unam-
(4) The principle of experimental rigor (based, at least from Galileo on,
and, within their limits, exhaust the experimental data they aim to account
in quantum physics the question of how one interprets its data is as crucial
as, and reciprocal with, that of how one interprets quantum theory. Much
more is to be said on this point as well (even leaving aside the question of
Physical laws would then be seen and defined as physical laws in accor-
case of quantum physics, one, according to Bohr, may need to accept the
must also abandon, at the level of the quantum world, certain other (episte-
tation, becomes rigorously correlated with, if not the condition of, its disci-
plinarity as physics, and thus maintains the continuity with classical physics,
which could otherwise be broken. It is true that one can technically practice
According to this view, the departure (still not quite absolute, given that
classical physics occurs at the level of epistemology, not at that of the char-
acter and the practice of physics as a mathematical science or, since, as Hei-
all, not anymore than any other description. I have stressed throughout that
science it continues to function within its proper limits and is often part of
that predicts these phenomena. This is why Bohr often speaks of "the epis-
The laws of quantum physics are now seen as the laws of nature only in
ever, quantum mechanics does not describe the nature or structure of these
that quantum theory, or, in view of its laws, conceivably any theory, can say
about these constituents as such. Ultimately, as we have seen, not even this
nature" from Galileo on. Quantum mechanics at the very least allows for,
was the first theory to do so. One might argue (Bohr does) that Einstein's
relativity would pose some of these questions already, perhaps, once all the
chips or (we may never have all) more chips are in, even all of these ques-
tions.
may be added, also applies to the relationships between the thought of these
thinkers and modern science in its nonclassical aspects. The list of these
figures is not random, and this type of argument may be more difficult in
some of the cases just mentioned, and it will not be applicable at all in still
other cases. As in Bohr's case, however, some of the most radical epistemo-
(b) one encounters what I called earlier the extreme disciplinary conser-
sense that their discipline (in either sense) in fact requires it. Indeed,
That both of these facts are commonly overlooked largely accounts for
for the misshapen nature of some of the recent debates in which they figure
prominently.
mechanics itself as a way to describe nature. (He did recognize its practical
of some of Bohr's radical (in the present sense) physics as "the highest musi-
cality in the sphere of thought."71 A very good violin player and an admirer
of Haydn, in particular, Einstein might have preferred and had tried to give
this music a more classical shape, or he at least urged others to do so. Nei-
ther he nor others ever succeeded. Bohr, although, by contrast, a bad piano
however, let alone Bach, Mozart, and Beethoven, may be much closer to
as the case may be, undescription of chaos in Paradise Lost in his Creation
Before thir [Satan's, Sin's, and Death's] eyes in sudden view appear
This extraordinary vision is, I would argue, much closer to that of the
Quasi-Mathematics
-RENE DESCARTES
interpretation.
-NIELS BOHR
-LACAN
Introduction
Although I had thought from time to time on the subject of this essay, the
following Paul Gross and Norman Levitt's book Higher Superstition: The
Academic Left and Its Quarrel with Science and physicist Alan Sokal's hoax
article published in the journal Social Text. Sokal and his coauthor, Belgian
physicist Jean Bricmont, had then just published their book, Impostures
intellectuelles, devoted to the misuse or, as the subtitle of the American ver-
by the authors as arguably the most notorious case of this alleged abuse.
Some of Lacan's statements that they cite were bound to attract special
some puzzlement on the part of those readers, including some in the sci-
and, if he was serious, whether there was any way to view this or certain
other statements by Lacan (and other authors attacked in Sokal and Bric-
bers (such as the square root of -1), for example, those he borrows from
topology and mathematical logic. I shall deal directly, however, only with
statement just cited, in "The Subversion of the Subject and the Dialectic of
organ" of Lacan's statement is not the same as the penis. The phrase may
not even quite be seen as referring to the phallus, defined by Lacan in the
same essay as "the image of the penis," but instead as the image of the phal-
lus-the image of the image of the penis. Indeed, whatever the word "penis"
ately indicate to an attentive reader, even one unfamiliar with Lacan's work,
Lacan's "equation." One need not, of course, choose to engage with this
Lacan's "equation."
the Science Wars debates, this epistemology is related to the key epistemo-
theory, but they do not play the same irreducible role in this case, and the
complex numbers can have the same "length" of this type.) Nor can we
both quantum mechanics (in its standard version) and complex numbers
may be, and may need to be (it may never be certain that they must be),
ever, is not the same as saying that the nonclassical epistemology of com-
plex numbers. It is difficult not to think (and both Bohr and Heisenberg,
among others, make some suggestions in this regard) that the nonclassical
and more complex argument, which can only be indicated rather than prop-
erly pursued here. None of my arguments and claims in this study depends
If, however, one must exercise considerable caution in the case of the
indeed when one moves to the relationships between mathematical and sci-
those of Lacan or Derrida, and the nature of their respective arguments and
claims. I have stressed this point from the outset, and I shall try to exercise
deals with such relationships. Boldness, of course, has its appeal and even
spoke; and it is a right balance of boldness and caution, of radical and con-
cult. On the other hand, one does not need exceptional boldness or great
formulation, "we are not dealing here with more or less vague analogies,
but with an investigation of the conditions for the proper use of our con-
hope, the "dream of great interconnections," that is, to limit one's argu-
I shall for the most part bypass the Science Wars debates in this chapter
potential problems in the work of Lacan and other authors under criticism,
the arguments against them by Gross and Levitt, Sokal and Bricmont, and a
few other recent critics in the scientific community cannot be seen as ethi-
specifically to the authors just mentioned and not to the views or opinions
Indeed, it is my view that such critics as Gross and Levitt or Sokal and Bric-
mont do not represent, and should not be seen as representing, science and
(a) their lack of the minimal necessary familiarity with the specific subject
matter, arguments, idiom, and context of many of the works they criticize;
cal and scientific ideas in these works; (c) their lack of general philosophical
tion, whether one sees them positively or critically; and d) their insufficient
cism by such critics virtually impossible. Indeed, some of their claims con-
are incorrect, which is far less excusable, and in a way more of a disservice
to mathematics and science in their case than in Lacan's, even assuming that
Lacan is as bad as Sokal and Bricmont make him out to be (which is far
mont are, nor does he criticize mathematics and science, let alone belliger-
ently attack them, the way Sokal and Bricmont, or Gross and Levitt, attack
tual network. It makes little, if any, sense without taking both and their con-
Lacan's ideas into a more accessible idiom, and certainly without taking into
tions are bound to retain considerable complexity for the general audience.4
requires no mathematics as such, which one can decouple from this argu-
into statements free from them. Indeed, sometimes one must do so, includ-
ing in the present case, insofar as "the square root of -1" of Lacan's state-
ment is, I shall argue here, in fact not the mathematical ' -1. (For this rea-
son from here on, I shall use the symbol J for the mathematical square root
and use (L)> for Lacan's "square roots" or other symbols of Lacan's "alge-
bra," such as (L)1, (L)-1, and so forth.) One might argue that Lacan should
have performed this subtraction himself and avoided the use of his quasi-
images together. Should he have said instead (with, as will be seen, much the
same meaning) something like the following, "the image of the erectile
between this image and the psychoanalytic reality behind it are profoundly
reality hitherto available here)"? Perhaps. Or, perhaps, were he to keep his
for example, have added the following to the just formulated statement:
cal complex numbers, where the square root of -1 generates the system of
114 * The Knowable and the Unknowable
ics, which is not to say that we can simply (or at all) identify such objects."
For whatever reasons, he did nothing of that nature. Accordingly, one can-
not especially object to, and could even expect, a degree of unease with
Lacan's statement. One would also do well, however, to recall that the
which the statement was made. Indeed, from the psychoanalytic perspective
one might ask why this conjunction appears incongruous or silly to some or
even most of us. One can also ask, in the psychoanalytic context, a number
ical ~ -1, let alone Lacan's "square root of -1," (L)E -1, to the erectile
organ. Conversely, one can question (and some have, of course) the value of
would not undermine my argument here; on the contrary, they would sup-
port it. My point is that, while one can (I would think rather easily) decou-
ple mathematics from Lacan's argument, the reverse cannot be done: one
statements, such as the one in question here. We may call such concepts and
Admittedly, the task of reading Lacan's texts is not easy even for an
complications. (These features in part result from the fact that one usually
confronts not written texts but transcripts of oral presentations, which pose
major editorial problems in their own right.) I need not deal with these
problems in any significant measure here. For, while I cannot fully bypass
Lacan's psychoanalytical argument, I need not fully spell out this argument
attack or even criticize Lacan, the ethics of the discussion gives me a greater
Versions of the Irrational * 1 15
sis and degree of support of our claims. On the other hand, it is not my aim
such a defense; indeed the present argument may even establish proper
grounds for a criticism of Lacan, which may even be more severe than that
found in his Science Wars critics but which may also be meaningful.5 I am
not a follower of Lacan and, while I do find his imagination and some of his
ideas appealing, I do not find all of his texts and arguments either especially
Lacan's arguments.6
the one hand, and literature and art, on the other. The term "concept" is,
not be able to consider them in detail here, although they are of course cru-
cial to Lacan's work and are indissociable from the overall architecture of
his concepts. Indeed, the psychoanalytical dimensions of his work are pri-
marily responsible for the particular role of the concept of the erectile organ
Spirit. Hegel is one of the key figures for and (if implicitly) subjects of the
essay, indelibly inscribed in the phrase "the dialectic of desire" of its title. It
116 * The Knowable and the Unknowable
sketch was made as Hegel was thinking through the structure of his dialec-
(in part via Alexandre Kojeve's reading) rereads this dialectic by using his
organ ("the square root of -1," (L)> -1 of Lacan's system), as the dialectic
to the spirit of Hegel or to the letter of Hegel's spirit, just as he argued that
Freud. (He uses his famous pun on "letter" as, among other things, an item
nections. He may also be seen and would have seen himself as trying to
two sides of Hegel's vision and then reglue them into a kind of Moebius
strip or, given the twisted structure of each movement, into Klein's bottle
(two Moebius strips glued together along their edges). I hope I will be for-
given for this (I would think mild) abuse of topology, especially since I am
merely trying to follow Lacan, and perhaps Hegel. Indeed, these topological
pictures or metaphors, while not without their uses, are rather impoverished
are, although one could, if one wants to go quite a bit further in choosing
bundles (objects, I simplify, that are locally straight, Euclidean, but globally
twisted) of which the Moebius strip is one of the simplest examples.8 Simi-
larly, however (and with the same implications), Lacan's topological dia-
grams are also (although not only) his redrawing and regluing of Kojeve's
concepts in the above sense is, I would argue, where Lacan's usage of math-
ematics most fundamentally belongs and perhaps the best perspective from
From this perspective, there is a way, at least one way, to argue that Lacan's
even a metaphor) between the erectile organ and the square root of -1,
(L)/ -1, make sense. Ironically, in order to pursue this argument one has
indeed to know something not only about Lacan but also about imaginary
and complex numbers and their history. On that score, at least judging by
their comments on the subject in their book, Sokal and Bricmont appear to
be rather less informed than they should have been and, even more ironi-
cally, in some respects perhaps less informed than was Lacan. They appear
objects, in the way some physicists indeed treat them in their work, which is
fine for all practical purposes. The situation, however, is more complicated
their crucial role in defining first (real) irrational numbers and then complex
numbers, square roots will be my primary focus here. Let us recall, first, that
the square root (/) is the mathematical operation reversing the square of a
a fact that will be significant here. I hope I will be forgiven for being so ele-
mentary, but I want even those who know nothing, or forgot everything,
complicated. Thus, X/2 is already a far more complex matter than /4, both
ward mathematical genealogy. One needs it if one wants to know the length
of the diagonal of a square. This is how the Greeks discovered it, or rather
its geometrical equivalent. If the length of the side is 1, the length of the
numerical value is. It does not have an exact numerical value in the way
rational numbers do: that is, it cannot be exactly represented (only approx-
called an "irrational number," and it was the first, or one of the first, of such
lier figures are also mentioned. It was an extraordinary and, at the time,
dal, of Greek mathematics. The diagonal and the side of a square were
also used in its direct sense. The discovery, made by the Pythagoreans
mony of the cosmos was expressible in terms of (whole) numbers and their
This discovery was also in part responsible for a crucial shift from arith-
matics (although some of the key ideas are found in earlier Greek figures,
such as Diophantus), entered the scene and in large measure defined mathe-
while the diagonal of the square was self-evidently within the limits of geo-
as the Greeks conceived of it. On the other hand, following Deleuze and
Guattari, one might also see such problems as paths to solutions, that is, as
light (as independent of the motion of the source); or Heisenberg and Bohr
Greek tragedy, which may allow one to consider the question of Lacan and
Versions of the Irrational * 1 19
subject would require a separate treatment. The source of the legend men-
Elements, which discusses the Pythagoreans and the discovery of the incom-
mensurables and irrationals and further deepens the meaning and the alle-
Euclid's Elements.
The scholium quotes . . . the legend according to which "the first of the
tigation of these matters [of the irrationals and the incommensurables] per-
erly concealed, and, if any soul should rashly invade this region of life and lay
it open, it would be carried away into the sea of becoming and be over-
whelmed by its unresting currents." There would be a reason also for keep-
ing the discovery of irrationals secret for the time in the fact that it rendered
tions addressed in the passage are far from irrelevant to the current (and of
so forth.
forth) at the limit of the rational (the accessible, the knowable, the repre-
ger and beyond, or in mathematics from the Pythagoreans and the diagonal
before him (and at least as profoundly as anyone since him), the extraordi-
nary critical potential of this situation has been powerfully utilized by such
ered here, gave especially remarkable shape to these relationships. The (non-
complex and radical than that confronted by the Greeks. The nature of the
scientific, concepts involved is not unsimilar. To "be carried away into the
current" still appears to some as (and perhaps indeed ultimately is) a great
crucial to the idea and history of complex numbers or other areas of math-
more aware of and attentive both to these concepts and their history them-
selves and to the philosophical thought (often in turn quite radical) of the
key mathematical and scientific figures involved than are their recent critics
numbers together with (real) irrational numbers (such as roots of all powers
I say "real irrational" numbers because all imaginary and complex numbers
are, by definition, irrational, since, not being real numbers, they cannot be
but, it appears, not by Lacan, whom they criticize for confusing irrational
can be zero (the latter, yet another invention of Arab mathematicians, was
Versions of the Irrational * 121
unknown to the Greeks, as were negative numbers).15s The main reason for
using this term is that real numbers are suitable for measurements, in par-
world around us-the world of things that are, or appear to be, real. It is
ize them as points on the continuum of a straight line.16 We can do all stan-
dard arithmetic with real numbers and generate new real numbers in the
0), and so forth. The same is true for rational numbers but (because of divi-
no problem. We can always mathematically define its square root and cal-
domain of real numbers the square root can be defined, can be given unam-
very simple reason (recall that the square root is the reversal of the square):
times -2 is also 4, and the same is true for 1 and -1-the square of both is
exist, at least in the way real numbers exist or appear to exist. This is why,
when introduced, they were called imaginary, and sometimes even impossi-
ble, numbers.
Why worry, then? First, from early on it appeared (correctly) that one
could operate with square roots of negative numbers as with any other num-
bers-add them, subtract them, multiply them, divide them, and so forth.
they were called at the time) made their first appearance during the Renais-
with -1. On the other hand, it was clear that such "numbers" could not be
accept the mathematical legitimacy, let alone reality (the mathematical real-
ity), of these new entities and rigorously to define them as numbers. Their
122 * The Knowable and the Unknowable
physics, where they eventually came to play a major role). The resolution
required a great and protracted effort and the best mathematical minds
tage point, but in truth, at least at the time, highly nontrivial stratagem-by
bers (for which the term "imaginary numbers" was retained) but are in gen-
been known for just as long as strictly imaginary numbers. Complex num-
bers in general are written in the form a + bi, where a and b are real num-
remains the simplest complex number, and it may be seen as the fundamen-
tal element, which, once it is formally adjoined to the old domain of real
Just as real (or rational) numbers do, complex numbers form what is in
needs to worry about how to properly define any roots or, more generally,
any solution of polynomial equations. For, unlike the case of real numbers,
they now all belong to the same field. (In mathematical terms, the field of
numbers is all one needs in order to have not only x2 + 1 = 0 but every poly-
exactly as many solutions as the (highest) degree of the polynomial. The lat-
ter statement constitutes the so-called main theorem of algebra, one of the
it also became possible to represent the whole system on the regular real (in
above zero on the vertical axis. In this representation the domain of com-
for it was the great (one of the greatest ever) German mathematician Karl
Friedrich Gauss who used it in his efforts to give legitimacy and perhaps
address on the fifty-year jubilee of his doctorate. In 1977 the German Post
discovery he, however, suppressed for twenty years for fear of being laughed
at by philistines or perhaps for the same reason that made the Pythagoreans
to think it wise to conceal the existence of the irrationals. Gauss did not
think, as it happened rightly, that the human world was quite ready to con-
front the non-Euclidean world. As will be seen, this and related geometrical
and topological aspects of Gauss's work are extraordinary and are highly
seen, Gauss might have been more aware than it might appear and which
are crucial for the present argument. Accordingly, I shall discuss these com-
plications at some length, even though I can still offer only a sketch rather
than a full argument. Lacan appears to have seen the epistemology of com-
plex numbers along these lines (even if in a more general and vague way)
tion here (L) J -1, into his argument concerning the dialectic of desire. I
at least possible and even reasonable, given the epistemological aspects of,
but not in themselves. That is, the Gauss-Argand plane diagram may not be
erties. Far more diverse and complex geometrical (and topological) means
observed that, while all visualization may need to be seen as in one way or
complex numbers and certain other "geometrical" objects, both spatial and
alization of some of their aspects are possible and are even necessary. A
much more complex interplay of algebra and geometry, along with analysis
mathematics, these two objects are not isomorphic, insofar as one can
assign, and one can, any algebraic structure to the real plane. The main rea-
son for this is that a real (in the mathematical sense) point on the two-
dimensional plane with Cartesian (now truly Cartesian rather than Carte-
braically, a point on the real plane can only be seen as a vector, usually "pic-
extending from the zero point to a given coordinate point. This structure,
sional real plane itself is seen as having or is given the structure of the so-
called vector space over real numbers, rather than being a field, as are com-
plex numbers or real (or rational) numbers. In the case of vector spaces, for
numbers over which a given vector space is defined, here that of real num-
bers. (Vector spaces can be defined over the field of complex numbers as
sional real plane.17 This absence defines the difference between the algebra
of vector spaces, such as that of the real plane naturally endowed with a vec-
tor-space algebraic structure, and the algebra of fields, such as that of com-
plex or real (or rational) numbers. In contrast to the field of complex num-
bers, the field of real numbers is represented by the (real) line as a field and
the real line. (As will be seen, epistemologically the situation may be more
complicated even in these cases.) In any event, in the case of fields both mul-
tiplication and division between their elements are defined as well (except,
again, the division by zero), and in the former case, again, they are not. We
sional plane by real numbers, but not among themselves. Thus the point
(1,0) can be multiplied by any given number, say, a, giving us the point
say, the point (0,1). This can only be done for the elements, such as 1 and i,
such coordinate points on the Argand plane but which are fundamentally
considered as a vector space, and thus is really not captured by the Argand
plane. Also, importantly, unlike real numbers or real vectors, complex num-
bers as such cannot be assigned lengths and, hence, cannot be used in mea-
surements. In this sense, they are indeed deeply "irrational," insofar as they
/z/ of this number defined (by virtue of the Pythagorean theorem) as Iz/ =
x + y2. Geometrically, the modulus is the distance from the point 0 to the
sured. The modulus is also equal to the product (x + yi) x (x - yi) of a given
It follows that the complex plane in fact carries two different, noniso-
morphic, algebraic structures: (a) that (an algebraically richer one) of a com-
real vector space (not a field) over real numbers. The latter is, by definition,
126 * The Knowable and the Unknowable
however, to give the real plane the structure of an algebraic field is to endow
it with the structure of the field of complex numbers, which would no longer
allow us to see it (in either sense) as a real plane. Indeed, the language of this
giving the real plane the structure of the algebraic field. This simply cannot
be done on the real plane but only on the complex "line," as it were, since
equal to one rather than to two. This structure, I argue, may ultimately not
even to some of its aspects appears to require much more complex means
than the Gauss-Argand plane. Indeed, the heterogeneity of these means may
exceed the capacity of any single geometrical object, let alone a diagram,
such as the Gauss-Argand plane, which, I argue, may be only, spatially and
I here, again, use the term "visualizable" in a sense the way it was used
tions, images, pictures, and so forth. As will be seen presently, the same term
lines. This double nature or the double structure of the topological space of
complex numbers is, I would argue, the source of the great-I would call it
matics of, or involving, complex numbers. One may define the Euclidean
(in the present sense) of such objects themselves and of their topological and
physics and its model, and indeed from Galileo on served as a source of such
(flat) Euclidean plane may be the only rigorous realization of this model
objects.19 The double nature of the complex plane appears to force one to
and are differently involved in, constituting the real and complex properties
tive-but ultimately not unifiable either in practice (as many would admit) or
(an idea that few are willing to entertain) in principle. We can see this het-
resentations may be more direct but partial or more complete but diagram-
usage of geometrical or, again, more generally spatial concepts, if not terms,
while in fact referring to the objects that can only be algebraically defined or
128 * The Knowable and the Unknowable
indeed rigorously conceived.20 These questions may even apply to the very
an obvious case here. There are also far more esoteric mathematical objects
available as examples. The point, however, appears to apply even in the case
may be argued that Rene Descartes's analytic geometry already poses certain
It is, of course, true that the spatial-geometrical metaphors and the very
There are even spaces without points altogether, as in the so-called topos
sion of the concept of topology (in the sense of "structure" rather than "dis-
that in all these cases, and even in the case of complex numbers, perhaps at
the limit all numbers, even natural numbers, we are in fact dealing with
"algebra," even if in the name of geometry. That is, in fact we ultimately use
the equivalents or analogues of the algebra associated with, and not of the
ics. This algebra does of course have a certain specificity that arises by virtue
of this association with space and geometry in the latter case and that,
spaces are non-Euclidean, even in the case of special relativity, and, hence,
such objects, we visualize only three- (and perhaps mostly two-) dimen-
sional configurations and then (or in the process) supplement them by alge-
try"-which names the field that is, arguably, the ultimate extension of the
geometry and developed from the late eighteenth century to the present, in
particular in the works of Andre Weil and Grothendieck and their follow-
These points may be more immediately apparent and more often sug-
gested in the case of the arcane objects just mentioned, perhaps especially
case where, as we have seen throughout this study, one has confronted this
spaces, or that of complex numbers (over which the vector spaces of quan-
dealing with things that are inconceivable in any terms and by any means,
gories (in de Man's sense), rather than only with that which is spatially
that the argument would apply everywhere beyond three dimensions and in
can conceive of space as its object, this science as such can perhaps only deal
with real spaces (of dimensions below or equal to three) and even, in all
rigor, only with Euclidean spaces. These spaces sometimes allow us to real-
ize partial or local models, more or less rigorous or more or less loose, of
as I argue here, the history of mathematics and science has taught us that we
need not be able even to conceive of, let alone spatially or pictorially visual-
examples, but, with respect to the rigorous conceivability of its objects, the
logically even more radical than in quantum physics. In the latter case we
can at least ascertain the material character of the efficacity of these effects
in nature, locating this efficacity beyond the furthest classically drawn "cut"
again, what is thus idealized (as quantum or, again, even as "constitution,"
this idealization. (The latter qualification is here used both in the sense that
it can be idealized nonclassically and in the sense that, even though non-
in the same way. Such objects may quite simply not exist anywhere, even
logically more radical or at least more complex than the case of quantum
terms of, say, material biology of the brain as a sufficiently ultimate efficac-
ity here, say, for the purposes of scientific investigation of the human mind,
does (evidence is, again, so far slim at best, even at the conjectural level).28
here. While the subject of lively investigations and debate, it is at the very
early stages even at the most rudimentary levels.29 The possibility of locat-
is almost infinitely far away, if possible at all, even by the most optimistic
views. Even this, however, may not get us to mathematical "objects" them-
selves, say, complex numbers, or for that matter any numbers, let alone
exist at all in any conceivable sense, in other words if they might be thought
cally, once, as I said, more chips are in, say, when we have some form of
material" efficacity (it could still depend on the materiality of our body, the
both situations even closer, but it does not quite erase the differences
tarity, one can approach its "inconceivable" objects only by means of math-
linked and work, with the help of experimental technology. This is one of
mology, which, however, also links differently nature, mind, and technol-
ogy. (For the sake of graphic comfort, I skip quotations marks, which can
type of object. I now mean this understanding in the sense of rigorous con-
ments, cited earlier, even though he may not see the situation in epistemo-
ject and is well beyond my scope here. The epistemological situation of rep-
epistemology outside the sciences in the work of most of the key figures con-
this mathematics subscribe to this view), or, again, at least, just as in the
able) way this mathematics depends on and benefits from so doing in view
mysterious way. But, luckily for us, it works, and often where classical the-
allels it with the rise and emergence of the complex numbers. The latter his-
of affairs, the story of the complex numbers. For centuries people assumed
that they could add to the real numbers a 'number' [i] such that i2= -1, and
they thus obtained many results without running into contradictions [a cru-
the 19th century people exhibit models of the complex numbers [just as
While there are some reasons for this conclusion (especially as far as the last
true, certainly not as far as the history of the debate about complex numbers
should speak of a revolution in both cases, even though both involved a fair
amount of "normal science," which, however, is usually the case in all rev-
olutions. Indeed, as I argue, both may be seen as part of the same non-
by -1 reversing the sign, then we are inclined to accept the 90-degree rotation
(of the plane containing the line of real numbers) as the square root of -1. All
this looks childishly simple, why do mathematicians make such a fuss around
it? How can one dare to compare this plain idea to profound philosophical
ergo sum" stayed unperturbed for more than three centuries, like a monu-
time, whilst the little speck of dust, the square root of -1, has been growing
geniuses like Cauchy, Gauss and Riemann, and turned into an evergreen
intensely alive vibrant tree supporting in its branches our sacred knowl-
world.32
leading as concerns Descartes and "Cogito ergo sum" (assuming that this
one may make virtually the same observation concerning it as that Gromov
phers, such as Kant, Hegel, Husserl, Heidegger, Levinas, and Derrida, and
Riemann, among them, and turning into an evergreen intensely alive vibrant
everything we see (and do not see) in this world. This is not to deny either
would only reinforce the present point. (It is not impossible to speak of a cri-
question here and in the thought of scientific figures just mentioned, begin-
ning with the work of Descartes himself (who, as will be seen presently, was
the first to use the term imaginary numbers) and extending to, among oth-
ers, Gauss, Riemann, and Bohr. It would be, I think, quite naive to think
Gromov is right to give complex numbers the place they deserve in intellec-
tual history, but it may not hurt, at this point, to give Descartes's Cogito the
ing of mathematics in terms of concepts and his more radical approach to the
concept of space. Riemann here follows both his teacher Gauss (more math-
Riemann left this area and constructed his spaces in a completely different
way."33 I think this is, by and large, correct. I would replace entirely with
often, especially given Gauss's work. I would further argue that Riemann's
and some earlier work, especially again that of Gauss, was so radically revo-
some degree, even physics (to which they made major contributions as well,
and not only of a mathematical nature) to, at least, the threshold of nonclas-
sical thought.
addressing the question of the complex plane itself from this perspective,
of which it was not initially possible to speak of, say, straight lines, distances
between points, and angles." In other words, in our terms, this is a purely
ical object would still remain a question. Even leaving these deeper consid-
was in question, even as late as 1849, fifty years after Gauss's dissertation,
proof of the theorem for which he was awarded his degree. Gauss says:
"The wording of the proof [of his theorem] is taken from the geometry of
developed, and one cannot move in it without the use of language borrowed
Leibniz, but more deliberately and decisively, and, one might say, more
"analysis situs" is still Gauss's term, as it is also that of Riemann and even
cipline similar to algebra and analysis or, of course, arithmetic and geome-
try, earlier. The term "topology" was introduced around Riemann's time,
136 * The Knowable and the Unknowable
however. It may be argued that already the Greeks had a kind of topology
Greeks went rather far in their topology, for example, in Plato's discussion
which I shall further comment in the next chapter. Topology (or what one
the spatial. However, for the Greeks, as for the moderns, geometry was the
only mathematics of space. Leibniz was arguably the first to envision a pos-
calculus. With Gauss (one can invoke several others among his contempo-
erations enters the history. Riemann takes it further, and for both topologi-
dimensional plane, the Gauss-Argand plane, and so forth are clearly sug-
From the perspective of the statement just cited, Gauss's view of geomet-
appears to indicate that for Gauss, too, the complex plane may have been
ultimately a diagram used for the sake of possible simplicity and intuitabil-
ity rather than a rigorous geometrical representation. That is, such would be
the case if we can call such a diagrammatic picture the Gauss-Argand plane,
which, at least for Gauss, may have meant a rather different topological/
above analysis in mind. The same type of argument may be made concern-
ing Riemann's view of both complex numbers and spatiality of space. There
still remains the question of how far both Gauss's and Riemann's views
reach epistemologically along the lines indicated earlier, that is, as concerns
tions would indicate, however, there may be some surprises in store as con-
Bohr's views are suggested by Gauss's passage, although one must, again, be
Versions of the Irrational * 137
ered the points of the plane given by real coordinates t, u and introduced an
'algebraic structure,' that of [the field of] complex numbers. Riemann was
with the preceding analysis. It is clear, however, that Gauss, who anticipates
try and topology and the emergence of complex numbers are parallel, as
Now, in the case of the complex numbers it is indeed the particular alge-
braic structure of the field that poses the main problem, in spite of the pos-
sense and for the reasons just delineated that I see the latter as a "diagram"
way the (real) line represents the real numbers, or perhaps appears to repre-
sent them. Ultimately, these considerations may pose more general and still
point (1,1) on the plane); or using a pair of similar triangles enables one to
locate the multiple of two complex numbers. This is what the Argand plane
plane. The Argand plane diagram as such gives us only this representation.
(topological) correspondence between the points of the real plane and com-
138 * The Knowable and the Unknowable
numbers.
would be naturally possible in the case of the real line and real numbers.
described earlier (i.e., with the complex "plane" as both the field of complex
numbers and the real vector space). Second, all arithmetic of real numbers
ever, the first point, my main point here, will be retained in all circum-
stances, the situation may be more complex even in the case of real num-
even in the case of the real numbers, even though multiplication appears to
Definitions 15 and 16 of Book VII of The Elements. The same may be said
think of are involved even in the case of the most familiar mathematical
cal construction of arithmetical operations with the rationals and the irra-
tionals on the plane and, of course, that in this case we are dealing with
intervals and their length, not with "real numbers." In other words, there
may be questions, ultimately along the more general lines indicated earlier,
real numbers in all their properties and in all of their aspects, and in partic-
ular the operations performed upon them. Adding two intervals (whatever
of course that the result is an interval of the length 6, and we know that it is
the same as to add the first interval to itself three times, or that this product
the case of the fractions and, then, the irrationals, for example). But would
(one is almost tempted to add) than in the case of complex numbers and the
Versions of the Irrational * 139
parallels with the case of complex numbers. In other words, the line may be
geometrically capturing only the additive but not the multiplicative proper-
ties of the real numbers. Hence, it captures only their vector space but not
their field structure, similarly to the way the real plane captures the vector
space but not the field structure of the complex numbers. In the latter case,
it does so, however, by virtue of the fact that as a vector space the "com-
plex" plane is isomorphic to the real plane, but as a field it is not, which is,
to think even in the simplest cases. This is in part why algebra is such a great
Gauss's algebra (his achievements in this field were phenomenal) and topol-
make possible the impossible (at least before Godel came along).
sentation or/as visualization. First of all, these differences are responsible for
the difference in attitudes toward the imaginary and complex (as opposed to
involved in the earlier history of complex numbers. Certainly, the more rad-
them on the part of the key figures involved in their discovery or creation.37
great mathematician in his own right, had reservations concerning the geo-
ered them as purely symbolic (algebraic) entities and at one point even
we have seen, even, or indeed in particular, Gauss's view of the situation has
complexities are involved in the case of real numbers, or indeed of all num-
bers. Lacan aside, the question may well be, Do we know what is J -1, or
the eighteenth century), or for that matter 1, which may take us all the way
Pythagoreans, as the opening of this question? Gottlob Frege once said that
it is scandalous that we do not know what numbers really are.39 The spatial
way, it would not change the point here defined by the double algebraic
real plane and the other (a field) not. As a result the representation/visual-
be seen only as the "real" diagram in the above sense, and in this sense as a
visualization of the complex numbers. The latter, this is my main point here,
geometrical objects. Their properties can be spelled out and rigorously com-
domains. Thus, ultimately, complex numbers may remain not only imagi-
nary but, at least geometrically, strictly unimaginable. They (in the ultimate
the square root of-1 or, more accurately, the signifier "-1" signals-"rep-
bers in, epistemologically, this type of way (as he appears to have), he had
complex numbers (rather than Lacan) these nuances and qualifications are
in a more rigorous way, which is what I have tried to do here. They would,
however, leave plenty of space for making Lacan's looser view of complex
with the erectile organ. However, given the argument just sketched and
is in fact not so strange. One must keep in mind the much looser (than in
nature of some of its specific features, used here, may, as I have stressed
from the outset, pose difficulties in turn. In this sense, at certain junctures at
analogy, the structure of his concept, and his view of complex numbers
(shaping this analogy), not the (more) arcane specifics of Lacan's concepts
and argument; and, hence, it will remain sufficiently tolerant of the difficul-
dimensional real plane-gives one a hint. "The erectile organ" may be seen
cally analogous to the signifiers one encounters in the case of complex num-
tion, any image, in particular visual image, of the erectile organ, including
signifier, not the signified. (I shall further comment on this point presently.)
limit, this signifier-that is, the ultimate structure of the signifier designated
this signifier is in fact or in effect unnamable, for example, again, as the erec-
tile organ, or the phallus, which, as I said, may not be the same as the erec-
tile organ within the Lacanian economy of subjectivity and desire. That is,
142 * The Knowable and the Unknowable
ciated with it, just as we can formally manipulate complex numbers within
not really know and perhaps cannot in principle conceive of what the erec-
tile organ really is as a signifier and what its properties are, if one can speak
sible signifiers within it.) The image of this signifier, and in particular its
such.40 The situation may even be subtler, insofar as one may need to deal
ally emerges. Let me stress that (whether one is within Ferdinand de Saus-
sure's or Lacan's scheme of signification) we are dealing here with the irre-
ducible inconceivability of the erectile organ as the signifier, not the signified
between the signifier and the referent within signification). Its signified and
its referent, whatever they may be, may be in a certain sense even more
ever, to think here in terms of the ultimate inaccessible nature (which is not
signified, and the referent, the latter of which should be considered in the
register of the Lacanian Real. That is, along the lines considered earlier, the
to think of the latter as an object in any way. It would, again, be more rig-
that are associated more directly (although still ultimately as effects) with
such an object rather than to attribute them to this object as such. It follows
that one can further split this situation into a potentially infinite chain, sim-
matic) views of signification, both of which are, along with Saussure's semi-
effects of the Real and the irreducible and irreducibly nonclassical random-
ness of these effects as considered earlier. This, of course, does not mean
mately the Real that is responsible for the nonclassical features of Lacan's
epistemology. I shall not pursue this argument here, but one can virtually
situation (even the role of formalism) into the Lacanian matrix.42 It may,
(since we cannot speak of the latter as such) of the strange effects of this
something more similar to "madness" than "reason," using both terms with
utmost caution. Unlike Freud, Lacan does apply the psychoanalytic machin-
ery to psychosis as well, a related but separate issue, which cannot be con-
sidered here. De Man, too, tropes the effects of irony as a kind of mad ver-
analytically (in either sense), to madness, that is, again, to the effects of
madness, only in terms of and through reason. Or, we might say, that we
we can only relate to the (to us, seemingly, "mad") "quantum world"
through its effects upon our classical, reasonable instruments, even though
the ultimate constitution of the latter may be quantum as well. But then,
be irreducibly random or, again, how the ordered effects are constituted of
random ones. In a certain sense, from Socrates and before on, this question
defines the history of our inquiries into human nature, and by the twentieth
Real operates in this regime and is seen as the ultimate and ultimately mate-
rial (in the sense of the preceding chapter) efficacity of all the effects in ques-
the erectile organ, (L)> -1, in the Lacanian psychoanalytic field, and J -1, i,
ties that enable an introduction of, and may be seen as structurally generat-
bols allows one to deal with problems that arise within previously estab-
lished situations but that cannot be solved by their means. The first situation
analytic situation in psychoanalysis, where one needed to, and in the previ-
ous regime could not, approach certain particular forms of anxiety. The sec-
needed to but could not rigorously define complex numbers in order, for
(a) the extent to which such systems represent or otherwise relate to,
(b) the extent to which the properties of such symbolic systems and of
situation.44
like language may or may not be seen as out of place (and while one might
serve as a "model." At least, there is some space for and value in parallel, if
not for crossing of the terms of discussion. As I said, one may still wonder
why one might prefer to avoid this crossing in this context (the present
Versions of the Irrational * 145
ter reaction, though, seems to me to be taken too far. It is not that big a deal
after all. As Lacan obviously did not have such misgivings, he offered us this
crossing as well, and if one is to address it, critically or not, one must try to
reflected in the term "imaginary numbers," are not altogether resolved even
Gauss might have worried more than them and more than they suspected.
Leibniz may well have given the problem its most glamorous expression:
"Imaginary roots are a subtle and wonderful resort of the divine spirit, a
non Ens Amphibio)."45 Perhaps Descartes should be given the last word
here. For, among the great many other things he was and the many other
"firsts" to his credit, he was one of the first to give serious consideration to
imaginary roots and their nature and, indeed, was the first to use the very
have seen, ventured into more complex relationships between algebra and
said, quite possibly both to his concept (also in Deleuze and Guattari's
(on both geometry and complex numbers) may be seen as a continuation of,
unless the last word is Lacan's, who said in "Desire and the Interpretation
sense of the term-and yet it must be conserved, along with its full func-
tioning" (29). This may need to be more precisely stated, but it is in essence
right, and it is this statement that grounds and guides my analysis here. It
the irrationals are born the impossible or imaginary quantities whose nature
tional." The passage is cited both in Sokal's hoax article and in Impostures
nary numbers. Lacan's usage, however, does not appear to me, due to his
ered earlier, and in the sense of modern algebra. This view is correct, and it
nary numbers in their book is hardly edifying as concerns the substance and
numbers. Indeed, they claim even more strongly that they have nothing to
ning with i, are irrational numbers as the latter are defined by Sokal and
fraction can be found to represent them, since no real number of any kind
sense that both require one to adjoin to the system square roots that do not
exist within it and for which the "rationality" (now in the sense of compre-
hensibility) of the original system provides. But, for the moment, this is sec-
latter may seem a minor point, and, while a strange oversight, one can
One might expect two theoretical physicists to be a bit more cautious, given
how these aspects of complex numbers define the way in which they func-
tion, even in strictly technical terms, in quantum mechanics, even if they are
not holding Sokal and Bricmont responsible for their treatment of numbers
other mathematical and scientific ideas are often sloppy and not very help-
science. (I shall give further examples later.) They are physicists, not mathe-
point-their responsibility to know those aspects of all three that they con-
sider in Lacan. Or, assuming that they do, it is their responsibility to care-
Versions of the Irrational * 147
fully consider and appropriately explain these issues as part of their argu-
The erectile organ of the Lacanian, or, Lacan argues (in part "against"
Freud himself), already Freudian, system is, then, analogous to the mathe-
cal) kind of rigor that the mathematical system of complex numbers does; it
is much "looser," although it is not without its own (form of) rigor. In this
or absorbs this enclosure.) While this enclosure plays its role in and is, in a
other). Certainly, insofar as one can see Lacan as borrowing from the math-
inal system is lost in the process. These circumstances, I argue here, indeed
decouple some or even most (although perhaps not all) mathematics in its
ever, the very same circumstances also make it impossible (certainly, inad-
The erectile organ, or, again, a certain formalization of it, must be seen as
(as defined by and as defining) "the square root of-1," (L)E-1 of the Lacan-
word, the erectile organ is the square root of -1, which I here designate as
the (L)> -1, of Lacan's system; the mathematical ' -1 is not the erectile
gies with the mathematics of complex numbers, most particularly the fol-
indeed the signifier (in Lacan's sense), belongs and gives rise to a psychoan-
alytical system different from the standard one or ones (based on misread-
148 * The Knowable and the Unknowable
tion that is more effective both conceptually and in terms of the ensuing prac-
or, again, the (Lacanian) signifier of this system, while and in a sense because
and its referent. At the limit it is inaccessible even as that which is absolutely
In order to explain the reasons for seeing Lacan's statement in this way,
I shall indicate some steps and elements of Lacan's logic and "algebra"
(mimicking the algebra of complex numbers). I shall not spell out the struc-
textual field. This, as I said, may introduce certain difficulties into my argu-
ment, which I shall try to mitigate as much as possible; it may not be possi-
ble to avoid them within my limits here. In this sense, the "summary" to fol-
and, arguably, the very reason for the epistemology or analogy in question.
used only in order to show that Lacan relates them to the epistemology in
question and will not be explicated or read as such. This, I realize, may be
equally (albeit for different reasons) frustrating both for those who know
Lacan and for those who do not. However, as I said, my argument essen-
implications of, and reasons for, Lacan's argument, not its psychoanalytical
in the way Sokal and Bricmont could, and perhaps should, have tried before
to unravel this definition (or virtually any proposition to be cited later) and
to give it a proper sense, which, however, is not important for the present
(represented by) the "1," (L)1, of the system, which Lacan is about to map
by (or relate to) his "algebra." This formalization would reflect a certain
ing respectively the "-1," (L) -1, and "the square root of-1," (L)/-1, of his
system. Indeed, he does not appear to see any configuration short of this
analytic theory and practice. What is, again, important for me here is the
system, "S" that is "the signifier for which all the other signifiers represent
(-1) in the whole set of signifiers" (316). This signifier is, thus, (symbolized
as) "the -1," (L) -1, of the psychoanalytical system in question in its "alge-
braic" representation (the "algebra" is, again, that of Lacan). This negation
tioning this point in order to indicate that at stake here is indeed Lacan's
system and its "numbers" rather than mathematical complex numbers; and
Lacan's map or graph of the "economy" of desire offered in the essay clearly
shows this as well. "As such, S is inexpressible, but its operation is not inex-
even have a certain visualizable or geometrical model (in the way real num-
manipulate such symbols, even if their (signified) content (or, of course, ref-
erents) or, in some cases, even their ultimate structure and nature as
ever, another signifier, "s"-the symbolic square root of -1, (L)> -1-is
"derived." This derivation takes the form of a symbolic analogy with the
relations are given the psychoanalytic content via the nature of S and s
argue, more radically, at the level of the corresponding element of the for-
There are, thus, two interactive but distinct levels of the functioning and
epistemology of the signifier in Lacan. The first is the more general concep-
tual level (subject, the phallus, lack, and so forth), which is not quasi-math-
with the erectile organ. First, Lacan only argues that a certain radically inac-
again, yet another formalization of it, as "the square root of -1," (L) J -1,
of this formalization, rather than the "1," (L)1, or "-1," (L) -1, (which is
"S") of the system. It is of course also responsible for the specific character
-1] is what the subject lacks in order to think himself exhausted by his cog-
ito, namely that which is unthinkable for him [although, we might add, it
and much of the philosophical content, and cite the statement only to indi-
cate Lacan's epistemological concerns. "But," Lacan asks next, "where does
this being, who appears in some way defective in the sea of proper names,
originate?" (317).
In order to answer this question, Lacan maps the passage from, in his
terms, the Imaginary to the Symbolic order, especially as regards the phallic
via the castration complex. Here one must keep in mind the difference
between Freud and Lacan insofar as the Lacanian economy of the signifier
phallus and the difference between the phallus (as a Freudian signified) and
what Lacan designates as the erectile organ as a signifier. The latter, more-
over, may need to be seen as formalized yet further as "the square root of
-1," (L) ' -1, thus adding yet another "more distant" level of signification.54
The jouissance [associated with the infinitude involved in the castration com-
plex in Freud] ... brings with it the mark of prohibition, and, in order to con-
stitute that mark, involves a sacrifice: that which is made in one and the same
This choice is allowed because the phallus, that is, the image of the penis,
is negativity in its place in the specular image. This is what predestines the
sacrifice, but which, at the same time, masks the fact that it gives it its instru-
ment.55ss
Lacan, that it is the erectile organ-the image or, better, the signifier as an
un-image of the phallus, and thus un-image of the image of the penis (in the
issue.56 It is, then, as such and only as such that the erectile organ is the
square root of-1, that is, (L) J/-1-that is, as "the square root" within, and
of, the Lacanian system itself in the Symbolic order of its operation or,
erectile organ comes to symbolize [again, also in Lacan's sense of the sym-
bolic] the place of jouissance, not in itself, or even in the form of an image,
it restores by the coefficient of its statement to the function of the lack of the
x2 = -1. The main point here is that "the signification of the phallus" so
shown that neither the signified nor the referent is simply suspended here,
ian Real. For as Lacan says, "if [the erectile organ's role], therefore, is to
bind the prohibition of jouissance, it is nevertheless not [only] for these for-
mately related to the Real and its epistemology. As I have indicated, the Real
any way that is or will ever be available to us), which only emerge at the
handled by Lacan's Real. In other words, the register of the Real follows the
material efficacity, which is the Real. In short, the Real is unknowable in the
of the erectile organ is a scandal-in either sense, but most crucially in terms
thereof. In this latter sense it is not unlike what ' -1 in mathematics was
the juxtaposition between the Symbolic and the Imaginary. In the register of
the new system (in the Symbolic register) the Lacanian signifiers themselves,
the dialectic of desire and castration, which enables the subject defined by
this system and/as the Lacanian analytical situation to function. The sym-
bolic object itself in question is given a specific formal structure, just as ' -1
is in mathematics, if, again, in Lacan a looser one. From this perspective, the
erectile organ is not a real unity or oneness, positive or negative, neither "1"
nor "-1," the (post-Parmenidean) One, positive or negative, nor even any-
that contains oneness, "1," (L)1, and the negative of oneness, "-1," (L) -1,
as terms but that makes the "solution" itself, while in a certain sense for-
Versions of the Irrational * 153
malizable, inaccessible even at the level of the signifier. To the degree they
offer us any image of it all, all our imaginaries and visualizations ineluctably
"miss" this signifier, along with the signified and the referent-the Real,
statements cited here, that they did play a role, however. He also knew
enough mathematics and mathematicians to draw this parallel and to use it.
some of the statements cited earlier would indicate, even if he did not actu-
ally derive his scheme from the epistemology of complex numbers. There
are other candidates for the sources of this epistemology of the inaccessible
more immediate semiotic terms, such as in the work of Saussure and Hjelm-
slef, on the one hand, and Peirce, on the other. In more philosophical terms,
one can see such sources in nonclassical philosophy in the wake of Kant and
Hegel, especially in Nietzsche, although one can, again, also trace some of
It is, then, only in the sense of the square root of -1, (L) ' -1 of Lacan's
numbers, that the erectile organ is the square root of -1. This argument
would clearly invalidate the kind of critique that Sokal and Bricmont level
at Lacan, were their critique to survive far lesser levels of scrutiny. Unwit-
tingly, Sokal and Bricmont's own comment in fact says as much: "Even if
Indeed! This is the point. It is clear even from the most cursory reading that
Lacan never says they are, although, as we have seen, mathematical num-
tence introducing the formula in question and cited by Sokal and Bricmont,
braic method used here."60 That is, Lacan is calculating here according to
point here. Indeed, Sokal and Bricmont cite other statements by Lacan
main reason why modern mathematics and science enter the work of Lacan
and other so-called postmodernist figures (I use this highly misleading term
154 * The Knowable and the Unknowable
or Derrida. There is nothing new here. The same has been the case through-
out the history of philosophy, beginning with Johannes Kepler, Galileo, and
Newton, or indeed with the mathematics of Plato's time, the irrational num-
phers, beginning with Plato, may and sometimes do get certain things wrong
and some of these things are not easy, technically or philosophically. These
science by these figures, their desire to parade their familiarity with subjects
by Gross and Levitt or Sokal and Bricmont. Lacan's statements and certain
then tried to use these ideas, however loosely. Subsequently he used topol-
ogy in a similar way, that is, one can say (and some often do), loosely
Lacan's construction considered here may have been designed primarily for
and the like. For example, on some occasions certain constructions of mod-
ern topology, such as the Moebius strip and the Klein bottle, serve Lacan to
find that which "we [can] propose to intuition in order to show" certain
nonvisualizable, algebra and geometry, and so forth; and the role of these
Versions of the Irrational * 155
sophical concept. The presence and role of such concepts in Lacan are, in
given sample of Lacan's text. As must be apparent from the passages cited
here, in "The Subversion of the Subject," imaginary numbers are only a por-
phy, or whatever.
As I said, this view shifts Lacan's usage of mathematics from the psycho-
analytic into the philosophical register, in accord with Deleuze and Guat-
ence, and art-Deleuze and Guattari omit psychoanalysis from this argu-
have of course been the subject of important recent investigations. One can
cially in The Post Card (where Lacan is the main subject, along with Freud
tions that philosophy and psychoanalysis are multiply and perhaps irre-
ent (and more fundamental) role that mathematical concepts play in philo-
sophical versus psychoanalytic thought and discourse. Indeed, one may see
more scientific or, with Freud, to affirm its scientific character. In the
their role in our culture. At the very least, he makes apparent the extraordi-
modern mathematics, even though he himself does not always takes advan-
tage of the richness of these resources either. The success of his deployment
ferent question, in part given the very nature of his thought, work, and
evaluates him in this role) rather than as a psychoanalyst, to the degree that
sis and philosophy in general, as just indicated. At the very least, the role of
understood.
-JACQUES DERRIDA
Derrida's essay "Structure, Sign, and Play in the Discourse of the Human
ernism, a history extending to the Science Wars and beyond. The essay was
shall cite here, appeared in 1970.) By then the controversy was on its way to
possible parallels between Derrida's ideas and modern mathematics and sci-
ence, specifically Einstein's relativity theory. In his reply, Derrida made the
following statement:
The Einsteinian constant is not a constant, is not a center. It is the very con-
cept of variability-it is, finally, the concept of the game [jeu]. In other
observer could master the field-but the very concept of the game which,
the Science Wars. Both Gross and Levitt's Higher Superstition and then
158 * The Knowable and the Unknowable
Sokal's hoax article in Social Text commented on the passage and origi-
nated this wider circulation.2 Amid the controversy, there followed the
appearance of Sokal and Bricmont's book and ripple effects in its wake.3 In
the (hopefully) final stages of the Science Wars other authors, such as Lacan
or Latour, moved into the foreground as figures of, and figures for, the
The case of "Derrida and relativity," however, has never disappeared from
the Science Wars radars. It also retains its paradigmatic significance for our
possible and that remain in place and at work, as they were long before the
Science Wars.
The circulation of Derrida's remark and the treatment that it received are
sical thinking, affects the current intellectual and cultural landscape, and
work and its treatment in the Science Wars debates can be made for other
and still others, and such arguments have been already made here and a few
more will be made later.6 While paradigmatic as far as its treatment is con-
narily misplaced, prominence of the statement just cited and of his work, or
rather name (the work is hardly addressed), in the Science Wars attacks. It
the reasons for this resistance, as explained earlier, Derrida has long been an
question was made to stand for all of the deconstructive or even postmod-
diminish the significance of this fact but, as I shall explain, instead amplify
and considers their significance in philosophy and elsewhere. His first book
an investigation that largely defines his work. In his reflections on the rela-
tionships between his work and mathematics and science, and in his actual
engagement with philosophical and literary texts for his work.9 One might
argue that mathematics and science play a more significant role in his work
matics and science, and he even argues that "science is absolutely indispens-
very least in relation to the role of mathematics and science for all modern
mathematics and science, and has been used by them, as, I shall argue in the
ence (or, conversely, the relevance of the latter for his work), however, nor
his caution in this respect has been considered by his critics in the scientific
question. These critics instead appear (this has not diminished during the
post-Science Wars debates) to base their views of Derrida's ideas, and those
ments in the context of his work and often ignoring even the most basic
rules of reading, let alone scholarly reading, of Derrida's and others' texts.12
Such a placement could enable one to give these statements their more
the same time, taking into account the circumstances under which some of
these statements, such as the one on "the Einsteinian constant," were made,
circumstances that account for their somewhat loose and even, on the sur-
who commented on them. (By now, one finds some defenders of Derrida
160 * The Knowable and the Unknowable
and other targets of the Science Wars criticism among the latter as well.) In
may not be possible to say what Derrida's statement on the Einsteinian con-
whether it has a strict meaning. It is, however, possible to argue that this
mately involving the most radical limits of this concept, as developed in and
following Derrida's and de Man's work. First of all, however, we are deal-
ing here with the most elementary and traditional norms of reading, such as
semination of) the meaning of Derrida's statements, any more than of any
other statement.14 The overall case considered here offers a powerful, if dis-
tressing, illustration of this point. Nor is it possible to claim that any given
reading (for example, the one offered here) is definitive. This does not mean,
however, that one should not read with utmost care, rigor, and respect the
context within which a given statement is made, or that one cannot argue
about such readings, or that one can simply disregard traditional norms of
best theory and the best practice of deconstruction, such as those of Derrida
does argue, of course, that such protocols, even if scrupulously adhered to,
what happens in these types of cases and why, and Derrida and others offer
many subtle explanations of them. But this is quite different from endorsing
remains pertinent, even though some among these circumstances and con-
texts have been pointed out by a number of commentators and partly recon-
"But It Is Above All Not True" * 161
idea in itself, and indeed (I could not be more serious) the removal of all
other chapters would further help the book. The authors' comments on the
the book, as Derrida's letter, "Sokal et Bricmont ne sont pas serieux" (Sokal
and Bricmont are not serious), to Le Monde (20 November 1997) would
or the lack thereof, of Sokal and Bricmont's book.) The book certainly con-
was part of Sokal's hoax article, republished in the book. Sokal continued
to use the quote in his lectures, including in his interview with National Pub-
lic Radio, for quite a while after the circumstances of the statement,
sons for ultimately not having such a chapter, became known to him (8).
Given these facts, the authors' explanation concerning the omission of Der-
rida from the list of alleged "abusers" in the book and in several exchanges
Le Monde (12 December 1997). Upon the publication of the book, a num-
such a chapter should have been included after all or, at least, that the book
ated some further debate in the press. Sokal and Bricmont's "arguments"
the book, already indicated in the previous chapter, but are worth repeating:
the failure to understand the relationships between such work and modern
mathematics and science; the failure to read the relevant texts themselves in
between good and bad work in the humanities that they target; and, finally,
the failure to handle the key historical and philosophical aspects of mathe-
matics and science and sometimes to adequately represent the latter them-
selves. These problems are even more transparent and more disturbing in
Gross and Levitt's response to their critics in the second edition of their
book.
matter of the truth of such accusations: hence my chapter title here, which
clearly their point) he is in fact totally ignorant (79). Beyond being blatantly
untrue, "above all not true," this is hardly an innocent and, if believed,
the ethical and political effects of their interventions are far from
insignificant, and the damage done by these books is far from inconsequen-
tial. This damage, I contend here, also includes that done to the positive
works are and should not be seen as representing science and scientists or
their views of science and culture. The exposition of mathematics and sci-
ence themselves in these works hardly does much service, and sometimes
seen throughout this study, at stake here is at least the possibility of a cer-
enclosure, which is at least in part classical, that they share with a broader
philosophical field. As I said, in the present case, this can only be done by
relating Derrida's statement to "Structure, Sign, and Play" and his work in
general and then relating the latter to the epistemology and conceptual
of work takes place within and is often necessary for mathematics and sci-
tists and in certain technical texts. Indeed, it sometimes happens, as, again,
relations, that this work reaches its most radical philosophical limits in cer-
tain technical texts and by so doing indeed challenges both our intellectual
"But It Is Above All Not True" * 163
At the same time, however, one cannot in a case like this simply dissoci-
ate the conceptual substance and ethics, and one must consider both and
and
earlier included).17 The latter is one of my central points, and part of the
to be suggested here.
This is not to say that I am not about to offer conceptual views, argu-
ments, or claims here, more or less consistent with and supporting my over-
all argument in this study. My point instead is that some of these arguments
are framed in a particular way and that this framing must be taken into
val invariant under the Lorentz transformations. (I shall explain both con-
164 * The Knowable and the Unknowable
cepts later.) This argument would remain largely intact if one maintains
also argue, not unlimited. Indeed, in view of the fact that this argument is
concerned with the possible relationships between Derrida's work and the
ever was said (or not said) in the Hyppolite-Derrida exchange, it would
remain intact even if this phrase or related statements were in fact meaning-
less or uninterpretable. I do not think that they are; quite the contrary, I see
It is far more important to show that there exists a view of the relationships
and support for them. The conceptual content and significance of this
ments on relativity and other areas of mathematics and science, or his work
of Derrida and others should take place. Scholars in the humanities should,
of course, exercise due caution as to the claims they make about mathemat-
ics and science and should respect the areas of their specificity, which, as I
have also stressed, has not always been the case. Reciprocally, however, sci-
"But It Is Above All Not True" * 165
entists and other nonhumanist scholars should exercise similar due care and
dealing with innovative and complex work, such as that of Derrida, all the
more so if they want to be critical about it. Derrida, I would think, would
cians and scientists. However, no such criticism has been offered, at least
not yet. In order for this to happen, reading, in Blanchot's words, must
become a serious task for all of us, scientists and nonscientists alike. On
another occasion (in conjunction with the controversy surrounding his hon-
I would be content here with a classical answer, the most faithful to what I
respect the most in the university: it is better, and it is always more scientific,
to read and to make a pronouncement on what has been read and under-
stood. The most competent scientists and those most committed to research,
history and to processes which modify the frontiers and established norms of
their own discipline, in this way prompting them to ask other questions,
other types of question. I have never seen scientists reject in advance what
even if that encouraged them to modify their grounds and to question the
fundamental axioms of their discipline. I could quote here the numerous tes-
what the scientists you mention [in conjunction with the Cambridge incident]
are saying.19
Derrida may here have proved to be more generous to scientists than sub-
sequent events merited-that is, to some scientists, for as I have argued here
the views and discourses of Gross and Levitt, Sokal and Bricmont, and their
like do not and should not be seen as representing the views, work, and
ideas of science and scientists. In any event, one can find many testimonies
engagement with Derrida's and other recent philosophical work on the part
of scientists is possible, too, and we might yet see it. This engagement may
even help them to ask those "other questions, other types of question" that
mathematics and science are likely to bring us in any event and, by so doing,
perhaps even to make this philosophical work obsolete. For these questions
166 * The Knowable and the Unknowable
may well prove to be more radical-more radically classical and more radi-
philosophies.
Among the many accusations and complaints of Gross and Levitt's Higher
Superstition, those against Derrida have a special role and significance. This
the second edition of their book, they claim to the contrary (279-80)-
without, however, offering any textual support for this claim. Of course,
event. A reading that would make the relationships between Derrida's work
(or that of other nonclassical thinkers) and mathematics and science mean-
and Levitt's book, as it is to Sokal and Bricmont's. They do not even com-
arguably the most explicit and most famous reference of that type in Der-
rida.20 They only speak of its general abuse by "postmodernists" (78). They
and, it appears, from a secondary source (265n.10), but famous ever since:
matters can be obtained from the following quotation, which also gives one
some sense of how seriously to take such claims: "The Einsteinian constant is
not a constant, [is] not a center. It is the very concept of variability-it is,
finally, the concept of the game. In other words, it is not the concept of
field-but the very concept of the game." The "Einsteinian constant" is, of
course, c, the speed of light in vacuo, roughly 300 million meters per second.
"But It Is Above All Not True" * 167
Physicists, we can say with confidence, are not likely to be impressed by such
verbiage, and are hardly apt to revise their thinking about the constancy of c.
Rather, it is more probable that they will develop a certain disdain for schol-
ars, however eminent, who talk this way, and a corresponding disdain for
other scholars who propose to take such stuff seriously. Fortunately for Der-
rida, few scientists trouble to read him, while those academics who do are,
for the most part, so poorly versed in science that they have a hard time
lowing sections, I shall only say here that nothing can be more misleading
have pointed out, the contrary is in fact true. In truth, all of Gross and
spite of the fact that the unacceptable looseness of their treatment was
are quite simply not true.21 Much else can be said about their "represen-
tation" of Derrida in their book. But it is above all not true ("Mais ce
n'est surtout pas vrai"). Gross and Levitt are self-evidently not among
those "few scientists [who] trouble to read" Derrida. One cannot help but
impending danger of disdain and, the second edition adds, "scorn" on the
criticism of his work without reading it, "this is also extremely funny."
But he added: "The fact that this is also extremely funny doesn't detract
I would now like to consider Gross and Levitt's second main example of
case. In various other Derridean writings there are to be found, for example,
used without definition and without any contextual justification. Clearly, the
intention is to assure readers who recognize vaguely that the language derives
from contemporary science that Derrida is very much at home with its mys-
teries. (79)
Once again, none of these assertions is true. Indeed, their claim of "assur-
ance" (no less!) notwithstanding, they offer no other examples of such "por-
168 * The Knowable and the Unknowable
tion.23 At least some familiarity with his work and contemporary (and much
earlier) philosophy would be necessary under all conditions, and this famil-
iarity is manifestly absent (on both counts) in Gross and Levitt's book.
one can respond more seriously. It is of course also the kind of charge that
Sokal and Bricmont. So, on these grounds (of the ethics of discussion), this
type of accusation requires a response. Gross and Levitt make much of their
We cannot resist the impulse to point out that in Derrida's usage the word
utive for a summer actuarial job he was asked: "What kind of mathematics
are you interested in?" "Topology," he replied. "Well, we don't have too
ties will no doubt wish to point out that topos (pl.: topoi) is a recognized
term within literary theory for a rhetorical or narrative theme, figure, gesture,
textual topoi. One's suspicions are reignited, however, when the term differ-
used to denote the study of the topological aspects of objects called "differ-
ingly (since Levitt is a topologist) uninteresting and will hardly give the
reader unfamiliar with the subject a real sense of this extraordinary branch
given their "roughly speaking." For one thing, not all two-dimensional sur-
faces are smooth. The relationships between smooth (differential) and not
(Roughly speaking, differential manifolds are the ones on which some form
surfaces are themselves differential manifolds; so one need not restrict the
"But It Is Above All Not True" * 169
matter, except, again, to the degree that this remark and their other com-
ments on science can become a source for a kind of confusion for scholars
in the humanities that should indeed be avoided. I also leave aside the banal
topology versus topography joke, and the overall inappropriate and unpro-
It is more difficult to leave aside the fact that Acts of Literature consists
edited, and the index compiled, by someone else. Moreover, Gross and
Levitt clearly did not check their own references to the index. The statement
that "in Derrida's usage the word topology seems to be virtually synony-
are related or used in similar contexts rather than are identical. Once one
checks the text, one sees that all these terms refer to a general sense of
"topos" as spatiality, which, as even Gross and Levitt admit, need not entail
matics the term" topology" carries a double meaning of the discipline and
ment itself, from his essay on Kafka, "Before the Law" (Devant la Loi):
guardian, within the polarity of high and low, far and near (fort/da), now
and later. The same topology without its own place, the same atopology
[atopique], the same madness defers the law as the nothing that forbids itself
needs to know and to read both Kafka's and Derrida's texts to make sense
of it. It deals, among other things, with the difficulties of applying any spa-
tial or topological concepts and metaphors to the domain where law oper-
ates as conceived in Kafka's work. This knowledge and a real act of reading
differential topology here. One can easily see, however, that Derrida says
ence is not audible in the spoken French, but it is visible on the page. This
170 * The Knowable and the Unknowable
was one of the reasons why Derrida introduced his neographism. In this
sense Gross and Levitt's mistake is ironic, and it unwittingly proves Der-
rida's point. When Derrida, in fact, uses topography a bit earlier in the same
of the "space" or/as "non-space" of the law in Kafka. This is why the edi-
tors list topography in the index. Primarily in question, however, is the con-
thus a double irony here.) Derrida does not even say "topology" here,
ously (it should be clear by now) in the general and perhaps quasi-mathe-
tial space or place, a certain topos or atopos, or atoposness. The French for
his field is topology and that the French is provided here, it is (or rather it
not pay attention to or did not bother to check this point. It is also inexcus-
able, given that his aim was to attack publicly Derrida's misuse of scientific
terms. In such circumstances one is obliged to treat with care even what one
perception of it as junk.
here in de Man's terms, which, as we have seen, entail the analogous "topol-
Kafka.) Any attempt to configure or even conceive of the "space" of law qua
space is bound to fail. We always fail if we think of the space of law, spatial
("before," devant) of, or at the gate to which (as in Kafka's parable), we can
cannot postulate that law, or more accurately that which produces the effect
"But It Is Above All Not True" * 171
of law, can be ascribed a spatial (or temporal, or any other) nature or struc-
ture, properties, and so forth. All of these, however, may in turn emerge as
effects of the "same" efficacity, that is (since the latter is never strictly the
can we postulate that this efficacity of law would exist as inaccessible but
ble from its effects and yet irreducibly inaccessible, as considered earlier. We
can infer the existence of this efficacity on the basis of the sum total of cer-
tain phenomena, which may also have particular topological, spatial effects.
the way Lacan invokes them (on this occasion perhaps in a more conven-
analytic situations. Something like the Moebius strip or the Klein bottle or
junction with Kafka, who may even have been familiar with them. They
then transferred to the "topology," the space, of the text and our relation to
it (the law of reading, as it were). Or rather (and this is itself a key point of
ogy" and epistemology of the case and, hence, also of literature, reading,
and law.
related throughout the essay and which bears essentially on other works just
mentioned, as well as "Structure, Sign, and Play," and on the possible con-
nections between the latter essay and relativity.25 This is why Derrida speaks
Plato's chora is read by Derrida as, at least, anticipating and (whether Plato
saw it this way or not) entailing what we may call the differance of space
and spatiality itself. That is, in accordance with the nonclassical view, chora
seen as the "site" of the spatial, while not being spatial "itself," or, again,
Ultimately, differance (it is, again, never the same differance) would be
space. As will be seen in detail later, these efficacious dynamics have con-
siderable implications for the relationships between Derrida's work and rel-
the connections that enable one to link Derrida's "Structure, Sign, and Play"
to relativity and (with qualifications given earlier) give meaning to the Hyp-
For the moment, my point is that, not unlike (but, again, not identically
topology. Indeed, one can say (with more caution and qualification) that it
from Leibniz, Gauss, and Riemann to Poincare and then into our own time,
extends from the work and ideas of these thinkers. From this perspective,
which, obviously, does not diminish the problem of Gross as Levitt's treat-
ment of Derrida here. Thus, on the one hand, there is no "differential topol-
ogy" or even a reference to it in the disciplinary sense of the term, and one
could indeed end (and some commentators did) one's argument against
"But It Is Above All Not True" * 173
Gross and Levitt's "criticism" of Derrida there. On the other hand, how-
ever, a stronger argument is that there are in fact deep conceptual and epis-
these connections, beyond their significance in their own right, may even
help us to illuminate and better understand each and possibly might also
Gross and Levitt did offer a "clarification" (if such is the word) in the sec-
ond edition of their book. This clarification, however, hardly requires alter-
It has ... been pointed out, quite correctly, that the term differential topol-
ogy does not appear in Derrida's Acts of Literature; what shows up is differ-
then, that Derrida did not authorize the translator's locution (but remember,
the Wolin affair demonstrated that he can be very finicky about translation
of his writings). The honor of the coinage then goes to translator Derek
often used by Sokal, with Bricmont and on his own, as well, is hardly a sub-
sis on the Wolin affair is irrelevant, if not meaningless.) Gross and Levitt fail
to mention that, as we have seen, the French original is given in the transla-
topology in mind. Either way it would not help Gross and Levitt's case
for example, along the lines considered here, it would not make Gross and
Levitt's criticism any more meaningful. In any event, one would have to
read Derrida's text to understand the phrase. To say, therefore, "we think
the point [made in their book] still stands, even if Attridge, rather than Der-
rida, is responsible for the example" (293) is scarcely more compelling than
The statement that bridges the two statements just cited is even less com-
174 * The Knowable and the Unknowable
pelling. Gross and Levitt say: "In any case, there has been an enormous
amount of comment from our critics on this one specific matter, which,
impression of rigor and congruity with 'cutting edge' science" (293; empha-
sis added). The statements cited there would clearly show (and would make
it difficult to forget) that this "specific matter" was nothing of the kind, the
illustration was not casual, and the point was not incidental. It certainly was
(and still is) a major point for Gross and Levitt, as it is for Sokal and Bric-
mont. Besides, this last claim in the passage just cited is still not true, "above
While, thus, Gross and Levitt accuse Derrida of "using" the term "dif-
work. The same can be argued concerning their treatment of the work of
many others whom they criticize in their book, whether these works, as
such, indeed deserve criticism along these lines (as some do) or not. Schol-
ogy, found in the immediate vicinity of just about every point of Higher
We all make mistakes and sometimes make absurd mistakes. Most crucial
here are the intellectually and scholarly inadmissible practices and attitudes
that pervade this irresponsible book. Gross and Levitt's warning concerning
"threats to the essential grace and comity of scholarship and the academic
life" (ix) becomes, in one of many bizarre ironies of this bizarre book, its
their book itself than of most of the works and ideas they criticize. (I have
not encountered this observation but would be surprised if others had not
made it as well.)
To be sure, and one must acknowledge this without false pretense or hes-
Sokal and Bricmont are at best embarrassing, as are, one might add, many
question. They offer the editors of scholarly journals and university presses
no more excuse for publishing them than Higher Superstition, The House
although one can think of some good "excuses," not excluding trusting in
the opinion of the present author, however, that Sokal's hoax article would
have probably been published somewhere in any event. The reasons for this
the humanities. One need not know any mathematics and science to reject
rida and other authors it comments on, and of course a minimally careful
back to the author as unacceptably incompetent, even if one did not per-
ceive the hoax nature of the piece. If this is what Sokal's hoax aimed to
prove, then it has achieved its goal. Hardly a revelation, however. There is
always some bad work in any field, including mathematics and science. The
problem is that, as I said, Gross and Levitt or Sokal and Bricmont are (by
virtue of their lack of expertise, or even superficial familiarity with the texts
nate between good and bad work.27 If they were, very different books
would ensue. The comedy of these books is that they say the worst things
about some of the best work and accept and sometimes praise-and draw
on-some of the worst. The tragedy is that some scientists, including some
among the best scientists, have taken it seriously and accepted its arguments
Lacan, or Deleuze, where Gross and Levitt or Sokal and Bricmont see abuse
there is no real abuse, even though one cannot deny that errors do occur.
This science and this role, however, the critics in question do not see, in
part, again, because they are entirely unequipped intellectually and philo-
sophically to read such texts in general. For, as I argue here, one can, in
principle, decouple mathematics and science from such authors, say Godel's
(their ideas, idiom, rhetorical and textual strategies and moves, and so
crude way of Gross and Levitt or Sokal and Bricmont. Derrida's, Deleuze's
176 * The Knowable and the Unknowable
ical ideas is not mathematics in its disciplinary or technical sense and cannot
this perspective, the whole criticism of these authors by Gross and Levitt or
Given the egregious nature of some of Gross and Levitt's errors it is sur-
prising that they were not discussed by reviewers immediately upon the pub-
lication of the book or indeed were not noticed before the book was pub-
extraordinary harm done by the book, that is, assuming that it should have
been published to begin with, given its flaws, which are unredeemable
regardless of the problems one might have with the authors discussed there.
The decision, even more unfortunate, to issue the second edition poses fur-
ther questions in this respect. By now some of these problems have been
argument here, often critics and even defenders of, say, Derrida still think
worst they are seen as inept or senseless) rather than understood in the con-
text of the relationships between philosophy and science.29 The same (with
suitable modifications) goes for other figures. The very critique of Gross and
Levitt's book often amounts to a "yes, but... ," such as "they got a few (or
even not so few) things wrong, but..."-not the kind of change of attitude
Nor, in my view, have these books, or Sokal's hoax, had as much value
have done so, of course. There are, however, better, much better, ways to
tionships between mathematics and science and the humanities (or the
social sciences), again with a very different set of intellectual attitudes and
lier, one would do well to argue that these works are taken far too seriously,
that is, again, as works rather than symptoms, are given far too much space,
and are trusted by too many scientists (or indeed nonscientists). Once one
emerge a very different sense of their work and of the relationships between
this work and mathematics and science. The true richness and complexity of
these relationships and the true possibility of reading these texts, including
"But It Is Above All Not True" * 177
The Einsteinian constant is not a constant, is not a center. It is the very con-
cept of variability-it is, finally, the concept of the game [jeu]. In other
observer could master the field-but the very concept of the game which,
To begin with, Derrida's last sentence, "which, after all, I was trying to elab-
because it indicates that the term "game" or "play" (in this context a better
translation of the French jeu, which carries both meanings) has a very
proper "editing" (of the type I suggested for Lacan in chapter 3) of Derrida's
deviating too much from their meaning (at least a possible and even plausi-
ble meaning), albeit at the expense of their great value for the Science Wars
"game" did not appear here, and Derrida had merely said "variability," the
statement would never have surfaced in the Science Wars debates, which
cept in the human sciences and its deconstructive potential, in view of this
differance (as well as related elements of Derrida's matrix). At the very least,
178 * The Knowable and the Unknowable
in his essay and understanding what Hyppolite and Derrida mean by "the
The accuracy of quotations from Derrida and other figures by their crit-
ics has been often stressed by scientists and others who welcomed and sym-
ever, even when they are accurate, their literal accuracy is meaningless if the
reader is not provided with the meanings of the terms involved (such as
establishing such meanings; or is free to construe them on the basis of, say,
famous c, the speed of light in a vacuum (as Gross and Levitt claim), if Hyp-
polite meant something else by it?32 It is true that Hyppolite's meaning may
uum, which is not surprising, given that this constancy defines (special) rel-
correlated with his concepts of variability and play, the concepts specifically
Derrida in his remark (a fact ignored by the Science Wars critics), indicates
(given that he refers to his lecture) is nothing of the kind. I shall return to the
terms of his essay, which the readings or unreadings in question lack, even
though there are, again, clear indications in his comment that such an expla-
Steven Weinberg's in "Sokal's Hoax," "I have no idea what this is intended
debates in question.
acknowledges that he did not initially pay much attention to the meaning of
Derrida's key terms and gives some consideration to the context of Der-
rity of Derrida's terms 'center' and 'game,'" and that he "was willing to
then says, "What bothered me was his phrase 'the Einsteinian constant,'
which I have never met in my work as a physicist" (56). It would seem nat-
ural to check what Derrida's "terms of art" could have meant, especially
sible meaning for the phrase and then to offer some comments on Derrida's
lite's own description of the (Einsteinian) "constant." Nor does he offer any
even risk a claim that some of Weinberg's own work on quantum field the-
ory (which brought him his Nobel Prize) carries in it certain philosophical
implications that are not that far away from the (nonclassical) epistemology
idea of decentering and play, and it is not surprising that this quotation was
"not much help" to Weinberg (56). The passage that Weinberg cites occurs
in the introductory portion of the essay, as part of the discussion of the joint
the event which I wish to mark and to define," indicates that Derrida is
play" emerges later in the essay, although a few sentences following the one
cited by Weinberg already give one a better sense of Derrida's ideas con-
The function of this center was not only to orient, balance, and organize the
all to make sure that the organizing principle of the structure would limit
what we might call the play of the structure. By orienting and organizing the
coherence of the system, the center of a structure permits the play of its ele-
ments inside the total form. And even today the notion of structure lacking
In short, those unfamiliar with Derrida's ideas would need a more exten-
terms. Therefore, more patience and caution may also be necessary before
me Derrida in context is even worse than Derrida out of context."34 The con-
texts and concepts at issue may well not be sufficiently familiar to most sci-
entists for them to be able to offer the kind of reading of Derrida's statement
ideas and contexts or have any obligation to engage them in any way. It is
irrelevant) one of the most open to radical and innovative theories in physics
itself-or many other physicists who commented on the subject. One might
the scientific community not only cite their comments out of context but vir-
following an oral presentation of his essay. The essay does not mention rel-
ativity, and the statement itself makes no substantive scientific claims. Rela-
Play." The conclusions may of course be different from those reached by the
cumstances that lead to them remain significant when such exchanges are
made clear by the editors of the volume.36 It is true that such statements are
and it is doubtful that it was done here; indeed it is virtually certain that it
was not. Hyppolite died before the volume (eventually dedicated to his
memory) went into production and did not even have a chance to edit his
own contribution, let alone his exchange with Derrida. One might, accord-
ingly, argue (and some have) that such improvised statements should be dis-
the interpretive problems that they pose, such statements and exchanges are
ent. I argue that the circumstances of these statements must be given special
however. First of all, they were far "too little, too late"-after several years
of relentless abuse, beginning with Gross and Levitt's book, and, as I have
means. Second, they do not appear to signal much change in the overall hos-
toward the work of Derrida and other figures on the part of the scientists
appear.)
With Einstein, for example, we see the end of a kind of privilege of empiric
menters who live the experience, but which, in a way, dominates the whole
construct; and this notion of the constant-is this the center [i.e., would it be,
given the improvised, tentative, and probing nature of his comments. It can
be read as compatible with special relativity, in particular the idea that the
distinction between space and time depends on the observer. Certain state-
on a specific reference frame, since the sequence can be reversed if seen from
Hyppolite does not actually use the phrase "the Einsteinian constant" as
lite's remark, rather than to any accepted scientific term, and is, in this
here may not mean-and to the present author does not appear to mean
tion of space [and] time" and something that "does not belong to any of the
the whole construct." ("The whole construct" here refers to the conceptual
It is also possible that Hyppolite has in mind this latter interpretation, while
mannian spaces) emerge along with their difference from classical spaces,
crucial, since both these notions are correlative, and both are correlative to
"But It Is Above All Not True" * 183
the constancy of the speed of light c in a vacuum and its independence of the
state of motion of the source. Also, both "constants" equally reflect key fea-
offered later) relativity, appear to make Derrida see the "constant" in ques-
tion as "not a constant," but as (that is, as reflecting) "the very concept of
variability" and, "finally, the concept of the [play]" (in Derrida's sense).
These features are most fundamentally at stake in Derrida's essay and led
It follows that one can read c in this way as well (all three readings are cor-
view the two conceptual readings of the Einsteinian constant just indicated
that can be established with more certainty from broader contexts (such as
tion may be possible, given the status of the text as considered earlier. At the
same time, interpretations of these statements are possible and may be nec-
essary, at least in the broader sense of relating them to the content of Der-
rida's essay. These statements have been interpreted without any considera-
them out to the degree possible, and to give these statements the most sensi-
observer could master the field, that is, the whole of space-time (if such a
standing, arising from a reading of Derrida's essay (such a reading is, again,
stant" are more conjectural, the meaning I suggest for Derrida's term "play"
(jeu) is easily supportable on the basis of his essay and related works. So is,
184 * The Knowable and the Unknowable
physics, the space-time of special, and even more so of general, relativity dis-
events. These two notions are, of course, not the same, and my "or" also
means "and" here. Both are disallowed, but the difference between them is
return to this question later. My point at the moment is that the broader
concerned) and, in its more radical implications, to Leibniz and his critique
motion of either the source or the observer. The "constancy," that is, invari-
this form by Minkowski. These considerations also led him to the concept
tivity theory, his theory of gravity. (I bypass the explanation of these more
technical terms, since they are not essential for my main point-the decen-
Hyppolite had in mind precisely this concept. The Einsteinian (concept of)
mind here. Hyppolite, of course, might have had space-time itself in mind as
time." (The special circumstances of both the exchange itself and of its tran-
scription, translation, and editing appear to have played a role here, how-
ory. Or, more accurately, "two Einsteinian constants," space-time itself and
the Lorentz distance, while both correlative to the constancy of the speed of
light and to each other, allow for two conceptually equivalent frameworks
of relativity.
Thus, while I do think that the conceptual (rather than numerical) inter-
pretations that I suggest are "more plausible," I do not think that reading
criticizing those who used c in the Science Wars debates, I do not mean to
first, that other interpretations are also possible and perhaps more likely, a
relativity and/as a decentered play, a concept that easily tolerates this read-
ing. Accordingly, if some, such as Richard Crew (in his exchange with the
present author), are right in arguing that c makes more or even most sense
would not change my overall argument.40 As I have stressed from the out-
set, my overall conceptual, let alone ethical, argument does not fundamen-
ence between the terms "constant" and "invariant" are justified, especially
val, invariant under the Lorentz transformations, could have been assimi-
"the play of the world," as opposed to play in the world. This shift indicates
which whatever "play" of events takes place. Instead we deal with dif-
ferance and play in which such effects as space and time, or space-time, or
rida posits a certain irreducible variability of the world itself and/as our con-
ground of events given once and for all, such as Newton's absolute space
(with absolute time) in classical physics. This idea was questioned already
sophically (his physics is another matter) not that far from Einstein's or
correlatively to the fact that in general relativity space can be seen as flat-
become equivalent) the curvature of the space measures gravity. The vari-
ability and "the play of the world" (in the present sense) are, however, not
only retained but enhanced as a result. One can see such spaces, manifolds,
the neighborhood of each point and transitions from one such neighbor-
That is, we deal here with nonclassical efficacity of the conceptuality or con-
forth-of the spatial (in mathematics and physics) or (in physics) also the
concept of chora proceed along these lines as well. This is, then, how dif-
ferance and its differantial (a)topology may be read, let us say, again, quasi-
can be extended to relativity, by virtue of the elements the latter itself shares
theory. According to the latter, space and time are not given independently
more complex way) theories. Instead, space and time, in any way we can
observation (and, again, in more complex way, of our theories) and indeed
key subjects of Derrida's essay (especially if read with the companion analy-
between that which is observed and that by means of which one observes.43
and so forth), or those of "the discourse of the human sciences," are posi-
tioned accordingly, and are all linked, at several levels, to (the concepts of)
siders Levi-Strauss in the same set of contexts. This dynamic (in general and
more complex and radical form in quantum physics and its epistemology. In
nounced, once one deals with more exotic gravitational phenomena, such as
neutron stars or black holes. Also, as the preceding analysis makes clear,
which were decisive for all of Bohr's epistemological ideas, were crucially
the spatiality of space and the temporality of time-as that which makes
time and temporal aspects of space-that is, they enable us to see time in
nonclassical perspective and indeed link both. As I have suggested from the
and its extension to the work of other authors discussed here, we may speak
rigorous deconstruction of, among others, the concepts of space and time
ential topology, and related fields), extending from Gauss and Riemann to
our own time, and modern physics, such as relativity, especially general the-
Einstein's relativity and other key areas of modern physics, such as quantum
these remarks can be read as consistent or, again, congruent with the philo-
"But It Is Above All Not True" * 189
appears to have sensed and suggested here is that part of the conceptual con-
tent of Einstein's relativity with its space-time may serve as a kind of model
for the Derridean concept of (de)centered play and related ideas. This sug-
gestion is neither surprising nor especially difficult for anyone who has read
Derrida's essay and has some knowledge of certain key ideas of relativity.
Derrida responds more or less positively but suggests that one needs a more
Derrida) to suggest.
tions, rather than questions of physics, and both Hyppolite's and Derrida's
here:
tion. But perhaps these dangers are averted if we restrict ourself to taking
This is not a quotation one is likely to find in recent critics of Deleuze in the
scientific community. I do not think, however, one can read Deleuze on sci-
ence without keeping this statement in mind (it is of course congruent with
and indeed correlative to his other statements on science cited here). Such
what these critics, and in particular Sokal and Bricmont, say about Deleuze,
could be easily shown that the same cautious attitude toward and utmost
respect for mathematics and science manifestly defines most of the "targets"
reflects concepts, including those of Hyppolite and Derrida, which are any-
190 * The Knowable and the Unknowable
are thought through by Derrida in the most rigorous way. Even more
significantly, they reflect and, at least in part, derive from the philosophical
questions arising in modern science itself. Here I would think that one is
introducing his question, Hyppolite says that "we have a great deal to learn
from modern science."46 And perhaps science has a few things to learn from
The very question of how casual such "casual" remarks are, or can be,
so forth and dismiss them on these grounds. As I have indicated, this type of
argument was advanced by some in defending Derrida and has even been to
Derrida here, this is a weak defense, that, I think, does not pay sufficient
work of these figures, whether specifically on science (or as concerns the rela-
tionships between their work and science) or in general, would be more inter-
esting and productive than any defense. The best position for making such a
ing of the relationships between modern mathematics and science and the
matics and science, are delivered to their projects via other trajectories (direct
ativity or other areas of mathematics and science, and even their misrepre-
ered, even when the thinkers in question do not offer such a consideration
cal, and that the trajectories involved are multiply entangled, although they
The possibility of such an argument in the present case should not be sur-
ever, leaving aside their general erudition, both have been the readers and
issues in and implications of relativity and are cited by many experts in the
or at least a crucial point in the history opened by this debate, which can in
niz's ideas on these subjects are not that far from those of Einstein (and one
might even argue that philosophically they go further). This may be espe-
cially said about Leibniz's critique of Newton's absolute space and absolute
time. Equally crucial is the accompanying idea that space and time (absolute
space and absolute time being no longer possible), rather than being a fixed
between such events. As we have seen, this idea opens a history of major
and time have crucial connections to both Kant's and Hegel's philosophy,
earlier philosophical thought from and before Plato on. Hyppolite's own
evant to some of his comments in his exchange with Derrida. Derrida's ideas
phy one finds even more pronounced links between Leibniz and relativity,
ics and science, which he headed for ten years (1954-63), Hyppolite had
the subject. He was previously a chair at the Sorbonne and a professor at the
est throughout his life. It is worth mentioning in this context that Hyppolite
was granted admission to the Ecole Normale on the basis of his ability in
philosophy and mathematics. Derrida, too, spent years of his career at the
Ecole Normale, first as a student (of, among others, Hyppolite) and then as
a professor, and had similar access to key ideas of modern mathematics and
science, although his background and interests in science are less extensive
ical affinities between relativity and Derrida's ideas in "Structure, Sign, and
Play" and beyond-if, once again, these affinities are indeed independent,
given the history just indicated. Einstein's relativity dates to 1905 and has
ematics and science) in relation to Derrida's essay is, then, not only
and arise from, and are given their significance by, an extraordinarily rich
work shapes the relationships between the work of Derrida (or that of
Deleuze and other authors mentioned here) and modern mathematics and
science. All this is entirely missed by their recent critics in the scientific
It is, then, hardly accidental either that Hyppolite, whose thought was
"But It Is Above All Not True" * 193
shaped by this type of understanding, invokes next still more radical con-
(at least it can be read in this way), to quantum physics and then, overtly, to
argues that we might yet need to rethink structuralism from the perspective
baki project.48 Andre Weil, one of the great mathematicians of this century
and a founding member of the Bourbaki group (and the brother of Simone
Kinship.49 Both Derrida and Hyppolite (or Serres) must have been aware of
Weil's article and might have been familiar with it. Derrida's "Structure,
Levi-Strauss and structuralism.s50 The essay does not consider this mathe-
from them, and some of these connections emerge more explicitly in other
essays in Writing and Difference and elsewhere in Derrida. At the very least,
It is, therefore, possible and even likely that Derrida has this point in mind;
194 * The Knowable and the Unknowable
and it is certainly a major question posed by "Structure, Sign, and Play," and
tual center (if any) of a theoretical structure, such as that of the relativity the-
ory, is, of course, quite different from invariance of a physical constant. One
even becomes a model for, the idea of decentering the overall conceptual
centering (centers are seen as local and multiple, but are not eliminated) and
Given Derrida's argument in "Structure, Sign, and Play" and the Der-
relativity could have primarily been, and would naturally have been, on
the speed of light c, which may indeed have been what Derrida had in mind
after all. Hyppolite might have also thought of the center in the sense of
stant" (however the latter itself is interpreted), when he spoke of the con-
these levels of "(de)centering" is possible, and these levels may mirror each
other, if one should speak of conflation here. Given that at stake is a gen-
the Derridean decentering or, better, "play." Both relativity and "play" in
Derrida's sense entail the great complexity of the relationships between cen-
"But It Is Above All Not True" * 195
centralization.
tualized via or metaphorized by the former? These are some among the
Derrida's concept of play and other ideas in "Structure, Sign, and Play."
ity, even if the philosophers of science may not address them in the form
that Derrida's work would suggest. On occasion, however, one does find
formulations in their work that are close to those that would emerge were
the exchange and "Structure, Sign, and Play," Richard Crew (in the
Crew's discussion may, even if, again, against his aims, also be seen as
Sign, and Play," concerning the relationships between continuity and rup-
ture in the reciprocal history of the concepts of structure and play. While, as
Crew correctly points out (par. 11), Derrida does speak of and consider a
rida's critique throughout his work, along with absolute origins, ends, and
so forth. This is suggested even by the way in which the very term "rupture"
event be then? Its exterior form would be that of a rupture and redoubling"
ocal rupture, at most something whose exterior form appears as and has the
either, or again, anything to the efficacity of these effects.55 A very long his-
tory of continuities and ruptures around the idea of structure (or again,
play) is at stake, as the second paragraph of "Structure, Sign, and Play" sug-
gests.56 From this perspective, one should expect some intimations, antici-
pations, and so forth of the "event" at issue, sometimes radical and striking,
even before Newton), and there are much earlier cases, for example, in Plato
Play" were likely of that nature.57 And yet, at the same time, the kind of
suggests does not come into the foreground in any significant measure
around Nietzsche's time.58 Derrida's key concept of "the play of the world,"
not the heyday of classical mechanics" (par. 9). The point is, however, that
10). (Let us assume for the moment that these possibilities are viable even if
tion, along the lines discussed in chapter 1.) As I have indicated, there are
indeed reasons for doing so, since already with Galileo and then Leibniz cer-
tain key features necessary here were in place. However, the history of clas-
sical mechanics in this respect is for the most part a history of physically
Maxwell on. So Derrida's genealogy is on the mark and not at all in conflict
One can, thus, read Derrida's comment in the following way (which, as
I shall explain presently, is not the same as finding out what Derrida meant
or even what he might have meant by his statement at the time it was
level of the conceptual structure of Einstein's theory itself. Both may be seen
as mirroring each other in this respect, with the qualifications indicated ear-
lier and keeping in mind the rigorous specificity of the mathematics and
stant" itself is left open, although it cannot be seen as arbitrary (and the
sense, as the effects of differance. The former signals the latter: that is, the
quasi-Derridean play.
This might or would have been an intriguing insight or a lucky guess (or
broader nonclassical field outlined earlier, rather than his knowledge of rel-
ativity or related mathematics, from Gauss and Riemann on. This is what
Derrida's remark in fact or in effect performs. That is, it might be that, pro-
not-that is, if this is not what he meant, or even could have possibly
meant-this is still how this statement may be read, in view of the nonclas-
floating terms and ideas, which may be read in the way just described. Even
thereby established allows one to interpret it in the way suggested here. This
orous-that is, the statement can be given a rigorous meaning on the basis
of both Derrida's text, specifically "Structure, Sign, and Play," and relativ-
ity. As I said, if this was what Derrida had in mind, it was likely only a lucky
guess, although such guesses are never purely a matter of luck, a dynamics
throughout his work. At the very least, the history of Derrida's ideas has
198 * The Knowable and the Unknowable
well as certain areas of modern biology and genetics, especially in their non-
classical aspects. As I argue here, in all of these theories one finds the philo-
whom he confronted in his work. One also finds related problematics else-
the philosophy of relativity. One must take responsibility for this reading
and claims, be they ultimately right and wrong, since Derrida makes none of
have argued here, more radical ideas than those found anywhere in contem-
part of the credit will still have to go to Derrida. If I am wrong, the fault, as
work in any substantive way) and gives only a brief improvised response to
in mind concerning relativity and his concept of play, quite possibly some-
thing vague and induced by Hyppolite's question. What that was is not eas-
ily, if at all, recoverable after thirty years. In any event, while, as some, such
as Crew or indeed Derrida himself, have pointed out, Derrida might not
indeed nearly everyone, on both sides of the Science Wars, admits at this
point, including (however lamely) Sokal, although not Gross and Levitt. In
short, there is little ground for any meaningful criticism, and certainly not
for the kind of attack to which Derrida was subjected during the Science
Wars. At most, if one is inclined to criticize, one can see Derrida's remarks
as irrelevant as far as relativity is concerned. In this latter case, one has from
matics and science, and the relationships between them and his own work.60
If, however, one does want to suggest connections between Derrida's work
and relativity or to establish them rigorously, one is free to do so. One has
the very least, this suggestion can be put on the table, and it would, I think,
Deconstructions
-Werner Heisenberg
As must be apparent from the preceding chapters and as this chapter will
nor concepts. Nor is it always easy to relate effectively some among these
strata to the relevant mathematical and scientific ideas. This is in part why
negotiations, let alone building bridges, between the fields and cultures
limits at certain points, at which the difference between the sides involved
(there may be and, in a certain sense, always are more than two) becomes
sary for a meaningful commentary? Some humanists do, and there is some
ground for this argument. As I have suggested from the outset of this study
in more general terms, I would not see the situation as altogether symmetri-
erature (or art) and mathematics and science. There are also significant dif-
ferences between the latter as well, which may, however, be less germane for
and were certainly able to read philosophy, and the work of some of them,
or in any event some of their work, can be read as philosophy, and is so read
in this study. It would be more difficult to find the reverse cases. Leaving the
Greeks aside, both Descartes and Leibniz, or differently, Galileo (but not
appears to the present author to be primarily the former, but one would
202 * The Knowable and the Unknowable
want to be cautious here. It may be added that the problem is not only and
not so much the lack of the necessary expertise in, say, Deleuze or Derrida,
on the part of their scientific critics, but their lack of even a minimal engage-
ment with the texts they criticize. The latter would easily enable them to
avoid the more embarrassing errors and ethical mistreatments of the works
that nonclassical thought can help us in these negotiations and bridge build-
together rather than hindering such efforts, as is often argued by its oppo-
nents. It does so at the very least in view of the nonclassical features found
cal thinkers in mathematics and science and elsewhere are especially sensi-
tive, while their scientific critics often miss or refuse to confront the non-
classical for the reasons that I have discussed here beginning with the
I further argue that, even if and to the degree that nonclassical authors
(speaking of those of them who are not mathematicians and scientists) are
latter. At the very least, sometimes through deliberate effort (as in Deleuze),
tion or of course in literature and art. But then, mathematics and science
can ever strictly delimit the latter (not an easy assumption, if possible at all).
we make concerning their findings. In other words, while Deleuze and Der-
about mathematics and science. Bohr, of course, both does physics, experi-
physics, too, aptly called it, and also tells us a great deal about it, as does
Heisenberg.3
mathematics and science or about their role in intellectual history and cul-
cal works. (The humanists might want to and, I would urge, should some-
times consult the latter as well, which is not as impossible as it may seem.)
ematicians and scientists cited in this study would demonstrate. One cer-
or, on the (more) classical side, Einstein and Schrodinger, which are espe-
cially crucial for this study. A list of such examples would be long and
would include a great many of the major figures whose work and ideas
and literature and art, and an even greater number of lesser-known figures.4
These works, however, are something altogether different from what one
finds in Gross and Levitt, Sokal and Bricmont, and several related writings.
science in the work of the Science Wars figures in question suffers from
problems that are not altogether unlike, or complicit with, those they
bemoan in the works in the humanities that they criticize. I do speak here of
where the problematic character of their treatment and the symmetry just
indicated are much more pronounced and have been extensively commented
not been much, if at all, considered). By "complicit" I mean that the prob-
lems in the exposition of science that one finds in Gross and Levitt (and
matics and science by the humanists. This exposition also contributes to the
especially on those humanists or intellectuals (there are some) who are, for
with the lucidity and precision necessary for nonspecialists, who want to
This discussion is also philosophically central to their book. Far from being
itself and of the debates concerning it, and I shall consider it in some detail
The second section, the final section of this study (a short conclusion is to
ern mathematics and science. In this case, I shall be concerned with the rela-
Derrida's work and relativity in the preceding chapter. I shall, however, pur-
sue this argument via Heisenberg's, rather than Bohr's, work and via
shall also argue that this deconstruction is in fact, or in effect, essentially the
same as Derrida's, which allows one to remove the quotation marks around
Derrida who read Heisenberg's 1929 text, which is factually unlikely but
reading may have been. Even though Derrida does not appear ever to refer
another, for example, via, among others, such figures as Bataille (for whom,
source), Blanchot, and Lacan, with whose work he has engaged extensively.
Wars, rarely bring additional clarity and often increase the confusion. The
sible, but is hardly sufficient, to justify the poor quality of these discussions,
especially on the part of the scientists involved. Consider Sokal and Bric-
Science Wars debates. I leave aside their hazy sense of the philosophical idea
of linearity (or of the philosophical aspects of the idea of the line, mathe-
physics are confusing and misleading as well. Thus, they say, "one may
speak of a linear function (or equation): for example, the functions f(x) = 2x
and f(x) = -17x are linear, while the functions f(x) = x2 or f(x) = sin x are
is indeed linear (and this fact is indeed crucial to quantum mechanics and its
ing to think about it, let alone of the so-called v-function that satisfies it, in
the terms just indicated by Sokal and Bricmont. This function, which bears
significantly on the nature of quantum theory, is itself not linear. The lin-
with respect to the functions to which it applies. That is, if you apply an
dealing with vectors in a Hilbert space, which is associated with the quan-
unless one has a substantial (at least for a non-physicist) substantive knowl-
that ordinary linear functions, such as those mentioned by Sokal and Bric-
mont, behave in the same way (obey the same linearity rules) with respect to
account for the experimental data of quantum mechanics are linear. The
task for "calling quantum theory nonlinear, though it is the only known
206 * The Knowable and the Unknowable
altogether correct even in physics, but a small matter here. Markley may or
may not be confused as concerns some of these questions. If he is, his con-
This is still leaving aside that some quantum field theories (for example, the
cerns different views of these matters even among physicists. These facts,
once not properly sorted (as they are often not in popular or even technical
ties. But Weinberg does not much clarify these matters either, as he could
and, I would argue, should have. Indeed, to some degree he even adds to the
mechanics may not account for the behavior of its objects in the same way
classical physics does, even though, as we have seen earlier and as Weinberg
this difference clear on this very occasion, albeit in part against the grain of
his argument.
In their long note on quantum mechanics, then, Gross and Levitt offer
The assertion that deterministic causality is still viable within the phenome-
nal world of quantum mechanics may come as a surprise to people far better
Diirr, Goldstein, and Zanghi [hereafter DGZ] shows that a large part of clas-
deterministic through and through, and in which the only "hidden" variables
This work amplifies and makes rigorous rather old ideas of the philoso-
Bohm briefly and skeptically.., but seems to be unaware that Bohm's pro-
gram is not just philosophical but involves a specific strategy, now seen to be
We emphasize, of course, that this particular work only rederives the clas-
the standard debates use this case as a touchstone. The moral, if one must be
whimsical, is that Occam's Razor may now cut through the leash that hereto-
than Centaur, Sphinx, and Hippogriff. In other words: (a) Einstein was right:
God does not play dice with the cosmos. (b) On the other hand, since indi-
vidually and collectively, we are not God, nor Laplace's Demon, nor any
the universe as something of a crap game. As far as (b) is concerned, note well
that even if the universe were purely Newtonian in the best eighteenth-cen-
the crap game would still be inevitable. This follows from the work of Poin-
care on classical mechanics and from that of his latter-day disciples, which
I omit, for the moment, their technical summary of DGZ's model, which
with mathematics and physics as such than with Gross and Levitt's claims.
event. They do see their summary as one "for readers with a little knowl-
course on how one defines "a little." In order for this summary to be useful
and had very good teachers. Indeed, one would need a good (technical)
{partial differential} equation] ... and gradient [involved in the ordinary dif-
are given in terms of a Riemannian metric scaled by the masses of the parti-
cles" (262 n. 9)? Even if one were, justly, to lament the dismal state of math-
ematical and scientific education in this country, this statement would seem
matics and physics. I am not saying that some of these things cannot be
explained otherwise; they can. They are just not so explained by Gross and
Levitt, who could have presented the key ideas involved in a more accessi-
ble form, and indeed much better and more lucidly and, most of all, more
accurately than they do here. (The phrase "the phenomenal world of quan-
and certainly requires further explanation, but we may let it pass.) If they
did, their energy would be far better spent than in their incompetent treat-
taste for public discussions of the latter. Perhaps they will learn to do it right
eventually.10
As the text stands, even readers with a little more than a little knowledge
of mathematics would only have Gross and Levitt's authority to assume, for
Bohmian and hence causal (262n.9). Well, these readers may not take this
authority for granted, and they should not. The assumptions in question
involved. Their physical justification and the overall status of DGZ's theory,
More generally, some (less informed) readers may get an impression here
but against Gross and Levitt's presentation of the situation. DGZ them-
shall discuss presently. Their actual claims for it are more circumspect than
those made for them by Gross and Levitt. One would easily see from, for
tion. The latter involve no less than the initial configuration of our universe,
obviously a wide open question. Cushing says: "The crucial question here is
singular event (i.e., the actual and only initial configuration of our uni-
to develop Bohm's views so as to give them more viability. There have been
ing, even though the attitude toward Bohmian theories in the physics com-
As I have noted earlier, some among Bohmian theories are different from
than only in terms of the physics they imply, such as causality or nonlocal-
accounting for the same data, rather than different interpretations of the
additional assumptions (and this is in part why these assumptions are intro-
theory, beginning with Bohm's original (1952) version. (This is why these
sion and its extensions (in particular to quantum field theories), Bohmian
theories possess certain (for most physicists) undesirable features that make
observed. Even in theses cases, however, this violation, and hence a conflict
with relativity, is a built-in feature of the theory, which is not the case in the
latter. This is arguably the main reason why these theories are suspect to
the skeptics was Einstein, who was on this (and other) grounds rather nega-
initiated by Einstein and brought into the foreground in the wake of Bell's
theorem, involves subtle arguments. Gross and Levitt claim that "Bell's the-
this form and without much further qualification, beginning with what they
it is true, those who argue for a general nonlocality of the quantum world.
any event, the existence and the nature of these arguments would confirm my
overall point here, as concerns the presentation of the state of quantum the-
ory and the debates concerning it. Certainly, nonlocality, manifest or not, is
This may be said, even if one leaves aside the problems of his exposition of
the Copenhagen interpretation itself, which hardly does justice to the com-
Bohm's own account and views. Bohm treats Bohr's arguments and his
thought with the greatest respect and sees it as crucial for his own views.
added).16 If this is the case, among those who misunderstood Bell's work
was certainly Bell himself. I was unable to find in any of Bell's articles, for
not think that one can find such an argument there. If anything, Bell sees the
Deconstructions * 211
more, and perhaps less, likely than Bohr's interpretation, however unap-
pealing the latter was to Bell.17 In any event, nowhere does he say that the
the Science Wars, depend on it and present it as a scientific fact rather than
in Science," published in Gross, Levitt, and Lewis, The Flight from Science
and Reason, cited earlier. This point itself is not wrong, or rather it can be
author among them), who not only have little in common but whose argu-
ments are also clearly in conflict with each other, which makes one doubt
mechanics and the thought of the key founding figures, both in general and
subscribe, and indeed, as we have also seen, he would not see quantum
of course a predictive tool, but it is also much more than that. His position
this study, and he makes it clear throughout his writings. Beller also appears
the opinion of the physics community than that offered by Goldstein and
Beller, or, again, Gross and Levitt. As Murray Gell-Mann and John Hartle's
alternatives to, say, Bohr's interpretation but of the nature of the claims
to either the actual state of quantum theory or the debates about it, either
physics. The support for the deterministic causality of the quantum world
does not seem (at least they say "seem") as strong as Gross and Levitt's
appears to cut pretty much the way it has for a while now, including as con-
cerns DGZ's "modest and unproblematic assumptions." That is, again, not
mechanics; quite the contrary, as we have seen throughout this study, this
debate continues. Indeed, it is this debate that makes the status of DGZ's
theory very different from what Gross and Levitt's account suggests. What
makes the invocation of Occam's Razor difficult in this case is the wide
sion, and of general philosophical views (often different even within each
ent status in the physics community in order for Gross and Levitt's claim to
tion, although the latter case may be made as well. This sloppiness is far
when mathematical and scientific ideas are used by humanists, and which is
and scientists do not always have proper expertise in areas outside their
own. This is, obviously, no sin; and nobody (naturally, the present author
included) is guaranteed from getting certain things not quite right or even
wrong altogether. But the situation does requires a very different type of
attitude toward all discourses involved from that of Gross and Levitt, Sokal
and Brictmont, and most of their fellow critics and supporters, or of course
Demon and Schrbdinger's cat are. Both ideas are pertinent to the argument
in question and the reader's ability to follow and access it. Laplace's demon
is a superhuman intelligence that would know all the positions of all the
particles (i.e., all the ultimate constituent atoms of the universe) and all the
forces that act upon them. Accordingly it would be able to predict strictly
(rather than in terms of probabilities) the future state of the universe at any
given point, by using the laws of the Newtonian mechanics or some version
at least given the actual laws of classical physics, as we know them, laws
that are, of course, demonstrably incorrect beyond certain limits.22 All cor-
unifiable, or even reconcilable with each other, even within classical physics
and leaving aside the limits, sometimes stringent, within which each of these
theories are applicable. Nor, in view of quantum theory and indeed already
relativity, does it appear, at least for the moment, that there is a classical-
like set of alternatives for different forces that will enable such reconcilia-
sical or not.) The so-called quantum gravity and other unifying programs
(such those emerging from string and, by now, "brane" theories) aim to do
principle. If they succeed, however, it may be, in its rigorous specificity, nei-
Schrbdinger's cat.
course, that this characterization may be applied by Gross and Levitt to the
214 * The Knowable and the Unknowable
cally) locked in a box inside which a certain quantum, and hence irreducibly
probabilistic, event may or may not take place. If it does, it will trigger a
device (also installed in the box) that will kill the cat. We would not know
the outcome before opening the box, which we can do long after the event
did or did not take place inside the box, although this time, too, is subject to
quantum event in question, the triggering device, the cat and all-is consid-
hand, it is indeed difficult to assume that the cat would not be definitively
dead or alive inside the box immediately after the quantum event in question
took place inside the box. The "paradox" (not everyone sees it as such)
requires a subtle analysis and remains a part of the debate concerning quan-
tum mechanics, a subtlety that need not be followed here. It may be observed
briefly that this subtlety has mainly to do with the relationships between the
quantum microworld and the classical macroworld, and the character of the
a quantum system. For example (this would be more or less Bohr's view),
event inside the box, and hence as dead or alive, depending upon whether the
event did or did not take place. In this view, there is no paradox.
from any account of the physical world. This would be the case whether we
both in practice and (in contrast to the classical chance) in principle, or clas-
sical chance, irreducible only in practice but leaving the causality in princi-
ple intact. The latter would be the case in classical mechanics, classical sta-
lel is far too loose. More generally, the status and roles of these theories in
Deconstructions * 215
our description of the physical world on different (macro and micro) levels
are quite different, even if one maintains the underlying causality in all
cases. These differences and nuances, blurred by Gross and Levitt, are
sion about science that bothers Gross and Levitt and their fellow critics.
Poincare's extraordinary findings, which are indeed one of the early inti-
in principle and even in idealized situations, this behavior (once the number
causal) laws of such behavior, specifically his law of gravitation, and can
write corresponding equations for it. In other words, the behavior of such
tions of such behavior are not possible even in principle, beyond certain lim-
That is, a small change in such conditions can lead to a big change in the
behavior of the system. In other words, in the case of the theories assembled
under the rubric of chaos theory, indeterminism arises from the nature of
In any event, the equations are assumed to map the behavior of these objects
as such, which may or may not be constituted by more or less large multi-
already three bodies would do; in the case of (long-term) weather predic-
indicated, actual systems are often chaotic. The point of chaos theory is that
some controversy.
ian, let alone as that of Einstein's general relativity, the complexity becomes
in principle two sets of descriptions: one maps the actual behavior of objects
involved, and the other does not, while it depends on this behavior in estab-
lishing the counting procedures that enable statistical predictions of the the-
the difference between the two types of systems just described. One of the
by chaos theory) is, beyond certain limits, unpredictable. In this sense, the
cal physics make statistical predictions. The theory does show certain com-
plex patterns or forms of order. One can see these patterns in, by now
would be the "order theory," the theory of certain complex and unpre-
In quantum theory (in the standard, rather than Bohmian, version and in
models, to exhibit causal (or indeed any describable) behavior; and individ-
ual quantum events are, in general, not subject to physical law. (Further
lier, must be kept in mind.) Nor, in dealing with quantum statistical multi-
and must be in the case of classical statistical physics (that is, we cannot
assume that we can mark, "flag," each and trace, even in principle, the indi-
possible and are made possible by quantum theory. This theory, accord-
Deconstructions * 217
theory describing individual objects and their behavior in the way classical
ries, such as DGZ's, are yet another story, and their relationships to other
(or the lack thereof) of quantum physics are introduced by quantum field
theories, the currently standard way of theorizing what we see as the ele-
One might say that the scientific and philosophical expectation, concern-
cal physics were defeated from two sides, that of chaos theory and that of
quantum mechanics. The first made apparent that the behavior of certain
tion (in the sense defined earlier), since individual quantum events are seen
These are well-known considerations, and I repeat them here for the fol-
picture of what is at stake here. Second, this picture further exposes the
statements cited earlier. One may say, as Gross and Levitt do, that "even if
tion, and the perplexities of quantum mechanics entirely avoided, the crap
first, the significant differences between different classical crap games, such
as those of chaos theory versus classical statistical physics, and, second, the
fundamental and irreducible difference between the two "crap games," clas-
technically (very different formalisms are required in each case), let alone
epistemologically. To say, however, as Gross and Levitt do, that "this [i.e.,
the unavoidability of the crap game in the classical case] follows from the
work of Poincare on classical mechanics and from that of his latter-day dis-
ciples, which goes under the fashionable name 'chaos theory,"'" is hardly
is, the statement appears to be very much in the "classical" tradition of con-
fusion, indeed common enough among humanists, about these matters; and
here again a statement like this is more likely to cause or increase this con-
fusion rather than help to avoid it. There are different "Newtonian uni-
tions of the behavior of celestial bodies (be they planets of the solar system,
true, since the crap game would then be that of statistical physics, possibly
with that of chaos theory. If such, or more elementary, objects are subject to
a Bohmian mechanics, such as DGZ's theory (which does not take into
We have very little sense at present how such different classical ways of
description can come together even in idealized cases (and hence leaving
physical data). The history of modern physics is the history of dealing with
and those of its correspondence with the available data, including the data
that eventually led to relativity and quantum physics. The success has been
limited on both scores. For a program of the Laplacian type to succeed one
since Gross and Levitt's elaboration is, after all, a ("casual"?) footnote. The
note and the argument, however, are hardly casual, even leaving aside that
Gross and Levitt's sloppiness is out of place, given their critical stance as
footnote in their book, and for obvious reasons. First of all, the subject of
deterministic causality, or the lack thereof, in the physical world (or, again,
saying that the outcome of this debate, whatever it is, has a direct bearing
on how one sees the world elsewhere, for example, in the humanities and
the social sciences. One does not need quantum physics to argue the non-
instructive. On the other hand, these are not some "postmodernists," but
classical philosophers and scientists, who have, since Newton at least, used
the classical behavior of nature both as a model of and as support for their
Levitt or Sokal and Bricmont do just the same, albeit in a much cruder way.
that these theories both incorporate classical theories and enable, better that
classical theories could in these circumstances, the disciplinarity and the dis-
cipline of the fields and projects that sustain, and are sustained by, classical
theories.
Nor is Gross and Levitt's (or, as we have seen, Sokal and Bricmont's)
thing, how are the humanists, in particular students, to correct their views
Levitt or Sokal and Bricmont appear to want them to do, if their own pre-
atic nature of Gross and Levitt's (or Sokal and Bricmont's) book and related
criticism or not) but also the nature and complexity, and sometimes strictly
their relation to culture. The authors may say that it was not their goal to
discuss the latter subject. The point, however-and this may well be the
broadest and the most important critical point one can make here-is that
possible otherwise.
Heisenberg's Deconstruction
My argument throughout this study has been that, in contrast to their Sci-
significant, and sometimes deep, about the nature of mathematics and sci-
ence. The point is more or less self-evident when one encounters such sci-
entific nonclassical thinkers as Bohr and Heisenberg, which is not to say that
this point is insignificant even in this case in view of the fact that in question
and science. We do, however, learn something about the philosophical and
about complex numbers from Lacan, and about topology and relativity
from Deleuze and Derrida. Reciprocally, we also learn a great deal about
not have the benefit of Bohr's and Heisenberg's arguments. Certainly quan-
dous "opportunity of testing the foundation and scope of some of our most
We are far from finished with this nonclassical testing, and perhaps new,
more severe and more unimaginable, tests and trials are in the offing. In
physics, while some of the currently available theories, such as much of clas-
plete within their limits, overall our theories remain manifestly incomplete
such as string theories, by now extended into "branes" theories; and each of
ories describe various macro and micro aspects of the physical world and of
these theories is far from established, and some of them are highly specula-
selves. This may or may not continue to be the case, once new theories or
not inconceivable either. For the moment we can at best correlate some
Deconstructions * 221
among the available physical descriptions and try to maintain their consis-
the century of physics that began with Planck's discovery of the quantum of
action in 1900. It is also possible that the future will produce not only as yet
gether, and of entirely other questions (if this, for now seemingly
ematics, biology, or computer technology, as all these fields invade and fuse
with each other ever more aggressively, without, it appears, worrying too
the one just given for physics. We do not know at the moment, but the
edge in these fields are at least as likely as any return to classical knowledge
its significance both in nonclassical knowledge and in its own right in many
mathematics and science, assuming, as I do, that it has done so in the past?
It would appear that at least in the near future it would; at least for the
has already been considered here from this perspective. I would like to close
poetry, calls "apophrades." He writes, "the uncanny effect is that the new
poem's achievement makes it seem to us, not as though the precursor were
writing it, but as though the latter poet himself has written the precursor's
makes this type of point in Milton), or Wallace Stevens has written certain
passages in Milton, even if not some of his poems. In the present case we
need not be strictly personalized or need not, and indeed cannot, be con-
tained within a single field either. As we have seen, the anxieties concerning
nonclassical theory are many and diverse. There are those pertaining to the
possibility that such theories are right after all, a difficult circumstance for
those who do not want them to be right or (it is rarely a question of choice)
those who cannot accept them as possibly ultimately right. There are also
anxieties that relate to the possibilities (often more important than fact
when anxieties are concerned) that somebody was there already, a while
ago. "Poets were here before me," Freud liked to say. To the degree that this
is true (not a simple question, but secondary for the moment), in our cul-
ture, in our "two cultures," it appears easier both for poets to be there first
and for others to confront poets who are ahead of them. "The shadows of
gigantic mirrors which futurity casts upon the present" and "the unac-
them easily could be on the same grounds that philosophers like Deleuze
and Derrida are. But then, as we have seen throughout this study, nonclas-
sical thinkers are rather willing to give mathematics and science, and math-
ematicians and scientists, the place Freud gives to poets and often place
classical or, as, among others, Lyotard suggests, indeed postmodern, first.
Shelley certainly would, and did, apply his view of poetry, as always the
poetry of the future, always coming from the future, to mathematic and sci-
entific discoveries. For him physics, chemistry, and (then emerging) biology,
too, were "the shadows of gigantic mirrors which futurity [casts] upon the
some of the problems of Bohr's Como version, in part through his more
physical basis of the lectures, is closer to Bohr's later views, the views in turn
the Quantum Theory." This appendix is much more than an appendix and
tum theory is, epistemologically, close to the view adopted by this study, in
We shall now sketch the deduction of the fundamental equations of the new
matical sense of the word, since the equations to be obtained form themselves
the postulates of the theory. Although made highly plausible by the follow-
108).
effects in the above sense, which, rather than properties of quantum objects
known and well documented, however, it was Einstein's argument that "the
224 * The Knowable and the Unknowable
theory decides what can be observed" that was guiding Heisenberg in his
work leading to his invention of quantum mechanics. His theory was not
involves their irreducible mutual reciprocity, which found its way and
famous, but not always carefully read, opening statement: "The present
261; emphasis added). "Relationships" is the key word here, and the title of
crucial too, for, no matter how theory-laden and how complicated the
in this or any other way quantum objects and their behavior beyond the
effects in question.
Thus, even leaving aside for the moment the theory-laden character of all
conceivable data (including that of classical physics), dealing with such, "in
them, and Heisenberg's paper does not conform to the latter conception.
analogy with Newton's calculus, which was both born from classical
physics and through which classical physics found its ultimate representa-
physics his calculus brought about. While working with the available data
of quantum physics (such as the Rydberg-Ritz formulas and the Bohr fre-
quency relations), his theory qua theory was founded above all on Bohr's
argue that for large quantum numbers the data becomes the same as it
description could, in all rigor, no longer be the same). The principle was also
used to argue, correlatively, that the equations should be formally the same
have turned out to be his greatest and (it is uniquely his) most original con-
variables of the theory. Both Dirac's and von Neumann's schemes are more
berg's stroke of genius of finding his matrices (not altogether unprepared by,
among others, Bohr, but a stroke of genius nevertheless) was itself a found-
ing theoretical move. That is, this arrangement of the relationships between
not an observation of nature, which does not arrange anything in this way.
(One might also recall that these matrices must be infinite in order to derive
matrices are not observable quantities. The latter is indeed a correct point,
never claimed they were. They were, however, linked to observable quanti-
in the statement cited earlier in this study that, while "Bohr was primarily a
our day and age carries weight only if its every detail can be subjected to the
physics, and to some degree even in physics itself. Heisenberg starts with
Thus it was characteristic of the special theory of relativity that the concepts
"measuring rod" and "clock" were subject to searching criticism in the light
assumption that there exists (in principle, at least) signals that are propagated
with an infinite velocity. When it became evident that such signals were not
to be found in nature, the task of eliminating this tacit assumption from all
logical deductions was undertaken, with the result that a consistent interpre-
tation was found for facts that had seemed irreconcilable. A much more rad-
ical departure from the classical conception of the world was brought about
tion of experiments in which both the gravitational constant and the recipro-
cal of the velocity of light may be regarded as negligibly small. (The Physical
Derrida's extended sense of the term. This critique may also be seen in (par-
these terms to some degree allow one to proceed otherwise than by means
the Physical Concepts of the Wave Theory." (Those familiar with decon-
out. I shall return to this point presently. Now, I shall address the two cri-
of the classical particle and wave theories, virtually in the strict sense of Der-
cular Theory of Matter"] the simplest concepts of the wave theory, which are
rect." They were taken as the basis of a critique of the corpuscular picture,
and it appeared that this picture is only applicable within certain limits,
which were determined. The wave theory, as well, is only applicable with cer-
tain limitations, which will now be determined. Just as in the case of particles
Broglie waves). For these theories the term "classical wave theories" will be
used; they are related to the quantum theory of waves in the same way as
critique of the wave theory concept has been added to that of the particle
due regard is paid to the limit of applicability of the two pictures. (47)
need not. Heisenberg virtually does it for me! The very essence of decon-
defined here, as well as, and indeed correlatively to, certain key aspects of
sion, or Heisenberg's own earlier views, ultimately, since, while there are
effective in Bohr's later work. On the other hand, it may also be argued that
allow one to avoid some of the problems of Bohr's Como argument, as con-
228 * The Knowable and the Unknowable
Bohr's Como argument, a task that Bohr himself accomplishes later rather
hold for matter and wave motion. The solution of the difficulty is that the
cles, the other of waves-are both incomplete and have only the validity of
analogies which are accurate only in limiting cases [where quantum and
classical physics give the same results]" (10). The main point, however,
under deconstruction and, second, their workings, against each other within
in question require.
There are, it is true, significant differences as well, even leaving aside the
the common metaphysical or, one might also say, classical base for views
the classical concepts of particle and wave, although of course not the com-
How bad, then, can Derrida's "commentary on physics" be, even though
Deconstructions * 229
and possibly because it is in fact something other than physics that is his
subject?
(more classically) expected from it. It would be difficult to discuss the rela-
work to show that it requires or (keeping in mind the analysis given ear-
we have seen in chapter 2. (In this respect, his initial hopes for his equa-
through juggling both wave and particle pictures and complementarity (in
out of the paradoxes, at least to make some key moves in this direction. As
Eventually, in the wake of the EPR argument, Bohr arrived at a more rad-
directly, although that was not so apparent at the time.30 This interpretation
does not depend on either wave or particle theories or properties, not even in
theory are retained at the level of the effects of the quantum (and hence ulti-
suring instruments upon the latter. Nor, accordingly, would one need to
this way, too, as Heisenberg does in his Chicago lectures, but one need not;
and in a way Bohr might have been (it is difficult to be certain) better off fol-
I cannot consider the subject in proper detail here, but it may be argued
made possible by this vision. (I continue to remain here within the linea-
otherwise my argument at the moment may not apply.) This vision itself
(if one may use such an expression) a radically deformalized form-that is,
way at all. It may not be humanly possible to do so, even though, in contrast
question, the (material) elements constituting these data are available to phe-
way configuratively, and indeed deal with, as de Man would have it, "the
Deconstructions * 231
their classical and hence configurable appearance (in either sense), of any-
thing that could possibly be mapped by a classical model, even though they
terms and hence of their manifest classical configurativity. For example, they
should not be seen either as points resulting from classically conceived colli-
At this stage, even the radical (Derridean) tracelike character of these marks
This suspension is necessary, and the vision that results is possible, to the
degree it is possible as a vision, for the following reasons. (Once again, the
such, or for that matter any material physical process in the way classical
character of the marks in question and the nonclassical nature of the inter-
the emergence of the manifest effects, the visible marks, involved, must also
232 * The Knowable and the Unknowable
to quantum data and the very possibility of configuring these data accord-
tions in the paper itself would suggest nearly as much. Heisenberg does not
of which he was only vaguely aware at the time. His main concern was to
in the situations where all previous attempts had failed. These conse-
work cited earlier, specifically his paper introducing the uncertainty rela-
than only Bohr, may, as I suggested earlier, be seen as the discoverer of the
tesy of Bohr, who also gave this epistemology a firmer philosophical base.
or, as against the classical view, reassembling this data through his matri-
mechanics may still apply. He was not even aware at the time that the cor-
the time, which was realized by Max Born, his teacher, upon reading the
paper. Heisenberg reinvented it. This new approach to the data is what
enabled him and others (this took a few months) to formalize these effects
upon the latter in a new way, ever since called quantum-mechanical for-
explained earlier, the new (as opposed to classical physics) kinematics was
defined by the data arranged into these tables, now used as variables of
Deconstructions * 233
his ideas, that is, on the basis of what principles and data (for example, for-
mally taking over the equations of classical mechanics and substituting new
arguments), the situation of his invention does not look quite so dramatic.
way poetic, for this, too, is "the highest form of musicality in the sphere of
thought."
the type Derrida reaches (more deconstructively in his earlier work, more
directly in his later works) in and through his analysis of writing. This argu-
mology and its relation to various, let us say, more technical deconstructive
works. In other words, the question is this. Do we need this type of techni-
classical epistemology, for example, that entailed by and entailing such Der-
ridean structures as differance, trace, writing, and so forth (we recall that,
while finite, this list does not permit termination)? Conversely, can we pro-
his later work appears to have done? Or how is deconstruction (in this
Chicago lectures and Bohr's Como approach, while earlier work of Heisen-
berg and Bohr's post-EPR work link nonclassical epistemology and comple-
which, however, may make these questions all the more significant.34 In any
event, in considering the quantum data, we deal with nothing less than writ-
ten traces in Derrida's sense and perhaps, as we have seen, with something
more radical. The "traces" one encounters in quantum physics may or may
Then, the question of how bad Derrida can be on physics (or in general)
indeed acquires a meaning rather different from the one it received in most
234 * The Knowable and the Unknowable
recent (and some no longer recent) criticism of Derrida's work. Rather than
being bad, as and in the way these critics contend, he may only not be good
enough.
Physics itself is yet another question. It does appear, however, that one
physics more than it has even been classically, under these nonclassical con-
ditions. This task may, as Einstein said, indeed require "the highest musi-
cality in the sphere of thought." It may even require the highest musicality
in the highest sphere of thought, where musicality and thought become one
and the same. But then perhaps this is what thought is, wherever we
in any way. It appears likely that this thought, just as modern mathematics
and science, can only be thought of, unthought of, nonclassically-if then
just as quantum objects. We may need to think of them in the same unthink-
able way.
Conclusion
In his 1927 Como lecture, Bohr speaks of the "inherent 'irrationality' "
Bohr's writing on quantum theory. The idea itself of the quantum irra-
over, that in using this term Bohr in the Como lecture had in mind the way
cal space and time (PWNB 1:76). It is true that he actually spoke of this
tained the last point itself and hence the symbolic character of both methods
on other grounds, which are indeed many, throughout the articles, includ-
ing at the juncture just cited. The essential dependence on, the irreducibility
of, the role of complex numbers in quantum mechanics became clear shortly
ical sense, as they are in their mathematical sense, thus, I would argue, pur-
236 * The Knowable and the Unknowable
ter 1, is a key nonclassical part of this view) will it ever be available to us.
ics, its mathematical formalism, which might, and to some does, take on the
objects but only enables us to predict the effects of their interactions with
measuring instruments upon the latter. This formalism, we also recall, irre-
Remarkably, all these elements appear and come together in the famous
picture that Bohr drew in 1962, one day before his death, on the blackboard
in his house (BCW 7:286). At the top there is the general formula for the
ber sheets. Riemann introduced this idea for dealing with such functions in
order to be able to have a single rather than double (and hence, ambiguous)
tions (defined in mathematics so that only a single value of the function cor-
responds to each value of the variable). Bohr tried to use Riemann's idea
ald Bohr, who was to become a great mathematician in his own right and
ald, Bohr described his emotional state during his early work on electron
is, for their effects upon measuring instruments. Of course, Bohr's sketch on
2:54-57). Thus, as has been often observed, Bohr's sketch represents the
beginning and the end or, rather, the last key stage of his work on comple-
mentarity. What does not appear to have been commented upon, however,
is that right underneath the formula for the functions of complex variables,
Bohr in fact writes J 2, the first irrational number! One might argue that it
could also be z, rather than 2, an often used symbol for complex variables
(x + iy). However, Bohr does have a "z" in his sketch. It is used as the des-
ignation just explained above his formula, and it looks very different in his
turies, indeed millennia, from perhaps the first scientific encounter with the
In the twentieth and by now the twenty-first century we have moved into
even speak of progress here, if only against some recent critics) can be traced
in other fields, such as those traversed here, and in the interactions among
these fields or between them and mathematics and science, and hence in the
eternally return, even though and perhaps also because, as I have argued
here, there are more than two cultures involved in these interactions and
confrontations, more that two and less than one. Something on the order of
the "two cultures" continues to invade this multiplicity and this less than
advice to Horatio upon the appearance of his father's ghost: "This is won-
more things in heaven and earth, Horatio / Than are dreamt of in your phi-
welcoming these strange things is in order. For now, we confront each other
more in the way Levinas and Blanchot describe in speaking of ethical rela-
always somewhere other than we are, not belonging to our horizon and not
It is nearly understood that the Universe is curved, and it has often been sup-
posed that this curvature has to be positive: hence the image of a finite and
will [man] ever be ready to receive such a thought, a thought that, freeing
him from fascination with unity, for the first time risks summoning him to
response would necessarily fall anew under the jurisdiction of the figure of
Riemann's ideas are used in a similar way (albeit somewhat more loosely)
and indeed via formulations echoing the ones just cited, specifically lan-
such, it suggests, along the lines indicated earlier in this study, the poten-
tially nonclassical nature of Einstein's general relativity, the basis of the cos-
interrogation is, for Blanchot, also, finally, the question of whether we are
capable of mathematics and science. Thus, the question of the "two cul-
tures" is also the question of the nonclassical culture or, irreducibly, cul-
tures, always more than two and less than one, within which we must think
differently the work, old and new, of both literature, and mathematics and
science, and their relationships. It is indeed not easy to confront such a task
240 * The Knowable and the Unknowable
throughout their history, or in our (in the broadest sense) ethical encounters
with strangers or with what is strange, which we often find in those whom
strangers coming from the "two cultures," to hear in Levinas's and Blan-
chot's words what they want us to hear: that invisible that "turns away
by Blanchot and which is the subject of this study. At this limit what we are
indeed be put to its ultimate test. We tend to steer away from this limit or
have argued here, with Nietzsche, they test our ethics, too, at least our self-
when we approach the strangeness of, and strangers from, other fields or
strangers from far afield, such as modern mathematics and science, on one
side, and French philosophical thought, on the other. It is not that we can
Perhaps the ultimate ethics or (since the ultimate ethics may not be pos-
in other cultures, those ideas that bring our own culture-say, science, on
one side, and the humanities, on the other-to the limits of both what is
reach the limits of both what is knowable and what is unknowable in one's
both of these limits, to tell the other culture or field both what we know and
Naturally, we can never be fully certain how our offerings will be received,
not even within our own culture, since we are often strangers there, too. Nor
can we be quite certain as to what we know and what we don't know. Nei-
Chapter 1
vols. (Woodbridge, CT: Ox Bow Press, 1987), 2:67-69; 3:1; Werner Heisenberg,
The Physical Principles of the Quantum Theory, trans. Carl Eckart and Frank C.
2. As early as the 1930s, Bohr (in collaboration with Leon Rosenfeld) extended
(hereafter QTM), ed. John Archibald Wheeler and Wojciech Hubert Zurek (Prince-
8:219.
6. This confluence may not be coincidental, even beyond the more general
Accursed Share: Volume 1, trans. Robert Hurley [New York: Zone, 1988], 191 n.
2). Indeed, a chapter on Bataille could have been easily included in this study. I have,
NC: Duke University Press, 1994) and more recently in "The Effects of the Unknow-
able: Materiality, Epistemology, and the General Economy of the Body in Bataille,"
8. Bohmian theories are different from the standard quantum mechanics, since,
in contrast to the latter, they expressly assign trajectories to particles, which the stan-
dard version does not do (Bohr's interpretation of course expressly forbids such an
the standard version. Accordingly, I shall collectively refer to Bohmian theories here
The latter is made possible by virtue of assuming certain additional physical quanti-
ties ("variables") that are, at least at this stage, unavailable to experimental detec-
sical and nonclassical features, as one can easily see from the introductory materials
in volumes 6 and 7 of BCW. For a representative example, see Henry Folse, "Niels
1987, ed. Pekka Lahti and Peter Mittelstaedt (Singapore: World Scientific, 1987),
161-80, on which I shall comment later. I have considered some among the promi-
10. Ultimately, there may be no physical theory that is fully free of interpreta-
tion, to the degree that we can unequivocally distinguish between theory and inter-
theory, such as quantum mechanics. Some do see quantum theory, or any physical
matically consistent scheme of the theory as a prerequisite for any meaningful inter-
pretations. For example, Bas C. van Fraassen takes this view in his exposition of his
View (Oxford: Clarendon, 1991). Van Fraassen, accordingly, does not see most of
Bohr, all my epistemological arguments can be properly correlated with Von Neu-
more elegant mathematics. Indeed, using the latter would make some of my key
arguments easier to make. However, the highly abstract nature of this formalism
would make my exposition far less accessible for the general reader.
Notes to Pages 13-16 * 245
11. Predictably, some among such arguments appeal to the limited scope of most
current theories, specifically quantum mechanics, and contemplate and pursue the
such as quantum field theories) that take into account higher-level quantum effects
or gravity. Among the attempts of this type are those by Roger Penrose, David
Deutsch, Giancarlo Ghirardi, Alberto Rimini, Tulio Weber, and several others. Most
they share with Bohmian mechanics (which inspired many of these proposals). See,
for example, Roger Penrose, Shadows of the Mind: A Search for the Missing Science
of Consciousness (Oxford: Oxford University Press, 1994), 331-46, which also sur-
veys some of these attempts, and Giancarlo Ghirardi, "Beyond Conventional Quan-
tum Mechanics," in Quantum Reflections, ed. John Ellis and Daniele Amati (Cam-
(New York: Random House, 1994), often refers to views that are classical in the pre-
sent sense, as are also some among the views associated with such denominations as
terms "mathematics" and "science" are not always as self-explanatory as they might
14. Friedrich Nietzsche, The Gay Science, trans. Walter Kaufmann (New York:
tarity, History, and the Unconscious (Gainesville, FL: University of Florida Press,
1993), 54-74.
17. This circumstance poses the question of whether the interpretations of clas-
sical theories pursued by such investigations are in fact nonclassical, rather than still
nonclassical theories they consider. If (and when) the latter is the case, the paradox
disappears. On the other hand, as has been pointed out (by, among others, Kant),
there will always be "savants" who would find anything in one or another of the
pre-Socratic philosophers, provided they are told what to look for. I am saying this
not in order to dismiss all such rereadings of old texts via new theories but to sug-
gest that new theories, whose interpretations (such as those offered by this study)
may of course be debated in turn, entail a complex balance of (re)reading both the
18. For Bell's view of quantum mechanics see his The Speakable and the
1987). The title of this study may be seen as an implicit argument with Bell's views
176-90.
246 * Notes to Pages 17-31
Bohr: His Life and Work as Seen by His Friends and Colleagues, ed. Stephan S.
23. I have addressed the latter subject in In the Shadow of Hegel, 54-83.
24. Friedrich Nietzsche, The Birth of Tragedy and the Case of Wagner, trans.
26. Gilles Deleuze and Felix Guattari, What Is Philosophy? trans. Hugh Tom-
linson and Graham Burchell (New York: Columbia University Press, 1993), 11-12,
24.
27. One can mention such recent endeavors as algebraic topology and its exten-
Chapter 2
1. I cannot consider here Planck's law and its history or key events following
Planck's discovery and leading to quantum mechanics and its interpretation. See
(Oxford and New York: Oxford University Press, 1978) for arguably the best
account of the early stages of this history. Bohr gives an excellent and conceptually
(Princeton, NJ: Princeton University Press, 1983), are exceptionally useful as well. In
the context of Bohr's work, see also introductory and historical material in volumes
2. This may be said, even if one were to leave aside the fact that the conven-
tional (nonrelativistic) quantum mechanics has only a limited scope in our, as yet
incomplete, understanding of nature at the level of its ultimate constitution and that
the epistemology of the higher-level quantum field theories remains as yet a largely
Theories (Princeton, NJ: Princeton University Press, 1995) is among a few recent
attempts and also contains useful references. By contrast, the literature (technical,
and even a list of classic works is long and not easily claimed to be definitive in view
Notes to Page 31 * 247
of the diversity of the views concerning virtually every aspect of quantum mechan-
mechanics itself. Even within the cluster of the standard or Copenhagen (or, as it is
range is formidable, even if one restricts oneself to such founding figures as (in addi-
tion to Bohr) Heisenberg, Born, Pauli, Dirac, Von Neumann, and Wigner. The two
main lines of thought within the Copenhagen cluster are defined by the argument of
the behavior of quantum objects themselves. The first line often follows Dirac's and
Dirac's and Von Neumann's versions see their seminal works: Paul A. M. Dirac, The
Principles of Quantum Mechanics (Oxford: Clarendon, 1995), and John Von Neu-
(Princeton, NJ: Princeton University Press, 1983). Their own positions on the sub-
ject are actually more complex and are closer to Bohr's view, especially in Dirac's
case (The Principles of Quantum Mechanics, 10-14). Indeed, this is not surprising,
even from a mathematical standpoint, since Dirac's and Von Neumann's versions
say it entails) the other view. Richard Feynman's version is still another story and
will not be discussed here. The profusion of new interpretations during recent
decades was in part motivated by the EPR argument. It may, however, be seen as
triggered by David Bohm's reformulation of the EPR argument in terms of spin and
impetus from Bell's theorem (1966) and related findings, especially the so-called
Kochen-Specker theorem and then from Alan Aspect's experiments (around 1980)
confirming these findings. Bell's theorem states, roughly, that any classical-like the-
ory (similar to Bohm's) consistent with the statistical data in question in quantum
late relativity theory. Bohm's theory does so explicitly, in contrast to the standard
quantum mechanics, which does not. As will be seen, these developments recentered
James T. Cushing and Ernan McMullin (Notre Dame, IN: Notre Dame University
Press, 1989), offers a fairly comprehensive sample, although it requires some updat-
ing. See also Ellis and Amati, eds., Quantum Reflections. David Mermin's essays on
the subject of quantum mechanics in Boojums All the Way Through (Cambridge:
Cambridge University Press, 1990) offer one of the better nontechnical, although
demanding, expositions of some of these subjects. By now, dealing only with non-
among others (and with variations within each denomination), the hidden-variables
tion. Among the most recent additions is Mermin's provocative proposal for what he
248 * Notes to Page 31
calls "the Ithaca interpretation," which maintains that only statistical correlations
considered by quantum theory. While, on the one hand, Mermin's approach aims
to ascribe physical reality to correlations alone, not correlata, as would be the case
ing to Tell Us?" American Journal of Physics 66, no. 9 (1998): 753-67, and refer-
ences there. Mermin credits such theorists as Arthur Fine, Paul Taylor, Carlo Rov-
elli, and Lee Smolin (the latter two contributed significantly to both Bohmian
ing one to nonclassical epistemology. The present argument develops and some-
times significantly refines several related works by the present author, in particular
bara H. Smith and Arkady Plotnitsky (Durham, NC: Duke University Press, 1997);
and Complementarity.
those associated with what happens on the microscale, established by Planck's law,
and can be extended to macro-objects, including the Universe, which may (there are
Their quantum "behavior" is, however, due to their ultimate quantum microconsti-
tution (regardless of how one interprets the latter), at least the way we understand
the situation now, an important qualifier to be applied to all claims made here con-
cerning physics and its epistemology. These complexities have largely to do with the
possibility of placing an observer outside the Universe, a fact that is sometimes used
tion, such as Bohr, has to be deficient on that scale. See, for example, Lee Smolin,
The Life of the Cosmos (New York and Oxford: Oxford University Press, 1997),
and, for a counterargument, a review by the present author, "The Cosmic Internet:
A Review of Lee Smolin's The Life of the Cosmos," Postmodern Culture 8.3 (May
1998) (published electronically). Smolin's more recent book, Three Roads to Quan-
tum Gravity (New York: Basic Books, 2001), avoids some of the problems of The
Life of the Cosmos. In particular, the book offers a useful argument for the observer
ing to this argument, while all observers relate to the universe as a whole, each
observer must be excluded from those parts of the universe this particular observer
can describe (48). The latter point is consistent with Bohr's argument concerning the
difference between quantum objects and measuring instruments. The concept of the
Notes to Pages 31-35 * 249
universe as a whole (i.e. as a wholeness), retained from The Life of the Cosmos, is
argue, in fact undermines the argument for the observer dependent logic. I
addressed this latter issue in "The Cosmic Internet." These considerations do not
affect the present argument. It may be observed, however, that it is not inconceiv-
able that, in view of its possible ultimate constitution (quantum or not), the "Uni-
verse" on this scale may not allow for a conception in terms of any attributes or,
Blanchot said in his The Infinite Conversation (trans. Susan Hanson [Minneapolis:
University of Minnesota Press, 1993], 52), which, as I said, may even be the case
already at the level of Einstein's general relativity at the cosmological scale (i.e.,
conception of "the unfigurable Universe" in the conclusion to this study. The con-
cept itself of the unfigurable universe, as just explained, would be consistent with
Bohr's scheme, without at the same time disabling our investigation of its physical
4. Such constituents, including those associated with the particles of matter and
radiation, such as electrons and photons, remain different in other respects. In addi-
tion, as will be seen, only some particle-like or wave-like properties could apply in
any event.
5. Most key (technical) papers on the subject are assembled in Sources of Quan-
tum Mechanics, ed. B. L. van der Waerden (Toronto: Dover, 1968), which contains
see Abraham Pais, Subtle Is the Lord: The Science and the Life of Albert Einstein
(Oxford: Oxford University Press, 1982), 402-14, and Niels Bohr's Times, in
Heisenberg considers spectra, but this does not change the epistemological consider-
ations in question. Indeed, this fact makes these considerations more pertinent, since
it makes the very term kinematical (if understood as something representing motion)
ing instruments, and thus translated Bohr's argument, which Bohr in effect has done.
Much of this work summarizes more complex arguments of his great earlier papers,
tions" (in B. L. van der Wareden, ed., Sources of Quantum Mechanics, 261-77) and
Wheeler and Zurek, eds., Quantum Theory and Measurement, 62-86), introducing,
of the second title is misleading and should read instead "on the representable (intu-
the role of the term "anschaulich," crucial for both Bohr and Heisenberg, later. The
while earlier papers are in fact closer to Bohr's post-EPR argument, on which point
10. Niels Bohr, interview by Thomas Kuhn, Aage Petersen, and Eric Riidinger,
11. The concept of phenomenon was introduced by Bohr in his Warsaw lecture
12. It may appear that, following Kant, one should more properly speak of
"noumena" here. Bohr is, however, right not to do so, since "noumena" would more
13. It is worth keeping in mind that the latter concepts are not independent of
the case of particles, for example, such attributes would include "position" or
tories are more or less immediately prohibited in quantum mechanics). In the case of
elementary particles of modern quantum physics, such attributes would also include
such "objects" are not seen as likely to exist in nature. As we have seen, the objects
of classical physics, too, are sometimes idealized, for example, as (massive) dimen-
itself. It is also worth observing that the "size" of the electron was a problem for the
well.
least to directly refer to, if not quite to describe, quantum processes themselves and
as it is in classical physics. Hence, such interpretations may be seen, and are often
Thus, we would have at least competing and, to many, epistemologically more palat-
referring to the hidden-variables theories and related nonlocal theories (these theo-
ries are expressly different from quantum mechanics) or the many-worlds interpre-
tation of quantum mechanics, as the latter do not retain the key features of quantum
mechanics here under discussion. I have in mind (in addition to some earlier ver-
Notes to Pages 41-43 * 251
University Press, 1994); and possibly (this case is more complex) Mermin's "Ithaca
interpretation." These are, arguably, among the closest to Bohr's among such inter-
1998), are more grounded in the concept of measurement. Both van Fraassen and
Omnes associate their interpretation with the Copenhagen interpretation; and Mer-
min's indicates further proximities between Bohr's and the histories interpretations
of Griffiths and Omnes (Mermin, "What Is Quantum Mechanics Trying to Tell Us?"
763). I shall refrain from an ultimate judgment concerning such claims and offer
only the following observation, related to what Bohr's or a Bohrian response would
be as concerns the possibility of such claims in principle, referring the reader to the
imply the following for the interpretations in question, given that the mathematical
formalism involved is that of standard quantum mechanics, that is, a highly sym-
ment is related to anything that can actually occur in space-time, Bohr's phenomena
16. This view of the ultimate nature of the physical world was recently advo-
cated by Julian B. Barbour in The End of Time: The Next Revolution in Physics
(Oxford: Oxford University Press, 1999). Barbour is right to argue that motion (or
change, differentiation, and so forth) may not apply at the ultimate level of the con-
stitution of nature, which is the present view as well. The present argument also sug-
gests, however, that what Barbour calls "Platonia" (the state of nature without
change and motion) need not follow (indeed its attribution is no less questionable
than that of its opposite), unless, of course, strong further evidence for such a view
is given, which Barbour does not appear to offer even at the level of conjecture.
Overall, his argument clearly and perhaps deliberately echoes that of Parmenides
17. Abraham Pais, "Einstein and the Quantum Theory," Reviews of Modern
Physics 51 (1979): 907. Cf. also David Mermin's commentary in Boojums All the
18. Marcel Proust, "The Guermantes Way," The Remembrance of Things Past,
trans. C. K. Scott Moncrieff and Terence Kilmartin (New York: Vintage, 1981),
1988), 2:366.
20. The potential inapplicability of these terms and concepts (rather than the
reported in Aage Petersen, "The Philosophy of Niels Bohr," Niels Bohr: A Cente-
nary Volume, ed. A. P. French and P. J. Kennedy (Cambridge, MA: Harvard Uni-
versity Press, 1985), 305. One must, of course, exercise caution in considering such
Bohr's works that he denies the existence of that to which "the quantum world"
would refer or, again, from (the effects of) which Bohr's "quantum objects" would
be, nonclassically, idealized as that to which no such attributes can be assigned. The
statement may instead be read, especially given the context (the question of whether
quantum mechanics actually represents the quantum world), by putting the empha-
efficacity of the effects in question. Instead, it indicates the inapplicability to the lat-
indivisibility (of quanta themselves), and so forth, or any other physical attributes,
processes-to the "quantum objects" of Bohr's interpretation. Beyond this level (of
Bohr's idealization), even this latter claim may, as I said, be inapplicable. At the
ual phenomenal effects is essential. Both have been at stake in quantum physics
from Planck on. In fact, Planck's law is incompatible with assigning identities to
defines quantum physics for Bohr, primarily in view of the nonclassical considera-
tions here in question. This fact has far-reaching consequences in quantum physics,
from Planck's law on. On the one hand, "identity" in the sense of the interchange-
ability of all particles of a given type (photons, electrons, and so forth) is crucial; on
the other, this identity also peculiarly manifests itself in the impossibility of assign-
ing particles individual identity in certain situations, or perhaps ultimately ever (for
ticles, one can no longer quite speak of the particles of the same type. An investiga-
other types of particles, conceivably all existing types of particles. This is the main
process was somewhat prolonged), which entails this situation, as one of the great-
est discoveries of modern physics, "perhaps the biggest of all the big changes in
arguably most radical available limits, even beyond those of the standard quantum
mechanics of Heisenberg and Schrodinger (31-35). The latter is a complex and lit-
Notes to Pages 45-56 * 253
tie developed subject, which cannot be addressed here without expanding the scope
of this discussion beyond its limits. The circumstances themselves in question, how-
ever, are not only fully consistent with but would reinforce the present epistemo-
logical argument.
21. I shall not address the question of models in physics, a major subject of inves-
tigation in its own right, which would require a separate treatment. The present dis-
22. This is a clear implication of Stephen Hawking's argument in his debate with
Penrose in Stephen Hawking and Roger Penrose, The Nature of Space and Time
24. This argument is indebted to, although different from, Arthur Fine's point in
"Do Correlations Need to Be Explained?" (in Cushing and McMullin, eds., Philo-
more formal viewpoint of quantum logic, most recently in his The Interpretation of
26. Bohr's sense of the situation is of some interest. He opens his remarks with a
as well, provided one redefines, as Bohr does, the phenomena in question accord-
ingly and, correlatively, rethinks what is in fact described and how by quantum
mechanics. However, he traces the emergence of the classical model just described
and the program it entails more specifically to Newtonian mechanics and subsequent
developments (PWNB 3:1). This view is astute in presenting this model both as a
version of Galileo's more general program and as departing from Galileo's own view
and practice.
27. For a relevant commentary see Lawrence Sklar, Physics and Chance: Philo-
tions are often pertinent in situations treated by means of classical statistical physics
alone and complicate such situations considerably, even though chaos theory itself
is classical.
29. See also Pauli's elegant commentary (whose essential points he attributes to
Bohr) in Max Born, The Born-Einstein Letters, trans. Irene Born (New York:
30. There exists the quantum-theoretical concept of state defined via the formal-
ism of quantum theory and specifically the so-called state-vector, a concept bound
by the uncertainty relations. This concept is more significant when the formalism of
Realism appears to remain the main aspiration of most such interpretations. Bohr's
interpretation does not assign, and does not allow one to assign, physical reality to
the state-vector.
254 * Notes to Pages 57-58
31. Arguably the best definitions are those of Bohr (PWNB 1:8) and Heisenberg
Tudor, 1949).
33. It is epistemologically much closer to, indeed virtually the same as (the con-
cept of "phenomena" as such was to come later, but the epistemology is in place) the
version presented in "Discussion with Einstein" than to the Como lecture. Some of
Bohr's critics did not fail to point out these inconsistencies, and some argue that they
undermine Bohr's project, the logic of complementarity, and even that of quantum
mechanics itself, for example, versus Bohm's theory. This type of argument is
1994), and, in part following Cushing, to whom and to whose work she acknowl-
edges a special debt, by Mara Beller as part of their defense of Bohm's theory, or (as
she might argue) in Beller's case, her advocacy of Bohm's theory versus Bohr and the
Bohr is at best a gloss, as Cushing nearly admits himself, and cannot, accordingly, be
of much help, even if one were to accept this gloss qua gloss, which would not be
easy for most who read Bohr (26). Beller's treatment of Bohr is, by contrast, exten-
sive, indeed it takes most of the book. In this author's view (and given the analysis
offered in this study), Beller's analysis of Bohr's key concepts lacks sufficient preci-
sion and discrimination of crucial nuances, which are all the more necessary in view
that grounds much of the argument here. (I leave aside Beller's poorly supported
commentaries on Bohr's personality, which are in obvious conflict with most known
accounts of Bohr, by critics and admirers alike, beginning with Einstein, who was
both.)
tarity (or that there is a conflict between Bohr's earlier and later arguments, as just
indicated), or that Bohr's argument for it (in whatever version) is not inevitable, as
some argue, but rather only consistent with the available data and the formalism of
versions of Bohm's theory, which changed considerably over the years, even in
Bohm's own work.) As I said, it is true that Bohr does not always help either, espe-
cially on the first question. He is, I would argue, more cautious as concerns the
involved (we are not quite finished with them even now), it is hardly surprising that
Bohr had to refine or even correct his argument. Glitches are found virtually until
"Discussion with Einstein," including in his reply to EPR, and I shall indicate some
of them. It is difficult to expect one to fully control such a text, a point often made
by Bohr himself, who famously never stopped revising his work and was reluctant to
publish at almost any stage. The "conflict" between the Como version and the post-
EPR one is that the first, while an extraordinary step, ultimately did not quite work
and had to be refined or even reworked, as just explained. The second ultimately
Notes to Page 59 * 255
does, which is quite enough, even if only as a local interpretation, which is more than
most and perhaps all others are able to offer; certainly it is something that Bohm's
theory fails to offer. This is an immense labor. Beller is right to argue for its multifar-
ious complexity, and she summons much valuable historical evidence and documen-
tation. Her interpretation of this process or, again, of Bohr's work and complemen-
tarity itself, is far less compelling. Admittedly, more needs to be said in order to
ground appropriately this view of her book and make it more than a view of this par-
ticular reader, which cannot be done here. So a view it must remain-on that score.
On the other hand, one can argue much more definitively that Beller's book con-
and related finding, which are, in any event, never properly, or even minimally,
explained in the book but are merely (and barely) referred to (along with some liter-
tum mechanics (or even the data in question themselves) and, hence, as erasing the
main problem of Bohm's theory versus the standard quantum mechanics, which is
local or may be interpreted as local. This view is among the factors that lead her to
nomena, which I shall consider in detail later (Quantum Dialogue, 252-59). Beller
reads this concept through the prism of Bohm's nonlocal quantum wholeness, which
entanglement] and 'nonlocality'" (253). This may be true, if one sees nonlocality (as
the conflict with relativity theory, found in Bohmian mechanics) as revealed by the
case so far. Indeed, Beller admits that Bohr's "reasoning makes sense in an anti-real-
part, something that she persistently (and I would argue, mostly mistakenly) denies
to Bohr; and indeed this point would be enough to justify my claims in this study or
Beller's strong realist agenda in the book; and, however much she tries to separate
herself from naive realism, it remains realism, nevertheless, which, as I said, is per-
haps always naive. She also (correctly) sees Bohr's argument as rejecting nonlocality.
On the other hand, she sees Bohm's nonlocal theory as an equally viable, and (it
related results. That, however, could only be argued were nonlocality (in the sense
cisely in view of Bohr's antirealist interpretation. As will be seen later in this chap-
ter, locality and realism may indeed be incompatible. In any event, nonlocality is not
physicists, beginning with Einstein, who, incidentally, did not care for Bohm's the-
ory, in part for this reason. There may be no alternative on that score, at least for
now.
34. In his The Physical Principles of the Quantum Theory (13, 63-65), Heisen-
256 * Notes to Pages 65-71
berg was rather closer to the mark and to Bohr's later argument in both his great
early papers, introducing, respectively, his matrix quantum mechanics and uncer-
tainty relations. To some degree he was also able to handle more successfully the
argumentations of the type of the Como lecture in his 1929 Chicago lectures, just
ment Paradox," Foundations of Physics, 18, no. 9 (1988): 940-41. There are a vari-
as such. These topics would require a separate discussion. Among the most interest-
ing approaches to these questions from the present perspective are those along the
lines of quantum information theory, further coupled to the so-called Bayesian ver-
tum Computation and Information Transfer, ed. Tony Gonis and Patrice E. A.
36. Henry Stapp is one of the most persistent advocates of this view, which, he
quences of Quantum Theory, ed. Cushing and McMullin, 154-74, and Henry P.
(1997): 300-304. For an effective counterargument, via Bohr's reply to EPR, see
37. Bohr illustrates this argument by his famous drawing showing only heavy,
tum level.
38. The assessment of the relative merits of both interpretations could change
were nonlocality to follow from the formalism of quantum theory or data them-
selves, as some argue. I would contend that such arguments are unconvincing so far
(which, of course, may change, too), as the Mermin-Stapp exchange cited earlier
39. See Bohr's comments in PWNB 3:5-6 and "On the Notion of Causality and
Complementarity" (BCW 7:335). Along these lines (there are considerable differ-
sible is physically possible in quantum mechanics. See Roland Omnes, The Interpre-
40. This concept, accordingly, has nothing to do, as some argue, with David
Atoms." Among very few authors who commented on the significance of Bohr's spe-
cial conception of atomicity is Henry Folse in, among other works, "Niels Bohr's
Modern Physics 1987), cited above. Folse's interpretation of this conception is dif-
ferent from the one offered here. In particular he argues that Bohr, in fact, did not
carry his program far enough insofar as he failed to provide an account of individ-
ual quantum objects themselves. According to the present reading, Bohr instead
argued that his conception of atomicity at the level of phenomena makes such an
said, crucial to all of quantum physics, the "identity" of individual quantum objects,
say, all electrons, is in all rigor only a shorthand for properly correlated individual
closer arguments is that offered in an excellent recent article of Ole Ulfbeck and
Aage Bohr, "Genuine Fortuitousness: Where Did That Click Come From?" Founda-
tions of Physics 31, no. 5 (2001): 757-74. In this case while the authors' handling,
their language, "clicks," is close to that of the present study, their argument, by con-
trast, would be different from the present one as concerns the role given to quantum
aspects of the efficacity of these effects. These aspects, I argue here, are crucial (espe-
cially in the argument concerning the locality of quantum mechanics), even though
and in part because this efficacity must be seen as nonclassically unknowable in the
present sense. In this sense Ulfbeck and Bohr's view appears to differ from Bohr's, as
here read (their article itself is not a reading of Bohr, but an independent argument
cal, rather than spatio-temporal, nature of the key elementary concepts, such as par-
ticles or atoms, especially from the point of view of symmetry and group theory.
This argumentation would also contrast with the present one, even though both
share the view that an attribution of such concepts (or, again, any conceivable con-
cepts) is, in all rigor, impossible at the quantum level. I am grateful to the authors for
42. Hawking and Penrose, The Nature of Space and Time, 60.
43. Abraham Pais, Inward Bound: Of Matter and Forces in the Physical World
45. Most of the better treatments are, unfortunately, technical. See, however,
Roger Penrose, The Emperor's New Mind: Concerning Computers, Minds, and the
Laws of Physics (Oxford: Oxford University Press, 1989), 236-42. The exposition
may require effort, but it is elegant and, in principle, does not require any specialized
(10-14), which, while technical, would convey the point even to a nonspecialist.
47. These considerations are significant even in the case of the so-called spin (a
particle, which has no equivalent or even strict analogue in classical physics), where
we deal with finite-dimensional Hilbert spaces, but especially in the case of standard
There are, however, alternative views as concerns the physical reality of this formal-
ism, found, for example, in Penrose's works cited here, as well as in his more acces-
sible summary in The Large, the Small, and the Human Mind (Cambridge: Cam-
bridge University Press, 1997). On this point, I permit myself to refer to my earlier
The Large, the Small, and the Human Mind, with a Glance Back at The Emperor's
New Mind, The Shadows of the Mind, and The Nature of Space and Time," Post-
Scientist, 681-82.
49. As I said, Einstein did not much care for Bohm's theory either, perhaps pri-
complete description would not agree with the laws of nature" (226). This view
echoes Schrodinger's sentiment that classical ideals and models "cannot do justice to
51. The literature on the subject is massive, and it would not be possible to
address it here. See especially the commentary by John Bell, The Speakable and the
in Ellis and Amati, eds., Quantum Reflections. As several essays in these collections
suggest, Bohr's views and writings encounter considerable resistance among the pro-
ponents of Bohm's approach, who find their inspiration in Einstein's and Bell's cri-
tiques, as, to give two recent examples cited earlier, in Beller's Quantum Dialogue
ter 5. As I have indicated earlier a propos Beller's book, to paraphrase Bohr on EPR's
argument, the trends of these arguments do not seem to me adequate to meet the
case that Bohr's writings present to us. Bohm himself had a more balanced view and
cogent sense of Bohr's argument and work, which he admired, as is clear from most
of his writings, for example, Wholeness and Implicate Order (London: Routledge,
1995). Bohm is, of course, also the author of the classic exposition of the standard
quantum mechanics, Quantum Theory (New York: Dover, 1979), originally pub-
Bohr's, as presented here. For arguably the final version of Bohm's theory itself see
David Bohm and Basil J. Hiley, The Undivided Universe: An Ontological Interpre-
have indicated, while they have attracted some public attention recently, the
Bohmian approach represents a minority view among the physics community (they
Naturally, this fact is not an argument against such views (the nonlocality of
Bohmian mechanics is, in my view) or against this or any criticism of Bohr. For dif-
ferent views of these questions, closer to the present position, see essays by Abner
Shimony, "New Aspects of Bell's Theorem," and Kurt Gottfried, "Does Quantum
Mechanics Carry the Seeds of Its Own Destruction?" in Ellis and Amati, eds., Quan-
tum Reflections, as well as several essays in Cushing and McMullin, eds., Philo-
Fraassen, Fine, and Folse ("Bohr on Bell," which, however, offers a reading of Bohr
that is rather different from the one offered here). See also Mermin's discussion of
these questions in Boojums All the Way Through, 81-185. Gottfried appears to
express the majority view in arguing that relativity in fact provides the experimental
evidence ("test") against Bohmian mechanics and that Bell's theorem(s) and related
theoretical findings further support this evidence in his "Inferring the Statistical
Interpretation of Quantum from the Classical Limits," Nature 405 (2000), 533-36,
a point that extends the argument of his "Does Quantum Mechanics Carry the Seeds
52. See, again, Mermin's essays on the subject in Boojums All the Way Through,
81-185.
53. The arguments for it, such as those by Stapp, mentioned earlier, appeared to
be unconvincing at best.
54. Stapp argues this on several occasions, for example, again, in "Quantum
Nonlocality and the Description of Nature," clearly attributing to Bohr terms and
sible, to sustain such an argument. Beller, too, appears to misconstrue Bohr's depen-
dence on the concept of disturbance, specifically in his reply to EPR (her argument is
different from Stapp's), which is less significant than she and others argue even in his
quantum objects and measuring instruments. While they are important in explaining
the reasons for Bohr's argument, they are fully consistent with the preceding analy-
56. See Fine, "Do Correlations Need to Be Explained?" (in Cushing and
57. See Mermin's comments in Boojums All the Way Through, 108-9.
58. Einstein, The Born-Einstein Letters, 209; see Mermin's discussion in Boo-
59. For a preliminary approach see an article by the present author, "Landscapes
of Sibylline Strangeness."
60. See Linda Wessels, "The Way the World Isn't: What the Bell Theorems Force
260 * Notes to Pages 98-106
Quantum Theory, 80-96, for a useful discussion of possible "options" given Bell's
and related theorems, which, however, does not discuss a "nonclassical" option in
61. See, for example, Shimony's "New Aspects of Bell's Theorem" in Ellis and
Mechanics Is Trying to Tell Us," and references there. Arguments concerning the
been made by others as well, including, interestingly, by Henry Stapp in his earlier
62. See Mermin, "What Quantum Mechanics Is Trying to Tell Us," 765 n. 31.
63. Hans Reichenbach, The Direction of Time (Los Angeles: University of Cali-
64. While the earlier qualifications, especially those concerning nonlocality, must
be kept in mind they would not affect the points made at the moment.
66. I refer primarily to the work of Thomas Kuhn, Imre Lakatos, and Paul Fey-
erabend, and following them the work in, as they became known, the social studies
of science (for example, in the work of David Bloor and his school) and related
developments, in particular the work of Bruno Latour, one of the primary subjects
of the Science Wars debates. I shall comment on these developments in this latter
context in chapter 4.
plexities mentioned earlier. However, the core of the present argument concerning
would be maintained, indeed, I would argue, all the more so once these complexities
68. On these questions I permit myself to refer to an article by David Reed and
65-92.
69. Martin Heidegger, What Is a Thing? trans. W. B. Barton Jr. and Vera
Deutsch (South Bend, IN: Gateway, 1967), 93. Quantum mechanics and comple-
mentarity, as considered here, obviously give new and more radical dimensions to
Heidegger's point, made by him, fittingly, in the context of Galileo's project, but not
70. As Sylvan S. Schweber argues in his QED and the Men Who Made It: Dyson,
1994), in the case of quantum electrodynamics (QED), at a certain point in the his-
tory of quantum physics, it was the persistence in keeping the existing framework,
that paid off. In the case of QED it was, ironically, Dirac, its founder, who gave up
on his creation and believed that yet another radical transformation, similar to that
sary. Schweber speaks of the "extreme conservatism" of the figures mentioned in his
Notes to Pages 106-10 * 261
title in this context and in this sense. From the present perspective, the extreme con-
the other hand, it cannot be seen as necessary in all conditions or at all points, even
conform to this view at the time of its emergence. We can never be certain what will
ultimately pay off. In some respects the creation of modern QED was quite radical
piece of quantum field theory ever since. So the creators or (it was mostly founded
by Dirac and several others earlier) perhaps "saviors" of modern QED, too, were
both extreme conservatives and extreme radicals, just as were the founders of quan-
Philosopher-Scientist, 45-47. It is true that Einstein here refers to Bohr's 1913 the-
ory of the atom, which appeared at the time to hold some promise for a classical res-
olution, even to Bohr then. As we have seen, however, certain nonclassical features
were in place in Bohr's work even then. Einstein's statement refers precisely to
tions, of which Bohr was himself acutely aware at the time as well. This is not unlike
72. Obviously, Milton's conception has its sources as well, especially Lucretius,
who is sometimes also argued, for example and in particular, by Michel Serres, to be
sophical ideas.
Chapter 3
1. Paul R. Gross and Norman Levitt, Higher Superstition: The Academic Left
and Its Quarrels with Science (Baltimore: Johns Hopkins University Press, 1994)
(the second edition with supplementary notes appeared in 1998); Alan D. Sokal,
tum Gravity," Social Text (spring/summer 1996): 217-52; Alan Sokal and Jean Bric-
2. Jacques Lacan, Ecrits (Paris: Seuil, 1966) and Ecrits: A Selection, trans. Alan
Sheridan (New York: W. W. Norton, 1977). Unless indicated otherwise, I refer here
and throughout this chapter to the English edition, Ecrits: A Selection. The follow-
ing other works by Lacan will be considered: "Desire and the Interpretation of
Desire in Hamlet," trans. James Hulbert, Yale French Studies 55/56:11-52; and "Of
The Languages of Criticism and the Sciences of Man: The Structuralist Controversy,
ed. Richard Macksey and Eugenio Donato (Baltimore and London: Johns Hopkins
4. Accordingly, the more strictly Lacanian part of the present essay may be
more difficult for the general reader than ideally desirable. This may be harder to
avoid than in the case of other figures discussed here, complex as their work may be.
sible without major distortions, thus defeating any attempt to give a nonspecialized
version of Lacan's arguments. In this sense Lacan's text may be irreducibly special-
ized-to Lacan. This quality would not be inconsistent with Deleuze and Guattari's
work, and applicable to other figures considered here, and this view is adopted here
however, in part for the reasons just explained. All one can do is to try to make one's
his ambient discourse. This is what I shall attempt here. On the other hand, my argu-
and is, thus, fully available to the general reader. I am grateful to Guy Le Gaufey for
productive exchanges and for permission to read his yet unpublished essay on Lacan,
with the relationships between Lacanian problematics and mathematics and science,
more direct defenses of Lacan, have appeared since the publications of these books,
and I shall comment on some of them later and in the next chapter. For example,
ber 2000: 352-57]) argues that Lacan's use of topology is not as bad or nonsensical
as Sokal and Bricmont or Gross and Levitt make it out to be. Krips sees this use in
allels considered in this study, which may indeed be more appropriate in the case of
books, however, and even their critics still tend to submit to their authority on math-
ematics and science, in my view, too readily, as I shall argue here and in the next two
chapters.
7. The secondary literature on the subject is vast (although more limited when
dealing with the connections between Lacan and mathematics and science) and can
only be minimally addressed in this essay, which centers on Lacan's own key quasi-
which makes one's task of navigating among both Lacan's own ideas and commen-
8. Just in case, I shall give here a brief explanation of the Moebius strip as a
mathematical object. The model of the Moebius strip is a strip of paper with its
edges glued together by turning one of them upside down. Locally, in the vicinity of
Notes to Pages 116-17 * 263
any given point, the Moebius strip is, thus, indistinguishable from a regular (cylin-
der) strip, which one obtains by gluing a strip of paper, while keeping both edges on
the same side. Each point is an intersection of a vertical interval (think of each such
drawn in the middle of the strip (think of it as a single long stalk to be tied to the
sheaf). We can now glue the end intervals of the strip either regularly or twist one of
them into a Moebius strip or tie our stalk accordingly (hence "sheaf" terminology).
No single vertical stalk, or its intersection with a long stalk, is affected in itself, and
the (tied) horizontal stalk is a circle in both cases. The two global objects are, how-
ever, different topologically. If you begin to paint on one side of the Moebius strip,
say, in blue, you will end up painting "both" sides the same color-that is, it only
has one side. Its boundary is topologically a single circle, while the boundary or,
rather, boundaries of a regular (glued) strip are two separated circles. (One can paint
its two sides in two different colors, say, red and blue.) A regular strip is also a sheaf,
but an untwisted one, a trivial sheaf, as mathematicians call it. In general, a circle
and a line interval are replaced by other, usually more complex, mathematical
objects. A horizontal circle or its equivalent is called the "base" of a sheaf, while ver-
tical intervals or their equivalents are called "fibers." The space as a whole is a bun-
dle of "fibers" (hence the "fiber bundle" terminology). Technically, the terms
objects, but these differences are not essential for the present purposes. The essence
is that one can glue a multiplicity of objects into a single object so that a global
"twist" may result, or conversely one can decompose an object such as the Moebius
strip into this type of configuration in order to distinguish it from the untwisted
ones. Leibniz appears to have had the earliest intuitions concerning such objects.
These objects play a major role in contemporary physics, and indeed most field the-
momentous.
Lacan and philosophy, and the philosophers, beginning with Hegel, is a broad and
complex subject, which is, in its full scope, well beyond the limits of the present
analysis. The reader may be referred, for example, to the collection Lacan avec les
philosophes, ed. N. Avtonomova (Paris: Albin Michel, 1990), and the proceedings of
the UNESCO colloquium under the same title. I would argue, however, that the pre-
philosophy.
tre national de la recherche scientifique, 1984), 2:1, 107-247 (which, along with
Study's article itself appears to shape most accounts of the subject by mathemati-
haus et al., Numbers (New York: Springer-Verlag, 1990), which follows Cartan
264 * Notes to Pages 118-25
closely; and David Reed, Figures of Thought: Mathematics and Mathematical Texts
(London: Routledge, 1995). I am also grateful to Barry Mazur, David Mermin, and
11. The method used in the proof-an argument based on the so-called excluded
middle (still the most common and effective form of mathematical proof)-was orig-
inally used by Parmenides and Zeno and formed one of the foundations of dialectic,
12. Maurice Blanchot, The Writing of the Disaster, trans. Ann Smock (Lincoln:
University of Nebraska Press, 1986), 103; emphasis added. Blanchot probably has in
13. I discuss de Man's allegory in this context (and in connection with Bohr's
temology, Quantum Theory, and the Work of Paul de Man," in Material Events:
Paul de Man and the Afterlife of Theory, ed. Barbara Cohen, Thomas Cohen, J.
2000), 49-92. For other aspects of the legend and relevant epistemological links,
especially in the context (highly relevant here) of Descartes's work, see Claudia
14. The relevant aspects of Godel's work and related areas of mathematical logic
ation. The standard collection containing key major works is Philosophy of Mathe-
matics, ed. Paul Benacerraf and Hilary Putnam (Cambridge: Cambridge University
(193-224).
16. This statement involves further complexities, which have a long history of
their own, extending to Richard Dedekind's and Georg Cantor's work, and then to
17. One can define two other operations in vector spaces, the so-called scalar and
vector multiplication. The first, however, always gives us numbers, not vectors; the
second does give us vectors, but ones no longer belonging to the original vector
space, here the real plane. The vector product of two plane vectors gives us vector
spaces, over the field of complex numbers and the so-called (probability) amplitudes,
which are complex numbers. In order, however, to establish correlations with mea-
dictions) one needs to use the squared moduli, which are always positive real num-
bers. Quantum mechanics provides definite rules for how to do this, such as Born's
sequences of this scheme. It is also worth recalling that, while it may deal with that
which cannot be visualized, quantum mechanics, too, uses visualization, either in the
more direct sense of this word or in the sense of intuitive pictures or representations,
closer to German Anschaulichkeit. The "pictures" of that type may be either partial
mind qualifications given in chapter 2), or diagrammatic, such as the so-called Feyn-
19. We can do so for the Riemannian geometries, which have positive curva-
tures. There is only one (flat) Euclidean geometry, but there are many non-Euclidean
21. These are much more entangled, more intricately mixed, than is commonly
well. The confluence of these two titles themselves is of some interest here, and, in
addition, both books deal with the question of textuality in Descartes and beyond.
22. I leave aside the questions of geometrical or topological definition and of our
spatial intuition concerning the idea of dimension itself. It was a subject of profound
investigations, first in the wake of Georg Cantor's introduction of the idea of sets
and, then, in early topology, in particular in the work of Henri Poincare and Luitzen
23. Richard Feynman made interesting comments on this point in describing his
own visual intuition in thinking about quantum objects. See Sylvan Schweber, QED
and the Men Who Made It, 465-66. The present argument obviously does not take
into account the relevant research in psychology and physiology (and to some
degree, sociology) of perception and cognition, which may complicate the situation
and even actually, but "imperceptibly," deploy, more complex perceptual or phe-
about. Conversely, we might not even "see"-that is, may not really be capable of
jects, the research concerning them remains at a relatively early stage and is often
inconclusive at best, and thus is difficult to utilize in the present context. It does not
appear to be in conflict with the present argument. Indeed, it is conceivable that this
266 * Notes to Page 129
research can eventually help to support this argument and even radicalize it. We may
well be ultimately less likely to visualize even two-dimensional real objects than to
be able to improve on our intuition concerning the actual geometry of the complex
plane, let alone on our intuition concerning the behavior of quantum objects. I shall
further comment on some of the complexities concerning real numbers and their
24. I am indebted to Barry Mazur for this particular point and for productive
quite the contrary, in mathematics and science or elsewhere it was crucial to various
aspects of nonclassical thought all along, beginning with the pre-Socratics. Nor, con-
physics) and language, both used nonclassically, play significant roles in quantum
physics as well. My point is that nonclassical thought, at the limit, always irreducibly
again, more accurately, what it relates to. This is why I associate nonclassical
thought here with "algebra" rather than "geometry," even though algebra, too, can
a crucial figure in this context (of the relationships between algebra, geometry,
analysis, and even topology), and his work would require a separate analysis,
specifically via Deleuze's work on Leibniz and the Baroque in The Fold: Leibniz and
1993). On the other hand, on the particular point in question at the moment, see
University Press, 1974), 75-81, and of "the linearity of the symbol" and its invasion
by nonlinear and, hence, nonclassical "writing" in Derrida's sense a bit further in the
book (85-87). It may, however, be argued that, while both algebra and geometry in
significant implications for the very understanding of what would count as rigor-
Kepler and Galileo on, or elsewhere. At the same time, it would be tempting to con-
sider the parallel histories of algebra, geometry, and representation (including alle-
gorical aspects of the latter, ultimately in de Man's nonclassical sense) from their
Medieval roots; to their role in the Renaissance, and the role of algebra (rather than
only geometry) in Galileo; to the invention of calculus in the late Renaissance and
the early Baroque by Newton and Leibniz; to the emergence of modern algebra (in
particular complex numbers and the theory of algebraic equations) in the work of
such figures as Gauss, Niels Henrik Abel, and Evariste Galois, contemporary with
physics, specifically quantum theory. Several works and analyses more directly rele-
Notes to Page 129 * 267
vant to the present argument may be considered from this perspective as well. They
ence and Repetition, trans. Paul Patton (New York: Columbia University Press,
situated at the very heart of Classical thought" (in Foucault's sense) in The Order of
Things, 57, 83. I have addressed some of these issues in the context of de Man's
This is of course an immense program, which can only be indicated here as a pos-
at stake, one can specifically mention investigations, in the wake of Kant, into the
nature of algebra as "the science of pure time." I think, in particular, of the work of
Sir Rowan Hamilton, who was closely aligned with Coleridge and his circle, whose
members not only had strong connections to German Idealism (an obvious point)
but were also specifically interested in questions of the type considered here. Hamil-
ton was especially interested in ideas of symbol and symbolic representation, both in
a more general sense and, I would argue, specifically in Coleridge's sense of the term.
quaternions. For a useful historical account of this invention (from a different per-
spective bypassing the epistemological questions considered here) see Andrew Pick-
ering, "Concepts and the Mangle of Practice: Constructing Quaternious" (in Smith
and Plotnitsky, eds., Mathematics, Science, and Postclassical Theory, 40-82). This
invention can be linked to the philosophical ideas just mentioned, even though some
a kind of "numbers") was the first example of, topologically, a mathematical object
of more than three (in this case, four) dimensions. As such, it in many ways
prefigured and indeed initiated both twentieth-century abstract algebra and such
dimensions), used in quantum mechanics. (Hilbert spaces are usually associated with
work and philosophical ideas were in turn crucial for twentieth-century mathemat-
ics and physics, including, again, in the present context. At the core of Hamilton's
work on quaternions are questions of the relationships between algebraic and geo-
metrical objects and their mutual representation. At the same time, fundamental
on the one hand, and the question of reality in more general philosophical terms, on
the other, were central to Hamilton's thinking on quaternions. The latter themselves
physics), specifically to quantum mechanics. Earlier (in 1834), Hamilton also intro-
duced modern or Hamiltonian, as it has been known ever since, formalism of classi-
cal mechanics, by exploring the analogy between mechanics and optics. It was, as we
268 * Notes to Pages 129-31
have seen, this formalism that served as the primary model for the mathematical for-
ideas were instrumental to the thought of Bohr, Heisenberg, Dirac, and other
questions at stake, and Gauss's and then Riemann's work, and their connections to
Kant and German Idealism, may well be even more fertile cases here than that of
Hamilton.
Gibbs Symposium, Yale University, 1989, ed. G. G. Caldi and George D. Mostow
26. This is yet another question, also having to do with the difference between
curved lines and surfaces in space and the curvature of space itself (with some con-
here. It is of some interest that the complex numbers allow one to construct more
ture (hyperbolic geometries). It can also be added that, according to this view and
the phenomenal world is not disproved merely by virtue of the discovery of non-
Euclidean geometries. One would need to mount a far more complex argument
mathematical Platonism (which postulates, in one way or another, with one degree
own right, which cannot be addressed here, beyond the implications of the present
analysis. The latter itself is reasonably conditional in the sense that, similarly to the
tation of the situation rather than offers an argument against the inevitable nature of
this interpretation, which is not required for my purposes here. As follows from the
preceding analysis, a nonclassical view, such as the one suggested here, cannot be
realist by definition. It may also be pointed out that the Platonist position, however
28. This area has, however, been a subject of intense investigation in the wake of
Penrose's proposal, advanced in his recent works cited earlier. It may be noted that
some of the claims advanced along these lines heavily depend on the arguments for
missing the conjectural nature of these arguments and the lack of experimental evi-
dence supporting them, as considered earlier and as will be further discussed in chap-
29. Gerald Edelman's work is perhaps best known and is, predictably, the sub-
Notes to Pages 131-42 * 269
ject of much controversy and counterargument. His Bright Air, Brilliant Fire: On the
Matter of the Mind (New York: Basic Books, 1992) is arguably his best popular
exposition that also contains key references to philosophical and (some) technical lit-
erature. It also takes issue with Penrose's proposal for the quantum nature of con-
sciousness in The Emperor's New Mind, as discussed earlier (216-18). For the most
recent exposition, see Gerald Edelman and Giulio Tononi, The Universe of Con-
sciousness: How Matter Becomes Imagination (New York: Basic Books, 2000), and
for the original (more) technical presentation, see Gerald Edelman, The Remem-
1990).
30. Thus, Penrose in the works cited earlier conjectures (and hopes) that such
32. Misha Gromov, "Local and Global in Geometry," preprint, October 29, 1999.
34. Laugwitz, Bernhard Riemann, 225-26. Gauss's paper itself, Beitrdige zur
1863-).
36. Euclid, The Thirteen Books of Euclid's Elements, ed. Thomas L. Heath, 3
37. Lacan appears to have been aware of some of these complexities, as is sug-
cited by Sokal and Bricmont in Fashionable Nonsense (27-36). These passages cause
Sokal and Bricmont much aggravation. In truth, however, they are at worst harm-
less, and often there is nothing especially wrong with them-especially, again, if one
tries to understand Lacan's actual argument where these passages are used. On the
other hand, Lacan and other radical thinkers under criticism in the Science Wars
concepts in question and of their history and of the philosophical thought of the key
mathematical and scientific figures involved than are their recent critics in the sci-
entific community. In Lacan's works one finds numerous references to the history of
mathematics and, specifically, algebra and the theory of equations, which indicates
this awareness, however one evaluates his deployment of these ideas themselves.
matical Thinking, trans. Theodore J. Benac (New York: Frederick Ungar, 1951),
107.
40. Here and later the term "analogon" may also be understood in its Greek
ogy, 27-73, which is one of the defining texts in the history of deconstruction.
42. Two recent commentaries address the relationships between Lacan and
quantum mechanics. The first is Slavoj Zizek's "Lacan with Quantum Physics," in
Analysis), ed. George Robertson et al. (London: Routledge, 1996). The second is
Configurations 7, no. 1 (winter 1999): 43-60. Krips follows Zifek's argument but
mechanics in terms of a confrontation with the Lacanian Real. Ultimately, the argu-
ments of both articles turn on how one reads Lacan's Real, which is pertinent. For
Lacan's Real is indeed crucial to Lacan's thought and to the relationships between it
and quantum mechanics. On the present view, both of these reading are ultimately
classical as concerns both Lacan and Bohr, and Krips in fact offers more a reading
Bohr's views, which are not sufficiently elaborated by Krips in any event, as well as
some among the views of Heisenberg (or of the exchange between the latter and
Bohr), or indeed some of the key elements of quantum mechanics. I further think
that the article equally misconceived the transition from the earlier stages of quan-
of quantum mechanics. First of all, some key aspects and refinements of Bohr's inter-
pretation, as considered earlier, emerged following this version, a point entirely dis-
regarded by Krips. Most of the Bohr-Einstein debate and arguments concerning the
EPR argument and nonlocality took place after it as well. Indeed, as I have indicated,
all of the key features of Bohr's interpretation can be reformulated in terms of Von
Neumann's Hilbert-space scheme, as Bohr was well aware, as are indeed most
authors who write on the subject. In fact, this reformulation helps Bohr's post-EPR
operators in Hilbert spaces) of Von Neumann's scheme. Thus, there is no break here
of the kind for which Krips argues, especially if one considers Bohr's post-EPR ver-
sion of complementarity (the difference by and large ignored by Krips). There seems
malism. Krips's claim that complementarity sees quantum mechanics as merely pro-
viding the rule for predicting the outcome of experiments is also mistaken, as this
study would suggest. It does that, of course, but this is hardly sufficient to charac-
terize Bohr's interpretation. Indeed, I was unable to locate the statement "the math-
opposed to meaningful truth claims]," which Krips attributes to Bohr either on page
60 of PWNB 2 (Krips refers to the original edition of which this is a reprint) or else-
where in Bohr. Perhaps the ellipses prevented me from locating it. In reading the arti-
cle one gets an impression that the Bohr quotations come from secondary sources,
since they do not always exactly correspond to the original, and passages from the
same essays, specifically "Discussion with Einstein," are cited from different publi-
Notes to Pages 143-47 * 271
cations (notes 6 and 7, pp. 44, 45). As stated, the above formulation is closer to Ein-
upon by Bohr there (PWNB 2:61). Phenomena in Bohr's sense are unsustainably
seen by Krips as referring to the attributes of quantum objects, and the role of mea-
suring instruments, decisive for Bohr, is ignored altogether, while it is, as we have
seen, crucial to the question of the relationships between complementarity and post-
certain aspects of Bohr's interpretations, despite its decline in theoretical and con-
ceptual terms (alleged by Krips) in, at least as presented in the article, naively psy-
choanalytic terms, is especially unfortunate, and it misses most key aspects of the
history of Bohr's ideas and of their impact. Indeed, while Krips offers this situation
made. Thus, if there is little, if any, Bohr in the article, there is hardly much more of
Lacan. Zizek's argument concerning Bohr is rather looser and, in essence, claims far
less in term of the specifics of Bohr's interpretations. It is, again, another question
44. One must keep in mind the difference between complex numbers, or indeed
any mathematical object, and the Lacanian system in question as concerns their
respective relationships with materiality (whether one sees the latter in terms of real-
ity in the classical sense or otherwise). The situation is in some respects analogous,
but not identical, to that of the difference between mathematics and quantum
mechanics, as considered earlier. In the case of the Lacanian system, the relation-
ships between the symbolic and the material are more immediately germane, some-
what similarly (although not identically) to the way mathematical models function
in physics. In the case of mathematics, the symbolic systems that mathematics uses
well, and, as mentioned earlier, there is of course still the question of nonmaterial
bers, 58.
25.
50. To the extent that one can speak of a metaphorical parallel, it operates at the
272 * Notes to Pages 149-51
level of the two systems themselves. This is a classic Lacanian move, and it is often
found elsewhere as well. For example, Poe's "The Purloined Letter" is read by Lacan
as textualizing the scene and indeed the field of psychoanalysis ("Seminar on 'The
writing in Derrida's sense in "Le facteur de la verite" (The Post Card: From Socrates
to Freud and Beyond, trans. Alan Bass [Chicago: University of Chicago Press,
1987]), as part of his deconstruction of Lacan. In the sense just explained, however,
one can also speak of a certain "repetition" of Lacan on Derrida's part, albeit a rep-
etition in the sense of Derrida's differance as the interplay of differences and simi-
larities, distances and proximities, and so forth. Sokal and Bricmont may be seen as,
at best, confusing the systemic metaphor, as just described, with a direct metaphor
applied to a single element of, or, more accurately, extracted from, the Lacanian sys-
tem. That is, they see Lacan's concept of the erectile organ as defined, metaphori-
cally, through the mathematical J -1, while missing or ignoring the Lacanian system
within which his concept is inscribed (as "the square root of J -1," (L) -1, of his
system) and, hence, also the systemic metaphor in question. They even profess a total
lack of familiarity with the Lacanian system. This indeed did not need to be a factor,
or at least as much of a factor, were their view of Lacan's "metaphor" correct, which
it is not.
"-1" and "the square root of -1"-can be read as "operations" upon the elements
of the "space" of the subject (thus giving the set of these operations itself a certain
53. See, however, the qualifications offered earlier. The question of "negativity"
arate discussion.
54. One can speak of "distancing" here only with caution. Although an
efficacious materiality of the Lacanian Real can be seen as, in a certain sense, more
signifiers can be unequivocally "measured." One can speak in these terms only pro-
visionally. Nor, as was suggested earlier, can the overall efficacity of the Lacanian
signification be contained by this materiality: this efficacity, along with its effects,
"the image of the image of the penis," discussed earlier, must be seen as designating
the "material" and the "phenomenal." However, indeed by the same token, other
55. Lacan, Ecrits, 319. In order to read this passage one would also need to con-
sider the question of sacrifice in Lacan, via Hegel, in particular in the Phenomenol-
sity Press, 1977]), and, then, via Alexandre Kojeve and Georges Bataille, both of
56. The difference between the erectile organ and the phallus would be inscribed
57. It is significant that Lacan says "equivalent" and not "identical," which,
again, suggests the difference between Lacan's "algebra" and that of the actual
mathematical complex numbers rather than a claim of their identity on Lacan's part.
58. It is worth noting, however, that certain key ideas, including key epistemo-
logical ideas, of the figures just mentioned-in particular, Saussure, Hjelmslef, and
26. The reader may be spared the rest of Sokal and Bricmont's sentence, equally
ironic in its confirmation of my point here and equally remarkable in its naivete and
blindness.
26.
63. Gilles Deleuze and Felix Guattari, Anti-Oedipus: Capitalism and Schizo-
phrenia, trans. Robert Hurley, Mark Seem, and Helen R. Lane (Minneapolis: Uni-
Chapter 4
"Nature Faking in the Humanities," Skeptical Inquirer 15, no. 4 (1991): 371-75. It
offers a criticism similar to that of Gross and Levitt, who borrow the passage itself
from this source, rather than from the original, with which they do not appear to
domains, including the popular press in the United States and Europe, on and
around Sokal's hoax has been staggering, even leaving the innumerable exchanges
on the Internet aside. Several Web sites, however, both serve as good bibliographical
sources and contain the texts of articles and exchanges themselves involved. A num-
ber of these sites are listed on Alan Sokal's Web page, <www.physics.nyu.edu/
has figured nearly uniquely throughout these discussions, even though, more
recently, Derrida has overtly moved into the background and other figures have
"Sokal's Hoax," New York Review of Books, August 8, 1996, 11-15, and Steven
Weinberg et al., "Sokal's Hoax: An Exchange," New York Review of Books, Octo-
274 * Notes to Page 158
ber 3, 1996, 54-56, on which I shall comment later. Subsequently Weinberg pub-
lished an article on Kuhn, "The Revolution That Didn't Happen," New York
Review of Books, October 8, 1998, which is a more general response to the current
stage of the drama, and sometimes comedy, of "the two cultures." The article is con-
cerned primarily with the nature of scientific knowledge rather than with the
Two other collections, inspired by Gross and Levitt's book and Sokal's hoax, and
containing contributions by them, are The Flight from Science and Reason, ed. Paul
R. Gross, Norman Levitt, and Martin W. Lewis (Baltimore: Johns Hopkins Univer-
sity Press, 1997), and A House Built on Sand: Exposing Postmodernist Myths about
Science, ed. Noretta Koertge (Oxford: Oxford University Press, 1998). Most of the
articles contained in these volumes carry the problems of Gross and Levitt's and
Sokal and Bricmont's books (exceptions are few and far between). See also The
Sokal Hoax: The Sham that Shook the Academy, ed. Lingua Franca (Lincoln: Uni-
Sokal affair, assembled by the editors of Lingua Franca, where Sokal's hoax was
originally disclosed. The characterization "Science Wars" has been objected to from
both sides on several occasions. I shall continue to use it here, even if only for con-
venience's sake. It is, however, not altogether out of place, and Sokal protests too
much on that score. His own commentaries have never managed to transcend a cer-
tain militancy, if not belligerence, even and sometimes especially when reasonable-
4. Latour's case has a special significance in this context, both on its own terms
and in the broader context of (constructivist) science studies, to which the debate in
question has shifted more recently. These questions would, however, require a sepa-
rate treatment. It would not so much concern commentaries by Gross and Levitt or
Sokal and Bricmont (or, for the most part, those in the partisan collections, cited ear-
lier). These do not have much interesting to say on these subjects, nor would they
require a significant modification of the main ethical points made here. I have in
these areas of the history and philosophy of science. Most of these discussions have
been conducted by prominent mathematicians and scientists, on the one hand (such
as Kurt Gottfried, Michael Harris, Jay Labinger, David Mermin, Stephen Weininger,
and Kenneth Wilson), and, on the other, leading representatives of the constructivist
school in science studies (such as, in addition to Latour, David Bloor, and several
and Simon Schaffer) in scientific journals, such as Nature and Physics Today, and in
Social Studies of Science. See also The One Culture?: A Conversation about Science,
ed. Jay A. Labinger and Harry Collins (Chicago: University of Chicago Press, 2001).
case, "What's Wrong with This Reading?" Physics Today (October 1997): 11-13,
and the subsequent exchange, with some predictable reaction from the prominent
April 1999: 70-71. I also refer to John Huth's "Latour's Relativity" (in Koertge, ed.,
Notes to Page 158 * 275
The House Built on Sand, 181-92), which, while (mostly pertinently and interest-
ingly) critical of Latour, is one of the few, if not the single, exceptions in Koertge's
collection. Otherwise the latter exemplifies some of the worst aspects of the Science
Wars and its aftermath, which is of course not surprising given the list of contribu-
tors. I would also like to mention Michael Harris's letter, "Science Wars or Wars of
and his unpublished essay on these subjects, "I Know What You Mean." See also
Latour's own recent Pandora's Hope: Essays on the Reality of Science Studies (Cam-
bridge, MA: Harvard University Press, 1999). A number of other prominent histori-
ans and philosophers of science contributed to these exchanges on both or, again, all
sides. Ian Hacking's recent The Social Construction of What? (Cambridge, MA:
Harvard University Press, 2000) discusses the Science Wars and the question of
interest as well and, while uneven, are of much higher quality than the Science Wars
collections cited earlier: Yves Jeanneret, L'affair Sokal ou la querelle des impostures
(Paris: Presses Universitaires de France, 1998), and Baudouin Jurdant, ed., Impos-
tures scientifiques: Les malentendus de l'affair Sokal (Alliage #35-36, 1998, coedi-
appears unlikely that some form of the problem of the "two cultures" did not exist
in Plato's time.
6. Obviously we must allow for the differences between these figures and
specific critical points concerning them; hence, I say analogous here. I shall indicate
7. Certain earlier events, such as Derrida's exchange with John Searle, are rele-
vant here. On the exchange with Searle, see Derrida, Limited Inc abc. .. (Evanston,
University Press, 1995), 399-456. Mathematics and science do bring new dimen-
sions to these debates and contribute to their public reverberations, in part in view
of the particular role science and scientists play in modern society. The latter is a
broader question, which would require a separate analysis. Substantively, one does
not find much new or enlightening in most of the arguments by the Science Wars sci-
entists (to some degree, in contrast to the discussions of the Science Wars them-
selves). In terms of their substance, many of these interventions (certainly, Gross and
Levitt's or Sokal and Bricmont's) are irrelevant. In cultural, ethical, and political
terms the case is, again, different, given the unfortunate implications of these inter-
ventions. There is no contradiction between this contention and the argument, given
in the main text, concerning the inseparability, within the present discussion, of
these and conceptual dimensions of the questions at issue. First, this inseparability is
part of a particular argument offered here, an argument involving both ethical and
conceptual issues (in contrast, say, to the previous chapters of this study, which are
276 * Notes to Page 159
more conceptually oriented). Second, while ethical (or political) and conceptual sub-
jects are linked and, perhaps, ultimately indissociable, they are not the same, and at
as those offered by Gross and Levitt or Sokal and Bricmont. In this case we need to
consider both aspects, ethical and intellectual, even though the intellectual contribu-
trans. John P. Leavey (Stony Brook, NY: Nicolas Hays, 1978). Beyond commen-
taries throughout his works, one can especially mention here the as yet unpublished
seminar "La vie la mort," concerned, in Derrida's words, "with a 'modern' prob-
Jacob, Canguilhem, etc.)"'" (Derrida, The Post Card, 259n. 1). A number of com-
and science have been published, including Complementarity by the present author.
It is true that Derrida does not, for the most part, engage mathematics and science in
overt ways, as, for example, Deleuze does. And yet, deeper and more meaningful
connections to mathematics and science are found in Derrida's work. The connec-
tions emerge, for example, by virtue of Derrida's more specific engagements with
humanist and scientific disciplines. They may also be implicit and unperceived by
physics, or algebra suggested by Hyppolite in his exchange with Derrida. Such con-
nections are inevitable and irreducible within what may be called the culture of sci-
entific modernity (it may, again, not be possible to think modernity otherwise), from
Kepler and Galileo (perhaps, again, from Plato and Aristotle, or even the pre-Socrat-
ics) on. As I stressed from the outset, philosophy is multiply interlinked with mathe-
matics and science within this culture or these cultures, and these links proceed from
mathematics and science provide concepts, ideas, and ways of thinking to philoso-
phy (and much else). That sometimes includes ideas of a fundamentally philosophi-
cal nature, as, for example, according to Deleuze, those that Abel and Galois intro-
some finessing. (He should, at this juncture, have also mentioned Lagrange, whose
work was crucial for both Abel and Galois.) His point, however, not only is funda-
mentally correct and well taken but is indeed of crucial importance. Conversely,
from the invention of the proof from contradiction by Parmenides and his school,
philosophy and other nonscientific (if they even are or have ever been strictly non-
scientific) disciplines, or literature, do the same for mathematics and science. These
relationships are and have always been reciprocal, which is, again, not to say that the
trans. Gary E. Aylesworth (Atlantic Highlands, NJ: Humanities Press, 1995), 52.
10. See, again, Conversations with French Philosophers, 52, and "As if it were
Possible: 'within such limits,'" Questioning Derrida: With His Replies on Philoso-
chaos theory" in his chapter "What the Social Text Affair Does and Does Not
Prove" (in Koertge, ed., A House Built on Sand, 9-22). I would welcome here an
Deleuze and Guattari are better or "worse in context than out of context," when
they speak about chaos in What Is Philosophy? which incidentally is an exact quo-
Sand, 12). Note that, unlike Sokal, I say chaos, not chaos theory, and I have good
To slow down is to set a limit of chaos to which all speeds are subject, so that they
form a variable determined as abscissa, at the same time as the limit forms a uni-
versal constant that cannot be gone beyond (for example, a maximal degree of
contraction). The first functives are therefore the limit and the variable, and ref-
erence is a relationship between values of the variable or, more profoundly, the
relationship of the variable, as abscissa of speeds, with the limit. (What Is Philos-
ophy? 118-19)
This may be a difficult passage to understand without reading the two-page elab-
oration of which it is the conclusion and without perhaps spending some time with
the book as a whole, and it may appear as nonsense to those who, having not done
so, do not understand it and in part because they do not understand it. Formidable
as this passage may be, however, even a cursory examination of the paragraph from
which this quotation is extracted by Sokal would show that he is in error in think-
ing this passage to be on chaos theory. The paragraph, including the part cited by
him, and much of the book is indeed about chaos, but it is decidedly not about chaos
theory. It is about a particular idea of chaos and how it functions differently in phi-
losophy and science-science in general, not chaos theory-and how this difference
helps us to understand the difference between philosophy and science in general. The
first sentence of that paragraph says: "The primary difference between science [sci-
attitudes toward chaos" (117). The next sentence expressly defines chaos in terms
that have self-evidently little, if anything, to do with "chaos theory." It defines chaos
an infinite speed of birth and disappearance [of forms]" (What Is Philosophy? 119).
278 * Notes to Page 160
Obviously, the passage would make no sense if one does not explain the key
terms used in it, such as chaos or "functives." The latter are defined in the preceding
page quite simply (nothing obscure or "postmodern" here) as the elements of which
the passage makes no sense as a comment on chaos theory because it is not a com-
ment on chaos theory. It would take a considerable amount of space (which I do not
have here) to explain Deleuze and Guattari's ideas and the elaboration itself, which,
as all good philosophy, requires fast thinking but slow reading. My point is that the
passage cited by Sokal is unquestionably not about "chaos theory," to which the
book briefly refers much later in a different context (206). However undeliberately,
Sokal's reading not only takes the quotation out of context, but it deprives it of its
content. There is nothing in the quotation itself that suggests chaos theory, except
The point seems so obvious that it is difficult for me to conceive how Sokal could
make an error here. Naturally, I would not presume that he himself has as hazy a
notion of what "chaos theory" actually is as, regrettably, do many humanists who
have commented on it. As I have indicated, chaos theory is, for the most part, a the-
most such systems are in fact causal, they can, in practice, appear random-
sense, catchy as it is, the term "chaos theory" is somewhat of a misnomer. It might
have been better called "order theory" or "complex-order theory," and it is indeed
close to the so-called complexity theory, another fashionable and much misused item
trast, are interested precisely in chaos and offer a specific new concept of it. There is,
of course, no single concept of chaos, assuming that we can conceive of it, to begin
with.
It would only take a little more argument to show similar problems throughout
the chapter on Deleuze in Fashionable Nonsense and throughout the book on most
other authors, and many of these problems were indeed made apparent by a number
of commentators and reviewers since the book appeared. Finally, I ought to mention
that the argument just offered was presented at the conference entitled "Debate on
Science, Science Studies, and Their Critics," at the University of California, Santa
Cruz, in May 1997, where the present author was paired with Sokal, who presented
the identical version of his comments just discussed. This, however, obviously did
not make him change these comments in the subsequent publication of his article.
13. On these issues in a more general context, see Derrida's analysis in "Limited
394 [1998]: 141-43) and Thomas Nagel ("The Sleep of Reason: Review of Alan
Sokal and Jean Bricmont, Fashionable Nonsense," The New Republic, October 12,
checking the original texts with any care (if at all) and without stopping to think
even minimally about the content of the quotations. These quotations are often
with the texts is pronounced, goes so far as to call Sokal and Bricmont's commen-
ish"). These commentaries are clearly nothing of the kind, regardless of whether or
not some of these postmodernist works are gibberish. Patience is what these authors
and these reviewers lack above all. Such reviews, just as the books under review,
depend on and perpetuate the ignorance of their readers and can hardly be of much
service to the latter or contribute to any productive debates, regardless of the prob-
lems of some among the postmodernist figures under discussion. To encounter this
from prominent scientists, philosophers, and scholars is sad but, unfortunately, not
surprising. This is not altogether new either, and earlier examples are not hard to
come by. There have been more balanced assessments of these books or, as I said, of
tists, without necessarily defending the "targets" of these books, which makes these
assessments all the more valuable in this context. In general, as I said, in question
here is not a defense of these targets as such, but a defense of the ethically and intel-
lectually proper conduct of the discussion, as well as an argument for more substan-
tive arguments concerning the key conceptual issues involved. As I have stressed
essential even if (by virtue of not being always sufficiently familiar with modern
mathematics and science) the "postmodernist" works miss some of the philosophi-
cal content and certain implications of mathematics and science. Such a critique
here-a very different form of exchange would then emerge. A characteristic exam-
on Bruno Latour's Versions of the Seventeenth Century" (in Koertge, ed., House
Built on Sand, 240-54). Jacob not only misses that, in Latour's We Have Never Been
Modern, trans. Catherine Porter (Cambridge, MA: Harvard University Press, 1993),
which she considers (she gives the wrong publication data [252n. 2]), Latour
expressly dissociates his framework from "postmodernism," as indeed his title indi-
cates, but she also misses that he criticizes Steven Shapin and Simon Schaffer's
Leviathan and the Air Pump: Hobbes, Boyle and the Experimental Life (Princeton,
NJ: Princeton University Press, 1985). This can hardly give one much confidence in
15. See also his response to the articles, "As if it were Possible, 'within such lim-
16. These relationships may be, and often are, seen in terms of metaphor rather
than, as here, in terms of conceptual (in Deleuze and Guattari's sense) junctures.
Some among recent treatments along these lines tend to suppress the complexity
ability. In fact, in both cases at stake is a specific concept that is quite different from
latter. On these subjects, the reader may be referred to well-known works by Der-
rida (such as "White Mythology," Margins of Philosophy) and de Man ("The Epis-
Nietzsche's great pioneering work, "On Truth and Lying in an Extra-Moral Sense."
Related commentaries are found in Rodolphe Gasche in The Tain of the Mirror
(Cambridge, MA: Harvard University Press, 1986) and The Wild Card of Reading:
On Paul de Man (Cambridge, MA: Harvard University Press, 1998) and by the pres-
17. I am, again, not saying that there have not been different discussions, such as
those I mentioned earlier, which ask legitimate and important questions concerning
the usage and treatment of mathematics and science in the humanities and the social
18. Accordingly, all that I claim with any definitiveness is the existence of such
philosophical relationships between Derrida's ideas and relativity-and that the pos-
nature of my argument was entirely missed both by Sokal and Bricmont (Fashion-
able Nonsense, 263n.113) and by Gross and Levitt (Higher Superstition [2d ed.],
293) in their comments on the earlier version of the present article, published in
themselves reflective, sometimes comically, of the Science Wars situation, but I shall
tarity, where, however, I have concentrated on Derrida's practice and the philo-
that one encounters here only a loosely metaphorical analogy rather than a rigorous
pleteness theorems, are used by Derrida. Many key features of the Gbdelian
tarity argues for a more comprehensive philosophical parallel between Bohr's com-
ingly that study and a companion work, "Complementarity, Idealization, and the
significance.
Notes to Pages 167-69 * 281
21. Gross and Levitt neither offer any support nor retract any of these accusa-
tions in responding to the criticism of their work, including in their response to the
23. All they manage to come up with in reply to this criticism in the second edi-
tion is a quotation on topology not by Derrida but by John Law, "an ally of Bruno
Latour" (which is not altogether accurate either). Law's comment is, it is true, not
altogether coherent but is basically benign. This quotation, moreover, is taken from
human mind. I studied differential topology at the University of Leningrad with two
cians would know these names and those of other figures just mentioned, and it is a
pity that nonmathematicians do not know them-a subject that would require a sep-
and development of this field was extraordinary, from such founding figures as
Henri Poincare to the extraordinary contributions, throughout the first half of this
century, of such figures as Elie Cartan, Jean Leray, Henri Cartan, Jean-Pierre Serre,
Rene Thom, and many others, and then by their younger followers up to the present.
I mention the French names (mathematicians from other countries also made major
contributions) because key developments to which they contributed took place when
Hyppolite, a key figure for my discussion, was first a student at the Ecole Normale
and then a professor at the Sorbonne, the Ecole Normale, and the College de France,
where many of these figures were Hyppolite's fellow students and then colleagues.
Derrida was a student at the Ecole Normale (where he studied with Hyppolite)
around the time of major breakthroughs in the field, which were widely discussed in
the intellectual community to which he, Hyppolite, and other philosophical figures
mentioned here, such as Michel Serres, belonged. This community also included
major historians and philosophers of science. Some of these figures (it is a well-doc-
umented fact) were also interested in contemporary philosophy and other disci-
plines, including Althusser's and Lacan's work, and discussed mathematics and sci-
ence with the latter figures as well. Michel Serres, for example, was not only an avid
reader of Jacob's and Monod's work (both Nobel Prize laureates in biology), but he
knew Monod well and discussed science extensively with him. This is leaving aside
that Serres was educated in mathematics, science, and (naval) engineering and that
at some point of his life taught mathematical logic at the Sorbonne. This is immea-
surably more than the likes of Gross and Levitt or Sokal and Bricmont can claim as
credentials in any field in the humanities, let alone what they refer to as "postmod-
able for anyone even remotely familiar with the intellectual environment just indi-
24. Derrida, Acts of Literature, ed. Derek Attridge (New York: Routledge, 1992),
208-9.
282 * Notes to Pages 171-78
25. See most specifically his "Differance" (in Margins of Philosophy, 1-28) as
Alan Bass (Chicago: University of Chicago Press, 1980), and Writing and Difference,
26. Jacques Derrida, "Khora," in On the Name, ed. Thomas Dutoit (Stanford,
27. Accordingly, the fact that some of the passages they cite are at best "junk"
thought is irrelevant. One could argue (and Sokal and Bricmont did at some point)
that one would need to argue against (or for) what they have to say, author by
author, point by point. I do not think, for the reasons just explained, that they, or
Gross and Levitt, have earned the right to expect or demand this, even if some of
their targets are worthy of criticism or even laughter (some are). Nothing in their
this assessment, and in fact some of the collections cited here are diminished by the
be. What was already cited and referred to here is, I think, quite enough to prove my
point. I would, accordingly, stand by my assessments of these authors. The reader is,
which may, again, be critical with the work of Derrida and other nonclassical
28. An extensive survey of such problems has been published by Roger Hart in
"The Flight from Reason: Higher Superstition and the Refutation of Science Stud-
ies," in Science Wars, ed. Andrew Ross (Durham, NC: Duke University Press, 1996),
259-92.
29. This attitude is found, for example, in Hart's article, just cited, which is
sometimes also problematic in its attempt to defend the authors, whose work on sci-
ence is, in my view, indefensible. Hart is of course right to argue that many of these
30. Both the essay and the discussion are in Languages of Criticism. The essay,
31. This point of the necessity of understanding both terms is clearly brought
into the foreground by Steven Weinberg in "Steven Weinberg Replies" (New York
Review of Books, October 3, 1996, 56). Weinberg there also qualifies his original
remarks on Derrida somewhat (without changing his view) and comments on the
my argument here. I also leave aside for the moment the problem of translation, even
though it is significant. Thus, the translation of Derrida's essay published in the con-
ference volume has several problems, and one is better off reading the version pub-
lished in Writing and Difference. In particular, the version in the conference volume
stantial context of the statements at issue. This context may make any claim con-
cerning these statements, including any claim to be offered here, irreducibly tenta-
32. The very disagreement between Sokal's and Gross and Levitt's interpreta-
tions suggests that a more careful reading may be necessary. Of course, Sokal's arti-
anything, and it can be shown that it misrepresents (deliberately or not) virtually all
Derrida's remark makes no sense whatsoever given Hyppolite's question and Der-
rida's essay. It is strange that several scientists appear to have accepted this interpre-
tation on the basis of an admitted hoax, especially since, as Weinberg points out, this
is not a standard term in physics. This makes Weinberg (in this case more under-
standably) puzzled about the phrase and makes him suggest the meaning of the
phrase as, again, referring to a numerical constant, that of Newton's constant figur-
ing in Einstein's theory ("Steven Weinberg Replies," 56). He does not appear to
attribute this meaning to Derrida, which indeed would not make any sense. Yet
another reading proposed by some scientists, that of the so-called cosmological con-
stant appearing in certain versions of relativistic cosmology, makes even less sense,
his early cosmological investigations but was quickly abandoned by him. (He even
spoke of its introduction as the greatest scientific mistake of his life.) It was resur-
invoked at the time of the Hyppolite-Derrida exchange in 1966. Nor does it appear
to make much sense as Hyppolite's reference, given what he says here or given Der-
38. I am grateful to Joshua Socolar for his suggestions in clarifying this particu-
lar point. It must be kept in mind that, in Bohr's formulation, "the space-time coor-
dination of different observers never implies reversal of what may be termed the
quantum physics, Einstein's relativity remains a causal and otherwise classical phys-
ical theory, at least special relativity does, since these questions-causality, reality,
and so forth-become more complex in the case of Einstein's general relativity, his
bewegter Korper" (On the electrodynamics of moving bodies), one may firmly con-
jecture the following on the basis of the available experimental evidence: "[T]he
same laws of electrodynamics and optics will be valid for all frames of reference for
which the equations of mechanics hold good. We will raise this conjecture (the pur-
port of which will thereafter be called the 'Principle of Relativity') to the status of a
postulate, and also introduce another postulate, which is only apparently irreconcil-
284 * Notes to Pages 185-88
able with the former, namely that light is always propagated in empty space with a
definite speed c which is independent of the state of motion of the emitting body"
sity Press, 1979], 281-82). Einstein's "reconciliation" of these two apparently irrec-
oncilable postulates within the framework of special relativity was his great achieve-
ment.
40. See the exchange between Richard Crew and the present author, "An
comments, which, while critical of Derrida and the present author alike, are in sharp
contrast to the standard Science Wars criticism (and some defenses) of Derrida. Sub-
41. There are in turn complex reasons, physical and philosophical, that com-
pelled Newton to introduce his notions of absolute space and absolute time.
43. These questions, as well as Derrida's early work, have, again, significant con-
what he calls the Other (Autrui), to which I shall return, via Blanchot (Levinas has
been crucial for Derrida as well), in the conclusion and which has been the subject
of many recent discussions. In a rough outline, following primarily his 1961 Total-
ity and Infinity, trans. Alphonso Lingis (Pittsburgh, PA: Duquesne University Press,
1990), arguably his most influential work, Levinas's phenomenology and epistemol-
ogy of the radical alterity of the Other may be viewed as follows. This radical alter-
ity, the emergence of the (effect) of the Other in, or even on the horizon of, the phe-
nomenal, of the radar of our thought, radically restructures the latter by the power
of its effects, especially in the ethical context. Indeed one may see the effects in ques-
tion as the opening of ethics for Levinas. The ultimate locus of the Levinasian other-
ness itself (if such or, again, any terms may ultimately apply) may remain indetermi-
with others (other individuals, cultures, and so forth), even though the role of the lat-
ter may be decisive in the emergence of these effects. One may speak of a certain
materiality in this respect, by analogy with quantum mechanics, which, for example,
originally made physicists rethink the nature of light along these lines. To some
degree this is true even as concerns (more classical) relativity, which alienated light
from the classical theory, or indeed from the (classical thought of the) Enlighten-
ment. Indeed this "alienation" of light in physics was happening just around the time
when Levinas, for whom light is a crucial figure (including in de Man's allegorical
sense), began to develop his ideas, originally through his encounters with Husserl's
phenomenology and Heidegger's work. See especially the section "The Violence of
nological effects that is more crucial, while the nature of the efficacity, even if ulti-
mately material (which is not always clear in this context), is best understood along
(or closer to) the lines of the nonclassical epistemology of effects as considered here.
Levinas's infinity entails apparently (and from the classical standpoint inevitably)
paradoxical imperatives. Most crucial of them is that of placing the radical, irre-
ducible alterity of the Other (Autrui) (that which is irreducibly unavailable to sub-
jectivity) in the position of something that is irreducibly linked to and indeed is the
efficacity of subjectivity rather than something that is absolutely excluded from it, is
(techno) phenomena, albeit, in Levinas, under the opposing name of infinity (vs.
the efficacity of effects in question. This argument opens or enters an immense field
of questioning. Such questions would, for example, concern how nonclassical Lev-
inas's thought ultimately is, that is, whether his particular concept of radical alterity
(a term used by others, specifically Derrida, via Levinas but in more determinately
to the way early Heidegger does, according to Derrida (Of Grammatology, 19-20).
responds to nonclassical theories and encounters with his works, such as Derrida's.)
Another central, and to some degree correlative, question would be how strictly phe-
just outlined is or can be, that is, whether it is deconstructible into an irreducibly
also crucial in this context. Finally, the connections between Levinas's ideas and new
science, such as relativity and quantum physics, while, at some level, inescapable,
require a careful analysis. These questions cannot be addressed here. Nor can I con-
sider here the, by now, massive literature on Levinas. Among more recent works,
Naas and Pascale-Anne Brault (Stanford, CA: Stanford University Press, 1999),
remains, arguably, the most cogent and effective, at least in the present context. My
and its nonclassical aspects and potential. From this point of view, it may well be
just as it happens (more overtly) in Heidegger, even if not in Husserl, that Levinas's
investigation ultimately arrives at, even if against himself, at least as he saw his pro-
ject initially or even by the time of Totality and Infinity. (It is more pronounced in
his later works.) On the other hand, Derrida's work offers us cases of more deliber-
ately techno-textual nonclassical theory, and, as I argue here, in all rigor, no theory
45. Gilles Deleuze, Cinema 2: The Time-Image, trans. Hugh Tomlinson and
47. This type of dialectical synthesis is sometimes (and not always accurately)
associated with Hegel, a subject that would, however, require a separate analysis.
48. Michel Serres and Bruno Latour, Conversations on Science, Culture, and
Time, trans. Roxanne Lapidus (Ann Arbor: University of Michigan Press, 1995), 35.
Bell, John R. von Sturmer, and Rodney Needham (Boston: Beacon Press, 1969),
221-27. I am grateful to David Reed for reminding me about this fact and, again, for
50. Here I use this, by now complicated, term "deconstruction" in the sense of
the analytical practice of Derrida's own (mostly earlier) work, such as "Structure,
Sign, and Play." This is not the place to consider the "continuities" and "disconti-
nuities" in Derrida's work over the last thirty years nor the differences in the ways
this work is received on different sides of the Atlantic. These factors are relevant to
framework, say, around a given concept, and the centering of the structure(s) con-
52. These questions of course also open the general problematic, introduced ear-
lier in this study, of how we interpret classical physics, which would require a sepa-
rate treatment.
55. On these points I permit myself to refer to my book In the Shadow of Hegel,
84-95, 380-88.
57. In any event, they seem to me more likely than those that Crew suggests (par.
10).
60. Cf., again, Derrida's comments in "Sokal et Bricmont ne sont pas serieux."
Chapter 5
1. These differences are one of the main subjects of Deleuze and Guattari's
nesota Press, 1984). He argues that, if one wants to follow mathematics and science
or what they tell us about nature and mind, in the way, say, the Enlightenment fol-
lows classical mathematics and physics, one might at least ask and examine carefully
what they are "telling us" itself becomes a matter of interpretation and debate, such
as those involving the "two cultures" or those of the Science Wars. The stakes are
ceptual architecture). One can, however, also think of Parmenides's invention, men-
tioned earlier, of the proof from contradiction as a great reciprocal example of both.
8. On the other hand, Weinberg's strong claim, "And both the philosopher
physics" (12), does not appear to be based on familiarity with their usage of the term
"time." In any event this claim is not supported by textual evidence. Indeed, it would
be naive to think that there is a simple or single view of time in modern physics itself,
10. If one judges by their more recent publications, however, such as Norman
Levitt's recent book Prometheus Bedeviled: Science and the Contradictions of Con-
temporary Culture (New Brunswick, NJ: Rutgers University Press,1999), that goal is
hardly in sight.
12. Still more recent work in these areas, including by DGZ, does not change
Gross and Levitt's treatment of the subject (and the latter treatment does not take
these subsequent developments into account in any event). For more recent refer-
ences, see Sheldon Goldstein, "Quantum Theory without Observers" (parts 1 and
2), Physics Today, March 1998, 42-46, and Physics Today, April 1998, 38-42, and
for the subsequent exchange see Physics Today, February 1999, on which I shall
comment later.
13. As I have pointed out, some argue that Schrodinger's equation is determinis-
tic. For the reasons explained already, such a view does not appear to me sustain-
able, at least not without much qualification, usually not provided. It can, perhaps,
288 * Notes to Pages 210-12
point, once the initial conditions are given. This is important and rightly stressed, for
from this fact that the behavior of quantum objects themselves is causal, in all rigor,
even if one assumes that the mathematical formalism of quantum mechanics, such as
Schrbdinger's equation, describes this behavior. Either one of these two assumptions
ably predictable manner, for example, by using Schrbdinger's equation. The out-
comes of the experiment are never certain but are constrained by uncertainty rela-
tions that are inherent in Schrbdinger's equation. Obviously, one must keep in mind
these qualifications would not change the present argument, which only concerns the
fact that the claim of the deterministic character of Schrbdinger's equation requires
15. This last point may also be made about Jean Bricmont's commentaries on
may, again, be noted that it appears that, among just about all of the adherents of
Bohmian mechanics, Bohm himself is the one who does justice to Bohr's views. I
leave aside the question of whether Goldstein's exposition may be seen as a definitive
ments. It is quite clear, at the very least, that not all Bohmian physicists or scholars
17. See, specifically, his commentary in "Bertlmann's Socks and the Nature of
Reality" (in The Speakable and the Unspeakable, 154-55), arguably his most
18. Mara Beller, "The Sokal Hoax: At Whom Are We Laughing?" (Physics
Today, September 1998, 29-34). The article is followed by the exchange in Physics
20. Ibid. I also permit myself to refer to Complementarity, where this point is
21. Robert Griffiths and Roland Omnes, "Consistent Histories and Quantum
might want to consult Roland Omnes's books cited earlier, as well as Griffiths and
Notes to Pages 213-39 * 289
Omnes's article itself, technical but still useful to a nonspecialist. It gives a far more
22. It is worth stressing that Laplace was, at most, skeptical as concerns the pos-
Wigner replaces the cat with his hypothetical friend, known as "Wigner's friend"
ever since.
24. See Carlo Rovelli's article, "'Incerto Tempore, Incertisque Loci': Can We
25. On some of these arguments see, for example, the works by Leggett,
28. Percy Bysshe Shelley, Shelley's Poetry and Prose, ed. Donald Reiman and
29. I cannot discuss these works here, but if one wants to look for other exam-
type) outside deconstruction, one can think of, among unexpected places, Godel's
more readily in Imre Lakatos's works, especially Proofs and Refutations: The Logic
30. Indeed, the point is rarely, if ever, made, including by Heisenberg himself,
who in his subsequent works, especially those philosophically oriented, moves in yet
other directions.
32. De Man, Aesthetic Ideology. I have discussed de Man's work in this context
33. We recall that the nature of quantum probability is in turn nonclassical and
Conclusion
2. The omitted part of the quotation would slightly modify but not undermine,
Avtonomova, N., ed. Lacan avec les philosophes. Paris: Albin Michel, 1990.
Barbour, Julian B. The End of Time: The Next Revolution in Physics. Oxford:
Bell, John. The Speakable and the Unspeakable in Quantum Mechanics. Cambridge:
1998, 29-34.
Bloom, Harold. The Anxiety of Influence. Oxford: Oxford University Press, 1997.
Bohm, David, and Basil J. Hiley. The Undivided Universe: An Ontological Interpre-
Press, 1987.
Born, Max. The Born-Einstein Letters. Translated by Irene Born. New York:
Walker, 1971.
Quantum Theory: Reflections on Bell's Theorem. Notre Dame, IN: Notre Dame
. The Fold: Leibniz and the Baroque. Translated by Tom Conley. Minneapo-
Translated by Robert Hurley, Mark Seem, and Helen R. Lane. Minneapolis: Uni-
1997.
Derrida, Jacques. Acts of Literature, ed. Derek Attridge. New York: Routledge,
1992.
his Replies on Philosophy, ed. Michel Meyer. London: Ashgate, 2000, 96-119.
Press, 1995.
1980.
. The Post Card: From Socrates to Freud and Beyond. Translated by Alan
1995.
Edelman, Gerald. Bright Air, Brilliant Fire: On the Matter of the Mind. New York:
Edelman, Gerald, and Giulio Tononi. The Universe of Consciousness: How Matter
Ellis, John, and Daniele Amati, eds. Quantum Reflections. Cambridge: Cambridge
Euclid, The Thirteen Books of Euclid's Elements, ed. Thomas L. Heath, 3 vols. New
of Modern Physics 1987, edited by Pekka Lahti and Peter Mittelstaedt, 161-80.
Gallo, Ernest. "Nature Faking in the Humanities." Skeptical Inquirer 15, no. 4
(1991): 371-75.
Gasche, Rodolphe. The Tain of the Mirror. Cambridge, MA: Harvard University
Press, 1986.
Gauss, Carl Friedrich. Werke. . . Herausgegeben von der K. Gesellschaft der Wis-
Gonis, Tony, and Patrice E. A. Turchi, eds. Decoherence and Its Implications in
Gottfried, Kurt. "Inferring the Statistical Interpretation of Quantum from the Clas-
Griffiths, Robert, and Roland Omnes. "Consistent Histories and Quantum Mea-
Gromov, Misha. "Local and global in geometry," preprint, October 29, 1999.
Gross, Paul R., and Norman Levitt. Higher Superstition: The Academic Left and Its
Quarrels with Science. Baltimore: Johns Hopkins University Press, 1994. 2d. ed,
Gross, Paul R., Norman Levitt, and Martin W. Lewis, eds. The Flight from Science
Hacking, Ian. The Social Construction of What? Cambridge, MA: Harvard Univer-
Hart, Roger. "The Flight from Reason: Higher Superstition and the Refutation of
Hawking, Stephen, and Roger Penrose. The Nature of Space and Time. Princeton,
Press, 1983.
Press, 1979.
Quantum Theory and Measurement, edited by John Archibald Wheeler and Woj-
ciech Hubert Zurek, 62-86. Princeton, NJ: Princeton University Press, 1983.
. The Physical Principles of the Quantum Theory. Trans. Carl Eckart and
. "Quantum Theory and Its Interpretation." In Niels Bohr: His Life and
Jeanneret, Yves. L'affair Sokal ou la querelle des impostures. Paris: Presses Universi-
Koertge, Noretta. ed. A House Built on Sand: Exposing Postmodernist Myths about
2000): 352-57.
Bibliography * 295
Labinger, Jay A., and Harry Collins. The One Culture?: A Conversation about Sci-
1977.
Whatever." In The Languages of Criticism and the Sciences of Man: The Struc-
Lacour, Claudia Brodsky. Lines of Thought: Discourse, Architectonics, and the Ori-
Lahti, Pekka, and Peter Mittelstaedt, eds. Symposium on the Foundations of Mod-
Lakatos, Imre. Proofs and Refutations: The Logic of Mathematical Discovery. Cam-
Latour, Bruno. Pandora's Hope: Essays on the Reality of Science Studies. Cam-
E.P.E.L., 1991.
Bell, John R. von Sturmer, and Rodney Needham. Boston: Beacon Press, 1969.
Lingua Franca, eds. The Sokal Hoax: The Sham That Shook the Academy. Lincoln:
Macksey, Richard, and Eugenio Donato, eds. The Languages of Criticism and the
Mermin, David. Boojums All the Way Through. Cambridge: Cambridge University
Press, 1990.
66 (1998): 920-24.
. "What's Wrong with This Reading?" Physics Today, October 1997, 11-13.
Milton, John. Paradise Lost. Edited by Scott Elledge. New York: W. W. Norton,
1993.
Nagel Tomas. "The Sleep of Reason: Review of Alan Sokal and Jean Bricmont,
Nietzsche, Friedrich. The Birth of Tragedy and the Case of Wagner. Translated by
1974.
Rhetoric and Language, edited and translated by Sander L. Gilman, Carole Blair,
and David J. Parent (New York: Oxford University Press, 1989), 246-57.
Press, 1999.
Pais, Abraham. "Einstein and the Quantum Theory." Reviews of Modern Physics 51
(1979): 861-907.
1991.
. Subtle Is the Lord: The Science and the Life of Albert Einstein. Oxford:
Penrose, Roger. The Emperor's New Mind: Concerning Computers, Minds, and the
. The Large, the Small, and the Human Mind. Cambridge: Cambridge Uni-
Petersen, Aage. "The Philosophy of Niels Bohr." In Niels Bohr: A Centenary Vol-
Theory, and the Work of Paul de Man." In Material Events: Paul de Man and the
2000.
H. Smith and Arkady Plotnitsky. Durham, NC: Duke University Press, 1997.
. "The Cosmic Internet: A Review of Lee Smolin's The Life of the Cosmos."
and the Human Mind, with a Glance Back at The Emperor's New Mind, The
Shadows of the Mind, and The Nature of Space and Time." Postmodern Culture
Plotnitsky, Arkady, and Richard Crew. "An Exchange: Richard Crew and
electronically.
Proust, Marcel. A la recherche du temps perdu. Paris: Gallimard, 1988, 2 vols. The
298 * Bibliography
Routledge, 1995.
Press, 1956.
Rovelli, Carlo. "'Incerto Tempore, Incertisque Loci': Can We Compute the Exact
Rozental, Stephan S., ed. Niels Bohr: His Life and Work as Seen by his Friends and
1949.
Schweber, Sylvan S. QED and the Men Who Made It: Dyson, Feynman, Schwinger,
Serres, Michel, and Bruno Latour. Conversations on Science, Culture, and Time.
Shapin, Steven, and Simon Schaffer. Leviathan and the Air Pump: Hobbes, Boyle,
and the Experimental Life. Princeton, NJ: Princeton University Press, 1985.
Shelley, Percy Bysshe. Shelley's Poetry and Prose. Edited by Donald Reiman and
Smith, Barbara H., and Arkady Plotnitsky, eds. Mathematics, Science, and Postclas-
Smolin, Lee. The Life of the Cosmos. New York and Oxford: Oxford University
Press, 1997.
217-52.
Sokal, Alan, and Jean Bricmont. Impostures intellectuelles. Paris: Odile Jacob, 1997.
Ulfbeck, Ole, and Aage Bohr, "Genuine Fortuitousness: Where Did That Click
don, 1991.
Van der Waerden, B. L., ed. Sources of Quantum Mechanics. Toronto: Dover, 1968.
Weinberg, Steven. "The Revolution That Didn't Happen." New York Review of
56.
Weinberg, Steven, et al. "Sokal's Hoax: An Exchange." New York Review of Books,
Wheeler, John Archibald, and Wojciech Hubert Zurek, eds. Quantum Theory and
book and are indexed only via "key discussions" pertaining to them, or via
subheadings.
287n. 4
as metaphor, 129
quantum objects
tics
Anaximander, 120
Physics, 55
Arts, xv, 25
Aspect, Alain, 89
Astronomy, xv
72-73
5, 48, 71-73
Baroque, 266n. 24
Beckett, Samuel, 25
260n. 60
Benjamin, Walter, 25
198,221,276n. 8,287n. 2
255n. 33,256n. 40
225
23
214-18
of, 121
Compton, Arthur, 31
history of science), 19
Continuum, 136
passim
143-44,160,170,220,230,264n.
ing, 160
Democritus, 72-73
222,225-28,233-34,245n.13,
217-18,287n.12
47
Diaphantus, 118
ferance
differential
18-20, 44
260-61n. 70
72-73
54
36, 286n. 46
130-31,142-43, 170-71,186-87,
mathematics, 130-31
mentary
89-90, 94
Epicurus, 73
cation, 138
surement
spaces, 125-26
259nn. 51, 56
308 * Index
plane, 124-27
280nn. 16, 20
283n. 32
Greenberger-Horn-Zeilinger experi-
ment, 98
288n. 21,289n. 25
281n.23
ics, 203-19
tainty relations)
Heraclitus, 13
Hume, David, 15
sciousness, 69
32
70, 130
imaginary numbers
mathematics), 118-19
244n. 10
Visualization
Irigaray, Luce, 22
Irreversible amplification, 63
Jordan, Pascual, 32
geometry, 268n. 26
249n. 9
unknown, 21
tion, 132-33
141, 154
265n. 21
287n. 4
268n. 25
207
43, 286n. 48
31-35
289n. 25
49
138-39
mechanics, 204-6
67; ordinary, 67
312 * Index
287n. 2
11
sion of
43, 45
numbers, 121
48-49, 54, 87
Index * 313
Nihilism, 10
44-45
theory, 14-15
ematics, non-Euclidean
ity
Noumena, 250n. 12
Number theory, 25
55
tors in
Optics, 267n. 24
257n. 43
Phenomenon (continued)
psychoanalysis, 155
Photon(s), 31
mental, 202
tion
20
266n. 3
186, 196
"Platonia," 251n. 16
play/game
Poetry, xv
250n. 13
249n. 4
passim, 272n. 50
Quantum computing, 88
Quantum cryptography, 88
Mills, 206
quantum
256n. 35
223,228-29,235,247n. 2
nological concept, 73
Quasi-mathematics, quasi-mathemati-
113-14
character of, 31
classical theories, 54
21
Romanticism, 266-67n. 24
quencies), 224
273n. 58
48, 261n. 71
science), 260n. 66
201-19
structuralism, 193
Space(s) (continued)
theory), 128
Spatialization, 123-27
53-55, 260n. 61
253n. 30
structure
Subjective, 23
Subjectivity, 285n. 44
Symmetry, 257
Theaetetus, 118
Thermodynamics, 1, 17, 44
Things-in-themselves, 43
168-69
ity), 136
mechanics), 78
Truth, 119
Index * 319
57
264n. 18
cal, 226
205
246n. 25
Woolf, Virginia, 26
ing
Zeno, 264n. 11
42