12

Marek Dobrzeniecki

M. Dobrzeniecki · The Conflicts of Modernity in Ludwig Wittgenstein’s Tractatus Logico-Philosophicus

The author offers a new look at one of the most influential books in the his-
tory of philosophy: Ludwig Wittgenstein’s Tractatus Logico-Philosophicus. He
presents the Tractatus as expressing the intellectual anxieties of its modernist
epoch. The most intriguing but usually unanswered question concerning the
Tractatus is why Wittgenstein had to think that only propositions of natural

The Conflicts
science have meaning. The author reviews the most popular interpretations of
the Tractatus and comes to the conclusion that the early Wittgenstein was an
ethical subjectivist. With this insight, he solves the tension between Tractarian

of Modernity in
theses that influenced neopositivism and its mystical part.

Ludwig Wittgenstein’s
Tractatus
Logico-Philosophicus

European Studies in T heolog y,

Philosophy and Histor y of Relig ions
Marek Dobrzeniecki is Lecturer in Philosophy at the Pontifical Faculty
of Theology in Warsaw and at the Centre for Thought of John Paul II in Edited by Bartosz Adamczewski
Warsaw.

ISBN 978-3-631-66780-4

EST 12_266780_Dobrzeniecki_AM_A5HCk PLE.indd 1 18.12.15 KW 51 16:21

 12
Marek Dobrzeniecki

M. Dobrzeniecki · The Conflicts of Modernity in Ludwig Wittgenstein’s Tractatus Logico-Philosophicus

The author offers a new look at one of the most influential books in the his-
tory of philosophy: Ludwig Wittgenstein’s Tractatus Logico-Philosophicus. He
presents the Tractatus as expressing the intellectual anxieties of its modernist
epoch. The most intriguing but usually unanswered question concerning the
Tractatus is why Wittgenstein had to think that only propositions of natural

The Conflicts
science have meaning. The author reviews the most popular interpretations of
the Tractatus and comes to the conclusion that the early Wittgenstein was an
ethical subjectivist. With this insight, he solves the tension between Tractarian

of Modernity in
theses that influenced neopositivism and its mystical part.

Ludwig Wittgenstein’s
Tractatus
Logico-Philosophicus

European Studies in T heolog y,

Philosophy and Histor y of Relig ions
Marek Dobrzeniecki is Lecturer in Philosophy at the Pontifical Faculty
of Theology in Warsaw and at the Centre for Thought of John Paul II in Edited by Bartosz Adamczewski
Warsaw.

EST 12_266780_Dobrzeniecki_AM_A5HCk PLE.indd 1 18.12.15 KW 51 16:21

The Conflicts of Modernity in Ludwig Wittgenstein’s
Tractatus Logico-Philosophicus
 EUROPEAN STUDIES IN THEOLOGY,
PHILOSOPHY AND HISTORY OF RELIGIONS
Edited by Bartosz Adamczewski

VOL. 12
 Marek Dobrzeniecki

The Conflicts of Modernity

in Ludwig Wittgenstein’s
Tractatus Logico-Philosophicus
Bibliographic Information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche
Nationalbibliografie; detailed bibliographic data is available in the internet
at http://dnb.d-nb.de.

Library of Congress Cataloging-in-Publication Data

Names: Dobrzeniecki, Marek, 1980- author.
Title: The conflicts of modernity in Ludwig Wittgenstein's Tractatus logico-
philosophicus / Marek Dobrzeniecki.
Description: 1 [edition]. | New York : Peter Lang, 2016. | Series: European
studies in theology, philosophy, and history of religions, ISSN 2192-1857;
Vol. 12
Identifiers: LCCN 2015050755 | ISBN 9783631667804
Subjects: LCSH: Wittgenstein, Ludwig, 1889-1951. Tractatus logico-
philosophicus.
Classification: LCC B3376.W563 T73253 2016 | DDC 192–dc23 LC record
available at http://lccn.loc.gov/2015050755

This publication was financially supported

by the Pontifical Faculty of Theology in Warsaw.

ISSN 2192-1857
ISBN 978-3-631-66780-4 (Print)
E-ISBN 978-3-653-06444-5 (E-Book)
DOI 10.3726/978-3-653-06444-5
© Peter Lang GmbH
Internationaler Verlag der Wissenschaften
Frankfurt am Main 2016
All rights reserved.
Peter Lang Edition is an Imprint of Peter Lang GmbH.
Peter Lang – Frankfurt am Main ∙ Bern ∙ Bruxelles ∙ New York ∙
Oxford ∙ Warszawa ∙ Wien
All parts of this publication are protected by copyright. Any
utilisation outside the strict limits of the copyright law, without
the permission of the publisher, is forbidden and liable to
prosecution. This applies in particular to reproductions,
translations, microfilming, and storage and processing in
electronic retrieval systems.
This publication has been peer reviewed.
www.peterlang.com
Table of Contents

Introduction................................................................................................................9

Chapter 1.  The Problem of Ontology in the Tractatus............................21

1.1 Anti-metaphysical interpretation of the Tractatus.......................................22
1.1.1 Anti-metaphysical interpretation and the tasks of
philosophy..............................................................................................22
1.1.2 The context principle in the Tractatus................................................26
1.1.3 In defence of semantic atomism..........................................................31
1.2 The Argument for Substance...........................................................................49
1.2.1 The argument from the false judgement or thinking
what is not..............................................................................................52
1.2.2 The argument from the determinateness of sense............................54
1.2.3 Zalabardo’s objections...........................................................................58
Summary......................................................................................................................67

Chapter 2.  The Simple Objects of the Tractatus.........................................71

2.1 Phenomenalistic interpretation of the simples.............................................73
2.1.1 The question of Russell’s influence on the Tractatus.........................74
2.1.2 Arguments in favour of the phenomenalistic
interpretation.........................................................................................77
2.1.3 Counterarguments................................................................................82
2.2 Materialistic interpretation of the simples.....................................................88
2.2.1 Simple objects as material points, point-masses or
physical atoms........................................................................................90
2.2.2 Arguments in favour of the materialistic interpretation....................95
2.2.3 Advantages of the materialistic interpretation................................. 107
2.2.4 Counterarguments............................................................................... 110
2.3 Resolute interpretation of the Tractatus...................................................... 113
2.3.1 The principles of the resolute interpretation................................... 114

5
 2.3.2 The notion of philosophy in the Tractatus...................................... 118
Summary................................................................................................................... 125

Chapter 3.  Wittgenstein’s Theory of Judgement..................................... 129

3.1 The context of Wittgenstein’s theory........................................................... 133
3.1.1 Russell’s theory of judgement............................................................ 134
3.1.2 The notion of the empirical self........................................................ 139
3.2 Wittgenstein’s criticism of Russell’s views on judgement.......................... 143
3.3 The Tractatus 5.54-5.5422............................................................................. 149
3.3.1 Conceptual clarifications................................................................... 149
3.3.2 The form of “A believes that p” according to Wittgenstein........... 151
3.3.3 Consequences of Wittgenstein’s theory of judgement................... 154
3.3.4 The repudiation of the existence of the complex soul
(TLP 5.5421)....................................................................................... 156
3.4 Other interpretations of TLP 5.54-5.5422.................................................. 159
3.4.1 Anscombe: TLP 5.54-5.5422 and the extensionality principle..... 160
3.4.2 Hacker: Hume’s influence on Wittgenstein’s theory...................... 162
3.4.3 Jacquette: the distinguishability problem........................................ 165
Summary................................................................................................................... 167
Chapter 4.  The Transcendental Self...................................................................... 171
4.1 The transcendental philosophy of Schopenhauer..................................... 172
4.1.1 Schopenhauer and the notion of the transcendental self.............. 172
4.1.2 Schopenhauer and the safeguarding of values................................ 178
4.2 The willing subject......................................................................................... 181
4.2.1 Examples of transcendental interpretations referring to
Wittgenstein’s ethics........................................................................... 182
4.2.2 Counterarguments............................................................................. 184
4.3 Solipsistic theses of the Tractatus................................................................. 192
4.3.1 The transcendental self as the owner of the
phenomenal world............................................................................. 193
4.3.2 The transcendental self as the linguistic soul.................................. 196

6
 4.3.3 Arguments in favour of Tractarian transcendental solipsism...... 200
4.4 Tractarian understanding of death.............................................................. 210
4.5 General remarks about the transcendental interpretation....................... 212
4.5.1 Is the nonsense of solipsism illuminating?...................................... 213
4.5.2 Transcendental reasoning................................................................. 217
Summary................................................................................................................... 219

Chapter 5.  Ethics in Wittgenstein’s Early Writings................................ 221

5.1 The nonsense of ethics in the eyes of early Wittgenstein.......................... 224
5.1.1 The absoluteness of moral values..................................................... 225
5.1.2 The sense of proposition and the idea of objectivity..................... 229
5.2 Early Wittgenstein’s subjectivism................................................................. 233
5.2.1 Ethical sentences as expressions of attitudes................................... 233
5.2.2 Ethical subjectivism and Tractarian silence.................................... 238
5.2.3 Wittgenstein’s subjectivism and early Russell’s emotivism........... 241
5.3 Solipsistic theses of the Tractatus under a subjectivist reading............... 246
Summary................................................................................................................... 251

Final Thoughts. The Defence of Human Values by

early Wittgenstein................................................................................................ 253

Bibliography........................................................................................................... 263

7
Introduction

One of the main questions in the following dissertation reads as follows: What
reasons did Wittgenstein have to think that only propositions of natural science
have meaning?
The totality of true propositions is the whole of natural science (or the whole corpus of
the natural sciences) (TLP 4.11)1.

One could expect such a statement from an admirer of natural science (of its pro-
gress, results, clarity or influence on everyday life). One could also expect this
statement to be the beginning of some philosophical programme in which the pro-
gress of all other branches of culture hinges on a scientific conception of the world.
Yet the Tractatus has nothing to do with these kinds of views. In a letter to his pub-
lisher, Ludwig von Ficker, Wittgenstein informed him that his work consisted of
two parts: the written part – the text that the reader has before his or her eyes, and
the unwritten part – topics about which Wittgenstein was intentionally silent2. To
Bertrand Russell, who believed that one should implement scientific methods into
the practice of philosophy3, he wrote: “How different our ideas are, for example,
of the value of a scientific work”4. He was explaining to the first English translator
of the Tractatus, Charles Kay Ogden, with respect to thesis TLP 6.5 (“The riddle
does not exist”), which could be interpreted straightforwardly as proof of Tractar-
ian positivism, that he did not wish “anything ridiculous or profane or frivolous
in the word when used in the connection ‘riddle of life’ etc.”5. Wittgenstein was

1 All English quotations from the Tractatus Logico-Philosophicus (later: the Tractatus)
are from the Pears and McGuinness translation (Revised edition from 1974).
2 “My work consists of two parts, the one presented here plus all that I have not written.
And it is precisely this second part that is the important point. For the ethical gets
its limit drawn from the inside, as it were, and I am convinced that this is the ONLY
rigorous of drawing those limits; (…) I have managed in my book to put everything
firmly into place by being silent about it” (cited in: ProtoTractatus, p. 16). Engelmann,
on the other hand, confirms that it was the ethical part which was of greater importance
to Wittgenstein than the logical theory: “It could be said with greater justice that
Wittgenstein drew certain logical conclusions from his fundamental mystical attitude
to life and the world” (Engelmann 1967, p. 97).
3 “It is not results, but methods, that can be transferred with profit from the sphere of
the special sciences to the sphere of philosophy” (Russell 1914b, p. 57).
4 Letters, p. 53.
5 Wittgenstein 1973, p. 36.

9
afraid that his views would be understood as merely negating the meaningfulness
of philosophy. According to Drury’s testimony, he once said: “Don’t think I despise
metaphysics or ridicule it. On the contrary, I regard the great metaphysical writ-
ings of the past as among the noblest productions of human mind”6 and, according
to Carnap’s recollection, the result of the Tractatus, i.e. the thesis that metaphysical
and ethical utterances are senseless, was “extremely painful for him emotionally, as
if he were compelled to admit a weakness in a beloved person”7. Therefore, there
is clearly tension in the Tractatus between theses that could as well have been ex-
pressed by the proponents of neo-positivism or scientism8 and its mystical part,
where Wittgenstein writes, among others, that “there are, indeed, things that can-
not be put into words. They make themselves manifest. They are what is mystical”
(TLP 6.522)9.
One of the explanations for such tension comes from Wittgenstein’s biography.
If it was not for his war experiences, the Tractatus would contain only considera-
tions on logic and language. As Ray Monk writes:
The Austrian Eleventh Army, to which Wittgenstein’s regiment was attached, faced [in the
June of 1916] the brunt of the attack and suffered enormous casualties. It was at precisely
this time that the nature of Wittgenstein’s work changed (Monk 1991, p. 140).

It was in the same month, on 11 June 1916, when Wittgenstein noted in his Note-
books the famous entry which begins with the question: “What do I know about
God and the purpose of life?” (NB 11.6.16. p. 72). The mystical-ethical part of the
Tractatus (TLP 6.4–7) is strictly connected with the religious conversion Witt-
genstein experienced during World War I. Distressed and depressed by the evil

6 Drury 1960/1967, p. 68.

7 Carnap 1964/1967, p. 36.
8 Apart from TLP 4.11, one can mention in this context thesis TLP 6.53: “The correct
method in philosophy would really be the following: to say nothing except what
can be said, i.e. propositions of natural science—i.e. something that has nothing to
do with philosophy—and then, whenever someone else wanted to say something
metaphysical, to demonstrate to him that he had failed to give a meaning to certain
signs in his propositions. Although it would not be satisfying to the other person—he
would not have the feeling that we were teaching him philosophy—this method would
be the only strictly correct one”.
9 “Anyone who has read Ludwig Wittgenstein’s ‘Tractatus’ is struck, indeed is usually
fascinated, by two apparently contradictory aspects of it. On the one hand, it seems
to confine all sensible talk to the propositions of natural science, and even puts the
propositions of logic into the category of senseless; on the other hand it itself embraces
extremely non-scientific positions” (McGuinness 2002, p. 55).

10
and malicious company of his fellow soldiers and faced with the danger of losing
his life, he started seriously considering problems which up to that point he had
thought to be “philosophical” in the worst possible meaning of the word. The
reference here to Wittgenstein’s biography, however, does not provide a sugges-
tion as to what the correct interpretation is of the solipsistic (TLP 5.6-5.641) or
the ethical theses of the Tractatus.
In order to address the aforementioned issues one has to put the Tractatus in
a broader context of the history of philosophy at the turn of the 19th and 20th cen-
turies. Among the philosophers of that time there prevailed convictions which
I will call in this dissertation ‘modernist’. The most important feature of modernist
thinking is granting science the primary role in the task of describing reality10.
Impressed by new physical achievements, such as Maxwell’s theory of electro-
magnetism (1873), philosophers and scientists began to believe in the possibility
of a unitary physical description of the world – one of such projects was taken up
by Heinrich Hertz in his Principles of Mechanics (1894). Moreover, Darwin’s theory
of evolution (1859) showed that we can explain events of the biological world ex-
clusively in terms of causes and effects without mentioning the notion of an aim,
which was always tinged with theological associations. Darwin’s theory explained
the rise of the human species in a fully naturalistic way, therefore the explanation
for the existence of the human being (with all of its magnificent mental abilities)
was done without reference to the special act of God’s creation11. In effect, some
philosophers believed that the progress of science would make philosophical and
theological doctrines, such as the philosophical doctrine of the immortal soul or
the theological doctrine of the creation of the world, useless. Summing up, the first
feature of the modernist way of thinking is ascribing to science (and especially
natural science) the primary role in searching for the truth about the world.
On the other hand, it was exactly this “rise of science” that resulted in anxiety
that the scientific worldview might flatten the complexity and sophistication of our
perception of the world. Many philosophers, who came from different traditions,

10 “That we are ‘scientific’ in our attitudes and live in a scientific age is wildly held to be
both a fact and a ground for rejoicing, an achievement to be celebrated and carried
further” (Midgley 1992, p. 3).
11 “Darwin and other biologists, particularly Thomas Henry Huxley, seemed to have
established that the existence of human beings had a naturalistic explanation and
required no special creation. Among physiologists the dominant opinion, especially
stemming from Germany, was equally uncompromising. The processes of life were
at root ‘mechanical’ and required no special life force to explain them” (Harré 2008,
p. 23–24).

11
began to wonder: “What is the place for religion and ethics in a worldview domi-
nated by scientific thinking?” and “How can such notions as the notion of a free
will, the notion of the self or the notion of a moral value be accommodated in the
scientific worldview?”12. I see, after Charles Taylor, posing these kinds of questions
as the second feature of modernism13. My main interpretational hypothesis of the
Tractatus assumes that in this book both of these characteristics of modernism are
present14. In this sense the Tractatus shared the intellectual interests and anxieties
of its epoch. If I am right then the main problem of the Tractatus reads as follows:
“How can one safeguard the world of human values from the claims of science?”.
I shall call up this question later in my work on the fundamental problem of the
Tractatus. The aforementioned tension between some of its formulations will then
find an explanation in the fact that the Tractatus is an example of the modernist
way of thinking. On the one hand, it acknowledges the progress and success of
natural science at the beginning of the 20th century and, on the other, it tries to find

12 In this context one can invoke the example of such different thinkers as the idealist
Josiah Royce (“Like other Anglo-American idealists, Royce attempted to find a place
for religion in a world of scientific facts” (Allard 2008, p. 57)), and the pragmatist
Charles Sanders Peirce (“Peirce saw positivists as committed to a flawed conception of
reality which led inevitably to scepticism; and he shared James’s hope for an empirically
grounded philosophy which would find room for values and religious belief ” (Hookway
2008, p. 77)).
13 “All this can help explain the particular form of the modernist turn to interiority.
Thinkers in the early twentieth century were exercised by a problem which is still posed
today: What is the place of Good, or the True, or the Beautiful, in a world entirely
determined mechanistically?” (Taylor 1989, p. 459).
14 One can find in the literature associations of early writings of Wittgenstein with
modernism, however, in a different meaning of this term. For instance, Janik writes:
“Wittgenstein’s effort to get straight about the limits of thought and language in all of
the stages of his development and thus to be fair to science, religion, and art account for
his place of honour among critical modernists” (Janik 2001, p. X), but he mentions as
critical modernists such thinkers as Kraus, Loos, Trakl or Weininger, who fought with
the Wiener Moderne – the cultural movement which “attached itself to an irrationalist
cult of subjective experience that sought thrill in everything ‘new’, especially in what was
obscure and ambiguous. Thus it was in most respects closer to our post-modernism
than any classical form of modernism except symbolism” (Janik 2001, p. 208). Michael
Fischer (Fischer 1993) also labels Wittgenstein as a modernist philosopher, but his aim
is to point out an analogy between his philosophy and modern art. Just as listeners of
modern music wonder if it is music at all, the readers of Wittgenstein either revere his
work or entirely reject it. According to Fischer, the uncertain reception of Wittgenstein’s
writings brings to mind T. S. Eliot’s anxiety about the inheritability of culture.

12
a way to express the problem of the meaning of life and of moral values in a world
governed by the laws of science.
The fundamental problem of the Tractatus in its present formulation needs
clarification. First, what do I mean by the world of human values? Roughly speak-
ing, it is the world described from the anthropocentric perspective15. Human
values, in this sense, are those objects which occur only in the anthropocentric
description of the world. Bertrand Russell, in On Scientific Method in Philosophy,
indicated that, for instance, if one describes the development of species from
protozoa through primates to human beings as progress, then one takes exactly
the anthropocentric perspective16. This is because from a strictly objective stand-
point there is no such thing as progress in the development of species, i.e. there
are no better and worse species. This means that the concept of progress refers
to a human value, and a description of the world which contains this notion is a
description from the anthropocentric point of view. In the narrow sense, human
values are those which address a group of issues which Wittgenstein named “the
problems of life” in the Tractatus17. In this meaning one can include moral and
aesthetic values to the human values, as well as values that make life worth living.
The next issue to clarify is why these values need to be defended from the claims
of science? By answering this question I shall point to the fact that the natural sci-
ences, as an effort to describe the world objectively, sub specie aeterni, contain no
concepts which are necessarily connected with the anthropocentric perspective.
This fact, combined with the acceptance of the authority of science as the only
source of truth about the world, posed a problem for some thinkers. The mechani-
cal worldview which emerges from the convictions that:
• the scientific description of the world contains ultimately only concepts of
material particles, and
• the scientific description of the world is complete; there is no aspect of reality
which cannot be captured by science

15 I am adopting here Bernard Williams’ point of view (Williams 2002), according to

which the relation between “the human” and “the anthropocentric” perspective is such
that “the human” perspective always assumes “the anthropocentric” one.
16 “Organic life, we are told, has developed gradually from the protozoon to the philosopher,
and this development, we are assured, is indubitably an advance. Unfortunately it is
the philosopher, not the protozoon, who gives us this assurance, and we can have no
security that the impartial outsider would agree with the philosopher’s self-complacent
assumption” (Russell 1914b, p. 62).
17 “We feel that even when all possible scientific questions have been answered, the
problems of life remain completely untouched” (TLP 6.52).

13
seemed to be too depressing. For instance, in the book which contributed the most
to Wittgenstein’s religious conversion18, The Gospel in Brief, Leo Tolstoy wrote:
When fifty years old, having questioned myself, and having questioned the reputed phi-
losophers whom I knew, as to what I am, and as to the purport of my life, and after get-
ting the reply that I was a fortuitous concatenation of atoms, and that my life was void of
purport, and that life itself is evil, I became desperate, and wished to put an end to my life
(Tolstoy 1896, p. 8).

In ethics we describe human beings as persons who possess dignity and rights.
Physics, on the other hand, describes human beings as a “concatenation of atoms”.
This view on the human being as a complex of its material elements, “unimportant
agents in an aimless and senseless universe that is ruled by blind natural forces”19,
amounts to a reification of persons, treating them like objects among other ob-
jects, which could lead, in the eyes of some of the commentators of Wittgenstein,
to catastrophic social and political consequences20. In this sense, human values
need to be defended from the claims of science.
In the second meaning, one needs to safeguard human values because of pos-
itivism’s mistake consisting in the conviction that both science and religion or
metaphysics have the same goal, which is to describe the world, with the differ-
ence being that science does so more accurately. Therefore, in order to acquire
the proper worldview, one has to overcome these temporary, i.e. religious and
metaphysical, stages of the intellectual development of humankind. Those who
want to adopt the scientific worldview must reject naive ethical and religious
convictions which are necessarily connected with false and unjustified anthro-
pocentrism. This was the claim of some philosophers, for instance, Russell and

18 As a biographer of Wittgenstein, Ray Monk notes: “The book captivated him (…) He
became known to his comrades as ‘the man with the gospels’. For a time he – who
before the war had struck Russell as being ‘more terrible with Christianity’ than Russell
himself – became not only a believer, but an evangelist, recommending Tolstoy’s Gospel
to anyone in distress” (Monk 1991, p. 115–116).
19 Sukopp 2007, p. 91.
20 “When we locate human subjectivity in the world there is always the danger that we end
up thinking of human subjects as things in the world just like other things, as objects
among objects. That way of thinking can have devastating moral consequences. It is
also likely to obscure the very nature of human subjectivity” (Sluga 1983, p. 136–137).

14
Schlick21, which provoked Wittgenstein to defend the right to hold ethical and
religious convictions22.
Apart from distorting the meaning of ethical and religious notions and the
claim that ethics, metaphysics or religion represent bad science, or sad testimonies
of periods of history when humankind was plunged into the darkness of igno-
rance, one can also discern the third reason for defending the world of human val-
ues from the consequences of the rise of science. One can express this reason, after
Wittgenstein, as the danger that a scientific description of the world impoverishes
our life and culture23. The basic reason for this impoverishment lies in the fact that
the problem of the meaning of life is not a scientific one:
Our conception on the contrary is that there is no great essential problem in the scien-
tific sense (CV, p. 20).

Apparently, for Wittgenstein, if our culture and education are dominated by scien-
tific thinking, then we will cease to ask questions about the meaning of life or about
moral goodness. This means we will cease to wonder about the most important
things in our lives24. It was in this sense that moral values needed to be protected
from the claims of science – the danger consisted, in the eyes of Wittgenstein, that
enchanted by the success of scientific explanations we will forget to pose questions
about the meaning of life.
The task, therefore, was to safeguard the notions of ethics and religion from
scientific distortion to defend the right of a rational person to hold ethical and
religious convictions and to justify the importance of posing the problems of
life. In the conclusions to this dissertation I hope to show in which respect Witt-
genstein’s solutions either fulfil or do not fulfil this task. What I find, however,

21 “The ethical element which has been prominent in many of the most famous systems
of philosophy is, in my opinion, one of the most serious obstacles to the victory of
scientific method in the investigation of philosophical questions” (Russell 1914b, p. 63).
22 Carnap reports the difference between Wittgenstein’s and Schlick’s position in the
following way: “Once when Wittgenstein talked about religion, the contrast between
his and Schlick’s position became strikingly apparent. Both agreed of course in the view
that the doctrines of religion in their various forms had no theoretical content. But
Wittgenstein rejected Schlick’s view that religion belonged to the childhood phase of
humanity and would slowly disappear in the course of cultural development” (Carnap
1964/1967, p. 35).
23 “Science: enrichment & impoverishment” (CV, p. 69).
24 “Our children learn in school already that water consists of the gases hydrogen &
oxygen, or sugar of carbon, hydrogen & oxygen. Anyone who does not understand is
stupid. The most important questions are concealed” (CV, p. 81).

15
the most intriguing in the Tractatus, and the reason I wrote this dissertation, is
that by answering the question: “What is the place of ethics and religion in the
scientific worldview?” Wittgenstein did not abandon his conviction, expressed
in TLP 4.11, that natural science is the only source of truth about the world. He
admitted the rightness of the most radical versions of scientism and naturalism,
and it was from this point of view that he tried to see how one can talk about
moral or aesthetic values. His point of interest was the question whether one,
without giving up his or her rationalism, can still search for the meaning of life
or whether one, without undermining scientific claims about the world, can
still make sense of his or her own religious experiences? By answering this fun-
damental problem he did not choose the easy way out, consisting in belittling
the value and possibilities of natural science. One could say that in fulfilling
the task of safeguarding human values he agreed with his possible positivistic
opponent regarding all of that opponent’s views on science and metaphysics.
In my opinion, this is what makes his effort so fascinating and worth analysing
also today. Wittgenstein reveals himself in the Tractatus as a doubly serious
philosopher – he acknowledges the progress and explanatory powers of science
and, at the same time, concedes the importance of safeguarding the values of
the human world.
This conviction about what is especially interesting in the philosophy of early
Wittgenstein determines the structure of this work. In its first part I concentrate
on proving that Wittgenstein, although he experienced some kind of spiritual
illumination during the World War I, did not withdraw from his scientism. In
the first chapter I shall discuss the basic, i.e. from the point of view of my work,
problem whether the Tractarian theory of meaning is an example of the real-
istic theories of meaning. I shall argue in favour of this thesis and against the
claims of the anti-metaphysical interpretation of the Tractatus. I find this prob-
lem crucial because if proponents of the anti-metaphysical interpretation are
right, and early Wittgenstein was indeed not interested in ontological topics,
and if the notion of a simple object was a purely formal one (that is, if Wittgen-
stein, when writing the Tractatus, had no concrete candidacy for the referent of
this concept in mind), then, obviously, there is no fundamental problem of the
Tractatus as I formulate it here. Then Wittgenstein could not pose the question
about the place of ethics in the scientific worldview simply because he did not
raise the problem of a proper worldview at all. The second chapter analyses the
two main candidacies for the referents of the Tractarian concept of a simple
object: simple units of experience (as the phenomenalistic interpretation of the
Tractatus argues) and the most elementary particles of matter (as the materi-
alistic interpretation proposes). I shall defend in that chapter the materialistic

16
interpretation of the Tractatus, i.e. I shall show that Wittgenstein’s position in
his early oeuvre could be classified as radical naturalism – a view according to
which the world consists only in physical particles and their movements.
According to Putnam’s description of scientism, this position not only assumes
“that science, and only science, describes the world as it is in itself ” independently
of an anthropocentric perspective but it also claims “that science leaves no room
for the independent philosophical enterprise”25. In my opinion, one can notice
in the Tractatus also this second aspect of scientism. Wittgenstein claims, among
others, that “philosophy is not one of the natural sciences. (The word ‘philoso-
phy’ must mean something whose place is above or below the natural sciences,
not beside them.)” (TLP 4.111). He also suspects that traditional metaphysical
concepts are empty26. Wittgenstein’s scepticism towards philosophy is directed at
metaphysics as a doctrine about the world27, i.e. a doctrine which attempts to de-
scribe the world more comprehensively than science or which assumes that there
is an aspect of reality which is a special subject-matter for philosophy. Wittgen-
stein fought with this conception of philosophy also in less obvious fragments of
his book. In this context I shall present in Chapter 3 Wittgenstein’s discussion of
the Russellian theory of judgement (TLP 5.54-5.5422). Obviously, the main topic
of this fragment of the Tractatus is to reconcile the existence of propositional atti-
tudes with a strong extensionality thesis28, but I shall argue that Wittgenstein’s so-
lution to this problem: “It is clear, however, that ‘A believes that p’, ‘A has thought
p’, and ‘A says p’ are of the form ‘ “p” says p’ ” (TLP 5.542) as its background has
Wittgenstein’s conviction that Russell did not confer any meaning in his theory
of judgement on the notion of the mind. I will also analyse this topic because it
strengthens the view that the author of the Tractatus was a naturalist. In my inter-
pretation, Wittgenstein’s theory of judgement does without the dualistic notion of
the self. In contrast to Russell’s theory it is able to explain the fact that propositions

25 Putnam 1992, p. X.

26 “The correct method in philosophy would really be the following: to say nothing except
what can be said, i.e. propositions of natural science—i.e. something that has nothing
to do with philosophy—and then, whenever someone else wanted to say something
metaphysical, to demonstrate to him that he had failed to give a meaning to certain
signs in his propositions” (TLP 6.53).
27 “Philosophy is not a body of doctrine but an activity” (TLP 4.112).
28 The view according to which every meaningful sentence is a truth-function on an
elementary sentence, including an extreme example of an elementary sentence which
is a truth-function on itself.

17
communicate content without referring to the realm of the mental, which is alleg-
edly distinct from the realm of the material.
In Chapter 4 I shall discuss the transcendental interpretations of the Tracta-
tus. The proponents of this kind of reading of Wittgenstein’s early work agree
that its fundamental problem consisted in finding a place for ethics and religion
in the world of scientific facts, but, in contrast to what I believe, they believe that
Wittgenstein, when addressing this issue under the influence of the philosophy
of Schopenhauer, accepted an idealistic or even a solipsistic point of view at the
expense of his naturalism. After a close examination of the arguments advanced
in favour of this interpretation and the fragments of the Tractatus which suppos-
edly speak in favour of this reading, I shall hope to prove that early Wittgenstein
was consequent in his naturalism and did not adopt transcendentalism. If he
wanted to “defend” the world of human values against the claims of science, he
had to do so in another way.
I shall try to reconstruct what his strategy and his answer to the fundamental
problem of the Tractatus were in the last chapter of my dissertation. The main
difficulty of this effort consists in the fact that an answer to the fundamental
problem belongs (invoking Wittgenstein’s letter to von Ficker) to the “unwrit-
ten” part of the Tractatus – the book which its author ends with a call to silence.
However, on the basis of what we know about the position of the Tractatus (for
instance, that one can classify its ontological position as materialism) and Witt-
genstein’s later remarks (especially on the basis of A Lecture on Ethics, which, as
I suspect, differs from the Tractatus with respect to views on ethics only in that
in his lecture Wittgenstein was less consequential and gave in to the temptation
of expressing necessarily nonsensical ethical convictions), I shall formulate the
hypothesis of Wittgenstein’s subjectivism with respect to ethics. This means that I
shall defend the view according to which Wittgenstein held ethical expressions to
be nonsensical and aiming to express a speaker’s attitude to the world as a whole.
In general, I have read the Tractatus as a modernist oeuvre29. I have read it
as a sign of the times when philosophers, having acknowledged the importance

29 This is exactly the opposite view to the one represented by Ronald Hustwit, who reads
the Tractatus as an anti-modernist book. As confirmation of his views he indicates theses
TLP 6.37-6.6.372, in which Wittgenstein compares the modern conception of physical
laws to the ancient myths of Fate or God. Both of these myths provide an illusion of the
ultimate explanation: “The modern view uncritically suppose that it holds no starting
points beyond the laws of nature, while holding the unacknowledged presupposition of
realism that the laws of nature are the description of reality” (Hustwit 2011, p. 568). In
contrast to Hustwit, I interpret Wittgenstein’s critique of the laws of nature as a sign of his

18
of the results of scientific research, began to question the status of traditional
philosophical doctrines. In my opinion, Wittgenstein, in his book, skilfully ma-
noeuvred between Scylla of neo-positivism and Charybdis of transcendental
idealism. It is a book worth reading, among others, for the consequence in draw-
ing morals from naturalistic positions it holds. Up until today it shows us what
the possible and most substantial position is of someone who, on the one hand,
accepts the ontological authority of natural science but, on the other, does not
want to see religion or metaphysics merely as past stages in the intellectual his-
tory of humankind.

strong naturalism. According to this view, in order to describe the world one needs only
the notion of spatial-temporal points and their intrinsic properties, and one does without
the laws that govern the causal interactions between these points (for the difference
between the strong and weak version of naturalism, see: Papineau 2008, p. 132–134).

19
Chapter 1. The Problem of Ontology
in the Tractatus

In the following chapters (Ch. 1–2) I shall analyse the hypothesis that TLP 4.11
(“The totality of true propositions is the whole of natural science (or the whole cor-
pus of the natural science)”) is a consequence of Wittgenstein’s materialistic views
on the world30. Briefly speaking, I intend to defend two hypotheses: in Chapter 1
the realist claim31 that in Wittgenstein’s opinion the division of reality into simple
objects is an intrinsic feature of reality; and in Chapter 2 the materialistic claim
that the simplicity of objects means their physical indivisibility. In my opinion,
the problem of thesis TLP 4.11 can be solved by answering the question about
the worldview entailed by the theses of the Tractatus. The question reads as fol-
lows: “What, in one’s view, does the world have to look like in order for one to be
convinced that the only truths about it are to be found in statements of natural
science?” This means that the first problem I have to address is the problem of the
ontology of the Tractatus.
But is it possible to draw ontological morals from the Tractarian theses at all?
There is a vast group of commentators who, for different reasons, claim that there
is no metaphysics in Wittgenstein’s early oeuvre32. Some say that the Tractatus puts
forward merely views on logic, language, the conditions it has to fulfil in order to
be meaningful, and that early Wittgenstein was simply not interested in meta-
physics. According to such a view the claim that simple objects exist is merely the
consequence of a linguistic requirement that any language can be analysed in the
simple elements. Others claim that the Tractatus belongs to the Kantian tradition
of philosophy, therefore, because we are imprisoned in our representations, one
cannot say anything about reality in itself33. At best one can say something about
the way we represent the world, for instance, that in order to describe the world we
need the notion of an object. It does not determine, however, if there are objects in

30 I shall use term “materialistic” as synonymous with “physicalistic”.

31 It is realism both in the epistemological sense allowing us to be able to know the intrinsic
features of reality as well as realism with respect to a theory of meaning claiming that
the referents of names are objects “ready” to be named, i.e. bits of reality which are not
formed by language or any other form of representation. In my interpretation I shall
focus on realism with respect to the theory of meaning.
32 McGuinness 2002, Atkinson 2009.
33 Meyer 1986; Bergmann 1973.

21
the world as it is, i.e. independently of any systems of representation. Referring to
the interpretations which favour the aforementioned claims, I shall use the terms:
“anti-metaphysical” or “formalistic” interpretation of the Tractatus.

1.1  Anti-metaphysical interpretation of the Tractatus

Accordingly, in the first part of the first chapter (1.1) I shall present the anti-met-
aphysical interpretation of the Tractatus by focussing on two pivotal points: first,
on the question whether Wittgenstein had any views on what there is, or, in other
words, if there is any ontology in the Tractatus (section 1.1.1)? The second point
refers to the anti-realist theory of meaning preferred by the proponents of the
anti-metaphysical interpretation of the Tractatus (section 1.1.2). The damaging
effect of a possible negative answer to the first point is clear: if Wittgenstein had no
views on the nature of objects, then he could not have been a proponent of materi-
alism. The connection between an anti-realist theory of meaning and my interpre-
tation of the Tractatus is less clear, and I will try to explain it later in section 1.1.2.
In section 1.1.3 I shall argue against the anti-metaphysical interpretation by:
(1) opposing the widespread opinion regarding Wittgenstein’s alleged lack of
interest in metaphysics
(2) defending the thesis that, according to the Tractatus, names have a reference
also outside the context of a proposition (the thesis which from now on I shall
call the ‘thesis of semantic atomism’).

1.1.1  Anti-metaphysical interpretation and the tasks of philosophy

I shall begin my presentation of the anti-metaphysical interpretation of the
Tractatus by considering its main supposition: early Wittgenstein was not in-
terested in metaphysics, hence considerations about the nature of the referents
of names would be entirely alien to him. In the Preface to the Tractatus, Witt-
genstein stated that we pose philosophical questions because we do not un-
derstand the functioning of logic and language34. Philosophers pose questions
such as, for instance: “Is there a world outside our ideas?”, which to ordinary
people seem a bit odd. They discuss whether we are trapped in the world of our
ideas or if we have evidence for an external reality that is independent in its

34 “The book deals with the problems of philosophy, and shows, I believe, that the reason
why these problems are posed is that the logic of our language is misunderstood”
(TLP, p. 3).

22
existent. The realists accuse the idealists of holding nonsensical positions, but
they tend to forget that if idealism is nonsensical and conveys no content, then
realism as contradicting nonsense is also nonsensical (how could one mean-
ingfully contradict something which is not even a thought?). Exactly this kind
of discussions are, according to McGuinness’ understanding of Wittgenstein,
an example of the misunderstanding of the functioning of language: “There is
already contained in language and thought the possibility of all objects that
are possible. All logical forms are logically possible within language, within
thought. No separate investigation of ‘reality’ is conceivable”35.
What is the point of doing philosophy if metaphysical questions are the result
of misunderstanding of the functioning of language? According to the defenders
of the anti-metaphysical interpretation, Wittgenstein’s aim was to analyse lan-
guage and reveal the logical form of a proposition which is hidden under the
surface of everyday language. “All philosophy is a ‘critique of language’ ” – says
Wittgenstein in the Tractatus36, and that statement, in the eyes of the proponents
of the anti-metaphysical interpretation, means that a philosopher does not need
to care about the nature of the world. Wittgenstein also claims that “logic must
look after itself ” (TLP 5.473)37. One of the consequences of this standpoint, for
the proponents of the formalistic reading, is that the result of the philosophical
inquiry, i.e. the revealed form of a proposition, is not determined by reality and
one cannot infer what kinds of objects exist from the analysis of language.
Therefore, if a reader of the Tractatus encounters in the book the notion of
an object, then it has to be a purely formal notion which should not awake in
the reader any metaphysical associations. McGinn, for instance, emphasises
that Wittgenstein asked if simple objects exist only in the context of logic: “Can
we manage without simple objects in LOGIC?” (NB 9.5.15, p. 46)38. This notion
is essential to the symbolism of logic39, just as the subject-predicate structure
of propositions belongs to symbolism. But just as the question whether there is

35 McGuinness 2002, p. 91.

36 TLP 4.0031.
37 The proponents of the formalistic reading indicate that this sentence is a frequent motif
in Wittgenstein’s early writings: NB 22.8.14, p. 2; 2.9.14, p. 2; 3.9.14, p. 2; 13.10.14, p. 11;
26.4.15, p. 43.
38 McGinn 2007, p. 210. “The simple objects (…) are not so much a kind of metaphysical
entity conjured up to support a logical theory as something whose existence adds no
extra content to the logical theory” (Ishiguro 1969, p. 40).
39 “What I want to argue is that the idea that the question about simples is to be understood
as a question about the essential nature of a symbolism” (McGinn 2007, p. 209).

23
something in the world which corresponds to the subject-predicate structure
of a proposition would have been a fatal mistake, the question as to what the
ultimate elements of reality are would also be a mistake. The way we represent
reality includes the fact that we describe it in terms of objects. We cannot de-
scribe it in some other way, and in this sense one can say that the concept of
the object is internal to the system of representation. For instance, according
to McGinn, if I truly say that the pen on my desk is red, then I do not say any-
thing about things in the world, particularly, I do not state the existence of such
things in itself, such as desks and pens. I simply correctly describe in a particu-
lar system of representation (by means of such logical devices as names) a state
of affairs40. McGinn concludes:
In all these cases, we are making the mistake of treating what is internal to a symbolism
in which we express propositions that can be compared with reality for truth or falsity,
as if it were a question of fact (McGinn 2007, p. 209).

In logic, the concept of the object designates everything which can go proxy for
a variable in a propositional function. When the proponents of the anti-meta-
physical interpretation claim that the notion of the object is a purely logical no-
tion41, they mean by that that according to Wittgenstein, an object is anything
that can go proxy for x in f(x) (or, in other words, an object is anything that can
be referred to by a third-person singular pronoun). If Wittgenstein says that re-
ality consists of objects, he simply means that the world consists of something42.
The question: “What is this something?” is not of the philosopher’s concern.
Ishiguro writes about the concept of the object that: “The ‘objects’ of the Trac-
tatus are not particular entities in any normal sense, but entities invoked to fit
into a semantic theory”43, and McGuinness adds in a similar mode: “An object
in the Tractatus which is the reference of a name or simple sign can be viewed
as simply truth-value potential of a certain expression”44.
The realist theory of meaning assumes that there are some ultimate and funda-
mental constituents of the world. It claims that there are links between bits of lan-
guage (names) and bits of reality. One can know the reference of a name without

40 “To say that the object for which the word ‘red’ stands is a constituent of an existing
state of affairs is just to say that the state of affairs is correctly described by means of a
proposition in which the word ‘red’ occurs” (ibid., p. 214).
41 Ishiguro 1969, p. 27.
42 Soin 2001, p. 35.
43 Ishiguro 1969, p. 21.
44 McGuinness 2002, p. 87.

24
knowing the reference of other names. This is so because it is reality which deter-
mines what the referents of names are: “It is not up to us what counts as a simple
part of reality; it is an intrinsic feature of reality itself ”45. On the other hand, ac-
cording to the anti-metaphysical reading of the Tractatus, the notion of the object
is a formal one – it is internal to symbolism. To claim that the thesis: “There are
simple objects” says something about the constituents of reality is the same as to
claim that the statement: “Today it is raining or it is not raining” helps me to make
the decision whether I should take an umbrella when I go out. For the commen-
tators who are more in favour of the anti-metaphysical reading of the Tractatus,
the independent order of reality is inaccessible to us. Consequently, according to
McGuinness, such a notion as “the realm of reference” is a myth and “the sort of
metaphysics [Wittgenstein] condemns”46. Ishiguro writes in a similar tone: “To ask
what kind of familiar entities correspond to the objects of the Tractatus seems to
lead us nowhere”47, and McGinn: “A transcendental realm with an intrinsic struc-
ture is an illusion”48. The proponents of this reading often refer to the entry from
the Notebooks: “It keeps on looking as if the question ‘Are there simple things?’
made sense. And surely this question must be nonsense” (NB 5.5.15, p. 45). It is
exactly because of this stance on Wittgenstein’s views on the role and aim of phi-
losophy that this interpretation is called an anti-metaphysical one49. Peter Winch
presented the most far-reaching consequences of this position:
What has become of the idea that to say something is to stand in relation to some indepen-
dently reality? We seem to be in a position not essentially different from that of Protagoras
or Gorgias. There is no reality; and if there were, no man could know it; and even if a man
could come to know it, he could not communicate what he knew to anyone else (Winch
1969a, p. 11)

As I have tried to show in the last paragraph, an anti-metaphysical interpreta-

tion raises the issue of Wittgenstein’s approach to philosophy; it puts emphasis
on Wittgenstein’s conviction that philosophy is a critique of language. If he truly
reduced the problems of philosophy to problems of language, then indeed the
discussion whether simple objects are sense-data or physical particles would be
pointless. Apart from discussing the general role of philosophy in the eyes of
early Wittgenstein, the anti-metaphysical interpretation brings up the question

45 Child 2011, p. 55.

46 McGuinness 2002, p. 84.
47 Ishiguro 1969, p. 47.
48 McGinn 2007, p. 216.
49 Ibid., p. 201.

25
whether, according to him, a name has a reference also outside the context of a
proposition. The representatives of this interpretation search for the answer to
this question in thesis TLP 3.3. In the next section I shall present its solution
and how it affects the conviction of the existence of objects that are independ-
ent from us.

1.1.2  The context principle in the Tractatus

I have reduced for the moment the problem whether there are any ontological
consequences of the Tractarian system to the pivotal question if Wittgenstein
in the Tractatus was a proponent of semantic atomism. I do so only because
when a name also has a reference outside the context of a proposition does it
make sense to ask: “What is the object to which I refer, independently of the
particular system of propositions or of a particular conceptualization?”. If the
answer to the pivotal problem was negative, then one could ask and obtain an-
swers about the referents of names only restricted to a given proposition or to
a system of propositions. For instance, the answer could read as follows: “The
world consists of mass-particles under the system of propositions of Hertzian
mechanics, but the world consists of human beings, animals, plants, chairs,
football games, justice, and so on under the system of propositions of ordinary
language. What are the constituents of the world in itself, independently of any
system of propositions, we simply cannot know”. If this was true with respect
to the Tractatus, then my aim of proving that the worldview of young Wittgen-
stein assumes ontological materialism would be fruitless right from the start.
That is why at the beginning of my disquisition about the objects of the Tracta-
tus I shall defend the view that the Tractatus involves semantic atomism.
The thesis of semantic atomism reads as follows:
(A)  A name has a reference outside a proposition.
In other words, a name refers to an object independently of the reference of other
names. It seems to me that the thesis of semantic atomism is well represented and
defended in the Tractatus50. In TLP 3.203 we read a statement that is the exact
formulation of this position:

50 This thesis is also well represented and defended in the Notebooks: “It is quite clear that
I can in fact correlate a name with the watch just as it lies here ticking in front of me,
and that this name will have reference outside any proposition in the very sense I have
always given that word” (NB 15.6.15, p. 60). “We feel that the WORLD must consist of
elements (…) The world has a fixed structure” (NB 17.6.15, p. 15).

26
 A name means an object. The object is its meaning. (‘A’ is the same sign as ‘A’.)
(TLP 3.203).

The Tractatus presents a simple vision of the world which consists of facts that
are independent from other facts (TLP 1; 2.06; 2.061). Facts are combinations
of objects (TLP 2.01). The world is, by virtue of the common logical form be-
tween facts and propositions, mirrored in language (TLP 2.12; 2.18). It presents a
view on language in which elementary propositions are the most basic bricks of
language – they depict and assert the existence of states of affairs (TLP 4.21). El-
ementary propositions are concatenations of names (TLP 4.22) and they picture
states of affairs exactly because of the relation of reference between names and
their bearers – objects. The pictorial relation between a proposition and a respec-
tive state of affairs exists because of correlations of the elements of the proposi-
tion with the elements of the state of affairs (TLP 2.1514). Wittgenstein compares
names to feelers (die Fühlers). It is because of these “feelers” that “the picture
touches reality” (TLP 2.1515). These scattered remarks are summed up by Witt-
genstein in theses TLP 4.0311-4.0312:
One name stands for one thing, another for another thing, and they are combined with
one another. In this way the whole group—like a tableau vivant—presents a state of af-
fairs. The possibility of propositions is based on the principle that objects have signs as
their representatives.

Semantic atomism is often connected with a realist theory of meaning51. On a re-

alist reading of the Tractatus, the possibility of the reference of names outside the
context of a proposition is grounded in the fact that reality consists of objects that
are “ready” to be named. This is because reality is already divided into objects, and
a name can have a constant referent. Language simply represents the structure of
reality (it does not participate in creating it)52. That is why it is enough to know

51 “On one well established interpretation, the Tractatus presents a realist theory of
meaning, which conceives the representing relation as consisting of a direct link
between bits of language (words) and bits of the world (objects)” (McGinn 2007,
p. 200). Pears defines realism with respect to the theory of meaning as an idea that:
“when a name is attached to a thing, the nature of the thing takes over and dictates
its subsequent use” (Pears 1987, p. 65). Morris describes realism as a claim that “the
nature of the world as it is in itself is altogether independent of anything to do with
any thought or representation of it” (Morris 2008, p. 96).
52 As Child puts it: “On the realist reading of the Tractatus, the division of reality into
simple objects is an intrinsic feature of reality (…). It is because reality has the structure
it does that any language adequate to represent reality must have the same structure”
(Child 2011, p. 56–57).

27
what the referents of names are in order to know what the world consists of. The
second claim of semantic atomism is, therefore:
(B) The reference of names is determined by objects, independent in their ex-
istence from any system of propositions (or any system of representation).
For Norman Malcolm (a noted proponent of the realist reading of the Tractarian
theory of meaning), this theory was so clear and obvious that he found it “difficult
to understand how anyone could study the Tractatus and come away with the im-
pression that objects owe their nature as simple to human convention or choice”53.
Despite Malcolm’s difficulties, there were many commentators who contested
the established, i.e. realist, interpretation of the Tractatus. In 1969 Hidé Ishiguro
wrote a paper titled Use and Reference of Names, in which she undermined the
thus common trust in the realist interpretation of the Tractarian theory of mean-
ing. Her objections to the prevailing reading of the Tractatus were soon shared by
Brian McGuinness and Peter Winch. Different doubts to the realist interpretation
were also raised by such scholars as Rush Rhees, Marie McGinn, William Child
and Hans Sluga.
The proponents of the anti-metaphysical interpretation of the Tractatus fight
not only with the idea of a “ready to be named” reality that consists of some basic
bricks – the simples – but also with the idea which is entailed by a realist theory
of meaning, i.e. with the claim that a name has a reference outside a proposition.
At the fore of Ishiguro’s argumentation comes the so-called context principle:
Only propositions have sense; only in the nexus of a proposition does a name have
meaning (TLP 3.3).

It was Frege who introduced the distinction between, using his terminology, sense
(Sinn) and meaning (Bedeutung). The meaning of a sign was, for him, an object
which is designated by a name; and the sense of a sign was a mode of presentation
of a designated object54. This is the terminology inherited in the Tractatus by Witt-
genstein55, so wherever I use in this work the concept of reference, it is synonymous

53 Malcolm 1986, p. 60–61.

54 “It is natural to think of there being connected with a sign (name, combination of
words, written mark), besides that which the sign designates, which may be called the
meaning of the sign, also what I should like to call the sense of the sign, wherein the
mode of presentation is contained” (Frege 1892/1984a, p. 158).
55 “Throughout the Tractatus Wittgenstein makes a distinction between Sinn and
Bedeutung in a manner roughly corresponding to Frege’s later works (…) and he took
‘Bedeutung’ in the sense of ‘reference’ ” (Ishiguro 1969, p. 23).

28
with the Tractarian notion of meaning. For instance, when Wittgenstein writes that
“Der Name bedeutet den Gegenstand” (a name means an object)56, under the con-
ceptualisation I am using he means that a name refers to an object (under my con-
ceptualisation the concepts of meaning and of sense are synonymous). It seems,
therefore, that in TLP 3.3, Wittgenstein states that only in the nexus of a proposi-
tion does a name have a reference. This claim would contradict (at least at first
sight) semantic atomism (which claims that a name has a reference also outside the
context of a proposition). How does it contradict semantic atomism in detail? To
answer this question one has to go back to the philosophy of Frege.
Frege continually reminds us that “what is being said concerning a concept
does not suit an object”57. For instance, a concept’s behaviour in a sentence is
essentially predicative and it cannot be replaced in a sentence by a name. What
is the connection between the object-concept distinction and a claim that a
name has a reference only in a proposition? Frege takes as an example the sign:
“Vienna”58. He claims that given only this sign, one cannot decide what its ref-
erence is, i.e. one needs the context of a proposition to determine it. “Vienna”
in the proposition: “Vienna is the capital of Austria” refers to an object, namely
to the largest city in Austria (“Vienna” is in the given sentence a proper name),
but it can turn out as well that it pertains to a concept, as in the proposition:
“Trieste is no Vienna”. In the second case “Vienna” is a concept-word like “city”.
This is the central reason why Frege introduced the context principle59.
On the other hand, as Ishiguro admits, in the Tractatus there is no distinction
between objects and concepts, and Wittgenstein did not think “in terms of satu-
rated and unsaturated, or complete and incomplete sense”60, so clearly the context
principle has to have a different function than that in Frege’s writings. According
to Ishiguro, the context principle in the Tractatus serves as evidence that Witt-
genstein was interested in the question: “How is the reference of a word fixed?”61
What secures the reference of a word? How can I be sure that when I refer to the

56 TLP 3.203.
57 Frege 1892/1984b, p. 190.
58 Ibid., p. 189.
59 “Als Grundsätze habe ich in dieser Untersuchung folgende festgehalten: es ist das
Psychologische von dem Logischen, das Subjektive von dem Objektiven scharf zu trennen;
nach der Bedeutung der Wörter muss im Satzzusammenhange, nicht in ihrer Vereinzelung
gefragt werden; der Unterschied zwischen Begriff und Gegenstand ist im Auge zu behalten”
(Frege 1884, p. X).
60 Ishiguro 1969, p. 24.
61 Ibid., p. 20–21.

29
apple tree in my garden with the name: “APTree”, I am referring to the tree and
not to its shape or to its bark?
The proponents of the formalistic reading of the notion of the object claim,
roughly speaking, that Wittgenstein had at his disposal two theories with respect
to the question of how one can identify an object of reference and secure refer-
ence of a name. One can identify an object either by a definite description or by
a demonstrative (for instance, by a mental act or by pointing at something). One
can identify an object by a description such as: “The highest tower in Fribourg”
or by the ostensive definition – pointing at the tower of the cathedral of Fribourg.
But because of the characteristics of the simple objects given in the Tractatus,
Wittgenstein could use neither of these ways to identify objects62. The simples of
the Tractatus are, for example, colourless (TLP 2.0232), in contrast to objects with
which one is acquainted or which one can point to. Therefore, the simples can-
not be identified by means of ostensive definitions63. A description, on the other
hand, pertains to the external features of an object and, according to the Tractatus,
external (material) features are produced by configurations of objects (just as, in
our example – one can speak of the highest tower only in comparison to other
towers), and as such they do not belong to the Tractarian substance of the world
(TLP 2.0231) made up of simples (TLP 2.021). By means of a description one can
at best identify a complex object, but not a simple one. There is no identification
of a simple object through an ostensive definition or through a description – this
is what the adherents of the existence of the context principle in the Tractatus con-
cluded. The only way one can identify an object is through the use of a name which
designates it64. It is the use of a name which secures the reference. If I asked at the
beginning of the paragraph what the guarantee is that if I name an apple tree in
my garden using the name “APTree”, I am truly referring to the tree and not to
its shape, Ishiguro answers: “There is no such guarantee”. One can recognise the
reference of “APTree” only by the later use of the name in question; for instance, if
I wanted to cut an APTree because it obscures the view from my window, then the
hypothesis that an APTree refers to a shape would be eliminated. One can observe
the use of the name – one can refute or confirm one’s hypotheses with respect
to possible referents of names – only in propositions, such as, for instance: “The
APTree obscures the view from my window. I have to cut it”. In Ishiguro’s opinion

62 Stokhof 2002, p. 161.

63 Ishiguro 1969, p. 28.
64 “In the Tractatus Wittgenstein is anxious to stress that we cannot see how the name
refers to an object except by understanding the role it plays in propositions” (ibid.,
p. 23).

30
this is what Wittgenstein had in mind when he wrote that a name gains a reference
only in the nexus of a proposition.
According to Ishiguro, problems connected with identifying objects of refer-
ence and the fact that there is no solution which semantic atomism could offer in
this respect should lead to the refutation of this position:
The Tractatus theory of names is basically correct, however, in so far as it is a refutation
of views which assume that a name is like a piece of label which we tag on to an object
which we can already identify (Ishiguro 1969, p. 35).

If the reference of a name were fixed by its use, then it would falsify the realist intui-
tion that referents are simply there in the world, waiting to be named. Peter Winch
indicated that the line of Ishiguro’s argumentation resembles “the beetle in the box”
argument from Philosophical Investigations (PI, 293). The fictitious proponent of a
private language has exactly the same problems with establishing the reference of
his or her private experiences as does the proponent of semantic atomism, in Ishig-
uro’s view, with establishing the reference of simple objects. According to Winch,
we should remember the lesson from “the beetle in the box” argument: “if we con-
strue the grammar of the expression of sensation on the model of ‘object and name’,
the object drops out of consideration as irrelevant” (PI, 293). Winch sums up his
commentary on Ishiguro in the following way: “The Tractarian objects are quite
unnecessary, an idle wheel, the intrusion of which is masking the true workings of
the mechanism”65. If Winch was right in his conclusions, then, naturally, the ques-
tion of the nature of Tractarian simple objects would be senseless.
The proponents of the thesis that the system of the Tractatus manifests a real-
ist theory of meaning are challenged by Ishiguro in two ways: first, they have to
propose a way to identify objects that would be in accordance with both a real-
ist theory of meaning as well as the requirements that Wittgenstein put on the
simples. Second, they have to offer a different reading of TLP 3.3, i.e. a reading
that would not assume that Wittgenstein in this thesis raised the issue of how one
fixes the reference of names (because if that was the topic of this thesis, then it
would be difficult to avoid Ishiguro’s conclusions).

1.1.3  In defence of semantic atomism

In the following section I want to refer to the aforementioned arguments and the-
ses of an anti-metaphysical reading of the Tractatus. I would summarise them as
follows:

65 Winch 1969a, p. 13.

31
1. The Tractatus is interested in logic and treats philosophy as a critique of lan-
guage. It does not have metaphysical ambitions. This means that, according to
this book, in order to set the limits of sense, a philosopher does not make use of
metaphysics.
2. According to the Tractatus, an object is a value of a variable. The notion of an
object is a purely formal concept.
3. One of the central topics of the Tractatus is the question as to how one fixes the
reference of names. The reference of a name is fixed by its use in propositions.

Ad 1. Is there no interest of early Wittgenstein in metaphysics?

Wittgenstein’s interest was the nature of the proposition. It is not my intention
to argue with this claim. As he writes in the Notebooks: “My whole task consists
in explaining the nature of the proposition” (NB 22.1.15, p. 39). However, my
intention is to show that in order to fulfil the task of explaining the nature of the
proposition, he had to direct his thoughts to reality. For example, exactly in the
same entry of the Notebooks he admits that explaining the nature of the proposi-
tion leads to an inquiry about the nature of facts, and, consequently, of all being:
My whole task consists in explaining the nature of the proposition. That is to say, in giv-
ing the nature of all facts, whose picture the proposition is. In giving the nature of all
being” (NB 22.1.15, p. 39).

He was aware of this unexpected metaphysical twist of his primarily logical work:
My work has extended from the foundations of logic to the nature of the world (NB 2.8.16,
p. 79).

One can indeed keep track of Wittgenstein’s development in the Notebooks. He

began with the repeatable remark that “logic must take care of itself ”66, and the
correction of the Fregean criterion of a meaningful sentence: “Frege says: Every
well-formed sentence must make sense; and I say: Every possible sentence is well-
formed” (NB 2.9.14, p. 2). But in the next month he introduced the picture theory
of meaning (NB 29.9.14, p. 7). One of the consequences of this theory on his views
on proposition is the conviction that: “In the proposition the name goes proxy for
the object” (NB 29.12.14, p. 37). From this statement there is a short path to the
topic of simples and complexes, with which Wittgenstein was absorbed in May of
1915. The first remarks with respect to the possibility of giving an example of a
simple object are sceptical. As the favourite entry of the proponents of the anti-
metaphysical interpretation states:

66 NB 22.8.14, p. 2; 2.9.14, p. 2; 3.9.14, p. 2; 13.10.14, p. 11; 26.4.15, p. 43.

32
 It keeps on looking as if the question ‘Are there simple things?’ made sense. And surely
this question must be nonsense” (NB 5.5.15, p. 45).

But soon Wittgenstein divorced himself from this anti-metaphysical attitude.

I think that he realised that in order to ensure a solid ground for his linguistic
system, one has to anchor it in reality; otherwise, without any example of a simple
object, one could accuse Wittgenstein that his ideal language – containing only
notions with sharp meanings – is merely a chimera. First, he seriously considered
the candidacy of the points in the visual field as the simples. He said, among other
things, that the simples are “the simplest things with which we are acquainted”
(NB 11.5.15, p. 47), but since he realised that in fact the things with which we are
acquainted are complex things, he rejected this view. He did not, however, aban-
don the search for the simples. At the same time he admitted that the fact that
we are acquainted with complex objects encourages us to seek the parts of which
these complexes are constructed:
Even though we have no acquaintance with simple objects we do know complex by ac-
quaintance, we know by acquaintance that they are complex. –And in the end they must
consist of simple things? We single out a part of our visual field, for example, and we see
that it is always complex, that any part of it is still complex but is already simpler, and so
on (NB 24.5.15, p. 50).

From this moment on Wittgenstein turned his attention to the particles of phys-
ics as the most probable candidacy for the simples (I shall track this change of
mind more carefully in Chapter 2.):
The division of the body into material points, as we have seen it in physics, is nothing
more than analysis into simple components (NB 20.6.15, p. 67)

I shall also argue there that the conviction that simple objects are the most basic el-
ements of matter had not left Wittgenstein throughout the time of his writing and
editing the Tractatus. This claim does not intend to suggest that Wittgenstein sud-
denly “discovered” that the world consists of material simple objects; it intends to
say that Wittgenstein realised that only the language of physics fulfils the require-
ments he put on meaningful language (for instance, the postulate of a determinate
sense or the postulate of the logical independence of elementary propositions).
Concluding, my thesis is that the metaphysical twist of Wittgenstein’s linguis-
tic inquiries took on the following form: The first stage included introducing the
picture theory of meaning and seeing a proposition as a picture of a state of af-
fairs (September of 1914). In other words, the central idea of the picture theory is
stated as: (1) Meaningful propositions mirror reality. The next steps were the con-
sequences of the first step. In June of 1915 Wittgenstein realised that: (2) Only the

33
language of physics fulfils the requirements of meaningful discourse67. From (1)
and (2) it follows that: (3) Only the language of physics mirrors reality. Combin-
ing it with a realist theory of meaning which claims that the names of a given lan-
guage refer to objects that are independent from us in their existence, one arrives
at the metaphysical conclusion that: (4) The world consists of physical particles.
Of course, it should not escape our attention that, according to Wittgenstein, we
discover the simples as a requirement of the logical theory: the concept of a simple
object is contained in the idea of analysis, and the best proof for that is the fact that
one can come to this idea without providing any examples of the simples68. But on
the other hand, the fact that we, so to speak, a priori assume the existence of the
simples did not prevent Wittgenstein from formulating, in June of 1916, the thesis
of semantic atomism:
It is quite clear that I can in fact correlate a name with the watch just as it lies here ticking
in front of me, and that this name will have reference outside any proposition in the very
sense I have always given that word (NB 15.6.15, p. 60)
We feel that the WORLD must consist of elements (…) The world has a fixed structure
(NB 17.6.15, p. 15)

McGinn thinks that the intellectual journey of young Wittgenstein led from a
realist theory of meaning to the formalistic conception of a name:
There are striking and significant differences between the remarks on simples that oc-
cur in the Notebooks and the corresponding sections on simples in the Tractatus and the
Prototractatus. In the Notebooks, Wittgenstein makes a number of remarks that strongly
invite the sort of realist reading I want to argue against (…) Wittgenstein ultimately finds
the source for the requirement that names stand for simples objects in the demand for
definiteness of sense and the possibility for complete analysis (McGinn 2007, p. 202–203).

This cannot be true: Wittgenstein wrote the remark by stating that the world
has a fixed structure just one day after he had admitted that the existence of the
simples is the consequence of the postulate of a complete analysis. This means
that holding the requirement of a determinate sense, at least for Wittgenstein,
was not in contradiction to a realist theory of meaning. And even if one supposes
that the requirement of a determinate sense stands in contradiction to a realist

67 I shall elaborate this point in Chapter 2.

68 “It seems that the idea of the SIMPLE is already to be found contained in that of the
complex and in the idea of analysis, and in such a way that we come to this idea quite
apart from any examples of simple objects, or of propositions which mention them,
and we realize the existence of the simple object – a priori – as a logical necessity”
(NB 14.6.15, p. 60).

34
theory of meaning (I personally do not think that there is any contradiction be-
tween these two claims), then one has to take into account that the remarks about
the fixed structure of the world follows in the Notebooks the remarks about the
requirement of a determinate sense. In other words, even if Wittgenstein saw the
contradiction in holding both of these groups of remarks, then the conclusion
should be that he, at some point, rejected the postulate of the definite sense and
stood by the realist convictions.
Both realists and formalists agree that the origins of the Tractatus go back to
logic, but they disagree when it comes to the question: “Did Wittgenstein take one
step further in the direction of metaphysics?”. The realists would stress the impor-
tance of the moment when the picture theory was introduced in the Notebooks69,
while the formalists would incorporate the picture theory as a part of the theory of
language and would not tie any metaphysical consequences to it70. Why is the mo-
ment of introducing the picture theory of meaning of such importance to the pro-
ponents of a realist reading of the Tractatus? It is because the picture theory claims
that “in order to tell whether a picture is true or false we must compare it with
reality” (TLP 2.223), and in the eyes of the early commentator of Wittgenstein’s
book, Otto Neurath, the idea of the comparison of a sentence with reality is purely
metaphysical71. If Wittgenstein truly wanted to avoid any metaphysical standpoint,
he should have rather said that in order to tell whether a picture is true or false, one
compares it with another picture or system of pictures72. Each new proposition,
according to the line of thought of the anti-metaphysical interpretation, should
be compared with the totality of previously coordinated existing propositions. But
that is exactly, as will be clear in section 1.2, when I shall discuss the requirement
of a determinate sense, which Wittgenstein did not want to say.
Summing up, I think that on the basis of the entries from the Notebooks one
can describe the development of Wittgenstein’s thought as being marked by a
constant growth of interest in metaphysical topics. Second, the argument that the
simple objects were first thought of as a requirement of a definite sense does not
contradict this observed interest in metaphysics – as we have seen, Wittgenstein

69 von Wright 1955/1967, p. 18; Stokhof 2002, p. 22–23.

70 Rhees 1969/1970, p. 25.
71 “Statements are compared with statements, not with ‘experiences’, ‘the world’, or anything
else. All these meaningless duplications belong to a more or less refined metaphysics and
are, for those reason, to be rejected” (Neurath 1959, p. 291).
72 “A statement is always compared with another statement or with the system of statements,
never with a ‘reality’. Such a procedure would be metaphysical; it would be meaningless”
(ibid., p. 292).

35
reconciled this requirement with claims that the world has a fixed structure. Fi-
nally, the picture theory of meaning, with its idea of a comparison of a proposi-
tion with a state of affairs instead of a comparison of a proposition with a system
of propositions, also betrays the metaphysical inclinations of early Wittgenstein.
That, I think, allows to falsify the claim of the anti-metaphysical interpretation of
the Tractatus that early Wittgenstein’s considerations on language are divorced
from reality.

Ad 2.) Is the concept of an object a purely formal concept?

Why should we reject the conception according to which an object is every-
thing which can be a value of a variable x in a propositional function? For me,
the most obvious answer is that it is in disagreement with Wittgenstein’s identi-
fication of propositions of natural science with the totality of true propositions
(TLP 4.11). The task of natural science is to discover the internal features of the
world (features that are independent from observers). But according to the for-
malistic conception of an object there can be true propositions about justice,
beauty or baseball (why not replace x in fx with the notions of an off-side, or a
truce?) – notions that are essentially connected with human activities which,
by definition, are not subject-matter for natural science. In other words, thesis
TLP 4.11, in the light of the formalistic conception of an object, seems to be
unjustified and dogmatic, and that is the conclusion one should try to avoid.
Moreover, the formalistic conception of the Tractarian object allows different
types of things to be referents of names: abstract objects such as economic
growth, objects of science such as chemical elements, objects of everyday life
such as knives and forks, particulars such as stones, and complexes such as the
Swiss nation. We have at our disposal evidence that Wittgenstein’s conviction
was exactly the opposite. Commenting on the Tractarian solution to the theory
of types he writes in one of his letters to Russell:
I think that there cannot be different types of things! In other words whatever can by
symbolized by a simple proper name must belong to one type (Letters, 16.1.13, p. 19)

Obviously, proponents of the formalistic interpretation of the Tractatus can de-

fend their position by saying that, of course, all examples of objects, as I men-
tioned before, belong to one type of things, i.e. to logical objects. In my opinion,
however, when Wittgenstein writes: “There cannot be different types of things”, he
cannot have in mind something such as: “There cannot be illogical objects”. His
letter to Russell is much more comprehensible if we understand under the type
of things such categories as abstract and concrete things or, to put it differently,
relations and individual things. This is because the letter is clearly a discussion

36
with Russell’s views, on which “atomic sentences at the bottom level of analysis are
combinations of names of ontological atoms of different types – individuals, prop-
erties of individuals, dyadic relations of individuals, etc”73. Johnston also notices
that for Russell a term was an entity which could appear in a complex, and that he
discerned two logical types of terms: universals (that is relations of all orders) and
particulars (terms which are not universals)74. Therefore, Wittgenstein’s restriction
is nothing more than saying that objects of reference belong either to the category
of abstracts or to the category of individual objects. Moreover, one can make a
probable case that TLP 3.1432:
Instead of, ‘The complex sign “aRb” says that a stands to b in the relation R’, we ought to
put, ‘That “a” stands to “b” in a certain relation says that aRb.’

opts for the rejection of the ontological category of relations75. It would turn out
that Wittgenstein’s conviction that there cannot be different types of things is
tantamount to the claim that there is just one category of objects – the category
of particulars, and that is the conclusion which stays in opposition to the for-
malistic notion of a thing as anything that can be referred to by a third-person
singular pronoun.
Which is not to dismiss that there is a grain of truth in the formalistic notion
of the object. This notion is, indeed, in the light of other Tractarian theses, trou-
blesome. This is because the sentence: “There are objects” is a perfect example of
Tractarian nonsense: anyone who truly understands it must recognise it as mean-
ingless (TLP 6.54):
So one cannot say, for example, “There are objects”, as one might say, “There are books”
(TLP 4.1272).
To ask whether a formal concept exists is nonsensical. For no proposition can be the
answer to such a question (TLP 4.1274).

The Tractatus treats the concept of the object as a formal one76. One can ask
neither meaningfully about formal concepts nor about the instances of these

73 Ricketts 1996, p. 69.

74 Johnston 2012, p. 16.
75 “Wittgenstein rejects the reality of relations (…). Relations are not things, are not
entities; relations cannot be labelled or designated. Unlike ‘a’ and ‘b’, ‘R’ is not a symbol
in ‘aRb’. Instead, roughly put, the holding of a relation over objects is symbolized by
the holding of a relation over names of those objects. But this way of thinking is itself
misleading for its use of ‘object’ and ‘relation’ as a contrasting pair of common nouns”
(Ricketts 1996, p. 72).
76 Formal concepts are those corresponding to formal properties. C.f. TLP 4.126.

37
concepts. One also cannot express formal concepts in propositions because, ac-
cording to Wittgenstein, formal concepts are expressed in propositions by vari-
ables (TLP 4.127), and variables show a form (a form cannot be represented – it
can only be shown) (TLP 4.1271). According to TLP 4.1272, a proposition:
(a)  There are tables
is meaningful, and the expression
(b)  There are objects
is meaningless. The fact that “object” denotes nothing is shown in the sentence
by the variables. According to TLP 4.1272, in the logical notation “x” is a proper
sign for the formal concept of “object”. The right analysis reveals the form of (b)
as follows:
(b’)  There are x
(b’) is neither true or false. It is a function whose set of values covers concrete
objects, but not the genus of them. To say: “It is meaningless” means that the
meaning of (b’) is not determined yet. The other example covers the pair of the
following propositions:
(c)  The table is brown
(d)  The table is an object
Sentence (c) is meaningful – expression (d) on the contrary77. Sentence (d) says
nothing, nothing is learnt by it, it gives no information. The concept of the object
is not supposed to give any information at all – it shows the form of the sentence,
therefore, it points out what kind of names could be values of x in the proposition.
The proper analysis reveals the form of the expression containing the concept of
the object:
(e)  x is brown.
When one expresses proposition (c), one has already put the propositional func-
tion (e) to use, and this means that one already knows the meaning of the formal
concept: “object”78. It is now clear that no new information about a table is added
in (d). What (d) tries to say is already given along with the proper uses of the name:
“table”.

77 “When something falls under a formal concept as one of its objects, this cannot be
expressed by means of a proposition” (TLP 4.126).
78 Block 1986, p. 141.

38
Nevertheless, the Tractaus contains theses such as TLP 2.02 (“Objects are simple”).
In the Notebooks, Wittgenstein considered different candidacies for the simples79.
We could repeat Sluga’s question: “If Wittgenstein is not seriously entertaining the
idea that visual points might be the simple objects he is looking for, then what is
the point of raising the question?”80. In the Tractarian system, expressions which
state the existence of something are, as we have seen, meaningless. This does not
mean, however, that there are no ontological commitments in that system. One
should not, however, ascribe them to existential propositions but to the function
of a name in an elementary proposition. For example, let us take an elementary
proposition:
(f) “AB”
where “A” and “B” are simple symbols which refer, respectively, to the simple ob-
jects A and B. In order to express the existence of A and B, one does not have to say:
(g)  “A and B exist”
because this is already contained in (f)81. (g) is meaningless because it is equiva-
lent to:
(h)  “AB exist exist”82
We can combine the thesis that the concept of the object is formal with the fact
that Wittgenstein’s work “extended (…) to the nature of the world” (NB 2.8.16,
p. 79) by accepting that expressing meaningless sentences counts among the nec-
essary costs of climbing up the Tractarian ladder in order to see the world aright
(TLP 6.54). A statement such as: “There are simple objects” is one of the steps on
the ladder. One who has climbed up sees the world aright. But there is no other
way to see the world aright than through the process of transcending the meaning-
less sentences of the Tractatus, and one of these could be the thesis of the existence

79 “It seems to me perfectly possible that patches in our visual field are simple objects”
(NB 18.6.15, p. 64); “The division of the body into material points, as we have it in
physics, is nothing more than analysis into simple components” (NB 20.6.15, p. 67).
80 Sluga 2012, p. 103. He admits the right to Ishiguro’s commentary. According to him,
behind the questions about the simples stands Wittgenstein’s concern if we recognise
that the analysis of a proposition is complete (Sluga 2012, p. 104).
81 “That M is a thing can’t be said; it is a nonsense: but something is shown by the symbol
‘M’ ” (NM, p. 109).
82 We could also prove its meaningless by analysing the negation of (f). (f*) “AB does not
exist”. In logical analysis of (f*) we get to proposition (f#) Ex (x is AB and AB does not
exist). There is something which does not exist.

39
of the simples83. This resembles Frege’s dilemmas with the concept of concept.
Because Frege discerned sharply between names and concepts, he was convinced
that what one asserts about a name cannot be stated about a concept84. One of the
consequences of this claim is, as Conant notices, the impossibility of saying any-
thing about concepts85. Every time one tries to ascribe to concepts certain features,
the only result is nonsense. For instance, the proposition: “The concept ‘horse’ is a
concept that is easily attained” is nonsense. The whole confusion is the result of the
fact that in the sentence: “The concept ‘horse’ is a concept that is easily attained”,
the linguistic sign: “The concept ‘horse’ ” functions as a name. Names refer to ob-
jects. To say that an object is a concept that is easily attained is neither true nor
false. It is nonsense but, nevertheless, it is elucidatory nonsense. We know what
one is trying to say by means of this proposition. It is only the awkwardness of
language, claims Frege, which hinders expressing ideas about concepts in a mean-
ingful manner86. Fregean perplexity is, I suppose, similar to the confusion with
Tractarian objects, about which one also cannot assert anything meaningful, and
yet those meaningless expressions such as “Objects are simple” serve as a medium
to show something substantial about the world. This also means that a proponent
of semantic atomism is bound to accept the hypothesis of substantial nonsense
(the idea combatted by the so-called New Reading of the Tractatus).
Summing up, although the notion of the object is, in the opinion of the author
of the Tractatus, formal, and it means that nothing in the world corresponds to
this concept, nevertheless an inquiry about the nature of objects is elucidatory. It
reveals the truth about the world. Expressions which assert something about ob-
jects, strictly speaking, are nonsensical, but they are a necessary stage of seeing the

83 We could also act on the hint of Anscombe and Black and say that we are dealing here
with the ambiguity of the concept of the object. “Object” from TLP 2.02 and “object”
from TLP 4.126 must not be taken in the same sense. The latter sense is just technical
and serves to explain some views on logic and, first of all, the difference between
showing and saying, while the former sense is metaphysical, and it serves to explain
the ontological structure of reality (Anscombe 1965, p. 122; Black 1964, p. 199).
84 “I do not want to say it is false to say concerning an object what is said here concerning
a concept; I want to say it is impossible, senseless to do so” (Frege 1892/1984b, p. 189).
85 “The idea that what such an attempt (to assert of a concept what can only be asserted
of an object) ends up saying is not merely false, but senseless, is one which runs
throughout Frege’s writings (…). Frege commits himself to the conclusion that what
he himself wants to say about concepts is also nonsensical” (Conant 2000, p. 224).
86 “It must indeed be recognized that here we are confronted by an awkwardness of
language, which I admit cannot be avoided, if we say that the concept horse is not a
concept” (Frege 1892/1984b, p. 185–186).

40
world aright. According to the materialistic reading, this necessary stage includes
asserting the feature of simplicity to objects. This claim, although nonsensical, al-
lows to see the world aright – as the totality of physical particles. The proponents
of the physicalist interpretation agree that there are no objects in the world (there
are physical atoms, but not objects), but nevertheless they do not consider expres-
sions about objects as pure nonsense that is comparable to childish babbling, or as
a manifestation of a belief in a metaphysical myth. In their opinion these expres-
sions form a necessary step in order to acknowledge the truth of materialism.

Ad 3) Is the reference of names fixed by their use in propositions?

According to Ishiguro, Wittgenstein was moved by the question as to how one
identifies objects of reference? This question assumes the existence of a particu-
lar point of view – the point of view of a user of a language who has to identify
the objects of reference. Hence, she must be thinking about early Wittgenstein as
interested in topics engaging subjective points of views. I do not agree with this
opinion: Wittgenstein tried in the Tractatus to look at the world sub specie aeterni
(TLP 6.45), which means at the world as a whole, without any special, distinctive
points of view, including the point of view of a user of language. The basic Tractar-
ian idea is to describe the world objectively, as it is itself   87. The lack of a subjective
perspective in Wittgenstein’s project of describing the world is especially clear in
the thesis TLP 5.631:
There is no such thing as the subject that thinks or entertains ideas. If I wrote a book
called The World as I found it, I should have to include a report on my body, and should
have to say which parts were subordinate to my will, and which were not, etc., this being
a method of isolating the subject, or rather of showing that in an important sense there
is no subject; for it alone could not be mentioned in that book.

In this thesis Wittgenstein examines a project which assumes the subjective per-
spective – the world as I have found it. But for him even if I tried to accomplish this
task and if I described this world meaningfully, then I would not mention a subject
in my book simply because a subject of representation does not exist. Wittgen-
stein’s striving for maximal objectivity is also clear in identification of meaningful

87 According to Shanker, Wittgenstein tried to replace “epistemological approaches to

the major issues that had troubled philosophy with a strictly logical point of view”
(Shanker 1986a, p. 28), Hacker sees “The Tractatus as an attempt to supply a complete
answer to these questions [How a description of the world by means of a system of
representation is possible? How symbols can say how the world is?] by developing a
rigorously realist theory of meaning” (Hacker 1972, p. 33).

41
discourse with the set of propositions of the natural sciences (TLP 4.11). It is clear
that the natural sciences describe the world sub specie aeterni without singling out
any special points of view, i.e. the world as it is, independently of human beings.
The aim of science is to deliver a description which is valid and accessible to all
rational beings.
Ishiguro accuses the standard interpretation of offering no solution to the
problem of the identification of objects. The following example will illustrate
that the realistic standpoint does not have to offer any solution in this respect88.
Let us imagine a room which one should describe without invoking the subjec-
tive perspective, so the way in which I come to the room and describe what I see
is excluded. Therefore, the qualities of objects would not occur in the descrip-
tion of the room. This is in agreement with the standpoint of the Tractatus, ac-
cording to which objects are colourless (TLP 2.0232) and do not have material
properties. I cannot use descriptions of objects (such as: “an object standing
next to the window”, “the heaviest object in the room”) either because descrip-
tions pertain to material (external) properties, which are contingent. According
to the Tractatus, objects form the substance of the world, and that substance
does not determine the material properties (TLP 2.0231). Apart from that, in
the description of the room one should pertain to existing objects, and if one
identifies them by descriptions referring to contingent properties, one is always
in danger of speaking about non-existing things. If one, in order to describe
the room in an objective manner (this means without invoking any subjective
perspective), cannot describe it without coming into it or by means of such
object-descriptions, then it seems that an objective description of the room is
impossible. Even if we managed to describe it properly, it would be by a pure
and inconceivable accident. Normally, we would just use our imagination and
the effect of our work would have nothing to do with the real room.
What method of description of the room is then left for the realist? He should,
so to say a priori, establish that in the description he wants to have names that
never lose their designates. As we will see in the next chapter, that condition is
fulfilled when one introduces simple objects as the basic structure of the world89.

88 The inspiration for the example comes from the Notebooks: “Let us imagine a white
surface with irregular black spots on it. We now say: Whatever sort of picture arises
in this way, I shall always be able to approximate as close as I like to its description by
covering the surface with a suitably fine square network and saying of each square that
it is white or is black” (NB 6.12.14, p. 35).
89 Anscombe 1965, p. 28; Black 1964, p. 28.

42
As far as I am concerned, this suits the spirit of Wittgenstein’s line of argumenta-
tion from the Notebooks:
It also seems certain that we do not infer the existence of simple objects from the ex-
istence of particular simple objects, but rather know them (…) as the end-product of
analysis (NB 23.5.15, p. 50)90.

Proper names are names which designate simple objects. The conclusion is that
a realistic, objective description of the room would include only proper names.
That is what the Tractatus exactly says, we should just replace the name “room”
with the name “the world”. The full description of the world consists in elemen-
tary sentences which are concatenations of names which designate simple ob-
jects. One could ask: “If we assume objects to be a requirement of the theory of
language, then is it truly a realistic interpretation?”. It is obvious that in order to
avoid this kind of accusation and in order to avoid going into (at least for me) the
death alley of a Kantian reading of the Tractatus, the next step is necessary. If we
said that Wittgenstein was not interested in the question as to what simple ob-
jects are, we would be left with a picture in which philosophy imposes some re-
strictions on reality, such as the existence of simple objects without considering
if there truly is such a thing as a simple object91. This would clearly be a postulate
of theoretical reason and we would represent a peculiar form of Kantianism92
instead of the coveted realism. What the realist, in my opinion, has to do – after
having established the requirements of the theory of language postulating the ex-
istence of simple objects – is to inquire if any factual language corresponds to the
requirements of the theory of language. In Timm Lampert’s opinion, Wittgen-
stein was not only aware of this problem, but also his answer was: “The language

90 According to Rom Harré, it also suits the practice of the physicists of the German
tradition, with which Wittgenstein was acquainted: “The older generation of physicists
held to a qualified realism, in that physics was importantly concerned with systems of
masses which we know must exist from certain conditions on the meaningfulness of
formulae expressing physical laws (…) In short the catalogue of elementary objects of the
world is necessitated not by experience but by the forms of the laws themselves” (Harré
2001, p. 214–215).
91 An example of such a position: “Central to his metaphysical conception is the conviction
that there must be such ultimate facts for there to be universe at all. But he will be wholly
unable to provide a specimen of an ‘atomic fact’: that there must be atomic facts, if there
are any facts at all, is known only a priori, through philosophical reflection” (Black 1964,
p. 28); “All we can really know about objects is that they exist” (ibid., p. 57).
92 Peculiar, because Kantianism assumes postulates of practical, not theoretical reason.

43
of mechanics suits the best the assumptions entailed by the theory of language”93.
Hence, I will defend in the next chapter the physicalistic reading of the Tractatus.
Going back to “the room example”, one must simply divide the space of
the room into the smallest parts, i.e. points, then coordinate these points with
names and finally determine the points by ascribing to them places in the space
of the room and in time. The final result, i.e. an elementary sentence, would be,
just as Wittgenstein wanted, a concatenation of names: a name for a material
point and a spatial and time point [Nn, Sn, Tn]. That would also explain why in
the system of the Tractatus there are facts (and not objects) which primarily ex-
ist and why simple objects necessarily exist in combination with other simples
(TLP 2.0121). These convictions, according to the line of reasoning I am pre-
senting here, would be a reflection of the Tractarian idea that a particle exists
necessarily in a given space and time94.
Let us now have a closer look at Ishiguro’s proposal, i.e. the thesis that the refer-
ence of a name is fixed in its usage in propositions. Ishiguro accepts the Fregean
position which ascribes priority to the concept of truth over the concept of refer-
ence95. According to her, early Wittgenstein shared the view that first one under-
stands the sense of a proposition and then, on the basis of this understanding,
one can grip the reference of names occurring in the proposition. The identifica-
tion of an object is secondary to the identification of the truth-conditions of the
proposition96. I reject the conviction according to which that was Wittgenstein’s
view. The indirect pieces of evidence in favour of such a rejection can be found

93 Lampert 2000, p. 228. “Mechanics determines the form of description of the world by
saying: All propositions in a description of the world by saying: All propositions in a
description of the world must be capable of being got in a given way from a number
of given propositions – the axioms of mechanics. In this way it supplies the stones for
building up natural science” (NB 6.12.14, p. 35).
94 “Just as we are quite unable to imagine spatial objects outside space or temporal objects
outside time, so too there is no object that we can imagine excluded from the possibility
of combining with others” (TLP 2.0121).
95 “What is distinctive about my conception of logic comes out first in that I give top
priority to the content of the word ‘true’ and then that I immediately introduce thoughts
as that concerning which the question of truth arises. I therefore do not begin with
concepts that I put together into thoughts or judgements. Rather, I obtain thought-
components by analysing thoughts” (G. Frege, Aufzeichungen für Ludwig Darmstaedter,
as cited in: Ricketts 2010, p. 155).
96 “The identity of the object referred to by a name cannot be settled prior to or independently
of the sense of the propositions in which they are used, and agreement about the truth of
some of these propositions” (Ishiguro 1969, p. 34).

44
in Philosophical Investigations, which were, as the author admitted, at least partly
intended as settling accounts with the claims of the Tractatus97. In PI, 115 Wittgen-
stein writes (most presumably) about his earlier standpoint that a certain “picture
[of language] held us captive”. The picture he describes at the beginning of his later
oeuvre:
These words, it seems to me, give us a particular picture of the essence of human lan-
guage. It is this: the words in language name objects a sentences are combinations of
such names. —– In this picture of language we find the roots of the following idea: Every
word has a meaning. This meaning is correlated with the word. It is the object for which
the word stands (PI, 1).

is strikingly similar to TLP 4.0311-4.031298. The rightness of later Wittgenstein’s

critique of semantic atomism is of no importance here; what counts is that Witt-
genstein himself understood his early work as an example of semantic atomism.
We have every reason to think that later Wittgenstein understood his earlier book
properly. There are many textual pieces of evidence from the Tractatus that suggest
that early Wittgenstein accepted the so-called compositionality thesis, according
to which the sense of a proposition is a product of the meaning of its constituents,
and one understands the proposition because one understands its elements. The
compositionality thesis stays in sharp opposition to the context principle. It is im-
portant to note that, according to semantic atomism, there are names which have
reference, and the proposition has meaning derivatively because its names have
secured reference. On the other hand, according to the anti-metaphysical reading,
the direction is reversed: the basic unit of sense is the proposition, and names have
reference because the process of analysis of the proposition was performed. This is
why the proponents of semantic atomism advocate in favour of the compositional-
ity thesis99, and the proponents of the anti-metaphysical interpretation back up the

97 “Then it suddenly seemed to me that I should publish those old ideas and the new
ones together: that the latter could be seen in the right light only by contrast with and
against the background of my older way of thinking” (PI, p. 4).
98 One name stands for one thing, another for another thing, and they are combined with
one another. In this way the whole group—like a tableau vivant—presents a state of
affairs. The possibility of propositions is based on the principle that objects have signs
as their representatives (TLP 4.0311-4.0312).
99 “Objects are the foundation for the theory of logical syntax of the Tractatus” (Malcolm
1986, p. 30).

45
context principle100. It seems to me that early Wittgenstein rather takes sides with
semantic atomism:
[A proposition] is understood by anyone who understands its constituents (TLP 4.024).
One name stands for one thing, another for another thing, and they are combined with
one another. In this way the whole group – like a tableau vivant – presents a state of af-
fairs (TLP 4.0311).
A proposition must use old expressions to communicate a new sense (TLP 4.03).

In the atomistic view, constituents of a proposition have reference independently

of the proposition. This idea seems to be confirmed in the Tractatus, when Witt-
genstein describes the process of translation:
When translating one language into another, we do not proceed by translating each prop-
osition of the one into a proposition of the other, but merely by translating the constituents
of propositions (TLP 4.025).

The opponents of semantic atomism would answer: how can a name refer out-
side the context of a proposition and how can the reference of a name be fixed
before fixing a truth-value of a proposition if Wittgenstein wrote something ex-
actly opposite in TLP 3.3 (“Only propositions have meaning. Only in the nexus
of a proposition does a name have a meaning”)?101
In my opinion, in order to avoid this kind of tension between the theses of the
Tractatus, one has to retain the compositionality thesis, which is better represent-
ed in the book and to re-interpret thesis TLP 3.3 itself so that it will not express
the context principle. One way of reconciling TLP 3.3 with TLP 4.24 is proposed
by Timm Lampert, and generally consists in putting TLP 3.3 in the broader con-
text. Lampert mentions that the alleged context principle is preceded by thesis

100 “We have seen that Wittgenstein favours a ‘top-down’ view, on which the correlations
between words and objects are affected by the ‘theory’ that is built into our use of the
language as a whole” (Chlid 2011, p. 35).
101 Child accuses the proponents of the compositionality thesis that they cannot avoid
regressus ad infinitum which would consist in the following line of reasoning: in
order for a name to pick out the object (α), one has to identify the object of refer-
ence. In order to identify the object, one would have to have a thought about it (β).
In order to have the thought (β) one would have to have another thought (γ), and
so on, ad infinitum (Child 2011, p. 33). I tried to show earlier that the best strategy
for an adherent of semantic atomism is to avoid considerations about the way one
identifies an object. Semantic atomism in the Tractarian version is connected with
the ideal of the impersonal representation of the world, a representation which does
not single out any particular point of view.

46
TLP 3.263, saying that “the meanings of primitive signs can be explained by means
of elucidations”, which suggests that Wittgenstein is in this fragment of the Trac-
tatus writing about propositions of ordinary language, and not about elementary
propositions. Accordingly, the claim that “a name has meaning only in the nexus
of a proposition” (TLP 3.3) would refer only to propositions of ordinary language.
It would be just an indication that sometimes one clarifies the reference of names
of ordinary language by a closer examination of the propositions in which a given
name occurs. One should not, however, associate any solutions with respect to
simple objects with this thesis102.
In my explanation of TLP 3.3 I would indicate the Tractarian isomorphism be-
tween language and reality. In order for language to depict reality, the logical syn-
tax of language must share its features with the features of reality. Thesis TLP 3.3
is, under such a reading, a linguistic formulation of thesis TLP 2.011, stating that
it is essential to things that they occur in states of affairs. But what is meant here
by “a state of affairs”? It should be brought to our attention that the concept as de-
fined by Wittgenstein differs from the most popular examples from philosophical
literature. At first glance one has the tendency to treat propositions such as:
(a)  A book lies on a table
as expressing states of affairs; in this case a state of affairs consisting of two objects
(a book and a table) and the relation between them (lying on). It is clear, however,
that one can think of a book independently of its lying on a table, therefore, when
Wittgenstein was writing about objects and their necessary occurrence in combi-
nations with other objects, he could not have had in mind states of affairs such as
(a). A better example for a state of affairs would be:
(b)  A yellow book
An object having a material property (being a yellow book is a material property
because the book could have a different colour and still be that book) is depicted
by a proposition. On the other hand, what is depicted by a proposition is a state of

102 Lampert 2000, p. 306. He also suspects that the context principle in the Tractatus is
not of Fregean provenience but betrays the engineering education of young Witt-
genstein, in the sense that in TLP 3.3 names are compared to parts of a machine, i.e.
parts which function and serve a purpose only as those parts, and not as independ-
ent objects (they, however, can exist independently or as parts of another machine):
“Sie sind wie Teile einer Maschine, die zusammenpassen. Sie allein bilden einen
Zusammenhang, ein Ganzes, weil jedes einzelne Teil schon die Form der andere
Teile enthält” (ibid., p. 278–279).

47
affairs, hence, an object having a material property is a state of affairs103. A material
point in space and time is a state of affairs. In language, one expresses this state of
affairs by means of an elementary proposition: [Nn, Sn, Tn]. In my view, TLP 3.3
expresses, then, in a formal mode what one expresses in a material mode when
one says that a material object must be situated in space and time104. This way of
reformulating TLP 3.3 has the one big advantage that it retains the coherency of
the Tractatus. Please note that it was Ishiguro who challenged thus such a com-
monly accepted realist interpretation. The challenge was two-fold. First, Ishiguro
asked what story a proponent of the realist theory of meaning can tell with respect
to fixing the reference of names. My response consisted in proving that the propo-
nent of the realist reading does not have to provide the way of identifying objects
of reference. Objects are, according to Wittgenstein, postulates of the theory of
meaningful language. What the realist should do is to search if any language corre-
sponds to Wittgenstein’s requirement with respect to meaningful language. If that
is true, then the reader of the Tractatus faces a dilemma: does Wittgenstein back
up the compositionality thesis, which is well documented in the book, or the con-
text principle, whose presence in the Tractatus is conditioned by a currently weak-
ened reading of TLP 3.3, and in which Wittgenstein was interested in the question
of the identification of objects of reference? In my opinion the choice is obvious
and favours the compositionality thesis. That also means that TLP 3.3 does not
express the context principle and the realist theory of meaning is not threatened.
The main problem of the chapter was to answer the question as to why Witt-
genstein identified the totality of true propositions with the propositions of natu-
ral science. My starting idea was the hypothesis that this claim follows from the
materialism of the Tractatus. The basic accusation against such an explanation

103 Later, Wittgenstein realised that there was confusion in the Tractatus concerning
the question: “what are facts: things like (a) or things like (b)?”. Later he began to
differentiate between complex objects and facts. According to Wittgenstein, the main
difference between facts and complexes are the following: (1) A complex can move
from one place to another, but a fact cannot (PR, p. 301); (2) A complex is spatial,
but not a fact (PR, p. 301); (3) A complex is composed of its parts, on the other hand,
it is completely misleading to say that, e.g. the fact that this ball is red is composed
of a ball and redness (PR, p. 302); (4) One can point to a complex, whereas one can-
not point to a fact (or, in other words, pointing to a fact, contrary to pointing to a
complex, means to assert something) (PR, p. 303).
104 Other commentators defending the realist reading of the Tractatus understand TLP 3.3
as a trivial statement that names are linguistic signs and as such are parts of proposi-
tions or language: “Wittgenstein’s thought may have been that names have no use
except as parts of propositions” (Black 1964, p. 118). Similar: Fogelin 1987, p. 33.

48
refers to the fact that the Tractatus is mainly a book on language and logic. There
is a suspicion that its author was not interested in metaphysical considerations,
including questions such as: “What does the world consist of?”. This suspicion
is enhanced if one reads thesis TLP 3.3 as expressing the context principle. One
of the consequences of this principle is that one cannot ask about the reference
of a name outside the context of the system of representation. One could at best
obtain the answer as to what objects are relatively to a given system of repre-
sentation. The reality that is independent from us is inaccessible. Accordingly, I
defended in section 1.1 the following theses:
1. Wittgenstein’s philosophical development shows a change of interest from
topics about language to topics about the relation between language and real-
ity. This means that early Wittgenstein did not avoid metaphysical questions
such as: “What does reality consist of?”.
2. Assertions about objects, strictly speaking, are nonsensical (they say some-
thing which cannot be said but only shown), but nevertheless they are a me-
dium to see the world properly.
3. In the Tractatus, Wittgenstein supported the compositionality thesis from
which it follows that propositions have sense because their components have
reference, and these components retain reference also outside a proposition.
The compositionality thesis supports semantic atomism and the realist the-
ory of meaning, according to which reality has a structure that is fixed and
independent from our systems of representation.
These results, so to speak, pave for me the way to strictly metaphysical questions,
such as: “What is there? What are the objects? What is their nature?” The validity
of these inquiries is based on my conviction that, first of all, Wittgenstein was
interested in such topics and, secondly, he believed that reality has a fixed struc-
ture, i.e. a structure that language mirrors and does not impose on the world. In
the next section, 1.2, I shall discuss the most famous example of the ontological
commitments of Wittgenstein’s views on language, i.e. his conviction that if sense
is determinate, there have to be simple objects. In other words, on the next pages
I shall analyse the Argument for Substance.

1.2  The Argument for Substance

The focus of my interest in the next sections will be the theory that is often called
metaphysical atomism. I shall elucidate what arguments Wittgenstein had in order
to support his views that not only must there be objects in order for propositions
to be meaningful, but also that these objects – the referents of names – have to be

49
simple. I shall present two arguments which we find in support of the existence of
the simples (sections 1.2.1-1.2.2). Then (in section 1.2.3), I will discuss Zalabardo’s
objections, as he argues that one of these arguments (the argument from thinking
what is not) has no textual support in the Tractatus105. In this chapter I assume
that, according to Wittgenstein, if a linguistic expression contains an empty name,
then this expression is meaningless. This is the assumption which is the cause of
Zalabardo’s opposition.
Let us begin our considerations about the ontological commitments of the
Tractatus with the theses called: “Argument for Substance”, or “The substance
argument”106:
2.02 Objects are simple.
2.0201 Every statement about complexes can be resolved into a statement about their
constituents and into the propositions that describe the complexes completely.
2.021 Objects make up the substance of the world. That is why they cannot be composite.
2.0211 If the world had no substance, then whether a proposition had sense would de-
pend on whether another proposition was true.
2.0212 In that case we could not sketch any picture of the world (true or false).

In the Argument for Substance, in thesis TLP 2.0212, there occurs the notion of a
picture:
[If the world had no substance], we could not sketch any picture of the world (true or
false) (TLP 2.0212).

According to the Tractatus’ picture theory of meaning, a proposition has sense

because it is a picture of a state of affairs – there is a parallel between the elements
of a state of affairs and the elements of the respective proposition (TLP 2.13); an
elementary proposition is a nexus of names (TLP 4.22) and names denote ob-
jects (TLP 3.203). Therefore, a proposition has meaning because its names reach
(or touch, using the Tractarian metaphor) objects which constitute a state of
affairs expressed by an elementary proposition (TLP 2.15121; NB 25.9.14, p. 6).
Names form a frame of a proposition. What is within the frame is irrelevant to
the sense of a proposition, but it is clear that for a proposition to be a picture
it has to have a frame. That briefly summarises the so-called picture theory of

105 “The line of reasoning under consideration supports the claim that the Tractatus
advances a version of the Empty-Name Argument with the contention that this is
the best way to interpret the substance passage. This exegetical claim is endorsed by
many interpreters, and no serious alternative has been put forward. My goal is to
argue that this reading is incorrect” (Zalabardo 2012a, p. 130).
106 Morris 2008, p. 355–363.

50
meaning. The following table will help to see that without objects names would
reach or touch nothing and a sentence would not depict a state of affairs.

Table 1107

linguistic expression relation between a proposition state of affairs

and a state of affairs
“ABCD” is true if ABCD
“ABCD” is false if ACBD
“ABCD” is nonsensical if ABC

Wittgenstein discerns between the falsehood of a linguistic expression and its

meaninglessness. A proposition is false when it expresses relations between ex-
isting objects but reflects them wrongly. A completely different situation takes
place when an expression refers to non-existing objects – then it is meaningless.
One can reformulate the Argument for Substance in the following way:
P1 In order for a proposition to be meaningful, there must be objects
P1.1 A meaningful proposition is a concatenation of names
P1.2 A name represents an object
P2 There are meaningful propositions
C There are objects (from P1 and P2)

The excerpt of TLP 2.02-2.0212 is meant to have the structure in which TLP 2.02 is
the thesis and TLP 2.021-2.0212 is the justification for the thesis. We saw that our
final conclusion, which on the grounds of the theory of representation seemed to
be plausible, reads as follows:
C There are objects

But thesis 2.02 is the following:

TLP 2.02: Objects are simple

The conclusion at which we have arrived is more moderate. It is something dif-

ferent to state:
A)  In order for a language to sketch the world there must be objects
than:
B)  In order for a language to sketch the world there must be simple objects

107 I shall defend later, during the discussion on Zalabardo, that this draft is a justified
picture of Wittgenstein’s views.

51
The most intriguing question of the analysed fragment is, then: “Why must ob-
jects be simple?”. Why did Wittgenstein exclude the possibility of the existence
of complex objects (TLP 2.021)? This exclusion is certainly contra-intuitive. If
a claim is contra-intuitive, it does not mean that it is false, but certainly one has
to give strong arguments to support it. In the next two sections I shall analyse
two arguments that Wittgenstein could put forward in order to justify his point
of view.

1.2.1 The argument from the false judgement, or thinking

what is not108
The reason for negating the existence of complex objects was provided by Witt-
genstein in Philosophical Investigations, 39:
The sword Nothung consists of parts combined in a particular way. If they are combined
differently, Nothung does not exist. But it is clear that the sentence “Nothung has a sharp
blade” has a sense, whether Nothung is still whole or has already been shattered. But if
“Nothung” is the name of an object, this object no longer exists when Nothung is shat-
tered into pieces; and as no object would then correspond to the name, it would have no
meaning. But then the sentence “Nothung has a sharp blade” would contain a word that
had no meaning, and hence the sentence would be nonsense. But it does have a sense; so
there must still be something corresponding to the words of which it consists. So the word
“Nothung” must disappear when the sense is analysed and its place be taken by words
which name simples. It will be reasonable to call these words the real names.

Wittgenstein in this fragment supposes that it is possible to have a proposition p

which
1) has sense and
2) contains a complex name (for example: “ABC”) which denotes a complex
object (for instance ABC).
But since abc is complex, it is always possible that it would decompose and cease
to exist (to, for instance, AB and C).

108 C.f. Malcolm 1987, p. 43. Zalabardo calls it “the Empty Name Argument” (Zalabardo
2012, p. 120).

52
TABLE 2

proposition p state of affairs

Before decomposition “ABC DEF” ABC DEF
After decomposition “ABC DEF” AB C DEF

We would have, then, a strange example of proposition p which would be the

nexus of names without a denotation. We have learnt from TABLE 1 that if
names denoted nothing, then p could have not depicted the world and could
have not been a picture of a state of affairs, so p after decomposition of the com-
plex ABC would have been meaningless (it would have ceased to be a proposi-
tion). The meaningfulness of p would have been a contingent fact.
One can accept this conclusion and state that the fact that propositions are
meaningful is contingent, or one could change the premise (2) and say that even
if complex names occur in propositions, they do not denote complex objects but
rather something more primary. Obviously, in the Tractatus Wittgenstein (as he
himself reported in PI, 39 and earlier in the Notebooks) chose the second solution.
Independently of the fact whether the sword Nothung exists or not, the expres-
sion: “Nothung has a sharp blade” is a proposition. In Investigations we read: “It is
clear that the sentence ‘Nothung has a sharp blade’ has a sense, whether Nothung
is still whole or has already been shattered”109, and in the Notebooks: “Obviously
propositions are possible which contain no simple signs”110.
One of the possible views on the proposition in the Tractatus is that it is some-
thing which is either true or false. From this it follows that a linguistic sign which
is neither true nor false is not a proposition. According to Wittgenstein’s take, lin-
guistic expressions about Nothung, a non-existing object, are nevertheless propo-
sitions. This means that they have a truth-value. One of the explanations for this
fact could be a straightforward conclusion that, apparently, in Wittgenstein’s view,
there are possible propositions containing empty names. The second explanation
retains the claim that the meaningfulness of a proposition depends on the fact that
all of its names “touch” reality. It states, however, that, apparently, linguistic expres-
sions containing the name “Nothung” are not about Nothung but about something
else, i.e. something that lasts despite the decomposition of Nothung. As we will
see, Zalabardo thinks that the Tractatus follows the first path of reasoning. I shall
argue that early Wittgenstein chose the latter option, according to which as long as
a denotatum is a complex object (susceptible to decomposition), there is always a

109 PI, 39.

110 NB 9.5.15, p. 46.

53
danger that a proposition will lose its sense, and that is the conclusion Wittgenstein
wants to avoid. In our analysis we should get to objects which are in no danger of
decomposition, i.e. which are indestructible111. These are the simple objects. In my
opinion, Wittgenstein came to the conclusion that the real reference of the name
“Nothung” is not a complex object such as the sword which bears this name but
the particles which Nothung is composed of. They last even though Nothung has
ceased to exist. We could say that the Tractatus in effect enacts the law which al-
lows to put in the right column of TABLE 2 only single letters (A, B, C, etc.).

TABLE 3

proposition p state of affairs

Before decomposition ABC DEF ABCDEF
After decomposition ABC DEF DACFBE

The solution presented in TABLE 3 shows that for the sense of p, decomposi-
tion of ABC is irrelevant. On the other hand, the cost of the solution is that only
simple objects are allowed in the right column by the theory of language. In other
words, entities which truly deserve the title of “object” are those that are insuscep-
tible to destruction, i.e. simple ones.

1.2.2  The argument from the determinateness of sense

The second argument is well summed up by Peter Hacker: “If the world lacked a
substance consisting of objects, then whether a proposition had a sense i.e. was
true or false, would depend upon whether another proposition were true i.e., a
proposition asserting the existence of a complex”112. Two notions are essential
to Wittgenstein’s second argument: the concept of bivalence and the concept of
the determinateness of sense; both are connected to each other. The principle
of bivalence says that every proposition has exactly one truth-value; it is either
true or false. According to TLP 2.21, “a picture agrees with reality or fails to
agree; it is correct or incorrect, true or false”. A proposition is a kind of picture
(TLP 4.01) and it is true or false only “in virtue of being a picture of reality”
(TLP 4.06). Hence, the bivalence principle is held in the Tractatus. In Wittgen-
stein’s system of language there are two kinds of propositions: truth-functions

111 Of course, one should not mistakenly identify a physical decomposition with the
metaphysical one. At least for now I leave the question if being physically indestruct-
ible is a sufficient condition for being a simple object aside.
112 Hacker 1986a, p. 79.

54
on elementary propositions and elementary propositions (which can also be
seen as truth-functions on themselves). The truth or falsehood of truth-func-
tions on elementary propositions depends on the truth-value of the respective
elementary propositions. On what does the meaning of elementary proposi-
tions depend? If the bivalence principle is to be held, then the meaning of an
elementary proposition is determined only by a state of affairs expressed by the
respective elementary proposition. That requirement could be called the re-
quirement of a determinate sense113. The meaning of an elementary proposition
cannot be dependent on other propositions’ meanings (TLP 5.134). If asked if
an elementary proposition “ABCD” is true, one cannot answer: “That depends
on what you mean by ‘A’ or ‘B’ ”? For that to be possible, as Wittgenstein puts it,
“a proposition must restrict reality to two alternatives: yes or no” (TLP 4.023).
What is the hidden condition imposed on reality by the requirement of a de-
terminate sense? As the Tractatus states it, it is the existence of simple objects:
The requirement that simple signs be possible is the requirement that sense be determi-
nate (TLP 3.23).

In other words, Wittgenstein’s demand consists in the idea that the truth-condi-
tions of any elementary proposition would never be determinate if reality were
not discrete114. I shall explain this on an example taken from everyday language.
According to Wittgenstein, all propositions of this language “are in perfect logical
order” (TLP 5.5563). This means that the bivalence principle and the requirement
of a determinate sense also apply to this language115. On the other hand, it is clear
(and it was clear also for Wittgenstein – c.f. TLP 3.323) that everyday language is
full of ambiguities and vagueness. Let us take, for example, the proposition:
(A)  This paint is green
There are, of course, cases when it is clear if we are dealing with green paint or not.
There are also situations when we would hesitate: “Is that shade of colour green or
perhaps yellow?” and we would ask for a definition: “That depends on what you
mean by green?”. That would also be a sign that the proposition (A) did not restrict
reality to two alternatives: yes or no, and the requirement of a determinate sense

113 In this work I shall use the term: “a determinate sense“ synonymously with the term:
“the definite sense”.
114 “The idea would be that the truth-condition of a sentence could never be perfectly
sharp unless reality were discrete” (Carruthers 1990, p. 47).
115 There cannot be any truth-value gaps in everyday language (Hacker 1986b, p. 117).

55
would not be fulfilled116. If we were always in a need of an additional explanation if
a given concept applied to a concrete situation, this would have unacceptable con-
sequences for Wittgenstein’s system. It would mean the impossibility of elementary
propositions, i.e. propositions whose meaning is independent of any other propo-
sitions (definitions, conceptual clarifications, hints as to how to apply concepts,
etc.) and whose truth depends only upon the agreement with reality117. That is why
Wittgenstein identified the requirement of a determinate sense with the existence
of the simples. The picture behind this identification is a fine-grained reality, where
it is ultimately decided where the last shade of yellow is and where the first shade
of green begins. Having once established the border between the colours green and
yellow, one avoids the danger of ambiguity. A fine-grained reality was, for Wittgen-
stein, the necessary condition for linguistic expressions to be sharp. This, in turn,
was the condition for a meaningful language, i.e. a language that is able to express
states of affairs. The idea of a fine-grained reality was developed later, in Some Re-
marks on Logical Form, by the notion of degrees of a quality of a given colour. One
degree of a quality of a colour could be compared with a simple object. This is just
an example, and I do not mean by this that simples are simple units of perception,
but it does show well what stands behind the idea of the determinateness of sense
and why Wittgenstein, in the Argument for Substance, passed from the anteced-
ent: “If the world had no substance” and by that he meant: “If the world did not
consist of simple objects” or “If the reality was not full grained” to the conclusion:
“then whether a proposition had sense would depend on whether another proposi-
tion was true” (TLP 2.0211).
It is quite easy to check if the substance consisting of complex objects would
fulfil the requirement of a determinate sense. A very simple example will show
that it would not. According to TLP 2.0201, every statement about complexes
can be resolved into statements about their constituents. If the name “ABC” de-
notes a complex ABC, then one can analyse the proposition p in which “ABC”
occurs by means of propositions in which there occur names denoting objects
A, B and B. For example, we could spell out the entailments of the proposition:
p: “This chair is wooden”

116 “If the proposition ‘The book is on the table’ has a clear sense, then I must, whatever
is the case, be able to say whether the proposition is true or false” (NB 20.6.15, p. 67).
117 “The truth of a proposition containing a complex expression will depend upon the
truth of the propositions into which it is analysable. But analysis must come to an
end (…) with propositions whose truth depends only upon agreement with reality”
(Hacker 1986b, p. 123).

56
in the following way:
q: “The backrest of the chair is wooden”
r: “The seat of the chair is wooden”
s: “The legs of the chair are wooden”
The truth-value of proposition p depends on the truth-values of propositions q, r
and s. This clearly shows that p is not an elementary proposition. If there were no
simple objects, we could say the same about q, r and s. Their truth-value would
depend on some other propositions. The analysis would never end and we would
never arrive at a proposition whose truth would depend on no other proposition,
but just on its eventual accordance with the respective state of affairs.
The other way to prove that the analysis has to have an end was borrowed by
me from Ernest Sosa118. His proof was put in quite a different context, namely his
considerations on the nature of subjects, but his main argument can still be inter-
preted as an objection to the idea that the world consists only of complex objects.
The greatest difference between us is that he takes the notion of simplicity to be
metaphysical, therefore, he believes that for every object x, when x is material,
then x is metaphysically complex, whereas I take the notion of simplicity to be
physical, therefore I leave the decision as to whether something is complex and
could be reduced and identified with a lower micro-level of reality to the scientists,
i.e. I also apply Wittgenstein’s tactics. Because Sosa excludes that there is such a
thing as a simple, material object, he uses his argument as an argument against
materialism119.
Having in mind this difference, let us take, as an example, a complex object:
Z.Z is in fact a complex of parts, let us say: X1, Y1, and Z1. One could say that
Z is in fact a fiction representing certain other entities: what there truly is are
objects: X1, Y1 and Z1. But if there are no simple objects then even though X1,
Y1 and Z1 are constituents of the complex, they are complexes too, and they are
as fictitious as object Z. From this it would follow that every object is fictional
and represents other entities. The problem is that there would be no entities
which would be represented by some complexes at a higher level of reality and
at the same time which would not be representatives for objects of a lower level.
The hierarchy of levels would be endless and there would be no genuine, non-
fictitious objects. This conclusion is a very high price to pay for an opponent
of the existence of simple objects. The argument shows, above all, that having

118 Sosa 1997.

119 Ibid., p. 74.

57
a choice between two worldviews: the first, assuming that there are only com-
plex objects, and the second, assuming that there are only simple objects; one
should choose the latter option.
To say that in order for a proposition to be meaningful there have to be objects
is quite trivial. After all, a proposition is a linguistic expression which is true or
false; the truth or falsity of a proposition consists in its accordance with a state of
affairs, which is a combination of objects. The demand for the existence of objects
seems to follow from the definitions that the Tractatus accepts. But in the last two
sections I have tried to show that Wittgenstein’s claim is much more substantial –
it says that reality consists of simple objects. If the linguistic theory of the Tractatus
is to work, then one has to accept quite a specific assumption about reality, e.g.
that it is discrete. The Argument for Substance is, hence, a good example of the
ontological commitments of the Tractatus. Moreover, it is also a crucial step in
our inquiries about the worldview of young Wittgenstein. If we have established
that the world is a fine-grained reality of simple objects, then the next question is
quite natural: “what is the example of the simple object?” Before I analyse the pos-
sible answers to this question, I want to respond to the objections with respect to
one of the arguments in favour of the existence of simple objects as raised by José
Zalabardo.

1.2.3  Zalabardo’s objections

Zalabardo contests the hypothesis that the Argument for Substance expresses the
Empty-Name Argument (ENA) or, as Malcolm called it, the argument from think-
ing what is not120 – the line of reasoning which makes use of the consequence that
is inacceptable to Wittgenstein that if objects of reference were destructible, then it
would be possible that propositions were neither true nor false – because for Witt-
genstein propositions are always either true or false, so then the only protection
from the unwanted consequence is establishing that objects of reference cannot be
susceptible to destruction. Zalabardo reconstructs the ENA in the following way:
(1*) “If the value that a name of a proposition actually receives under the proxy
mapping did not exist, then the proposition, with the sense it actually has,
would not receive a truth-value”121. (Premise)

120 “There is no good reason for thinking that the Tractatus puts forward a version of
the Empty Name Argument, and, in particular, that the substance passage should
not be read in this way” (Zalabardo 2012a, p. 120).
121 Ibid., p. 128.

58
where the proxy mapping is “the mapping of names onto constituents of states of
affairs”122. From (1*) follows the following thesis:
(2*) “If the names of a proposition receive contingently existing values under the
proxy mapping, then there will be possible situations that make the proposi-
tion neither true or false”123.
Together with the second premise:
(3*) “If a proposition has sense, then there is no possible situation that would
fail to make the proposition either true or false”124.
they make up an argument for the simples: the names of a proposition have to
receive necessarily existent values.
According to Zalabardo, we have no reason to claim that in the Tractatus Witt-
genstein held claim (1*). The argument itself is valid, but it depends on the validity
of the premise (1*). For many Wittgensteinians this premise is already contained
in the picture theory of meaning. It states that in a proposition the names of the
proposition are coordinated with objects of the respective state of affairs. If the
elements of the proposition are combined in the same way as the elements of a
state of affairs, then the state of affairs is the sense that the proposition in question
expresses. Zalabardo calls this rule the Proxy Principle125, but he is convinced that
the Proxy Principle is neutral with respect to the question whether an expression
with empty names is false or neither true nor false126. He discerns between the gap-
py and the gapless account. According to the gappy account, in a possible situation
if constituents of an actual state of affairs would cease to exist, this situation would
make the proposition expressing the actual state of affairs neither true nor false.
According to the gapless account, in a similar situation the proposition would be
false. The gappy account is, in contrast to the gapless account, in accordance with
(1*). Zalabardo however, points out that in contrast to the gappy account, it is the

122 Ibid., p. 121.

123 Ibid., p. 127.
124 Ibid., p. 126.
125 “Proxy principle: The sense of an elementary proposition p is constituted by the
state of affairs in which the values of the names of p under the proxy mapping are
combined in the same way in which the names are combined in p” (Ibid., p. 122).
126 “The principle has no direct consequences concerning how the pairing of a proposition
with a state of affairs determines the truth conditions of the proposition” (Ibid., p. 128).

59
gapless account that finds confirmation in the texts of the Tractatus127. Above all,
he cites in this context thesis TLP 3.24:
A proposition that mentions a complex will not be nonsensical, if the complex does not
exist, but simply false.

This quotation suggests that Wittgenstein was not troubled by the idea of a name
without a referent. If the Tractatus supported the gapless account, then it could
not have stated what the premise (1*) expresses. If it were true, then we would
have to accept that Wittgenstein in the Tractatus made no use of the Empty-Name
Argument.
One possible solution for Zalabardo’s adversary is to limit TLP 3.24 as refer-
ring only to propositions about complexes, and to claim that TLP 3.24 says noth-
ing about elementary propositions. Then the case whether the gappy account
holds for elementary propositions would be open. But, responds Zalabardo, this
restricted reading of TLP 3.24 is probably motivated only by the desire to save
the gappy account with respect to elementary propositions. No independent rea-
son speaks in favour of the restricted reading of TLP 3.24 – the gapless account
is still the most probable interpretation of Wittgenstein’s views128. The other solu-
tion is to claim that (1*) is proved by the Argument for Substance. But, accord-
ing to Zalabardo, and I agree with him in this respect, it is a helpless solution.
It is rather – and one can notice it also in my interpretation – that the premise
(1*) is the silent assumption of the argument; the thesis which helps us explain
the unobvious dialectics of TLP 2.02-2.0212 rather than the conclusion that was
being proved by the excerpt in question. But do we truly have reasons to see
the hidden premise (1*) behind the formulations of the Argument of Substance,
asks Zalabardo? What is the motive that inclines us to interpret TLP 2.02-2.0212
in accordance with ENA? Again, in the opinion of Zalabardo there is no such
independent reason129. On the contrary, the best way to read this fragment of

127 “It is undeniable that the Tractatus endorses a view that is strongly reminiscent of
the gapless account” (Ibid., p. 128).
128 “Defending this strategy would require providing support for the claim that Witt-
genstein actually endorsed the gappy account, and for the corresponding limitation
on the scope of 3.24. The crucial point here is that the requisite support cannot be
supplied by Wittgenstein’s commitment to the Proxy Principle. The Proxy Principle
is irrelevant to the scope of 3.24 as to the truth of 1*” (Ibid., p. 129).
129 “We have reached a situation in which we have no independent reason for claiming
that the Tractatus endorses the Empty-Name Argument. This claim will be unsup-
ported unless we can establish that the best way to read the substance passage is as

60
the Tractatus, according to his interpretation, is in the context of Wittgenstein’s
polemics with Russell’s theory of judgement.
Let us analyse Zalabardo’s defence. My question is whether TLP 3.24 truly con-
firms the gapless account? It was already answered by David Pears that TLP 3.24
rather states the problem as to why propositions about complexes which do not ex-
ist are false in contrast to propositions about simples which are nonsensical when
they contain empty names. Or, in other words, why do propositions about a com-
plex breach the wall of a theory of representation which assumes the existence of
referents for all names of a proposition? According to Pears, the answer is similar
to what we read in PI, 39. The surface contradiction is the consequence of the fact
that the names of complex things in fact refer to respective simple objects: “The
existence of the complex simply gave ‘a’ [a name of a complex] a short cut to its
sense, which truly depends on the existence of the simple objects mentioned in its
complete analysis”130.
Zalabardo answers that this restricted (in a sense that it is limited only to prop-
ositions about complexes) lecture already assumes what should be proved – the
truth of (1*) and the existence of ENA in the Tractatus. I do not think that Pears’
interpretation assumes the truth of (1*); it merely shows that TLP 3.24 is neutral
with respect to the gappy and the gapless account. Both the adherent of the gap-
less account, like Zalabardo, as well as the adherent of the gappy account, like
Pears, are able to reconcile this thesis with their reading of the Tractatus. In the
case of the interpretation that I defend in this dissertation, there is nothing sur-
prising in the claim that a proposition which mentions a complex that does not
exist is false, and not nonsensical. This is the solution that someone who accepts
the conclusion of ENA, i.e. that only simple objects are the objects of reference,
should expect. For instance, the proposition: “Nothung has a sharp blade” after
decomposition of Nothung is false. The name: “Nothung”, according to the view
which assumes the gappy account, refers to the particles that constitute the sword.
These particles exist even after the decomposition of Nothung. However, they
stand in different relations to each other than previously. The name “Nothung”
after decomposition of the sword still refers to existing objects, but it assumes a
combination of these objects that does not exist. Hence, used in a proposition it
makes it a wrong model of reality – a false proposition.

advancing the Empty-Name Argument. I am going to argue that this is not the case”
(Ibid., p. 132).
130 Pears 1987, p. 78.

61
 At best, TLP 3.24 is neutral with respect to the gappy and the gapless account.
In my opinion, however, the burden of proof is on Zalabardo’s side and not on
Pears’. He should prove that the gapless account is the better reading of TLP 3.24.
It is simply because TLP 3.24 when read at face value refers only to propositions
that mention complexes. The most apparent interpretation is the restricted lecture
of this thesis, which points out that Wittgenstein’s statement (that along with the
disappearance of the referent of a name a proposition which contains the name
does not suddenly become meaningless but is false) applies only to propositions
about complexes. The restricted reading finds its support in the broader context of
the Tractatus: thesis TLP 3.25 that follows directly after TLP 3.24 refers to an anal-
ysis into elementary propositions (“A proposition has one and only one complete
analysis”). Hence, Wittgenstein’s interest in this part of the Tractatus concentrates
around propositions about complexes and the question as to what differentiates
them from elementary propositions (and as Pears’ interpretation states, these dif-
ferences are: I. Propositions about complexes have indeterminate sense, and II. A
proposition about a complex in a situation when the referent of a complex sign
does not exist is false and not nonsensical131). There is no need, in my opinion,
for Pears to additionally back up his limited reading of TLP 3.24 and not take the
formulations of TLP 3.24 at face value. It is rather Zalabardo who should show
that TLP 3.24 also refers to elementary propositions, and that would be difficult
if we take into account the Tractarian theory of representation (which Zalabardo
calls the Proxy Principle). Perhaps this is why he softens the meaning of the Proxy
Principle:
The Proxy Principle specifies how an elementary proposition is paired with the state of
affairs that determines its sense. Hence it has consequences concerning whether a propo-
sition would have, in non-actual situations, the sense it actually has. But the principle has
no direct consequences concerning how the pairing of a proposition with a state of affairs
determines the truth conditions of the proposition (Zalabardo 2012a, p. 128).

If I am right in supposing that the Proxy Principle in Zalabardo’s paper is the coun-
terpart of the Tractarian picture theory of meaning, then one has to notice that the
concept of a proposition as a picture or as a model entails the gappy account that is
contested by Zalabardo. As I have said, the picture can mirror reality only because
of the “feelers” with which a proposition “touches” reality132. Wittgenstein says that
a proposition “is laid against reality like a measure” (TLP 2.1512). This metaphor
makes sense only under the condition that “the end-points of the graduating lines

131 Ibid., p. 76.

132 “That is how a picture is attached to reality; it reaches right out to it” (TLP 2.1511).

62
actually touch reality” (TLP 2.15121), i.e. speaking less metaphorically, only if the
bearers of names exist. Obviously, Zalabardo is aware of the Tractarian conception
of the proposition as a picture that includes not only a combination of signs but
also pairings of these with their bearers; it is the notion of proposition which as-
sumes that a proposition has its sense necessarily133. He admits that this view on
the proposition confirms premise (1*), but, in his opinion, if one accepts this no-
tion of a proposition, then ENA does not work. The premise (1) of ENA is the fol-
lowing: “If the value that a name of a proposition actually receives under the proxy
mapping did not exist, then the proposition would not have the sense that it actu-
ally has”134. The second premise of ENA, according to Zalabardo, states that propo-
sitions cannot have their senses contingently. In Zalabardo’s opinion, if one takes
the notion of proposition on which a proposition has its sense necessarily, then
one should rather formulate the premise (1) as stating that if a bearer of a name
of proposition p did not exist, then there would be no proposition at all135. Under
this reading, however, there is no disagreement or contradiction between premises
(1) and (2), and one does not have to conclude that the objects of reference have
to be simple objects. The paradox is established only when one reads premise (1)
in a way that does not assume the notion of proposition which includes not only
combinations of signs, but also their pairings with their bearers. The problem for
a proponent of the presence of ENA in the Tractatus consists in the fact that if he
decided, in order to save the validity of ENA, to resign from this understanding
of a proposition, then he would lose the independent reason to think that early
Wittgenstein accepted the gappy account.
The dilemma that Zalabardo presents could be summed up in the following
way: if one wants to defend the presence of ENA in the Tractatus, then one has
to show that Wittgenstein accepted the premise (1*) of ENA and, consequently,
the gappy account. One finds the argument in favour of the gappy account in the
Tractarian theory of proposition, but if one accepts the notion of a proposition
as it is in the picture theory of meaning, then ENA is not valid and its conclusion
is not entailed by its premises; there is nothing paradoxical in the premises that
would demand a special solution, such as the requirement that only simple ob-
jects can be referents of names. On the other hand, if one takes a different notion

133 “Some passages suggest a conception of propositions according to which they in-
clude, not only the combinations of signs, but also the pairings of these with their
images under the proxy mapping” (Zalabardo 2012a, p. 124).
134 Ibid., p. 122.
135 “In the counterfactual situations in which the images of the names of p under the
proxy mapping do not exist, p, thus constructed, does not exist either” (Ibid., p. 124).

63
of the proposition (which suggests that a proxy mapping is a result of an arbitrary
decision), then ENA is valid – the conclusion follows from the premises, but there
is no reason to accept the truthfulness of the premise (1*) and, consequently, the
gappy account.
In response to Zalabardo, I would generally answer that Wittgenstein’s later as-
sessment of his own earlier argument should not be so easily disposed136. The piec-
es of evidence taken from the aforementioned PI, 39, but also from PR, 72137, give
us not only the essence of ideas standing behind the Tractatus but also a hint of
how one should solve the dilemma with which Zalabardo is facing the proponents
of ENA. As far as I understand Zalabardo, he complains that there is no contradic-
tion between the following premises:
1) If the value that a name of a proposition actually receives under the proxy
mapping did not exist, then there would be no proposition.
2) Propositions cannot have sense contingently.
Which is a paradox that would demand a special restriction as to what objects
can be the bearers of names. I think that PI, 39 shows what the paradox con-
sists in. The value of a name in the example considered by later Wittgenstein
is Nothung. According to the premise 1) and to the conception of proposition
that follows from the picture theory, if Nothung did not exist, then there would
be no proposition containing the name “Nothung”. The paradox consists in the
fact that these propositions clearly exist! According to the gappy account, as
long as Nothung exists, such a linguistic expression138 as: “Nothung has a sharp
blade” is a proposition, but after decomposition of Nothung the linguistic ex-
pression: “Nothung has a sharp blade” is no longer a proposition. That conclu-
sion seems to be for Wittgenstein unacceptable. Disagreement with exactly this
consequence led him to the postulate of the existence of simple objects. There-
fore, I would formulate the premises of ENA in the following way:

136 Zalabardo claims that “later” Wittgenstein does not present “a line of reasoning” of
early Wittgenstein but a later explanation of an earlier held and unsupported convic-
tion (C.f. Zalabardo 2012a, p. 148).
137 “What I once called ‘objects’, simples, were simply what I could refer to without
running the risk of their possible non-existence; i.e. that for which there is neither
existence or non-existence, and that means: what we can speak about no matter what
may be the case” (PR, p. 72).
138 I use the term: “linguistic expression” as including both meaningful combinations
of signs (propositions) as well as meaningless combinations of signs (nonsense).

64
1) If the value that a name of a proposition actually receives under the proxy
mapping did not exist, then there would be no proposition.
2) Linguistic expressions cannot have sense contingently (this premise excludes the
possibility that a proposition just ceases to exist, it becomes merely nonsense)139.
These two premises do contradict each other. They demand a solution, namely
blocking the possibility of the antecedent of the first premise, which is the possibil-
ity that the value of a name ceases to exist. This is exactly what ENA concludes: the
referents of names should be indestructible, simple objects. Under my proposal,
ENA is not only valid, but the notion of the proposition it uses is such that it gives
us reasons to accept the premise 1).
Finally, I want to argue against Zalabardo’s conclusions, indirectly, by showing
the consequences of the point of view in which the Tractatus favours the gapless
account. Nonsense, according to Conant’s description, consists in a case in which
linguistic signs do not produce any sentence140. One can ask what can go wrong
in constructing propositions? When the linguistic signs fail to produce any sense?
It seems that if we accept Zalabardo’s remarks (that a proposition in which empty
names occur is nevertheless meaningful), then the only thing that can go wrong
with respect to the sense of a proposition is when a linguistic sign is used “other-
wise than in its proper semantic role”141. It seems that Wittgenstein’s examples of
nonsense – “Socrates Plato” or “Socrates is identical”142 – confirm that theory. In
order not to speak nonsensically, it is enough to be careful with the semantic roles
of expressions. For example, “( ) is identical” is a two-argument predicate; it marks
a symmetry relation that holds between objects. In Wittgenstein’s example: “So-
crates is identical”, the role of this particular predicate is violated. “( ) is identical”

139 The other steps of ENA are the following (I reformulate the argument given by:
Zalabardo 2012a, p. 122–123): 3) If a bearer of a name of a proposition was suscep-
tible to decomposition, then it would be possible that a linguistic expression has
sense contingently. 4) (from 2, 3) Therefore names of a proposition cannot refer to
objects susceptible to decomposition. 5) Complexes are susceptible to decomposition
(premise) 6) (from 4,5) Therefore names refer to simple, indestructible objects.
140 “Mere nonsense – a string composed of signs in which no symbol can be perceived,
and which hence has no discernible logical syntax” (Conant 2000, p. 191). One of
the examples of a symbol was, for Wittgenstein, a sentence (TLP 3.31). That is why
I write about nonsense as a collection of signs which does not produce any sentence.
141 McGuinness 2002, p. 89.
142 NM, p. 115; TLP 5.4733.

65
pretends here to look like a relation of a name (Socrates) falling under a concept.
Hence the speaker of this expression does not succeed in producing a thought143.
But if this is all one has to give attention to in order to speak meaningfully, then
why, according to Wittgenstein’s advice, should we restrict ourselves to express-
ing propositions of natural science (TLP 6.53)? Why does only natural science
succeed in producing thoughts? If a sentence is constructed in accordance with
logical syntax, then, for example, a sentence taken from a work of fiction such as:
“Odysseus was set ashore at Ithaca while sound asleep” should be treated as mean-
ingful. Indeed, it made sense to Frege. It did not refer to anything (in Fregean ter-
minology this means that it lacks truth-value) but, according to him, the lack of
truth-value has no impact on the sense of a proposition144. Clearly Wittgenstein,
who for some reason dismissed even perfectly logically constructed sentences of
ethics and aesthetics as senseless, did not adopt the Fregean attitude in this re-
spect. This is why, in my opinion, we have to complement the Tractarian theory of
sense so that it contains, apart from the requirement of being an expression con-
structed in accordance with logical syntax, also the requirement of the existence
of referents of all elements of a proposition. In other words, I do not agree with
the pervasive interpretation of the Tractatus according to which the only source of
nonsense is the usage of expressions not in accordance with their roles in logical
syntax. I emphasise also the (frequently omitted) second source of nonsense, i.e.
usage of empty names in propositions145. One reads about this source of nonsense
in TLP 5.4733. According to this thesis, the fact that a proposition does not have
sense could also be caused by the fact that “wir einigen seiner Bestandteile keine
Bedeutung gegeben haben” (TLP 5.4733). There is no reason why we should not
understand here “Bedeutung” as in the rest of the text, i.e. as the concept that is
synonymous with the concept of reference. In other words, theses like TLP 5.4733

143 Or, as Hacker puts it: “The reason why ‘Socrates is identical’ is nonsense is that we
have given no meaning to ‘identical’ as an adjective” (Hacker 2000, p. 358).
144 “The sentence ‘Odysseus was set ashore at Ithaca while sound asleep’ obviously has
sense. But since it is doubtful whether the name ‘Odysseus’, occurring therein, means
anything, it is also doubtful whether the whole sentence does” (Frege 1892/1984a,
p. 162).
145 In the recent literature also Michael Morris underlines the importance of this second
condition: “Languages depend for their meaningfulness on correlations between
certain linguistic items, on the one hand, and extra-linguistic terms, on the other
(…) The extra linguistic items with which linguistic items have to be correlated for
languages to be meaningful are items in the world (objects)” (Morris 2008, p. 60–61).

66
or TLP 6.53146 indirectly provide us independent reasons to think that the Tracta-
tus holds the gappy account. If there are such reasons, then one should not reject
the premise (1*) of ENA (under Zalabardo’s formulation). Although Zalabardo’s
objections against the presence of the Empty-Name Argument in the Tractatus
are very challenging and demand a closer look at what Wittgenstein truly had to
say in the Argument for Substance (from which one can only profit), they cannot,
however, question the claim that the argument from thinking what is not is one
of the reasons for which Wittgenstein stated that reality consists of simple objects.

Summary
One of the main theses of my interpretation of the Tractatus is that one should
take the Tractarian identification of true propositions with propositions of natu-
ral science seriously. In this chapter I tried to present what the consequences are
of a serious reading of TLP 4.11. It is, for instance, admitting to the fact that early
Wittgenstein was a proponent of ontological materialism. In this chapter I have
attempted to prove the truthfulness of this hypothesis.
In the first part of the chapter I showed that in contrast to popular opinion,
Wittgenstein was interested in metaphysical topics (sections: 1.1.1; 1.1.3). This
interest in the “nature of being” was the consequence of his initial logical inquiries
into the nature of the proposition. According to the picture theory of meaning,
a proposition is a model of the respective state of affairs. This theory introduces
the idea of a comparison between a proposition and reality and, consequently, the
idea of reality (hence the presence in the Tractatus of such notions as the notion
of fact, state of affairs or the substance of the world). Naturally, the claim that the
picture theory of meaning brings reality into the scope of interests of young Witt-
genstein depends on our decision as to how we classify this theory: is it anti-realist
(agnostic to the question as to what objects of reference are in themselves) or real-
ist (assuming that objects are ready-to-be-named bits of reality that are independ-
ent of any system of representation)? Only if one considers the picture theory of
meaning as a realist one is one entitled to say that metaphysical problems were
of early Wittgenstein’s concern. That is why in sections 1.1.2 and 1.1.3 I took up
Ishiguro’s challenge, according to which the slogan: “meaning is use” was present
already in the early writings of Wittgenstein. I hope to show that, in contrast to

146 In the Notebooks, on the other hand, Wittgenstein wrote: “We must be able to con-
struct the simple functions because we must be able to give each sign a meaning”
(NB 15.4.16, p. 71).

67
Ishiguro’s opinion, Wittgenstein in the Tractatus held in fact the compositionality
thesis which supports a realist reading of the picture theory of meaning.
In the second part of the chapter (section 1.2) I tried to show what the ontologi-
cal commitments of the Tractarian theory of proposition were. I wanted to answer
the question as to what it means that Wittgenstein raised ontological topics in the
Tractatus. In my opinion, it means that he realised that in order for his theory of
language to work (a theory that entails that language consists of elementary propo-
sitions that are logically independent from one another and their truth-functions),
reality has to be determined in a certain way; it has to consist of what he calls: “sim-
ple objects”. I also think that because he believed that he had correctly described
the functioning of language, he inferred that the world indeed consists of simple
objects.
In support of my hypothesis I invoked two arguments in favour of the ex-
istence of simple objects (sections 1.2.1 and 1.2.2). The discussion regarding
Zalabardo, who rejects the idea that the Empty-Name Argument is in the Trac-
tatus, brought in effect another ontological commitment of early Wittgenstein’s
theory of meaning, namely the requirement that in order for a linguistic ex-
pression to convey sense, all of its constituents have to refer to existing objects
(section 1.2.3). This means that the requirement that all constituents of a lin-
guistic expression occur in their proper syntactical roles is not enough to state
the meaningfulness of the given linguistic expression. In other words, ontology
enters into Tractarian considerations through the fact that in order to state that
an expression conveys content, one has to reach beyond propositions and look
out into reality.
The intention of this chapter was to explain why Wittgenstein identified all true
propositions with propositions of natural science. My hypothesis includes the fact
that Wittgenstein was an ontological materialist. I defended the thesis according
to which he indeed fought with metaphysical questions. My investigations in this
chapter have resulted, briefly speaking, in the conclusion that, according to the
Tractatus, reality consists of simple objects. If only we could prove that simple ob-
jects are objects of physics, then we could rightfully conclude that all true propo-
sitions ultimately express physical states of affairs, and all true propositions are
propositions of physics. Only these propositions have secured reference (simple
objects of physics are not susceptible to decomposition), so only they fulfil one of
the requirements that was put on linguistic propositions in order for them to con-
vey content – they do not contain empty names. If other propositions, e.g. proposi-
tions of everyday language, have sense, it is only because they are about physical
states of affairs; facts which under a complete analysis of language would be more
precisely expressed by propositions of natural science.

68
 In order to conclude that the hypothesis of ontological materialism and the nat-
uralistic worldview of the Tractatus is true, one has to check if there is any evidence
in Wittgenstein’s early writings that would allow to presume that one can identify
Tractarian simple objects with physical atoms (the most elementary particles of
matter). This topic will be the subject-matter of the next chapter. The current point
of my concern reads as follows: “What are the simple objects of the Tractatus?”.

69
Chapter 2.  The Simple Objects of the Tractatus

The main question in the following chapter reads as follows: is the ontology of the
Tractatus materialistic? Are the simple objects of the Tractatus material particles,
or something else? If we accept a realist theory of meaning, and if we assume that
it is possible to ask about the reference of names outside the context of a proposi-
tion, then there are two positive and probable (considering Wittgenstein’s early
writings) answers to the question: “what is the reference of simple names?” at
our disposal, i.e. that there are simple units of perception (the phenomenalistic
interpretation of Tractarian objects147), or that they are physical particles (the
materialistic interpretation)148. In this chapter I shall present arguments in favour
of and against both of these proposals (sections 2.1 and 2.2). I am also going to
argue that the materialistic interpretation offers stronger arguments in its favour
(sections 2.2.2 and 2.2.3). In section 2.3 I will analyse the resolute interpretation.
I have decided to analyse this way of reading the Tractatus because of the concep-
tion of philosophy that it ascribes to Wittgenstein. In my opinion, the resolute
interpretation mistakenly identifies the target of Tractarian critique of philoso-
phy, thus a discussion on the resolute interpretation gives me an opportunity to
introduce the second hallmark of Tractarian scientism, i.e. Wittgenstein’s claim
that in the task of solving “the problems of life”149, or in the task of defending the
values of the human world, philosophy is of no avail.

147 In Wittgensteinian literature we also find the term: “phenomenological” (Lampert

2000, p. 14; Bradley 1992, p. 69).
148 Obviously, in the literature about the Tractatus one can find other interpretations of
Tractarian simples: For Keyt and Hacker (although the latter author suspects that the
notion of a simple object is incoherent), the simples of the Tractatus resemble Platonic
forms (Keyt 1973, p. 295; Hacker 1986a, p. 81). Copi represents the view on which Trac-
tarian objects are bare particulars: “with the possible exception of Parmenides, I know
of no historical philosopher who discussed absolutely bare particulars” (Copi 1973,
p. 184). Klemke thinks about the simples as of “peculiar metaphysical objects – objects
which can never be apprehended by any experience, but which nevertheless are real
and which form the substance of the world” (Klemke 1971a, p. 117). Morris suspects
that in the notion of a simple object Wittgenstein wanted to comprise the Aristotelian
intuition that to be an object is to be independent in its existence from other things
(Morris 2008, p. 28). All of these conceptions are less represented in the commentaries
about Wittgenstein, and I shall not analyse them in detail.
149 C.f. TLP 6.52–6.521.

71
 Before I start analysing arguments and counterarguments of different interpre-
tations, I owe a clarification as to what it means when one claims that Wittgenstein
was an ontological materialist. When one discusses the physicalistic interpreta-
tion, one should not commit the mistake of imposing notions taken from modern
discussions within the philosophy of mind on an understanding of materialism
during the time when the Tractatus was written. The most significant difference is
the fact that what is nowadays seen as a challenge to physicalism was not seen this
way in the times of early Wittgenstein. For example, Kim writes:
It is perhaps no accident that what many regard as the most important obstacle to physi-
calism is called the problem of “explanatory gap”. The problem (…) is that of explaining
(…) phenomenal consciousness, or qualia, in terms of physical/biological phenomena
(Kim 2008, p. 96).

Kim indicates that physicalism has problems with explaining why conscious-
ness accompanies mental events150. According to Chalmers, the lack of logical
supervenience of consciousness on the physical (and he argues in favour of
that thesis with help from arguments of the logical possibility of zombies, of
the possibility of the inverted spectrum, from the epistemic asymmetry of the
first- and the third-person perspective, from the so-called knowledge argu-
ment and from the absence of analysis) tells us that the explanatory reduction
of consciousness has no chance of success. In other words, according to Chal-
mers, God, having created the physical facts and the natural laws, had to create
something more, namely conscious states connected with the respective physi-
cal laws151. A non-physicalistic approach in the contemporary philosophy of
mind takes, therefore, often the form of defending non-reductive explanations.
It is claimed that as long as there are facts that cannot be explained reductively,
physicalism, as a project of reducing everything to a material basis, fails.
On the other hand, some maintain that an explanatory reduction is something
different than an ontological reduction152. In the intellectual circle of Wittgenstein
it was also not decided if a reductive explanation (understood as a reduction of

150 Chalmers 1996, p. 47; p. 106.

151 Ibid., p. 41.
152 Ontological reduction of A to B, according to Crane, “involves the claim that A=B”,
whereas explanatory reduction is “a relation between theories (…). One theory (…)
gives us an insight into the underlying mechanisms which explain how the entities
of the reduced theory work” (Crane 2001, pp. 54–55). Crane also notices that one
could have just an ontological reduction without the explanatory reduction. It would
entail holding “an identity theory of the entities in question without holding that the
theories of these entities can stand in an explanatory relation” (ibid., p. 55).

72
laws of psychology and biology to physical laws) is possible. In order to accept that
a physicalistic reduction is successful, it was sufficient to argue in favour of the the-
sis that “all psychological statements refer to physical events (viz. physical events
in the body, especially the central nervous system of the person in question)”153.
Wittgenstein and some members of the Vienna Circle could therefore agree with
Chalmers that physicalism does not explain the occurrence of consciousness and
phenomenal states, but the reductive explanation was simply not in the scope of
their interests. Maybe it is true that such a reductive explanation is unachievable,
but materialism, in their eyes, was a successful project, as long as we are not forced
to introduce non-physical objects into our explanations. When we talk about con-
sciousness, perhaps we cannot explain why phenomenal states have to accompany
some states of the brain but, nevertheless, consciousness supervenes on states of
the brain and the neural system and, hence, it is still something of a material nature
that we are talking about. If everything in the range of scientific and philosophical
research is of a physical nature, then materialism, in the eyes of its early 20th-cen-
tury adherents, is fine and intact. If an anti-materialist accepted this response to
the problems of a reductive explanation but still wanted to reject early 20th-century
physicalism, he would have to undermine the conviction that there are only physi-
cal objects.

2.1  Phenomenalistic interpretation of the simples

Up until the 1950s the prevailing answer to the question of the nature of Tractari-
an simples was phenomenalistic. “Most philosophers took it for granted that both
Russell and Wittgenstein were talking about the same sorts of atoms, namely,
sense-data”154. It is also one of the most popular interpretations among scholars

153 Carnap 1931/1995, p. 71. One has to be very careful in ascribing ontological reduc-
tion to Carnap because of his rejection of metaphysical questions as such (the mind-
body problem was for him, for instance, not a question about the relation between
two spheres of reality but a question about the relation between the terms of physics
and psychology). His physicalism consisted in theses that 1) all statements in sci-
ence can be translated into physical language, and 2) physical language is a universal
language, i.e. it can describe every state of affairs. Carnap also claimed that although
every biological and psychological statement refers to a physical state of affairs, it
is not certain that we will find a translation of biological and psychological laws
(which he understood simply as a generalisation derived from singular statements)
into physical law. In other words, he saw the claim that the explanatory reduction is
unachievable as possibly true (Carnap 1931/1195, p. 66–69).
154 Bradley 1992, p. 3.

73
nowadays, e.g. M. & J. Hintikka (1990), J.W. Cook (1994) and P. Frascolla (2007),
to name but a few representatives. The main thesis of the phenomenalistic inter-
pretation is that simple objects of the Tractatus are simple units of experience,
named also sense-data.

2.1.1  The question of Russell’s influence on the Tractatus

Phenomenalism is a metaphysical theory that says that reality consists of sense-da-
ta and that there are no material objects “behind” them. If the phenomenalistic in-
terpretation of the Tractatus is right, and Tractarian simple objects are sense-data,
then there are only sense-data which build up the substance of the world. There
is nothing besides sense-data. Phenomenalism is a consequence of an empirical
epistemology which claims that sense-data are the only entities whose existence
we can be sure of. The rest (i.e. for example, material objects, Cartesian Ego, Kan-
tian things in themselves) is speculation. A phenomenalist is rather interested in
the epistemological question: “what are the foundations of our knowledge?”, than
in the ontological question: “what is there?”155. Among the phenomenalists there is
contention whether epistemological empiricism is compatible with the ontological
belief in the existence of material objects. Locke, for example, thought that there
has to be a cause for the existence of sense-data and postulated the existence of
things independent in their existence from the human mind. Things are the causes
of the existence of sense-data. Could an empiricist be pleased with the Lockean
theory? Not entirely. The tenet that the appearance of sense-data is caused by a
physical thing is not empirical. By definition, it is impossible to be acquainted with
one term of the relation between sense-data and things. One is acquainted only
with sense-data, and external things are always beyond the perceptive capacities of
human beings. Therefore, the Lockean tenet is not empirical – it should be treated
by an empiricist as an unverifiable theory.
One of the trials to resolve this deadlock and to reconcile the common-sense
claim that there are material objects with epistemological empiricism was taken
up by Bertrand Russell, who tried to show that objects postulated by physics could
be exhibited as functions of sense-data156. He believed that hypotheses of physics

155 “Although it is primarily a metaphysical view, phenomenalism is a theory with import

ant epistemological implications and, indeed, it is often accepted by philosophers for
its alleged epistemological advantages” (Fumerton 2010, p. 586).
156 “We may lay down the following definition: Physical things are those series of appear-
ances whose matter obeys the laws of physics. That such series exist is an empirical
fact, which constitutes the verifiability of physics” (Russell 1914a, p. 22).

74
which go beyond sense-data are not in opposition to the empirical language of
sense-data but, on the contrary, they are in concordance: one can interpret the
language of physics in terms of sense-data language: “If your atom is going to serve
purposes in physics, as it undoubtedly does, your atom has got to turn out to be a
series of class of particulars [sense-data]”157. The same procedure goes with objects
postulated by everyday experience; for example, what we call a “table” is not, in
fact, an independent and separate external object but a series of classes of particu-
lars. The particulars – sense data – are the real things, the table – a logical fiction158.
The table presents at each moment a number of different appearances (it has dif-
ferent appearances for different people), so at a given moment all the appearances
make up a class of appearances. In a certain time all the appearances from that
period of time make up a series of classes. When one says that he or she sees a
real thing, as opposed to a phantom, what he or she truly means is that a class of
appearances of an alleged object which is given to him or her at a given moment
connects with the other classes of appearances in such a way that it answers his
or her expectations159. In general, physical things “are those series of appearances
whose matter obeys the laws of physics”160. Russell was trying to show in this way
that propositions and hypotheses of physics could be translated into the phenom-
enalistic language of sense data and that these languages do not contradict one an-
other. With the help of this method, on the one hand, physics could find its basis in
the sense-data language and, on the other hand, one could save the requirements
of the empiricist epistemology (often accused of being unrealistic for the ordinary
practice of scientific research).
For Russell, phenomenalism was also a part of his philosophical position
which he called: “logical atomism”. Just like Wittgenstein, Russell saw in the
logical analysis of language the most important, or maybe even the only, task
of philosophy. Many philosophical crazy, irrational theories and alleged para-
doxes are, according to Russell, the result of misunderstanding and confusion
about the morals one should draw from semantics or about the functioning of
the syntax of language. The most typical example of linguistic confusion, says
Russell, is supposing that every noun stands for an object. In the case of abstract
nouns this often leads to the acceptance of the existence of universals, such as

157 Russell 1918, p. 274.

158 “The things that we call real, like tables and chairs, are systems, series of classes of
particulars, and the particulars are the real things, the particulars being sense-data
when they happen to be given to you” (ibid., p. 274).
159 Ibid., p. 275.
160 Russell 1914a, p. 22.

75
triangularity or rationality161. A misunderstanding of the syntax is displayed
when a philosopher thinks that all propositions in a language are reducible to
the most basic form of a proposition, which is a subject and a predicate. This
form, simply a feature of Indo-European languages, has, according to Russell,
an enormously bad influence on philosophy, especially when a philosopher
who has succumbed to the impression of the reducibility of all forms of propo-
sitions to the subject-predicate form, thinks that this form reveals something
essential about reality. Generally, it can be said that this bad influence is ex-
pressed by the unexpectedly great popularity of ontological monism162. Famous
thinkers such as Leibniz, Spinoza and Hegel were overcome by the idea of a
basic form of a proposition and the metaphysical conclusions one should draw
from it. In the time of Russell’s philosophical activity, ontological monism was
confessed by Hegel’s followers from Great Britain (e.g. F.H. Bradley, E. Caird,
T.H. Green). They maintained that although common sense tells us that there
are many different and independent from one another things, in fact there is
only one object – Reality (the Absolute), and the rest are just mere appearances.
According to Russell, this statement is a result of the completely false Hegelian
logic. It states that all propositions are of the form: a subject plus a predicate.
A predicate, according to this theory, stands for a quality. Secondly, according
to Hegel’s doctrine, in order to understand a thing, one has to have knowledge
about all the qualities of that thing. Moreover, an adherent of Hegelian logic
also interprets relational propositions as propositions about objects and their
qualities. A disciple of Hegelian logic sees all relations of an object as its quali-
ties. Because from a certain point of view one thing is ultimately, directly or
indirectly connected (stands in relation) with the whole of the universe, he
or she jumps to the conclusion that in order to know a particular, one has to
know everything163. In the order of cognition Reality is first, and then follow
its appearances which in everyday language are called things. The direction of
Russell’s thought is opposite. If one wants to understand the totality of things,
one has to understand its constituents in the first place. First, one has to be
acquainted with particulars. The role of a philosophical analysis is to reveal

161 “I do not wish to maintain that there are no universals, but certainly there are many
abstracts words which do not stand for single universals –e.g. triangularity and ra-
tionality. In these respects language misleads us both by its vocabulary and by its
syntax” (Russell 1924/1971, p. 331).
162 Ibid., p. 331.
163 “There is, as you know, a logical theory according to which, if you really understand
any one thing, you would understand everything” (Russell 1918, p. 204).

76
what the simplest objects are with which one is acquainted164. Fundamental to
Russell’s atomism was, therefore, the defence of the common-sense belief that
there are many separate things165.
The answer to the ontological question: “What are the simples?” in Russell’s phi-
losophy boils down to the epistemological question: “What are the simples with
which one is acquainted?”. The logical analysis helps in the philosophical consid-
erations in such a way that it reveals proper names as logical atoms166. A proper
name names a simple object – which is trivial, but Russell also asked: “Under what
condition can a speaker name something?”, and he answered: “Only when he is
acquainted with it”. This claim has peculiar consequences: since nobody in the
world is at the moment, for example, acquainted with Socrates, then nobody can
name him, and therefore “Socrates” is not a name. In Russell’s theory this is a de-
scription167. The only names in the logical sense are, for Russell, expressions such
as: “this” and “that”. In Russell’s view, these expressions stand for sense-data. Ac-
cording to him, we do not have a direct acquaintance with objects postulated by
physics, such as electrons, or suggested by everyday experience, such as a piece of
chalk. What we truly perceive are the appearances of the aforementioned things –
such things as “little patches of colour or sounds, taste, smells, etc.”168. They are the
simples – the bricks of which the world prima facie consists of.

2.1.2  Arguments in favour of the phenomenalistic interpretation

The considerations above about Russell’s philosophy provides the first and one
of the most important arguments in favour of the phenomenalistic interpreta-
tion of the Tractatus. It reads as follows: Wittgenstein started his philosophical
work in the time of Moore’s and Russell’s fervent discussions on idealism. Russell

164 That explains the label: “atomism” for his theory. He calls it “logical atomism” because
the simplest objects with which one is acquainted with are the last residue in logical
analysis – they are logical atoms (ibid., p. 179).
165 “The logic which I shall advocate is atomistic, as opposed to the monistic logic of the
people who more or less follow Hegel. When I say that my logic is atomistic, I mean
that I share the common-sense belief that there are many separate things” (ibid., p. 178).
166 Ibid., p. 200.
167 Ibid., p. 201. Diamond showed that Russell was convinced that it is because of the
theory of description that one can answer to the idealistic challenge. Using Diamond’s
example, I do not have access to the object which Bismarck denotes with the pronoun
“I”, nevertheless I can know Bismarck by means of the propositions that describe him
(Diamond 2000b, p. 265).
168 Russell 1914a, p. 5.

77
challenged idealism by advocating in favour of ontological pluralism. At the core
of his argumentation stood logical atomism, which assumes the phenomenal-
istic ontology with simple units of experience as simple objects. Wittgenstein,
as a disciple of Russell and one of the protagonists of the refutation of idealism,
simply adopted his master’s view in the Tractatus. Phenomenalism was the back-
ground for all logical and semantic solutions of early Wittgenstein.
There are many similarities between Wittgenstein’s and Russell’s atomism169. Ac-
cording to both:
–– the world consists of facts (TLP 1.1; Russell 1918, p.  183), which are com-
plexes of simple objects (TLP 2.01; Russell 1918, p. 189);
–– objects subsist in the way that Aristotelian substances subsist (TLP 2.021,
2.024; Russell 1918, p. 202);
–– the existence of simple objects is established by logical considerations, but
establishing what exactly simple objects are belongs to empirical inquiries;
–– the meaning of a name is an object for which it stands for (TLP 3.203; Russell
1918, p. 187), the meaning of a proposition is a fact, but propositions are not
names for facts. It is impossible to name a fact – it could be merely expressed
(TLP 3.221; Russell 1918, p. 186–187);
–– the meaning of a proposition is a result of the meanings of its elements (the
compositionality thesis) (TLP 4.024; Russell 1918, p. 193);
–– the truth or falsity of a proposition depends on obtaining or not the facts that
are expressed by propositions (TLP 4.062; Russell 1918, p. 182);
–– molecular propositions are truth-functions on atomic propositions (TLP 5;
Russell 1918, p. 210);
–– it is meaningless to say that an object exists (TLP 4.1274; Russell 1918, p. 232–
233).
The adherents of the phenomanalistic interpretation claim that if Russell influ-
enced the author of the Tractatus in so many ways, then he probably also influ-
enced him with respect to the view that simple objects are sense-data170.
J.W. Cook says that it is better to describe the philosophical inspirations of
Wittgenstein’s atomism as being indebted to the neutral monism professed by
William James and Ernst Mach171. Neutral monism can be summarised as the
view according to which:

169 I use here the list of similarities as prepared by Raymond Bradley (Bradley 1992,
pp. 6–9).
170 Hintikka 1990, p. 92.
171 Cook 1994, p. 14.

78
• the world consists only of our sensations
• sensations are not mental or “in” mind
• there is nothing that is subjective
• the material constituting the mental is the same as the material constituting
the physical172.
With respect to the topic of my interest, it is irrelevant if Wittgenstein’s atom-
ism bears the stamp of neutral monism or of Russell’s logical atomism. I reject
both of these proposals173. In my view, both theories acknowledge that simple
objects are sense-data, that sense-data are the basic bricks of the world and that
things of everyday experience are nothing other than classes of sense-data. I think
that Cook misunderstood Russell by imputing to him that he held the notion
of sense-data according to which sense-data are something mental, private and
incommunicable. Russell stressed very clearly that sense-data are not mental174.
The mind does not produce sense-data; it merely adds awareness to sense-data.
He himself did not see his own ontology of logical atomism in opposition to the
ontology of neutral monism. On the contrary, in The Philosophy of Logical Atom-
ism he wrote: “I feel more and more inclined to think that [neutral monism] may
be true”175, and in The Relation of Sense-Data to Physics: “What I have to say in
the present paper is compatible with their [Mach and James] doctrine and might
have been reached from their standpoint”176. Therefore, independently of whose
debtor Wittgenstein truly was, i.e. Russell’s or Mach’s and James’, the first argu-
ment in favour of the phenomenalistic interpretation reads as follows: Tractarian
objects are sense-data, because Wittgenstein inherited his ontology from his great

172 Ibid., p. 8.

173 I do not want to say that Cook’s interpretation has no point of originality. I intend to
stress, however, that it is not different from the classic phenomenalistic interpreta-
tion with respect to the interpretation of simple objects of the Tractatus. It may be,
however, though I do not want to ponder that question here whether Cook offers a
better explanation for some theses of the Tractatus (TLP 5.542, TLP 5.64) than the
phenomenalistic sense-data theory.
174 Russell 1914a, p. 7.
175 Russell 1918, p. 279.
176 Russell 1914a, p. 8. Of course, Russell did not entirely agree with the neutral mon-
ists. His greatest concern was the theory of beliefs and desires – the paradigmatic
examples of mental phenomena. Usually, neutral monists reduce such phenomena
as beliefs to bodily behaviour, and behaviourism implicated by neutral monism was
for Russell a difficulty that hindered him from full endorsement of neutral monism
(Russell 1918, p. 221–222).

79
predecessors. He did not mention explicitly sense-data as simple objects because
it was obvious what kind of metaphysics prevailed in his intellectual environment.
Obviously, this is not the only argument in favour of the phenomenalistic inter-
pretation. Evidence from Wittgenstein’s texts can be grouped into three categories:
(a) Although sense-data are not mentioned explicitly in the Tractatus, there are
many hints in this work – claim the adherents of this interpretation – that sense-
data are the only possible candidates for that label. (b) Apart from that, there are
some more direct indicators in works which were never meant to be published, as
in the Notebooks. (c) Finally, there are some strong arguments in later works, e.g.
lectures and discussions (mostly from the 1930s), in which Wittgenstein inter-
preted the Tractatus himself.
Ad (a) The passage TLP 5.62-5.63 contains famous remarks on solipsism:
For what the solipsist means is quite correct; only it cannot be said, but makes itself mani-
fest. The world is my world: this is manifest in the fact that the limits of language (of that
language which alone I understand) mean the limits of my world. The world and life are
one. I am my world. (The microcosm.) (TLP 5.62-5.63).

If we keep in mind that the simples make up the substance of reality, and if simul-
taneously we want to make sense of the above-mentioned remarks, i.e. that life
and the world are one, or that the world is my world, we have to assume – claim
the proponents of the phenomenalistic interpretation – that the world consists of
my experiences (which also build my life). If the world consists, on the one hand,
of simple objects and, on the other hand, of experiences, then it is highly probable
that we should identify the simples with simple units of experience. One can read
TLP 6.431 (“So too at death the world does not alter, but comes to an end”) in a
similar way. Death means the end for possibilities of experiencing. If the end of
the flow of experiences means the end of the world, that entails that Wittgenstein
identifies the world with experiences. Some commentators, therefore, conclude
that the solipsism of the Tractatus supports the phenomenalistic interpretation
of the simples177.
With the help of the phenomenalistic interpretation it is also easier to under-
stand why, according to Wittgenstein, solipsism coincides with realism. As we
remember, for Russell sense-data were the basic bricks of the universe and the
physical things were the logical constructs – series of classes of appearances. It
seems that Wittgenstein shared this particular view in the Tractatus178. He could

177 For instance, Hintikka 1990, p. 86–89.

178 In favour of such a thesis, according to Hintikka (Hintikka 1990, p. 86), speaks
TLP 5.552: “The ‘experience’ that we need in order to understand logic is not that

80
sum up the contention between realists and solipsists in the following way: real-
ists were wrong in their claims that there are objects outside one’s experiences
(it is even nonsensical to ask if there are any objects beyond experience since
experiences make up the world), but they were right in their claims that the ex-
istence of objects is independent of the existence of a subject (for Wittgenstein
there was no such thing as a subject of experience), and here is the point where
solipsism and realism coincide. There is no difference between the realists and
solipsists since the only thinkable difference (i.e. that the realists believe in the
external world) is not even expressible in language.
Ad (b). It is certain that Wittgenstein was not clear and straightforward in the
Tractatus with respect to the metaphysical category of simple objects. He never
gave any example of the simples there. Fortunately, he was more direct and specific
in the entries of his Notebooks, where he tried different solutions to problems and
often illustrated them with examples. May and June of 1915 were laden with en-
tries of interest to us:
But how am I imagining the simple? (…) As examples of the simples I always think of
points of the visual field (just as parts of the visual field always come before my mind as
typical composite objects (NB 6.5.1915, p. 45)
We single out a part of our visual field, for example, and we see that it is always complex,
that any part of it still complex but is already simpler (NB 24.5.1915, p. 50)
It seems to me perfectly possible that patches in our visual field are simple objects, in that
we do not perceive any single point of a patch separately; the visual appearances of stars
seem certainly to be so (NB 18.6.1915, p. 64).

The examples are clear: during the process of creating the Tractatus, Wittgen-
stein if not accepted then at least seriously considered the candidacy of such
things like points of the visual sense (sense-data) for the simples.
Ad (c). The other group of testimonies is formed of Wittgenstein’s retrospection.
The most popular example of Wittgenstein’s flashback comes from his lectures
in the early 1930s. One of Wittgenstein’s students, Desmond Lee, asked directly
about Tractarian simple objects and got the following answer: “Objects, etc. is here
used for such things as a colour, a point in visual space, etc.”179. The other excerpt
comes from Philosophical Remarks, where Wittgenstein admitted that: “I do not
now have phenomenological language, or ‘primary language’ as I used to call it, in

something or other is the state of things, but that something is: that however is not
experience”.
179 Lectures 1930–1932, p. 120.

81
my mind as a goal. I no longer hold it to be necessary“180. The latter quote is less
convincing and it cannot be used as direct evidence because we do not know what
period of interest in the phenomenological language Wittgenstein was mention-
ing: did he have in mind the time of writing the Tractatus or his conversations with
Friedrich Waismann or Moritz Schlick in Vienna in the late 1920s?

2.1.3 Counterarguments
This set of arguments sounds quite solid. Can we then assume that the simples
of the Tractatus are sense-data? The case is not so simple. Let us keep a close eye
on the first argument. It is truly difficult to decide who had influence on whom.
Hintikkas and Cook make their judgement too easily by assuming that the influ-
ence was not reciprocal. Obviously, Wittgenstein as a young student, although an
astonishingly independent thinker for his age, took advantage of Russell’s or Frege’s
philosophy, but he was very critical of his mentors, too181. Cook claims that Witt-
genstein’s philosophy is rooted in Mach’s neutral monism182. He seems to forget
what Wittgenstein truly thought of Mach’s works (he shared his view with Russell
in one of his letters to him):
I was very interested to hear your views about matter, although I cannot imagine your
way working from sense-data forward. Mach writes such a horrid style that it makes me
nearly sick to read him (Letters, p. 20).

This does not sound like an act of admiration; and it challenges, by the way, the
aforementioned arguments that only the phenomenalistic interpretation makes
sense of Tractarian solipsism. The basis for this argument was the conviction that
Wittgenstein regarded, just like Russell, physical things as functions of sense-data.
As we can see, at least in 1913 (when he wrote the letter quoted above) he did not.
For the sake of the argument the adherents of the phenomenalistic interpreta-
tion focus only on Russell’s works prior to the Tractatus, i.e. mainly on: The Relation
of Sense-Data to Physics (1914) and The Philosophy of Logical Atomism (1918). But
what about works written after publishing the Tractatus (1921)? In 1924 Russell
wrote a paper titled Logical Atomism, in which he clearly changed his views with
respect to the simples. He abandoned his conviction that one has to be acquainted
with simple objects (“When I speak of ‘simples’ I ought to explain that I am speak-
ing of something not experienced as such, but known only inferentially as the limit

180 PR 1, p. 51.
181 Expressions such as: “In contrast to Russell” (NB 30.10.14, p. 21) are nothing unusual
in his works.
182 Cook 1994, p. 14–28.

82
of analysis”183). Philosophy, according to Russell, should build its theories on sci-
ence. And what is important is that he had in mind physics, not psychology184; con-
structions of psychology are for him, as compared to physics, “purely provision”185.
Sense-data are not bricks out of which the world is built186; they are replaced by
something which Russell called “event-particles”187. And all of these views are pre-
ceded by the remark:
I am much indebted to my friend Wittgenstein in this matter. See his Tractatus Logico-
Philosophicus. I do not accept all his doctrines, but my debt to him will be obvious to
those who read his book (Russell 1924, p. 333).

We have two narrations then: according to the first, Russell’s logical atomism in-
fluenced Wittgenstein’s early work, and we infer from that that Tractarian simples
are sense-data. But, on the other hand, we could say that the work of early Witt-
genstein influenced Russell in such a way that he ceased to believe that sense-data
are the simple objects. I am not going to advocate on behalf of the second narra-
tion. My aim is simply to show that the first argument in favour of the phenom-
enalistic interpretation is too weak, and we have too little data to decide without
a doubt what the actual directions of influence were between the two famous
philosophers. Certainly we are not entitled to judge that in the Tractatus Wittgen-
stein was indebted to early Russellian ontology.
What about the second argument? It surveys three groups of texts, i.e. hints
in the Tractatus, direct indications in the Notebooks and Wittgenstein’s memo-
ries. With respect to the third group I have already said that we will never be
sure if his statements from the 1930s are not distorted by the topic of his inter-
est from that time, which was the possibility of a phenomenal language188. In
almost every book devoted to the problem of early Wittgenstein’s metaphysics
we find a record of his conversation with Norman Malcolm, in which he admit-
ted that at the time of writing the Tractatus he thought of himself as a logician,

183 Russell 1924, p. 337.

184 “Psychology is scientifically much less perfected than physics” (ibid., p. 330).
185 Ibid., p. 330.
186 “Such pure empiricism, exemplified by Hume [and early Russell – one would like to
add], leads straight to skepticism rather than to support of scientific doctrines, as it
supposed to lead” (ibid., p. 323).
187 Ibid., p. 329.
188 In fact, Anthony Kenny was convinced that Wittgenstein misrepresented his own
thoughts in his own later statements about the Tractatus (Kenny 1986, p. 67).

83
so deciding what a simple thing is was for him “purely empirical matter”189. This
is a different statement than in his response to Lee, because the natural sciences
refer to physical objects and not to units of perception. Perhaps what neutralises
the previously mentioned excerpt from the dialogue between Wittgenstein and
Desmond Lee most effectively is the testimony about Tractarian simples taken
from Philosophical Investigations. In his later work Wittgenstein often criticised
his former ideas, including the idea of simple objects. He asked, among others,
“what lies behind the idea that names truly signify simples?” (PI 46). After a
longer quotation from Plato’s Theaetetus, where Socrates considered the doc-
trine of primary elements of reality, he recalls:
Both Russell’s “individuals” and my “objects” (Tractatus Logico-Philosophicus) were like-
wise such primary elements (PI 46).

This means that he contrasted the simples of the Tractatus with what was thought
by Russell to be the primary elements of reality. In the later phase of his philo-
sophical work he threw doubt on the meaningfulness of the concept of the ab-
solute simple object, but nevertheless he discerned the Tractarian simples from
simple units of experience. Later, Wittgenstein admitted that he committed a
mistake with respect to the simples, but the mistake did not consist in the accept-
ance of the phenomenalistic ontology.
As far as we are concerned with the direct indications from the Notebooks,
one has to put attention to the character of this work. Wittgenstein never meant
to publish his notes, i.e. it was rather a draft in which he tried different posi-
tions and solutions. He changed his opinions a number of times, as statements
written in the Notebooks occur reformulated in the Tractaus, so, for instance,
we read in the Notebooks:
The simple thing for us IS: the simplest thing that we are acquainted with (NB 11.5.1915,
p. 47).

But only 12 days later:

But it also seems certain that we do not infer the existence of simple objects from the exist-
ence of particular simple objects, but rather know them – by description, as it were – as the
end-product of analysis, by means of a process that leads to them (NB 23.5.1915, p. 50).

189 “I asked Wittgenstein whether, when he wrote the Tractatus, he had ever decided
upon anything as an example of a ‘simple object’. His reply was that at that time his
thought had been that he was a logician; and that it was not his business, as a logician,
to try to decide whether this thing or that was a simple thing or a complex thing, that
being purely empirical matter” (Malcolm 1984, p. 70).

84
The change is great. First, Wittgenstein acknowledges that we know the simples
by acquaintance and this is a confirmation of the phenomenalistic interpreta-
tion, but then he claims that it is by a description that we identify the simples
(and that is problematic in the light of this interpretation). Apart from the frag-
ments quoted by the adherents of the phenomenalistic interpretation, one can
also find such records as the following:
The division of the body into material points, as we have it in physics, is nothing more
than analysis into simple components (NB 20.6.1915, p. 67).

It seems that Wittgenstein allowed (at least in this fragment) for a physical analy-
sis of complex things into simples to be an example of logical analysis. It does
not prove that Wittgenstein contradicted himself, but it does prove that the Note-
books are just a draft and that one should be very careful with inferring strong
conclusions on its basis.
We are left with hints of solipsism in the Tractatus. This is probably the strongest
argument in favour of the phenomenalistic interpretation. I will defer the task of
arguing against the way in which the phenomenalistic interpretation understands
the so-called solipsistic theses of the Tractatus until Chapter 5, but I think that
in order to refute this interpretation it is enough to point to the characteristics of
simple objects in the Tractatus. For example, Wittgenstein attributes to the simples
the quality of timelessness. Objects are “unalterable and subsistent” (TLP 2.027).
The eternality of sense-data is a very strange conception; the attribution of time-
lessness rather eliminates sense-data as candidates to Tractarian simples. Then
there is the attribution of colourlessness to the simples (TLP 2.023). Patches of
colours are phenomenalistic paradigms of simple objects190. Instead of this, Witt-
genstein thought that being coloured is a form of an object (TLP 2.0251). His
characteristics are a great challenge to the phenomenalistic interpretation191. The
proponents of the phenomenalistic interpretation sometimes defend their views
by claiming that the Tractatus presents a philosophical journey: the Tractatus be-
gins with the characteristics of the simples that rather excludes a phenomenalistic
reading, but it ends with stating the truthfulness of solipsism. It was not the point

190 Locke 1690/1999, p. 104.

191 “If Hintikkas had paused to reflect on 2.0232: ‘Roughly speaking: Objects are colour-
less’ they would surely have seen that this is appropriate not to entities that are or
could be perceptible, but to objects of physics (…). Following the Hertz/Wittgenstein
line we would be obliged to drop even that connection to what is perceptible. 2.0232
alone should suffice to show that the Tractatus is not in any way phenomenological
nor are its simple objects among those known by acquaintance” (Harré 2001, p. 224).

85
of Wittgenstein’s book to present a consistent system of claims but rather to con-
front us with a certain path of thinking. The particular stages are not important – it
is the path itself which is important, and the goal towards which this path leads.
According to this reading, the early theses of the Tractatus about simple objects are
cancelled by the later section about solipsism. I would answer, however, that even
if we accept this reading of the Tractatus, then one has to notice that the Tractatus
ends not with asserting solipsism but with the claim that the correct method of
philosophy is to say nothing more than is said in natural science (TLP 6.53). Sci-
entists would be very astonished if they heard that the ultimate subject-matter of
their inquiries are sense-data. In order to save the phenomenalistic interpretation,
its proponent would have to clarify his claim in a Russelian way: the natural sci-
ences deal with physical objects, but since physical objects are nothing more than
series of classes of sense-data, then, in fact, the natural sciences deal with sense-
data. A philosopher would then tell a scientist what the real object of his interest
is. This is intelligible for Russell, who saw the task of philosophy in criticising and
clarifying notions that are fundamental to science192, and he found nothing special
or shocking in philosophy giving new meaning to the terms of science193. It seems
to me, however, that such an intervention of philosophy in natural science is radi-
cally contradictory to Wittgenstein’s intentions, as he also liked to see philosophy
as an activity of clarifying propositions194, but at the same time he wanted to sepa-
rate philosophy from science:
Philosophy is not one of the natural sciences. (The word ‘philosophy’ must mean some-
thing whose place is above or below the natural sciences, not beside them) (TLP 4.111).

According to him, philosophy has nothing to do with science, it should then not
decide what the subject-matter of scientific research is. Therefore, even if we ac-
cept this way of reading the Tractatus, under which the latter theses of the book
invalidate the earlier claims, the final conclusion of the Tractatus does not sup-
port the phenomenalitic interpretation.
At last, I want to mention Anscombe’s famous argument that “there is hardly
any epistemology in the Tractatus”195. As we remember, Russell’s logical atomism
responds to the challenge which reads as follows: “How do you know that there
is an external world; that there are independent objects; that there is something
behind sense-data, if the basis of your knowledge are merely sensations?” Logical

192 Russell 1924, p. 341.

193 Ibid., p. 339.
194 TLP 4.112.
195 Anscombe 1965, p. 27.

86
atomism showed how the objects of physics are constructed from the simplest
units of experience. The whole project was thought of as a defence of scientific
knowledge from a sceptic’s attacks; its aim was to indicate the foundations of our
knowledge. Russell himself admitted that “the special importance of sense-data is
in relation to epistemology, not to metaphysics”196, and that “if we could construct
an impersonal metaphysics, independent of the accidents of our knowledge and
ignorance, the privileged position of the actual data would probably disappear”197.
Anscombe argues that there is neither trace of fight against scepticism in the
Tractatus nor of any other traditional epistemological problems198; there are no
remarks on acquaintance with objects or on sensible verification. Moreover, adds
Anscombe, it seems that in TLP 4.1121199, Wittgenstein explicitly expressed the
intention of avoiding the theory of knowledge “by cutting it dead and by concen-
trating on logic”200. On the other hand, the phenomenalistic interpretation is sus-
tainable under the condition that the Tractatus is concerned with epistemological
problems201. Since this is not the case, we have no reason to presume that Tractar-
ian simples are sense-data.
In the last section I argued against the phenomenalistic interpretation of sim-
ple objects. The arguments it presents to support its principles do not hold. Both
the characteristics of the simples in the Tractatus as well as Wittgenstein’s later un-
derstanding of them (expressed in Philosophical Investigations) contradict the the-
sis that sense-data are the primary elements of reality. Moreover, the assumptions
one has to accept in order to advocate in favour of the phenomenalistic interpre-

196 Russell 1914a, p. 7.

197 Ibid., p. 6.
198 Wittgenstein simply dismissed scepticism: “Scepticism is not irrefutable, but obvi-
ously nonsensical, when it tries to raise doubts where no questions can be asked”
(TLP 6.51).
199 “Psychology is no more closely related to philosophy than any other natural science.
Theory of knowledge is the philosophy of psychology. Does not my study of sign-
language correspond to the study of thought-processes, which philosophers used to
consider so essential to the philosophy of logic? Only in most cases they got entangled
in unessential psychological investigations, and with my method too there is an analo-
gous risk”. (TLP 4.1121).
200 Anscombe 1965, p. 152.
201 It is also admitted by the proponents of the phenomenalistic interpretation such as John
Cook, who writes: “Better sense can be made of the Tractatus if we recognize that its
author was very much concerned with epistemology, especially skepticism, and that his
linguistic doctrines were intended to subverse his epistemological convictions” (Cook
1994, p. xv).

87
tation (the most fundamental questions in philosophy are epistemological ones;
philosophy reveals the real subject-matter of scientific research) are incompatible
with what the Tractatus says about the relation between philosophy and science,
and with early Wittgenstein’s interests in the area of philosophy.

2.2  Materialistic interpretation of the simples

In the next sections I shall present a view according to which although early Witt-
genstein did not explicitly give an example of simple objects, he nevertheless had
an idea what these objects had to be. In the last section I rejected the interpretation
which says that these objects are simple units of experience. My own answer is that
Tractarian simples are of a physical nature. They are physical atoms in the sense of
the most basic and indivisible units of matter. This answer explains why there is no
example of a simple object in the Tractatus. The reason is exactly the same as the
one expressed during Wittgenstein’s conversation with Norman Malcolm: it is the
job of the physicists to determine what the primary element of matter is. Perhaps
scientists are close to the definite answer, but it is not a problem a philosopher can
assess or judge. Secondly, the answer to the question of the simples given by the
materialistic interpretation has the advantage that it is the only one that provides a
clear answer to the question as to why early Wittgenstein identified the totality of
true propositions with propositions of natural science. The only true sentences are
sentences of natural science, because reality consists only of physical atoms (other
“objects” are complexes of these simple elements), and meaningful sentences con-
tain names which refer only to existing objects. Somebody who wants to know the
truth about the world should ask not a logician, a philosopher, a poet, an ethicist
or a priest about it, but a physicist.
In recent times a number of new approaches has emerged to the Tractatus.
The most famous interpretation is labelled “the resolute interpretation”, “the New
Wittgenstein interpretation” or the “American school of interpretation”202. Apart
from it we are witnessing the growth of “another new”, as Alfred Nordmann put
it, way of reading the Tractatus. In this attempt one underlines the engineering
and scientific background of Wittgenstein’s education. Nordmann discerns three
ways of showing how this background shaped the Tractatus203. The first emphasises

202 I am going to challenge the theses of this interpretation later in the chapter (section
2.3).
203 A detailed elaboration of all three of these interpretations in: Nordmann 2002,
pp. 358–382.

88
that Wittgenstein received an engineer’s education with a thorough knowledge of
physics. Kelly Hamilton, for example, recognises in Wittgenstein’s conception of
the relation between language and reality the mind of an engineer. She traces, for
example, the beginnings of the Tractarian theory of unalterable things and change-
able states of affairs in the education of designer engineers. In order to design a ma-
chine (a counterpart of the Tractarian state of affairs), an engineer had to know the
alphabet of elements of that machine (the counterparts of Tractarian objects), and
had to have the ability of imagining how those elements can combine with one an-
other. An engineer when designing a machine had to use a large amount of visual
thinking based on his or her knowledge of the basic elements204. This engineering
practice gives us insight into how Wittgenstein understood his dictum that we use
a proposition “as a projection of a possible situation. The method of projection is
to think the sense of a proposition” (TLP 3.11). The representative of the second
group of commentators – David Hyder – seeks the influence of Wittgenstein’s edu-
cation on the Tractatus somewhere else, and he finds it in Helmholtz’s and Hertz’s
conviction that mind and nature are dynamic models of one another. According
to Helmholtz, in order to model the reality, an experience has to have the same
multiplicity as an event in the world. He and his successors (with Hertz among
them who, as Nordmann writes, “turned Helmholtz’s perceptual manifolds into a
mathematically refined and epistemologically purified space of representation”205)
developed the representational device of spatial manifolds. It was this work, claims
Hyder, that inspired the Tractarian theory of isomorphism of language and the
world: “[According to Hertz] the semantic relation at the heart of scientific models
consists in a mapping relation between purely geometric theories of matter and
appearances in a perceptual manifold. This relation is indeed a form of ‘picturing’,
and I claim that this is the interpretation that Wittgenstein gave to his theory of
logical space once he extended it beyond its logicist origins”206.
I am not going to discuss the above theses that a better understanding of
pre-Russelian inspirations of young Wittgenstein helps to better understand
the basic ideas of the Tractatus. In the context of the discussion about the status
of simples I am interested only in the third group of the aforementioned family

204 “This alphabet of objects, whose configurations produced a variety of working

inventions, gave the engineer working knowledge of the basic component parts of
machines and the principles underlying the forms of machines. This allowed an en-
gineer to visualize these elements recombined into new configurations” (Hamilton
2001, p. 70).
205 Nordmann 2002, p. 381.
206 Hyder 2002, p. 14.

89
of interpretations, namely the one which argues in favour of the thesis that the
answer to the question as to what simple objects of the Tractatus are lies in
Hertz’s Principles of Mechanics. Tractarian simples are supposed to be exactly
the same as the primary elements of the Hertzian system207. This interpreta-
tion, developed by Gerd Grasshof (1997) and Timm Lampert (2000, 2003),
says, therefore, that the simples are material points or mass-particles. Accord-
ing to this interpretation, Wittgenstein in the Tractatus represented physical-
ism and, hence, with respect to this interpretation I will use the terms: “the
physicalistic interpretation” or “the materialistic interpretation”.
As I have said before, I sympathise with this interpretation although I do not
support it completely. In the following passage I shall present the differences be-
tween Grasshof ’s and Lampert’s interpretation as well as the differences between
my point of view and their version of the materialistic interpretation. Next I
shall discuss the arguments and counterarguments in favour of the materialistic
interpretation. I shall also analyse its advantages.

2.2.1 Simple objects as material points, point-masses or

physical atoms
I shall start the presentation of the materialistic interpretation with a conceptu-
al clarification. Every author who advocates in favour of this interpretation has
slightly different ideas when it comes to the question: “what are the simple objects
of the Tractatus?”. The basic difference between my view on Tractarian simples
and the materialistic interpretation of Grasshof and Lampert is that whereas I
claim that the simples of the Tractatus are of a material nature, they are the indi-
visible parts of matter, although it is the matter of scientific progress to determine
what particles are truly elementary, Grasshof and Lampert think that Wittgen-
stein literally inherited Hertzian ontology and what was thought by Hertz to be
the elementary particle of matter is at the same time the simple object of the Trac-
tatus208. I claim that under the weight of evidence one has to admit that early

207 With respect to Hertzian mechanics, Harré writes: “Here we have a much more plau-
sible source of Wittgenstein’s picture theory and the doctrine of simples in a point
of view with which he must have been very familiar than from anything Russell had
to offer him at that time” (Harré 2001, p. 217).
208 “The Prinzipen der Mechanik of Heinrich Hertz were Wittgenstein’s leitmotif and
philosophical stimulus. With a full grasp of its metaphysical content, Wittgenstein
used it as the foundation for the philosophical architecture which is built in close
connection with the logical theory proposed by Russell and Frege” (Grasshof 1997,
p. 87). Another difference between my own and Lampert’s interpretation consists

90
Wittgenstein was clearly impressed by the achievements of contemporary physics
and that he saw in the language of mechanics a language capable of mirroring the
complexity of reality, but he did not necessarily have to see Hertzian mechanics
as a finished system (which neither thought Hertz). He could, for example, as-
sume that on the basis of what one knew in his times about physics, one could
determine that elementary physical objects are the only ones which fulfil the re-
quirements of the postulate of a determinate sense, but this does not mean that he
excluded the possibility of scientific progress, in result of which it would turn out
that the elementary particles of physics are something other than mass-particles.
My understanding of the Tractatus assumes that its worldview is materialistic and
at the same time agnostic with respect to the exact determination as to what the
simple objects of physics are. It is specifically work for the physicists to establish it.
It is no wonder then that Wittgenstein did not fix the reference of the simples and
left this task to empirical research209. With respect to the simples of the Tractatus,
I would use the notion of a physical atom (tantamount to the notion of a physical
particle or the most elementary particle of matter). In my view, the role of this
notion is merely to indicate the physical character of the simples. In this sense my
reading of the Tractatus is close to the reading of Griffin (1964), who also thought
that the simples are material particulars and who was also aware of the fact that
Wittgenstein could not, as a philosopher, determine what the ultimate elementary
constituent of the universe is210.

in the fact that for Lampert, and in contrast to me, the Tractarian thesis about the
existence of simple objects is not motivated by the principle of a determinate sense
(Lampert 2000, p. 329). It also seems that his physicalistic interpretation assumes
only that the propositions of other languages are translatable into the language of
physics, but he does not tie with this thesis any metaphysical claim that, for instance,
the world consists of mass-particles: “Unter einer physikalistischen Interpretation
wird vielmehr in Anlehnung on Carnaps Begriff des Physikalismus verstanden, nach
der Wittgenstein im Tractatus die Auffassung vertritt, sämtliche Sätze seien unter
Voraussetzung bestimmter Übersetzungsregeln in Sätze einer physikalischen Sprache
zu übersetzen” (Lampert 2000, p. 14).
209 “Philosophy gives no pictures of reality, and can neither confirm nor confute scientific
investigation. It consists of logic and metaphysics, the former its basis” (NL, p. 93).
210 “Never in the Tractatus does he give or intentionally suggest an example of an el-
ementary sentence or of an object, and the Tractatus is neutral on all questions which
would require knowledge of an example for their solution; furthermore, he does at
places give certain very general forms of objects and of elementary sentences” (Griffin
1964, p. 164); This position is also similar to Bradley’s interpretation (1992). How-
ever, although Bradley thinks that the simples of the Tractatus are material points,

91
 In contrast to this opinion, Lampert and Grasshof are convinced that Witt-
genstein thought that simple objects are material points (that is the opinion of
Grasshof) or mass-particles (Lampert’s thesis). What is the difference between
these two views? Both of these concepts – “material points” and “mass-particles” –
come from Hertz’s Principle of Mechanics. According to Hertz’s definition, a mass-
particle (ein Massenteilchen) is a space-time location with a particular property; it
is a property of space and time211. As Grasshof underlines, it follows from this that
a mass-particle is not an object in space and time, and one should not associate
with the notion of a mass-particle, for example, the idea of being heavy212. These
associations are allowed, however, in the case of another basic concept of Hertz-
ian ontology – the concept of mass. The mass of any given space s1 is defined by
Hertz as the number of mass-particles in any space s1 compared with the number
of mass-particles in some other space s2 at a fixed time213. Finally, a material point
(ein materieller Punkt) is mass contained in an infinitely small space214. From this

he discerns a metaphysical “material point” from physical atoms. The former is such
that its division into smaller parts is inconceivable (in contrast to physical atoms):
“I’m not saying that Wittgenstein’s metaphysical simples are physical atoms, or any
other material particles. He chooses his words, here, rather carefully, speaking of
points rather than particles. The ‘material points’ – or ‘point-masses’ – of physics are
different from both atoms (as traditionally conceived) and from particles (as currently
conceived). They are such that one cannot even conceive of their being divided into
parts. They are extensionless points, and hence, for purely logical reasons, they are
indivisible” (Bradley 1992, p. 78).
211 “Ein Massenteilchen ist ein Merkmal, durch welches wir einen bestimmten Punkt des
Raumes zu einer gegebenen Zeit eindeutig zuordnen einem bestimmten Punkte des
Raumes zu jeder anderen Zeit. Jedes Massenteilchen ist unveränderlich und unzer-
störbar. Die durch dasselbe Massenteilchen gekennzeichneten Punkte des Raumes
zu zwei verschiedenen Zeiten fallen zusammen, wenn die Zeiten zusammenfallen”
(Hertz 1894, p. 54).
212 “The function of mass-particles at this point is just to mark uniquely a space-time
location, so that such points are countable. That is only what is required to define a
concept of mass” (Grasshof 1997, p. 105).
213 “Die Zahl der Massenteilchen in einem beliebigen Raume, verglichen mit der Zahl
der Massenteilchen, welche sich in einem festgesetzten Raume zu festgesetzter Zeit
finden, heißt die in dem ersteren Raume enthaltene Masse” (Hertz 1894, p. 54).
214 “Eine endliche oder unendlich kleine Masse, vorgestellt in einem unendlich kleinen
Raume, heißt ein materieller Punkt” (ibid., p. 54).

92
definition Hertz draws the moral that a material point consists of mass-particles
connected with one another215.
Grasshof points out that in the Tractatus or in the Notebooks, Wittgenstein
uses the notion of a material point:
Wir dürfen nicht vergessen, dass die Weltbeschreibung durch die Mechanik immer die
ganz allgemeine ist. Es ist in ihr z. B. nie von bestimmten materiellen Punkten die Rede,
sondern immer nur von irgendwelchen (TLP 6.3423).
Die Zerlegung der Körper in materielle Punkte, wie wir sie in der Physik haben, ist
weiter nichts als die Analyse in einfache Bestandteile (NB 20.6.15, p. 67)216.

That is why he thinks, given that Hertz’s ontology influenced early Wittgen-
stein’s worldview, that there are material points which play the role of the sim-
ples in the Tractatus. Moreover, he notices that Tractarian simples are external
objects, and whenever Hertz referred to external objects he used the notion of a
material point. In contrast to this concept, the notion of a mass-particle was an
a priori notion in Hertzian mechanics. This means that in defining what is mass,
one presupposes the existence of such entities as mass-particles. Mass-particles
are only features of space-time. They are like, as Andreas Blank says: “factors of
particularity that allow one to associate a given point in space at a given time
with another point in space at another time”217. The notion contains no empiri-
cal content and that, in Grasshof ’s view, would speak against mass-particles as
candidates for Tractarian simple objects.
On the other hand, as Keyt and Blank argue218, taking into account the Hertzian
definition of a material point which states that material points consist of a number
of mass-particles, one should rather say that the former notion better suits the
notion of Tractarian states of affairs. This is because material points, in the under-
standing of Hertz, display complexity. Perhaps it is for this reason that Lampert
is inclined to choose mass-particles as the simples of the Tractatus219. First, the
characteristics of the concept of mass-particles entail no complexity. The other

215 “Ein materieller Punkt besteht also aus einer beliebigen Anzahl mit einander ver-
bundener Massenteilchen” (ibid., p. 54).
216 The Pears-McGuinness translation uses in this context a confusing notion of
“point-mass”, which one can mistake for the Hertzian notion of Massenteilchen –
mass-particles.
217 Blank 2007, p. 252.
218 Keyt 1965, Blank 2007.
219 “Wittgenstein’s ideas of a physical language were more specific than Carnap’s: it is
the physical language based on Heinrich Hertz’s definition of a mass-particle in the
first part of his book The Principles of Mechanics” (Lampert 2003, p. 286).

93
reason is that the characteristics of simple objects in the Tractatus correspond
to what Hertz ascribed to mass-particles. For example, Hertz writes that mass-
particles are invariable and indestructible (Hertz 1894, p. 54), and Wittgenstein
writes similarly that objects are unalterable and subsistent (TLP 2.027). One of
the consequences of the Hertzian definition of a material point is that it requires
the existence of more than one mass-particle. The concept of a single mass-parti-
cle has no empirical sense. It acquires empirical content only in connection with
other mass-particles (in connection with other mass-particles it forms a material
point, and the notion of a material point is empirical). Wittgenstein by echoing
that conception says that “there is no object that we can imagine excluded from
the possibility of combining with others” (TLP 2.0121). It is true that the concept
of a mass-particle is a priori, but so was the concept of a simple object in the
Tractatus. After all, Wittgenstein did not discover the simple objects with a micro-
scope, but he asserted their existence as the end product of the logical analysis of a
proposition. Moreover, in Chapter 1, in what I called “the room example”, I tried to
show that the fact that the notion of a simple object is a requirement of the picture
theory of meaning does not shatter the conviction that this theory falls under the
type of realist theories of meaning. It turns out that the fact that the notion of a
mass-particle gains an empirical meaning only in connection with the notion of
a material point is not an argument against a mass-particle as a candidate for a
simple object, as Grasshof thought. Quite the opposite – it suits even better the
Tractarian description of the simples.
Despite the differences in opinions as to what the simples of the Tractatus
are – physical atoms as elementary particulars of physics or maybe constituents
of reality described as in Hertz’s system, i.e. material points or mass-particles –
what is common to all representatives of the materialistic interpretation are the
following theses:
(1) The simples are of a material kind (even if the notion of a mass-particle itself
is an a priori notion, mass-particles in Hertz’s mechanics form, after all, ob-
jects of physics).
(2) A physical analysis serves as a paradigmatic example of the Tractarian analy-
sis of complex objects.
(3) The language of physics is the only one that possesses the multiplicity that is
needed to mirror in language every change in the world.
(4) The language of physics is the language of science. This means that every
scientific statement can be translated into a statement of physics.
(5) Wittgenstein’s elementary propositions express material states of affairs and
as such are translatable into the propositions of physics.

94
2.2.2  Arguments in favour of the materialistic interpretation
The materialistic interpretation argues in favour of identifying the simple objects
of the Tractatus with the elementary constituents of material reality. It is hardly
imaginable that Wittgenstein, having constructed such a meticulous and detailed
system of meaningful language as in the Tractatus, did not consider what would
best fill the empty brackets of the notion of an object (at least in the Notebooks
the inability to mention a simple object was for Wittgenstein a difficulty220). In
the Notebooks he considered two main candidates for the simple objects: simple
units of experience (or sense-data) and material points. Because the characteris-
tics of the simples from the Tractatus are in disagreement with the candidacy of
sense-data, the hypothesis that the Tractarian ontology is materialistic becomes
very appealing. What are the arguments in favour of such a hypothesis?
The presentation is divided into three parts: arguments one could draw on
the basis of the lecture of the Notebooks, arguments which point at materialistic
assumptions of some of the Tractarian theses, and Wittgenstein’s accusation of
plagiarism against Rudolf Carnap. 

2.2.2.1  Arguments from the Notebooks

In this section I shall focus on entries of the Notebooks coming from June of 1915.
As we know, one of Wittgenstein’s main concerns during the time of his writing
the Tractatus was which requirements one should put on language in order for it
to be meaningful. The requirement of the determinateness of sense says that prop-
ositions of a meaningful language “restrict reality to two alternatives: yes or no”
(TLP 4.023). In order for a proposition to have a definite sense (and in this way to
depict reality), its notions have to have sharp meanings (“It seems clear that that
what we MEAN must always be sharp” (NB 20.6.15, p. 68)) – only then would the
truth-conditions of a proposition be so determined that it would allow a “Yes” or
“No” answer to the question: “Is this proposition true?”, whatever the case was. We
also know that the consequence of the requirement of a determinate sense is the
postulate of simple objects – the constituents of propositions have sharp meaning
only if they refer to the simples221 – but this particular consequence is not of my

220 “Our difficulty was that we kept on speaking of simple objects and were unable to
mention a single one” (NB 21.6.15, p. 68). This entry follows after two months of
considerations about the nature of the simples. The idea that Wittgenstein was un-
interested in ontological questions is untenable.
221 “The demand for simple things is the demand for definiteness of sense” (NB 18.6.15,
p. 63).

95
immediate concern in the context discussed here. What is important is that Witt-
genstein searched for a language whose propositions have a determinate meaning.
The crucial step in these inquires was undertaken by Wittgenstein on 24 May
1915. On that day Wittgenstein rejected the hypothesis that it is the language of
sense-data which fulfils the requirement of a determinate sense. He realised that
sense-data are always given to us as complex parts that can be divisible into sim-
pler parts. Hence, units of experience cannot be simple objects (what is simple,
concludes Wittgenstein, cannot, therefore, be known by acquaintance), and the
phenomenological language, i.e. language whose names refer to sense-data, can-
not have a determinate sense. Its propositions will always be more or less vague:
Even though we have no acquaintance with simple objects we do know complex objects
by acquaintance, we know by acquaintance that they are complex. – And that in the end
they must consist of simple things? We single out a part of our visual field, for example,
and we see that it is always complex, that any part of it is still complex but is already sim-
pler, and so on (NB 24.5.15, p. 50).

In this respect the language of physics has a clear advantage over the phenom-
enalistic language. Its constituents refer ultimately to elementary constituents of
reality, which are simple:
The division of the body into material points, as we have in physics, is nothing more than
analysis into simple components (NB 20.6.15, p. 67).

As Timm Lampert sums up: “A physicalistic analysis involves analysing the world
in the greatest of detail. It is impossible to make any further distinction after hav-
ing specified points or infinitesimally small units of a physicalistic analysis”222. The
language of physics fulfils the most basic condition for being the candidate of lan-
guage whose constituents have definite meanings, i.e. the condition that names of
such a language refer to simple and indestructible objects.
Wittgenstein does not want to, however, admit too quickly that it is only the
language of physics which is capable of picturing the world and that the physical-
istic analysis is a model of an analysis into simpler parts. After all, this would mean
that only propositions of physics have meaning, and that conclusion would be
very strong, especially when one takes into account that we normally understand
perfectly clear the propositions of everyday language, and that, in turn, suggests
that also everyday language is a meaningful language. Does this mean, therefore,
that an analysis of the propositions of everyday language into simple signs reveals
another type of simple objects? Could it be that the referents of the simple signs of

222 Lampert 2003, p. 295.

96
everyday language are the simple objects next to the elementary particles of phys-
ics? This depends on the question whether it is possible to conduct a full analysis
of a proposition of everyday language, i.e. an analysis which results in a proposi-
tion with a determinate sense. This is why Wittgenstein considers the problem
if the proposition (a) “The book is lying on the table” has a clear sense as “an
extremely important question”223. The possible negative answer would ultimately
weigh in favour of the physicalistic worldview. These are the topics of the extended
(and unfortunately also slightly tangled) entries from 20 until 22 June 1915.
It seems that Wittgenstein provides at least one argument in favour of the thesis
that propositions of everyday language can have perfectly clear sense. This argu-
ment refers to the fact that in the case of vagueness, the meaning of a proposition
can always be precisified by means of an ostension:
If someone were to drive me into a corner in this way in order to shew that I did not
know what I meant, I should say: “I know what I mean. I mean just THIS”, pointing to
the appropriate complex with my finger (NB 22.6.15, p. 70).

Wittgenstein, however, claims that the elements of propositions precisified by

means of ostensive definitions are not simple. The precisification of sense by
means of the ostensive definitions amounts, for Wittgenstein, to the claim that
in the proposition (a) the referent of the name: “book” functions as the simple
object. This means that for the purposes of such and such a conversation, for
such and such interlocutors, the name “book” refers to a simple object. It is a
relatively simple object: “This object is simple for me!” (NB 22.6.15, p. 70). In
the thesis which is a summary of Wittgenstein’s disquisition we read:
It always looks as if there were complex object functioning as simples, and then, also
really simple ones, like the material points of physics (NB 21.6.15, p. 69).

Therefore, in Wittgenstein’s view there are objects which sometimes function as

simple (books, watches, Socrates), and objects which are truly simple (the physi-
cal atoms – the end-products of physical analysis). But why did he decide to as-
cribe the feature of relative simplicity to the end-products of analysis of sentences
of everyday language? After all, if the proposition: “The book is on the table” has
the definite sense, then should we not consider the referents of the names of this
proposition to be just simple? According to Wittgenstein, however, the proposi-
tions of everyday language, despite the precisifications we pursue, do not have a
determinate sense:

223 “When I say, ‘The book is on the table’, does it really have a completely clear sense?
(An EXTREMELY important question)” (NB 20.6.15).

97
 It is obvious that a proposition like “This watch is lying on the table” contains a lot of in-
definiteness (NB 21.6.15, p. 69).

Sometimes we are not certain how to call certain positions of the watch, or it is
doubtful that the notion of “lying on” describes the relation between the watch
and the table in the best possible way. Wittgenstein believed that even if for the
sake of a conversation one stipulates the meaning of the proposition (a) by an
ostensive definition, it would not change the character of the objects one refers
to by expressing (a):
Now when I do this and designate the objects by means of names, does that make them
simple? All the same, however, this proposition is a picture of that complex (NB 22.6.15,
p. 70)

It turns out that for Wittgenstein what counts, apart from the requirement of a
determinate sense, is also the nature of objects. He has certain metaphysical views
on objects that result in the conviction that a book could at best function as a
simple object.
I shall try to reveal these views. One can think about objects in two ways.
Following Michael Jubien’s distinction, we can see reality as consisting of objects
such as horses, statues and puddles224. Horses and statues are naturally physical
objects but, according to this way of thinking, being a horse is a more fundamen-
tal feature of an object than being composed of physical particles. In the second
view one can see reality as made up of physical objects which merely happen to
be, for example, horses or statues. The metaphysics of the first point of view is
silently assumed in the everyday language. The metaphysics of the second point
of view is the metaphysics of physicalism225. The deficiency of the metaphysics of
everyday language, from Wittgenstein’s point of view, consists in the fact that it
assumes that objects instantiate essences. The world, under this conception, con-
sists of matter (the physical material of an object) and essences (what makes an
object what it is – a horse, a statue or a puddle). The notion of essence belongs to
the old Aristotelian tradition of metaphysics, which describes the world from our
human perspective. In contrast to this tradition, the Tractatus tried to describe
the world sub specie aeterni, i.e. how it is independently of any points of views226.

224 Jubien 2007, pp. 108–109.

225 “Helmholtz, Hertz and Boltzmann (…) like their predecessors in England, during
the seventeenth and eighteenth centuries, took the really real world to be a domain
of masses in motion. That was all there was to the world” (Harré 2001, p. 225).
226 Therefore, I agree with Sluga that the Tractatus pursued the project of describing the
world in entirely objective terms, and this meant that it rejected “the old Aristotelian

98
Note also that Wittgenstein believed that typical metaphysical concepts have no
meaning at all:
The correct method in philosophy would really be the following: to say nothing except
what can be said, i.e. propositions of natural science—i.e. something that has nothing to
do with philosophy—and then, whenever someone else wanted to say something meta-
physical, to demonstrate to him that he had failed to give a meaning to certain signs in his
propositions (TLP 6.53).

These arguments should incline us to think that Wittgenstein rather professed

the metaphysics of the second type. This metaphysics is also the one which makes
possible Wittgenstein’s project of constructing a language whose propositions
would have a determinate sense.
Let us summarise: the fact that Wittgenstein discerns between relative and ab-
solute simple objects suggests that he indeed was interested in the nature of objects
and, secondly, that the condition that objects are the referents of the names of a
proposition with a determinate sense is not enough to ascribe to those objects the
feature of simplicity. It seems that Wittgenstein had another condition in mind,
namely an absolute indivisibility of objects. Only in the case of absolute simple ob-
jects is further analysis into simpler parts impossible, and the reason for vagueness
of propositions is eradicated for good. It follows then that he accepted a material-
istic worldview, and that by “the simples” he understood the elementary constitu-
ents of matter. In this section I have also tried to show, with the help of Jubien’s
distinction, what Wittgenstein’s worldview could look like (that it assumed that the
world consists of physical particles that just happen to be objects of familiar kinds).
From these considerations it follows that indeed such a worldview suits better early
Wittgenstein’s conception of philosophy because it is a worldview which assumes
only the existence of objects of science and does without metaphysical essences.

2.2.2.2  Arguments from the Tractatus

In this section I shall analyse two arguments in favour of the materialistic inter-
pretation which are based on the text of the Tractatus. As I have mentioned earlier,
entries from the Notebooks are simply hints, Wittgenstein tried there different po-
sitions and one should not put too much weight on the formulations. They are not
decisive when it comes to an interpretation of the Tractatus. Therefore, in the first
argument I shall try to show that the vision of the world as consisting of physical

conception of the world, which, it is said, interpreted things in human and, hence,
subjective terms” (Sluga 1983, p. 125). I disagree, on the other hand, with Levi’s opinion
that “Wittgenstein’s Tractatus is indeed Aristotelian in spirit” (Levi 1964/1967, p. 370).

99
objects that just happen to be, for example, horses, statues and books, is also pre-
sent in the Tractatus. In my opinion, this aspect of early Wittgenstein’s thought
did not disappear in the process of his crystallising his views. The second argu-
ment aims to prove that one should ascribe to Wittgenstein physicalism in the
Carnapian sense, i.e. the view that every meaningful sentence is translatable into
the language of physics (not in the meaning that every psychological or biological
law has its counterpart in a physical law, but rather that every meaningful sentence
expresses physical states of affairs; states of affairs that are also describable in the
language of physics). Physicalism as defined by Carnap does not necessarily have
to be a metaphysical standpoint declaring which objects exist. A Carnapian physi-
calist can simply claim that the language of physics has the multiplicity suitable to
express all kinds of facts; that in this language one can predict future events in the
most accurate way, without going into considerations on what there is: mass-parti-
cles, sense-data, essences or moral values. But if I was correct in defending the the-
sis that Wittgenstein was a proponent of semantic realism, one could indeed draw
metaphysical conclusions from the thesis that the Tractatus advocates in favour of
Carnapian physicalism. If the referents of the names of a meaningful proposition
have to exist in order for that proposition to be meaningful and if the propositions
of physics are meaningful, then objects postulated by physics must exist.
I. In the previous section I interpreted a fragment of Notebooks 20-22.6.15
as considerations about the proper worldview. Does reality consist of physical
objects that happen to be horses, statues or books or does it consist of objects
instantiating their essences – objects which are fundamentally their essences? Ac-
cording to my understanding, Wittgenstein opted for the atomistic view, accord-
ing to which reality consists of various complexes of mass-particles (mereological
parts of these complexes). My current question is whether in the Tractatus one
could find traces of this ontological conviction? Answering this question, I would
like to invoke the thesis TLP 5.631:
There is no such thing as the subject that thinks or entertains ideas. If I wrote a book
called The World as I found it, I should have to include a report on my body, and should
have to say which parts were subordinate to my will, and which were not, etc., this being
a method of isolating the subject, or rather of showing that in an important sense there
is no subject; for it alone could not be mentioned in that book (TLP 5.631).

I argue that this position is exactly what we would expect from a proponent of the
atomistic view of the world. In this world there are only physical particles arranged
in a certain way. That is why if one describes the world from one’s own perspec-
tive (and we assume that Wittgenstein requires that one does so meaningfully),
the description, when it comes to the person describing the world, will contain

100
only the mentioning of one’s body. From a materialistic perspective this concrete
arrangement of physical particles which forms my body just happens to be me.
Perhaps I have gained weight, so there was another arrangement of particles which
formed my body yesterday, but my current body and yesterday’s body are two dif-
ferent objects. Naturally, intuition tells us that something has not changed, namely
the fact that both objects are my bodies. But in order to express this conviction
one needs the metaphysics of essences. Then one would be able to say that this
object instantiates the essence of this particular human being that does not change
through the flow of time (we could refer to this essence with the help of the notion
of a subject. Obviously, I assume that the notion of the subject in TLP 5.631 refers
to the essence or to the core of the human being. My conviction is grounded in
the fact that Wittgenstein clearly juxtaposes here this notion with the notion of the
body. He had to, therefore, understand in TLP 5.631 a subject as something differ-
ent than a body), and then the world would consist not only of physical atoms, but
also of subjects. Now, I do not say that this view is wrong. What I want to prove
is simply that early Wittgenstein thought it was wrong. In his view, a proper de-
scription of the world would mention no subjects (and we can with all likelihood
assume that also no essences of other objects), but just arrangements of particles –
bodies. It follows from this that his worldview during the time of editing the final
version of the Tractatus agrees with the atomistic worldview he discussed in June
of 1915. According to it, reality consists of various complexes of physical particles.
These particles are truly the substance of the world – they are bricks that form all
objects of the world. If that is so, then one must accept the materialistic interpreta-
tion as that which best explains what the simple objects of the Tractatus are.
II. The second argument, which is formulated by Timm Lampert, centres
around the theme of the model of analysis that was employed in the Tractatus.
According to Lampert, the basis of the Tractarian analysis was the physical
language – the one which was familiar to him via Hertz’s The Principles of
Mechanics227. The line of Lampert’s reasoning is the following: if the language
of physics is indeed the basis of the Tractarian analysis of propositions, then
Wittgenstein treated meaningful propositions as translatable into the language
of physics, and this thesis lies at the heart of physicalism as defined by Carnap.
Lampert shows that with respect to the sense-data language, Wittgenstein

227 “For [Wittgenstein], a physical language served as a basis of analysis and the distinc-
tion between sense and nonsense. (…) It is the physical language based on Heinrich
Hertz’s definition of a mass-particle in the first part of his book The Principles of
Mechanics” (Lampert 2003, p. 286).

101
made use of psychophysical analysis of colours in the Tractatus at least two
times. In TLP 4.123 he wrote:
This shade of blue and that one stand, eo ipso, in the internal relation of lighter to darker.
It is unthinkable that these two objects should not stand in this relation.

Here, Wittgenstein excludes that “being brighter than” is the external relation,
i.e. the relation that one can state by comparing two objects. Brightness, there-
fore, is the internal feature of an object and, according to Lampert228, one can
make sense of this rather paradoxical statement only under the assumption of the
physicalistic notion of a colour, which pervaded during the time of writing the
Tractatus229. If “being brighter than” were an external relation, then one would
determine the truth-value, for example, of the proposition (c): “This shade of
blue is darker than this one” simply by comparing two colours. It would mean,
however, that the proposition (c) would function as an elementary proposition
whose elements are simple signs. In other words, one would treat patches of
colours as simple objects. A proponent of the phenomenalistic interpretation
of the Tractatus would be very pleased with this conclusion, but Wittgenstein
states something quite opposite. A feature of brightness is, according to him, an
internal one. One of the practical consequences of this thesis is that one cannot
fix the truth-value of the proposition (c) simply by looking at the state of affairs
that is expressed by (c), but by further analysing the proposition (c). This analy-
sis, suspects Lampert, is the psychophysical analysis of colours230. According to
the views of physicists contemporary to Wittgenstein, colour was a plurality of
shades-units caused by light of a certain duration of oscillation. According to
Lampert, the proposition: “A bright shade of blue is brighter than a dark shade
of blue” is meaningful. It is a truth-function on certain elementary proposi-
tions, and one has to reveal what these elementary propositions are. I disagree
with Lampert. I am convinced that a more accurate formulation of the problem,
more faithful to Tractarian principles, is the following: The expression: “A bright
shade of blue is brighter than a dark shade of blue” is, according to Tractarian

228 “Wittgenstein schließt in 4.123 explizit aus, dass die Helligkeit eine externe, unana-
lysierbare Beziehung zwischen einfachen Gegenständen ist, und setzt hierbei die
physikalische Farbanalyse voraus” (Lampert 2000, p. 185).
229 This conviction was explicitly expressed by Wittgenstein in the Notebooks: “That the
colours are not properties is shown by the analysis of physics, by the internal relations
in which physics displays the colours” (NB 11.9.16, p. 82).
230 “Eine Farbe besteht aus Farbtoneinheiten, die durch ein bestimmtes Licht verursacht
werden und für die gilt, dass sie sich an einem Punkt im Gesichtsfeld befinden”
(Lampert 2000, p. 179).

102
criteria, nonsense because a negation of this expression is unthinkable. What it
tries to say could only be shown by means of other propositions, for instance,
one which ascribes to object A a certain number of shades-units and the sec-
ond which ascribes to object B a certain number of shades-units. Whatever the
proper formulation of the problem is, thesis TLP 4.123 is understandable only
under the assumption of a physical analysis of colours. It seems then that in the
Tractatus a physical analysis serves as the model for analysis of the language
of sensations: one makes sense of propositions about sensations by translating
them into propositions of physics. This conclusion is even more clear in thesis
TLP 6.3751, in which Wittgenstein tackles the colour-exclusion problem:
For example, the simultaneous presence of two colours at the same place in the visual field
is impossible, in fact logically impossible, since it is ruled out by the logical structure of
colour. Let us think how this contradiction appears in physics: more or less as follows—a
particle cannot have two velocities at the same time; that is to say, it cannot be in two
places at the same time; that is to say, particles that are in different places at the same time
cannot be identical. (It is clear that the logical product of two elementary propositions can
neither be a tautology nor a contradiction. The statement that a point in the visual field
has two different colours at the same time is a contradiction.) (TLP 6.3751)

When one states:

(d):  “This patch is blue”,
one can conclude, and conclude with certainty, that, for example:
(e):  “This patch is not yellow”.
It seems that
(f):  “That patch is blue and, therefore, not yellow”
is necessarily true. It poses a problem for the Tractarian system because in the
Tractatus the only necessary truths are logical ones. However, it seems that (f) is
neither a logical truth nor a contingently true proposition. In TLP 6.3751, how-
ever, Wittgenstein states that (f) is indeed a logical truth, and he analyses colour
propositions in physicalistic terms. In other words, if only under the physicalistic
interpretation one can hold the essential division of true propositions into necessary
logical truths and contingent empirical truths, then in order to read the Tractatus
as a consistent system one has to accept its physicalistic interpretation231.

231 According to Lampert, one needs an additional condition in this line of reasoning, i.e.
the condition that one proves the logical impossibility not only by demonstrating the
contradiction between two statements, but also by demonstrating the incompatibility

103
 Wittgenstein’s line of reasoning is the following: he reformulates (f) as a propo-
sition about two different colours existing at the same moment at the same place:
(f ’): “Two colours are simultaneously present at the same place in the same visual
field”
He states that (f ’) is logically impossible. Now Wittgenstein takes a decisive step
and proposes a physicalistic analysis of (f ’): “Let us think how this contradiction
appears in physics: more or less as follows – a particle cannot have two velocities at
the same time” (TLP 6.3751). Only by means of such an analysis does he manage
to prove the logical impossibility of (f ’). (F’) is translated into:
(f ’’):  “A particle has two velocities at the same time”
Wittgenstein refers here to Helmholtz’s definition of colour, according to which
colours are sensations which are caused by ether parts oscillating with a certain
velocity232. The fact that (f ’’) is a translation of (f ’) indicates that Wittgenstein
held the former proposition to be more fundamental. It also supports the hypoth-
esis that Wittgenstein employed the psychophysical analysis with respect to the
language of sensations, and this, in turn, suggests his physicalism. Wittgenstein
continues his argument in the following way:
A particle cannot have two velocities at the same time; that is to say, it cannot be in two
places at the same time; that is to say, particles that are in different places at the same time
cannot be identical (TLP 6.3751).

This thesis contains two steps. In the first step (“a particle cannot have two veloc-
ities at the same time; that is to say, it cannot be in two places at the same time”),
Wittgenstein used, as Lampert suspects, the definition of momentary velocity
(v = dx/dt). (F’’) allows different values of dx for one particle (because there can
also be two values of v for one particle), i.e. it does not exclude the case of one

of the proposition in question with the basic definitions of mass-particles. In other

words, by demonstrating that a statement cannot be formulated in the language of
physics: “Wittgenstein’s proof of colour exclusion does not result in a demonstrating
contradiction using rules of analysis, but as a demonstration of how the statement
that two colours are at the same point in the visual field at the same time is incom-
patible with Hertz basic definition of particles (…) This shows that Wittgenstein’s
criterion of logical possibility lies in the compatibility with the mechanical world
description according to Hertz’s definition of a mass-particle” (Lampert 2003, p. 311).
232 “In the purely mechanical interpretation of the wave theory of light, which both
Helmholtz and Hertz held, simple colours are caused by ether particles oscillating
with a certain velocity” (ibid., p. 294).

104
particle being in two places at the same time. Let us call this consequence the
consequence of (f ’’) (in short (Cf ’’)). In the second step, Wittgenstein applied
Hertz’s aforementioned definition of a mass-particle (c.f. footnote 211):
(HD):  “Particles that are in different places at the same time cannot be identical”
(Cf ’’) is contradictory to (HD). This is enough for Wittgenstein to state:
The statement that a point in the visual field has two different colours at the same time
is a contradiction (TLP 6.3751).

At this point of his philosophical career, Wittgenstein was satisfied with his solu-
tion of the colour-exclusion problem (his later dissatisfaction is of no interest here
because my investigation refers only to a proper interpretation of the Tractatus).
Independently of the fact whether the presented proof is the answer to the chal-
lenge of the colour-exclusion problem, TLP 6.3751 betrays an interesting, from my
point of view, conviction. Wittgenstein manages to show the logical impossibility
of (f ’) – first of all, under the condition of an analysis of sensations in purely physi-
cal terms; secondly, under the assumption of definitions that imply the physicalis-
tic worldview. In other words, if Wittgenstein did not accept these definitions, he
could not prove the logical impossibility of (f ’). Thesis TLP 6.3751 is intelligible
only under the assumption that Wittgenstein accepted the standpoint according to
which reality consists of physical atoms.

2.2.2.3  The argument from Wittgenstein’s testimony

One testimony confirms that Wittgenstein thought of the Tractatus as physicalistic
in spirit. It comes from 1932, so it has to be treated with reservations for, as we re-
member, from that period there come also Wittgenstein’s testimonies that suggest
a phenomenalistic reading of his work. In that year Wittgenstein became famil-
iar with Rudolf Carnap’s article Die physikalische Sprache als Universalsprache der
Wissenschaft. In this paper, Carnap propagated ideas of physicalism. He claimed
there that “all statements in science can be translated into physical language”233.
He believed that all terms of natural science (for example, biology, but also psy-
chology) are ultimately reducible to the terms of physics234. The physical language
describes every state of affairs, and all states of affairs are of one kind235. In Carnap’s

233 Carnap 1931/1995, p. 28.

234 “Every statement in biology can be translated into a physical language (…) All psy-
chological statements refer to physical events” (ibid., p. 70–71).
235 “Science is a unity, all empirical statements can be expressed in a single language, all
states of affairs are of one kind and are known by the same method” (ibid., p. 32).

105
view, “because the physical language is the basic language of Science the whole of
Science becomes Physics”236. The physical language is the only intersubjective lan-
guage, and Carnap also proposed an analysis of sentences of the protocol language
(reports of a subject’s experiences) into sentences of the physical language; for in-
stance, a report concerning a note of such and such pitch, timbre and intensity
could be expressed in the physical language in terms of “material oscillations of
such (specified) basis frequency with superimposed additional frequencies of such
(specified) amplitudes”237. Thanks to such translations, Carnap could ultimately
conclude that experiences are nothing more than physical facts.
Wittgenstein read Carnap’s paper and reacted with rage. He accused Carnap
of committing plagiarism. Carnap answered this accusation in a manner that is
similar to what is claimed by the proponents of the anti-metaphysical interpreta-
tion of the Tractatus – that Wittgenstein never spoke in favour of physicalism and
that there is no trace in the Tractatus of the claim that we can identify elemen-
tary propositions with physical ones or that simple objects are physical atoms. In
response to this defence, Wittgenstein wrote a letter to Moritz Schlick in which
he complained: “That I had not dealt with the question of ‘physicalism’ is untrue
(only not under that – horrible – name) and I did so with the brevity with which
the whole of the Tractatus is written”238.
Following the hint of this letter, one should not expect direct and unambigu-
ous confirmations of physicalism in the Tractatus, and one, of course, will not
find the term “physicalism” in the book, but, still, this does not mean that there
are no physicalistic ideas contained there. Wittgenstein’s conviction that Carnap
had simply copied his thoughts could also help us understand why, if early Witt-
genstein was convinced that the world consists of physical atoms, he identified
the totality of true propositions with propositions of natural science and not
with propositions of physics. The explanation could be that he, just as Carnap,
believed in the unity of science, so that there was no need for him to contrast
physics with other branches of science.
Opponents of the physicalistic reading of the Tractatus are convinced that
the accusation of plagiarism refers to Wittgenstein’s views from the period of
1929–1932239, but this claim clearly does not stand the force of evidence of
Wittgenstein’s letter to Schlick. The letter explicitly mentions the Tractatus in

236 Ibid., p. 97.

237 Ibid., p. 57.
238 cited in: Monk 1990, p. 324.
239 Haller 1989, p. 29.

106
the context of plagiarism. Similarly, according to Stern, Wittgenstein’s resent-
ment towards Carnap was caused by the fact that Carnap did not mention him
as the guiding spirit of the idea of the nature of philosophy, or of the ideas that
the ostensive definition “does not lead us outside language”, and that philo-
sophical pseudo-problems are eliminated “by means of the formal mode of
speech”240. In his view, the accusation could not refer to Carnap’s ontology be-
cause the Tractatus did not defend – in contrast to Carnap’s paper – the thesis
of the primacy of physical language241. In my opinion, the arguments provided
by Lampert as presented earlier in the chapter prove that in the Tractatus the
language of physics is indeed considered to be the most fundamental one, and
Wittgenstein analysed propositions reporting sensations in terms of physics
contemporary to him, hence Stern’s views are wrong.

2.2.3  Advantages of the materialistic interpretation

In the previous section I gave arguments in favour of a materialistic interpreta-
tion of the Tractatus. In short, I indicated:
A) The fact that some excerpts of the Notebooks as well as of the Tractatus sug-
gest that Wittgenstein saw the world as consisting of complexes of physical
particles.
B) The fact that Wittgenstein held psychophysical analysis to be the model anal-
ysis for the language of sensations (Lampert’s argument).
C) The accusation of plagiarism against Carnap who advocated in favour of phys-
icalism.
The materialistic interpretation claims that the simple objects of the Tractatus are
the most elementary particles of matter. In the opinion of the advocates of this
interpretation, this thesis suits well the whole Tractarian system:
1. First, it gives a concrete form to some Tractarian ideas. Without specifying
the reference of the simple signs, Wittgenstein’s early oeuvre would remain
merely an intellectual exercise, an interesting and sophisticated one, but only
an exercise. For example, Wittgenstein’s view on tautology (TLP 4.46-4.4661),
held to be one of the greatest achievements of the Tractatus242, rests on the

240 Stern 2007, p. 322.

241 Ibid., p. 329.
242 Shanker, taking the point of view of contemporary philosophy, sees the Tractatus as a
set of “curiosities, offering interesting critical problems, but hardly serious positions
that any modern philosophers would want to defend or pursue” (Shanker 1986a,

107
 division between senseless logical truths (TLP 4.461) and contingent truths
(TLP 4.464). But in order to maintain this sharp division one has to make, as
the colour-exclusion problem shows, certain assumptions about reality. One
needs ontology in order to understand Wittgenstein’s book. The materialistic
interpretation provides it.
2. As we saw earlier, the phenomenalistic interpretation has trouble coincid-
ing sense-data with the attributes ascribed in the Tractatus to the simples.
In contrast, physical atoms fulfil the requirements put on the simple objects
that could not be fulfilled by the simple units of experience. According to
Hertz, mass-particles were invariable and indestructible, and that resembles
the Tractarian characteristic of the simples: thesis TLP 2.0271 claims that
objects are “unalterable and subsistent”. Physical atoms are unalterable and
subsistent in the sense that although I am 33 years old, the physical particles
of which I am composed are as old as the universe. The proponents of the
materialistic interpretation also underline that such a criterion of simplic-
ity such as indivisibility discloses the most elementary physical particles as
perfect candidates for the simples (because from the definition this Holy
Grail of the physicists’ investigations is indivisible).
  According to the physicalistic interpretation of the Tractatus, the form of
an elementary proposition looks as follows:
 (*) MST,

where M goes proxy for the most elementary particle of matter (for instance
in the system of Hertz’s mechanics M goes proxy for a mass-particle), S for a
point in space and T for a moment in time. This corresponds to the Tractarian
definition of an elementary proposition which is a concatenation of names
(TLP 4.22). Moreover, if one assumes this form of an elementary proposi-
tion, one also gains an explanation for one of the most difficult dictums of the
Tractatus, which says that objects cannot be thought of independently of their
connections with other objects (TLP 2.0121). In the same thesis, Wittgenstein
simply explains that one cannot think of an object outside of space and time
(TLP 2.0121). According to the materialistic interpretation, this statement
does not provide an example or an analogy that one can afterwards apply
to other objects. It is a straightforward illustration of the claim that one can
think about objects only in connection with other objects.

p. 17), and he believes that it is exactly the Tractarian view on tautologies that could
explain the enormous popularity and constant influence of the Tractatus on modern
philosophy (ibid., p. 19).

108
   The physicalistic interpretation also explains why an elementary proposition
cannot contradict any other elementary proposition (TLP 4.211). For example,
no proposition can contradict: (**) M1S1T1. If one said that the proposition
(***) M1S2T1 is in contradiction to (**), then a proponent of the materialistic
interpretation would answer that one, on the grounds of the language of phys-
ics, cannot even formulate (***). (***) breaks the rules of the logical syntax of
this language and is simply nonsense.
3. Another advantage of the materialistic interpretation is that the candidacy
of physical particles as the simples responds to the requirement that sense
must be determinate. This requirement, expressed by TLP 4.023: “A propo-
sition must restrict reality to two alternatives: yes or no”, is the basis of logi-
cal atomism. In order to restrict the reality to yes or no, a proposition has to
have a determinate sense. Metaphorically speaking, it has to ask reality the
question which excludes answers such as: “It depends, what you mean by…”.
Why do proponents of the materialistic interpretation claim that material
simples satisfy the requirement of a determinate sense better than units of
experience? It is best to explain this on the example of colours243. The spec-
trum of colours is a continuum without sharp boundaries between different
shades. It is then difficult to judge if a given colour is a shade of yellow or
already a shade of green. If the continuum of the spectrum of colours were
indefinite, then colours would overlap and there would be situations where
the decision if something is yellow or green would be purely conventional.
On the other hand, if the spectrum of colours was not indefinitely continu-
ous, and if it was possible to end the analysis at the smallest shades of col-
ours, then it would be possible to establish differences between colours. We
would be able to decide what the colour of a given thing is: is it the last shade
of yellow or the first shade of green? The requirement of a determinate sense
would be kept. Under the phenomenalistic interpretation such an analysis
of colours is difficult to imagine. On the other hand, physics, with its notion
of colour, where the notion of wavelength plays a central role (colour is a
property derived from degrees of stimulation which light of different wave-
lengths gives to different types of cone cells in the retina), seems to provide
the necessary tools. It seems, then, that if one wants to preserve the require-
ment of a determinate sense with respect to reports of one’s experiences, one
has to accept the materialistic interpretation of the simples.

243 This example is contained in: Carruthers 1990, p. 88.

109
4. In the last two chapters I was searching for the answer to the question: “Why
did Wittgenstein identify the totality of true propositions with the propositions
of natural science?”. The materialistic interpretation gives me the answer. It is
because Wittgenstein considered the world as primarily consisting of physical
objects (complexes of the most elementary particles of matter) that just happen
to be objects of certain kinds. It is a world described by physics. One of the con-
ditions of the truthfulness of a proposition is that all the names of this proposi-
tion have referents. The referents of names are objects in the world. If ultimately
the only objects of the world are physical particles, then only propositions of
physics can fulfil this condition of the truthfulness of a proposition. TLP 4.11
becomes intelligible.

2.2.4 Counterarguments
The last advantage of the materialistic interpretation that I mentioned above
is sometimes considered to be its greatest drawback. One can question: “What
about the propositions of everyday language?”. It follows from the materialistic
interpretation that the proposition: “The book is lying on the table” is not true
even if the book is lying on the table. If only propositions of natural science are
true, then it follows that propositions of everyday language are not244. The answer
of a proponent of the materialistic interpretation could take on the following
form: the proposition: “The book is lying on the table” is meaningful, and one
shows this by means of physical analysis, which reveals that when one speaks
about books and tables, in fact one is speaking about arrangements of particles
standing in a certain relation to one another (“The complete physical proposi-
tion does after all deal with things, relations and so on” (NB 20.6.15, p. 67)). This
means, however, that (assuming that the meaning of a proposition is a state of af-
fairs expressed by the proposition) most of the time we are not aware what we are
speaking about. It seems that we refer to objects such as trees, cars and watches
when in fact we are referring to basic elements of matter. It seems that we express
the states of affairs we perceive, such as that the book is on the table, when in fact
we are expressing states of affairs most of us are not aware of:

244 Max Black regards the quest for the simples to be futile, and he consequently thinks
that thesis TLP 4.11 is unacceptable: “The identification of science with the totality
of contingent truths obliterates any distinction between science and common sense,
does not square with the most sophisticated account of scientific language supplied
later (e.g. at 6.341), and is therefore unacceptable” (Black 1964, p. 186).

110
 On Grasshof ’s account, neither “Einstein” nor “Berne” are names in the sentence “Einstein
is in Berne” – these names vanish in the course of analysis. But when a sentence is used to
locate a person in the city, it is not locating a spatio-temporal concatenation of molecules
in respect to buildings, streets, let alone bricks or the other molecules that the bricks are
made of. None of these are properly elements of the thought that is to be expressed by the
sentence (Nordmann 2002, p. 376).

This consequence of the materialistic interpretation seems to be unacceptable

to some of the commentators. Nordmann indicates the basic contradiction be-
tween this conclusion and theses TLP 3.2-3.201245:
In a proposition thought can be expressed in such a way that elements of the proposi-
tional sign correspond to the objects of the thought. I call such elements “simple signs”,
and such a proposition “completely analysed”.

For Nordmann, it is clear that elements of the thought about Einstein’s stay in
Berne are, among others, Einstein and Bern. Let us take the proposition:
(g): “Einstein rented the flat on the second floor of Kramgasse 49 from 1903
to 1905”
According to thesis TLP 3.2, the elements of the proposition (g) correspond to the
elements of the thought about Einstein’s stay in Berne. But if the elements of this
thought are Einstein and Berne, then it follows that the elements of the proposi-
tion (g) are: “Einstein” and “the flat on the second floor of Kramgasse 49”. These
elements, according to thesis TLP 3.201, are simple signs. It seems that this conclu-
sion stays in contradiction with the materialistic interpretation.
What arguments could the proponent of the materialistic interpretation give in
order to defend his way of reading the Tractatus? He could, for example, refer to
the text of the Tractatus, which states that in a proposition a thought can be, but
not necessarily is expressed in such a way that elements of the propositional sign
correspond to the objects of thought. If it can be expressed this means that there
are cases in which the elements of the propositional sign do not correspond to
the objects of thought. Wittgenstein does not make precise in what percentage of
cases there is a correspondence between the elements of thought and the elements
of propositions. This percentage could be very small, and it could refer only to the
thoughts of molecular physicists.
This reply sounds slightly artificial. It is hardly imaginable that Wittgenstein
suddenly thought about the molecular physicists and their thoughts about their
professional interests and decided to include in the body of the Tractatus a

245 Nordmann 2002, p. 376, Carruthers 1990, p. 125.

111
thesis referring to them. Therefore, advocates of the materialistic interpretation
need to assume a different strategy of defending their views. For example, a pro-
ponent of the materialistic interpretation could admit that it is “impossible to
analyze each sentence to the level of its atomic components”246. Then he could
make use of Wittgenstein’s aforementioned distinction between simple objects
and objects that function as simple. Grasshof provides here an analogy from
the practice of physicists who, for the sake of the comfort of a calculation (for
example, a calculation of forces that influence the body, or of the kinetic energy
of the body) treat material bodies as material points. In this analogy, Einstein
and Berne are like extended material bodies which for the sake of convenience
are treated as material points – the referents of the simple signs. Therefore, we
are allowed to say that Einstein and Berne are the elements of thought about
Einstein’s stay in Berne and, hence that “Einstein” and “Bern” are simple signs,
but only if one is aware that “Einstein” is a shortcut for “a composition of physi-
cal molecules”; a shortcut we employ for the sake of the convenience of com-
munication. In this way one can save the materialistic interpretation despite the
challenge of theses TLP 3.2-3.201247.
In Chapter 1 I defended the realistic theory of meaning. According to it, apart
from breaking the rules of logical syntax, the other source of nonsense is the lack
of reference of the names of a proposition. If we accept the realistic theory of
meaning, then there are two major candidates for Tractarian simple objects: sim-
ple units of perception and physical atoms. I discussed both of these hypotheses.
I believe that there are more convincing arguments on the side of the materialistic
interpretation (these arguments are, among others, the fact that material points
are compatible with the Tractarian description of the simples, the way in which
Wittgenstein solved the colour-exclusion problem, his accusation of plagiarism
against Carnap). That is why, according to my hypothesis, the materialistic inter-
pretation of the simples is the best way to understand the Tractatus. The simples
are of a material nature, the substance of the world consists of physical parti-
cles, only these entities could be rightfully considered as objects, and meaningful
propositions can by translated into the language of physics. That point of view
could be rightly named physicalism.

246 Grasshof 1997, p. 104.

247 Naturally, this line of defence does not satisfy the adversaries. It is because TLP 3.201
does not say that the elements of propositions that refer to elements of thought func-
tion as simple signs, but that they are simple signs.

112
2.3  Resolute interpretation of the Tractatus
Before I continue my considerations about the Tractatus, I shall refer to one
more proposal of reading this book. Its basic assumption is the alleged fail-
ure of all the previous trials to understand the Tractatus. An anti-metaphysical
reading sees the Tractarian object as a purely formal notion. Its main assump-
tion says that early Wittgenstein was primarily a logician who was not inter-
ested in metaphysical problems. This corresponds well with the fact that the
postulate of simple objects was in fact the consequence of the semantic views
of young Wittgenstein (his demand that the sense of a proposition should be
definite); it reflects Wittgenstein’s anxiety, expressed in the Notebooks, that he
is unable to give an example of a simple object but, on the other hand, it is in
contradiction with Wittgenstein’s desire to know the nature of being, or with
semantic realism, which I believe is confessed by early Wittgenstein. According
to Wittgenstein, in order for a proposition to be meaningful, its constituents
have to have referents. Therefore, the semantic considerations have to be com-
pleted with the metaphysical question: “What could be a referent of a name?
What exists?”. The phenomenalistic interpretation claims that there are simple
units of experiences – sense-data – that exist. A very strong argument in fa-
vour of this interpretation is Wittgenstein’s identification of the world with my
world, his conviction that what solipsism means is quite correct but, on the
other hand, the characteristics of Tractarian simples (that they are unalterable
and subsistent, that they can exist only in states of affairs) are incompatible
with the characteristics of sense-data. Finally, the materialistic interpretation
claims that the simples of the Tractatus are the most elementary constituents of
matter. The description of physical particles corresponds to what Wittgenstein
wrote about the simples, and some theses of the Tractatus assume the necessity
of the translation of propositions of the language of sensations into a language
of physics, but on the other hand, one of its conclusions is that elements of
thought are not what we think they are. According to the materialistic inter-
pretation, Wittgenstein was convinced that in our thoughts we refer to physical
particles instead of, for instance, tomorrow’s shopping, Andy Murray’s victory
in Wimbledon or Einstein’s stay in Berne.
The fact that each of the above interpretations has its merits and flaws raises
a suspicion that the whole dispute about the simple objects is a complete mis-
understanding of the aims of the Tractatus. Recently, many of Wittgenstein’s
commentators have been inclined to regard the discussions I presented in the
last two chapters as, in the view of the author of the Tractatus himself, hope-
lessly fruitless and senseless. This interpretation, often called the New Reading,

113
or the resolute interpretation of the Tractatus, has recently gained recognition
among scholars (Diamond, Conant, Crary, Ricketts, Ostrow), so in the light of
the topics I am discussing here it cannot be ignored. It decidedly (as the name
of the interpretation states) challenges the solutions I defend. That is why I
shall refer to the most common arguments of the resolute interpretation in the
following sections of the chapter. I shall present three hallmarks of this reading
(section 2.3.1), i.e. its thesis that there are no theses in the Tractatus, its rejec-
tion of the conception of substantial nonsense, and its views as to what, ac-
cording to Wittgenstein, was the primary reason for our failures in philosophy.
In section 2.3.2 I shall respond to the claims of the New Reading. At the centre
of my attention is the discussion as to what, according to Wittgenstein, is the
reason for philosophical nonsense: is it the pursuit to fix the limits of sense (as
the New Reading claims) or is it competing with science in the task of provid-
ing the most comprehensive worldview.

2.3.1  The principles of the resolute interpretation

The most recognisable mark of the resolute interpretation is serious treatment of
the result of the Tractatus, i.e. Wittgenstein’s claim:
My propositions serve as elucidations in the following way: anyone who understands me
eventually recognises them as nonsensical, when he has used them—as steps—to climb
up beyond them. (He must, so to speak, throw away the ladder after he has climbed up
it.) (TLP 6.54).

According to the proponents of the resolute interpretation, Wittgenstein wanted

to communicate in this thesis that “in the end, all the pronouncements of [Witt-
genstein’s] text are just gibberish”248. One could say, in other words, that there is
just one thesis in the Tractatus: that the theses of the Tractatus are nonsensical.
The only meaningful claim are Wittgenstein’s instructions as to how to approach
the book249 or, as Conant puts it, the claim that we are exposed to temptations of
ridiculous questions250. There are no theses, doctrines or even arguments in the
Tractatus. The New Interpretation negates that “Wittgenstein is throughout the
text attempting to provide arguments for controversial and disputable theses”251.
This attitude critically assesses interpretations which say that not only did Witt-

248 Ostrow 2002, p. 5.

249 Morris & Dodd 2009, p. 249.
250 “The only ‘insight’ that a Tractarian elucidation imparts, in the end, is one about the
reader himself: that he is prone to such illusions of thought” (Conant 2000, p. 197).
251 Ostrow 2002, p. 5.

114
genstein consider the conditions for a meaningful proposition, but also that he
gave a good deal of answers (for example, that the reason for the senselessness of
expressions is using names in discord with their roles in logical syntax).
In one place, in section 1.1.3, I claimed that although it is true that, strictly ac-
cording to Tractarian rules, one cannot say that there are objects (strictly speaking
it is nonsense), nevertheless the thesis “There are objects” allows us to see the world
aright. The thesis is a useful tool in the process of acquiring the proper world-
view. We cannot meaningfully express theses about objects (just as Frege could
not meaningfully describe concepts) but, nevertheless, these expressions somehow
transmit what we should think about objects252. This conception – that nonsensical
expressions convey content – is the object of the fierce attack of the resolute inter-
pretation. According to its representatives, it allows for the unacceptable distinc-
tion between mere nonsense and substantial nonsense253, whereas there is only one
kind of nonsense: sentences such as Diamond’s “piggly wiggle tiggle”, which has
no content at all254. The Tractatus, from this point of view, is simply a set of piggly
wiggle tiggles. If one recognises an expression as nonsense, this means that one also
recognises that nothing was said255. One cannot pretend, therefore, that there is an
ineffable reality which cannot be expressed, but in spite of this one can have some
access to it. The fight with the idea of substantial nonsense is the second hallmark
(apart from a serious treating of Wittgenstein’s claim that all theses of the Tractatus
are meaningless) of the resolute interpretation.
According to the proponents of the New Interpretation, if we accept that there
are Tractarian theses which make use of nonsense in order to convey a truth about

252 “The sentences of the Tractatus, though nonsensical, are used by Wittgenstein to
bring us to see the ineffable truths which explain why this is so. Such nonsense sen-
tences communicate truths by getting the reader to grasp the truths lying behind his
words” (Morris & Dodd 2009, p. 249).
253 Conant 2000, p. 176.
254 “When you ascribe to someone the thought that p, this involves you in using a sentence
giving the content of the thought, a sentence that you understand, a sentence of some
or other language that you understand. You are not ascribing a belief to someone if
you say that she believes that piggly wiggle tiggle, if ‘piggly wiggle tiggle’ is nonsense
(although it may be that she has been hypnotized and somehow or other made to think
that when she says ‘piggly wiggle tiggle’ out loud or to herself she is making sense.) If
‘piggly wiggle tiggle’ is nonsense, then ‘Mary thinks that piggly wiggle tiggle’ or ‘Mary
says that piggly wiggle tiggle’ is nonsense” (Diamond 2000a, p. 151).
255 “To recognize a Satz as nonsensical [Unsinn], for the Tractatus, is not a matter of rec-
ognizing that it is attempting to say something that cannot be said, but rather a matter
of recognizing that it fails to say anything at all” (Conant 2000, p. 194).

115
the world, this would mean that we ascribe to Wittgenstein exactly the same mis-
take he fought against256, i.e. that we can take a position outside of a language. In
the opinion of the proponents of the New Reading, it is the position the adherents
of the standard interpretation must hold. According to the standard interpreta-
tions (both anti-metaphysical and metaphysical; phenomenalistic or materialistic),
Wittgenstein in the Tractatus tried to set the limits of meaningful thought; but in
the opinion of the adherents of the resolute interpretation the concept of the limit
of sense suggests that there is something outside the limit since there is something
a thought can express (within the limits of sense) and there is something a thought
cannot express (outside the limits of sense). In the eyes of the proponents of the
New Interpretation, there is nothing a thought cannot express. Either we succeed
in conveying content or we do not succeed in producing a thought at all, whereas
the standard interpretations make use of the idea of something a thought cannot
express. According to the resolute interpretation, there is just one kind of non-
sense, something one could compare with childish babbling, and in this light the
pursue for setting the limits of sense is contradictory and self-refuting:
”Wittgenstein’s point here, and throughout the Tractatus, is that the idea of a limit to
thought is self-undermining in the sense that there is no intelligible way in which one
can be drawn. Doing so, Wittgenstein is saying here in the Preface, would require hav-
ing some grasp of what lies beyond such a limit; indeed, the very suggestion of a limit
implies that there is something which is being excluded, which lies outside of the range
of thought and so cannot be reached” (Cerbone 2000, p. 293).

Under the New Interpretation the aim of Wittgenstein’s criticism, when he says
that, for instance, “most of the propositions and questions to be found in philo-
sophical works are not false but nonsensical” (TLP 4.003) is, therefore, the Frege-
an and Russellian project of fixing the limits of what is meaningful and what is
nonsensical:
In the transparent vacuity of this culminating statement [TLP 6.54] we are meant to
see the vacuity of this Frege/Russell logic, of any attempt to specify a priori the limits of
thought and language (Ostrow 2002, p. 114)

Frege-Russell’s project of the philosophy of language aims to guarantee philoso-

phy a special place in human culture. According to this view, philosophy is the

256 “They [the representatives of the resolute interpretation] charge that standard inter-
pretations in effect represent Wittgenstein as undertaking the very type of metaphysi-
cal project which, even according to the interpretations themselves, he is repudiating”
(Crary 2000, p. 3).

116
most general discipline embracing the other disciplines of science257. Its specific
subject-matter is a region of necessity, in contrast to natural science whose sub-
ject-matter is the sphere of the contingent. It states truths independently of what
the world actually looks like. The New Interpretation rejects this view of philoso-
phy. It accuses this way of philosophising of accepting the groundless assumption
that there is a point of view from which one sees the world as a whole258. Such an
assessment of philosophy (that it unjustifiably assumes the existence of the point
of view from which we can view the world as a whole) results in seeing all of its
traditional discussions as empty. I discussed, for instance, the problem whether
the simples are simple units of experience or material points. These questions, ac-
cording to the resolute interpretation, are nonsensical. It is a dispute between the
realists and the empiricists. In Diamond’s opinion, empiricism is a mirror side of
realism259. Diamond argued that although we know what it means to have a realis-
tic attitude to the world, what it means to be a realist in politics and what realism
in literature consists in260, in philosophy there is a struggle when it comes to un-
derstanding what is contrasted with the realistic standpoint. A realist, for example
Russell, says that he can talk about the external world; the world independent of
my ideas or experiences. He claims that he can cross the limits of the subjective
world of private experiences. An antagonist of a realist says, on the other hand,
that this is impossible. We are imprisoned in our own worlds. Ultimately, in the
opposition to realism, therefore, stands solipsism:
Solipsism rejects the Russellian idea that we can get beyond the “limit of private experi-
ence” but keeps its conception of that limit: it precisely does give us one of what Russell
had given us two of. The solipsist does not rigorously follow out his solipsism; if he did,
it would lead him to a non-Russellian realism. A one-limit view self-destructs (Diamond
2000b, p. 282).

257 “On Frege’s view, logic, as the maximal general science, does not concern itself di-
rectly with cognition and cognizers: its topic – universality establishes a framework
for all of science” (Ricketts 2010, p. 169).
258 “Take the opening sentences of the Tractatus after the preface: ‘The world is all that
is the case. The world is the totality of facts, not of things.’ With those sentences we
imagine a point of view from which we can consider the world as a whole. That idea,
not recognized as an illusion, characterizes the practice of philosophy as it has gone
on” (Diamond 2000a, p. 160).
259 “Empiricism is something we get into in philosophy by trying to be realists but going
about it in the wrong way, or not hard enough” (Diamond 1974–1982/1995, p. 39).
260 Ibid., p. 40–41.

117
According to Diamond’s reading of the Tractatus, one cannot even formulate the
idea of solipsism. This means that one cannot understand the philosophical stand-
point one hoped to make clear by contrasting it with solipsism, i.e. the standpoint
of realism. It turns out that the discussion between both rivalling interpretations
of the Tractatus, i.e. the phenomenalistic and materialistic one, is nothing more
than idle chatter261. It seems that it is impossible to find more contrasting views
than realism and solipsism, but under closer scrutiny it turns out that both of
these standpoints share the same conviction about the existence of a limit be-
tween the subjective and objective sphere of reality. According to Diamond, an
opinion with respect to one’s ability to cross the limits of the subjective world is
secondary as compared to the fact that neither of these two standpoints can make
sense of the idea of a limit between what is subjective and objective. Summing
up, the third feature of the resolute interpretation is a view on which the Tractaus’
criticism of philosophy was aimed at a philosophy that sets limits, i.e. between
what is meaningful and nonsensical, or between what is subjective and objective.

2.3.2  The notion of philosophy in the Tractatus

As we can see, the New Interpretation in a radical way (1) rejects the supposition
that there are doctrines in the Tractatus, (2) rejects the conception of substantial
nonsense, and (3) attacks the project of setting the limits of meaningful discourse.
How could one answer the challenge of the resolute interpretation? I cannot as-
pire to do this better and in a more comprehensive way than Peter Hacker or Ian
Proops did, who, on the basis of Wittgenstein’s writings, proved that he could
not hold the position that the resolute interpretation ascribes to him262. All his-
torical evidence (Wittgenstein’s letters, later lectures, notes, records of discussions
with other philosophers), meticulously listed by the aforementioned authors263,
suggests that Wittgenstein himself held the Tractatus to be a book containing
substantive philosophical doctrines. I shall just concentrate on the argument that
the thesis that there is just one claim in the Tractatus, i.e. that the theses of the
Tractatus are nonsensical, is untenable.
Such authors as Hacker, Proops, Child or Horwich point out that if the resolute
interpretation sees the theses of the Tractatus as nonsensical, it has to assume that

261 “There can be no question (…) of adopting a standpoint from which we can, once for
all, assess the ‘real character’ of our utterances about objects” (Ostrow 2002, p. 77).
262 Hacker 2000; Proops 2001. Other important responses to the resolute interpretation:
McGuinness 2012; Sullivan 2004.
263 Proops 2001, pp. 382–395; Hacker 2000, pp. 360–382.

118
we understand that they are nonsensical: a reader of the Tractatus, according to the
proponents of the New Reading, is led through different tempting philosophical
positions up to the point of becoming aware of the futility of philosophy and, final-
ly, to the decision of throwing it away. The rejection of philosophy entails that one
becomes aware of its nonsense. It assumes that at some point one understands that
the claims of philosophy are nothing other than “piggly wiggle tiggle”264. But the
same proponents of the New Reading say that there is just one kind of nonsense
in which there is no content to grasp because nonsense fails to produce a thought,
and yet they describe the intentions of the Tractatus as assuming that its reader will
realise that its theses are nonsensical. But if one sticks to the theory of just one kind
of nonsense, it seems that the reader of the Tractatus cannot become aware that
its theses are nonsense. After all, if one is confronted with “piggly wiggle tiggle”,
he does not have to realise that it is nonsense. Exactly because there is no content
to grasp, there can be no process of reading, seeming to understand, and realising
that there was nothing to understand. Clearly, the proponents of the New Reading,
trying to stick to both a serious reading of TLP 6.54 as well as to the theory of one
kind of nonsense, are tangled up in hopeless contradictions.
According to the standard interpretations, the problem of “throwing away the
ladder” is easy to solve. Wittgenstein, under this reading, tries to set the limit of
sense and then, on the basis of the theory of proposition defended by the Tracta-
tus, one rejects some linguistic expressions as not fulfilling the requirements put
on propositions. This explanation is inaccessible to the resolute interpretation,
which claims that there are no doctrines in the Tractatus. Moreover, it claims that
the project of setting the limits of sense is alone the main source of philosophical
nonsense. It makes the position of the resolute interpretation even more perplex-
ing. If fixing the limits of sense is senseless, how could one possibly know that the
theses of the Tractatus are nonsensical?265
In order to respond to such arguments, Morris and Dodd tried to improve the
resolute interpretation. They called the Diamond-Conant interpretation “The Not-
All-Nonsense View” because it says that there is only one claim in the Tractatus.

264 “It would seem rather that, although there are not supposed to be any fully cogent
formulations of those ideas, still Wittgenstein’s formulations do somehow manage
to steer us in the right direction” (Horwich 2012, p. 94).
265 “On the conventional view, the nonsensicality of the Tractatus’ own propositions
follows directly from the positive theory of language that Wittgenstein is advancing.
But if we insist that the Tractatus embodies no positive theory of language at all,
there seems nothing to say about why Wittgenstein judges his own propositions to
be nonsensical” (Child 2011, p. 71).

119
The Not-All-Nonsense View assumes the existence of a frame in the Tractatus. To
the frame belong, allegedly, the Preface and thesis TLP 6.54. The frame should be
read as containing genuine propositions which guide a reader through the book.
It is The Not- All-Nonsense View that is entangled, in the opinion of Morris and
Dodd, in the aforementioned problems as to how one can know that the theses of
the Tractatus are nonsensical since there is no positive theory of language in the
book266. In contrast to this version of the resolute interpretation, Morris and Dodd
proposed “The No-Truth-At-All View”, which says that Wittgenstein did not want
to communicate any truth whatsoever. He even did not want to say that the theses
of the Tractatus are nonsensical267. I am not sure if Morris and Dodd make any
favours to the resolute interpretation. The conclusion of their view amounts to
the claim that there is no difference between the Tractatus and a blank sheet of
paper. After all, a blank sheet of paper conveys no truths either. They would obvi-
ously not agree with this criticism. According to them, the text of the Tractatus
has a certain aim. This goal cannot be to convey truths (since there are none in
the text) but it is “to bring us to adopt another perspective on life altogether; and
this other perspective, we suggest, is the perspective of mysticism”268. Even if one
accepts this explanation and agrees that The No-Truth-At-All View does not re-
duce the Tractatus to a blank sheet of paper, the following problem arises: “What
is the difference between The No-Truth-At-All View and Conant-Diamond’s The
Not-All-Nonsense View?”. Morris and Dodd could be asked how they know that
Wittgenstein wanted us to adopt a mystical point of view since there are alleg-
edly no truths in the Tractatus, just like Conant and Diamond are asked how they
know that the theses of the Tractatus are nonsensical if there is no doctrine there?
In the end, the proponents of the New Reading themselves cannot be faithful to
the principles of their interpretation. Hacker, and rightly so, indicates that Conant
and Diamond do not hold all the theses of the Tractatus as gibberish. When we
read their papers we have a feeling that they are quoting theses TLP 4.126-4.1272,
TLP 5.437 or TLP 5.25 with approval, as if these theses contain a true doctrine.
They agree with the author of these claims and refer to these theses in order to

266 “The Not All-Nonsense View simply does not have the resources to explain why it
should matter to Wittgenstein that nonsense comes to be nonsense for the particular
reason he provides” (Morris & Dodd 2009, p. 256).
267 Ibid., p. 251.
268 Ibid., p. 261.

120
justify their views269. This is a strange tactics for commentators who should rather
say that these theses did not succeed in formulating a thought.
I shall move to the second hallmark of the resolute interpretation, i.e. the claim
that there is just one kind of nonsense. In my opinion, the reason for the aforemen-
tioned inconsistency with respect to the claim that there is just one thesis in the
Tractatus could be reduced to the fact that in trying to understand the Tractatus
one cannot avoid the distinction between substantial and mere nonsense. The pro-
ponents of the New Reading are correct when they see the danger of abusing the
doctrine of substantial nonsense. It suggests that there is an ineffable reality (since
substantial nonsense is about something we cannot put into words) and that it
could lead to some groundless conclusions such as, for instance, that Wittgenstein
believed in the existence of the transcendental self. As we will see in Chapter 4, I
reject usage of the doctrine of substantial nonsense with respect to the existence
of the self. I am also against associations of the doctrine of substantial nonsense
with mysticism, but this does not mean that we have to throw the baby out with
the bathwater and negate any hint of the existence of the doctrine of substantial
nonsense in the Tractatus. One can prevent the abuse of the usage of this doctrine
simply by stating that, according to Wittgenstein, nonsense is substantial when we
can show what it is triysing to say by means of other propositions. For example,
when one says that there are objects in one’s room, we can show what this non-
sense is trying to say by saying, for instance, that there are chairs in one’s room270.
One shows what the expression: “Dark blue is darker than light blue” is trying to
say by means of propositions in which one ascribes to the dark blue colour a cer-
tain amount of shades-units (for instance, 5), and to light blue another amount of
shades-units (for instance, 3). And, using the Tractarian example, “what the axiom
of infinity is intended to say would express itself in language through the existence
of infinitely many names with different meanings” (TLP 5.535).
As we can see, we have convincing evidence that Wittgenstein agreed that not
every nonsense is gibberish. Sometimes we manage to show what nonsense tries to

269 “Diamond holds that Wittgenstein really did think that the signs ‘p’ and ‘~p’ can say
the same thing (Tractatus 4.0621), that his criticisms of Frege in 4.063 are not ‘plain
nonsense’ but genuine, powerful criticisms, as are his criticisms of Russell’s theory
of judgement. With this one must agree, but wonder whether this is not a case of
trying to have one’s cake and eat it” (Hacker 2000, p. 361).
270 “What you want to say by the apparent proposition ‘there are 2 things’ is shown by
there being two names which have different meanings” (Letters, p. 126).

121
say, and in this case nonsense is elucidatory271. Additionally, accepting the doctrine
of substantial nonsense can free us from questions such as: “If all nonsense is like
childish babbling, then how can one come to the conclusion that the theses of the
Tractatus should be rejected?” After all, it is impossible to understand childish bab-
bling”, I do not see any reason to reject this doctrine. Its acceptance allows to out-
line the following dialectics of the Tractatus. The Tractatus sets the limits of sense.
In the previous chapters I discussed what are, in my opinion, the two requirements
put forward by Wittgenstein regarding a meaningful proposition. First, it has to
consist of names with referents, and secondly, all constituents of a proposition
have to occur in it according to their roles in logical syntax. On the other hand, on
the basis of the Tractarian theory of sense, the limiting of sense is nonsensical. We
saw the reason for this: expressing the requirements put on a meaningful proposi-
tion demands using formal concepts, which in the ideal logical notation would be
represented by the variables, and this means that they cannot occur in the fully
analysed, meaningful propositions (because the latter are concatenations of real
names). This does not mean, however, that we should assess the whole project of
fixing the frame of meaningful discourse as rubbish, because theses which express
the limits of sense belong to the category of substantial nonsense.
At last I tackle the third claim of the resolute interpretation, i.e. that for early
Wittgenstein the primary examples of philosophical nonsense were the views
of Russell and Frege. I think that in claiming this the New Reading does not
take into account the historical background of the Tractatus. According to the
resolute interpretation, early Wittgenstein in his criticism of philosophy targets,
above all, the philosophy of Bertrand Russell and Gottlob Frege. That seems,
to me, highly unlikely. In the summer of 1911 Wittgenstein travelled to Jena
to meet Frege. He hoped that this meeting would help him decide if he should
begin work in philosophy or if he should continue his work in aeronautics.
Despite the fact that Frege, according to Wittgenstein’s own remark, “wiped the
floor with him”, he decided to develop his new interest272. He met Russell, and
that meeting was decisive for Wittgenstein’s later career273. If Frege and Russell
represented all “evil” and “delusions” of philosophy, young Wittgenstein would
not be so impressed by their thought, he would not change his entire life, and he

271 “It might help to suppose on Wittgenstein’s behalf that his ‘senseless propositions’
should not be equated with ‘gibberish’, but should be taken to involve some less
extreme defect – one that does not preclude their somehow being ‘elucidatory’ ”
(Horwich 2012, p. 91–92).
272 Monk 1991, p. 36–37.
273 Waugh 2010, p. 71.

122
would not give up his engineering studies. It is safe to say that there were rather
the modern tools of logic as developed by Russell and Frege which helped to
fight the philosophical nonsense of idealism that attracted young Wittgenstein.
In these personas he found mentors who could help him express criticism
against the traditional philosophy he knew, mainly from Schopenhauer’s books.
At least this is the picture of philosophical development (from Schopenhauer’s
idealism to Russell’s realism) we find in Wittgenstein’s Notebooks:
This is the way I have travelled: Idealism singles men out from the world as unique, sol-
ipsism singles me alone out, and at last I see that I too belong with the rest of the world,
and so on the one side nothing is left over, and on the other side, as unique, the world.
In this way idealism leads to realism if it is strictly thought out (NB 15.10.16, p. 85).

I agree with the representatives of the resolute interpretation about the negative
attitude of the Tractatus towards philosophy. I do not agree, however, with the
idea of what the aim of Tractarian criticism was. Diamond and Conant would in-
dicate at Frege and Russell, with their goal of setting the limits of sense. They can
answer to the biographical facts I have mentioned that it is comprehensible that
by the time of finishing or publishing the Tractatus and having worked for many
years with Russell, Wittgenstein came to the conclusion that his initial excitement
about Russell’s and Frege’s writings was groundless; that, in fact, these writings
are no more meaningful than the oeuvres of Kant, Hegel or Schopenhauer. I do
not think, however, that this possible response withstands the textual evidence
of the Tractatus. The proponents of the resolute interpretation claim that in the
eyes of early Wittgenstein the scandal of philosophy consists in a project of fixing
the limits of sense. But in the Tractatus we find entries that state that exactly this
should be the task of philosophy:
[Philosophy] must set limits to what can be thought; and, in doing so, to what cannot
be thought. It must set limits to what cannot be thought by working outwards through
what can be thought. It will signify what cannot be said, by presenting clearly what can
be said (TLP 4.114-4.115).

The reason for Wittgenstein’s criticism with respect to philosophy has to be dif-
ferent. In my opinion, and I shall analyse this in the next two chapters in order
to prove the claim, for Wittgenstein the example of nonsense in philosophy is
ontotheology. By that I mean, following Putnam’s description, philosophy which
searches for proofs for the existence of God, an immortal soul or free will274. In a

274 “A reason often given for the contemporary question concerning the need of or
role for philosophy is that philosophy for so long (…) was heavily interested in two

123
broader sense the object of Wittgenstein’s critique is a theoretical philosophy, i.e.
a philosophy which formulates doctrines, theories and claims that aspire to rival,
contest or complete the theories and claims of natural science. It is a philosophy
which – with all its proofs, argumentations, appearances of precision and own
method – is disguised as science. It is a philosophy which claims that there are
phenomena whose study is its special business. An example of such a philosophy
could be Cartesian dualism (which aims to prove the existence of the immortal
soul and God only on the basis of speculation) or the metaphysics of will by
Schopenhauer (which tries to oppose the results of mechanics, physics or biology
by delivering a broader and fuller picture of the world which includes the reality
of things-in-themselves). In the Tractatus, philosophy is contrasted with natural
science. The danger to philosophy, in the eyes of the author of the Tractatus, con-
sists in its constant trying to be like science. What is fatal for philosophers is their
trying to compete with scientists in their task of explaining events in the world.
Philosophy is not equipped for this task – it does not have a special domain of
reality to investigate and, moreover, its traditional concepts, such as the concepts
of self, essence, object, and so on, are empty. If the philosophical work of Russell
suits the description of ontotheology, then it is also targeted by the Tractatus, but
not because of the project of demarcating the meaningful and nonsensical ex-
pressions. If we assume that it is the theoretical philosophy which is the subject of
Wittgenstein’s criticism, then Tractarian comments on philosophy become clear:
(1) philosophy is not science; philosophy does not compete with the natural
sciences – it is not “beside” them (TLP 4.111).
(2) philosophy does not formulate any doctrines. The result of philosophical ac-
tivities are not any claims: “Philosophy is not a body of doctrine but an activ-
ity” (TLP 4.112).
(3) there is no special domain of philosophy. A philosopher is not allowed to say
anything more than the natural sciences do (TLP 6.53).
I owe an explanation to point (2). Earlier I stated that Wittgenstein had in mind
the answer to the question: “what exists?”. How could one reconcile this view with
the conclusion that philosophy should not contrive any theories? The materialis-
tic interpretation simply states that it is not the matter of philosophy to fix what
objects exist. In this sense philosophy indeed does not produce its own doctrines.
Nevertheless, it indicates where one should search for meaningful answers when

‘ontotheological’ ideas, namely, (1) the idea of God (…), and (2) the idea of the im-
materiality of the soul” (Putnam 2010, p. 39).

124
one is interested in these kinds of questions. The Tractatus indicates natural sci-
ence. In this way one could ascribe to the book certain metaphysical convictions –
one can label them as ontological materialism – and simultaneously one can give
justice to what Wittgenstein writes about the limits of philosophy. One could
respond that this answer of the proponent of the materialistic interpretation is
still unsatisfactory because materialism is a concrete philosophical doctrine, and
Wittgenstein states in the Tractatus that philosophy contains no doctrines and
claims. How does this view stand against the materialistic interpretation of the
Tractatus? Does it contradict my views? I would answer negatively. I sustain both
that, according to early Wittgenstein, philosophy is not a theory and that in the
Tractatus he represented materialism. The solution of this paradox lies in the fact
that the claim: “Objects are material points” is, strictly speaking, not a proposi-
tion but nonsense. It is substantial nonsense which allows us to see the world
properly. One could say that Wittgenstein’s philosophy contains no thesis of ma-
terialism, nevertheless the Tractatus allows us to see what the world consists of.
Summing up, in the last passages (section 2.3) I presented the main assump-
tions and the biggest problems of the resolute interpretation. This New Reading of
the Tractatus stands in sharp opposition to my reading, mainly because I argue in
favour of the doctrine of substantial nonsense. It is true that formulations such as:
“Simple names refer to material points” are nonsensical, but because of them one
gains insight into the nature of the world. On the other hand, the fact that Tractar-
ian requirements put on meaningful propositions can be expressed only nonsensi-
cally allows Wittgenstein to consistently claim that philosophy should not contain
any doctrines and theories. The proponents of the resolute interpretation are seri-
ously challenged by the numerous pieces of textual evidence from Wittgenstein’s
writings. In my opinion, Proops is correct when he writes that “until this challenge
is answered – and answered convincingly – the historical credibility of the New
Reading will remain in doubt”275.

Summary
In the last chapter I analysed three of the possible understandings of the Tractatus
(phenomenalistic, materialistic and resolute). In section 2.1 I discussed the claim
according to which the simples of the Tractatus are simple units of experience.
This interpretation refers to the common ontological opinions that prevailed in
the intellectual environment of young Wittgenstein – especially to the views of

275 Proops 2001, p. 398.

125
Russell and the fact that, at least at first sight, the Tractatus confirms the truth of
solipsism. Some of the recent publications, however, point to a different source of
inspiration for the metaphysical views of early Wittgenstein. According to them,
one should dig deeper into the engineering education of the Austrian philoso-
pher and his fascination with Hertzian mechanics. This fascination is displayed
in the Tractatus, mainly in the way in which Wittgenstein analysed the language
of sensations. His solution to the colour-exclusion problem (crucial with respect
to sustaining the fundamental distinction between contingent and logical truths)
betrays that he did not hold the language of sense-data to be the most basic lan-
guage as, for instance, Russell did. On the contrary, in order to save the afore-
mentioned distinction between contingent and logical truths, one has to treat
colours in accordance with the physical analysis, i.e. as sensations caused by ether
parts oscillating with a certain velocity. That is why in this chapter I advocated
in favour of the materialistic interpretation of the Tractatus (section 2.2). It states
that, according to this book, the world consists of the most elementary particles
of matter. Wittgenstein could find a good candidacy for these particles in the no-
tion of the mass-particle, taken from Hertzian mechanics. I think that one of the
advantages of the materialistic interpretation is that it best answers the question
I posed at the beginning of this work: “Why did Wittgenstein reduce meaningful
discourse to scientific discourse?” According to the materialistic interpretation,
all true elementary propositions contain names which ultimately refer to physi-
cal particles, exactly like the propositions of physics, hence meaningful language
should be identified with the language of physics. The materialistic interpretation
gives an answer to the questions avoided by the proponents of other interpreta-
tions: Why did Wittgenstein in fact reduce true propositions to propositions of
natural science? Why did he think that a proper method in philosophy should
consist in repeating what scientists say about the world (TLP 6.53)? If one looks
at the world as consisting of objects that are primarily physical entities that just
happen to have certain essences, then one has to conclude that all questions about
the world are of a scientific nature276; that the so-called “problems of life” (to the
problems of life one could include the question of moral values, aesthetic values
(TLP 6.421), the immortal soul, or the existence of God (TLP 6.4312)) one can-
not even express277, and this inability indicates a solution to these problems:

276 “When all possible scientific questions have been answered (…) then there would be
no questions left, and this itself is the answer” (TLP 6.52).
277 “When the answer cannot be put into words, neither can the question be put into
words. The riddle does not exist” (TLP 6.5).

126
 The solution of the problem of life is seen in the vanishing of the problem (TLP 6.521).

The materialistic interpretation states that the ontology of the Tractatus repre-
sents strong or radical naturalism:
1) it consists only of a space-time and point particles and rejects the existence
of natural laws or the object of common sense and folk psychology. In order
to describe reality it is enough to determine the position of a particle at a
given time278.
2) it makes science the arbiter of truth. It assigns a cognitive value to science
alone.
This means that Wittgenstein, despite his reluctance to the scientific problems279,
was considering the problems of life from the position of a scientist, i.e. somebody
who acknowledges the role of science in describing and explaining reality and
does not resort to a strategy consisting in downgrading its meaning or proclaim-
ing some kind of irrationalism. In my opinion, this position, taking into account
that one cannot classify Wittgenstein as a positivist, makes his views on ethics
particularly interesting.
In the last section of the chapter (section 2.3) I tackled problems posed by the
resolute interpretation, which accuses the anti-metaphysical, phenomenalistic
and materialistic readings of the Tractatus of a fundamental misunderstanding
of the book. All of these interpretations, in the eyes of the proponents of the New
Reading, forget that for Wittgenstein the problem of philosophy consisted in its
hopeless project to set the limits of sense. Allegedly, this project assumes that one
can take a point of view “outside” a language, i.e. from which one judges which
expressions are meaningful and which are not. On the resolute interpretation
this assumption is not only senseless in itself but was also the subject of a fierce
attack in the Tractatus. In contrast to this attempt to apprehend Wittgenstein’s
views on philosophy, I claim, together with the so-called standard interpreta-
tions, that to set the limits of meaningful discourse was precisely the goal of
early Wittgenstein’s philosophy. Philosophy fails when it tries to be like science,
when it perceives itself as rivalling with science in the task of describing the

278 Perhaps one should also specify, as Barry Loewer suggests, a momentum of a particle
and forces acting upon a particle: “The state of the universe at a time t [under the ontol-
ogy of classical mechanics] is given by a specification of the positions and momenta of
the particles at t and a specification of the forces on the particles” (Loewer 2008, p. 154).
279 As Carnap recalls: “All of us in the Circle had a lively interest in science and math-
ematics. In contrast to this, Wittgenstein seems to look upon these fields with an
indifference and sometimes even contempt” (Carnap 1964/1967, p. 37).

127
world. I labelled this kind of philosophy ontotheology, and I mentioned in this
context dualism of mind and matter and transcendental philosophy. In the next
two chapters I will show that Wittgenstein indeed combatted the dualistic and
the transcendental notion of the self.

128
Chapter 3.  Wittgenstein’s Theory of Judgement

In the previous chapter I argued in favour of the materialistic interpretation ac-

cording to which simple objects of the Tractatus are physical atoms. The claim
that Wittgenstein was a proponent of physicalism (“only not under that horrible
name”) could be surprising, especially when one takes into account his scepti-
cism about science: “In order to marvel human beings – and perhaps peoples –
have to wake up. Science is a way of sending them off to sleep again” (CV, p. 6).
It is true that Wittgenstein’s aversion towards science grew over the years280, but
already in the Tractatus he displayed a certain amount of sensitivity towards, as
he put it, “the mystical” or “the problems of life”281, which we would search for in
the works of the neo-positivists in vain. How could one explain the existence of
these two contradictory tendencies in the thinking of young Wittgenstein?
In my view, there are two morals one should draw from the fact that such a
sceptic philosopher towards scientism such as Wittgenstein, nevertheless, ad-
vocated in favour of ontological materialism. First, it means that Wittgenstein
acknowledged the ontological authority of natural science. By that I understand
the fact that, in Wittgenstein’s view, there are the natural sciences which answer
the question: “What is there?”. It is the task of physics to determine what the
simple objects are. In other words, in Wittgenstein’s eyes an honest philosopher,
when asked: “What is there?”, should send the asking person to a physicist.
Early Wittgenstein was genuinely interested in the problems of life: in his Note-
books he asks about the moral value of suicide, the meaning of belief in God, or
what a happy life consists in282. The fact that Wittgenstein accepted physicalism in

280 He, for instance, criticised his Tractarian thesis that all true propositions are proposi-
tions of natural science: “Now why am I so anxious to keep apart these ways of using
‘declarative sentences’? an attempt to see that every usage gets its due. Perhaps then
a reaction against the overestimation of science. The use of the word ‘science’ for
‘everything that can be said without nonsense’ already betrays this over-estimation”
(CV, p. 70).
281 “There are, indeed, things that cannot be put into words. They make themselves
manifest. They are what is mystical” (TLP 6.522); “We feel that even when all pos-
sible scientific questions have been answered, the problems of life remain completely
untouched” (TLP 6.52).
282 “To believe in a God means to see that the facts of the world are not the end of the
matter. To believe in God means to see that life has a meaning” (NB 8.7.16, p. 74); “In
order to live happily I must be in agreement with the world” (NB 8.7.16, p. 75); “If

129
the Tractatus means that in solving the aforementioned problems he did not want
to fall back to the strategy according to which although one admits that science is
making great progress in the task of describing the world, and on the basis of sci-
entific research one can comparatively make the most precise predictions about
future events, but nevertheless one claims that not every aspect of reality is ac-
cessible to scientific research; there are some aspects of human life where science
is of no avail, and exactly with respect to these aspects of life one should turn to
metaphysics, religion, ethics, or poetry. That would be the easy way out from the
challenge of the modern development of science and, apparently, Wittgenstein
did not want to use it. He accepted the modernist worldview according to which
only scientists are able to tell us what the world looks like. In this respect he is in
agreement with the representatives of scientism. Taking into account his interest
in “the problems of life” and in the defence of so-called “human values”, it makes
his solutions with respect to these issues even more interesting.
The second moral one could draw from Tractarian physicalism refers to Witt-
genstein’s ideas about the role of philosophy. If the ontological authority is due to
natural science, this means that there is no separate region of reality which could
be the domain of philosophy (such as, for instance, the domain of necessary
truths, or the domain of essences). Philosophy is helpless in answering tradi-
tional metaphysical questions or in solving the problems of life. I think that it is
essential for early Wittgenstein’s worldview that in the age of rapid development
of scientific research, philosophy cannot pretend to be a branch of science; it will
always lose the competition283. In the next two chapters of this dissertation I am
going to defend this interpretation of Wittgenstein’s views on philosophy.
At the end of the Tractatus Wittgenstein states that a possible proof for the ex-
istence of an immortal soul “completely fails to accomplish the purpose for which
it has always been intended” (TLP 6.4312), and that the question of God’s exist-
ence has nothing to do with what the world looks like284. There is a close analogy,
in Wittgenstein’s view, between Christianity and philosophy. Just as Christian-
ity, philosophy should not be a doctrine about the world on the basis of which

suicide is allowed then everything is allowed. If anything is not allowed then suicide
is not allowed. This throws a light on the nature of ethics, for suicide is, so to speak,
the elementary sin” (NB 10.1.17, p. 91).
283 In the Notebooks, Wittgenstein expressed this conviction by saying about Russell’s
philosophy: “Russell’s method in his ‘Scientific method in philosophy’ is simply a
retrogression from the method of physics” (NB 1.5.15, p. 44).
284 “How things are in the world is a matter of complete indifference for what is higher.
God does not reveal himself in the world” (TLP 6.432)

130
one could explain past events and predict future ones285. And just as Christianity
would revolt against its own vocation if it was just a theory, so philosophy would
rebel against its basic task of defending the world of spiritual values against the
claims of natural science286 if it was a theory. It would become a cold wisdom that
in fact would kill what is truly important287. The morals one should draw from
materialism of the Tractatus is, therefore, the conclusion that it was ontotheology
that was the object of Wittgenstein’s attack on philosophy. In his opinion, such a
philosophy would not only be a “bad science” but it would also wrongly serve the
aim of defending the spiritual values of the human world. The most explicit of
Wittgenstein’s arguments against metaphysics is that it uses empty names:
The correct method in philosophy would really be the following (…): whenever some-
one else wanted to say something metaphysical, to demonstrate to him that he had failed
to give a meaning to certain signs in his propositions (TLP 6.53).

Earlier I indicated Cartesian dualism and Schopenhauer’s metaphysics of will as

examples of the wrong way of doing philosophy. In the next two chapters I shall
argue that, indeed, in the Tractatus such a criticism takes place; and that indeed
philosophy which makes use of concepts such as “the empirical self ”, “mind” or
“the transcendental self ” fails to give meaning to these concepts. In Chapter 3
I shall argue that early Wittgenstein rejected the dualism of mind and matter.
This rejection took place in the context of Wittgenstein’s discussion with Rus-
sell’s theory of judgement which, as we will see, defended dualistic positions.
Then, in Chapter 4 I shall argue that Wittgenstein rejected the notion of the
transcendental self; the notion that is connected with the philosophy of Scho-
penhauer (it is almost certain that it got to Wittgenstein through Schopenhauer’s
The World as Will and Representation), where it served, as I shall try to show, to
safeguard moral values from the claims of a purely scientific worldview. It fol-
lows then that Wittgenstein’s critique of the notion of the transcendental self is
equivalent to the critique of safeguarding values the way Schopenhauer did. The
conclusion of the subsequent chapters will be: if human culture is endangered by
the progress of natural science, it is not theoretical philosophy that can defend
it. There has to be a different strategy for an adversary of scientism in his or her
pursuit to defend non-scientific worldview.

285 “Christianity is not a doctrine; I mean, not a theory about what has happened and
will happen with the human soul, but a description of an actual occurrence in human
life” (CV, p. 28).
286 “Philosophy sets limits to the much disputed sphere of natural science” (TLP 4.113).
287 CV, p. 53–56.

131
 The subject-matter of this chapter are theses TLP 5.54-5.5422. In this fragment
Wittgenstein aims to reconcile the strong extensionality thesis (the view according
to which every meaningful sentence is a truth-function on an elementary sen-
tence) with the existence of propositional attitudes such as “A believes that p”, in
which the modal operator B occurs, which indicates non-extensional contexts in
language. Simplifying, Wittgenstein’s solution, as I shall show, consists in the con-
viction that the problem of the alleged incoherence between the strong extension-
ality thesis and the propositional attitudes arises when, one, just like Russell did in
his theory of judgement, assumes the existence of the Cartesian self. Once one gets
rid of this metaphysical notion and shows that in a propositional attitude no mean-
ing was given to the notion of self (or mind), the problem disappears. The solution
has a naturalistic and an anti-metaphysical overtone, and there is much truth in
Bergman’s claim that in TLP 5.542 Wittgenstein took “the decisive step towards
materialism”288, with the reservation that Wittgenstein in the Tractatus did not take
steps towards materialism – I do not agree that problems with Russell’s theory of
judgement led Wittgenstein to a rejection of the dualism of mind and matter. Ma-
terialism was rather a tacit assumption right from the beginning of the first pages
of the Tractatus, and the problems discussed in theses TLP 5.54-5.5422 are simply
another confirmation of the rightness of the materialistic interpretation. Wittgen-
stein’s very strong conclusion that “there is no such thing as the subject that thinks
or entertains ideas” (TLP 5.631) supports such a reading. The failure of Russell’s
theory of judgement alone would not entitle Wittgenstein to state such a thesis, so
under my hypothesis thesis TLP 5.631 is not the conclusion of the discussion with
Russell but rather its background and assumption.
As I have said, theses TLP 5.54ff have an anti-dualistic overtone. Because of
that many commentators – among others Peter Hacker and David Stern – think
that Wittgenstein’s argument is a modern version of the Humean argument that
in the stream of consciousness one cannot find an idea of the self (or if one does
not like the sense-data language that among things there is no such thing as the
self or mind). What, in my opinion, is distinctive in my interpretation is the
claim that Wittgenstein in TLP 5.54ff tried to avoid jumping from the frying pan
of Russellian mind-matter dualism into the fire of Humean positivism. I find the
argument for such a suspicion in Wittgenstein’s interjection that “a composite
soul would no longer be a soul” (TLP 5.5421). One of the most tempting con-
clusions from the thesis that there is no immaterial subject of ideas, beliefs, de-
sires, etc. would be the claim that there are simply ideas, beliefs, desires, etc. and,

288 Bergmann 1973, p. 354.

132
therefore, what the dualists call “the soul” is nothing more than a complex array
of physical or psychic objects. That opens the way to such reductionist claims “as
the soul is nothing more than a flow of neurons in a brain”. But in his repudia-
tion of a complex subject Wittgenstein clearly rejected this idea. The reason for
this one finds, I think, in his project of setting the limits for natural science. This
attitude, in my opinion, differentiates his position from positivism and is the
reason why I call the Tractatus a modernist rather than a positivistic book. I shall
develop these ideas later in the chapter. Summing up, in what follows I want to
defend the following claims:
–– Theses TLP 5.54-5.422 are a discussion with Russell’s theory of judgement.
Wittgenstein reconciles the strong extensionality thesis with propositional at-
titudes by rejecting the existence of what I call the empirical self (I shall later
provide a description of what I mean by such a term)
–– Wittgenstein’s solution confirms his naturalistic and anti-metaphysical attitude.
–– The overtone of the conclusions is not only anti-dualistic but also anti-positiv-
istic. Theses TLP 5.54-5.5422 would confirm my modernist interpretation of
the Tractatus.
The plan of the chapter is the following: first, I shall present Russell’s theory of
judgement since it is the background of theses TLP 5.54-5.5422 (section 3.1.1).
I shall also characterise the concept of the self as assumed by Russell’s theory (with
respect to this concept I will use terms such as: “the empirical self ” and “mind”)
(section 3.1.2). In the next sections (3.2–3.3) I shall prove that Wittgenstein’s own
theory of judgement does without the empirical self, and this allows him to avoid
incoherence between the strong extensional thesis and the propositional attitudes.
Finally, I shall compare my interpretation with those occurring in various com-
mentaries (section 3.4), especially in those which put an emphasis on the con-
text of Wittgenstein’s trial to save the extensionality thesis (section 3.4.1), those
which in the anti-Cartesian overtone of TLP 5.54ff see the neo-Humean argument
against the self (section 3.4.2), and those which explain TLP 5.54ff by Wittgen-
stein’s alleged lack of interest in psychology (section 3.4.3).

3.1  The context of Wittgenstein’s theory

Let me start by quoting a relevant fragment of the Tractatus:
5.54 In the general propositional form propositions occur in other propositions only as
bases of truth-operations.
5.541 At first sight it looks as if it were also possible for one proposition to occur in an-
other in a different way. Particularly with certain forms of proposition in psychology, such

133
 as ‘A believes that p is the case’ and ‘A has the thought p’, etc. For if these are considered
superficially, it looks as if the proposition p stood in some kind of relation to an object A.
(And in modern theory of knowledge (Russell, Moore, etc.) these propositions have actu-
ally been construed in this way.)
5.542 It is clear, however, that ‘A believes that p’, ‘A has the thought p’, and ‘A says p’ are
of the form ‘ “p” says p’: and this does not involve a correlation of a fact with an object,
but rather the correlation of facts by means of the correlation of their objects.
5.5421 This shows too that there is no such thing as the soul—the subject, etc.—as it is
conceived in the superficial psychology of the present day. Indeed a composite soul would
no longer be a soul
5.5422 The correct explanation of the form of the proposition, ‘A makes the judgement
p’, must show that it is impossible for a judgement to be a piece of nonsense. (Russell’s
theory does not satisfy this requirement.).

There are two concepts in the text above which demand an explanation: the con-
cept of the self (the subject), identified in the tradition of empiricism with the
mind, and “the modern theory of knowledge”. Wittgenstein meant by the latter the
part of Russell’s theory which refers to judgements. Therefore, the next two sec-
tions will be devoted to Russell’s theory of judgement and, later, to the concept of
the self as assumed by this theory.

3.1.1  Russell’s theory of judgement

The problem faced by Russell reads as follows: “What is the difference between
a proposition and a mere collection of names?”289. His answer consisted in the
conviction that in the former case there is some sort of unity, whereas the latter
case lacks it. In a judgement there is something that unites its various objects, and
in this way Russell introduced the thinker, a factor uniting a proposition, and its
supplement, which by the act of judging completes the incomplete symbol (i.e. a
proposition)290. In the following section I shall present the views Russell adopted

289 Morris 2008, p. 83; Puhl 1999, p. 25.

290 “Some context is necessary before the phrasing expressing a proposition acquires a
complete meaning” (Russell 1913/1984, p. 109). Johnston thinks that Russell’s was
not the unity of a thought but the unity in a thought. In other words, Russell was not
interested in a subject that unites different elements of a proposition in a judgement
but in the synthesis of a proposition (Johnston 2012, p. 26). I do not agree with this
account precisely because Russell treated a proposition as an incomplete symbol.
Johnston treats a proposition as a complex (of terms), and it seems that he allows, in
contradiction to Russell, a proposition to have a sense on its own. Secondly, even if a
form of a proposition is responsible for a unity within a proposition, there is still the

134
in 1913, because these are the views which Wittgenstein was confronted with.
It is clear that over the years Russell developed various versions of the theory of
judgement and, probably around 1918, he abandoned dualism291.
This general characteristic of Russell’s position needs to be developed. I shall
begin with a clarification of the most important notions used in the text. Russell
held the concept of a judgement and of a belief to be synonymous, and he defined
them as a relation between a subject and a proposition. A judgement/belief is a
certain attitude of a subject to a proposition. Russell held a judgement/belief to be
a dated, particular event which could be studied empirically292. There are many
propositional attitudes apart from beliefs, such as an assertion, question, volition
or doubt – Russell calls them propositional thoughts, but in his considerations
he was concentrated most of the time on the concept of understanding. First, he
considered the relation of understanding to be the most general propositional
attitude. Every aforementioned propositional attitude is a kind of understanding.
For instance, both questioning what is expressed by a proposition as well as de-
siring what is expressed by a proposition assumes that the subject of a particular
propositional attitude understands the proposition in question. Second, Russell
hoped that a proper analysis of understanding a proposition would simultane-
ously (because of the akin logical form293) answer the question as to what a belief
in a proposition consists in. Therefore, he hoped that a proper analysis of under-
standing a proposition would provide the right theory of judgement.
The concepts of a judgement and of a belief assume the concept of a proposi-
tion. Whereas Wittgenstein in the Tractatus defined a proposition simply as a
perceptible sign projecting a possible situation (TLP 3.11), for Russell this was a
much more difficult task294. It seems that he hoped that his reader would intuitive-
ly grasp the idea of a proposition when he grasped what the following sentences:

problem of what is responsible for uniting the constituents of a proposition exactly

with this and not the other form.
291 Carey 2007, p. 12.
292 “When I speak of ‘belief ’, I mean the same kind of facts as is usually called ‘judge-
ment’. I prefer the word ‘belief ’, because it has much more definitely the suggestion
of a particular dated event which may be studied empirically by psychology” (Russell
1913/1984, p. 136).
293 “Understanding and belief are closely akin as regards logical form and raise the same
logical problems” (ibid., p. 108).
294 “It is perhaps easier to discover what is meant by ‘understanding a proposition’ than
to discover what is meant by a proposition” (ibid., p. 107).

135
“Beggars are riders.”, “Beggars would be riders.”, “Are beggars riders?”, “Beggars
shall be riders.” have in common295.
The point of departure is the claim that since understanding as a propositional
attitude is the relation of a subject to a proposition, then we could treat a propo-
sition as an object (this is the proposal of Meinong who named a proposition
Objektiv). If this were true then it would turn out that understanding is (just like,
for instance, perception) a dual relation of acquaintance. Russell gives two argu-
ments against Meinong’s hypothesis. First of all, a proposition is true or false,
and objects of acquaintance are what they are – they are neither true nor false.
Second, Russell seeks a theory of judgement which would be indifferent to the
truth or falsehood of propositions involved in the act of judging. Since he rejects
the reality of false propositions, he is forced to see true propositions as also un-
real296. Hence, in Russell’s opinion, propositions cannot be objects of acquaint-
ance. He claims that propositions are incomplete symbols. Simplifying, a phrase
expressing a proposition requires a context in order for the proposition to acquire
meaning. According to Russell, the conclusion from the above considerations is
that “when we understand or assume or believe a proposition, what is involved is
not a dual relation of the subject to a single entity, such as Meinong’s Objektiv”297.
Understanding cannot be a kind of acquaintance.
For Russell, acquaintance (like sensations, memories and imaginations) is a
dual relation (of a subject with just one relatum), whereas in understanding the
proposition: “A precedes B” (where A and B are particulars), “we must have ac-
quaintance with A and B and with the relation ‘preceding’ ”298. We also have to be
acquainted with the general form of a dual complex, but this is still not enough
because it does not allow us to discern between: “A precedes B” and “B precedes
A”. Hence, Russell, under Wittgenstein’s influence, introduced the notion of a
general form. It was Wittgenstein who indicated the importance of the differ-
ence between the two sentences above. Although the proposition: “Socrates is

295 Ibid., p. 107. We can notice that all of the mentioned sentences have different mean-
ings, and that is the reason why Russell did not want to identify a proposition with
the meaning of a given phrase.
296 “We must say that, in the sense in which propositions are involved in believing and
in propositional understanding, there is no difference, as regards reality, between
true and false propositions. And in this turn, since it is repugnant to admit the real-
ity of false propositions, forces us to seek a theory which shall regard true and false
propositions as alike unreal” (ibid., p. 109).
297 Ibid., p. 110.
298 Ibid., p. 111.

136
mortal” makes perfect sense, this is not the case with the proposition: “Mortal-
ity is Socrates”. If understanding a proposition would involve standing in rela-
tion just to objects and the relations between them, nothing would prevent us
from making nonsensical propositions like the last one299. Russell agreed with
Wittgenstein’s criticism, but his response to the problems raised by his disciple
was Platonic in spirit300. In order to solve the problems he introduced a general
form of a state of affairs. In the case of a dual complex the general form could be
indicated by “xRy”. For example, in order to understand the proposition: “The
reign of King Henry VIII precedes the reign of King Edward VI”, one has to be
acquainted with the general form: “Something precedes something”. Moreover,
according to Russell, one has to be acquainted with the general and pure form to
be in a position to understand a particular proposition.
Russell, for the sake of clarity, analyses the symmetrical relation (such as “A re-
sembles B”, “A and B are similar”) in which the positions of A and B in a sentence
are not essential (“A and B are similar” and “B and A are similar” express exactly
the same proposition). In order to understand the proposition: “A and B are simi-
lar”, one has to be acquainted with objects A and B, with the relation of similarity
and with the general form of symmetrical dual complexes301. It is plain to see
that this understanding, under such a conceptualisation, cannot be a dual but is a
comprehensive relation. This means that a subject in understanding a proposition
is acquainted with multiple elements. It is also plain that the understanding does
not consist in separate acquaintances with each of these elements (even if coexist-
ing in one momentary experience) but that it brings all these elements “into rela-
tion with each other, so that all become parts of one complex”302. In other words,
although according to Russell the general form is responsible for synthesis within
a proposition (in our example the synthesis of objects A and B and the relation of

299 Wittgenstein noticed that without the concept of a form under Russell’s theory (which
treats the particulars, universals, relations as terms of a proposition) one can substi-
tute, for instance, “bigger than” with “penholders” (both of these expressions appear in
a proposition in the same mode of a term). One of the consequences of Russell’s theory
would be then the claim that since: “The table is bigger than the book” expresses a
possible situation, so “The table penholders the book” (NL, p. 103). To prevent this
consequence, Russell introduced the notion of a form, which would block the pos-
sibility of a substitution of a universal by a particular and vice versa.
300 According to Pears, Russell’s thoery is a “remarkable Platonic theory” (Pears 1977,
p. 181).
301 Russell 1913/1984, p. 112.
302 Ibid., p. 112.

137
similarity), there is a further problem of a unity of judgement that involves, apart
from the terms of the proposition, also a subject and an act of judgement.
I will put emphasis on this fragment of Russell’s text because it clearly differen-
tiates his position from Wittgenstein’s. When Russell answers the question as to
what the difference is between a proposition and a mere collection of objects, he
comes back to his earlier remark that it is the fact that propositions are true or false.
Being true or false is a hallmark of the sphere of the mental; hence propositions are
features of mental events. But the current description of a proposition (the objects
and relations falling under a certain general form) does not satisfy the restriction
that propositions are true or false. It is my hypothesis that in order to solve this
problem Russell had to refer to the mind. As we remember, Russell takes proposi-
tions to be incomplete symbols. What completes the meaning of a proposition is
a mental act (such as judging or understanding); hence it is the connection to the
mind which makes propositions the sort of things which are true and false303. Peter
Hanks claims that “the real point of Wittgenstein’s objection [to Russell] is that
what is judged must be capable of being true or false, and a disunified collection
of objects, properties, and relations (…) lacks that capacity”304. I do not agree with
this interpretation. As I have tried to show, Russell was fully aware of the require-
ment mentioned by Hanks, and he addressed the issue. The reason for Wittgen-
stein’s dissatisfaction was the way the issue was addressed, not the fact that Russell’s
theory did not ascribe the feature of being true or false to propositions.
Russell’s solution betrays his dualism of mind and matter, where the region of
the mental is discerned from the region of the material (atoms, molecules, elec-
trons, etc.) and the psychological (emotions, sensations, etc.) by the possibility of
ascribing to the mental entities features of truth and falsehood. Russell’s answer
to the question as to what differentiates a collection of elements (objects: A and B,
the relation of similarity S, and the general form xRy) from the proposition that A
is similar to B reads as follows: the latter is a mental entity (in other words, it is true
or false). This answer introduces a subject as a necessary element of the theory of
judgement (it would be perhaps more accurate to speak about the introduction of

303 Of course, Russell writes that in order for a proposition to be a complete symbol,
an actual subject which has mental relations to the elements of the proposition is
not necessary: “It is not necessary to our definition that there should actually be a
subject which has one familiar mental relations to the objects” (ibid., p. 115), but at
the same time he admits that since there is no proof of the existence of propositions
not thought by anybody, “we cannot know of the existence of propositions other
than those that have been actually thought of ” (ibid., p. 116).
304 Hanks 2012, p. 39.

138
the mind). The benefit of my interpretation of Russell’s theory is that it explains
why one of the conclusions of Wittgenstein’s argument against Russell is the rejec-
tion of the existence of a subject.
Since in describing a proposition one has to refer to the mind, then in the final
formulation of what a judgement consists in the symbol that represents the sub-
ject has to occur. Accordingly, we read in the Theory of Knowledge that an under-
standing of the proposition: “A is similar to B” has the following logical structure:
U {S, A, B, similarity, R (x,y)}
Where U goes proxy for a mental act of understanding a proposition, S goes
proxy for the subject that understands a proposition, A and B go proxy for the
objects, and R(x,y) for the general form of the dual complex of similarity. Let us
call a proposition that A and B are similar proposition p. According to Russell, a
belief (tantamount to a judgement) has a similar logical form to an understand-
ing. The proposition of everyday language: “S believes that p” can be formalised
in the following way:
B {S, A, B, similarity, R (x,y)}
We can notice at once that in Russell’s formalisation of beliefs occurs the symbol
S that goes proxy for a subject. Therefore, taking a closer look at Russell’s notion
of subject is crucial to understanding Wittgenstein’s discussion with Russell.

3.1.2 The notion of the empirical self

In the following section I shall discuss the idea of the self as standing behind the
presented theory of judgement. Russell belongs to the tradition of empiricism
represented by such philosophers as Hume and Mill. This thesis does not con-
tradict my previous remarks about the dualistic solutions one finds in Russell’s
Theory of Knowledge. The superficial contradiction is a result of the fact that
the body-mind problem is usually presented in terms of the Cartesian doctrine
(the relation between different kinds of substances), but one has to remember
that it has its counterpart in empiricism, called by D.M. Armstrong “bundle
dualism”305. In the tradition of empiricism the category of self was identified with
the category of the mind, and the latter was identified with a series of impressions

305 “In the second place, we have what may be called ‘Bundle’ Dualism, the term ‘bundle’
echoing Hume’s notorious description of the mind as a ‘bundle of perception’. This form
of Dualism characteristically arises out of reflection on the difficulties of Cartesian
Dualism” (Armstrong 1968, p. 7).

139
(or appearances)306. The body-mind problem referred to impressions under such
a conceptualisation. According to Putnam, arguments in favour of dualism or
in favour of non-materialism were solely arguments against the identification of
sense-data with something material307. Therefore, I shall understand under the
concept of the empirical self the concept in which the self (tantamount to the
mind) is the container (reservoir, bundle) of impressions (appearances).
I will trace this concept of the mind by two prominent representatives of empir-
icism: David Hume and Bertrand Russell. I will also indicate the most important
difference between these two standpoints and I shall show how this comparison
should make a reader of the Tractatus suspicious of the common thesis that Witt-
genstein applied in this book a kind of neo-Humean argument against the self.
In A Treatise of Human Nature, Hume wrote, among others:
What we call a mind, is nothing but a heap or collection of different perceptions, united
together by certain relations, and supposed, though falsely, to be endowed with a perfect
simplicity and identity (Hume 1739/2001, p. 137).

One finds in this quotation two hallmarks of the concept of the empirical self:
identification of the self with the mind and the latter with a collection of percep-
tions. Similarly, in Logical Atomism, Russell represented the view that the con-
tents of one’s mind at one time consist of all of his or her sensations, images,
memories and thoughts308. For both of them the content of the mind is some-
thing different than material objects. Hume argued that we ascribe quite different
features to material objects and to impressions. For example, we ascribe to mate-
rial objects a distinct and continued existence, “since all impressions are internal
and perishing existences”309. That entails, in his opinion, an immaterial character
of impressions. Russell took immediate data of senses (certain patches of colour,
sounds, tastes, smells, etc.) to be prima facie very different from the objects of the
physical world: molecules, atoms or electrons (which have no colour, no taste,
make no noise, etc.)310. To be exact, Russell takes the above-mentioned sensibilia
to be psychological311, and propositional attitudes such as beliefs to be mental

306 Putnam 1986/1994, p. 5.

307 Ibid., p. 8.
308 Russell 1924, p. 341.
309 Hume 1739/2001, p. 129.
310 Russell 1914a, p. 5.
311 Ibid., p. 8.

140
entities312 (this means that since beliefs are true or false, sensibilia are what they
are – they are neither true nor false).
We come to the decisive difference between both philosophers, which will al-
low us to indicate the second feature of the Russellian concept of the mind. Hume
was convinced that all knowledge about the world comes from impressions, and
that one should not ascribe to reality more than is contained in only available
evidence about reality (the example of ascribing more would be saying that real-
ity contains material objects which are distinct and continued existences)313. As
a result, in his system the existence of the physical became doubtful, and though
it was still true that impressions and ideas are not material things, the mere idea
of comparison of what is mental with what is physical became meaningless. On
the other hand, Russell, on the basis of a real difference between beliefs and af-
fections, advocated for the realness of both relata of the relation: the mental and
the physical. The difference, roughly speaking, consists in the fact that the for-
mer has content and meaning. According to Russell, if one does not recognise
the difference between the mental on the one hand and the psychological or the
physical on the other, then one is tempted to reduce beliefs and judgements to
some bodily behaviours such as emotions314. Russell was afraid that this is a direct
path leading to logical behaviourism. One could suspect that what Russell had in
mind was the following reasoning: psychological statements are statements about
physical processes in the body (presumably in the central nervous system). But
because the current state of physiology does not allow us to reduce beliefs to con-
crete physical-chemical processes in the human body, then the first phase of this
reduction should be the translation of psychological statements into statements
about observable behaviours. For instance, instead of statements about excite-
ment, one could formulate statements about raised pressure and pulse, a rapid
heartbeat, a blush on a face, etc. But if there is no difference between statements
about beliefs on the one hand and statements about excitement on the other, then
one could also translate statements about beliefs into statements about pressure
and pulse, a rapid heartbeat, and so on. This position would deserve the name
of “logical behaviourism”. Russell indicated in this context that this is a mistake

312 Russell 1918, p. 221.

313 Hume 1739/2001, p. 128.
314 It is exactly the consequence of Hume’s point of view: “a belief may be most accurately
defined as a lively idea related to or associated with a present impression” (ibid.,
p. 67).

141
committed by “some of the American realists, the school whom one calls neutral
monists”315.
According to Russell, a proper answer to the question as to what the difference
is between an excitement and a belief would indicate the fact that a belief is an act
of the mind. But if one describes the mind as Hume does, i.e. solely as a container
of appearances, one will still not grasp the difference between beliefs and excite-
ments. Under such a conception of the mind there is no essential difference be-
tween beliefs and emotions – both are appearances. One, therefore, cannot content
himself with the Humean characteristic of the mind; the mind is not only a screen
that passively receives images. Russell writes little about this additional function of
the mind but, indeed we find in his Theory of Knowledge a short section where he
talks about “uniting” or “synthesizing the terms of proposition into one complex”.
The general form is responsible for the process of this synthesis, but there is an
additional process of synthesis in which a subject unites a proposition in an act of
judgement. Although the mind is not mentioned directly as the prime mover of
this process of unification, it is clear that there is no other candidacy (for example,
Russell draws a graph of a belief as a five-term complex. In the graph the lines
binding the terms meet in the subject). We can sum up the description of Russell’s
notion of the empirical self in the following way:
1) The category of the empirical self is tantamount to the category of the mind.
2) The empirical self is a container of sensations such as beliefs (judgements),
desires, understandings, etc.
3) The empirical self unites the terms of a proposition and in this way it forms
judgements.
That is the concept of the self which was criticised by Wittgenstein. From the
description I have proposed above it follows that:
(A) there are indeed differences between Russell’s and Hume’s conception of the
mind
(B) Russell did not think of the mind as an immaterial substance, hence:
If Wittgenstein’s critique is directed solely against the Cartesian notion of the
self, then it is not directed against Russell’s notion of the self. This cannot be
true because in TLP 5.54-5.5422 Wittgenstein directly referred to Russell’s the-
ory of knowledge. This leads me to the conclusion that Wittgenstein’s argument
is generally anti-dualistic, but if that is the case then one should also include

315 Russell 1918, p. 218.

142
here Humean bundle-dualism. Wittgenstein’s argument is not, as many com-
mentators claim, a new form of the Humean critique of Descartes.

3.2  Wittgenstein’s criticism of Russell’s views on judgement

In the previous section I analysed the background of the theses of the Tractatus
which are interesting to me. In the following section I shall point at four reasons
of Wittgenstein’s disagreement with Russell’s theory of judgement. Wittgenstein’s
own view on judgement will emerge from this critique. We know that Wittgen-
stein was unsparing in his criticism of the theory of judgement which Russell had
presented to him. In 1913 Russell was working on a new book on epistemology,
with the working title Theory of Knowledge, and he was hoping that it would be his
major contribution to this area of philosophy. After writing six chapters he pre-
sented the results of his efforts to Wittgenstein, but he came in for harsh criticism.
As a result of that criticism Russell withdrew publication of the book, although, as
he admitted, he had not understood all of Wittgenstein’s remarks. He just felt that
his disciple had to be right316. The manuscript remained unknown until 1967 and
waited for an edition until 1984. Probably until his death, because of Wittgenstein’s
criticism, Russell did not like the text and wanted to forget about it317. We do not
know the exact course of the discussions between these two philosophers318, but in
the light of Wittgenstein’s position this is what one could have expected his criti-
cism to be:
a)  The first objection is clearly expressed in the Tractatus:
The correct explanation of the form of the proposition, ‘A makes the judgement p’, must
show that it is impossible for a judgement to be a piece of nonsense. (Russell’s theory does
not satisfy this requirement.) (TLP 5.5422).

316 Carey 2007, p. 1–2.

317 In 1967 Russell sold a bundle of manuscripts (including the one containing Theory of
Knowledge) to the Russell Archives at McMaster University in order to collect money
for the International War Crimes Tribunal, but even then he remained silent about
it: archivists intrigued by the discovery of an unknown text from 1913 wrote a letter
to Russell about it, but he never answered it (Carey 2007, p. 2).
318 The ordinary source of information about the discussions between Russell and Witt-
genstein – Russell’s letters to Ottoline Morrell – remain silent about the matter. A
hypothetic course of the talks is given in: Johnston 2012, pp. 29–31.

143
There are different views as to why Wittgenstein thought that Russell’s theory
does not prevent one from thinking nonsense319. According to Johnston, this was
because it did not account for a thought’s synthesis320. Russell himself was con-
vinced that with the notion of the general form he was addressing Wittgenstein’s
accusation that he had expressed already in Notes on Logic:
Every right theory of judgement must make it impossible for me to judge that this table
penholders the book. Russell’s theory does not satisfy this requirement (NL, p. 103)

The accusation from Notes on Logic refers to the fact that in the period before
Theory of Knowledge Russell did not discern between particulars and universals as
far as their role in a proposition was concerned. He held both of them as appearing
in a proposition in the same mode of a term. According to Russell, one could sub-
stitute in a meaningful proposition a term appearing in a mode m with any other
term appearing in mode m and still obtain a meaningful proposition. But if, for ex-
ample, one substituted the term “love” with the term “Iago”, one would obtain from
the meaningful proposition: “Desdemona loves Cassio” plain nonsense such as:
“Desdemona Iagos Cassio”. That is the edge of Wittgenstein’s criticism, and Russell
corrected his views by introducing the notion of a form321. On the other hand, this
correction did not change Wittgenstein’s assessment: he still thought that Russell’s
theory allows one to think nonsense. Why? In my opinion this is because Russell’s
account of a judgement assumes an active role of the subject. If a belief consists in a
relation between a thinker and various objects of a proposition, and the role of the
thinker is to synthesise these objects so that he is acquainted with all constituents
of the thought, then he can unite in his mind objects and properties however he
wants (for example, objects number 3 and a sandwich and the relation of eating

319 In this context a nonsensical expression is an expression in which its constituents are
going against its entity types. For instance, in the expression: “The table penholders
the book”, “penholders” is going against its entity type and plays the role of a universal.
320 “Wittgenstein’s claim that Russell’s multiple relation theory is consistent with the exist-
ence of nonsense judgement was intended as a demonstration that that theory, even
in its 1913 version, does not account for thought’s synthesis” (Johnston 2012, p. 31).
321 Understanding the proposition “A is similar to B” assumes understanding the form:
“something is similar to something”. Only one who sees that in the place of relata oc-
cur only particulars and in the place of the relation only the universal can judge that
A is similar to B. In this sense, judging nonsense in which elements of a proposition
are going against their types (for example a particular playing the role of a universal)
is blocked.

144
with the general form: “something eats something”. The result of such a unifica-
tion: “Number 3 eats a sandwich” does not make any sense)322.
b) The Russellian theory of judgement entails an inconvenient consequence for
Wittgenstein’s theory of a meaningful proposition, which says that every mean­
ingful proposition is a truth-function of an elementary proposition (with the
extreme case of an elementary proposition which is the truth-function of
itself)323. If both Russell’s analysis of beliefs as well as Wittgenstein’s principle of
extensionality are true, then beliefs cannot be meaningful because judgements
of the form: “A believes that p” are not the truth-functions of p (let us suppose
for a moment that p is an elementary proposition). The truth-value of “A be-
lieves that p” is independent of the truth-value of an elementary proposition p.
This is also why Wittgenstein wanted to propose his own theory which would
retain the extensionality principle.
c) Russell mentions that among the objects of acquaintance there is “a form
of complex”. For example, in the case of a similarity between particulars A
and B, one is acquainted not only with particulars A and B and the relation
that connects both the particulars, but also with the form of this dual com-
plex. The pure form of a complex is described, according to Russell, in the
expression: “something is similar to something”324. Exactly this view caused
Wittgenstein’s disagreement. According to him, there are no logical objects
(TLP 5.4); one can do without introducing abstract (in contrast to concrete)
entities in order to explain propositions.
Moreover, Wittgenstein could not agree with Russell’s conviction that one describes
the form of a state of affair. In a state of affairs its elements stand in some relations
to one another (TLP 2.14) – thus a state of affairs has a structure (TLP 2.15). The
possibility of having a structure is called in the Tractatus a form – the logical form
or the pictorial form (TLP 2.151). Thesis TLP 4.121 expresses what is essential in
Wittgenstein’s views on form:

322 Of course, this would mean that in comparison to Notes on Logic, Wittgenstein
changed the meaning of the concept of nonsense. Under an earlier conception,
“Number 3 eats a sandwich” is a meaningful proposition. Each of its constituents is
in line with its entity type. Nevertheless, this expression, popularly understood, is
nonsense. Wittgenstein speaks then (in TLP 5.5422) about nonsense in this popular
meaning of this concept.
323 “All propositions are results of truth-operations on elementary propositions” (TLP 5.3).
324 Russell 1913/1984, p. 133.

145
 Propositions cannot represent logical form: it is mirrored in them. What finds its reflec-
tion in language, language cannot represent. What expresses itself in language, we cannot
express by means of language. Propositions show the logical form of reality. They display it.

Gilbert Ryle explained in a very simple and convincing way why the form itself
cannot be depicted: let us take a sheet of graph paper in a hospital. The dots of the
graph represent the temperatures of a patient. Ryle asks a reader to imagine the
following situation: “suppose we now asked the nurse to depict on a second sheet
of graph paper, not the course of the patient’s temperature, but the rules for rep-
resenting his temperature by dots on graph paper, she would be baffled. Nor can
the rules and conventions of map-making themselves be mapped”325. Because, ac-
cording to Wittgenstein, what can be shown (and that is, among others, the logical
form of a proposition) cannot be said (TLP 4.1212), Russell’s efforts to describe it,
were, in his eyes, futile. In one of the entries of his Notebooks Wittgenstein wrote:
The reality that corresponds to the sense of the proposition can surely be nothing but
it component parts, since we are surely ignorant of everything else (NB 20.11.14, p. 31).

According to Wittgenstein, when one describes a fact, the elements of propo-

sitions refer only to the component parts of the respective fact, i.e. to simple
objects. What one can do at best in describing forms of facts is to describe the
constituents of the fact: what they are and in what relations to one another they
are standing. There is nothing more to describe326. For Wittgenstein, describ-
ing pure forms as something existing apart from particulars sounds peculiar. In
his opinion, it unnecessarily introduces the second realm: the Platonic-Fregean
realm of thought327. Russell was aware of the fact that his Platonic ontology could
drive away his readers from accepting his theory, and he tried to explain the fact
that we normally pay attention only to particulars by referring to the popular
version of the theory of natural selection. He claimed that there are particulars
which have a vital role for our sustainability as a species. Particular things can be
good to eat or they can kill us; hence we are sensitive only to the existence of par-
ticulars and not universals. According to Russell, only eccentrics, relieved from
the struggle for existence, also pay attention to universals. This does not mean,

325 Ryle 1951/1967, p. 121.

326 In Wittgenstein’s opinion, expressed in one of the letters to Russell, it was the main
point of Wittgenstein’s theory of judgement that his mentor “had not really got hold
of ” (Letters 19.8.19, p. 71).
327 According to Pears, Wittgenstein’s solution to the problem of a form “is Aristotelian
in spirit. The forms that Russell had placed in a Platonic world were treated by Witt-
genstein as essential features of objects” (Pears 1977, p. 188).

146
however, that ordinary people are not acquainted with pure forms328. In the light
of Wittgenstein’s views, we could say that this explanation could serve him at
best as an example of a superficial psychology one cannot mix with philosophy.
d) Although the aforementioned differences are great, there is also, in my opin-
ion, a more basic discrepancy which reaches to the roots of both systems, and
that is the concept of meaning. The theory of judgement must appreciate the
fact that, using Hanks’ formulation, “what is judged [a proposition] must be
capable of being true or false”329. In section 3.1.1 we saw that Russell’s theory
fulfils this requirement by putting a proposition in a relation to the mind. A
proposition in itself is not an actual entity, it is an incomplete symbol. One
cannot prove the existence of propositions other than propositions which are
(or were) actually thought by somebody330. In contrast to these views, Witt-
genstein states that for a proposition to be true or false it is sufficient to be a
picture of a state of affairs331. For something to be a picture or a model of real-
ity is to fulfil three conditions:
1) The complexity condition. It has to share the complexity of the depicted
state of affairs, and this means that every element of a state of affairs has to
be represented by an element of a picture (TLP 2.13-2.131).
2) The structure condition. Condition 1) alone is not enough. Apart from eve-
ry object of the state of affairs being represented by an element of a picture,
a picture has to have the same structure as the respective state of affairs. By

328 Russell 1913/1984, p. 132–133.

329 Hanks 2012, p. 39.
330 Russell 1913/1984, p. 116.
331 David Shier would disagree with my opinion. According to him, “mere pictures are
not yet propositions” (Shier 1997, p. 69). His main argument says that the same picture
can serve to express two contradictory propositions, and that “the picture by itself, qua
picture, neither affirms the pictured state of affair, nor denies it” (ibid., p. 71). Shier
claims that in order to make a picture from a proposition, one needs a mental act, for
example, an act of thinking the sense of a proposition (ibid., p. 73). In his view, a mental
act is indispensable if we account for propositions, and in this respect, according to
him, there are no significant differences between Russell and Wittgenstein. I think that
Shier is forgetting that the ambiguity of a picture refers only to what the picture shows:
it is true that one proposition could be used to show two contradictory facts (to show
what the state of affairs obtains and to show what the state of affairs does not obtain).
But Wittgenstein underlined that what the proposition says is quite determinate – it
says how things would stand if the proposition were true (C.f. TLP 4.022).

147
 the structure of a state of affairs Wittgenstein understood the connection of
its elements (TLP 2.15).
3) The logical form condition. A picture has to have a pictorial form. In oth-
er words, there have to be some rules of projection combining different
objects into the relation “a model – a state of affairs”.
I shall explain the third condition a little more. In the Tractatus we read, among oth-
ers: a “Pictorial relationship consists of the correlations of the picture’s elements with
things” (TLP 2.1514). Wittgenstein says that such different objects as written notes
and a gramophone record could stand in this relation (TLP 4.014), but to make this
happen one needs rules according to which it is “possible to derive the symphony
from the groove on the gramophone record” (TLP 4.0141). As I have mentioned, the
possibility that “things are related to one another in the same way as the elements
of the picture” (TLP 2.151) was called by Wittgenstein the pictorial form, hence it is
the pictorial form which makes a state of affairs a picture of another state of affairs.
As we can see, not only propositions fulfil these three conditions and, therefore,
not only propositions are pictures and, therefore, not only propositions realise in-
formation (to use Chalmer’s notion332). If a gramophone record is a picture of writ-
ten notes, it realises the information contained in those notes; it transmits the sense
of the written notes. A proposition is a kind of picture (TLP 4.021), and even here
the concept of a proposition is used in a broader sense, i.e. one that is not limited
to linguistic signs. Wittgenstein could easily imagine propositions composed of
books, chairs and tables (TLP 3.1431).
But what about Russell’s question? What is the difference, then, between a
proposition and a mere collection of names? As we remember, Russell, in order to
explain the mental character of propositions, introduced a relation to a subject –
the factor that was missing in the case of a mere collection of names. Of course,
Wittgenstein, as Russell’s disciple, was aware of the problem:
A proposition is not a blend of words. – (Just as a theme in music is not a blend of
notes). A proposition is articulate. Only facts can express a sense, a set of names cannot
(TLP 3.141-3.142)

The solution to the problem was, however, different. It could seem that a random
collection of three objects on my desk is not in itself a picture of anything. But if
I put next to them three other objects and say that the object on the right of the
original collection of objects (C1) goes proxy for the object on the right of the new
collection of objects (C2), the object in the middle of C1 goes proxy for the object

332 Chalmers 1996, pp. 276–310.

148
in the middle of C2, and the object on the left of C1 goes proxy for the object on
the left of C2, then nothing has changed in C1 – but together with the rules of pro-
jection it is now more than just a mere collection of objects – it is a picture of C2.
How is it possible that there were no changes in C1 but now it is a picture? When
answering this question one should ask why one could introduce such rules of
presentation as “the object on the right goes proxy for…” in the first place? There
had to be something in C1 that allowed us to make it a picture of C2 – it was the
fact that in C1 objects stood in some relations to one another – and this is what
Wittgenstein called a logical form. The three aforementioned conditions are suf-
ficient for any collection of objects to be a picture. An additional condition – the
condition of the identity of logical forms – is a sufficient and necessary condition
for a picture to be a picture of a concrete state of affairs.
If a fact is to be a picture, it must have something in common with what it depicts. There
must be something identical in a picture and what it depicts, to enable the one to be a
picture of the other at all. What a picture must have in common with reality, in order
to be able to depict it—correctly or incorrectly—in the way it does, is its pictorial form.
(TLP 2.16-2.17)

The answer to the Russellian problem is, therefore, the following: what differenti-
ates a mere collection of names from a proposition is the fact that the latter has a
logical form. Simply by the fact of having a logical form is a proposition a unity.
I want to emphasise the fact that Wittgenstein’s solution does without the concept
of the mind. A proposition is true or false because it is a picture. It is a picture on
the basis of having a logical form.

3.3 The Tractatus 5.54-5.5422

After explaining both Russell’s and Wittgenstein’s theories of proposition, we can
better understand the argument contained in TLP 5.541-5.5422. In the following
section I shall analyse, thesis by thesis, the fragment in question. The text con-
tains important notions which in the Tractatus have a slightly different meaning
than in Russell’s writings, so I will start with short conceptual explanations. The
task of this excerpt of the Tractatus is, just as for Russell, to analyse judgement
and to show that this analysis does not put the extensionality thesis at risk. My
task in this section will be to reconstruct Wittgenstein’s theory of judgement.

3.3.1  Conceptual clarifications

Russell held the concept of a judgement to be synonymous with the concept
of a belief. Wittgenstein, additionally, thought that the concept of a thought is

149
another synonym for the concept of a judgement. Frank Ramsey, who as a stu-
dent visited Wittgenstein (at that time working as a school teacher in the vil-
lages of Austria) complained in one of his letters to his mother that Wittgenstein
would forget the meaning of what he had said five minutes before, that some of
the Tractarian claims are deliberately ambiguous and have two meanings, and,
moreover, that their author believes in both of them333. The concept of a thought
is a good example of the aforementioned ambiguity. In the Tractatus one can
single out two meanings of this concept334:
A) A thought as the content of a proposition: in TLP 4 Wittgenstein characteris-
es thought as “a proposition with a sense”. In the same context he complains
that “language disguises thought” (TLP 4.002). The concept of a thought
does not occur in this meaning in TLP 5.54-5.5422.
B) A thought as an act of thinking. For instance, Wittgenstein claimed that a
“thought can never be of anything illogical, since, if it were, we should have
to think illogically” (TLP 3.03). The English translation does not give exactly
what Wittgenstein meant because of the reification of the verb “denken” in
the translation. In the original the author states: “Wir können nichts Unlo-
gisches denken, weil wir sonst unlogisch denken müssten”. From now on by
the term “thought” I will understand an act of thinking; a dated, temporal
event.
The main proof for the claim that Wittgenstein in TLP 5.54ff considered a
thought a dated, temporal event can be found in a letter in which Wittgenstein,
when asked by Russell what the constituents of a thought are and what their rela-
tion is to the pictured state of affairs, responded as follows:
I don’t know what the constituents of a thought are but I know that it must have such
constituents which correspond to the words of Language. Again the kind of relation of
the constituents of thought and of the pictured fact is irrelevant. It would be a matter of
psychology to find it out (Letters 19.8.19, p. 72)

In contemporary literature one often juxtaposes beliefs with thoughts by say-

ing that whereas the former are dispositions, and therefore states, the latter are
mental acts, and therefore events335. But for Wittgenstein there is no difference

333 Waugh 2010, p. 199.

334 As Wittgenstein later wrote: “ ‘Thought’ sometimes means a particular mental process
which may accompany the utterance of the sentence and sometimes the sentence
itself in the system of language” (PG, p. 51).
335 Crane 2001, p. 103.

150
between a belief that p, a judgement that p, or having a thought that p. In all of
these cases he is talking about temporal and dated events.
Wittgenstein rejected the Russelian notion of a proposition336. As we remem-
ber, in the Theory of Knowledge Russell held that propositions are dependent on
mental acts; for Wittgenstein, in contrast, propositions are pictures337 (in the pre-
vious section I showed that being a picture of reality was a sufficient condition to
be a bearer of truth and falsity). Wittgenstein discerned two meanings of the no-
tion of a proposition: a propositional sign (a token of an utterance or an inscrip-
tion) and a symbol. A symbol is what all propositional signs expressing the same
sense have in common338. For instance, “The cat is on a mat” and “Kot jest na
macie” are two different propositional signs (in English and in Polish), but they
express the same symbol. In the fragment analysed in this chapter of the Tractatus
Wittgenstein used the notion of a proposition in the meaning of a symbol.

3.3.2  The form of “A believes that p” according to Wittgenstein

After these conceptual clarifications we are in a position to analyse Wittgenstein’s
response to Russell’s theory of judgement. In TLP 5.541 we read:
At first sight it looks as if it were also possible for one proposition to occur in another
in a different way. Particularly with certain forms of proposition in psychology, such as
‘A believes that p is the case’ and ‘A has the thought p’, etc. For if these are considered
superficially, it looks as if the proposition p stood in some kind of relation to an object
A. (And in modern theory of knowledge (Russell, Moore, etc.) these propositions have
actually been construed in this way.)

Wittgenstein introduces here Russell’s way of dealing with beliefs. According

to Wittgenstein’s interpretation of Russell’s theory, a belief consists in a relation
between a subject A and a proposition p. This is not exactly what Russell says
because in his theory the relation in question is not a dual one; it is a relation
between a subject and objects (particulars and universals), but this omission

336 “Frege said ‘propositions are names’; Russell said ‘propositions correspond to com-
plexes’. Both are false, and especially false is the statement ‘propositions are names
of complexes’ ” (NL, p. 97).
337 “A proposition is a picture of reality. A proposition is a model of reality as we imagine
it” (TLP 4.01).
338 “I call any part of a proposition that characterizes its sense an expression (or a sym-
bol). (A proposition is itself an expression.) Everything essential to their sense that
propositions can have in common with one another is an expression. An expression
is the mark of a form and a content” (TLP 3.31).

151
suggests which aspect of Russell’s theory Wittgenstein is concentrated on. He is
obviously interested in two things:
(1) that a judgement consists of a relation. One term of the relation is a subject.
According to Russell, (J): “A believes that p” is of the form
(F1)  B (A, a, b, R, γ)339
(2) and that this interpretation of Russell’s theory could put the Tractarian prin-
ciple of extensionality in danger.
However, it turns out that the danger mentioned in (2) vanishes simply because
(J) is not of the form (F1). So, in the next thesis (TLP 5.5421) we read:
It is clear, however, that ‘A believes that p’, ‘A has the thought p’, and ‘A says p’ are of the
form ‘ “p” says p’: and this does not involve a correlation of a fact with an object, but rather
the correlation of facts by means of the correlation of their objects.

In other words, a proper interpretation of (J), construed with the support of the
proper theory of judgement, should reflect what truly takes place in (J). As I
pointed out earlier, what Wittgenstein and Russell both agree on is that a propo-
sition is true or false. This fact was explained differently by both philosophers.
Russell’s formulation (F1) contains the reference to a subject, which in his theory
is responsible for uniting the elements of a proposition in a judgement.
In the previous section I presented Wittgenstein’s theory of a proposition and
I underlined that he goes without the concept of a subject. From the fragments of
the Tractatus quoted above we can give the following examples of propositions:
–– a gramophone record (picture) and the written notes (depicted state of affair –
TLP 4.014)
–– furniture in a room (could be a picture just like a written sentence – TLP 3.1431)
One can then explain why a proposition has content in terms of models, and this
means in terms of sharing the same complexity, structure and form with the de-
picted state of affairs. The examples given here show that one does not need any
subject to “complete” the propositions – they are independent entities. Therefore,
Wittgenstein’s interpretation of (J) looks as follows:
(F2):  “ ‘p’ says p”.
According to Wittgenstein, a judgement consists not of a relation between a subject
and a proposition but of a relation between two facts. The task for the interpreter is

339 “B” stands for a belief, “A” stands for a subject, “a” and “b” for particulars, “R” for a
relation holding between particulars and “γ” for a general form of a complex “aRb”.

152
to unravel which facts exactly Wittgenstein had in mind. Assuming the physicalistic
spirit of the Tractatus, according to my hypothesis, roughly speaking, Wittgenstein
held a judgement to report not a relation between a subject and a proposition but
a relation between an event in the brain and a state of affairs340.
There is one thing to ask: “What does a judgement consist in?”, and another to
ask: “What proposition does a reporting judgement (such as a proposition (J))
inform about?”. Wittgenstein’s answer, expressed in (F2), reveals two states of
affairs: quoted and unquoted p standing to each other in the relation of saying.
The use of the notion of saying alludes to thesis TLP 4.022, in which Wittgenstein
explains the relation of depicting states of affairs and claims that “a proposition
shows how things stand if it is true. And it says that they do so stand”. A proposi-
tion says that a state of affairs obtains. In other words, Wittgenstein, by revealing
the true form of (J): “A believes that p” by (F2): “ ‘p’ says p” claims that:
(1) “p” stands to p in the same relation as a picture stands to a depicted state of
affairs, and that
(2) a proposition reporting a judgement states the existence of a state of affairs.
What state of affairs? Some authors (Hanks, Jacquette, Puhl) talk in this context
about the language of thought or mental sentences341. Although some views that
Wittgenstein held with respect to language, such as semantic atomism or the
compositionality thesis, suggest that he could accept the language of thought
theory, such a proposal seems to impose a contemporary philosophical vocabu-
lary and solutions on the Tractatus. Given that the argumentation of the defend-
ers of the materialistic reading of the book is true and, hence, that Wittgenstein
saw the world as consisting of physical atoms, it seems more plausible to me
that, according to Wittgenstein, a proposition reporting judgement (J) informs
about the existence of an event in the brain. I prefer to speak about “an event in
the brain” more than about “mental sentences” also because the first expression

340 As far as I understand Peter Hanks’ interpretation (Hanks 2012), my proposal is

similar to his. Dayton’s interpretation also goes in the same direction: “The Tractatus
tends in the direction of materialism (…) The projective relation by means of which
a proposition pictures a fact essentially involves the human organism in its internal
causal complexity” (Dayton 1976, p. 282). Also, Elisabeth Anscombe, instead of
speaking of A’s beliefs or judgements, prefers to say: “There occurs in A or is produced
by A something which is (capable of being) a picture of p” (Anscombe 1965, p. 88).
341 “If judgement sentences say anything, they say only that certain physical facts exists,
the mental sentences that are depicted by the quoted sentences in their subject posi-
tions. Any other information provided by judgement sentences, including informa-
tion about the senses of these mental sentences, is merely shown” (Hanks 2012, p. 59).

153
is more general (after all, mental sentences, if there are any, are also events in the
brain), and it better suits Wittgenstein’s scepticism towards philosophy trying to
decide purely empirical matters.
When A judges that p (note that we are talking about judgements as dated,
temporal events), something happens in the brain of A. In (F2) an event in the
brain is represented by an unquoted p. What is left in the Tractarian analysis of a
judgement is a quoted proposition “p”. In my opinion, it is a proposition as a sym-
bol (in the meaning I gave at the beginning of the section). Of course, one has to
bear in mind how broadly Wittgenstein understood the notion of a proposition.
It could be a token of an inscription or of an utterance; it could even be a specific
arrangement of objects. These propositions express (show) what the content of a
judgement is and say that a certain event in the brain obtains.
Let us take the following example: a mother tidies up the room of her teenage
son. She does so because she believes that he has a mess in his room. This belief does
not need to be linguistically expressed. In this case a term of the judgement is the
actual way the son’s room looks. The mother’s belief consists in a relation between
an event in her brain (it has a sufficient complexity to depict a state of affairs) and
the arrangement of objects in her son’s room. If we take then a report of this belief
(J1):  “The mother believes that there is a mess in her son’s room”,
then, according to my reading, Wittgenstein would interpret this utterance as in
fact informing about the existence of a certain event in the brain that is capable of
depicting a state of affairs. The depicted state of affairs, in our case the mess in the
room, shows what the content of the mother’s thoughts is.

3.3.3  Consequences of Wittgenstein’s theory of judgement

If my interpretation is correct, then, according to Wittgenstein, a judgement con-
sists in a relation of picturing between a state of affairs and an event in the brain.
What does one gain from this interpretation? First, it is compatible with the ma-
terialism of the Tractatus. It is not committed to intentional states or processes
endowed with casual powers. It admits that beliefs are causes, for instance, that an
explanation for my taking an umbrella for a trip is my judgement that it will rain,
but beliefs themselves are explained in a naturalistic way. Therefore, this interpre-
tation, assuming that only matter is endowed with casual powers, is in accord-
ance with the physicalism of the Tractatus and it harmonises well with the general
conviction of the commentators that the overtone of TLP 5.54ff is anti-dualistic.
Secondly, the proposed interpretation explains a close connection between
Wittgenstein’s criticism of Russell’s theory of judgement and the repudiation of

154
the existence of the self by the end of the argument. Since Wittgenstein goes with-
out the notion of the self in the explanation of a belief, then the concept of the
self is unnecessary, and what is the unnecessary sign, according to Wittgenstein’s
reading of Occam’s razor, means nothing342. That is why in the following thesis
(TLP 5.5421) Wittgenstein considers Russell’s solution as absurd:
This shows too that there is no such thing as the soul—the subject, etc.—as it is con-
ceived in the superficial psychology of the present day. Indeed a composite soul would
no longer be a soul.

The absurdity consists in conferring no meaning on the sign: “A” in (F1): “B (A, a,
b, R, γ)”. Of course, the notion of the soul refers here to what I described in the pre-
vious sections as the empirical self. Otherwise, the conclusion would be entirely
unconvincing. The plain possibility of naturalisation of the intensional contexts
does not mean that the self does not exist. If Wittgenstein was inclined towards
such a strong thesis, this could mean two things: either he thought that the theory
of judgement is the only way for arguing in favour of the existence of the self (this
hypothesis I reject because, as we will see in the next chapter, Wittgenstein saw, at
least at some point of writing the Notebooks, the possibility of arguing in favour
of the existence of the self on the basis of the existence of moral judgements),
or the thesis “there is no such thing as the soul” is not a result of Wittgenstein’s
dissatisfaction with Russell’s views on judgement but rather a confirmation of
Wittgenstein’s assumptions. It seems, then, uncontroversial to say that TLP 5.5421
concludes that the notion of the empirical self is empty and that the conclusion is
a confirmation of the hypothesis that the self does not exist.
Thirdly, my interpretation shows that Wittgenstein’s theory of judgement elimi-
nates a possible danger for the extensionality principle. Intensionality of beliefs
consists in the fact that it is possible that, although proposition p expresses the same
state of affairs as proposition q, subject A believes that p, but he does not believe
that q. For instance, Oedipus believes that he is married to Jocasta, but he does not
believe that he is married to his own mother, despite the fact that the propositional
signs: (p) “Oedipus is married to Jocasta” and (q) “Oedipus is married to his own
mother” express the same proposition. We have a paradoxical situation in which
Oedipus believes and does not believe in the same proposition at the same time.
Frege solved the paradox by introducing the notion of the way of representation.
Because of a different way of representing “Jocasta” and “The mother of Oedipus”

342 “Occam’s maxim is, of course, not an arbitrary rule, nor one that is justified by its
success in practice: its point is that unnecessary units in a sign-language mean noth-
ing” (TLP 5.47321).

155
to Oedipus, one can say that (p) and (q) are in fact two different beliefs. The conse-
quence of this solution is connecting intensional contexts with a subject’s particu-
lar point of view. The connection is the following: intensional contexts express the
fact that a subject is directed at something and that things are presented to him or
her in a certain way. Wittgenstein’s hope that the elimination of a subject from a
theory of judgement would save the extensionality principle is probably the conse-
quence of an approach which claims that intensional contexts are the result of the
fact that subjects have particular points of view on the world. At the same time,
he solves the paradox of affirming and rejecting the same proposition at the same
time simply by saying that (p’) “Oedipus believes that he is married to Jocasta” and
(q’) “Oedipus believes that he is married to his own mother” inform about two dif-
ferent events in Oedipus’ brain and, hence, one cannot speak about a paradox but
about two different cases. Propositions p’ and q’, since they report the existence of
states of affairs, could be regarded as elementary propositions. The extensionality
principle is not threatened.

3.3.4 The repudiation of the existence of the complex soul

(TLP 5.5421)
At the end of the analysis of Wittgenstein’s theory of judgement I would like to
pay attention to an often overlooked but at the same time very startling remark.
In the second part of TLP 5.5421 Wittgenstein claims: “Indeed a composite soul
would no longer be a soul” (TLP 5.5421). Why did he add this remark? What is
its sense? The complex self would not be a subject of philosophy anymore but
of “superficial psychology” or neuroscience, and Wittgenstein clearly wanted to
avoid a conclusion such as: “The self is nothing more than an array of events
in the brain”, which would mean that there is no difference between his and
Russell’s theory of judgement343. But, of course, there has to be more in Witt-
genstein’s approach than the worry of distinguishing himself from his mentor.
In my opinion, the real reason for Wittgenstein’s position is his intention to set
the limits for natural science.
In the introduction to this chapter I mentioned that the remark that a soul
cannot be complex is a sign of Wittgenstein’s modernism in contrast to the

343 Russell says that a judgement consists in a relation between a subject and a proposi-
tion, Wittgenstein, on the other hand, claims that it consists in a relation between
an event in the brain and a state of affairs (and propositions are also states of affairs).
When one identifies a subject with events in the brain, then Wittgenstein’s theory is
different in comparison to Russell’s.

156
standpoint of positivism. I shall now develop this hypothesis. Suppose that a
philosopher acknowledging Wittgenstein’s arguments against Russell’s theory
of judgement concludes that a soul is nothing more than an array of states of a
brain. He could continue his considerations with the claim that with the death
of a brain the soul also dies. Since human brains die, human souls die also.
Our philosopher could be tempted to say that since one of the most important
elements of religious beliefs is the tenet that a human soul is immortal, then
science repudiates religious beliefs. Such a conclusion would show that the phi-
losopher in question believes in a positivist conception of science as able to
replace nonsensical theological views. In contrast to positivism, Wittgenstein’s
conclusion that a complex soul would no longer be a soul shows that scientific
research says nothing about a soul (neither that it is mortal nor that it is immor-
tal). Saying that science repudiates religious beliefs or the claim of the immor-
tality of a soul is, in the eyes of Wittgenstein, unjustified. On the other hand, he
was definitely not an apologist of religion and its notions. He held the notion of
the immortal soul to be meaningless, in the sense that it does not help to solve
any problems: not only philosophical or scientific but also existential ones:
Not only is there no guarantee of the temporal immortality of the human soul, that is to
say of its eternal survival after death; but, in any case, this assumption completely fails to
accomplish the purpose for which it has always been intended. Or is some riddle solved
by my surviving forever? Is not this eternal life itself as much of a riddle as our present
life? (TLP 6.4312).

But if Wittgenstein says that propositions containing the notion of a soul are non-
sensical, then what is the difference between his standpoint and the positivistic
one? In my opinion, the difference is significant: someone who sets the limits
of natural science in the way that Wittgenstein did cannot dream a positivistic
dream of a future in which traditional philosophical and religious questions will
be replaced by scientific ones. That difference was noticed by such an attentive,
positivistic and contemporary to Wittgenstein reader of the Tractatus as Otto
Neurath, who analysed the Tractarian metaphor of the ladder one has to throw
away in order to see the world aright (TLP 6.54). According to him, if one uses the
metaphor of a ladder, one should understand it as a picture of the constant pro-
cess of rejecting metaphysical theses in favour of scientific ones: “Every individ-
ual, in order to arrive at scientific knowledge, has temporary need of meaningless
word-sequences for ‘elucidation’ ” (my emphasis)344. But instead of a positivistic

344 Neurath 1959, p. 284.

157
image of science replacing metaphysics in the passage of time, Wittgenstein’s im-
age (modernist, as I call it) see metaphysics as a constant tendency of the human
mind345; a tendency that leads only to nonsense but, on the other hand, a tendency
which science cannot remove.
Note that the proposal that one should understand the notion of a soul as
standing for a complex of physical objects assumes that there is an aspect of
reality that the notion of a soul tries to capture, and that this aspect of reality is
better (more accurately, allows more precise predictions, etc.) expressed by sci-
entific notions. The positivistic picture assumes that philosophers try to explain
certain phenomena, such as, for instance, transtemporal identity, with the help
of the notion of self, and that this phenomenon could be better explained by sci-
entists studying the brain. For Wittgenstein, this conviction would be an exam-
ple of a distortion of philosophical notions which are not supposed to describe
or explain anything. For him, the notion of the immortal soul expresses “one’s
feeling that one has duties from which one cannot be released, even by death”346.
From the scientific point of view this formulation has no meaning at all: how
can one have duties even after one’s death? But this is exactly what Wittgenstein
tried to say: ethical and religious notions are not just empty notions – they are
necessarily empty and, therefore, hopelessly nonsensical notions. This, however,
means that not only are the expressions of somebody who believes in life after
death nonsensical but, in this respect, also the expressions of a scientist. We
stand before the choice: either we believe in some kind of progress from reli-
gious beliefs, through ontotheology up to the point of modern science, but then
one cannot see metaphysical expressions as nonsensical but simply as false, or
we sustain the view in which religion or ethics is nonsensical and we accept the
consequence that there is then nothing to be replaced, falsified or repudiated by
science. Either the notion of a soul refers to an object in the world, and it is pos-
sible to describe this object more precisely by means of scientific tools, or this
notion is necessarily empty; but then scientific investigations have nothing to
say with respect to the soul. If the above interpretation is correct, then not only
does thesis TLP 6.52 (“We feel that even when all possible scientific questions

345 “The running against the walls of our cage is perfectly, absolutely hopeless (…). But
it is a document of a tendency in the human mind” (LE, p. 12).
346 Malcolm 1984, p. 59. Interestingly, Russell also used the notion of the soul in this
non-philosophical meaning. Gustaw Herling-Grudziński quotes his conversation
with Russell about Joseph Conrad, during which Russell described Conrad as a
man who saw the horror of our soulless world earlier and better than anyone else
(Herling-Grudziński 2011, p. 258).

158
have been answered, the problems of life remain completely untouched”) limit
the authority of natural science and expresses the modernism of young Witt-
genstein. It turns out that modernist remarks are spread throughout the book,
including Wittgenstein’s discussion with Russell’s theory of judgement.
In the last section (3.3), I analysed Wittgenstein’s theory of judgement. Accord-
ing to it, one does not have to postulate the necessary relation to the mind in order
to explain the fact that propositions are true or false. What is truly needed is a
proper theory of meaning according to which in order to have a content a proposi-
tion has to be a picture and, more precisely, it has to have a logical form. Because
the notion of the empirical self is superfluous in the explanation, this means that
one did not confer any meaning on “A” (allegedly referring to the self) in the Rus-
sellian formula: “B (A, a, b, R, γ)” and, hence, the conclusion that the self – under-
stood as in the empirical tradition – does not exist. According to my interpretation
of the Tractatus, a belief consists in a relation between an event in the brain and a
state of affairs. Such a solution allowed Wittgenstein to be in accordance with the
physicalistic conviction that only matter has casual powers, and with the common
sense claim that beliefs explain, at least partly, human behaviour. At the same time,
my interpretation can explain the fact that Wittgenstein linked the theory of judge-
ment with the problem of the extensionality of language. According to my reading,
he must have held intensional contexts to be essentially and mutually combined
with a subject’s perspective on the world. By eliminating the category of a subject
he wanted to get rid of intensional contexts as well (section 3.3.3). In section 3.3.4
I sought the answer to the question as to why Wittgenstein put so much empha-
sis on the condition of the simplicity of the self. According to my hypothesis, the
reason for this is Wittgenstein’s intention to limit the authority of natural science.
As a concluding remark of the previous considerations I would like to pay
attention to the fact that in TLP 5.54ff there co-exist two projects: one is of the
naturalisation of beliefs and the second of limiting the range of application of
science. One aims to show the emptiness of philosophical categories, the second
aims to show that they nevertheless cannot be replaced by the concepts of natu-
ral science. Together they characterise the early 20th-century informal movement
of modernism. That is why I hold fragment TLP 5.54-5.5422 to be one of the best
arguments in favour of the modernist interpretation of the Tractatus.

3.4  Other interpretations of TLP 5.54-5.5422

As an appendix to Chapter 3 I want to compare my interpretation of TLP 5.54-
5.5422 with the others occurring in the literature. I have classified them into three
groups. The first group of commentators is interested in the question whether

159
Wittgenstein truly dealt with the problem of the intensionality of propositional
attitudes as such (section 3.4.1). The second group considers the problem if Witt-
genstein’s criticism of Russell’s theory of judgement resembles Hume’s critique of
Cartesian dualism of substances (section 3.4.2). Finally, the third group of com-
mentators ask if TLP 5.54ff is not a sign of Wittgenstein’s neglect of psychology
(section 3.4.3)?

3.4.1  Anscombe: TLP 5.54-5.5422 and the extensionality principle

Elisabeth Anscombe set the tone for the discussion if in TLP 5.54ff Wittgenstein
reconciled his requirement that all meaningful propositions have to be either
elementary propositions or truth-functions on elementary propositions with the
existence of intensional contexts in language. The expression: “A believes that p”
could not be, in Wittgenstein’s eyes, a form of a meaningful proposition because
it is not a truth-function on an elementary proposition. Wittgenstein solved this
paradoxical conclusion by saying that “A believes that p” is in fact of the form: “ ‘p’
says p”. If Wittgenstein’s analysis of “A believes that p” is the proper answer to the
challenge, then, as Anscombe points out, “ ‘p’ says p” has to be a truth-function
on an elementary proposition347. For the sake of the argument she concedes that
it is a proper solution. This means then that “ ‘p’ says p” is a meaningful proposi-
tion. In consequence, the discussion among scholars concentrated on the ques-
tion if “ ‘p’ says p” is truly a form of a meaningful proposition?
Anscombe thought that it is, although at first sight it seems that: “ ‘p’ says p”
is always true, hence not contingently true, hence meaningless. She goes back to
thesis TLP 3.1432, according to which the sign “aRb” says that a stands in relation
R to b. We can transform then “ ‘p’ says p” into “ ‘aRb’ says aRb” and then ask what
exactly in the propositional sign “aRb” says that aRb? Is it the fact that in “aRb”
“a” is written in italics and “b” in a Roman letter? This could be true, but we know
(by means of knowledge of linguistic conventions) that this is not the case. It is
the fact that in “aRb” “a” stands to the left and “b” to the right of “R” that says that
aRb. These two proposals (one false and one correct), in Anscombe’s eyes, show
that the form of a proposition: “ ‘p’ says p” can be understood in different ways.
The truth or falsity of: “ ‘p’ says p” depends on our choice of understanding. “The
fact that in ‘aRb’ “a” is written in italics and “b” in a Roman letter says that aRb”

347 “If Wittgenstein has not been careless, it must fit his general account of propositions –
that is, it must have true-false poles” (Anscombe 1965, p. 88).

160
is a false proposition348. “The fact that in ‘aRb’ “a” stands to the left and “b” to the
right of “R” says that aRb” is a true proposition. That entails that “ ‘p’ says p” “is a
genuine proposition with true-false poles”349.
On the other hand, such philosophers as Dayton or Cohen show dissatisfac-
tion with Wittgenstein’s analysis. According to them, “ ‘p’ says p” is a pseudo-prop-
osition. Michael Cohen says that “A believes that p” purports to say what it is that
A believes. Wittgenstein had to reject such propositions because, in the Tractarian
system, one cannot say what it is that A believes, but only show it350. One shows
what it is that A believes by indicating A’s utterances. That is why Wittgenstein
proposed the form: “ ‘p’ says p”. Unfortunately, according to Cohen, this is also
not the form of a significant proposition. As Eric Dayton pointed out, “ ‘p’ says p”
“is not in Wittgenstein’s technical sense of the word, a proposition; that “p” says
p should rather be shown, and therefore it cannot be said”351. “ ‘p’ says p” is an
example of Tractarian nonsense because it “cannot be a fact; it is used to convey
(strictly speaking, an inexpressible) picturing of one fact by another”352. But if: “ ‘p’
says p” is not a form of a significant proposition, then it is not the proper answer
to the challenge of the extensionality principle either.
What is my response to the above discussion? Is: “ ‘p’ says p” a form of a propo-
sition? When answering this question I want to stress the fact that I do not agree
with the common assumption of Cohen’s and Dayton’s interpretations that both
Russell and Wittgenstein were focussed on the problem as to how to express the
fact that propositions have meaning. Of course, taking into account Wittgenstein’s
views on meaning one cannot say what the meaning of a proposition is. If Russell’s
and Wittgenstein’s problem was how to express the fact that propositions have
meanings, then both solutions fail. When analysing the above-mentioned inter-
pretations, one has to remember that until 1967 the existence of the text with
which Wittgenstein argued, i.e. the manuscript of Russell’s Theory of Knowledge,
remained unknown, and until 1984 it was inaccessible. This had an impact on the
deficiencies of the above commentaries – Anscombe’s An Introduction to Wittgen-
stein’s Tractatus was originally published in 1959, Dayton’s and Cohen’s responses

348 “Although it contains a true description of the propositional sign as here occurring,
it is a false statement; for it is not, as it happens, this fact, but the fact that ‘a’ stands
to the left and ‘b’ to the right of ‘R’ that says that aRb” (ibid., p. 89).
349 Ibid., p. 89.
350 “ ‘A believes that p’ purports to say what is it that A believes. But this cannot be said”
(Cohen 1974, p. 444).
351 Dayton 1976, p. 277.
352 Ibid., p. 280.

161
come from the 1970s. In this chapter I tried to show that it was the Theory of
Knowledge which served as a background against which Wittgenstein formulated
his own thoughts. The problem that both philosophers were dealing with was the
question: “What does judgement consist in?”. Secondly, a comparison of fragment
TLP 5.54-5.5422 with the respective fragments of the Theory of Knowledge shows
that “ ‘p’ says p” is not Wittgenstein’s answer to the alleged difficulties of the form:
“A believes that p”, but to Russell’s formula: “B (A, a, b, R, γ)”. Therefore, his reason
for rejecting the propositional form: “A believes p” was not the impossibility of
saying the meaning of p but, as I think, incoherence of the concept of the empirical
self. In: “ ‘p’ says p” Wittgenstein encapsulated his own theory of judgement which
says that a judgement consists of a relation between an event in the brain and a
state of affairs. This theory manages to deal with the problem of the alleged con-
tradiction of the form: “A believes that p” with the extensionality principle, simply
by stating that there is no such thing as the self (the solution assumes, obviously,
that intensional contexts are internally related to the point of view of the subject).
Hence, in contradiction to Anscombe, Dayton or Cohen, I think that the problem
of the meaningfulness of “ ‘p’ says p” has very little to do with the question whether
Wittgenstein managed to defend the extensionality principle in his early oeuvre.

3.4.2 Hacker: Hume’s influence on Wittgenstein’s theory

The second group of commentators consists of philosophers who claimed that
Wittgenstein in TLP 5.54ff opposed the Cartesian conception of the self. For
example, Rosalind Carey, whose book is entirely devoted to Wittgenstein’s objec-
tions to Russell’s theory of judgement, claims that Russell defended Cartesian
dualism of mind and matter, and hence Wittgenstein’s remarks have anti-Car-
tesian significance353. Hans Sluga is convinced not only that Russell had ar-
gued in Theory of Knowledge for a Cartesian conception of the self, but that “it
was precisely this theory which Wittgenstein attacked in sections 5.54ff of the
Tractatus”354 (my emphasis).
I do not agree, however, with philosophers who understand Russell’s dualism as
substance-dualism and, hence, they see in the anti-Cartesian overtone of the Trac-
tatus 5.54-5.5422 a “re-make” of Hume’s classic argument that among things one
cannot find the self. For instance, David Stern thinks that Wittgenstein’s theory is
mostly negative and that its biggest achievement is a repudiation that the self is to
be found “within immediate experience be means of introspection, a ‘thinking,

353 Carey 2007, p. 4.

354 Sluga 1996, p. 324.

162
representing self ’ ”355. Jesse Prinz agrees with him and claims: “[Wittgenstein] can
be read as saying there is no self to be found in introspective experience – no phe-
nomenal I. When we think or perceive, we experience the contents of our thoughts
and perceptions, but not a subject of them”356. Dale Jacquette writes about “Witt-
genstein’s neo-Humean thesis about the no-ownership of psychological experi-
ences as a conclusion from his method for eliminating propositional attitudes
contexts”357.
Very often these claims occur together with another claim, namely that Witt-
genstein was also a proponent of a composite self. The most representative exam-
ples are the works of Peter Hacker, who writes, among others, that “Wittgenstein
was willing to adopt more or less Humean analysis of the empirical self ”358. Hack-
er argues that the relation of representing a fact consists in a correlation of two
facts rather than a fact and an object. Because facts are composite, only composite
things can “say” something. TLP 5.541 entails, according to Hacker, the claim that
The person A is not an object but a complex array of physical objects. ‘A believes p’ is
analyzable into a series of elementary propositions such that existence of the physical
constituents which corresponds to the constituents of the fact p is specified. These con-
stituents are related in some contingent way to whatever other facts constitute the person
A. (Hacker 1971, p. 165)

Hacker concludes from the premise:

“representations are composite”

that
“the person is composite (of physical objects)”

but for that to be correct one has to identify the flow of representations with the
self. In other words, from the fact that a representation is composite he concludes
that the self – as the container of representations – is also composite.
I would like to point to two problems of this interpretation. First, we should
take seriously Wittgenstein’s thesis that there is no such thing as a soul and that
a composite soul would no longer be a soul (TLP 5.5421). As I wrote earlier, ac-
cording to Wittgenstein it is better to say that there is no self at all than to claim
that the self is composed of material objects (even if we could successfully reduce
the flow of representations to something material). Of course I do not think that

355 Stern 1995, p. 73.

356 Prinz 2011, p. 147.
357 Jacquette 1992/1993, p. 199.
358 Hacker 1971, p. 165.

163
Hacker simply omitted TLP 5.5421. I suspect rather that his line of thought was
the following: in TLP 5.5421 Wittgenstein understood “soul” as the Cartesian
self – the immaterial subject of experience, its owner. It would indeed be absurd
to conceive such an object as composite. According to Hacker, Wittgenstein re-
jected in TLP 5.5421 only the Cartesian self – the subject of representations – but
not the Humean concept of the self. That is the second point of my disagreement
with Hacker. In my view, TLP 5.54-5.5422 is not a polemics with the Cartesian
notion of the self (understood as an immaterial substance), but with the notion
of the empirical self – the opinion about the self represented, among others, by
Russell, whose name is explicitly mentioned in the Tractatus. Moreover, if we ac-
cepted the Cartesian concept of the self – which includes that the self is the owner
of experiences of a mental nature – then no arguments used by Wittgenstein in
the analysed fragment of the Tractatus would be directed against such a concept.
An adherent of Descartes could easily accept all that Wittgenstein says about the
logical form as a necessary condition of a picture; he can agree that one does not
have to accept the existence of a mind synthesising raw sense-data material, and
he can still hold his dualistic views. Simplifying, if Wittgenstein’s argument says
only that among empirical reality there is no such thing as the self, then substance
dualists can reply that the self exists outside the empirical reality. Indeed, many
of the proponents of the thesis of Hume’s influence on TLP 5.54ff, including Peter
Hacker, are simultaneously adherents of the transcendental interpretation of the
Tractatus. Anthony Kenny and Hans Sluga are perfect examples, too; both admit
that Wittgenstein rejected the empirical self, but they also think that it did not
interfere with his belief in the existence of the metaphysical self, existing apart
from the material world. Kenny learns from Wittgenstein that the empiricist hy-
pothesis of the empirical self does not explain why propositions are true or false359
(the thesis I argued in favour of), but it does not prevent him from believing that
Wittgenstein ascribed the function of conferring a meaning on “the pure will of
the extra-mundane solipsistic metaphysical self ”360. Similarly, Sluga emphasises
Wittgenstein’s conclusion of the argument that “the notion of the soul or subject
is altogether incoherent and, consequently, there cannot be any such thing”361 (it
is also the thesis I argued in favour of), but nevertheless that leaves him open to
the question if perhaps there is a soul, but outside the world; not as a part of it362.

359 “The Tractatus itself did not think, as the British empiricists did, that impressions
and ideas could themselves confer meaning” (Kenny 1981, p. 147).
360 Ibid., p. 147.
361 Sluga 1983, p. 129–130.
362 Ibid., p. 130.

164
 In other words, some philosophers suppose that even if Wittgenstein thought
that the notion of the self as conceived in the empirical tradition is empty, this
does not mean that he rejected the notion of the self as empty in itself. It is pos-
sible, according to them, that Wittgenstein simply preferred the transcendental
understanding of this concept. I think that this hypothesis is wrong. Naturally, I
agree that in TLP 5.54ff the main aim of Wittgenstein’s argument was the concept
of the empirical self, and that does not exclude another understanding of the con-
cept of the self, but I think that the other theses of the Tractatus show that also
the transcendental notion of the self was under Wittgenstein’s attack. There is no
room for the task of proving this claim in this chapter, but I shall go back to it in
Chapter 4.

3.4.3  Jacquette: the distinguishability problem

The last group of interpretations raises the question whether Wittgenstein’s cri-
tique of Russell’s theory of judgement precludes a distinction between different
psychological states such as beliefs or doubts. One of the assumptions of Dale
Jacquette’s interpretation is the claim that Wittgenstein’s goal was to eliminate
propositional attitudes from a logical notation. According to Jacquette, “ ‘p’ says
p” is a “metalinguistic pseudoproposition purporting to describe the officially
unspeakable one-one correspondence relations between elementary proposi-
tions and atomic facts guaranteed by the picture theory semantics”363. It seems
that, according to Jacquette’s interpretation of Wittgenstein, a judgement con-
sists in a relation between an elementary proposition and a state of affairs. It also
seems that Jacquette holds a judgement to be the most general propositional
attitude, and beliefs or doubts to be different kinds of a judgement364. Beliefs
and doubts have the same form (“p” says p) but different psychological contents,
and that is why, in Jacquette’s eyes, Wittgenstein’s theory does not allow to dis-
cern between beliefs, doubts, desires, etc. Jacquette does not want to ascribe to
Wittgenstein an omission, therefore, he suspects that Wittgenstein intention-
ally left psychological contents outside the scope of his interests. He speculates
that Wittgenstein thought that discerning different kinds of judgements was
“an empirical task for the science of psychology, not a matter for philosophical

363 Jacquette 1992/1993, p. 195.

364 “Consider the sentences: ‘A believes that p’, ‘A doubts that p’. It is counterintuitive to
suppose that the analysis of these sentences could result in the very same reduction
in any correct theory of judgement” (ibid., pp. 195–6). Later, he writes about “the
contents of judgements that distinguish believing from doubting” (Ibid., p. 196).

165
semantics”365. It seems that, according to Jacquette, one can analyse A’s belief
that p as being composed of a psychological fact, which in Wittgenstein’s exam-
ple is a belief, and of a proposition with a content (in Wittgenstein’s example it
is a proposition “p” with a content p); hence the analysis of “A believes that p”
into: “Belief + [“p” says p]”. In fact, claims Jacquette, Wittgenstein was aware that
a judgement consists in two elements: a form – the subject-matter for seman-
tics, and a psychological state – the subject-matter for psychology366. Jacquette
claims that Wittgenstein probably thought that the problem of psychological
states is an empirical one and, therefore, of no interest to a philosopher. At the
same time, claims Jacquette, this conviction is the reason why Wittgenstein’s at-
tempt to “eliminate propositional attitude contexts from language at its deepest
level of analysis is faulty”367. It can work in simple cases such as “A believes that
p”. Wittgenstein tackles this propositional attitude by isolating a psychological
state from a propositional content: “Belief + [“ ‘p’ says p]”. But what about more
complex cases such as: “A believes that B doubts that q”? By analogy with the
former case, the form of this propositional attitude looks as follows: “Belief +
[“ ‘q’ says q]”, but because “q” stands for “B doubts that p”, we obtain in fact:
“[“[‘p’ says p] + Doubt” says [‘p’ says p] + Doubt] + Belief ”. It would turn out
that Wittgenstein did not get rid of the psychological (empirical) states from the
analysis of propositional attitudes at all368.
In response to Jacquette’s commentary, I would like to point to the fact that
Wittgenstein’s rejection of the propositional form “A believes that p” was not
entailed by his dismissal of psychology but was the consequence of his convic-
tion that it falsely interprets what beliefs consist in. Moreover, he thought that
propositional forms such as “A believes that p” are misleading in the sense that
they suggest the empirical self as being a term of relation with a state of affairs.
My objection to this interpretation is that it discerns between beliefs and judge-
ments, which is quite common in contemporary literature369 but unjustified with
respect to Wittgenstein. He held the notions of a judgement and of a belief to be

365 Ibid., p. 196.

366 Puhl thinks similarly, and he indicates at Wittgenstein’s intention of discriminating
psychology and philosophy as the motif for TLP 5.542. According to Puhl, Wittgen-
stein’s propositions of the form: “A believes that p” are the subject-matter for psychol-
ogy; propositions of the form: “ ‘p’ says p” are the subject-matter for philosophy (Puhl
1999, pp. 26–28).
367 Jacquette 1992/1993, p. 195.
368 Ibid., p. 210.
369 Crane 2001, pp. 102–3.

166
synonymous. In the Tractatus they pertain to dated and temporal events. Much
of Jacquette’s interpretation is based on this false discernment between judg-
ments, on the one hand, and beliefs, doubts and desires, on the other. It gives the
impression that Wittgenstein was interested only in abstract forms of proposi-
tions and their semantics. I do not think that Wittgenstein aimed in TLP 5.54ff at
a correct logical notation. His aim was rather to contrive a theory of judgement,
and if we accept that by “judgement” he understood a dated and temporal event,
then it turns out that he was interested in the question: “What does this psycho-
logical and mental act of judgement consist in?”. There is no reason to supple-
ment “ ‘p’ says p” by a psychological state; it simply leads to unsolvable problems.
This means, obviously, that I have a different view on what: “ ‘p’ says p” is. Ac-
cording to Jacquette, this expression purports to describe a relation between an
elementary proposition and a fact, but in fact it is a pseudoproposition, whereas
I do not think that “ ‘p’ says p” tries to describe anything. The proposition “aRb”
does not try to describe the form of the fact aRb. It says that aRb obtains and it
shows its form. Similarly, “ ‘p’ says p” says – under my interpretation – that an
event in the brain obtains. According to Jacquette, “ ‘p’ says p” gives only the form
of a judgement but not its content, whereas my interpretation allows to state that
“ ‘p’ says p” shows what the content of a judgement is.

Summary
In this chapter I analysed Wittgenstein’s rejection of the concept of the self pre-
vailing in empiricism. We can sum up the dilemma of empiricism as follows:
either we admit that beliefs belong to the region of the mental and we explain
in this way their truth or falsity but on the other hand we reject physicalism
(because it would turn out that some objects in the world are not ontologically
reducible to matter), or we defend materialism, reject the mental character of the
propositional attitudes and claim that there is nothing distinct between proposi-
tions and perceptions or emotions. We saw that Russell was the proponent of the
first solution and Hume of the second.
In the chapter I analysed the details of Russell’s and Wittgenstein’s theory of
judgement (sections 3.1 and 3.2). According to Wittgenstein, judgement consists
of a relation between an event in the brain and a state of affairs (section 3.3.2). In
my opinion, he killed in this way two birds with one stone:
I. The basic result of the Tractarian solution is that Wittgenstein succeeded in
avoiding the dilemma: he neither had to give up his materialism nor the truth-
values of beliefs. He solved the challenge of the physicalistic position, namely,
that rationality has a normative character that is non-reducible to matter by

167
means of his picture theory of meaning and the concept of the logical form that
was crucial for this theory. Pictures (or representations) are true or false not
because of the synthesising activity of the mind but just because of sharing the
same form with the depicted state of affairs. In order to explain that some enti-
ties (such as beliefs) are true or false, one does not have to introduce the mind.
Hence, Wittgenstein’s conclusion that a soul (understood as the empirical self)
does not exist. This claim is slightly ambiguous: it could suggest that after some
investigations one did not find an object corresponding to the characteristics
given in the concept. I do not think that this is what Wittgenstein had in mind in
his rejection of the empirical self; I believe that he wanted to say that one did not
confer any meaning on this concept (section 3.3.3).
II. Secondly, Wittgenstein showed how one can reconcile the thesis that all
meaningful propositions are truth-functions on elementary propositions with
the meaningfulness of propositions expressing propositional attitudes. Accord-
ing to Russell’s view, the former propositions are the results of operations on
propositions. The problem is that the modal operator B is not a truth-functional
operator. On the other hand, according to Wittgenstein’s solution, propositions
expressing propositional attitudes report the existence of an event in a brain,
so it is imaginable that they are elementary propositions. The tension between
the extensionality principle and the meaningfulness of propositions expressing
judgements does not arise (section 3.3.3).
In section 3.3.4 I also tried to show that the discussion with Russell’s theory
of judgement allows us to outline the strategy in which philosophy fulfils its task
of defending the world of human values against the impoverishing influence of
science. Speaking metaphorically, Wittgenstein thinks that there is one efficient
weapon left in the arsenal of philosophy, and that is setting the limits of mean-
ingful discourse:
It must set limits to what can be thought; and, in doing so, to what cannot be thought.
It must set limits to what cannot be thought by working outwards through what can be
thought (TLP 4.114).

In my opinion, the short remark: “A composite soul would no longer be a soul”

(TLP 5.5421) is an application of the above claim. What differentiates Wittgen-
stein from the standpoint of positivism is his opinion regarding the limitations
of science. This is why I think that, on the one hand, the fragment of the Trac-
tatus, 5.54-5.5422, is a good example of the existence of the book’s modernist
tendency. It shows that accepting modernism as an intellectual background of
the book helps in understanding its sometimes intricate and surprising formu-
lations. I think that these conclusions shall even be strengthened in the next

168
chapter, which is devoted to the great rival of materialism, i.e. transcendental
philosophy.
Finally, I shall try to present the “big picture” standing behind the results of this
chapter. I hope that, despite the fact that in the next remarks I shall go beyond the
exact results of my analysis of fragment TLP 5.54ff, I shall still say something that
is in accordance with the Tractarian spirit. In this chapter my considerations are
centred around the theme of dualism. Descartes, who was a great representative
of this philosophical tradition, had in Meditations on First Philosophy two goals:
to prove the existence of the immaterial soul and to prove the existence of God.
Descartes hoped that by achieving this goal he would provide strong fundaments
for religious beliefs against those who refused to see in the soul something differ-
ent than matter370. I am not going to analyse Descartes’ oeuvre; what is interesting
to me is that, according to him, it was precisely the task of philosophy to guarantee
the fundamentals for faith and moral values. Faith itself confronted with sceptics
is unable to do so:
For although it is sufficient for us Christians to believe by faith that the human soul
does not perish with the body and that God exists, yet it seems certain that unbelievers
cannot be convinced of the truth of religion, and scarcely even of any moral values, un-
less these first two truths are proved to them by natural reason. (…) And although it is
completely true that we should believe in the existence of God because it is taught in the
holy scriptures, and by the same token that we should believe the holy scriptures because
we have them from God (…) there is no point in asserting this to unbelievers, because
they would call it arguing in a circle [Descartes 1641/2008, p. 3].

In this context, the Tractatus with its fierce conviction that metaphysical propo-
sitions are nonsensical, that traditional philosophical concepts (such as the con-
cept of the soul) are empty, and that the world is completely described by natural
science is a witness to times when it is becoming clear that philosophy cannot
fulfil the task that Descartes put upon it. Ontotheology is helpless when it comes

370 “And as regards the soul, even though many authors have judged that it is very dif-
ficult to discover its nature, and some have even dared to say that human reasoning
convinces us that it perishes along with the body, and that we believe the contrary by
faith alone, nonetheless because the Council of the Lateran held in the reign of Leo X
condemns these people, and explicitly enjoins Christian philosophers to refute their
arguments, and to make every effort to prove the truth, I did not hesitate to tackle
this issue as well. Besides, I know that most of the impious refuse to believe that God
exists and that the human mind is distinct from the body, for no other reason than
that they say that these two points have never been proved by anybody up to now”
(Descartes 1641/2008, p. 4).

169
to defending the existence of the values of the human world in an age of the pro-
gress of science. I hope to prove this claim even more convincingly in the next
chapter when I analyse the influence of Schopenhauer’s transcendentalism on
Wittgenstein’s early oeuvre.

170
Chapter 4.  The Transcendental Self

Metaphysics fails in the task of finding a place for the soul, good or beauty in the
modern, scientific worldview, which is, according to my hypothesis, the claim of
the Tractatus. In the last paragraph we saw a critique of dualistic metaphysics that
sees the soul as a thing among other things. Wittgenstein in this respect belongs
to a large group of philosophers rejecting the Cartesian notion of the self. Some
of those philosophers undertook another philosophical effort to justify the exist-
ence of a non-reducible to matter self, and in this way to safeguard ethics and the
world of human values. In the history of philosophy this effort is called transcen-
dental idealism, and it will be the subject-matter of the following chapter.
In what follows I shall begin with the terminology, i.e. I shall summarise what
I mean under the term “the transcendental interpretation of the Tractatus”, and
under the concept of the transcendental self. Then I shall mention an impressive
group of scholars – proponents of the transcendental interpretation and their
arguments in favour of this interpretation. Extremely simplifying, the transcen-
dental interpretation claims that early Wittgenstein adopted from Schopenhauer
transcendental idealism. Hence, throughout the chapter I shall compare the phi-
losophies of these two philosophers (sections 4.1.1 and 4.1.2).
If the transcendental interpretation were true, then it would turn out that
Wittgenstein actually used metaphysics of the self in order to safeguard the
world of values and meaning. This is contradictory to my reading of his early
book, so I shall try to prove that arguments in favour of this interpretation
taken from the text of the Tractatus are not sufficient to justify such a reading.
These three fragments of the text are its ethical considerations (section 4.2), the
so-called solipsistic theses (section 4.3), and remarks on death (section 4.4). In
the conclusion of the chapter I intend to argue that, according to the Tractatus,
transcendentalism, similarly to dualism, fails to give meaning to the concept of
the self and, hence, fails to safeguard the values of the human world in the age
of scientism (section 4.5). This conclusion is in line with the main hypothesis
of this inquiry: that Wittgenstein belongs to a wide intellectual movement of
modernism and that he acknowledged that science replaces philosophy in the
task of describing reality, and although he sought an answer to the question of
the place of human values in a world governed by the laws of mechanics, he was
sceptical about the possibilities of philosophy in delivering answers. He rather
saw the solution in a radical turn to an inner, subjective life.

171
4.1  The transcendental philosophy of Schopenhauer
The transcendental interpretation is an interpretation which attributes to Witt-
genstein the thesis that the transcendental self exists, or the thesis that although
the claim about the existence of the transcendental self is nonsensical, it is some-
how an illuminating nonsense. I shall take the use of the term “the transcenden-
tal self ” to mean an entity which fulfils conditions (1) and (2) or an entity which
fulfils conditions (1) and (3):
1. An entity which is not a part of the empirical world but still there is a sense
according to which we can talk and think about it371
2. An entity to which the world is coordinated (this means that one cannot build
a wholly objective and materialistic worldview without assuming the existence
of a cognising subject), which is tantamount to saying that it has its world; an
owner of a point of view of the world as a whole372.
3. An entity which conditions the existence of the world. The world is depend-
ent in its existence on the transcendental self.
The term “the transcendental self ” is tantamount to the various other notions
used by commentators: “the solipsistic self ”, “the metaphysical self ”, and “the
willing subject”. By “transcendental” I shall mean every philosophy which claims
that the transcendental self exists. Terms synonymous with “transcendentalism”
in the following chapter are: “transcendental idealism”, “transcendental solip-
sism” and “idealism”.

4.1.1  Schopenhauer and the notion of the transcendental self

In the following section I shall develop the aforementioned ideas about the tran-
scendental self on the example of the philosophy of Schopenhauer.

371 For example, Bas van Frassen writes about the ego: “I exist, but I am not a thing
among things. I am neither a physical object, nor a mental substance or abstract
entity, nor a compound thereof. I am not animal, mineral, or vegetable. Nor am I a
thing constituted by or composed of things of that sort taken together. I am not some
piece of furniture of the universe” (van Frassen 2005, p. 87).
372 “No truth is more certain, more independent of all others, and less in need of proof
than this, namely that everything that exists for knowledge, and hence the whole
of this world, is only object in relation to the subject, perception of the perceiver”
(Schopenhauer 1844/1966a, p. 3).

172
 Ad 1) Schopenhauer claimed that the self is not a part of the empirical world:
The world as representation (…) has two essential, necessary, and inseparable halves.
The one half is the object, whose forms are space and time, and through these plu-
rality. But the other half, the subject, does not lie in space and time (Schopenhauer
1844/1966a, p. 5).
What is not in space or in time cannot be object; therefore the being or existence of
things-in-themselves can no longer be objective, but only of quite a different kind, namely
a metaphysical being or existence (Schopenhauer 1844/1966b, p. 7).

At first the idea of an object which is not a part of the empirical world but never-
theless exists as well as synonymous ideas of “non-factual existence”, “something
which exists, but not as a thing” and “something which exists outside the world”
could be confusing. When forced to explain what the concept of existence outside
the world means, I would use the Kantian category of things-in-themselves. Ac-
cording to transcendental thinkers, all that we know and perceive are just appear-
ances. The empirical world consists of appearances, so things-in-themselves – the
way objects are even if there is no cognising subject – transcend it. One should
not even use the word “object” when one refers to them. The category of an object
is misleading because the only objects we know are by definition appearances.
On the other hand, transcendentalists claim that they are sure of the existence of
things-in-themselves because only their existence explains the way the world of
appearances looks to us.
In the idealistic tradition the self is one of the things-in-themselves. Transcen-
dentalists claim that the self is not a part of the empirical world because its exist-
ence is different than the existence of empirical things, and in the more radical
versions of transcendental idealism, such as solipsism, its existence is even “great-
er” than the existence of empirical things because it is a source of the existence
of empirical things. On the other hand, when asked: “How do you know that the
self exists?”, the transcendentalists would answer that it is the only way to explain
certain phenomena. For example, we will see in the subsequent passages that the
justification for the postulate of the existence of the willing subject comes from
the need to explain the fact that we experience moral values. The concept of the
willing subject explains it because it captures the aspect of the self as a bearer of
moral values. Moral values enter the empirical world through the willing subject.
Ad 2) Schopenhauer endorsed the thesis that the material world is coordi-
nated with the subject in the sense that one cannot build a wholly objective and
materialistic worldview without assuming the existence of a cognising subject:
“No object without subject” is the principle that renders all materialism for ever impos-
sible. Suns and planets with no eye to see them and no understanding to know them can

173
 of course be spoken in words, but for the representation these words are a sideroxylon,
an iron-wood (Schopenhauer 1844/1966a, p. 29–30).

The next characteristics of the self in the idealistic tradition are that the world is
coordinated with the self. According to Glock, “Kant had refuted the Cartesian
doctrine of a soul-substance, but introduced two other notions: the ‘transcenden-
tal unity of apperception’ a formal feature of judgements, namely that they can
be prefixed by ‘I think’; and a noumenal self, the locus of free will and the moral
law”373. I explained the notion of a noumenal self in the previous point. What does
“transcendental unity of apperception” mean? For our purposes it is not neces-
sary to go into the details of Kantian philosophy because this notion was trans-
formed by Schopenhauer into the formula “the world is my representation”374,
and in this form it was filtered to Wittgenstein. The most important consequence
of Schopenhauer’s catchword is that if the world is a representation, then there
must be someone who perceives it – a subject375. The world, under this concep-
tion, is coordinated to the subject. By the concept “coordination” I understand
the fact that the mere indication at the existence of the world signals the existence
of the subject who perceives this world.
The transcendentalists would agree with the anti-Cartesian thesis of      TLP
5.5421 that the self is not a part of the empirical reality, so in order to express
the relation between the self and the world one needs a different concept than
“being a part of ”. It is the concept of ownership or of subordination. The relation
of ownership is not like the relation between Roman Abramovich (the owner)
and the Chelsea London football club (the property). Both Roman Abramovich
and his football club are parts of the empirical world, and both Roman Abra-
movich and Chelsea London would exist independently, without any relation
holding between them. In order to express the relation of ownership between
the self and the world, I shall make an analogy to the relation of substance and
accidents in Aristotle’s metaphysics. Substance, according to this tradition of
metaphysical thinking, is something which can exist on its own; accidents are
something that exists only instantiated by the substance376. Let us make a horse
an example of a substance (horses do not need any other things in order to ex-
ist, so they are a good example) and its properties, e.g. shape, size, properties of

373 Glock 1996, p. 348.

374 Schopenhauer 1844/1966a, p. 3.
375 “The whole of this world, is only object in relation to the subject, perception of the
perceiver, in a word, representation” (ibid., p. 3).
376 Martin 1980/1999, p. 37.

174
character and so on as examples of accidents. What is important in this analogy
is that although the properties of the horse are parts of the perceivable world
(I can see the size and the shape of a particular horse), the horse in itself – the
bearer of the properties – is not. Aristotelian substances are not perceivable. If
we replace “perceivable” by “empirical”, “substance” by “the self ”, and “proper-
ties” by “the empirical world”, then we will get the intuition of transcendental
idealism. Although the empirical world is perceivable and, hence, is a subject-
matter for scientific research, its bearer – the self – is not. What could also be
made of this analogy is the relation of coordination between a substance and its
properties: both are conceivable only together. We cannot conceive accidents
hanging in the air without the support of a substratum (one cannot conceive
the accident “tall” in itself without the connection with things that have this
feature), but, on the other hand, we cannot conceive pure substrata either –
things without any accidents at all. The same goes, according to the transcen-
dentalists, for the relation between the self and the world. Thinking about the
self independently of the subordinated world would be tantamount to repeat-
ing the mistake of the dualists, but on the other hand the idea of an external
world existing independently of the self would mean to commit a mistake of
materialism. According to the transcendentalists, we cannot conceive matter in
any other way than in relation to the subject. Having an idea of the matter is al-
ways conditioned by the existence of the subject that has that idea377. Hence the
conclusion that: “Every object always and eternally presupposes a subject”378.
Ad 3) Schopenhauer very clearly formulated the thesis of the dependence of
material things in their existence on the subject and the intellect:
We must absolutely deny to the dogmatist the reality of the external world, when he
declares this to be its independence of the subject. The whole world of objects is and

377 “Everything objective, extended, active, and hence everything material, regarded by
materialism as a solid basis for its explanations, all this is something that is given
only very indirectly and conditionally, and is therefore only relatively present, for
it has passed through the machinery and fabrication of the brain” (Schopenhauer
1844/1966a, p. 28).
378 Ibid., p. 95. In this respect to these theses Schopenhauer is a debtor of Berkeley,
whom he held in a high esteem: “The whole of this world, is only object in relation
to the subject, perception of the perceiver, in a word, representation (…). This truth
is by no means new (…). But Berkeley was the first to enunciate it positively, and he
has thus rendered an immortal service to philosophy, although the remainder of his
doctrines cannot endure” (Ibid., p. 3).

175
 remains representation, and is for this reason wholly and for ever conditioned by the
subject; in other words, it has transcendental ideality (Schopenhauer 1844/1966a, p. 15)
Matter and the course of nature become mere phenomenon, conditioned by the intellect;
for the phenomenon has its existence only in the representation of the intellect (Scho-
penhauer 1844/1966b, p. 13).

The epistemological thesis that the world is given to a subject only through experi-
ences, so the only world the subject knows is the world of the subject’s experiences,
starts to have peculiar consequences when expressed in ontology by the identity:
“the world = my world”, which one can also find in the Tractatus379. Jointly, these
theses do not only state that the self has its world, but they also comprise the
core of transcendental solipsism, i.e. the conviction that the world (the totality of
things, reality, the empirical) is dependent in its existence on the self, and that the
self is a condition of the existence of the world.
There is one thing that is troubling and confusing in an otherwise helpful
analogy, i.e. the analogy between the transcendental self and the Aristotelian
substance. It should be clear that the analogy works only insofar as it involves
the explanation as to what the relation between the transcendental self and the
empirical world consists in, but not in explaining the nature of the transcen-
dental self. As little as we can say about the transcendental self, one thing is
certain: because it exists outside the realm of empirical objects, and because
it does not subsist on its own, the category of a substance does not apply to it.
We are left only with metaphors with respect to the transcendental self; for ex-
ample, Schopenhauer compares the self with an eye380; similarly, Wittgenstein
uses an analogy of the eye and its visual field381. This picture serves to present
the self as an extensionless point rather than a particular object (which is al-
ways in space and time): “The I of solipsism shrinks to an extensionless point
and what remains is the reality co-ordinate with it” (NB 2.9.16, p. 82).
The other metaphor is the metaphor of a perspective or of a point of view sub
specie aeternitatis. There is nothing idealistic or transcendental in speaking about
different points of views and perspectives in philosophy when it discusses, for ex-
ample, intentionality382. What is exceptional in a point of view sub specie aeternitatis

379 TLP 5.621–5.63.

380 “The world as representation certainly begins only with the opening of the first eye,
and without this medium of knowledge it cannot be, and hence before this it did not
exist” (Schopenhauer 1844/1966a, p. 31).
381 TLP 5.6331; NB 12.8.16, p. 80.
382 For example, Crane claims that: “What the daffodil lacks and the ‘minded’ creature
has is a point of view on things or a perspective. The minded creature is one for which

176
is that it is a point of view on the world as a whole and, hence, placed outside the
world: “The usual way of looking at things sees objects as it were from the midst of
them, the view sub specie aeternitatis from outside” (NB 7.10.16, p. 83). Combined
with the assumption that whenever there is a point of view there is someone who
has this point of view, the claim that a point of view is outside the world leads to
the conclusion that there is someone outside the world – the transcendental self.
One can also find in the commentaries of the Tractatus the notion of an ethical
point of view. Why can this term bring to mind transcendental associations? To
answer this question one should go back to Schopenhauer’s claim that an act of
will and an action are two sides of the same coin383. For Schopenhauer, the unity
of the act of the will and the action meant that we are given to ourselves in two
different ways: as acting we perceive ourselves as objects among other objects,
and bond with them by the laws of nature. On the other hand, as willing, desir-
ing, etc., we are given to ourselves directly and immediately as the will in itself, i.e.
the subjects of the will384. Although only the first point of view is translatable into
the language of natural science, both of them are equally important. The second
point of view is even more valuable because it refers to “the original forces”385, i.e.
the vital forces which explain the rules and laws of mechanics, physics and chem-
istry, such as “the forces of impenetrability, gravitation, rigidity, fluidity, cohesion,
elasticity, heat, light, elective affinity, magnetism and electricity”386. Without an
explanation in terms of the original forces (for Schopenhauer it is obviously the
will), they remain mere phantoms and qualitates occultae387.
We have two points of view from which we have access to reality and to our-
selves; from each point of view we learn something different. In ethics we talk
about ourselves as willing and desiring. It is then ethics which discovers reality

things are a certain way: the way they are from the creature’s perspective” (Crane
2001, p. 4). This claim, naturally, does not entail any kind of idealism.
383 “The act of will and the action of the body are not two different states objectively
known, connected by the bond of causality; they do not stand in the relation of cause
and effect, but are one and the same thing, though given in two entirely different
ways” (Schopenhauer 1844/1966a, p. 100).
384 “To the subject of knowing, who appears as an individual only through his identity
with the body, this body is given in two entirely different ways. It is given in an intel-
ligent perception as representation, as objects among other objects, liable to the laws
of these objects. But it is also given in quite a different way, namely as what is known
immediately to everyone, and is denoted by the word will” (ibid., p. 100).
385 Ibid., p. 124.
386 Ibid., p. 122.
387 Ibid., p. 122.

177
that is inaccessible to natural science, i.e. the kingdom of will. The owner of the
ethical point of view – the willing subject – is, therefore, often said to be the
transcendental self.

4.1.2  Schopenhauer and the safeguarding of values

In the last paragraph I described the notion of the transcendental self. This self is
not a part of an empirical world, it is rather a point of view which takes the world
as a whole (and is, hence, placed outside the world). From this point of view we
gain deeper insight into the nature of the world.
According to its proponents, one of the main reasons to accept the transcen-
dental interpretation of the Tractatus is the claim that it reflects Wittgenstein’s
desire to safeguard the values of the human world from the standpoint of scient-
ism. Although I agree that the Tractarian point of departure involves the ques-
tion of the place of moral values, aesthetics and the soul in the world of scientific
facts, I do not agree that the transcendental interpretation rightly reflects Witt-
genstein’s answer to this question. But firstly I want to present, on the example
of the philosophy of Arthur Schopenhauer, what defending the world of human
values against the claims of science could look like in the case of transcenden-
tal idealism. What is the connection between the doctrines of idealism and the
modernist pursuit of finding a new way of expression for human values apt to
the requirements of the modern age? Schopenhauer held science as limited by
the principle of sufficient reason. He adopted Kantian thinking regarding this
principle. Therefore, in his eyes, it is not a law of nature; it rather encapsulates our
way of organising empirical data. In consequence, according to Schopenhauer,
what is governed by the principle of sufficient reason are not things-in-them-
selves but appearances, i.e. not noumena but objects as we perceive them. In this
sense objects depend in their existence on the cognising subject. In other words,
Schopenhauer saw the deficiency of science in the fact that it refers only to ap-
pearances. It describes the relations holding between them, the laws that govern
them, or the connections that take place between them388, but it never reaches
the core of reality or, in other words, the sphere of noumena. Perhaps we can
find in the Tractatus a vague echo of this point in the thesis where Wittgenstein
criticised the “modern conception of the world” for idolising the laws of nature

388 “All these, the common name of which is science, therefore follow the principle of
sufficient reason in its different forms, and their theme remains the phenomenon,
its laws, connexion, and their relations resulting from these” (ibid., p. 184).

178
and the illusion that they explain natural phenomena389, or in A Lecture on Eth-
ics, where Wittgenstein imagined what a scientific explanation in the case of a
miracle could look like. All that science could do in such a case would be to state
a sequence of events: in t1 a state of affair A, then in t2 a state of affair B, and in t3
a state of affair C. Because all a scientist could say would be 1) “A is the cause of
B, and B is the cause of C”, and 2) “We have never encountered such a sequence
of events before”390, it is thus, for Wittgenstein, an example of the limitation of
science by the principle of sufficient reason.
Many shades and aspects of reality slip through the scientific conceptual net.
Science loses the uniqueness and meaning of some events. According to Scho-
penhauer, the situation of somebody who would, nevertheless, want to describe
aspects of reality that are non-attainable to science is not helpless. It is enough for
him to abandon a futile, in this respect, language of science and to employ a dif-
ferent kind of discourse. One possibility is art because it grasps what is essential
in objects and their eternal ideas391. Another way to “lift the veil of deception”
is, according to Schopenhauer, a mystical experience392. For the sake of the argu-
ment I shall concentrate only on the third source of knowledge about the reality
of things-in-themselves, and that is metaphysics.
To be precise, Schopenhauer thought that every philosophy, even one deny-
ing the discernment between appearances and things-in-themselves, or one fully
trusting the cognising capacities of science, has its own metaphysics, in the sense
of a theory of what exists. The precursors of modern physicalism, such as Lamarck,
Hollbach and Cabanis, as mentioned by Schopenhauer, who take all phenomena
to be physical are also metaphysicians, with the exception that their metaphysics is

389 “The whole modern conception of the world is founded on the illusion that the so-
called laws of nature are the explanations of natural phenomena (…) People today
stop at the laws of nature, treating them as something inviolable (…) The modern
system tries to make it look as if everything were explained” (TLP 6.371).
390 LE, p. 10.
391 “What kind of knowledge is it that considers what continues to exist outside and
independently of all relations, but which alone is really essential to the world, the
true content of its phenomena (…)? It is art, the work of genius. It repeats the eternal
Ideas apprehended through pure contemplation, the essential and abiding element
in all the phenomena of the world” (Schopenhauer 1844/1966a, p. 184).
392 “Mysticism is every guidance to the immediate awareness of that which is not reached
either by perception or conception or generally by any knowledge (…) The mystic
starts from his inner, positive, individual experience, in which he finds himself as
the eternal and only being, and so on” (Schopenhauer 1844/1966b, p. 611).

179
identified with physics393. Schopenhauer had some bitter words to say about such
metaphysics (“it would look almost like Holberg’s theatrical pot-house politician
who was made burgomaster”394), and the most important reason for his outrage is
that such metaphysics would eliminate ethics. This, in consequence, would have
a damaging effect on morality395. This threat led Schopenhauer to formulate his
own credo:
We can therefore set this up as the necessary credo of all righteous and good men: “I
believe in a system of metaphysics” (Schopenhauer 1844/1966b, p. 175).

As I mentioned earlier, according to Schopenhauer a scientific explanation is

limited because it pertains only to appearances, and that is why a “physical expla-
nation (…) requires one that is metaphysical, which would furnish the key to all
its assumptions”396. Metaphysics in this second meaning is not a theory of what
exists but a theory passing beyond the phenomenal appearance. The possibility
of passing beyond is grounded in the fact that we experience ourselves not only
as appearances, i.e. as bodies, but also as things-in-themselves, i.e. the will. The
self has a special status in Schopenhauer’s metaphysics because it possesses in-
sight into the core of reality, which stands behind all appearances:
Man carries the ultimate fundamental secrets within himself, and this fact is accessible
to him in the most immediate way. Here only, therefore, can he hope to find the key to
the riddle of the world, and obtain a clue to the inner nature of all things (Schopenhauer
1844/1966b, p. 179).

Summing up, Schopenhauer regards scientific explanations, though they explain

the whole realm of appearances, as insufficient with respect to the moral aspect
of human life. Secondly, he thinks that for this reason we need metaphysics,
which could pass beyond what is empirically perceived. Thirdly, he hopes that
we – the selves – have access to what is beyond appearances – to the will397.

393 “Such an absolute system of physics as described above, which would leave no room
for any metaphysics (…) would be physics seated on a throne of metaphysics” (ibid.,
p. 175).
394 Ibid., p. 175.
395 “Certainly such a system would necessarily be destructive for ethics” (ibid., p. 175).
396 Ibid., p. 173.
397 It is the metaphysics of the will that enables meaningful discourse on ethics and jus-
tification of moral tenets. The same goal is unachievable to science, simply because
the condition of morality – the will – is a thing-in-itself and, hence, out of the scope
of scientific interest.

180
 What, in my opinion, is the most tempting reason to accept the transcenden-
tal interpretation of the Tractatus is the recognition that there occurs the same
dissatisfaction as presented by Schopenhauer:
We feel that even all possible scientific questions have been answered, the problems of
life remains completely untouched (TLP 6.52),

and perhaps the same cure for the disease, i.e. metaphysics of the self. One can
simply point at the well-known admiration that Wittgenstein had towards Scho-
penhauer’s philosophy to conclude that Wittgenstein accepted the transcenden-
tal solution to the problem of the imperfection of science. The question I want to
ask in this chapter is whether Wittgenstein used the same way of “safeguarding”
the special position of human beings in the world or the existence of moral val-
ues? Is the occurrence in the Tractatus of such notions as the notion of the meta-
physical self, or the notion of a point of view sub specie aeternitatis, explained
by the fact that Wittgenstein wanted to cross the limits of science and wanted to
suggest that we somehow have access to the metaphysical realm?
In this section I have concentrated on the possible main reason to accept the
transcendental interpretation – Wittgenstein’s desire to safeguard values. The
advantage of the transcendental interpretation consists in the fact that it can
both reflect this desire as well as show the way out for those who are unsatisfied
with the results of natural science. In the subsequent sections I shall analyse
particular versions of the transcendental interpretations. I will group them into
three categories. First, I will analyse the interpretations based on the ethical
theses of the Tractatus (4.2). Then I shall mention these referring to solipsism
of the Tractatus (4.3) and, finally, I shall refer to one interpretation underlining
Wittgenstein’s remarks about death (4.4). Every presentation will be concluded
with my response to the arguments. Finally, I shall present some general re-
marks about the failure of the transcendental interpretation (4.5), and in the
next chapter (5) I will present my own point of view on how to interpret claims
which raise suspicion regarding Tractarian transcendentalism.

4.2  The willing subject

First, I shall refer to a group of authors who claim that Wittgenstein’s ethical
views betray that he, during the period of writing the Tractatus, was a transcen-
dental idealist. They find confirmation of such a hypothesis in the parts of the
book dealing with ethics. The central notion of this interpretation is the con-
cept of the willing subject. As the proponents of the interpretation prove, this
concept fulfils all requirements put on the transcendental self: its referent is not

181
a part of the empirical world, but at the same time it guarantees the existence
of moral values, and in this sense it conditions the world as we know it. They
would also add that we have evidence in Wittgenstein’s writings that he be-
lieved in the existence of the willing subject. Therefore – conclude the adherents
of the transcendental interpretation – we have no other option than to accept
transcendentalism in the work of early Wittgenstein. I shall begin the section
with a short review of authors who represent such a way of thinking about the
Tractatus. Then I shall refer to the argument saying that since the concept of
the willing subject occurs in Wittgenstein’s texts, and the concept is defined in
such a way that it fulfils all the requirements put on the concept of the transcen-
dental self, then one has to accept a transcendental reading of the Tractatus.
My response consists in, to simplify things, pointing out that the concept of
the willing subject occurs only in the Notebooks, so at best one could talk about
transcendentalism in the Notebooks. I shall also argue that a comparison made
between the respective theses of the Tractatus and entries from the Notebooks
suggests that Wittgenstein ultimately withdrew from transcendentalism with
respect to ethics.

4.2.1 Examples of transcendental interpretations referring to

Wittgenstein’s ethics
According to Hidé Ishiguro, Wittgenstein borrowed from Schopenhauer a model
according to which we have at our disposal two different points of view of the
world, i.e. the scientific-objective-materialistic one and, the second, the personal-
ethical one. She speaks about the discovery of the ethical perspective by Witt-
genstein in the empirical and material world, with which I agree. This is a typical
way to describe a modernist point of view. On the other hand, I do not agree
with her conviction that the cost of this double-model is the acceptance of the
existence of the subject – the will, i.e. an individual that “chooses to view the
whole to which the object or event belong and to which he adopts an attitude”398.
Her reasoning is the following: since the willing subject has a perspective on the
world as a whole, then the self must withdraw from it. Hence, the willing subject
resides outside the world. We recognise in Ishiguro’s paper the following form
of reasoning: there are moral values. The existence of moral values is incompre-
hensible on the basis of the scientific-objective-materialistic point of view. Then
there must be another point of view and an owner of this point of view that is not
a part of the empirical reality. Hence the willing subject is the transcendental self

398 Ishiguro 1981, p. 459.

182
and one must accept the transcendental interpretation of the Tractatus. A simi-
lar position is taken by John Kelly. He also very correctly describes the point of
departure for Wittgenstein’s considerations on ethics: “Wittgenstein is concerned
to understand how there can be meaning and worth in the seemingly nihilistic
world of modern science where the order of nature has been divorced from any
conception of value”399. In my opinion, he falsely concludes that Wittgenstein, in
order to prevent us from scientific nihilism, uses metaphysics, and his answer to
the challenges of the modern era is to contrive a theory of the metaphysical self:
“Wittgenstein’s position is that the existence of ethical meaning and value is also
the result of the constituting activity of the metaphysical subject”400.
Martin Stokhof also notes the central opposition in the Tractatus between two
points of view: in his case there is one which apprehends the world as a complex of
facts and objects and one which apprehends the world as a whole. Stokhof claims
that “on the level of subject this distinction reappears as that between the world
from the point of view of the individual, psychological subject, which is in the
world as a subject among other subjects, and the world viewed by the metaphysi-
cal subject, which is not in the world, but constitutes its limits”401. From the first
point of view there is no necessary connection between my will and the events
in the world: “this is one sense in which one may uphold that will and world are
independent”402. If the world is independent of will, we are forced to accept deter-
minism, and that has a damaging effect on ethics because it makes sense to ascribe
to an action the value of good or evil only under the condition that a subject had
an influence on the action in question. This causes tension between the points of
view as mentioned above. To solve the impasse and to safeguard moral values, as
we could expect, Stokhof ’s interpretation introduces a metaphysical notion: “in
the case of Wittgenstein it is the will as the bearer of ethical values that escapes the
logical contingency of the world of language and thought”403. The will is apparently
identified by Stokhof with the metaphysical self. It is not the will in an empirical
sense (the will as experienced in desires) but “the will which presents the ethical
aspect of action”404.

399 Kelly 1995, p. 571.

400 Ibid., p. 573; Rosenberg shares this standpoint: “The existence of the metaphysical
self (…) is required by Wittgenstein’s doctrines of the will and of good and evil”
(Rosenberg 1968, p. 170).
401 Stokhof 2002, p. 233.
402 Ibid., p. 208; TLP 6.374.
403 Stokhof 2002, p. 208.
404 Ibid., p. 209.

183
 Stokhof ’s position could be summed up as follows: ethical values are intrinsic
aspects of our actions, they are an intrinsic dimension of reality (and as aspects
or dimensions they cannot be depicted, and hence ethics is senseless according to
Tractarian criteria of meaning). Moreover, the notion of an aspect of a thing or of
an aspectual shape requires the existence of a certain point of view. There is noth-
ing transcendental up to this moment: one can naturally talk about some aspects
of reality that we notice only when we have a certain attitude to the world, without
any ontological implications in mind. But in the case of Stokhof ’s interpretation it
is a point of view sub specie aeternitatis, which requires that an owner (we refer to
this owner by using the notion of the metaphysical subject) of this point of view
is positioned outside the world and, hence, his interpretation could be rightfully
named as transcendental405.
What should be clear after this short review is insistence on the fact that Witt-
genstein’s transcendentalism is an upshot of his observation that science impov-
erishes the human world of values and that he hoped to find a solution to this
problem in the metaphysics of the self. Now I shall examine what arguments the
proponents of such an interpretation use in order to support their hypothesis.

4.2.2 Counterarguments
In order to support the transcendental interpretation, its proponents often indi-
cate certain entries from the Notebooks. I will argue that there is a clear influence
of Schopenhauer on Wittgenstein’s thinking in the Notebooks – one can inves-
tigate a form of transcendentalism there. But at the same time one observes a
definite change in Wittgenstein’s ethical views in the Tractatus. Even if there is a
resemblance between the ethics of the Tractatus and Schopenhauer’s ethics, it is
not enough to state that the former oeuvre represents the position of transcen-
dental idealism.

4.2.2.1 Transcendentalism of the Notebooks and the shift in

Wittgenstein’s thinking about ethics
We saw earlier that for Schopenhauer the empirical reality is morally indiffer-
ent and the way to safeguard the position of moral values in our worldview is to
introduce the notion of the subject as a bearer of moral values. One could suc-
cessfully argue that this kind of reasoning is present in the Notebooks, in which

405 For now I leave aside the question if the inference from viewing the world as a whole
to the existence of an owner of this point of view who exists outside the world is right.

184
Wittgenstein claimed that the act of a will and the action are not two separate
states406 – a thought which, as I have mentioned, led Schopenhauer to contrive
a theory about two different points of view of ourselves and the world.
Secondly, in the Notebooks Wittgenstein identified the concept of the subject
with the concept of the willing subject407, and the description of the concept of
the willing subject is such that it fulfils all the conditions put on the transcen-
dental self.
Ad a) The willing subject as the bearer of ethics is not a part of the empirical:
Good and evil only enter through the subject. And the subject is not part of the world,
but a boundary of the world (NB 2.8.16, p. 80).

Moreover, in the Notebooks there is a conviction that, although apart from the
world, the self, the I, or the will – all of these concepts seem to be for Wittgen-
stein synonymous with the concept of the willing subject – somehow exists:
The thinking subject is surely mere illusion. But the willing subject exists (NB 5.8.16, p. 80).
There are two godheads: the world and my independent I (NB 8.7.16, p. 74).

and acts:
My will penetrates the world (NB 11.6.16, p. 73).

Ad b) The willing subject is also the subject of a particular experience, namely

the moral one. It is the bearer of moral values:
What really is the situation of the human will? I will call “will” first and foremost the bearer
of good and evil. Let us imagine a man who could use none of his limbs and hence could, in
the ordinary sense, not exercise his will. He could, however, think and want and communi-
cate his thought to someone else. Could therefore do good or evil through the other man.
Then it is clear that ethics would have validity for him, too, and that he in the ethical sense is
the bearer of the will (NB 30.7.16, p. 76–77).

Ad c) Finally, Wittgenstein straightforwardly writes that the willing subject, and

in consequence ethics, is the precondition of the existence of the world. The sub-
ject is also called in this context “the centre of the world”:
The subject is not the part of the world but a presupposition of its existence (NB 2..8.16,
p. 79).

406 “The act of the will is not the cause of the action but the action itself ” (NB 4.11.16,
p. 87).
407 “The subject is the willing subject” (NB 4.11.16, p. 87); “The thinking subject is surely
mere illusion. But the willing subject exists” (NB 5.8.16, p. 80).

185
 If the will did not exist, neither would there be that centre of the world, which we call the
I, and which is the bearer of ethics (NB 5.8.16, p. 80).
Ethics does not treat of the world. Ethics must be condition of the world, like logic (NB
24.7.16, p. 77).

Are the above quotations enough to secure the validity of the transcendental in-
terpretation of the Tractatus? Is the willing subject the Holy Grail of the transcen-
dental interpretation? When answering this question it should not escape our
attention that they are to be found only in the Notebooks, in which Wittgenstein
admitted that he was not certain about his ideas (“As for what my will is, I don’t
know yet” (NB 8.7.16, p. 74).) and that the Notebooks’ formulations are vague:
directly after a passage about the willing subject as the limit of the world and
directly before the passage about the subject as the precondition of the world’s ex-
istence, Wittgenstein notes: “I am conscious of the complete unclarity of all these
sentences” (NB 2.8.16, p. 79). To state the obvious, the basis for the interpretation
of the Tractatus is the text of the Tractatus and not of the Notebooks.
In comparison to the Notebooks it is easy to notice important changes in the
text of the Tractatus with respect to ethics. What is the most significant sign of
the shift in Wittgenstein’s thinking about ethics is that from all the quotations
from the Notebooks that confirm he was thinking about the willing subject as the
transcendental self, literally NONE occurs in the Tractatus. He disposed of ALL
of them. There is no mention of the concept of the willing subject, no mention of
any kind of bearer of ethics. On the contrary, now we read that “it is impossible
to speak about the will in so far as it is the subject of ethical attributes” (TLP
6.423). There is no mention that it is the will which is good or evil, no mention
that an act of will and an action are one. A person reading only the Tractatus
would have no idea that Wittgenstein was interested in the problem: “how do
good and evil enter into the world?”; not to mention that he claimed that the
willing subject is the condition of the existence of the world. In this respect one
finds in the Tractatus no evidence supporting the transcendental interpretation.
Perhaps a comparison of some theses from the Notebooks and their counterparts
from the Tractatus would be the most convincing; for example, the thesis, which
I have already quoted, reads in the Notebooks as follows (I shall give the original
German text in order to show the parallelism between the respective texts from
the Notebooks and the Tractatus):
Good and evil only enter through the subject. And the subject is not the part of the
world, but a boundary of the world. In the original: Gut und Böse tritt erst durch das
Subjekt ein. Und das Subjekt gehört nicht zur Welt, sondern ist eine Grenze der Welt.
(NB 2.8.16, p. 79).

186
But in the Tractarian counterpart one reads only that:
The subject does not belong to the world; rather, it is a limit of the world. In the original:
Das Subjekt gehört nicht zur Welt, sondern es ist eine Grenze der Welt. (TLP 5.632)

In another example we could compare the following entry from the Notebooks with
TLP 5.631:
The thinking subject is surely mere illusion. But the willing subject exists. In the original:
Das vorstellende Subjekt ist wohl leeren Wahn. Das wollende Subjekt aber gibt es. (NB
5.8.16, p. 80)

Very interestingly, in the Tractatus Wittgenstein gave up the second part:

There is no such thing as the subject that thinks or entertains ideas. In the original: Das
denkende, vorstellende, Subjekt gibt es nichts. (TLP 5.631)

Instead of speculations about an entity which somehow exists and acts although
it does not belong to the world, in the Tractatus we simply get information that
the subject is a limit of the world, which disposed of the background of consid-
erations about how good and evil enter the world, could just be read as a claim
that one cannot speak meaningfully about the subject of the will.
What happened then? What does the shift between the Notebooks and the final
formulations in the Tractatus consist in? In my opinion, in the Notebooks the in-
fluence of Schopenhauer is clear: not only did Wittgenstein agree that according
to the scientific worldview reality contains no moral values, but he also agreed
that safeguarding ethics reaches fruition by going beyond the reality of facts and
postulating an entity which would, on the one hand, secure the existence of good
and evil and, on the other hand, would not belong to the world of empirical facts.
Just as Schopenhauer, he identified this entity with the self, namely the willing
subject.
My supposition is that by the time of composing the final formulations of the
Tractatus, Wittgenstein realised that this kind of “safeguarding” ethics is just a
void gesture, a form of wishful thinking. If, according to the Tractarian theory of
meaning, the only meaningful propositions belong to natural science, then ethics
is nonsensical, and we should treat this result seriously. To support this hypothesis
I would indicate that, in contrast to Russell’s accusation, according to which Witt-
genstein in the Tractatus managed to say a good deal about ethics although he rec-
ommended silence about it408, one can trace a real effort to keep silent with respect
to ethics. In the Notebooks Wittgenstein thought about suicide as an elementary sin

408 Russell 1922, p. XXIII.

187
because if “suicide is allowed then everything is allowed” (NB 10.1.17, p. 91). In the
Tractatus he keeps silent about this topic. In the Notebooks there are vast entries
where Wittgenstein considers what it means to be happy409. The Tractatus contains
no such indications, apart from the statement that the world of a happy person is
different than the world of an unhappy person (but still no remarks as to what be-
ing happy consists in)410. It seems that he took seriously his own counsel to be silent
about things one cannot meaningfully express.
The gist of the above considerations reads as follows: the Notebooks take from
Schopenhauer a transcendental point of view. There is a willing subject and good
and evil enter into the world through the willing subject, and that perspective
(the point of view of the willing subject) allows to carry out concrete ethical con-
siderations. In Wittgenstein’s final formulations, in the official presentation of
his views, i.e. in the Tractatus, we observe in this respect a change. He withdraws
from using the controversial term of the willing subject and from contriving
a theory as to how good and evil enter the world, and he avoids talking about
specific ethical issues. Instead, in the foreground of the Tractatus is the thesis of
the nonsense of ethics and the conviction that an opinion according to which in
ethics we acquire access to a higher reality is deceptive:
It is impossible for there to be propositions of ethics. Propositions can express nothing
that is higher. It is clear that ethics cannot be put into words. Ethics is transcendental
(TLP 6.42-6.421).

The last proposition should not mislead us. It does not mean that Wittgenstein
ultimately kept thinking about ethics in transcendental terms. The significance
of the last sentence is not, as Anscombe claims, that ethical propositions “show
something that pervades everything sayable and is itself unsayable”411. In the
light of the shift between the Notebooks and the Tractatus, I think it is much
more probable to regard “transcendental” as tantamount to “attacking the limits
of sense”. In my view, according to Wittgenstein, ethics is an example of a hu-
man’s desire to cross the limits of sense. An attempt which fails. Hence, despite
the fact that in the Tractatus Wittgenstein describes ethics as transcendental, one
should reject the transcendental interpretation of the Tractatus as far as ethics
is concerned. This means that although its proponents rightly recognise in the
Tractatus the modernist suspicion that the progress of science leaves no room

409 “In order to live happily I must be in agreement with the world. And that is what
‘being happy’ means” (NB 8.7.16, p. 75).
410 TLP 6.43.
411 Anscombe 1965, p. 166.

188
for moral values, they are wrong in thinking that Wittgenstein’s response was to
adopt transcendental idealism. Naturally, proponents of the transcendental in-
terpretation could respond to my critique that, nevertheless, despite the changes
in the formulations about ethics, there is still much evidence in the text of the
Tractatus which confirms a transcendental reading of the book. I am at odds
with this kind of defence. In sections 4.3 and 4.4 we will come to see that a me-
ticulous analysis of the fragments given as typical examples of Tractarian tran-
scendentalism shows that we cannot rightfully ascribe to early Wittgenstein the
label of a transcendentalist.

4.2.2.2  The naturalisation of ethics

In this section I am going to argue that if there is any influence of Schopen-
hauer’s ethics on the Tractatus, it finds its expression not in the alleged tran-
scendentalism of this book but in the process I call the “naturalisation of ethics”.
Some of the ethical notions seem to suggest dualism of mind and matter, or the
existence of a supernatural reality. For example, the concept of free will seems to
invoke dualistic associations with the self which is a cause and in full control of
the action of the body. The concept of justice assumes, since we do not observe
moral justice in this world, the realness of another world; the beyond in which
human beings are punished or rewarded for their actions. Otto Neurath wrote
that “in its origin, ethics is the discipline which seeks to determine the totality
of divine injunctions”412. As a naturalisation of ethics I will mean a philosophi-
cal undertaking which aims to present such an understanding of ethical notions
which would not require any dualistic or supernatural connotations. I will argue
that this undertaking is present in Wittgenstein’s ethics and is a constant ele-
ment of Schopenhauer’s influence throughout the early phase of Wittgenstein’s
philosophical career. The naturalisation of ethics inherited by Wittgenstein
from Schopenhauer is displayed, among others, in:
1. The conviction of the illusionary character of free will which does not have any
influence on the world of empirical facts. Schopenhauer wrote that:
The dispute as to the freedom of the individual action (…) really turns onto the question
whether the will resides in time or not. If, as Kant’s teaching as well as the whole of my
system makes necessary, the will as thing-in-itself is outside time and outside every form
of the principle of sufficient reason, then not only must the individual act in the same
way in the same situation (…), but, as Kant says, if only the empirical character and the

412 Neurath 1959, p. 305.

189
 motives were completely given, a man’s future actions could be calculated like an eclipse
of the sun or moon (Schopenhauer 1844/1966a, p. 292),

and Wittgenstein, respectively:

The world is independent of my will (TLP 6.373). I cannot bend the happenings of the
world to my will: I am completely powerless (NB 11.6.16, p. 73).

The difference between these two philosophers is that for Schopenhauer the
claim of a unity between the will and an action led to a “discovery” of the core
of reality, one that is inaccessible to natural science. But it is clear that for young
Wittgenstein it meant only that an act of will (elusive to natural science) is not a
cause of actions: “The act of the will is not the cause of the action but is the action
itself. One cannot will without acting” (NB 4.11.16, p. 87)413. We do not change
facts through the exercise of will:
If the good or bad exercise of the will does alter the world, it can alter only the limits of
the world, not the facts – not what can be expressed by means of language. In short the
effect must be that it becomes altogether different world. It must, so to speak, wax and
wane as a whole (TLP 6.43)

The intention is clear: there are no two different causes of one event, i.e. mental
(synonymous with free will) and material (a precedent fact), but just one, i.e.
material. We should read TLP 6.43 as an anti-Cartesian remark against the thesis
that immaterial entities can directly influence events in the empirical world.
2. Reluctance towards the concept of the justice of retribution. The fact that we
are held accountable for our actions is a necessary condition for morality; we
take responsibility for what we do. The fact that we sometimes get away with
what we do wrong or that we sometimes are not appreciated for what we do
right could be seen as dangerous to our feeling of justice. In order to preserve
our respect of justice sometimes one postulates that, independently of the cur-
rent course of events, we are always held accountable. We will eventually be

413 Wittgenstein clearly connects the notion of free will with the notion of a cause of
events in the world. In this sense there is no free will (as independent of actions;
as their driving force). This claim also refers to actions which are often regarded
as purely mental, such as beliefs. As we saw in Chapter 3, Wittgenstein was of the
opinion that beliefs consist of the relations between events in the brain and states
of affairs. These events are not, therefore, in his view, purely mental. They are also
happenings in the world. In these cases one is also powerless, one also cannot say
that an action is preceded by an act of free will. Also, to these cases Wittgenstein’s
dictum is applied: the act of the will and the action are one.

190
 rewarded or punished for what we do – but after this life, in another, super-
natural world. This kind of combining the notions of reward and punishment
with the elementary feeling of justice was unacceptable for both Schopenhauer
and Wittgenstein. In their view, the concept of justice is not endangered by
cases in which the course of events does not correspond to our feeling of jus-
tice because both philosophers saw a reward and a punishment as inherently
connected to a deed. Schopenhauer claimed that “the concept of the retaliation
implies time, therefore eternal justice cannot be a retributive justice (…) Here
the punishment must be so linked with the offence that the two are one”414, and
this remark was later repeated in the Tractatus in thesis TLP 6.422:
Ethics has nothing to do with punishment and reward in the usual sense of the terms
(…) There must indeed be some kind of ethical reward and ethical punishment, but they
must reside in the action itself.

Just as the concepts of reward and punishment are demythologised and, so to

speak, turned back from the realm of eternity to the earthly reality, so the concept
of eternity itself is naturalised and deprived of a supernatural meaning. Eternity,
according to Schopenhauer, is not something that awaits us after our death but
rather resides in the present: “The will to live manifests itself in an endless present
(…) This is temporal immortality”415. One finds a similar remark with respect to
an eternal life in the Tractatus (TLP 6.4311): “If we take eternity to mean not in-
finite temporal duration but timelessness, then eternal life belongs to those who
live in the present”.
A reader of the Tractatus can sometimes have a feeling that although Wittgen-
stein recommends silence about ethics, he speaks much in this respect. What I
attempted to show is that this is not in contradiction with Tractarian intentions.
As one of the most important results of his book Wittgenstein held the discern-
ment between meaningful and senseless discourses. When somebody wants to
say something metaphysical, one should answer, claims Wittgenstein, by show-
ing the adversary that he did not confer any meaning on the signs he used. That
is how we should interpret the remarks about eternity, punishment, reward and
free will. If somebody uses these concepts in a traditional way (suggesting some
kind of dualism), he attaches no meaning to these ideas. A meaningful discourse
containing these concepts refers only to human deeds (instead of a reward and
a punishment in the spirit world), the present (instead of eternity) and to move-
ments of the body (instead of free will).

414 Schopenhauer 1844/1966a, p. 350–351.

415 Schopenhauer 1844/1966b, p. 479.

191
 Secondly, the naturalisation of ethics proves that commitment to the thesis of
the great influence of Schopenhauer on Wittgenstein’s thinking does not compel
one to accept the transcendental interpretation of the Tractatus. I can admit that
the Tractatus, with respect to ethics, is inspired by the views of Schopenhauer,
however, in my opinion this inspiration is restricted only to the naturalisation of
the ethical notions.

4.3  Solipsistic theses of the Tractatus

In the previous paragraph I analysed interpretations which emphasise Wittgen-
stein’s ethics as allegedly speaking in favour of the transcendental interpretation.
In the following section I shall refer to theses TLP 5.6-5.641, which are some-
times called (as I shall prove incorrectly) the solipsistic theses of the Tractatus. In
TLP 5.62 Wittgenstein surprises the reader of his book (which up to this moment
is a very consistent, logical and analytical text about facts, propositions and the
theory of meaning) with his sudden approval of solipsism – one of the most scan-
dalous philosophical positions:
This remark provides the key to the problem, how much truth there is in solipsism. For
what the solipsist means is quite correct; only it cannot be said, but makes itself manifest.
The world is my world: this is manifest in the fact that the limits of language (of that
language which alone I understand) mean the limits of my world (TLP 5.62).

Another astonishing claim states the identification of the world with life:
The world and life are one. I am my world. (The microcosm.) (TLP 5.621-5.63).

Moreover, although the reader of the Tractatus read earlier that there is no such
thing as a soul (TLP 5.5421), now he finds that there is a sense in which we can
talk about the self in philosophy and that the self is the limit of the world:
The subject does not belong to the world: rather, it is a limit of the world.(…) Thus
there really is a sense in which philosophy can talk about the self in a non-psychological
way. What brings the self into philosophy is the fact that ‘the world is my world’. The
philosophical self is not the human being, not the human body, or the human soul, with
which psychology deals, but rather the metaphysical subject, the limit of the world—not
a part of it (TLP 5.632; 5.641).

“What the solipsist means is quite correct” – says Wittgenstein (TLP 5.62). The
following review of interpretations covers a range of positions which try to fig-
ure out what, according to Wittgenstein, is rightly meant by solipsists. What is
common to these commentaries is the fact that their authors deduce the tran-
scendental idealism of the Tractatus not from a modernist disagreement with
the scientific worldview but they see this position as a consequence of, in one

192
case, Wittgenstein’s views on the simples (Frascolla) and in the second case of
Wittgenstein’s view on language (Hacker).
In the next two subsections I will present the views of two typical representa-
tives of the transcendental interpretation: Pasquale Frascolla (section 4.3.1) and
Peter Hacker (section 4.3.2). After a discussion of their views I shall present my
critical response. The result of the critique aims to show that these proposals
of reading the Tractatus have some intrinsic difficulties. They struggle to con-
vince the reader that transcendentalism truly follows from the assumptions of
Wittgenstein’s ontology or the theory of language. A possible response to that
critique could be to point at the respective fragments of Wittgenstein’s writings
and to say that even if there are no logical inferences between some of Wittgen-
stein’s views, he nevertheless saw the connection, and that is why he ultimately
admitted the correctness of solipsism. In response I shall analyse fragment
TLP 5.64-5.641 (section 4.3.3). I hope to show that what at first sight looks like
an approval of transcendental solipsism is in fact a critique of this standpoint.
Therefore the proponents of the transcendental interpretation of the Tractatus
have no support for their hypotheses.

4.3.1  The transcendental self as the owner of the phenomenal world

For Pasquale Frascolla, the so-called solipsistic theses of the Tractatus are confir-
mation of a phenomenalistic interpretation of this book (an interpretation which
claims that Tractarian simples are simple units of perception)416. According to
him, conceding that the simples are sense-data is the only way to make sense of
Wittgenstein’s claims about the subject who is at the same time the owner of both
language and the world: “The thesis that language is my language and that the
world is my world (…) must be a consequence of some of the pivotal principles of
the picture theory. And its source can in fact be easily found in the phenomenal-
istic ontology which, according to my conjecture, underlies language”417. Frascol-
la claims that Wittgenstein when writing the Tractatus had in mind some basic
principles. One of these is the phenomenalistic thesis that the world consists of

416 “The fact that Wittgenstein himself introduces the theme of solipsism as if it were an
obvious corollary of his theory proves that his ontology was phenomenalistic from
the beginning, and that the sections at the outset of the Tractatus are to be construed
as putting forward, even if to a certain extent implicitly, a phenomenalistic ontol-
ogy” (Frascolla 2007, p. 206). We also find a similar interpretation to Frascolla’s in:
Kannisto 1986.
417 Frascolla 2007, p. 205.

193
qualia. Qualia, on the one hand, make up the world and, on the other, constitute
my experiences, and, therefore, they constitute my life. Phenomenalism regards
life as identical to the totality of phenomena with which the subject is acquaint-
ed. The world is defined by phenomenalists in exactly the same way. Therefore,
somebody who is aware that the Tractatus silently presupposes the truthfulness
of phenomenalism should not be surprised by Wittgenstein’s identification of the
world with my world and the world with life.
Solipsism of the Tractatus should be taken then at face value: what is correct
in the thinking of a solipsist is the fact that we are trapped in the world of our
phenomenal experiences and the only objects whose existence one can be cer-
tain of are qualia. Wittgenstein admits that one cannot meaningfully express the
claims of solipsism but this is only because, as Frascolla claims, that he generally
bans expressions that try to grasp what the world is taken as a whole418.
The first pillar of Frascolla’s commentary is the conviction that the simples of
the Tractatus are units of perception. The second pillar of his interpretation is
the transcendental reasoning he detects in Wittgenstein’s work. The first step in
this reasoning consists in stating the existence of a particular point of view – an
attitude which takes the world as a whole. According to solipsism, the world as
a whole is a totality of experiences. In the next step Frascolla draws the moral
that there must be a bearer of this world: “The phenomenal world is inconceiv-
able without an owner”419. Wittgenstein, as Frascolla rightly notices, denies the
existence of the subject within the phenomenal world but, according to this
commentator, this “does not mean to deny that the phenomenal world, taken
as a whole, has an owner, nor does it mean to adopt the view that the phenom-
enal world, taken as a whole, is a subject-less world”420. He identifies the owner
of the world with the transcendental self421. Frascolla finds confirmation of his

418 “Since nothing in general can be meaningfully said about those limits, the solipsistic
principle, that the world is my world, cannot be formulated in a genuine proposition”
(ibid., p. 206).
419 Ibid., p. 208.
420 Ibid., p. 208. Also Heikki Kannisto, who is a proponent of the phenomenalistic in-
terpretation of the simples, draws the moral that the consequence of this position is
the acceptance of the transcendental self: “On the one hand, both subjects [Kant’s
and Schopenhauer’s] function as that to which the world is ultimately given. On the
other hand, they themselves lie outside the forms of representation, in the way even
the empirical subject cannot be said to do so” (Kannisto 1986, p. 151).
421 “It is the transcendental nature of the self which is owner of the world” (Frascolla
2007, p. 208).

194
conclusion in thesis TLP 5.633, where Wittgenstein compares the self and its
world to an eye and its visual field. Just like an eye is not a part of the visual field
but conditions the existence of it, in the same way the transcendental self is not
a part of the empirical world but conditions its existence and forms its limits.
Hence Wittgenstein was right to say both that the self is the limit of the world
and that the world is my world.
I will argue in the next section (4.3.3) against these conclusions from the
analogy of an eye and its visual field, but first I shall refer to the substance of
Frascolla’s interpretation. As I said before, it has two pillars. I am at odds with
both of them. First, Frascolla’s understanding of the solipsistic theses assumes
the identification of Tractarian simple objects with sense-data. From this iden-
tification it follows that the rightness of solipsism should be read at face value.
In section 2.1.3 I discussed the problems of the phenomenal interpretation.
There is no room here for repeating arguments against the candidacy of qualia
as the simples. Let me just say that the solipsistic theses are supposed to be the
main argument in favour of this candidacy. Frascolla seems to draw his cer-
tainty about the Tractarian simples from somewhere else. However, whereas
the phenomenalistic interpretation was doubtful even with the support of the
solipsistic theses, then without this support it is convincing to an even lesser
extent. The first pillar of Frascolla’s interpretation seems to collapse.
In the second important claim Frascolla argues that the phenomenal experi-
ence of the world as a whole has to have an owner. My problem with that argu-
ment is that I do not understand the conclusion that a subject of the phenomenal
world has to be a transcendental self existing outside the world. I assume that,
according to Frascolla, every experience has its subject. This seems cogent. But
from the claim that every particular experience has a subject it does not follow
that the totality of experiences has one and the same owner, nor that it transcends
the empirical world. It seems that someone who accepts this kind of reasoning
has already asserted the truth of transcendentalism. Transcendental reasoning
cannot function as an argument supporting the transcendental interpretation
of the Tractatus.
Concluding, Frascolla’s conviction that the Tractatus supports the existence
of the transcendental self is unjustified. It is based on two doubtful theses: for
instance, one cannot say that the phenomenalism of the Tractatus supports the
hypothesis that Wittgenstein was an adherent of solipsism since there are solip-
sistic theses which usually support the phenomenalistic interpretation. The only
way to save the transcendental reading is to say that the solipsistic theses of the
Tractatus speak for themselves. Their transcendental morals are so clear that they
do not need further support from other Tractarian theses. But again, I am at odds

195
with this claim. I shall examine TLP 5.6-5.641 in section 4.3.3 with the intention
of showing that these theses even when read at face value do not support the
transcendental reading. In other words, I shall show that even if the Tractatus
consisted only of these theses, it would still be a transcendental oeuvre.

4.3.2  The transcendental self as the linguistic soul

Perhaps problems with the phenomenalistic interpretation are the reason for which
some of the commentators who sympathise with the transcendental interpretation
(Peter Hacker, Hans-Johann Glock, Eric Stenius) do not go as far as Frascolla and
they do not take Wittgenstein’s stance to be phenomenalistic. Nevertheless, they
also think that Wittgenstein’s approval of solipsism means that he accepted in his
early work a form of transcendentalism. According to Frascolla, a language, which
is the topic of early Wittgenstein’s philosophy, is the language of sense-data. This
thesis is rejected by Hacker; however, he wants to maintain the feature of privacy
when referring to a language as described in the Tractatus. Instead of the language
of sense-data, Hacker prefers to speak about the egocentric language which, fol-
lowing Putnam’s accurate description, “is a language in which I can speak of things
as existing only when they are observed by me”422.
Hacker detects in the egocentric language of the Tractatus a remote ancestor
of the theory of the language of thought (LOT). Indeed, we saw in Chapter 3 that,
according to Wittgenstein, thoughts are facts, that the constituents of thoughts are
correlated with words, and that the relation of isomorphism holds not only be-
tween linguistic signs and reality but also between thought and reality. Simplify-
ing, thoughts can also represent states of affairs. This conclusion is in accordance
with LOT theory, which claims that mental states have a representational charac-
ter. There are other elements of the Tractarian theory of language, not mentioned
by Hacker, which make his hypothesis that we find LOT in the Tractatus more
probable. For example, Fodor argues that the representational theory of the mind
requires pure referentialism423, which is the claim that names have no senses just
reference. I have argued (in Chapter 1) that that is exactly the Tractarian position.
We could also argue in favour of the hypothesis that early Wittgenstein believed
in the compositionality thesis, i.e. the conviction that the content of a thought is
entirely determined by its structure and the contents of its constituents. Moreover,
in the previous chapter we saw that Wittgenstein naturalised intentional psychol-
ogy, which is impossible, according to Fodor, when we do not accept one of the

422 Putnam 2008, p. 10.

423 Fodor 2008, p. 16.

196
principles of LOT theory, i.e. the priority of thought over language. Referential-
ism, the compositionality thesis, naturalisation of propositional attitudes – all of
these features of the Tractarian system make Hacker’s view plausible. The first
pillar of Hacker’s transcendental interpretation is established: the Tractarian view
on language resembles what we nowadays call the language of thought.
In the second step of his reasoning, Hacker claims that for Wittgenstein the
obvious consequence of such a theory of language was acceptance of the tran-
scendental solipsism. Why such conclusions? As I mentioned before, according
to referentialism the meaning of a name is its referent. Hacker is interested in
the questions: “How does this correlation between signs and objects in the world
take place? What binds the elements of a sentence or thought with elements of
reality?”, and he claims that according to Wittgenstein it is through “some mental
act of meaning or intending a certain word to signify an object one has in mind.
It is an act of will which correlates word to signify an object one has in mind”424.
It is through the machinery of the mind that language gains its flesh and blood.
We reach out to the world through a mental act of an ostensive definition. For
Hacker it seems legitimate to ask: “Who is the agent of the correlation between
words and objects?, Who performs these mental ostensive definitions?”. Follow-
ing the hint of theses TLP 5.6ff, he thinks that it is the metaphysical self which
“injects meaning or significance into signs, whether in thought or in language425.
One might call this conception “The Doctrine of the Linguistic Soul”, for it is
the soul that is the fountainhead of language and representation426; the linguistic

424 Hacker 1972/1986, p. 73. A similar position is taken by Anthony Kenny: “The Tracta-
tus itself did not think, that impressions and ideas could themselves confer meaning
unaided. In the Tractatus meaning is conferred by pure will, the pure will of the extra-
mundane solipsistic metaphysical self ” (Kenny 1981, p. 147). The same conclusion is
expressed by Eric Stenius: “[For Wittgenstein] the form of experience is ‘subjective’
in the transcendental sense, the metaphysical subject being the ‘subject’ which uses
and understands language” (Stenius 1960/1996, pp. 220–221).
425 Hacker interprets the notion of the willing subject from the Notebooks also as a
counterpart of the notion of the transcendental self. Consequently, Hacker identifies
the willing subject with the linguistic soul. The transcendental self under his inter-
pretation confers meaning on linguistic expression and confers values on objects.
426 Hacker 1972/1986, p. 75. Hans-Johann Glock could agree with the doctrine of the
linguistic soul. He assumes that every representation is linguistic in the Tractatus,
and that entails a special kind of transcendentalism, namely linguistic solipsism.
Wittgenstein identified the world with “my world” and, moreover, the limits of my
world with the limits of my language, the only language one understands. Glock
understands these theses in connection with theses TLP 3.11-3.12, which state that

197
soul which confers meaning on linguistic expression by means of mental osten-
sive definitions. Hence, the user of a language uses his own private language.
Only he can understand the meanings of its expressions and this, for Hacker,
perfectly harmonises with thesis TLP 5.62:
The world is my world: this is manifest in the fact that the limits of language (of that
language which alone I understand) mean the limits of my world.

Solipsism in this respect is right: the private language entails that the world is my
world. We are trapped in our own languages and understandings. In contrast to
Frascolla’s interpretation, I do not neglect the first pillar of this transcendental in-
terpretation (that one can find in the Tractatus a remote ancestor of LOT), but
similarly to my doubts with respect to Frascolla I shall argue that there is no logical
connection between the belief in LOT and the belief in the existence of the lin-
guistic soul. The crucial question is if one truly needs to introduce the notion of
the linguistic soul in order to make the LOT theory consistent. If there is another
explanation for the correlation between signs and objects at our disposal, then one
is not bound to accept transcendental conclusions from the fact that Wittgenstein
probably approved LOT. If we find it, then we will falsify the reasoning that Hacker
adheres to.
I will just notice that one can find an explanation for the correlation between
words and objects which does not assume the existence of a non-empirical being
in the texts of the classic proponent of LOT, i.e. Jerry Fodor. Fodor argues in fa-
vour of a form of nativism – the form which tries to reconcile the typical claim of
nativism that we possess an innate inventory of concepts with the common sense
belief that mind-world interactions can change our repertoire of concepts427. The
fact that concepts are innate means that we have them because of our neurology.
Our neurology is the result of the phenotype, and the phenotype is the result of
the genotype. Hence, we have the concepts that we have because of our genotype.
All of the concepts we possess are built in our organisms, but we need something

signs become symbols only when we – the users of language – think and use them.
Linguistic symbols become meaningful only through my intentions or other mental
acts. There must be a subject who confers his intentions and meanings on symbols.
This is Glock’s explanation for the concept of the metaphysical self which we find in
the Tractatus: “The relation between the I and reality is replaced [in the Tractatus]
by the relation between the sentence and reality. And that relation depends on the
metaphysical subject, a linguistic soul which breathes life into mere signs” (Glock
1996, p. 350).
427 Fodor 2008, p. 131.

198
to trigger the concepts we already possess. For example, let us imagine a situation
when one sees a dog for the first time. This event triggers in a person the innate
concept of a DOG428. From this moment on, one uses the concept of a DOG to
refer to dogs. As we can see, Fodor does not need to employ the transcendental
concepts (such as the concept of the linguistic soul) in order to explain that by
using words we mean something. Instead, his theory speaks about genotypes and
ordinary interactions with the world.
It is not my task to decide whose explanation – Hacker’s or Fodor’s – is better.
In order to falsify Hacker’s reasoning it is enough to show that there are other
alternatives to the Doctrine of the Linguistic Soul. By indicating a different pos-
sible explanation for LOT one shows that Wittgenstein, being a proponent of
LOT, was not forced to accept transcendental solipsism. One can adhere to the
conception of LOT and one can negate that the linguistic soul has an impact on
the meaning of the concepts we possess.
Concluding, even if Wittgenstein believed in LOT, that fact alone could not
make him a transcendental solipsist. Hacker’s defence, therefore, should be ana-
logical to Frascolla’s: perhaps the belief in the transcendental self is the conclu-
sion from the theory of language as accepted by Wittgenstein. Nevertheless, he
made this mistake, and evidence of this one can find in the so-called solipsis-
tic theses of the Tractatus. Hacker indicates at remarkable similarities between
the formulations, pictures and metaphors that Wittgenstein used in TLP 5.6ff
and those one can find in the idealistic system of Schopenhauer: “Wittgenstein’s
metaphors are identical with Schopenhauer’s. There can be little doubt that the
last of the three extant notebooks was written while Wittgenstein was re-reading
Schopenhauer”429. These similarities should ultimately convince the reader of the
Tractatus that he has in his hands an idealistic oeuvre. Perhaps the arguments in
favour of transcendentalism are weak, says Hacker, but, nevertheless, Wittgen-
stein, being under the impression of Schopenhauer, seems to hold them as valid.
In the last two sections I investigated the interpretations which ascribe to Witt-
genstein the thought that transcendental solipsism is the consequence of his other

428 “We have the concepts we do because we have the neurology we do; we have the
neurology we do because we have the phenotype we do; and we have the phenotype
we do because we have the genotype we do (…) For concepts to be genotypically
specified is one thing; but for the genotypically specified information to be pheno-
typically expressed—for it actually to be accessible to a creature’s mental nativism
processes—is perhaps quite another. Maybe experience is what bridges the gap be-
tween a genetic endowment and its phenotypic expression” (ibid., p. 146–147).
429 Hacker 1972/1986, p. 88.

199
Tractarian theses (such as the alleged identification of the simples with sense-data,
or the acceptance of LOT theory). I am not interested if Frascolla or Hacker think
that Wittgenstein was right in his conclusions, but just in their conviction that he
did indeed infer his solipsism from his other standpoints. I argued in the last sec-
tions (section 4.3.1) against the claim that Wittgenstein could conclude that the
view of a solipsist is correct from his ontological phenomenalism. I also rejected
the hypothesis that he asserted the rightness of solipsism as a consequence of the
LOT theory (section 4.3.2). In other words, if Wittgenstein was a solipsist, this has
to follow solely from the solipsistic theses (TLP 5.6-5.641). In the next section I
shall investigate the validity of this claim by analysing these theses.

4.3.3  Arguments in favour of Tractarian transcendental solipsism

The conclusion of the last two sections is that, for instance, LOT theory alone would
not suffice to state Wittgenstein’s transcendentalism. But a proponent of the solip-
sistic interpretation could answer that there are direct proofs in the so-called solip-
sistic theses of the Tractatus that its author was truly a proponent of transcendental
solipsism. What is crucial for this argument is the claim that Wittgenstein inher-
ited ontological views from Schopenhauer: “In claiming the relativity of the world
or, better, of logical space, for a subject, and in providing his semantic version of
solipsism, Wittgenstein develops themes which one can find in Schopenhauer”430.
I shall argue in this section that, naturally, the so-called solipsistic theses “are bet-
ter understood in the light of Schopenhauer than any other philosopher”431, but,
on the other hand, this does not mean that the Tractatus adopted transcendental
solipsism. On the contrary, a comparison with Schopenhaeur’s system will reveal
that in fact Wittgenstein rejected solipsism.
The structure of the section is the following: first, I shall point at two of the
most important pieces of evidence from the text that allegedly confirm the tran-
scendentalism of the Tractatus. These are the identification of the world with my
world (section 4.3.3.1) and the metaphor of the eye and the visual field (4.3.3.2).
Next (section 4.3.3.3), I shall indicate two theses from fragment TLP 5.6-5.641

430 Frascolla 2007, p. 207. Another example of this claim: “These apparently Schopen-
hauerian influences suggest that Wittgenstein’s remarks about the illusoriness of the
thinking self are, like Schopenhauer’s unoriginal criticism of Descartes, directed
against the conception of a Cartesian res cogitans (…). Schopenhauer’s distinction
between the illusory Cartesian self and the transcendental self was taken over by
Wittgenstein” (Rosenberg 1968, p. 170).
431 Anscombe 1965, p. 12.

200
that suggest a rather polemical overtone of the whole passage to solipsism. The
point is to prove that there are neither substantive (sections 4.3.1 & 4.3.2) nor
derived from the text of the Tractatus arguments (current section) in favour of
the transcendental interpretation.

4.3.3.1  “The world is my world. I am my world” (TLP 5.62-5.63).

“The identification of the individual consciousness with the microcosmos, and the
microcosmos with the macrocosmos, is a central Schopenhauerian thesis”, claims
Hacker when commenting on the above quotation from the Tractatus432. Frascolla
claims, on the other hand, that the above assertions “unquestionably” confirm his
reading of early Wittgenstein433. First, I would answer these claims that, to be exact,
the Schopenhauerian thesis identifies the world with my representation and not
with my world434. It is clear that Hacker and Frascolla identify the term: “my world”
with what is experienced by the subject, but in this way they assume what should
be proved. I think we have important reasons to disagree with such an identifica-
tion. In thesubsequent passages I shall argue in favour of a different understand-
ing of the expression “my world”. What is important is that the thesis: “I am my
world” (TLP 5.63) is preceded in the Tractatus by the claim: “The world and life
are one” (TLP 5.621). In the Notebooks we can find an entry in which Wittgenstein
explains that he does not have to take the notion of life to mean a psychological
life435, and having that in mind for early Wittgenstein psychology meant what we
would rather call today phenomenology; thus it becomes clear that TLP 5.62 does
not identify the world with my experiences, as Hacker and Frascolla suggest. It is
not an ontological thesis stating that the world consists of my representations nor
an epistemological claim that we are acquainted only with sense-data. If we take
into account that Wittgenstein explained the concept of “life” as the “consciousness
of the uniqueness of my life”436, we should revise, all the more, the first associations
one could have after reading TLP 5.62. Being aware that one’s life is unique sug-
gests that Wittgenstein, when he talks about the world as my world, is more inter-
ested in rather existential questions than in epistemology.
The next argument supports my point. We read in TLP 5.632 that the meta-
physical subject is the limit of the world, but first we get to know the consequences

432 Hacker 1972/1986, p. 93.

433 Frascolla 2007, p. 206.
434 Schopenhauer 1844/1966a, p. 1.
435 “Psychological life is of course not ‘Life’ ” (NB 24.7.16, p. 77).
436 NB 1.8.16, p. 79; NB 2.8.16, p. 79.

201
of this claim in TLP 6.43. And these are not (as they should be if the claims of
TLP 5.6ff reveal Wittgenstein’s ontology or epistemology) a sceptic’s theses that
one can never be certain of the existence of other subjects (just the opposite:
according to the Tractatus (TLP 6.51), scepticism is obviously nonsensical), but
that “the world of the happy man is a different one from that of the unhappy
man” (TLP 6.43). This trivial thought is nonetheless astonishing to the reader of
the Tractatus, hence the world is supposed to consist of facts. They are the same
for all. From a certain perspective one could say that both happy and unhappy
people live in the world, where X won the lottery, travelled around the world and
helped the poor, and in a world where Y was diagnosed with cancer, his house
burnt down in a fire and so on – it is the same world. What is different is the fact
that X won the lottery has dramatically different importance for X than it does
for Y. It has a different meaning not because it refers to a different state of affairs,
nor because X has different access to this fact than Y (to win the lottery is a pub-
lic event), but simply because it has an existentially different meaning: the same
event could help X live a life of ease, whereas it does not change Y’s situation at
all. In this way the limits of the world of X have changed. As I have tried to prove,
when Wittgenstein talks about the limits of the world or of my world he does not
mean to advocate in favour of an epistemological or an ontological position. He
rather uses these concepts in the existential meaning.
I could also support my view by referring to the philosophy of Schopenhauer.
There are fragments in The World as Will and Representation in which Scho-
penhauer identifies the world with my world. For example, he claims that the
microcosmos is the macrocosmos. What I want to stress is the fact that he does
this in the context of remarks on the egoism of individuals who want everything
for themselves. Characteristically of Schopenhauer’s pessimism, individuals are
ready to sacrifice everything else in order to maintain themselves. They treat
themselves as the centres of the world, and only in this sense does Schopenhauer
identify the world with my world437. Every individual can take the place of the
centre of the world and, hence, there is nothing solipsistic entailed in the above
considerations. Schopenhauer uses the notions of the world and my world to
point out the quite obvious observation that, relatively to the point of view, we
estimate ourselves as extremely important (if we take into consideration a more
subjective attitude to reality when we think about ourselves in relation to our

437 “Every knowing individual is therefore in truth, and finds himself as, the whole will-
to-live, or as the in-itself of the world itself, and also as the complementary condition
of the world as representation, consequently as a microcosm to be valued equally
with the macrocosm” (Schopenhauer 1844/1966a, p. 332).

202
interests, pleasures, fears, goals, family and friends) or as quite ordinary (if we
look at ourselves as one of the countless representatives of one of many species
on one of countless planets in the universe)438. Concluding, in Schopenhauer’s
philosophy the claim: “The world is my world” betrays rather his pessimism
than solipsism, and even if Wittgenstein borrowed this claim from him it should
rather imply an existential reading of the Tractatus than a transcendental one.
There is another argument saying that one should not read TLP 5.62 as a par-
aphrase of Schopenhauer’s dictum that the world is my representation. It follows
from the analysis of the motivations that stood behind the Schopenhauerian
thesis. Wittgenstein rejected these motivations, so he had no reason to accept
the conclusion. For Schopenhauer to talk about the world as identified with
“my representations” meant that the empirical world is not a thing-in-itself: “All
knowing is essentially a making of representations; but my making of represen-
tations, just because it is mine, can never be identical with the being-in-itself
of the thing outside me”439. It is, in his opinion, necessary to discern between
appearances and things-in-themselves, because otherwise we are doomed to
the purely physical, chemical and mechanical explanations he despised440. The
misery of materialism is shown, according to Schopenhauer, in the absurd up-
shot of the investigations of the natural sciences. Among the ridiculous theories
of science contemporary to him he mentioned physical atomism441, corpuscular

438 “By looking inwards, every individual recognizes in his inner being, which is his
will, the thing-in-itself, and hence that which alone is everywhere real. Accordingly,
he conceives himself as the kernel and centre of the world, and considers himself
infinitely important. On the other hand, if he looks outwards, he is then in the
province of the representation, of the mere phenomenon, where he sees himself
as an individual among an infinite number of other individuals, and consequently
as something extremely insignificant, in fact quite infinitesimal” (Schopenhauer
1844/1966b, p. 600).
439 Ibid., p. 194.
440 “Let us not be deceived by dogmatic utterances and brazen assurances that these
matters are decided, settled, and generally admitted. On the contrary, the entire
mechanical and atomistic view of nature is approaching bankruptcy, and its advo-
cates have to learn that something more is concealed behind nature than thrust and
counter-thrust” (ibid., p. 311).
441 “In truth the atoms are a fixed idea of French savants, who therefore talked about
them just as if they had seen them” (ibid., p. 302). “I am no more obliged to think of
the mass of a body as consisting of atoms and of the spaces between them, in other
words, of absolute density and absolute vacuum, but I comprehend without difficulty

203
theory of light442, and the theory of biogenesis (which is directed against the
theory of spontaneous generation)443. Hence there are two motivations for the
Schopenhauerian identification of the world with my representation:
–– it allows to differentiate appearances from things-in-themselves
–– it protects us from “the continuance of the absurdities of the atomistic und
purely mechanical physics”444 because one can always see scientific explana-
tions as restricted only to the sphere of appearances. It opens the possibility
for metaphysical explanations of the world.
Wittgenstein rejected the above theses, and, therefore, there is no reason to as-
cribe to him the conclusion that the world is my representation; quite the op-
posite, the Tractatus begins with a polemics with this claim. On the first page of
the Tractatus we read that the world is the totality of facts (TLP 1.1)445, and on
the last page the claim that physical theories based on atomism and mechanics
(regarded by Schopenhauer as absurd) are the only meaningful propositions one
can express (TLP 6.53).
Concluding, the thesis that the world is my world cannot be seen as con-
firmation of Wittgenstein’s transcendental solipsism. It would be even strange
and inconsistent for him to disguise in this claim principles of Schopenhauer’s
philosophy, because in Schopenhauer’s system this thesis was thought to be pro-
tection against the claims of mechanics. If there is, however, a grain of truth in
the thesis of Schopenhauer’s influence on the analysed theses of the Tractatus,
then we should see TLP 5.62 as an existential claim rather than an epistemologi-
cal (for example: “I am acquainted only with sense-data”) or an ontological (for
example: “Basic bricks of reality are phenomenal units of my experience”) one.
The next paragraph of the Tractatus that is invoked by proponents of a tran-
scendental reading is thesis TLP 5.633, in which Wittgenstein compared the

those two phenomena as constant continua, one of which uniformly fills time, and
the other space” (ibid., p. 303).
442 “The constructions of light from molecules and atoms which have come from the
French are a revolting absurdity” (ibid., p. 302).
443 “Despite the most recent objections to it, I regard generatio aequivoca as extremely
probable at very low stages, and above all in the case of entozoa and epizoa [parasites]”
(ibid., p. 310).
444 Ibid., p. 303.
445 “The first proposition of the Tractatus is an echo of Schopenhauer’s, but at the same
time a seeming denial of it. The world is not my idea but all that is the case” (McGuin-
ness 2002, p. 133–134).

204
relation of the metaphysical self to its world to the relation of an eye to its visual
field. In the next passage I shall examine what the conclusion of this analogy
should be. I am not going to present my own views in the next passage because
I believe that David Pears sufficiently proved that TLP 5.633 is in fact a rejection
of solipsism. I shall focus, then, on presenting his argument.

4.3.3.2  The eye and the visual field (TLP 5.633)

Wittgenstein asks an imaginary interlocutor: “Where in the world is a metaphys-
ical subject to be found?” (TLP 5.633). There is a dispute over the meaning of the
answer. Peter Hacker, a proponent of the transcendental reading of the Tractatus,
sees in this thesis confirmation of his interpretation. In the analogy of the eye
and the visual field, he notices the transcendental thesis of a subordination of the
world to the subject. The fact that one does not see the eye in the visual field con-
firms, according to Hacker’s understanding of the text, that the transcendental
self is not a part of the empirical world. It also goes well with Hacker’s suspicion
that early Wittgenstein professed transcendental solipsism. It is because the eye
is the condition of the existence of the visual field just as, according to the solip-
sists, the subject is the condition of the existence of the empirical world446.
The other group of commentators (Pears, Stern, Puhl) has quite an opposite
view on the discussed thesis which, according to them, is in fact a polemics with
solipsism. Under their proposal, the analogy is an example of the nonsense of
consequent solipsism. In Pears’ eyes, the proponents of the transcendental inter-
pretation forget what morals Wittgenstein himself drew from the comparison of
the metaphysical subject with the eye and its visual field. The conclusion reads
as follows: “You do not see the eye. And nothing in the visual field allows you to
infer that it is seen by an eye” (TLP 5.633). In other words, the right conclusion
for a commentator of the Tractatus should be that, according to Wittgenstein,
a solipsist, if he thinks out his position strictly, has no argument to back up his
belief in the existence of the transcendental self.
David Pears explained why TLP 5.633 is the thesis saying that consequent sol-
ipsism is a nonsensical venture. Under Pears’ interpretation, Wittgenstein wanted
to say that “solipsism loses its intended meaning on the way to achieving the truth

446 “The metaphysical subject is the bearer of good and evil. Why is it not part of the world?
Wittgenstein merely hinted at an argument by way of analogy. The metaphysical subject
is related to the world as the eye is related to the visual field. Nothing in the visual field
entitles one to infer that it is seen by an eye. The eye of the visual field (…) is the source
of the visual field, not a constituent of it” (Hacker 1972/1986, p. 86–87).

205
that it aspires to achieve”447. Solipsism claims that the world is the subject’s world
because the limits of the world “are fixed from the inside by their relation to the
subject”448. According to Pears, Wittgenstein fights with this thesis. He betrays the
polemical frame of the mind exactly in the question directed at the imaginary sol-
ipsist: “Where in the world is a metaphysical subject to be found?” (TLP 5.633).
Wittgenstein’s adversary has two answers at his disposal. He could say that the sub-
ject is in the world, but that answer would for a solipsist be obviously self-refuting.
He is then forced to say that the self is outside the world. The best he can do, in
Wittgenstein’s opinion, is to use the metaphor of the eye and the visual field and
claim that there is an analogy between this picture and the relation between the
transcendental self and its world. But, according to Pears, this is exactly where
the solipsist falls into a trap, because by removing the subject from the world he
cannot verify his belief in the existence of the transcendental self. His claim that
the subject is outside the world is empty. The solipsist could naturally answer that
this accusation does not apply in the case of the subject of experiences, because
the identity of the subject of experiences is transparent for the solipsist. The sub-
ject is directly experienced, and hence the questions: “Which subject?” or “Whose
experiences?” do not even arise. Pears thinks that “it is against this defence that
Wittgenstein uses the analogy between ego and eye”449. If one sticks to the analogy
consequently, then one notices that nobody experiences his own seeing directly
and, therefore, if the transcendental self is the analogue of the eye, then it is not
accessible in introspection either. Moreover, Wittgenstein learns from the analogy
of the eye and the visual field that “the self of solipsism shrinks to a point without
extension, and there remains the reality co-ordinated” (TLP 5.64). This makes the
solipsist’s attempt to point at himself as the way of identifying the transcendental
self an empty gesture; as empty as the theory of solipsism. If Pears’ intuitions are
true, it would turn out that what at first sight looked like an iteration of Schopen-
hauer’s transcendental idealism is in fact a critical response to it.

4.3.3.3  “There is no a priori order of things” (TLP 5.634).

I sympathise with Pears’ interpretation because his conclusion that the analysed
fragment of the Tractatus is polemical with the idealistic tradition has strong
confirmations in two theses that follow the analogy of the eye and the visual field.
First, in thesis TLP 5.634, Wittgenstein admits that:

447 Pears 1993, p. 59.

448 Pears 1987, p. 164.
449 Pears 1993, p. 60.

206
 This is connected with the fact that no part of our experience is at the same time a priori.
Whatever we see could be other than it is. Whatever we can describe at all could be other
than it is. There is no a priori order of things.

Why does the fact that whatever we see could be other than it is weighs in favour
of the anti-transcendental reading of the analogy of the eye and the visual field? In
my opinion the most obvious answer refers to the fact that, according to transcen-
dental idealism, a part of our experience is a priori just because the transcendental
self constitutes experience. For example, we know a priori that every effect has
its cause because this is the way the subject cognises the world450. Everything we
experience is placed in time because time is the form of the inner sense451. From
somebody who believes in the transcendental ideality of time (and Hacker as-
cribes this view to Wittgenstein452) we would expect rather an acknowledgement
that there are some aspects of experience which cannot be other than they are.
And yet, Wittgenstein says quite the opposite. The reason for that is, in my view,
the fact that in the previous thesis he rejected the existence of the transcendental
self. If nothing imposes in advance the shape and form of experience, and if noth-
ing determines reality, then, in fact, everything could be different than it is, and we
have no guarantee that we capture the necessary features of experience. Talking
about an a priori order of things is unjustified.
In the next thesis (TLP 5.64), Wittgenstein claims that:
Here it can be seen that solipsism, when its implications are followed out strictly, coin-
cides with pure realism (TLP 5.64).

For Peter Hacker this thesis means that for Wittgenstein there would be “no
practical disagreement [between a solipsist and a realist] (…), nor will they

450 “The only genuine and convincing proof that we are conscious of the law of causality
prior to all experience is actually found in the very necessity of making a transition
from the sensation of the senses, given only empirically, to its cause, in order that per-
ception of the external world may come about” (Schopenhauer 1844/1966b, p. 37).
451 “We ask what time is; we investigate time as though it were something quite objective.
And what is this objective thing? (…). It exists only in the heads of beings that know,
but the uniformity of its course and its independence of the will give it the right and
title to objectivity. Time is primarily the form of the inner sense” (ibid., p. 35).
452 “Wittgenstein’s solipsism was inspired by Schopenhauer’s doctrines of transcendental
idealism (…) It involves a belief in the transcendental ideality of time (and presum-
ably space), a rather perverse interpretation of the Kantian doctrine of the unity of
apperception together with the acceptance of Schopenhauer’s quasi reification of the
unity of consciousness, and other related and obscure theories about ethics, the will,
aesthetics, and religion” (Hacker 1972/1986, p. 99–100).

207
quarrel over truth-values of propositions of ordinary language”453. The tran-
scendental truths such as: “Every experience is necessarily my experience” and
“The world is my world”, which differentiate solipsism from realism are non-
sensical for both of these standpoints. A solipsist and a realist agree that they
have no truth-value at all. There is nothing a realist truly and meaningfully
claims that could not also be affirmed by a solipsist.
Cora Diamonds reads TLP 5.64 as also equating solipsism with realism. Both
of these standpoints fail to say anything; both are equally nonsensical. Why such a
conclusion? According to Diamond, in the background of the solipsistic theses of
the Tractatus is the discussion with Russell’s theory of description. As I mentioned
in the previous chapters, for Russell, in order to understand a judgement one had
to be acquainted with all of its elements. I cannot be acquainted with, for example,
Bismarck’s experiences but, nevertheless, I can talk about (reach them indirectly)
them by means of a description454. The theory of description, however, depends on
Russell’s conception of quantifiers. For example, one reaches indirectly to others’
private experiences when x goes proxy for private objects in the formulas: “(Ǝx)
Fx” or “(x) Fx”. But, according to Diamond, “sentences with quantifiers are not
seen by Russell as they are by Wittgenstein, as constructions from sentences which
do not contain quantifiers”455. According to Wittgenstein, a proposition containing
a quantifier is a complex proposition – a conjunction of elementary propositions.
The consequence of Wittgenstein’s view on quantifiers is that one can understand
a sentence containing quantifiers only if one understands all the elementary sen-
tences that constitute it. If one cannot refer to the private experiences of Bismarck
in an elementary proposition, then one cannot refer to them by means of a descrip-
tion either. Under Russell’s proposal, speaking metaphorically, quantifiers offer a
bridge from the world of private objects to the world of public objects. But if Witt-
genstein is right in thinking that one cannot express what solipsism means (that
we cannot cross the boundaries of private experience), as claims Diamond, then
the whole idea of a bridge leading to objects “outside” our experience becomes
meaningless. If the concept of a private experience, on which solipsism is based, is
meaningless and plays no role in language, so the concept of objects is independent
of our experience456.

453 Ibid., p. 104.

454 “You or I or anyone else can think about Bismarck only via some description; we are not
directly acquainted with the object which he denotes by ‘I’.” (Diamond 2000b, p. 265).
455 Ibid., p. 270.
456 The solipsist does not rigorously follow his solipsism; if he did, it would lead him to a
non-Russellian realism. A one-limit view [limit of our experiences] self-destructs; we

208
 Diamond’s interpretation is based on a dubious premise. She claims that a re-
alistic standpoint is inconceivable without the conceivability of solipsism; that a
realist can explain his position only in opposition to the theory assuming the ex-
istence of private experiences. In Diamond’s eyes, Wittgenstein regarded realism
as a mirror reflection of solipsism, and he expressed this conviction in TLP 5.64.
This, in my opinion, is dubious, especially if we take into account an extended
version of TLP 5.64, i.e. an entry from the Notebooks. It turns out that Wittgen-
stein’s thesis is in fact an autobiographical remark about his own philosophical
development:
This is the way I have travelled: Idealism singles men out from the world as unique, sol-
ipsism singles me alone out, and at last I see that I too belong with the rest of the world,
and so on the one side nothing is left over, and on the other side, as unique, the world.
In this way solipsism leads to realism if it is strictly thought out (NB 15.10.16, p. 85).

It is a description of the intellectual journey which ends in accepting realism – the

claim that on the side where, according to solipsism, is the transcendental self,
there is nothing, and that I (a person who expresses this standpoint) belong to
the world. It is rather that than the observation that both positions – solipsism as
well as realism – are nonsensical. Wittgenstein started his philosophical career as
a proponent of idealism (probably inherited from Schopenhauer) and then, as he
reports, he had a solipsistic period of thinking. But solipsism places the self outside
the empirical world. It cannot identify its transcendental subject nor provide an
answer to the question of the relation between the transcendental subject and the
subject of everyday experience. The transcendental subject shrinks to a point. We
are left only with the empirical world. If that is the conclusion placed just after the
analogy of the eye and the visual field, then Pears is right in saying that TLP 5.64
implies that Wittgenstein was a realist457.
Concluding, I have tried to argue that theses proclaiming that there is no a
priori order of things and that solipsism strictly thought out leads to realism
prove that the metaphor of the eye and the visual field is in fact a polemics with
transcendental solipsism and the philosophy of Schopenhauer. Wittgenstein,
against the proponents of the transcendental interpretation, rejects solipsism
in TLP 5.6-5.641. In sections 4.3.3.1 and 4.3.3.2 I was arguing in favour of the
interpretation that the real significance of Wittgenstein’s claim: “the world is
my world” is existential. Fragment TLP 5.6-5.641 is often labelled the solipsistic

are not left, at the end of the Tractatus, with a philosophical view about a “far side” of
the “limit,” but merely with there being the sentences of our language (ibid., p. 282).
457 Pears 1987, p. 188.

209
part of the Tractatus. The main conclusion of section 4.3.3.3 is that this label is
obviously wrong.
In section 4.3 I analysed those interpretations which, on the basis of Wittgen-
stein’s remark that what solipsism means is quite correct, claim that he approved
solipsism and hence that the transcendental reading of the Tractatus is the most
appropriate. In sections 4.3.1 and 4.3.2 I analysed specific forms of this interpre-
tation and I pointed at their deficiencies. At best, they make early Wittgenstein’s
philosophy inconsistent. Even if Wittgenstein truly accepted a kind of egocentric
language theory, it does not follow that the transcendental self exists. One can
always answer that even if the accusation is correct, a belief in the existence of the
transcendental self is simply there, in the text of the Tractatus. Accordingly, I tried
to reject this response in section 4.3.3 by showing that there are no textual pieces
of evidence supporting transcendental solipsism. Concluding, there are neither
substantial nor textual arguments in favour of the transcendental interpretations.
Of course, one cannot forget that Wittgenstein claims that what solipsism means
is quite correct (TLP 5.62), i.e. that life and the world are one (TLP 5.621), that the
subject is a limit of the world (TLP 5.632), and that there is a sense in which a phi-
losopher can talk about the self (TLP 5.641). It is a job left to the opponent of the
transcendental interpretation to explain positively how to understand these theses
without invoking transcendental associations. I shall take on this task in Chapter
5. Before I do so I shall address the third kind of transcendental interpretation, i.e.
the one emphasising Wittgenstein’s remarks on life, death and the end of the world.

4.4  Tractarian understanding of death

Ireneusz Ziemiński straightforwardly approves the transcendental interpretation
of the Tractatus. The originality of his interpretation consists in taking at face value
Tractarian theses 6.431 that “at death the world does not alter, but comes to an end”
and 6.4311 that “death is not an event in life”. According to Ziemiński: “Such an
understanding of death indicates a discrepancy between objectivity (the world of
facts) and subjectivity (the self aware I)”458.
Strange as they may sound, these theses are quite comprehensible in the light
of Wittgenstein’s other remarks, and they do not have to result in the conclusion
that the existence of the world is conditioned by the existence of the self. We could
say, for example, that the death of an individual person is the end of the world for
that person and the end for the world of that person, and there would be nothing

458 Ziemiński 2007, p. 53.

210
odd in saying this. As I mentioned before, I understand the thesis: “I am my world”
(TLP 5.63) not ontologically but existentially. There is no reason not to think the
same way about TLP 6.431-6.4311.
Ziemiński reads Wittgenstein much more strongly. It seems that he reads
TLP 5.621 as an ontological identification of my world with the world. Saying
that my world – the world possessed by the self – is the only existing world
proves, according to him, that the existence of the self is a “prerequisite for the
world”459, and only by ascribing this conviction to Wittgenstein can we explain
why he wrote that together with the death of the self also comes the end for the
world. If the self is a condition for the existence of the world, then the disap-
pearance of the self means annihilation of the world: “Death as an absolute an-
nihilation of the subject and world has a profoundly metaphysical dimension
and constitutes final triumph of nothingness over existence”460. According to
Ziemiński, death is not an event in the world because it belongs to the history
of the self and not the history of the world. Because the self is not a part of the
world, then death is not a part of the world either.
I do not agree with these strong conclusions. As I noted before, we should in-
terpret TLP 6.431-6.4311, although they appear mysterious and “profoundly met-
aphysical”, as in fact quite trivial existential remarks about the ends of our lives.
Who is right? The difference consists in understanding the identity: “my world =
the world”. Ziemiński claims that the above equation should be read ontologically.
There is only one world and it is my world, the world of which I – the self – is the
owner. I have already provided arguments supporting the claim that TLP 5.621 is
an existential claim. I think that my interpretation is confirmed by the fragment of
the Tractatus, in which Wittgenstein, in preceding the TLP 6.431-6.4311 remarks,
compares the worlds of a happy and an unhappy person. He clearly does not
mean that since there are two worlds of individuals, then there are two objective
worlds, and each of them will come to an end with the death of their owners but,
rather, he speaks about two different attitudes towards the same world. Most of
all, in thesis TLP 6.4311, directly in the next sentence after: “Death is not an event
in life”, he explains that “we do not live to experience death”, which I think is an
unfortunate translation of the German original text: “Den Tod erlebt man nicht”,
which sounds like consolation to somebody who is afraid of death, and not as a
remark about the end of the world. The biographical context also speaks in favour
of the existential reading of TLP 6.431-6.4311. One should remember that the

459 Ibid., p. 53.

460 Ibid., p. 63.

211
theses in question were written when Wittgenstein fought at the fronts of World
War I. It is much more probable that at that time he was seeking reassurance in
the face of possible death rather than was deep in abstract thought regarding the
ontological dependence of the world on him461.

4.5  General remarks about the transcendental interpretation

In the last three sections I presented three kinds of a transcendental interpreta-
tion one can come across in Wittgensteinian literature. They refer to Tractarian
ethics (4.2), the solipsistic theses (4.3) and remarks on death (4.4). Of course, the
division between these three kinds is not sharp, and it is easy to find a transcen-
dental interpretation which refers to the ethical as well as to the solipsistic theses
of the Tractatus. In the following section I want to sum up the results of the previ-
ous discussions.
I would like to bring our attention to two points. In the last three sections I
argued that Wittgenstein rejected transcendentalism. But this could be answered
by saying that although it is true that Wittgenstein held, for example, solipsism
as nonsensical, he thought at the same time that this nonsense gives new insight
into reality. Literally false transcendentalism conveys at the same time an inex-
pressible truth. A discussion regarding this claim shall be the first topic of the
summary of chapter 4.5.1. Secondly (4.5.2), I want to go back to the claim with
which I started my analysis, that the main motivation for accepting the transcen-
dental interpretation is the fact that it reflects the modernistic anxieties of early
Wittgenstein. After examining the specific forms of transcendental interpreta-
tions, we are now in the position to claim not only that Wittgenstein did not use
the metaphysics of the self to safeguard the values of the human world, but that
he also could not use it. Hence, if it is true that one of his main problems dur-
ing the period of writing the Tractatus was to find a place for these values in the
scientific worldview, then one has to indicate Wittgenstein’s different solution.

461 Waugh cites two of Wittgenstein’s letters to his family coming from the time of the
Russian offensive in June of 1916: “Ich habe fruchtbare Szenen erlebt (…) ich fühle
mich sehr schwach und sehe keine äussere Hoffnung. Wenn es mir jetzt zu Ende geht,
so möge ich einen guten Tod sterben eingedenk meiner selbst. Möchte ich mich nie
selbst verlieren”, and two days later: “Wir sind in unmittelbarer Nähe des Feindes (…)
Jetzt wäre mir die Gelegenheit gegeben, ein anständiger Mensch zu sein, denn ich
stehe vor dem Tod Aug in Aug. Möge der Geist mich erleuchten” (cited in: Waugh
2010, p. 136). Exactly from this period of time come his thoughts about the meaning
of life and about death.

212
4.5.1  Is the nonsense of solipsism illuminating?
What does the phrase: “illuminating nonsense” mean? I will explain this expres-
sion on the example of a typical representative of the transcendental interpre-
tation – Peter Hacker. His thought proceeds along the following lines: first, we
should take the concept of the metaphysical subject seriously, for Wittgenstein
took moral values seriously, and because moral values were not illusory for
him, so their bearer was not either462. Hacker makes all of these claims by be-
ing fully aware of the fact that, according to Wittgenstein, and in contradiction
to Schopenhauer, transcendental solipsism cannot be meaningfully expressed.
But the literal absurdity of solipsism does not mean that it nonetheless does not
convey knowledge about the world. A group of scholars admit that, according
to the Tractarian criteria of a meaningful proposition, transcendental theses are
nonsensical, but they respond that they give us insight into the real nature of
the world; they help us to see the world aright. For example, Rush Rhees admits
that in the Tractatus there are no meaningful ethical propositions but, never-
theless, according to him Wittgenstein “still thinks that speaking of good and
evil means something”463. A.W. Moore is convinced that Wittgenstein’s tran-
scendental idealism consists in the belief that “our rational engagement with
things – that part of our engagement with things which is made possible by the
fact that we are rational, thinking beliefs – is not exhausted by whatever finds
expression in propositions with sense”464; that rational engagement with things
“is not just a matter of discursive knowledge”465 (as non-discursive engagement
with things he mentions evaluation, practice of philosophy, and understanding
of propositions). In another example, according to Birk, although solipsism
is inexpressible it is nevertheless the right attitude of a subject to the world466.
In my opinion, the same claim of the illuminating nonsense is expressed by
Atkinson when he writes that in order “to make sense of what cannot be said,
one must see the world from a viewpoint that is outside the world”467. The case
with Atkinson is slightly complicated because he regards his own position in
opposition to the transcendental interpretation. But his declaration seems to

462 Hacker 1972/1986, p. 67.

463 Rhees 1965, p. 17.
464 Moore 2007, p. 184.
465 Ibid., p. 185.
466 “[Solipsismus ist die] ethisch richtige Haltung des kontemplativen Subjekt zur logisch
strukturierten Welt” (Birk 2006, p. 80).
467 Atkinson 2009, p. 90.

213
me to be lip service since he claims, for example: “Conant’s view is that simply
there is no ‘it’ to be either said or shown”468. By “it” is understood here the tran-
scendental reality existing outside the empirical world, so Conant’s opinion
seems to be the one that one should accept if one wants to reject the transcen-
dental reading. But Atkinson comments Conant’s view as follows: “In contrast
to Conant, I believe more plausible position is that ‘it’, as he calls it, cannot be
put into words and cannot be said either to exist or not to exist”469. It seems as
if he distances himself from the simple rejection of the transcendental reality
in favour of a statement that the transcendental reality (including its existence)
cannot be meaningfully expressed. It still resembles the transcendental reading
of nonsense which nonetheless reveals something.
How does one respond to authors who, wildly simplifying, claim: “Yes, we
know that literally transcendental idealism cannot be true. But nevertheless it
reveals something ‘deep’ about the nature of the world”?
In contrast to the proponents of the resolute interpretation, I think that there
are several kinds of nonsense470. This means that I allow the possibility that al-
though a proposition says nothing, it still shows something. For example, strictly
speaking, the proposition: “There are at least two objects in the room” is non-
sense because “object” is a pseudo-concept. What this sentence tries to express
could nonetheless be meaningfully shown by a proposition: “There are at least two
books in the room”471. I will repeat the thesis I defended in Chapter 2, i.e. the dif-
ference between substantial nonsense and mere nonsense consists in the fact that
in the case of substantial nonsense one can show what nonsense aims to express
by means of other meaningful propositions. But in my opinion the transcenden-
tal claims are examples of mere nonsense, and the reason for that, as TLP 6.42
says, is that these propositions aim to express something higher than facts – the
world of values. To be exact, Wittgenstein speaks in TLP 6.42 about the ethical

468 Ibid., p. 105.

469 Ibid., p. 105.
470 For example, Child indicates three kinds of nonsense: propositions about logical
features of the world, metaphysical propositions and ethical-evaluative propositions
(Child 2011, p. 69). On the other hand, a proponent of the resolute reading, Ostrow,
thinks that a reader of the Tractatus should “realize that, in the end, all the pronounce-
ments of [Wittgenstein’s] text are just so much gibberish” (Ostrow 2002, p. 5).
471 “Wherever the word ‘object’ (‘thing’, etc.) is correctly used, it is expressed in concep-
tual notation by a variable name. For example, in the proposition, ‘There are 2 objects
which. . .’, it is expressed by ‘(Ǝx, y)…’. Wherever it is used in a different way, that is as
a proper concept-word, nonsensical pseudo-propositions are the result” (TLP 4.1272).

214
theses but, as we remember, the aim of the whole transcendental venture was to
tear off the veil of deception and to look outside the sphere of appearances. In this
sense, Wittgenstein’s critical remarks refer not only to ethics, but one could also
employ them to other kinds of transcendental interpretations.
The argument that ethical propositions are nonsensical because they do not
aim to describe reality was repeated in Lectures on Ethics. Wittgenstein talked
there about the metaphor of someone who would like to pour a gallon of water
(an ethical proposition) into a cup (a descriptive form of a proposition)472. Eth-
ical pseudo-propositions are not on the same level as descriptions and, there-
fore, one cannot give an example of a proposition which would show what an
ethical proposition purports to say. For example, even a detailed description of
a murder in terms of facts cannot show what the commandment: “Thou shalt
not kill” purports to convey:
If in our world-book we read the description of a murder with all its details physical and
psychological, the mere description of these facts will contain nothing which we could
call an ethical proposition. The murder will be an exactly the same level as any other
event, for instance, falling of a stone (LE, p. 6).

Also, transcendental claims are not on the same level as descriptions because
they aim to express the limits and conditions of the world. Hence, Wittgenstein’s
idea that sometimes one could show what nonsense intends to say has no appli-
cation in the case of transcendental interpretation. Sometimes the proponents of
a theory that assumes the importance of transcendental nonsense indicate “the
mystical” part of the Tractatus. But what is mystical, according to Wittgenstein,
is the fact that the world exists473. One cannot express directly the truth of what is
mystical (The sentence expressing the mystical experience: “It is astonishing that
the world exists” is nonsense because one can be astonished only by something
that exists contingently. I cannot imagine a non-existing world, hence being as-
tonished by the existence of the world is senseless), but the mystical astonish-
ment is shown in every true sentence (because every true sentence assumes the
existence of something). And even if one does not accept this explanation, it
is clear that the astonishment: “Why is there something rather than nothing?”
belongs rather to ontological considerations or to the philosophical proofs of the

472 “Our words used as we use them in science, are vessels capable only of containing and
conveying meaning and sense, natural meaning and sense. Ethics, if it is anything,
is supernatural and our words will only express facts; as a teacup will only hold a
teacup full of water and if I were to pour out a gallon over it” (LE, p. 7).
473 “It is not how things are in the world that is mystical, but that it exists” (TLP 6.44).

215
existence of God (mentioning Leibniz and his Principles of Nature and Grace,
Based on Reason should be sufficient here) than to transcendental considerations
about the limits of the world. Therefore, thesis TLP 6.522 cannot be used in sup-
port of the claim of illuminating nonsense.
The proponents of the resolute interpretation compare nonsense to childish
babbling. An infant confers no information, communicates nothing. But if an in-
fant in its babbling imitates grown-ups, we could say that it is trying to say some-
thing. Similarly as in the case of transcendental discourse. All one could state is the
need to cross the boundaries of sense, or “an urge to transcend our limitations”474 –
a tendency which Wittgenstein respected but nonetheless saw it as “hopeless”:
This running against the walls of our cage is perfectly, absolutely hopeless. (…) But it is
a document of a tendency in the human mind which I personally cannot help respecting
deeply and I would not for my life ridicule it (LE, p. 12)

The transcendental interpretation invokes those formulations of early Wittgen-

stein which are reminiscent of the formulations of a transcendentalist. But their
presence in Wittgenstein’s writings does not mean that their author accepted
transcendental philosophy. This presence could be more easily explained by the
respect that Wittgenstein had for the tendency of the human mind to attack the
limits of sense; the tendency that he held as hopeless. Why hopeless? For exam-
ple, in the Tractatus one reads: “When an ethical law of the form, ‘Thou shalt…’ is
laid down, one’s first thought is ‘And what if I do not do it?’ ” (TLP 6.422). In con-
trast to the Road Runner’s conviction that he is not affected by the law of gravity
because he has never learnt about it, there is nothing paradoxical or amusing in
not being affected by ethical laws. The physical laws are grounded in something
objective, i.e. in facts, and, therefore, one cannot even question: “And what if I do
not follow the laws of physics?”. In contrast to them, ethical laws are grounded in
nothing objective; nor do they express ineffable features of reality. Hence, every
effort to express something ethical in language results in an infant’s babbling.
Concluding, the theory of illuminating nonsense was, so to speak, the last re-
doubt of the transcendental interpretation. The defence proceeded along the fol-
lowing lines: one can reject the theses of the transcendental idealism; these are
not the theses which truly count but are the right attitude to the world which
idealism somehow shows. In the light of Wittgenstein’s writings a transcendental-
ist could be compared to a child imitating a (scientific) conversation of grown-
ups. In contrast to nonsensical expressions containing names referring to logical
pseudo-objects, there is nothing he aims to say. He cannot, therefore, show what

474 Moore 2007, p. 178.

216
the right point of view of the world is. The proponents of the transcendental in-
terpretation rightly claim that in the Tractatus we observe dissatisfaction with the
achievements of science (TLP 6.52). But they are wrong if they think, as it seems
they do, that the problems of life could be solved by childish babbling.

4.5.2  Transcendental reasoning

In the last section of the chapter I want to collect the results of the previous dis-
cussions. In my first remark about the transcendental interpretation I said that its
appeal consists in two things: (1) it reflects Wittgenstein’s intention to find a place
for moral values and a meaning of life in the materialistic worldview of natural
science, and (2) it offers a specific solution with respect to (1), which has the ap-
pearance of plausibility due to numerous similarities between the philosophy of
Schopenhauer and the Tractarian system (sections 4.1).
Then I argued in sections 4.2-4.4, when reviewing specific transcendental in-
terpretations, that in fact (2) is false. One has no proofs that Wittgenstein applied
the transcendental solution to the modernist challenge. In the conclusion of the
chapter I want to further emphasise this result. I claim that on the basis of what we
know about the system of the Tractatus, Wittgenstein not only did not, but also
could not believe in the existence of the transcendental self. This claim is already
contained in the previous sections. In what follows I shall not say something sub-
stantially new, rather, I will extract conclusions from the previous discussions.
The main thesis reads as follows: Wittgenstein could not be a transcendentalist
because transcendental idealism makes usage of transcendental reasoning (later in
the text called TR)475. By TR I understand reasoning of the following form:
Premise 1. There is an ineffable feature of reality – feature (x) (in many differ-
ent versions of transcendentalism, x goes proxy for moral values or a phenom-
enon of subjectivity or a phenomenon of a linguistic meaning).
Premise 2. A pure factual, descriptive discourse does not capture the notion of
(x). The existence of (x) remains unexplainable on the basis of factual discourse.
Conclusion: There must be a bearer X of a feature (x), and the existence of X
explains the existence of x (in some versions it guarantees the existence of x) (In
different versions of the transcendental interpretation X goes proxy for the notions

475 It is a form of reasoning introduced by Kant. As a Kantian scholar, Ralph Walker, writes:
“Kant’s transcendental arguments start from the premise that we have experience, or
knowledge.(…) Kant’s transcendental arguments typically have a second premise, to
the effect that p is a condition of the possibility of experience. This is tantamount to a
conditional of the form ‘If there is experience, then p.’ ” (Walker 1993, p. 61–62).

217
of the willing subject, the metaphysical self, the solipsistic self or the user of lan-
guage; all of which refer to the transcendental self).
We recognise this form of reasoning from the review of Schopenhauer’s phi-
losophy and from the review of the transcendental interpretations. In what fol-
lows I want to argue that TR could not occur in Wittgenstein’s writings. First of
all, in the Tractatus there cannot be any TR simply because it would speak against
the main conclusion and result of the book, i.e. to discern meaningful and sense-
less discourse by saying nothing more than the propositions of natural science:
The correct method in philosophy would really be the following: to say nothing except
what can be said, i.e. propositions of natural science—i.e. something that has nothing to
do with philosophy—and then, whenever someone else wanted to say something meta-
physical, to demonstrate to him that he had failed to give a meaning to certain signs in
his propositions. Although it would not be satisfying to the other person—he would not
have the feeling that we were teaching him philosophy—this method would be the only
strictly correct one (TLP 6.53).

A proponent of the presence of TR in the Tractatus has problems reconciliating

TR with TLP 6.53. The intention of the former is to overcome the limitations
of science and to gain insight into the world behind appearances. The intention
of the latter is to get these desires under control. In my opinion, holding both
TR and TLP 6.53 is untenable; hence there can be no transcendental reasoning
in the Tractatus.
Another reason to push ahead is that, according to the Tractatus, meaning
and value cannot be meaningfully expressed. A proponent of the transcen-
dental interpretation would claim then that the inexpressibility of human val-
ues is Premise 1 of TR. But why does a transcendentalist accept this premise?
After all, famous predecessors of transcendental idealism – Schopenhauer and
Kant – would not have accepted it. It is strange even for a proponent of the
transcendental argumentation – Peter Hacker: “The argument in support of the
ineffability of ethics is tenuous to say the least”476. Why did Wittgenstein limit
meaningful discourse to discourse referring to empirical facts? Of course, there
is a simple answer – the one which the proponent of the materialistic interpreta-
tion of the Tractatus would give. It is because ultimately only material objects
exist. At the same time, it is the answer that a proponent of the transcendental
reading cannot accept because, in his view, the Tractatus allows for the existence
of entities ontologically irreducible to material points. The transcendental inter-
pretation accepts a claim (ethical truths are nonsensical) whose acceptance is

476 Hacker 1972/1986, p. 83.

218
unexplainable in terms of the transcendental assumptions. A transcendentalist
says that ethical propositions are nonsensical from the point of view of natural
science. In contrast to this view, Wittgenstein claims that ethical propositions
are simply nonsensical. Just this difference should indicate that there cannot be
any TR in the Tractatus.
Thirdly, there is also something confusing and intractable in the conclusion of
TR. If in the conclusion we affirm the existence of a bearer of ineffable features
of reality, then the following question arises: “What is the relation between a
transcendental self and a self of everyday life that is the embodied individual?”.
It seems that it is not the relation of identity. Wittgenstein precisely says that the
metaphysical subject is not the human being; it is a limit of the world, not a part
of it (TLP 5.641). But if we cannot identify the selves of everyday experience with
transcendental selves, and if the transcendental self lies outside the empirical
reality, then we do not have any means to identify such a self (so, for example,
we do not know if, in fact, there is one or perhaps more metaphysical selves
coordinated with one empirical reality). It follows that the concepts of the meta-
physical or the willing subject are empty ideas. This is a conclusion with which
a transcendentalist should agree – after all, he claims that all we know about the
transcendental self is that it is a condition of the empirical world, and that we do
not even know if the category of an object applies to it. This metaphysical prac-
tice of conferring no meaning on signs was exactly the target of Wittgenstein’s
condemnation (TLP 6.53). He clearly distanced himself from this way of “mak-
ing” philosophy. Again, it seems impossible that Wittgenstein could employ TR
in his philosophy.

Summary
The result of the analysis of the chapter is that the transcendental interpretation
fails: its principles fail and it fails in providing arguments in its defence from the
text of the Tractatus. In my opinion, if Wittgenstein was concerned with the un-
contested triumph of science, then he did not search for the means to safeguard the
world of human values in the transcendental philosophy. This philosophy, as well
as the dualism of mind and matter that was analysed in Chapter 3, was rejected by
Wittgenstein. Both of these philosophies pretend to say something, whereas they
are nonsensical. They do not confer any meaning on the concepts they use. The
modernistic anxiety of Wittgenstein must have found a different way of comfort.
In the next chapter I shall show what this way was, and I shall do so by explaining
what, in my opinion, Wittgenstein truly had in mind when he talked about the
truth of solipsism and the world which coincides with my world.

219
Chapter 5. Ethics in Wittgenstein’s
Early Writings

In the previous chapters I advanced arguments in favour of the thesis that the
Tractatus grants science exclusive authority in discovering the truth about reality;
that the simple objects of the Tractatus are physical atoms and that, on the other
hand, the metaphysics of the self (independently whether defined in the dual-
istic or the transcendental tradition) fails to safeguard the world of values and
meaning from the claims of scientism. This interpretation of the Tractatus places
early Wittgenstein in the wider intellectual movement (pervading philosophy,
literature and theology) of the first decades of the 20th century called modernism.
This interpretation, however, leaves some questions unanswered. In this con-
text I mentioned in Chapter 4 the Tractarian remark that:
what the solipsist means is quite correct; only it cannot be said, but makes itself manifest
(TLP 5.62).

The second claim with which the naturalistic view on the ontology of the Tractatus
has interpretational difficulties is thesis TLP 5.641, in which Wittgenstein states
that (although it seemed that he rejected both the dualistic as well as the transcen-
dental conception of the self) there is a specific, philosophical sense of the notion
of the self. The sense of this notion falls, however, outside the scientific and psy-
chological definition, and it is connected with the fact that the world is my world:
There really is a sense in which philosophy can talk about the self in a non-psychological
way. What brings the self into philosophy is the fact that ‘the world is my world’. The philo-
sophical self is not the human being, not the human body, or the human soul, with which
psychology deals, but rather the metaphysical subject, the limit of the world—not a part
of it (TLP 5.641).

These claims, i.e.: (a) the correctness of solipsism, (b) identification of the world
with my world, and (c) the claim that the self of which philosophy is talking about
is the limit of the world, should make us aware of the dialectical situation of the
Tractatus. Wittgenstein accepts in his book, on the one hand, methodological sci-
entism (TLP 4.11) and, at the same time, he holds scientific solutions as insufficient
answers to the problems of life (TLP 6.52). This tension between the basic claims of
the Tractatus led its author to the conviction that there are two aspects of the book:
one aspect consists in what is explicitly claimed and the second consists in what
was purposely left unsaid. As Wittgenstein admitted in his letter to his publisher,
Ludwig von Ficker,:

221
 My work consists of two parts, the one presented here plus all that I have not written.
And it is precisely this second part that is the important point. For the ethical gets its limit
drawn from the inside, as it were, and I am convinced that this is the ONLY rigorous of
drawing those limits; (…) I have managed in my book to put everything firmly into place
by being silent about it (cited in: ProtoTractatus, p. 16)

If we looked at the Tractatus through the eyes of its author, then we would notice
that his main concern and problem was to reconcile naturalism (expressed in the
convictions that the correct method of philosophy is to say nothing except what
the natural sciences say, or that the fact that we cannot meaningfully express the
problems of life suggests that the riddle does not exist, etc.) with defence of the
views of those who are sensitive to moral, aesthetic and religious values. I called
this main concern the fundamental problem of the Tractatus. The hypothesis that
in the Tractatus one observes two contradictory tendencies, and the explanation
for this observation consisting in a conjecture that probably Wittgenstein wanted
to find a place for ethics and religion in a worldview shaped mainly by natural
science, is nothing new in Wittgensteinian literature477. However, what is original
in my interpretation is that I try to answer the fundamental problem after having
fully acknowledged the ontological materialism of the Tractatus478. I admit that
early Wittgenstein fully embraced the consequence of the naturalistic worldview,
according to which the world consists only of masses (or whatever will turn out to
be the most elementary particle of physics) and their movements. I claim that this
is the standpoint from which Wittgenstein tried to secure the legitimacy of moral
judgements or religious experience. In my conclusions I assume that Wittgen-
stein’s views with respect to the fundamental problem of the Tractatus have not
changed, at least until the delivery (in November of 1929) of a lecture on ethics to
the Heretics Society at Cambridge University479. My answer to the fundamental

477 “Wittgenstein is concerned to understand how there can be meaning and worth
in the seemingly nihilistic world of modern science where the order of nature has
been divorced from any conception of value” (Kelly 1995, p. 571); a similar view is
represented by, among others: Birk 2006 and Stokhof 2002.
478 Most of the time there are the proponents of the transcendental interpretation who
undertake this effort.
479 One confirmation of this could be the fact that in 1929 Wittgenstein still held the
basic principle of the Tractarian system, i.e. its atomism – although, as Peter Hacker
wrote, it was “the last gasp of Wittgenstein’s atomism” (Hacker 1986a, p. 85). In the
Lecture on Ethics, Wittgenstein repeats his earlier conviction that a true and complete
description of the world would consist only of propositions of science; that to be an
omniscient person it is enough to have knowledge about movements of bodies and
states of minds (“Suppose one of you were an omniscient person and therefore knew

222
problem assumes that Wittgenstein in A Lecture on Ethics expressed explicitly
what was left unsaid in the Tractatus. In this chapter I shall chiefly concentrate on
Wittgenstein’s account of ethics.
One should keep in mind that, according to Wittgenstein, ethics is the enquiry
into what:
–– is valuable
–– is truly important
–– is the meaning of life
–– makes life worth living
–– is the right way of living480.

all the movements of all the bodies in the world dead or alive and that he also knew
all the states of mind of all human beings that ever lived…” (LE, p. 6)). He admits
that the scientific propositions contain no ethical concepts (for instance, the concept
“good” in its absolute meaning). He rejects the idea (exemplified by Shakespeare’s
Hamlet) that states of mind are facts which could be described as good or bad: “What
Hamlet says seems to imply that good and bad, though not qualities of the world
outside us, are attributes to our states of mind. But what I mean is that a state of
mind, so far as we mean by that a fact which we can describe, is in no ethical sense
good or bad” (LE, p. 6). He concludes, therefore, that a state of affairs with an ethical
quality is a chimera (“I want to say that such a state of affairs is a chimera. No state
of affairs has, in itself, what I would like to call the coercive power of an absolute
judge” (LE, p. 7)). An ethical sentence is sensu stricto nonsensical; every time we
try to express our experiences of absolute (ethical) values, we formulate something
incomprehensible (“The first thing I have to say is, that the verbal expression which
we give to these experiences is nonsense! If I say ‘I wonder at the existence of the
world’ I am misusing language” (LE, p. 7). “But it is nonsense to say that I wonder at
the existence of the world, because I cannot imagine it not existing” (LE, p. 8).). The
existence of these motifs in the Lecture on Ethics suggests that there is no substantial
change with respect to ethics in Wittgenstein’s thinking from the time of finishing
the Tractatus. The only difference consists in the fact that this time Wittgenstein does
not follow his own advice and he speaks about what cannot be said. He succumbs to
the temptation to cross the limits of language and he tries to express what an attitude
to the world as a whole could be – what one claims when one says: “X is good”.
480 “Now instead of saying ‘Ethics is the enquiry into what is good’ I could have said
Ethics is the enquiry into what is valuable, or, into what is really important, or I could
have said Ethics is the enquiry into the meaning of life, or into what makes life worth
living, or into the right way of living. I believe if you look at all these phrases you will
get a rough idea as to what it is that Ethics is concerned with” (LE, p. 5).

223
Accordingly, as moral values I shall understand not only the traditionally men-
tioned values of good and evil in this context, but also, for instance, the meaning
of life or the values which make life meaningful. Wittgenstein’s views on religion
are not the subject-matter of this presentation. I shall, however, use these views as
a convenient analogy to his views on ethics. Consequently, I shall use the term: “an
ethical sentence (expression, utterance)” as referring to those sentences which aim
at expressing the claims of ethics (in the meaning accepted by Wittgenstein481), and
as synonymous with the term: “an axiological sentence (expression, utterance)”.
In the following passages I shall be interested in Wittgenstein’s reasons to
claim that ethical sentences are meaningless (section 5.1). On the basis of the
conclusions from these considerations I shall next formulate the position I call
subjectivism with respect to ethics. In short, I will defend the interpretation ac-
cording to which early Wittgenstein supported the following theses (section 5.2):
(A) “X is good” expresses the attitude of the speaker
(B) An attitude is subjective, which means it is not based on reason. By reason
I understand something which is (1) a fact that is accessible independently
of a subjective point of view and (2) has persuasive power. A rational person
confronted with such a defined reason has to change his or her mind or at-
titude. The aim of this description of reason is to exclude from the scope of
reasons mystical experiences and Kantian imperatives.
(C) The subjective turn in Wittgenstein’s philosophy consists in the fact that an
attitude comes from an individual decision, a point of view or a recognition
of what “the world as a whole tells me”.
Next I shall show how subjectivism helps us to reconcile the materialistic inter-
pretation with the Tractarian theses that: what solipsism states is quite correct,
that the metaphysical self is the limit of the world and that the world is my world
(section 5.3). In this way I hope to defend the ontological materialism of the
Tractatus and, at the same time, I intend to deliver a probable answer to the fun-
damental problem of this book.

5.1  The nonsense of ethics in the eyes of early Wittgenstein

The answer to the fundamental problem of the Tractatus can be found in the let-
ter to von Ficker that I quoted earlier: Wittgenstein defends the world of human

481 “I am going to use the term Ethics in a slightly wider sense, in a sense in fact which
includes what I believe to be the most essential part of what is generally called Aes-
thetics” (LE, p. 4).

224
values by recommending silence about things that, in his eyes, truly matter. There
is a realm of facts which one can express in propositions, and there is an aspect of
the world which can only be shown. And because it can only be shown, it cannot
be said482. If it cannot be said, then one should keep silent about it. Of course, this
is the proper answer in accordance with the text of the Tractatus but, on the other
hand, we feel that without a closer look at the presented line of reasoning it sounds
slightly mysterious and dogmatic. We cannot understand how silence about values
can protect them from the claims of natural science. In the following sections I
shall elaborate on Wittgenstein’s claims.
According to the Tractatus, the sentences of ethics are meaningless:
It is impossible for there to be propositions of ethics. Propositions can express nothing
that is higher. It is clear that ethics cannot be put into words. Ethics is transcendental
(TLP 6.42-6.421).

In the previous chapters I indicated two criteria discerning propositions from any
linguistic expressions. First, the elements of a proposition have to occur in it ac-
cording to their roles in logical syntax. This condition is fulfilled by the ethical
sentences. Secondly, the referents of all elements of a proposition must exist. In
Chapter 1 I advanced arguments in favour of the hypothesis that early Wittgen-
stein adhered to a realistic theory of meaning, which assumes that there are “bits”
of reality existing independently of our cognising capacities or of the systems of
representations by means of which we represent reality. If Wittgenstein claims that
there are no ethical propositions, then clearly he thinks that among the objects of
the world there are no objects that would make an ethical sentence true or false.
Indeed, the Tractatus seems to adhere to a sharp distinction between facts and
values:
In the world everything is as it is, and everything happens as it does happen: in it no
value exists—and if it did exist, it would have no value (TLP 6.41).

But what does it mean that there are no values in the world? A very attractive hy-
pothesis suggests that it means that values are transcendental, i.e. existing “outside
the sphere of what happens”.

5.1.1  The absoluteness of moral values

The hypothesis that ethical sentences are meaningless because ethical notions are
empty, and they are empty because moral values belong to the realm of what is
higher as opposed to the realm of empirical facts, seems to be supported by some

482 “What can be shown, cannot be said” (TLP 4.1212).

225
of early Wittgenstein’s utterances. In the Tractatus we read that “if there is any
value that does have value, it must lie outside the whole sphere of what happens
and is the case” (TLP 6.41). The thesis that there are no ethical propositions seems
to be connected with the thought that “propositions can express nothing that is
higher” (TLP 6.42), which could suggest that ethics aims to express something
that is higher. Finally, Wittgenstein claims that “ethics is transcendental” (TLP
6.421). Later, in A Lecture on Ethics, we also find expressions that bring to mind a
transcendental solution to the problem of finding a place for values in the world of
scientific facts. For example, with respect to the ethical values, Wittgenstein uses
a metaphor of a teacup (propositions of science) into which one wants to pour a
gallon of water (ethical values). He writes:
Our words used as we use them in science, are vessels capable only of containing and
conveying meaning and sense, natural meaning and sense. Ethics, if it is anything, is
supernatural and our words will only express facts; as a teacup will only hold a teacup
full of water and if I were to pour out a gallon over it (LE, p. 7).

The choice of words – ethical values are supernatural and absolute – is peculiar:
again, it could suggest that the existence of values is somehow “higher” than the
existence of facts, which is reported by means of scientific propositions capable of
conveying only “natural” sense. It could be reminiscent of the practice of the propo-
nents of the apophatic theology who claim that the reason for which we cannot ex-
press propositions about God is the fact that His being is too perfect and transcends
everything, and that His perfection entails that our words cannot refer to Him.
Interestingly, commentators who pay attention to the religious awakening of
Ludwig Wittgenstein during the time of World War I and his openness to the
problem of values usually interpret his conversion as a turn in the direction of
transcendental philosophy. A typical example in this respect is Monk’s biogra-
phy of Wittgenstein, where the mystical part of the Tractatus is summed up as
a restatement of Schopenhauer’s transcendental idealism483. Wittgenstein, once
freed by Frege’s and Russell’s realism from the enchantment of idealism, accord-
ing to this interpretation, “apparently relapsed back into it”484. In Chapter 4 of
this dissertation I analysed which fragments of Wittgenstein’s writings could in-
voke associations with transcendentalism, so I will not repeat these arguments

483 “Wittgenstein’s remarks on the will and the self are, in many ways, simply a restate-
ment of Schopenhauer’s ‘Transcendental Idealism’ ” (Monk 1991, p. 144).
484 “Frege, the thinker Wittgenstein credited with freeing him from his earlier Scho-
penhauerian idealism, was not, apparently, told of his relapse back into it” (ibid.,
p. 144–145).

226
here. I shall just recall that, in my opinion, a closer look at the process of writing
the Tractatus reveals that there is indeed a transcendental stage (which finds
its reflection in the Notebooks) in Wittgenstein’s thinking, but it was overcome
by the time of the final composition of the text. One can point at the moment
of withdrawal from the metaphysics of the transcendental self – it must have
been the winter of 1917/1918. From that period on we have at our disposal cor-
respondence between Wittgenstein and his friend, Paul Engelmann. In one of
the letters, Engelmann noticed another change of attitude in Wittgenstein: “It
seemed to me as if you – in contrast to the time you spent in Olmütz, where I
had not thought so – had no faith”485. Wittgenstein confirmed his friend’s con-
jectures. In his response he wrote:
I am far too bad to be able to theorize about myself; in fact I shall either remain a swine
or else I shall improve, and that’s that! Only let’s cut out the transcendental twaddle when
the whole thing is as plain as a sock on the jaw (Englemann 1967, p. 12; mine emphasis)

Wittgenstein’s moral self-examination is of lesser interest to me. What interests

me is the apparent aversion to “the transcendental twaddle”. This state of mind
could explain why, by the time of having finished the Tractatus, all transcendental
ideas of the self existing outside the empirical world, the dichotomy of the world
of noumena and the world of appearances, etc. vanished from Wittgenstein’s book.
Are moral values the last vestiges of transcendentalism in Wittgenstein’s early phi-
losophy?
In my opinion, the choice of words should not mislead us: for example, in the
Tractatus not only ethics but also logic is described as transcendental (TLP 6.13).
In the latter case this can mean that logical truths have nothing to do with what
the world looks like (“Logic is prior to every experience – that something is so”
(TLP 5.552)) or that they are senseless (“Tautologies and contradictions lack
sense” (TLP 4.461)). Similarly, in my opinion the point of using the notion “tran-
scendental” with respect to ethics is to underline the dichotomy between facts and
values. It also means, as I argued in Chapter 4, that in the eyes of Wittgenstein,
ethical expressions “attack the limits of language”; they transcend, so to speak, the
limits of what one can meaningfully express.
Some of Wittgenstein’s formulations, as I mentioned before, seem to link the
idea of the meaninglessness of ethics with the view that propositions express facts
of the empirical world, and as such they are not capable of conveying the absolute-
ness of values and the necessity of moral laws. It is true that with respect to values
Wittgenstein uses the predicate “absolute”, but it does not indicate the nature of

485 Cited in: Monk 1991, p. 152.

227
the values486. In A Lecture on Ethics he elaborates on the subject and on the basis of
his remarks one can discern two meanings of the concept of the absolute value. In
the negative sense the claim that moral values are absolute says that they are not
relative. It is not the same as to say that early Wittgenstein opposed moral relativ-
ism (understood as a claim that the truth and justifications of moral judgements
are absolute). In fact, I do not think that his position prevents one from accepting
relativism. Wittgenstein intends to discern two uses of the predicate “good”, i.e.
the trivial and the relative sense of “good”, and an ethical and absolute sense of
“good”487. The predicate “good” is used in a relative way, according to Wittgen-
stein’s description, when it is possible to express a sentence containing evaluative
notions with the help of a sentence containing purely descriptive notions. Let us
take, for example, the expression: “Usain Bolt is a good runner”. Obviously, this
expression, strictly according to Tractarian rules, is nonsense because there is no
such quality of objects as “good”, but it is a substantial nonsense. One can show
what it is trying to say by means of the proposition: “Usain Bolt runs regularly 100
metres under 10 seconds”. On the other hand, to say that moral values are absolute
amounts to saying that one cannot express ethical sentences by means of mean-
ingful expressions. One cannot give any example of a proposition which would
show what the expression: “Helping others is good” is trying to say.
The notion of an absolute value in its positive sense expresses the binding char-
acter of some experiences. According to early Wittgenstein, the absoluteness of
moral experience (or of moral values) means that in this experience one evaluates
something, and this evaluation necessarily changes one’s behaviour or attitudes.
For example, one can evaluate the ability to play tennis as something good, and, at
the same time, not have a need to improve one’s tennis play. One, however, cannot
stay indifferent with respect to moral evaluations: if I evaluate that helping others
is morally good, then necessarily I will feel obliged to help others (it is a different
story if I indeed do so). There is definitely something strange and paradoxical in
the utterance: “I evaluate feeding starving children as morally valuable but still I
do not feel any inclination to do so”488. When Wittgenstein says that ethical values

486 In A Lecture on Ethics, Wittgenstein uses the expression: “a judgement of absolute

value”, but he introduces the distinction absolute-relative values using the term: “an
absolute judgement of value” (LE, p. 5).
487 LE, p. 5.
488 “Supposing that I could play tennis and one of you saw me playing and said ‘Well,
you play pretty badly’ and suppose I answered ‘I know, I’m playing badly but I don’t
want to play any better,’ all the other man could say would be ‘Ah then that’s all right.’
But suppose I had told one of you a preposterous lie and he came up to me and said

228
are absolute, he means that if one evaluates something as morally right or wrong,
then it necessarily affects one’s attitudes. However, this does not mean that moral
values are transcendental or that they belong to the higher and objective order
of things. Moral values even cannot belong to the transcendental and objective
order of things, because if they did, then just by being confronted with them or
by experiencing them we would have to react in a particular way. There would
be no differences in people’s evaluations and, consequently, in their conduct and
attitudes. And Wittgenstein was aware that that is not the case:
And similarly the absolute good, if it is a describable state of affairs, would be one which
everybody, independent of his tastes and inclinations, would necessarily bring about or
feel guilty for not bringing about. And I want to say that such a state of affairs is a chimera
(LE, p. 7).

In early Wittgenstein’s ethics there are no contradiction between the absolute-

ness of a value and its subjectivity or moral relativism. Even if I am aware that it
is my subjective evaluation that it is good to feed a starving child and, perhaps,
other people evaluate the same state of affairs differently, this does not stop me
from feeding it. Still, my evaluation inclines me to behave in a certain way.

5.1.2  The sense of proposition and the idea of objectivity

Ethical sentences, in the assessment of the Tractatus, are nonsensical. This means
that they contain empty names. Wittgenstein expresses this thought by saying that
there are no values in the sphere of what happens or in the sphere of what is the
case. In the last section I tried to argue in favour of the thesis that these statements
do not necessarily mean that he puts moral values in the higher (one could as
well use such adjectives as transcendental, non-empirical or supernatural) order
of things. In the following section I shall test the hypothesis that the claim that
there are no ethical propositions is a result of the dichotomy between facts and
values. In this view, ethics does not deal with facts but with the evaluations or in-
terpretations of facts. Because evaluations and interpretations are subjective, then
sentences aiming at expressing these evaluations are meaningless. Correctness of
such a hypothesis would mean that one could classify early Wittgenstein’s position
in ethics as subjectivism.

‘You’re behaving like a beast’ and then I were to say ‘I know I behave badly, but then I
don’t want to behave any better,’ could he then say ‘Ah, then that’s all right’? Certainly
not; he would say ‘Well, you ought to want to behave better.’ Here you have an absolute
judgment of value, whereas the first instance was one of a relative judgment” (LE, p. 5).

229
 Young Wittgenstein clearly connects the concept of a meaningful expression
with the idea of objectivity. In thesis TLP 4.06 we read:
A proposition can be true or false only in virtue of being a picture of reality.

A proposition has to correspond to something which is objective, independent of

the mind or of a system of representation (including language), something which
is accessible to more than one cognising subject; something to which we can
refer in case of a discussion or disagreement, and something which one simply
states without considering what its meaning or value is: “A proposition shows how
things stand if it is true. And it says that they do so stand” (TLP 4.022). One could
argue in favour of the truthfulness or falsity of a given proposition because there is
something that both sides of a contention can agree on – it is the existence of sim-
ple objects – an existence which is not dependent on any subjective preferences or
predilections, individual knowledge or emotions of both sides of a discussion or
their particular point of views: “In order to tell whether a picture is true or false
we must compare it with reality” (TLP 2.223). In the system of the Tractatus the
sense of a proposition is identified with what is pictured by a proposition: “What
a picture represents is its sense” (TLP 2.221). A proposition pictures a concatena-
tion of simple objects which, if I was correct in Chapter 2, are physical atoms.
The only thing to settle is the answer to the question whether physical atoms are
connected with one another in such a way as is mirrored by a given proposition.
If that is correct, i.e. if the notion of sense in the Tractatus is truly connected
with the idea of objectivity, then it gives an indication of the status of ethics. From
this perspective, if an expression is nonsensical then it can mean, among other
things, that it expresses something subjective, irrational, partial, biased, or ac-
cepted only on the basis of a subject’s own point of view or predilection. In my
interpretation of the Tractatus this is the case for moral judgements. Sentences of
ethics are nonsensical because their aim is not to express facts but evaluations and
interpretations of facts.
The larger picture of Wittgenstein’s account is expressed in thesis TLP 6.43:
If the good or bad exercise of the will does alter the world, it can alter only the limits of
the world, not the facts—not what can be expressed by means of language. In short the
effect must be that it becomes an altogether different world. It must, so to speak, wax and
wane as a whole. The world of the happy man is a different one from that of the unhappy
man (TLP 6.43).

It seems that Wittgenstein says here: everything in the world that has value, has it
because of will. It is the will that permeates with meaning an otherwise cold world
of brute, empirical facts. The world is described in the Notebooks as “a world in
which there is only dead matter” (NB 2.8.16, p. 79). Wittgenstein repeats twice

230
that: “The world in itself is neither good or evil” (NB 2.8.16, p. 79). If we describe
states of affairs in moral terms it is because of the relation of things to the will:
“Things acquire ‘significance’ only through their relation to my will” (NB 15.10.16,
p. 84)489. The quotation marks around the word “significance” are important. They
indicate that it is not the case that my will makes things valuable, or that my will
is a special way of looking at things which discovers features that are hidden from
scientific research. I think that Wittgenstein is saying here simply that we evaluate
things as significant to us. The expression: “Killing is wrong” has no factual mean-
ing, it is incapable of conveying any content, it is neither true or false; it expresses
only the attitude of the speaker of that expression, one’s disapproval, terror or
some other negative emotion towards an event in the world, or one’s negative
evaluation of an event. The expression: “Killing is wrong” is not a proposition but
rather a substitute for an avowal. Confirmation of this understanding of Wittgen-
stein’s views on ethics I find in A Lecture on Ethics, in which he analyses the case
of a murder. According to his position, a description of events referring only to
facts would contain nothing indicating the moral meaning of the events. In this
respect there would be no difference between a deed we call a murder or a deed
we call watching a film. Descriptions of both of these events would contain only
propositions referring to movements of physical things, and on the basis of such
a description one could not conclude which event is the subject of moral disgust:
If for instance in our world-book we read the description of a murder with all its de-
tails physical and psychological, the mere description of these facts will contain nothing
which we could call an ethical proposition. The murder will be on exactly the same level
as any other event, for instance the falling of a stone (LE, p. 6)

A murder is definitely devastating to the family and friends of the victim. This
event makes their world miserable, but if they said to the murderer: “You should
not have done it”, they would simply be evaluating his deed. A purely factual
description of the event would not even contain the notion of a murder but the
movements of bodies, such as taking a stone, throwing it at the victim, hitting
the victim on the head, etc. The world seen sub specie aeternitatis is indifferent
to the moral values of facts.
According to my interpretation of Wittgenstein, it is through individual evalu-
ations of a subject that things acquire moral meanings. This means that, according
to subjectivism (similarly to transcendentalism), values enter the world because
of the activity of the subject (the transcendental version of this claim says that

489 As Wittgenstein once noted: “The human gaze has the power of making things pre-
cious” (CV, p. 3).

231
values enter the world through the activity of the willing subject). The advantage of
subjectivism, however, consists in the fact that it does not assume any metaphysics
of the self. It does not place the self outside the world of empirical facts. The prob-
lem of the transcendental interpretation was to show the connection between the
willing subject and the subject of everyday experience. Subjectivism, on the other
hand, can simply identify the evaluating subject with the subject of an everyday
experience. It was Ludwig Wittgenstein – a human being – who wondered about
the existence of the world or felt safe whatever happened in the world, or whatever
happened to him. These experiences, in this sense, are nothing exceptional490, they
are not transcending the realm of empirical things and they do not indicate at the
existence of some outer reality. They are simply the experiences of a human being
who was known to the world as Ludwig Wittgenstein.
Secondly, the transcendental interpretation claims that Wittgenstein, in se-
curing the position of values in our worldview, made use of ontotheology; that
he tried to sketch a picture of the world which would be competitive and more
comprehensive to the scientific one. In my opinion, this is in disagreement with
Wittgenstein’s thoughts on the role of philosophy. Its role was to guard the limits
of sense. Subjectivism does justice to this conception because its account of ethics
is a direct consequence of the considerations on what can be said. The authority
of science refers to the realm of facts. Moral judgements, because they express
nothing more than one’s attitudes towards the world (in other words, one’s evalu-
ations), cross the limits of sense, and Wittgenstein, consequently, counsels silence
with respect to them. Someone who is a proponent of subjectivism in ethics does
not compete with the scientific worldview; he does not claim that they are aspects
of reality that escape scientific research and can admit that all that can be said is
said by the natural sciences. He only states that we interpret reality, react to real-
ity and evaluate it in different ways; that, for example, the totality of facts says
something to some of us. In the next section we will see that reality was saying to
Wittgenstein something such as: “Nothing can happen to me” or “The existence of
the world as a whole is mysterious”, which, obviously, for naturalistic or neo-pos-
itivistic sensitivity is quite confusing but, on the other hand, it is also something
that subjectivism accepts by admitting that ethical utterances convey no content.

490 The exceptionality of ethical expressions consists in their effort to convey the neces-
sary binding of moral values; that one having evaluated something as a moral value
is inclined to change an attitude or a behaviour.

232
5.2 Early Wittgenstein’s subjectivism
In my opinion, if we reflect on the question as to how one could defend the views
of those sensitive to moral or aesthetic values, since the world is nothing more than
the totality of physical particles, then early Wittgenstein’s answer could be classi-
fied as subjectivism. In this section I shall develop the topic of Wittgenstein’s sub-
jectivism491 (section 5.2.1). I will often refer to Wittgenstein’s A Lecture on Ethics
and his views on religion. I shall also show that subjectivism explains why the Trac-
tatus ends with the cousel of silence (section 5.2.2). At the end I shall compare my
understanding of Wittgenstein’s position with Russell’s emotivism (section 5.2.1).

5.2.1 Ethical sentences as expressions of attitudes

According to my interpretation, with respect to ethics early Wittgenstein held
the following theses:
(A) the aim of an expression: “X is good” is to inform about an attitude to the
world (or, in other words, when A says that X is good, he or she evaluates X as
valuable492).
(B) an attitude to the world has nothing to do with the facts; it is a subjective
interpretation of what the world looks like to me. Earlier I connected the no-
tion of reason with the notion of fact. Hence, if attitudes or evaluations have
nothing to do with facts, then they are not based on reason.
AD (A) Earlier I pointed to the Tractarian theses and notes from the Notebooks
which prove that Wittgenstein believed in the dichotomy between facts and val-
ues. One of the consequences of his standpoint was that there are evaluations of
a subject that cast moral values on the world of empirical facts. In the talk given
to the Heretics Society at Cambridge University, Wittgenstein was more specific
when it comes to ethical questions. He described, for example, ethics as a desire to

491 The term “subjectivism” is not fortunate. It can suggest that early Wittgenstein formu-
lated a theory similar to those adhered to by Ayer or Stevenson. I shall later compare
the views which I ascribe to early Wittgenstein with emotivism propagated in the same
period by Russell. For now I just want to stress the fact that Wittgenstein obviously did
not contrive any ethical theory. I simply extract from different fragments of Wittgen-
stein’s early writings his silently accepted assumptions. There is no explicit confirmation
of subjectivism in the Tractatus but, in my opinion, if we interpret its position on ethics
as subjectivism, we are able to make Wittgenstein’s views consistent – we can show how
one can reconcile ontological materialism with a mystical attitude to the world.
492 For the purposes of this chapter I think of evaluations as of a kind of attitude to the
world.

233
say something about the meaning of life, the absolute good or the absolute value.
In his view, ethics mirrors not what happens in the world493 but our attitudes to
the world. This hypothesis – that ethics reflects our attitudes to the world – is
strengthened when one takes into account what Wittgenstein compares the expe-
riences of the absolute values with. When pushed to express what experience of
the absolute good consists in he mentions: a feeling of wonder at the existence of
the world, a feeling of absolute safety and its negative – a feeling of guilt.
I believe the best way of describing it is to say that when I have it I wonder at the exist-
ence of the world. And I am then inclined to use such phrases as ‘how extraordinary that
anything should exist’ or ‘how extraordinary that the world should exist.’ I will mention
another experience straight away which I also know and which others of you might be
acquainted with: it is, what one might call, the experience of feeling absolutely safe. I mean
the state of mind in which one is inclined to say ‘I am safe, nothing can injure me whatever
happens.’ (LE, p. 8)

I think that I am not distorting Wittgenstein’s thinking when I reconstruct his

thought into the form of a subjectivist theory that (A): when one says: “X is abso-
lutely good”, one expresses an attitude to the world (in the case of Wittgenstein this
would be his attitude of wonder and trust, or of evaluating something as worthy
of his trust and admiration). My reading of early Wittgenstein’s writings is in this
respect similar to Anscombe’s. She provides a metaphor of the world as a face. The
look of a face is a matter of facts. What is true (or false) about the world is also a
matter of facts. But just as one can interpret a face in different ways (it could seem
happy or sad to us, grave or grim, scary or funny), we can also interpret the facts
of the world494. In Wittgenstein’s view a religious believer and a non-believer can
agree when it comes to stating the facts; the difference consists in the conclusions
they draw from the same set of facts. In my opinion, Wittgenstein wants us to say
that the conclusion has nothing to do with the facts495. He himself experiences

493 “What [ethics] says does not add to our knowledge in any sense” (LE, p. 12).
494 “There is a strong impression made by the end of the Tractatus, as if Wittgenstein
saw the world looking at him with a face; logic helped to reveal the face. Now a face
can look at you with a sad or happy, grave or grim, good or evil expression (…). And
so he speaks of the world ‘waxing or waning as a whole’, i.e., in terms of my analogy,
as having more or less expression, or a good or evil expression. The world thought
of, not as how things are, but as however they are – seen as a whole – is the matter
of logic; thought of as my life, it is the matter of ethics” (Anscombe 1965, p. 172).
495 “How things are in the world is a matter of complete indifference for what is higher.
God does not reveal himself in the world. The facts all contribute only to setting the
problem, not to its solution” (TLP 6.432-6.4321).

234
this almost religious wonder at the existence of the world, but he definitely did
not conclude that the world’s existence as a whole is mysterious on the basis of
facts. Hence, the second claim of Wittgenstein’s subjectivism is: (B) An attitude is
subjective that is not based on reason.
AD (B). Wittgenstein admits that although he chose three of the aforementioned
feelings as expressing absolute values, he does not exclude that other people would
choose other ones. With respect to attitudes to the world as a whole, Wittgenstein
says:
In my case, it always happens that the idea of one particular experience presents itself to
me which therefore is, in a sense, my experience par excellence” (LE, p. 8, mine emphasis).

He is aware that his attitudes to the world as a whole are his own interpretation
of the facts. It is in his case that when facing the world he experiences wonder
at its existence. Others could have different interpretations and evaluations and,
therefore, different attitudes.
Important clues as to what Wittgenstein thought of ethical judgements are
contained in his remarks about religion. It was Wittgenstein himself who admit-
ted that there is an analogy between ethical and religious judgements. He admit-
ted, for instance in A Lecture on Ethics, that all experiences of the absolute values
have their counterparts in the religious language. The wonder at the existence of
the world is a counterpart of a religious belief that the world was created by the
omnipotent and loving Being; the feeling of being safe whatever the case and the
feeling that nothing that happens in the world can hurt me finds its counterpart
in the religious belief that God protects me, that one is saved by the grace of God,
and that salvation of the soul is what truly matters, and, finally, the feeling of being
guilty corresponds to the religious belief that some of our deeds are considered by
God as sinful and that God sometimes condemns one’s conduct496. If, in the eyes
of Wittgenstein, there is a close similarity between religious and ethical utterances,
then we can suspect that what Wittgenstein said about religion also sheds light on
his views on ethics.
It was Leo Tolstoy’s The Gospel in Brief that shaped Wittgenstein’s thinking
about religion. Wittgenstein was probably fascinated by Tolstoy’s focus on Jesus’

496 “This allegory also describes the experience which I have just referred to. For the first
of them is, I believe, exactly what people were referring to when they said that God
had created the world; and the experience of absolute safety has been described by
saying that we feel safe in the hands of God. A third experience of the same kind is
that of feeling guilty and again this was described by the phrase that God disapproves
of our conduct” (LE, p. 9–10).

235
moral preaching and his lack of interest in Jesus’ miracles or in the theologi-
cal disputes about the divine and the human nature of Jesus497 which definitely
helped such a naturalistic-oriented mind as Ludwig Wittgenstein to accept the
evangelical teachings498. One can easily recognise the influence of Tolstoy on the
Tractatus. Tolstoy tried to find a solution to the problem of life499, which was also
Wittgenstein’s concern. After Tolstoy, Wittgenstein understood the problem of
life as a question whether life has a meaning500. Finally, in the thought of both
of them we find a modernist anxiety that scientific progress somehow deprives
things of their meanings501. The topic of Tolstoy’s influence on Wittgenstein is
interesting on its own, but with respect to the subject of this chapter I shall con-
centrate on two features of Tolstoy’s views on religion. First, for Tolstoy, Chris-
tianity was not a doctrine. It describes nothing objective; it is not a competition
for a scientific worldview. What it offers is not another standpoint of what the
world looks like but making one’s life meaningful502. One can sum this up in the
following way: a sentence of a religious language p does not inform us about
facts in the world but about the way a person interprets those facts. One who
believes in God interprets facts in such a way that one sees one’s life as having
meaning. Conversely, an atheistic profession of a lack of faith amounts not to
(the question if it is a justified view on atheism is secondary) denying that a

497 “I sought a solution to a problem of life, and not of a theological or historical ques-
tion; and that is why I was indifferent to know whether Jesus Christ is or is not God,
and from whom proceeds the Holy Spirit” (Tolstoy 1896, p. 10).
498 Obviously, the circumstances of time also played a huge role in Wittgenstein’s fascina-
tion. He bought a copy of the book in Tarnów during the time he was feeling depressed
by the company of his evil and malicious fellow soldiers. Ray Monk even claims that
reading the book saved Wittgenstein from committing suicide (Monk 1991, p. 115).
499 Tolstoy 1896, p. 10.
500 “I sought to find the sources of light. And I found them in the Gospels (…). And when
I reached this source of light I was dazzled with its splendour, and I found there full
answers to my questions as to the purport of the lives of myself and others” (ibid., p. 9).
On the other hand, Wittgenstein writes: “I know that this world exists (…). Something
about it is problematic, which we call its meaning” (NB 11.6.16, p. 72–73).
501 Tolstoy: “After getting the reply that I was a fortuitous concatenation of atoms, and
that my life was void of purport (…) I became desperate and wished to put an end
to my life” (Tolstoy 1896, p. 8). Wittgenstein, on the other hand: “As a thing among
things, each thing is equally insignificant (…) A stone, the body of a beast, the body
of a man, my body, all stand on the same level” (NB 8.10.16, p. 83; NB 12.10.16, p. 84).
502 “The great mass of men who hold to faith and are uncorrupted by wealth, possess
the meaning of life” (Tolstoy 1896, p. 8).

236
certain object exists, but expresses a rather depressing attitude to one’s life as
being void of meaning.
Secondly, according to Tolstoy, faith is something subjective not only because
it expresses individual attitudes, but also because it is accepted not on the basis of
reason, evidence or arguments. He was aware that he could not provide any argu-
ments advancing his own religious point of view. What, in his opinion, could con-
vince his reader to accept his interpretation of the Gospel were the features of his
disquisition: unity, fullness, and its harmony with the inner feelings of his readers:
The justness of a conception of this kind is better proved, not by arguing particular
points, but by its own unity, clearness, fullness, as well as by its harmony with the inner
feeling of all who seek truth (Tolstoy 1896, p. 6).

Wittgenstein saw religion in a similar way. Also, for this Austrian philosopher
Christianity was not a matter of doctrine:
Christianity is not a doctrine, not, I mean, a theory about what has happened & will
happen to the human soul (CV, p. 32).

He analysed religious utterances as expressing attitudes to the world:

To believe in God means to see that life has a meaning (NB 8.7.16, p. 74).

The sentence: “God exists” is, in other words, a proclamation of someone’s evalu-
ation of life as meaningful. The second aspect of the subjectivity of religious be-
liefs – the fact that one cannot justify religious convictions on the basis of reason
or argue in favour of them – is also visible in Wittgenstein’s thinking:
And faith is faith in what my heart, my soul, needs, not my speculative intellect. For my
soul, with its passions, as it were with its flesh & blood, must be redeemed, not my ab-
stract mind. Perhaps one may say: Only love can believe the Resurrection. Or: it is love
that believes the Resurrection (CV, p. 38–39)503.

For Wittgenstein, faith is a matter of heart, which I understand as a claim that faith
is something felt or chosen by an individual, not something which is deduced by
intellect. Religious tenets have meaning, but only as expressing my attitudes to
the world of empirical facts which, of course, is a confusing claim for someone –
indifferently if for a believer or for an atheist – who takes religious utterances as

503 C. f.: “Religion says: Do this!–Think like that! but it cannot justify this and it only need
try to do so to become repugnant; since for every reason it gives, there is a cogent
counter-reason. It is more convincing to say: ‘Think like this! – however strange it
may seem or: ‘Won’t you do this? – repugnant as it is’ ” (CV, p. 34).

237
describing a supernatural reality504. For such a person an utterance about the Last
Judgement is an utterance (true for a believer and false for an atheist) about a future
event. In contrast to this objectivist view on religious beliefs, Wittgenstein claims:
A historical proof is irrelevant to belief. This message (the Gospels) is seized on by a
human being believingly (i.e. lovingly): That is the certainty of this “taking-for-true”,
nothing else. The believer’s relation to these messages is neither a relation to historical
truth (probability) nor yet that to a doctrine consisting of “truths of reason” (CV, p. 38).

In other words, for Wittgenstein, if one holds some religious beliefs, then one does
so not on the basis of some rational arguments. The utterances of believers inform
us of how they see the world, but not about what belongs, according to them, to the
region of facts. Wittgenstein also admits that such a view on faith makes it incom-
prehensible and, hence, nonsensical for non-believers. It is because one simply has
a religious attitude to the world or not; one feels safe and under divine protection
or not. There is nothing here one could learn or argue in favour of.
If the analogy between Wittgenstein’s thinking about religion and ethics holds,
then in his remarks on Christianity we would find another confirmation of his
subjectivism with respect to ethics. According to this argument, in Wittgenstein’s
view, ethical judgements express people’s attitudes to the world (thesis (A)). More-
over, ethical attitudes to the world are not justified by facts. There is nothing with
respect to facts that confirms or rejects the standpoint of a moral nihilist as well
as the standpoint of somebody who evaluates certain deeds as morally right or
wrong (thesis (B)).
In the subsequent paragraphs I want to compare my subjectivist interpreta-
tion with the commentaries of others. I especially want to analyse the differences
between what I call Wittgenstein’s subjectivism and Russell’s early views on eth-
ics, which are usually labelled emotivism. Before that, however, I would like to
analyse how my interpretation of Wittgenstein’s views on ethics could illuminate
the conclusion of the Tractatus, i.e. the Tractarian incentive for silence with re-
spect to things about which we cannot speak (TLP 7) and the claim that one
cannot express ethics (TLP 6.42-6.421).

5.2.2 Ethical subjectivism and Tractarian silence

My thesis reads as follows: the assumption that for Wittgenstein ethical judge-
ments are subjective (in the mentioned meaning that ethical sentences express

504 For instance, Anscombe notices: “The truth of the Tractatus theory would be death
to natural theology” (Anscombe 1965, p. 78).

238
somebody’s attitudes and that they are not justified by facts) explains why he
counsels silence with respect to things one cannot meaningfully express. This is
because one of the consequences of my interpretation is the futility of express-
ing ethics. If one does not confer any content on ethical sentences, then why
would anyone bother to engage in discussions about the moral values of some
deeds? Of course, the classical proponents of emotivism, such as Ayer and Ste-
venson, defended the practice of formulating moral judgements because they
thought that there is always a chance to persuade an interlocutor, to incline
him or her to change his or her views505. An adversary of the emotivists could
respond that what inclines us to adopt somebody else’s moral views is the truth
of these views. Therefore, the accusation claims that, according to the emotiv-
ists, there is nothing in moral judgements that could attract other people to
change their minds. The question: “Why bother to engage in moral conflicts or
discussions? Why should I bother about somebody’s subjective evaluations?”
still holds. On the other hand, Wittgenstein admits that if ethical sentences are
neither true or false, if they convey no content, if they simply express subjective
attitudes and are not justified by facts (one cannot therefore argue rationally
in favour of them or against them), then there is truly no point in engaging in
moral disputes. Hence, one should keep silent with respect to his or her moral
convictions:
Is not this the reason why those who have found after a long period of doubt that the
sense of life became clear to them have then been unable to say what constituted that
sense? (TLP 6.521)

Let us suppose that somebody discovered what the sense of life consists in. For
Wittgenstein, this is exactly the kind of discovery that is accomplished by the heart
and not by intellect. There is no surprise, then, that one cannot put it into words
and substantiate it rationally. One can, of course, express sentences such as: “I feel
safe even in the face of death”, but because this does not express anything objec-
tive, one cannot point at any facts that could support or explain one’s view. There
is nothing in the aforementioned sentence with which another person could agree
or disagree. The only thing to say is perhaps: “I feel the same way” or “I feel differ-

505 According to Stevenson, the method of settling ethical disagreements “is persuasive,
not empirical or rational; but that is no reason for neglecting it” (Stevenson 1963/1975,
p. 29). He also writes: “Ethical terms are instruments used in the complicated interplay
and readjustment of human interests (…) Being suited for use in suggestion, they are
a means by which men’s attitudes may be led this way or that” (ibid., pp. 17–18).

239
ently”. Wittgenstein fully embraces the conclusion of the meaninglessness of ethi-
cal judgements, and this is why in the last thesis of the Tractatus he states:
What we cannot speak about we must pass over in silence (TLP 7).

Wittgenstein notices the need to express moral or aesthetic judgements:

This running against the walls of our cage is perfectly, absolutely hopeless. Ethics so far
as it springs from the desire to say something about the ultimate meaning of life (…)
can be no science. What it says does not add to our knowledge in any sense (LE, p. 12)

But, at the same time he acknowledges that when one tries to fulfil this desire to
communicate one’s attitudes with respect to the meaning of life or moral evalua-
tions of certain facts, one fails. It is then better to be silent about them.
Although ethical expressions are examples of nonsense, we find in the Tracta-
tus the following thesis:
There are, indeed, things that cannot be put into words. They make themselves manifest.
They are what is mystical (TLP 6.522).

Before I proceed in my interpretation, I should warn that the things I am about to

say in the following sentences are a matter of speculation. Wittgenstein himself was
not clear on these topics. But, on the basis of the aforementioned thesis, I incline
to think that Wittgenstein indeed thought argumentation as futile with respect to
ethics. I am also inclined to think that despite the fact that he saw a way of “con-
vincing” others to accept certain moral standpoints; only this time not by means
of rational arguments, and definitely not by means of propaganda, but by means of
an example. It is an example of life, a fascination with somebody’s conduct, a desire
to be like an admired hero which will convince people to change their moral views.
We know that Wittgenstein was deeply fascinated by the Gospel (through
Tolstoy’s version) during the time he was finishing the Tractatus, and there is
indeed something from the evangelical: “Come and see”506 in his description of
Christianity:
Amongst other things Christianity says, I believe, that sound doctrines are all useless. That
you have to change your life. (Or the direction of your life.) That all wisdom is cold; & that
you can no more use it for setting your life to rights, than you can forge iron when it is cold.
For a sound doctrine need not seize you; you can follow it, like a doctor’s prescription.- But
here you have to be seized & turned around by something.- (I.e. this is how I understand
it.) Once turned round, you must stay turned round. Wisdom is passionless (CV, p. 61).

506 “Nathanael said to him, ‘From Nazareth? Can anything good come from that place?’
Philip replied, ‘Come and see’ ”. John 1, 46.

240
and in his description of ethics (that one finds in Wittgenstein’s conversation with
Feigl):
It was necessary to get rid of metaphysics in order to make room for den Ernst des Lebens,
the seriousness of life. Deep commitment was required, not rational justification, and
whoever had the former would not worry his head about the existence of the world or the
ultimate substance of it (cited in: McGuinness 2002, p. 191)

If it is true that for the moral cognitivists one needs to show the truth of some
moral judgements in order to incline others to change their moral standpoints,
and if it is true that for the emotivists, such as Russell or Stevenson, one needs
rhetoric skills, then for Wittgenstein one needs an example of the saints.
Summing up, in the previous sections I tried to show the consequences of the
thesis that expressions of ethics are nonsensical. I ascribed subjectivism to early
Wittgenstein (section 5.2.1). Moral judgements, in this view, express nothing more
than the attitude of the person who formulates those judgements. As such, these
utterances are subjective – one cannot argue in favour of them rationally. There-
fore, as I noticed in section 5.2.2, Wittgenstein, at the end of the Tractatus, counsels
silence. If I am correct in my reading of the early writings of Wittgenstein, one can
show what solutions to the problems of life consist in only by the example of life.
The subjectivist hypothesis, therefore, fits nicely in the system of the Tractatus; it
respects its final conclusion and the importance of the showing/saying distinction.

5.2.3  Wittgenstein’s subjectivism and early Russell’s emotivism

At the end of section 5.2 I would like to make my interpretation of early Witt-
genstein’s views on ethics clearer by comparing it to those of an author who was
a contemporary of Wittgenstein and to whom one very often ascribes emotiv-
ism, i.e. to the views of Bertrand Russell. Interestingly, one can notice a similar-
ity between the paths of both Russell and Wittgenstein that led them to their
ethical views. Russell was before World War I a famous logician and mathemati-
cian, and with respect to ethics he was a proponent of Moore’s intuitionism. He
believed in the objectivity of moral values such as goodness or badness. There
were horrifying events of the war that forced Russell to abandon the convic-
tion that moral values are obvious. Both sides of the barricade held as obvious
contradictory moral judgements – the Germans held killing the British to be
morally good and justified; the British, naturally, held killing the British to be
morally appalling but, on the other hand, maintained that killing the Germans

241
was morally justified507. Since it seemed that no arguments could convince peo-
ple to abandon their views, and since it seemed that in Europe there could be no
moral agreement on what is good and bad, then – concluded Russell – perhaps
the whole discussion was not about something objective? Perhaps the only thing
conveyed by both sides of ethical contention are emotions? Similarly, young
Wittgenstein, who before World War I was fervently enagaged in specialist dis-
cussions on logic, evolved spiritually because of his wartime experiences. If I
am correct that Wittgenstein was a proponent of subjectivism, then after the
war Russell and Wittgenstein – both of them now interested in practical phi-
losophy and both subjectivists – should have gotten along very well. We know,
however, that their reunion resulted in a disaster – they could not understand
one another. To Russell’s horror, Wittgenstein turned out to be some kind of a
mystic508. If I claim that Wittgenstein was an ethical subjectivist, then how can I
explain this feeling of alienation between these two philosophers? Obviously, in
my opinion, the similarities in their ethical views are superficial. It is true that
they both held moral judgements to be of no truth-value, that in the scientific
project of describing the world they did not see a place for values509, that they
would both agree that moral judgements express attitudes or feelings towards
the world taken as a whole510, and that neither of them contrived a theory of

507 “Russell began to understand that the good was not as obvious as Moore’s intuition-
ism made it out to be. Those around him, in England and all over the world, appealed
to similar values, similar rules, and made similar accusations against one another.
Each thought good was embodied in their own actions and causes, evil in the others.
The British regarded the Germans as wicked, and thought is good to slaughter them.
Germans felt likewise about the British” (Potter 2006, p. 12).
508 In a letter to Lady Ottoline he wrote about his first meeting with Wittgenstein after
the war: “I had felt in his book a flavour of mysticism, but was astonished when I
found that he was a complete mystic. He reads people like Kierkegaard and Angelus
Silesius, and he seriously contemplates becoming a monk” (Letters, p. 82).
509 “So too it is impossible for there to be propositions of ethics. Propositions can ex-
press nothing that is higher. It is clear that ethics cannot be put into words” (TLP
6.42-6.421). “The opposition between a philosophy guided by scientific method
and philosophy dominated by religious and ethical ideas may be illustrated by two
notions which are very prevalent in the works of philosophers, namely the notion
of the universe, and the notion of good and evil. A philosopher is expected to tell
us something about the nature of the universe as a whole, and to give grounds for
either optimism or pessimism. Both these expectations seem to me mistaken” (Rus-
sell 1914b, p. 57).
510 “I believe the best way of describing it is to say that when I have it [the experience of
the absolute good] I wonder at the existence of the world” (LE, p. 8); “What is valuable

242
meta-ethics (in the case of Russell at least not at this stage511) because their in-
terests were focused on solving particular ethical dilemmas rather than on the
theory of what is good512.
Despite these similarities, the difference between the attitude of both of these
philosophers to ethics is enormous. There are, for instance, different morals they
draw from their standpoints. As I said before, Wittgenstein embraces the con-
sequence of the meaninglessness of ethical judgements – the impossibility of
formulating one’s own ethical positions – so, at the end he counsels silence. Ac-
cording to my interpretation of Wittgenstein, one can argue in favour of an ethi-
cal judgement only by the example of a decent life. For Russell, silence was not an
option. He was in favour of moral activism even despite the fact that he thought
that moral judgements had no truth-value. In this view, even if a moral activist
has no rational arguments in his or her arsenal, he or she can still convince others
to share his or her views by means of rhetoric, propaganda or persuasion: “Russell
has no difficulty accepting the claim that, in a moral dispute, we become preach-
ers or proselytisers, using rhetoric and other persuasive methods to propagate
our attitudes”513.
Secondly, I would point to the different aims lying behind these two ethical
views. For Wittgenstein, saying that moral judgements are nonsensical is also a
way of underlining the difference between ethical and other values, i.e. the differ-
ence between absolute and relative values.
Now when this is urged against me I at once see clearly (…) not only that no description
that I can think of would do to describe what I mean by absolute value, but that I would
reject every significant description that anybody could possibly suggest, ab initio, on

is the indication of some new way of feeling towards life and the world, some way
of feeling by which our own existence can acquire more of the characteristics which
we deeply desire” (Russell 1914b, p. 64).
511 I compare Wittgenstein’s view with the early form of Russell’s emotivism expressed
in On Scientific Method in Philosophy (1914).
512 According to Rhees, in one of his private conversations with Wittgenstein his men-
tor expressed astonishment with why books on ethics do not contain “genuine or
moral problems” (Rhees 1965, p. 21), where moral problems are those which have at
least two probable solutions, such as, for instance, a dilemma that a scientist could
face: “Should I sacrifice my family and continue my research on cancer or should I
sacrifice the research and help my wife who is terminally ill?” (ibid., p. 23). On the
other hand, with respect to Russell, Potter notices: “The fact that Russell spends most
of his time at this stage applying his theory might explain why he spent so little time
explaining it” (Potter 2006, p. 73).
513 Potter 2006, p. 71.

243
 the ground of its significance. That is to say: I see now that these nonsensical expres-
sions were not nonsensical because I had not yet found the correct expressions, but that
their nonsensicality was their very essence. For all I wanted to do with them was just to
go beyond the world and that is to say beyond significant language (LE, p. 11).

In the eyes of Wittgenstein, what is distinct by moral values, and what makes
moral experiences absolute (whatever the progress of science will be), is that one
who evaluates something as morally good is bound to change one’s attitude and
behaviour. This feeling of obligation to change one’s behaviour is a part of moral
evaluation. Opposite to this project, Russell’s conclusion that moral truths are not
objective clearly aims to fight the idea that one, after having evaluated something
as morally good, should necessarily feel inclined to change one’s attitudes. Accord-
ing to Potter, before the World War I Russell had “a growing conviction that belief
in absolute value leads people to become hateful and intolerant”514. Up to this day
there are people who are ready to kill others just because, in their opinion, they
hold a wrong conception of God. On the contrary, although a discussion about,
for instance, the superiority of Chopin’s music over List’s could be fervent, it is
hardly imaginable that one could be ready to die in defence of Chopin’s Mazurkas.
It seems that disagreements in tastes are violent to a lesser extent than discussions
about truth, and as far as I understand Russell, his saying that moral judgements
express only our evaluations aims to reduce the level of hatred and intolerance. It
would, however, mean that, in contrast to Wittgenstein who wanted to emphasise
the unique character of moral judgements, Russell tried to present moral judge-
ments merely as another kind of opinion. The difference does not consist in the
fact that one of them believed in relativism and the other did not. In fact, both of
them rejected the connection between moral judgements and truth-values. The
difference lies in the role played by the fact that moral judgements express subjec-
tive evaluations. Wittgenstein, who formed his views in the trenches of the World
War I battles and who despite witnessing evil and maliciousness tried to preserve
his own decency and courage, insisted on the fact that others’ behaviour does not
matter. Still, if I evaluate an action or a conduct as morally good, it obliges me
to behave in a certain way. In ethics he was, so to speak, addressing himself and,
therefore, he insisted on the binding character of moral values (or, using his vo-
cabulary, on the absolute character of moral values). On the other hand, Russell,
who also formed his early emotivism during the World War I, and who, as a civil-
ian, experienced particularly badly the evil and maliciousness of public opinion,
was addressing mainly the so-called ordinary people. Having experienced that

514 Ibid., p. 10.

244
being blindly bound to one’s own moral evaluations can have a disastrous effect,
he postulated a greater distance to our own subjective convictions. One way to
achieve this is to stop ascribing a distinctiveness to moral judgements and to start
seeing them as simply mere tastes in the same way we usually treat musical tastes.
This biographical interjection could also explain the third difference between
Russell and Wittgenstein. According to Russell, our moral instincts are a result
of our being social animals:
Ethics is essentially a product of the gregarious instinct, that is to say, of the instinct to
cooperate with those who are to form our own group against those who belong to other
groups. Those who belong to our own group are good; those who belong to hostile groups
are wicked (Russell 1914b, p. 63).

Ethics, in his view, is an effort to present our own herd as justified515. These
thoughts would be extremely alien to Wittgenstein, as he never thought of eth-
ics as directed against some group of people. Ethics by no means is a result of a
gregarious instinct; it is a result of our personal investigations into the meaning
of life. It is not society which tells us what the meaning of life consists in but it is
the task of an individual to discover or recognise what his or her purpose in life
is. In other words, whereas Russell presents a rather pessimistic view according
to which values are held by people as a result of their conformism and an ability
to adapt to the requirements of the group, Wittgenstein sees people more opti-
mistically. In moral evaluations he saw a way to defend one’s independence from
the influence of the social environment or even from happenings in the world. If
I were forced to express this difference in the form of a theory, I would say that
whereas for Russell the moral judgement: “X is good” expresses the attitude of
the whole group to which a speaker of this judgement belongs, then for Wittgen-
stein the same moral judgement expresses just the attitude of a given speaker.
This particular difference between Russell and Wittgenstein reveals an impor-
tant feature of Wittgenstein’s views on ethics. In the aforementioned Gospel in
Brief, Tolstoy interprets the temptation of Christ in the desert as a metaphor for a
struggle between a body (which represents the empirical world) and a soul: “Then
temptation ceased, and Jesus knew the power of spirit”, concludes Tolstoy516. This
motif of the power of spirit – the internal core of a human being consisting in his
or her most intimate and important beliefs – which can rest intact against the

515 “When the animal has arrived at the dignity of the metaphysician, it invents ethics
as the embodiment of its belief in the justice of its own herd” (Russell 1914b, p. 63).
516 Tolstoy 1896, p. 27.

245
influence of the external world (all adversities of fate, hostile people, etc.) or even
in some cases master it, is a frequent motif in Wittgenstein’s writing:
How can man be happy at all, since he cannot ward off the misery of this world? (…) The
life of knowledge is the life that is happy in spite of the misery of the world. The only life
that is happy is the life that can renounce the amenities of the world (NB 13.8.16, p. 81)
I can only make myself independent of the world – and so in certain sense master it – by
renouncing any influence on happenings (NB 11.6.16, p. 73)
I can make myself independent of fate (NB 8.7.16, p. 74)

Up to this moment when I was talking about the subjectivism of early Wittgen-
stein, I had in mind his conviction that (A) ethical sentences express the attitudes
of their speakers and that (B) there cannot be arguments based on facts in favour
of or against moral judgements. This comparison of Wittgenstein’s thought with
Russell’s emotivism reveals that there is also another meaning of Wittgenstein’s
subjectivism, which I describe as his “subjective turn”. In this meaning, subjectiv-
ism refers to the fact that the source of the meaning of life and of happiness is in
us. It is the claim of the radical dichotomy between facts and values. Whatever the
case, in Wittgenstein’s view it is irrelevant to us feeling happy or unhappy, if we see
our lives as meaningful or not:
If the good or bad exercise of the will does alter the world, it can alter only the limits of
the world, not the facts—not what can be expressed by means of language (TLP 6.43).

One could juxtapose his position with a view which would welcome the suppo-
sition that, for example, the theory of evolution could scientifically explain why
people evaluate facts and events the way they do. When Wittgenstein claims that
our attitudes have nothing to do with facts, he is also referring to the facts from
the natural history of people. This feature of Wittgenstein’s thinking adds an air
of mysticism to his philosophy – one speaks here about the power of spirit, an
internal core, a source of meaning and happiness – and probably this element of
Wittgenstein’s conversion worked so discouragingly for Bertrand Russell. Hence,
although there are superficial similarities between his and his pupil’s ethics, then
in the details these standpoints are so distant from one another that the opinion
that Wittgenstein’s subjectivism is of a very original kind seems to be justified.

5.3 Solipsistic theses of the Tractatus under a

subjectivist reading
In the previous sections I presented what in my opinion was early Wittgenstein’s
view on ethics. I defended the thesis that on the basis of the mystical part of the

246
Tractatus, A Lecture on Ethics and Wittgenstein’s views on religion, one could
ascribe subjectivism to his early ethics. The sentence: “X is good”, according to
such an interpretation, expresses the attitude of the speaker. The second claim
of my interpretation was that moral evaluations are independent from the facts.
Because of this, one can always prove the nonsensicality of ethical judgements.
In my opinion, this was the reason for which Wittgenstein advised silence with
respect to what truly matters at the end of the Tractatus. Finally, I pointed to one
of the consequences of Wittgenstein’s division between facts and values. If moral
attitudes and valuations are independent from the facts, then they are held just
upon the strength of the individual’s personal decision.
I hope that we are now in a position to go back to the initial questions of the
chapter: how does someone who sees the Tractatus as defending ontological ma-
terialism and as being reluctant to theoretical philosophy understands the Trac-
tarian theses that (a) what solipsism means is quite correct, (b) the world is my
world, and (c) that the self enters into philosophical considerations as the limit of
the world?
I have already pointed out in Chapter 4 that these theses are similar to the
ideas contained in The World as Will and Representation, and that a scrutiny of
Schopenhauer’s text suggests that the aforementioned problematic theses should
not be read as ontological or epistemological ones, but rather as existential and
sapiential truths. If we read the solipsistic theses of the Tractatus this way, then
subjectivism can indeed account for their rightness.
1. The world is my world. In the Notebooks Wittgenstein writes:
What has the history to do with me? Mine is the first and only world (NB 2.9.16, p. 82)

The world consists of the totality of what is the case. We are all living in a world
where Spain won the world championship in football in 2010, where Henry
Dunant founded the Red Cross, and where Jorge Mario Bergoglio was elected
pope on 13 March 2013. The subjective interpretation does not deny this; it states
only that from an individual point of view besides the facts there are also interpre-
tations and evaluations of these facts which, using the Tractarian metaphor, “wax
and wane” the world as a whole and make the world different for different people.
Because Wittgenstein’s position in ethics assumes that these interpretations are
personal and subjective, then the world seen from an individual point of view will
always be something specific – it will be “my world”, as Wittgenstein describes it.
The subjective interpretation in contrast to the solipsistic one does not have to as-
sume the existence of non-natural beings, such as the self. Likewise, the problem
of the belief in the existence of other subjects does not even arise.

247
2. What solipsism means is quite correct. Two attitudes dominate in the litera-
ture with respect to the solipsistic theses of the Tractatus: the first takes the
Tractarian dictum at face value. Wittgenstein was truly a solipsist in the mean-
ing ascribed to this position by philosophical dictionaries517, but he could not,
according to the Tractarian theory of meaning, express this position meaning-
fully (just as Christians who claim that the dogma about the Trinity is true but
every expression of its truth leads to nonsensical and paradoxical statements).
The second group of commentaries acknowledges the fact that Wittgenstein
did not take solipsism at face value. It assumes that in the eyes of early Wittgen-
stein the importance of solipsism lies not in its direct formulations but in what
solipsism intends to express. We have seen an example of a religious believer
who thinks that the sentence: “God exists” describes the world, whereas, in fact,
in Wittgenstein’s opinion it expresses the believer’s hope that life has meaning.
Why should we not think that Wittgenstein had the same attitude to philo-
sophical theories? He treated the great works on metaphysics and theology with
reverence as books which contain important insights, but, as was made clear
on the example of religious utterances, in his view their authors were mistaken
when it comes to the aims of their judgements. This is an attitude according to
which great thinkers of the past were consciously stating ontological and epis-
temological truths but, in fact, unconsciously they were expressing insights of a
different kind. This could be the case with solipsism. What it shows is correct,
but only if one does not take solipsism as describing the world or as stating an
ontological truth. What solipsism states “consciously” is the claim that there is
only one mind in the world. The “unconscious” truth of solipsism would be an
expression of the helpless loneliness of somebody trying to justify his or her
most precious and intimate convictions. According to Wittgenstein’s subjectiv-
ism, one using only rational arguments cannot convince or persuade others to
accept one’s moral point of view. We are forced, in this respect, into silence. We
are closed in our own worldviews, unable to communicate things which are im-
portant for us. In this sense a solipsist is right: in our worlds we are alone. And
again, the fact that in the eyes of the author of the Tractatus solipsism is correct
does not mean that he doubted in the existence of other people518. It expresses

517 “Wittgenstein seems to be committed to the strongly idealist view that the nature of
the world depends upon me” (Morris 2008, p. 297).
518 On the contrary: “Scepticism is not irrefutable, but obviously nonsensical, when it
tries to raise doubts where no questions can be asked. For doubt can exist only where
a question exists, a question only where an answer exists, and an answer only where
something can be said.” (TLP 6.51).

248
 a despairing solitude which is a result of the fact that we are forced to be silent
with respect to what truly matters to us – with respect to the problems of life519.
Secondly, solipsism taken at face value states that everything is dependent in
its existence on me; “unconsciously”, however, solipsism would express the fact
that it is up to us what meaning we ascribe to different things in the world and
what values are the most important for us. And this is why, under the subjectiv-
ist interpretation, solipsism is, for Wittgenstein, quite correct.
3. The subject is the limit of the world. Wittgenstein, under the influence of Tol-
stoy’s book, regards as ideal a situation when we are independent in our atti-
tudes from the influence of the world. It is possible to feel happy even in the face
of death. It is the subject through which things acquire meaning. Its gaze makes
things precious (CV, p. 3).
On the other hand, “the world of the happy man is a different one from that of the
unhappy man” (TLP 6.43). Metaphorically speaking, these worlds have different
limits but not different contents: the same content is simply differently waxed. The
content of the world stands in this metaphor for the facts, and waxing of the world
stands for evaluating and interpreting. What is the meaning of the expression: “the
limit of the world”, then? In my opinion it stands for subject. That is why in TLP
5.632 we read that the subject is a limit of the world, and in TLP 5.641 we read that
the non-psychological talk of the self is connected to the fact that the self is a limit
of the world, and not its part.
Again, one does not have to read these formulations as the ontological claim
stating the existence of an extra-mundane object. The interpretation which sees
Wittgenstein’s theses as existential claims that in the identification of the self with
the limit of the world, Wittgenstein expresses the evaluating activity of subjects.
There are subjects that colour the world that is otherwise brute and indifferent. It
is a matter of my individual attitude if life has meaning for me. It does not assume
anything non-natural.
The idea that Tractarian remarks about the correctness of solipsism should not
be taken at face value is not new. This approach is represented by, among others,

519 Ernest Gellner was the author who first indicated the solitude of a person as the
consequence of Wittgenstein’s standpoint: “The other and more conventional path
to despair is by solitary confinement, imposed on the investigative self as it exam-
ines its data to see how far they will allow one to proceed, only to and that the data.
themselves are the limit of the world, and that self and world are coextensive and
identical. This theme is also conspicuously present in the Tractatus. It too guarantees
total solitude, emptiness and despair” (Gellner 2004, pp. 60–61).

249
Michael Kremer520. However, his interpretation of solipsism is quite different from
my reading. First of all, he thinks that the Tractatus offers the reader a precise
“path for life, a way to be followed”521, whereas in my view, in which Wittgenstein
held ethical judgements to be substantially nonsensical, one could not put into
words this kind of moral instruction even if one wanted to. I indicated earlier that
the consequence of Wittgenstein’s standpoint with respect to the problems of life
was ultimately silence (“What we cannot speak about we must pass over in silence”
(TLP 7)). In Kremer’s view, “at the end of the Tractatus, we are left with ourselves.
We are left, in particular with Ludwig Wittgenstein, the author of this book, and
his readers – human beings, capable of communicating with, and understanding,
one another”522. In my opinion, this conclusion is in sharp disagreement with the
Tractatus which underlines, among others, that those who have discovered the
sense of life are unable to convey what this sense consists in523. Secondly, I think
that solipsism by early Wittgenstein points at the solitude of a person in some rel-
evant aspects of his or her life. Kremer, on the other hand, believes that the solip-
sistic stance amounts to spiritual counsel: “Act as if you are alone in the world”524.
Again, I cannot agree with this interpretation, not only because it sounds slightly
arbitrary, but mostly because I think, just like Frascolla, that for Wittgenstein there
was no doctrine solving the riddle of life525. It is true that upon my reading Witt-
genstein presents a view according to which happiness is independent of fate or
events in the world, but I also claim that, according to Wittgenstein, how one un-
derstands happiness is an individual matter. On Kremer’s reading there is a very
clear answer to the problem of life; and this answer consists in self-abandonment:

520 “I would prefer to say that solipsism is an intellectual, moral and mystical exercise
aimed at bringing about a change in one’s spiritual life” (Kremer 2004, p. 59). Andrea
Birk sees, on the other hand, in Tractarian solipsism an invitation to vita contempla-
tiva – a contemplation of the logical structure of the world (Birk 2006, p. 80–95).
521 “ ‘Truth’ as something which one can do is not something we might be tempted to
think of as expressible in a proposition. It is, rather, a way to be followed a ‘path’ for
life. Insofar as the Tractatus communicates a ‘truth’ it is by demonstrating to us, in
practice, how to follow such a path” (Kremer 2004, p. 63).
522 Ibid., p. 78.
523 “Is not this the reason why those who have found after a long period of doubt that
the sense of life became clear to them have then been unable to say what constituted
that sense?” (TLP 6.521).
524 Kremer 2004, p. 68.
525 “There is no doctrine that can explain what the solution of the so-called riddle of life
is” (Frascolla 2007, p. 213).

250
 In contrast, I will argue that the ultimate point of the Tractatus’ discussion of solipsism is
left behind. This rejection of all forms of self-assertion, including the self-assertion found
in certain misguided forms of asceticism, piety and false humility, is what the Tractatus
really aims at (Kremer 2004, p. 60).

In my opinion, the original sin of Kremer’s reading, i.e. not taking into account
that Wittgenstein ends the Tractatus with the counsel of silence, results in what I
would describe as the latitude of his interpretation. Although I agree with Kremer
that the solipsistic theses of the Tractatus should not be regarded as ontological
ones, I do not think that Wittgenstein, by means of these theses, wanted to propose
a concrete model of conduct. Even if he had an idea as to what a decent life consists
of, it is the Tractatus itself which states that such ideas are incommunicable.

Summary
I tried to show in this chapter how the proponents of the naturalistic reading of
the Tractatus can read its solipsistic theses. The presence of these theses has caused
many scholars to ascribe to early Wittgenstein solipsistic transcendentalism526, so it
is a vital point of my interpretation to reconcile them with the conviction that the
worldview of the Tractatus was materialistic. I hope that I have achieved this goal by
interpreting the mystical part of the Tractatus as betraying Wittgenstein’s subjectiv-
ism in ethics. According to my reading, ethical sentences in the Tractatus express
individual and subjective attitudes to the facts in the world: it is up to me if I see the
world as created by God, or if I see some events in the world as miraculous or not,
or if the goal of my life is wealth or winning the Nobel Peace Prize. In the Notebooks,
directly after the solipsistic remark that I am my world, Wittgenstein noted:
Things acquire “significance” only through their relation to my will (NB 15.10.16, p. 84)

This claim, however, is not an expression of ontological or epistemological solip-

sism but of, so to speak, existential solipsism expressing the inability to convey our
convictions on the meaning of life. It seems to me that this solution allows, on the
one hand, to retain the naturalistic reading of the Tractatus and, at the same time,
it does justice to Wittgenstein’s wartime conversion resulting in his interest in the
problems of life.
The last and at the same time the most basic question in the dissertation reads
as follows: how did Wittgenstein reconcile the values of the human world with
the scientific worldview? In the conclusions I shall see if the solutions discussed
in this chapter bring us any closer to the answers.

526 Zemach 1973, Engelmann 1967, Kannisto 1986, Stern 1995, Atkinson 2009.

251
Final Thoughts. The Defence of Human Values
by early Wittgenstein

On the last pages of this dissertation I shall try to summarise its results. I wrote
this work because I was interested in the question as to how Wittgenstein, who
fully accepted the scientific worldview, solved the problem of ethics, i.e. the group
of issues concerning moral goodness or evil, the meaning of life or aesthetic
beauty. If, according to the naturalistic point of view, only physical particles ex-
ist – and by that Wittgenstein grants correctness to a proponent of the most far-
reaching scientism – then how can we talk about the values of the human world?
Wittgenstein’s answer seems to me to be interesting exactly because he does not
resort to shortcuts consisting of saying that although natural science has made
progress, science is somehow limited, and that there are other, equally valuable
points of view on reality. He assumes that it is basically possible that a set of true
utterances about the world is tantamount to the totality of propositions of sci-
ence. He asks from this standpoint: what is the role of philosophy? Is there a place
in such a worldview for our strong intuitions that some of our deeds are right or
wrong; or that some things or events in our lives present themselves as valuable?
In my opinion, those interpretations of the Tractatus which emphasise its nat-
uralism do not dedicate much attention to the so-called solipsistic and mystical
parts of the Tractatus, or even if they do, it seems that their authors are slightly per-
plexed by the presence of the mystical theses in Wittgenstein’s early book. Russell,
in his famous Introduction to the Tractatus, admitted that the mystical part left him
“with a certain sense of intellectual discomfort”, because although Wittgenstein as-
sures that one cannot talk about ethics, “nevertheless, he is capable of conveying his
ethical opinions”527. James Griffin, one of the first commentators who interpreted
the simple objects as physical atoms, passes over the mystical part or solipsism of
the Tractatus in silence. Timm Lampert, who recently renewed the physicalistic in-
terpretation of the Tractatus, claims that the concept of my world indicates merely
the fact that I experience the world from the first-person perspective528, but he does
not provide any convincing reason why such a trivial remark should account for
the label of solipsism?

527 Russell 1922, p. XXIV.

528 “Der Ausdruck ‘mein Welt’ bringt zum Ausdruck, dass ich einen bestimmten Stand-
punkt in der Welt habe, der Ausdruck ‘die Welt’ nicht” (Lampert 2000, p. 309).

253
 On the other hand, interpretations which try to do justice to the mystical and
solipsistic theses do so very often at the expense of taking Wittgenstein’s natural-
ism seriously. One group of these commentaries – the transcendental ones – claims
that when Wittgenstein was saying that science describes reality completely, he had
in mind an empirical reality which does not exclude his acceptance of the existence
of a higher order of things529. The second group is represented by McGuinness,
who rightly notices that, according to Wittgenstein, religion and ethics are not
fact-stating discourses or that propositions of science distort the nature of ethics530,
but because he asserts at the same time that Tractarian simples are “the truth-value
potential of a certain expression”531, then he finds an exclusion of ethical utterances
from the set of meaningful expressions “somewhat arbitrary”532. What is common
to both of these strategies is denying that early Wittgenstein was a naturalist. But
if he was not, then why did he propagate methodological scientism in TLP 4.11?
Why did he say that ethical expressions necessarily cross the limits of language?
And if someone answered that because these expressions try to convey neces-
sary truths, then the question reads as follows: why then did Wittgenstein identify
true meaningful propositions with contingent truths?533 If he truly believed in the
existence of the transcendental self, and he truly had some arguments in favour
of this thesis, then why did he claim that philosophy has merely two functions:
to repeat what scientists say and to guard the limits of meaningful expressions
(TLP 4.114)? As we saw in the work, interpretations which do not accept the thesis

529 “This is his one and ever-recurring thought: that the higher sphere, values, God do
not form part of the contents of the world, are not something within the world, to
be found in it and proved to exist; but are something manifested by the world seen
from the outside” (Engelmann 1967, p. 98).
530 “The real point of isolating so clearly the nature of science turns out to be that any
attempt at a scientific approach to human problems – moral, social, religious, or exis-
tential – involves a misconception of these problems (…). It is perhaps, a further point
that problems in ethics, for example, are not to be dealt with like ones in science, using
hypotheses, generalizations, thought-experiments” (McGuinness 2002, p. 128).
531 Ibid., p. 87.
532 Ibid., p. 78.
533 Instructive in this respect is Flanagan’s interpretation which states that Wittgenstein
was a non-naturalist in ethics because he believed that the metaphysics of morality
proclaims necessary truths (Flanagan 2011, p. 186). At the same time, Flanagan
admits that Wittgenstein did not back up this position with conclusive arguments,
and he passes over in silence the problem as to why Wittgenstein thought that only
contingent truths are meaningful.

254
that Wittgenstein granted ontological authority to natural science have their own
insolvable problems.
In my dissertation I tried, first of all, to find a balance between these contra-
dictory views on the Tractatus. I tried to treat seriously the thesis identifying true
propositions with the propositions of natural science. Second of all, I tried to see
how this “balanced” take on the Tractatus can answer the fundamental problem
posed by this book. Because I think that the naturalistic interpretations of the
Tractatus are rather in defence, I made an effort in my work most of all to defend
the view that Wittgenstein acknowledged the results and progress of natural sci-
ence with all its consequences for our views on what exists. In the first chapter
I defended the view, against the anti-metaphysical interpretation, that there is in-
deed in the Tractatus tension between the scientific worldview and our intuitions
about ethics (the anti-metaphysical interpretation denies that there is any world-
view in the Tractatus). I showed that Wittgenstein was interested in metaphysical
topics (in the understanding of metaphysics as a theory of what exists). I stressed
the fact that Wittgenstein himself admitted that his initial considerations on the
nature of a proposition evolved into an inquiry into the nature of being. I argued
that the picture theory of meaning assuming the idea of a comparison between a
proposition and a respective state of affairs, so to speak, “introduced reality” into
the scope of interests of early Wittgenstein. I was also advancing arguments in
favour of a claim that the most fundamental ontological thesis of the Tractatus is
TLP 2.02, which states that objects are simple.
In the following chapter I tried to figure out what the simples of the Tracta-
tus are. I defended the materialistic interpretation according to which Tractarian
simples are physical atoms – the most elementary particles of matter (I defended
this view against the phenomenalistic interpretation which sees the simple units
of perception in the role of simple objects). One of the most important argu-
ments in favour of the physicalistic interpretation indicates the way Wittgenstein
analysed the language of sensations in the Tractatus. This analysis assumes the
truthfulness of the physical concepts of colours and the priority of the language
of physics over the language of sensations. In Chapter 2 I also addressed the issue
concerning the nonsensicality of expressions which state the existence of objects.
The problem reads as follows: if Wittgenstein held the sentence: “There are two
objects in this room” to be nonsensical, then how can I sustain that he believed
that the sentence: “Simple objects are the most elementary particles of matter”
was true? By answering this accusation I made use of the distinction between
substantial and mere nonsense. Utterances about simple objects belong to the
first category, and this means that one can show what they aim to say by means of
other propositions. In my view, only a materialistic interpretation of the Tractatus

255
can explain thesis TLP 4.11 (the identification of true propositions with the prop-
ositions of natural science). If Wittgenstein thought that all one can meaningfully
say is said by physics, then there is no other option than to ascribe ontological
materialism to him.
In Chapter 3 I continued my considerations by showing how the materialistic
worldview affects, in Wittgenstein’s eyes, the activity of philosophers. In the fa-
mous thesis TLP 6.53, he counsels philosophers to repeat what scientists have to
say. If a philosopher wanted to break this rule, he counsels that that philosopher
has not conferred any meaning on the (presumably non-scientific) signs that phi-
losopher is using. In my opinion, Wittgenstein applied this strategy in TLP 5.54-
5.5422. His theory of judgement stating that a belief consists in a relation between
an event in the brain and a state of affairs aims, among others, to prove that Russell
did not confer any meaning in his theory of judgement on the concept of the mind.
On the example of this fragment of the Tractatus one can notice Wittgenstein’s
scepticism towards ontotheology, i.e. philosophy which tries to guarantee the fun-
damentals for faith and moral values by providing competitive theories describing
the world or by claiming that there is an aspect or part of the world – subject-
matter characteristic only of philosophy. Russell’s theory of judgement – belong-
ing to the tradition of dualism – is a good example of such philosophy. Its silent
assumption is the conviction that the realm of the mental is special subject-matter
for philosophical inquiry and its goal is to guarantee the distinctive character of
the mental. The second example of ontotheology is transcendental philosophy. In
Chapter 4 I analysed the views of those who think that Schopenhauer’s influence
on early Wittgenstein was so great that Wittgenstein himself could be classified as a
transcendental idealist. What the transcendental interpretations capture correctly
is the question of the fundamental problem of the Tractatus. However, I claim that
the medicine the transcendental interpretation applies to the disease of modernity
is harmful. Transcendentalists try to infer from the imperfections of the scientific
worldview (which were also noted by Wittgenstein534) that there is a part of reality
that is not captured by the scientific description of the world. For instance, if a sci-
entific description of the world does not contain a reference to moral values, then
one has to postulate the existence of a transcendental object (such as the willing
subject) through which values enter into the world. In Chapter 4 I aimed to show
that this was not and that it could not be Wittgenstein’s solution. According to the
Tractatus, even if we experience the imperfections of science with respect to the

534 “We feel that even when all possible scientific questions have been answered, the
problems of life remain completely untouched” (TLP 6.52).

256
problems of life, the only moral one should draw is that there are no meaningful
answers to these kinds of problems at all: “Of course there are then no questions
left, and this itself is the answer” (TLP 6.52).
The results of these chapters are mostly negative: young Wittgenstein did not
defend a transcendental view, but he was not a proponent of neo-positivism either
(he did not intend to deny the importance of ethics). What then was his position?
How did he reconcile ontological materialism with sensitivity to the problem of
the meaning of life? In Chapter 5 I described his view as subjectivism. Remember-
ing that for Wittgenstein, ethics includes not only issues connected with morality,
but also with aesthetics and the meaning of life, I put forward a hypothesis that for
him ethical utterances express the attitude of a speaker to the world. These atti-
tudes are irrational, i.e. not based on facts. The events of the world have no impact
on the attitudes one holds. For Wittgenstein, obviously, the expressions of these
attitudes are nonsensical and one should, consequently, be silent about them. This
solution, in my opinion, has an advantage over the transcendental interpretation
because it does not contradict naturalism. Someone who accepts subjectivism is
not bound to accept the existence of non-natural objects belonging to the higher
sphere of reality. A proponent of subjectivism, in opposition to a transcendental
philosopher, can grant natural science ontological authority. For him it is not the
sphere of facts which is the subject-matter of discussion between people holding
two different ethical opinions but rather the way they interpret facts.
The subjectivist philosophy of the Tractatus does not compete with the scien-
tific description of the world. It merely states that one can have different attitudes
to what is described by science. At the same time, it admits that these attitudes
(for instance that Somebody is taking care of me in this world, that my fate is
in His hands) cannot be put into words meaningfully – they are mere nonsense
under the Tractarian standards of sense.
How then did this subjective turn address the fundamental problem of the
Tractatus? Did early Wittgenstein succeed in the task of safeguarding values in a
world that is observing a rapid influence of science on our thinking?
In the Introduction I pointed at three aspects of this defence. One should pro-
tect human values from the claims of science because:
1. Science distorts ethical and religious notions and it can lead to treating hu-
man beings as objects.
2. Science treats religious and ethical convictions as premature views on the
world. This entails that a rational person educated in the 20th or 21th century
should reject naive ethics and religious beliefs.
3. The progress of science results in our lack of interest in the problems of life.

257
With respect to the last point I would say that it is an interesting question for so-
ciology to decide what factors have the greatest impact on people’s interests. Does
the lecture of the Tractatus stimulate ethical considerations in its readers? Obvi-
ously, this is not a philosophical question, and we do not know if Wittgenstein in
this sense defended the world of human values.
With respect to the first point, there are commentators who underline that the
significance of the Tractatus also consists in its prohibition of seeing human beings
as objects among other objects. Usually, these interpretations assume the transcen-
dental reading of the book535. According to these interpretations the self is not an
object of the empirical reality but its limit. Hence, one cannot apply the notion of
an object to the self. The problem is that I do not think that the transcendental
interpretation in general is right. I expressed my doubts if, according to the Trac-
tatus, there is such a thing as the self. The only meaningful view, according to the
Tractatus, with respect to human beings is to see them as bodies (TLP 5.631). This
view alone can lead to treating human beings as objects. Moreover, if the subjectiv-
ist interpretation is right, then it is possible to see human beings as persons pos-
sessing value and dignity, but at the same time it limits the validity of this view just
to my individual perspective. It does not give me any tools to criticise the view of
somebody who, for example, does not value black people as persons. His view, ac-
cording to Tractarian subjectivism, is just as nonsensical as mine. It turns out that
one of the costs of Tractarian naturalism is that it does not give me any intellectual
weapons in the fight with racism and other appalling moral views (remember, for
early Wittgenstein the only efficient weapon in ethics with respect to convincing
others is the infectious example of good behaviour).
In my opinion, Wittgenstein’s defence of values works only with respect to the
second point, but even here at a price which, for instance, the rational theists are
not eager to pay. I would even say that the whole of Wittgenstein’s defence of the
values of the human world could be reduced to the question of having the right
to hold certain moral and religious judgements; the right which is supposedly
endangered by scientific development. In my opinion, Wittgenstein’s tactics to de-
fend the world of human values consists in two claims. First, if I am correct in my
reading, moral judgements express attitudes and evaluations. Secondly, Wittgen-
stein believed in the dichotomy between facts and valuations. I argued in Chapter
5 that for this position there is no connection between interpretations and evalu-
ations and facts.

535 Sluga 1983; Klagge 1989.

258
 Let us take the notion which clearly belongs to the human point of view, i.e.
the notion of a miracle. A proponent of neo-positivism or scientism rejects the
existence of miracles because the notion of a miracle assumes the intervention of a
supernatural power536, and according to the naturalist worldview there is no place
for such entities537. According to him, a believer commits a mistake when he or
she believes, for example, that a recovery from a serious illness is a miracle. Sci-
ence can provide a more accurate description of the sequence of events that led
to the recovery. It is exactly this conviction which is the object of Wittgenstein’s
fierce attack. The notions of ethics or religion under a subjectivist understand-
ing are not supposed to describe the world or the sequence of events, so someone
who accepts subjectivism can agree with the proponent of scientism with respect
to what happened in the case of a recovery. He or she can not only concede the
value of scientific explanation, but also agree that only naturalistic explanations are
authentic. But the interpretation or the evaluation of the natural chain of events
is a different thing. I underlined in Chapter 5 that this interpretation is subjec-
tive – it has nothing to do with the facts – not only the present facts but also facts
from natural history (so the possible discoveries of evolutionary biologists do not
have to have any influence on people’s evaluations, attitudes and interpretations).
If it is not necessarily connected with the facts, then one has the right to hold the
attitudes one holds. For example, one has a right to see in a particular recovery a
miracle. If that is so, then, as Wittgenstein asserts:
It is absurd to say: ‘Science has proved that there are no miracles’. The truth is that the
scientific way of looking at a fact is not the way to look at it as a miracle (LE, p. 11).

In other words, when a philosopher says that one, by believing in miracles, holds
false convictions about the world, he is simply distorting the notion of a miracle.
Wittgenstein’s view entails that nothing that science claims can force a believer
to abandon his or her attitudes to the world. In a positivistic view, religious or
ethical statements are provisory descriptions of the world, which one should re-
ject when one acquires scientific knowledge538. On the other hand, according to
Wittgenstein, one has the right to hold his or her attitudes to the world, attitudes

536 Vollmer 2007, p. 39.

537 “Strong naturalism asserts that the distinction between nature and a realm over or
beyond nature is preposterous. ‘World’, ‘cosmos’ or ‘universe’ include every actually
existing ‘thing’. There is no place (and space) for supernatural entities” (Sukopp 2007,
p. 79).
538 “It is my belief that the ethical and religious motives, in spite of the splendidly im-
aginative systems to which they have given rise, have been on the whole a hindrance

259
according to which one evaluates the events in the world as miraculous, valuable,
good or evil. In this sense, in my opinion, Wittgenstein succeeds in safeguarding
the world of human values.
If propositions of science express states of affairs, then interpretations of facts
(especially interpretations of the totality of facts) are not scientific. This means,
however, that interpretations of facts cross the limits of meaningful discourse. If a
believer says that it seems to him or her that the world is telling him or her that his
or her existence was intentionally planned by a loving and omnipotent Being, then
his or her utterance is nonsensical. It is exactly as nonsensical as the utterance of
a positivist who says that the world tells him or her nothing (but is this a consola-
tion for believers?). Both of them cross the limits of sense. If both utterances are
equally nonsensical, then it is up to me which nonsense I choose. Although in the
eyes of the author of the Tractatus it would be best to keep silent in this case, he
himself could not despise the efforts of those who, nevertheless, cross the limits
of language:
My whole tendency and I believe the tendency of all men who ever tried to write or talk
Ethics or Religion was to run against the boundaries of language. This running against the
walls of our cage is perfectly, absolutely hopeless. (…) But it is a document of a tendency
in the human mind which I personally cannot help respecting deeply and I would not for
my life ridicule it (LE, p. 11–12).

This means that if I am correct in my interpretation, early Wittgenstein defends

ethics and reconciles it with the naturalistic worldview at an undeniably high
price. According to the Tractatus one has the right to hold ethical convictions at
the expense of their nonsensicality. Whereas scientism claims that religious ut-
terances are false, early Wittgenstein claims that sentences of religion, ethics and
aesthetics are meaningless – one does not communicate any content by means
of them. The consequence of the nonsensicality of ethics is the irrationality of
ethical views. From the point of view of someone who tries to secure the world
of human values from the demands of science, it has a bright side. In principle,
nothing from the outside can force one to change one’s attitudes. This means that
one is immune to what science can tell with respect to ethics. On the other hand,
the immunity of our attitudes against the influence of the outer world makes us, in
the eyes of Wittgenstein, similar to Leibnizian monads. In my opinion, this is why
he says that what solipsism tries to express is correct (TLP 5.62). This view has a
depressing practical consequence. Recall that the final conclusion of the Tractatus

to the progress of philosophy, and ought now to be consciously thrust aside by those
who wish to discover the philosophical truth” (Russell 1914b, p. 57).

260
states that what cannot be said – and the problems of life are included under this
rubric – should be passed over in silence (TLP 7). This means that in the eyes of
the author of the Tractatus we should not talk with one another about the most
important problems of our lives.
Finally, my hypothesis of subjectivism of early Wittgenstein has another unpal-
atable consequence for those who would want to defend their right of expressing
ethical or religious convictions. It particularly refers to religious utterances. Ac-
cording to Wittgenstein, someone who says: “I believe that God exists” in fact ex-
presses his trustful attitude to the world. He interprets facts in the world in such a
way that for him life has meaning539. We would probably find many believers who
would agree that religious statements are not in contradiction with the proposi-
tions of science, and that religion has a different function than science (the former
confers a meaning on lives; the latter describes the world). We would probably
find many believers who conclude that God exists not on the basis of facts but
simply by interpreting the world as a whole – they have this kind of sensitivity
which allows them to look at things in the world and conclude: “Somebody has
to stand behind all of this”. But exactly the last conclusion is not in accordance
with the subjectivism I ascribe to Wittgenstein. Notice that a religious expression,
according to subjectivism, expresses the speaker’s interpretation of the world but
not his ontological convictions. In other words, faith means for Wittgenstein that
when being confronted with the world one states that one feels safe, or that life has
meaning; but not that being confronted with the world one concludes that there
is a Creator of it540. Probably for this reason he was not seen by Englemann or
Malcolm as a religious person in the traditional meaning of the word541. I am not
sure if many believers would be ready to accept that they have the right to their
religious convictions but at the expense of accepting that these convictions say

539 “To believe in God means to see that life has a meaning” (NB 8.7.16, p. 74).
540 “If the believer in God looks around & asks ‘Where does everything I see come from?’,
‘Where does all that come from?’, what he hankers after is not a (causal) explana-
tion; and the point of his question is that it is the expression of this hankering. He is
expressing, then, a stance towards all explanations” (CV, p. 96–97).
541 “Was Wittgenstein religious? (…) The idea of God in the sense of a Bible, the image
of God as a Creator, hardly ever engaged Wittgenstein’s attention” (Engelmann 1967,
p. 77). “I do not wish to give the impression that Wittgenstein accepted any religious
faith – he certainly did not – or that he was a religious person” (Malcolm 1984, p. 60);
von Wright wrote: “I do not know whether he can be said to have been ‘religious’ in
any but a trivial sense of this word” (von Wright 1955/1967, p. 27).

261
nothing about the existence of God. In other words, Wittgenstein’s subjectivism
comes close to the accusation of distorting the nature of religious beliefs.
This accusation is clear with respect to religious discourse. It is doubtful that
religious persons would agree that their religions say nothing about God, the af-
terlife, etc. It is less clear (maybe it is the subject-matter for sociology) with respect
to ethics. Would people agree that they have a right to hold their moral judge-
ments, but only under the condition that they do not understand them as saying
anything about objective moral values or moral laws? If they did not, then it would
be another argument against Wittgenstein’s defence of the world of human values.
The costs of reconciling the right to hold ethical and religious views with the
naturalistic worldview are indeed great, but at least in my opinion only an interpre-
tation which assumes such a subjective turn in the early writings of Wittgenstein is
able to find a balance between the naturalism of the Tractatus and its modernistic
anxiety about the impact of scientific progress on the world of human values. To
put it in Wittgenstein’s words, it finds a balance between the written and the un-
written part of the book. However, it seems that Wittgenstein himself did not hold
for very long this subjectivist answer to the fundamental problem of the Tractatus.
We saw that already in 1929 in A Lecture on Ethics he gave in to the temptation to
express the nonsense of ethics. He acknowledges there that ethics is a constant and
invincible tendency in the human mind. In the later development of his philosoph-
ical career he tended to concede that ethical and religious expressions convey, after
all, content – one just cannot cross them off as nonsensical. From his Tractarian
starting point he had two possible paths to follow: he could choose the transcen-
dental path, i.e. to sustain that meaningful expressions are only those expressed by
science, but simultaneously admitting that there are objects and aspects of reality
which science cannot capture, or he could deny that only propositions of science
are meaningful. In my opinion, in Philosophical Investigations he chose the latter
possibility.
Dissatisfaction with the solution to the fundamental problem of the Tracta-
tus was strictly connected with the idea as to what it means to convey content
(or, in other words, with the idea of one final analysis of linguistic expressions –
c.f. PI, 91). It is possible, then, that the fact that the bulk of Wittgenstein’s later
work was devoted to the problem of the sense of propositions could easily be
explained by the hypothesis that he was constantly searching for the answer to
his fundamental problem; however, the development and justification of this
hypothesis does not belong to the subject-matter of this dissertation.

262
Bibliography

Ludwig Wittgenstein’s works

CV, “Culture and Value”, G.H. von Wright & H. Nyman (ed.), trans. by P. Winch,
Oxford: Blackwell, 1998
LE, “A Lecture on Ethics”, The Philosophical Review 74 (1965), no. 1, pp. 3–12
Lectures 1930–1932, “Lectures 1930–1932. From the Notes of John King and
Desmond Lee”, D. Lee (ed.), Oxford: Basil Blackwell, 1982
Lectures & Conversations, “Lectures and Conversations on Aesthetics, Psychol-
ogy and Religious Belief ”, compiled from notes taken by Y. Smythies, R. Rhees
and J. Taylor, ed. by C. Barrett, Berkeley and Los Angeles: University of Cali-
fornia Press, 1967
Letters, “Letters to Russell, Keynes and Moore”, G.H. von Wright (ed.), Oxford:
Basil Blackwell, 1974
NB, “Notebooks 1914–1916”, trans. from German by G.E.M. Anscombe; G.H.
von Wright & G.E.M. Anscombe (ed.), Oxford: Basil Blackwell, 1979
NL, “Notes on Logic”, in: NB, pp. 93–107
NM, “Notes dictated to G.E. Moore in Norway” in: NB, pp. 108–119.
NT, “Notes on Talks with Wittgenstein taken by F. Waismann”, The Philosophical
Review 74 (1965), no. 1, pp. 12–16
PI, “Philsophische Untersuchungen. Philosophical Investigations”, trans. by
G.E.M. Anscombe, P.M.S. Hacker & J. Schulte, Oxford: Wiley-Blackwell,
2009
PG, “Philosophical Grammar”, trans. by A. Kenny; R. Rhees (ed.), Berkeley &
Los Angeles: University of California Press, 1974
PR, “Philosophical Remarks”, R. Rhees (ed.), trans. by R. Hargreaves & R. White,
Oxford: Basil Blackwell, 1975
ProtoTractatus, “ProtoTractatus—An Early Version of Tractatus Logico- Philo-
sophicus”, trans. by D.F. Pears & B.F. McGuinness; B.F. McGuinness, T. Nyberg,
G.H. von Wright (eds.), London: Routledge & Kegan Paul, 1971
TLP, “Tractatus logico-philosophicus”, German text: “Logish-philosophishe Ab-
handlung”, in: Ludwig Wittgenstein, „Werkausgabe Band I“, pp. 7–85, Frankfurt
am Main: Suhrkamp Verlag, 2006; English translation: „Tractatus logico-phil-
osophicus“, trans. by D.F. Pears & B.F. McGuinness, London: Routledge, 2001
Wittgenstein, L. 1973, “Letters to C.K. Ogden”, Oxford: Basil Blackwell and
London: Routledge & Kegan Paul

263
Wittgenstein, L. 2003, “The Voices of Wittgenstein. The Vienna Circle”, trans.
by G. Baker, M. Mackert, J. Connolly & V. Politis; G. Baker (ed.), London:
Routledge
WVC, “Wittgenstein and the Vienna Circle” B. McGuinness (ed.), trans. by
J. Schulte & B. McGuinness, Oxford: Basil Blackwell, 1979

Others
Allard, J. 2008, “Idealism in Britain and the United States”, in: Baldwin 2008,
pp. 43–59
Armstrong, D.M. 1968, “A Materialist Theory of The Mind”, London: Routledge
& Kegan Paul
Anscombe, G.E.M. 1965, “An Introduction to Wittgenstein’s Tractatus”, New
York: Harper Torchbooks
Atkinson, J. R. 2009, “The Mystical in Wittgenstein’s Early Writings”, New York –
London: Routledge
Ayer, A. J. (ed.) 1959, “Logical Positivism”, Westport: Greenwood Press Publishers
Baldwin, Th. (ed.) 2008, “The Cambridge History of Philosophy 1870–1945”,
Cambridge: Cambridge University Press
Beard, R.W. & Copi, I.M. (eds.) 1973, “Essays on Wittgenstein’s Tractatus”, New
York: Hafner Press
Bergmann, G. 1973, “The Glory and the Misery of Ludwig Wittgenstein”, in:
Beard & Copi 1973, pp. 343–358
Birk, A. 2006, “Vom Verschwinden des Subjekts. Eine historisch-systematische
Untersuchung zur Solipsismusproblematik bei Wittgenstein”, Paderborn:
Mentis
Black, M. 1964, “A Companion to Wittgenstein’s Tractatus”, Ithaca: Cornell Uni-
versity Press
Blank, A. 2007, “Material Points and Formal Concepts in the Early Wittgen-
stein”, in: Canadian Journal of Philosophy” 37 (2007), no. 2, pp. 245–262
Bradley, R. 1992, “The Nature of All Being. A Study of Wittgenstein’s Modal
Atomism”, Oxford: Oxford University Press
Carey, R. 2007, “Russell and Wittgenstein on the Nature of Judgement”, London:
Continuum
Carnap, R. 1931/1995, “The Unity of Science”, transl. by M. Black, Bristol: Thoe-
mmes Press
Carnap, R. 1959, “Psychology in Physical Language”, transl. by G. Schick, in:
Ayer 1959, pp. 165–198

264
Carnap, R. 1964/1967, From his “Autobiography”, in: Fann 1967, pp. 33–39
Carruthers, P. 1990, “The Metaphysics of the Tractatus”, Cambridge: Cambridge
University Press
Cerbone, D. R. 2000, “How to Do Things with Wood? Wittgenstein, Frege, and
the Problem of Illogical Thought”, in: Crary & Read 2000, pp. 293–314
Chalmers, D. J. 1996, “The Conscious Mind. In Search of a Fundamental Theory”,
New York: Oxford University Press
Child, W. 2011, “Wittgenstein”, London – New York: Routledge
Cohen, M. 1974, “Tractatus 5.542”, in: Mind 83 (1974), no. 331, pp. 442–444
Cook, J. W. 1994, “Wittgenstein’s Metaphysics”, Cambridge: Cambridge Univer-
sity Press
Conant, J. F. 2000, “Elucidation and Nonsense in Frege and Early Wittgenstein”,
in: Crary & Read 2000, pp. 174–217
Copi, I. M. 1973, “Objects, Properties, and Relations in the Tractatus”, in: Beard
& Copi 1973, pp. 167–186
Crane, T. M. 2001, “Elements of Mind. An Introduction to the Philosophy of
Mind”, New York: Oxford University Press
Crary, A. 2000, “Introduction”, in: Crary & Read 2000, pp. 1–18
Crary, A. & Read R. (eds.) 2000, “The New Wittgenstein”, London & New York:
Routledge
Dayton, E. B. 1976, “Tractatus 5.54-5.5422”, Canadian Journal of Philosophy 6
(1976), no. 2, pp. 275–283
Descartes, R. 1641/2008, “Meditations on First Philosophy. With Selections
from the Objections and Replies”, transl. by M. Moriarty, Oxford: Oxford
University Press
Diamond, C. 1974–1982/1995, “Realism and Realistic Spirit”, in: Diamond
1995, pp. 39–72
Diamond, C. 1995, “The Realistic Spirit. Wittgenstein, Philosophy and the
Mind”, Cambridge MA: MIT Press
Diamond, C. 1995a, “Wittgenstein and Metaphysics”, in: Diamond 1995,
pp. 13–38
Diamond, C. 2000a, “Ethics, Imagination, and the Method of Wittgenstein’s
Tractatus”, in: Crary & Read 2000, pp. 149–173
Diamond, C. 2000b, “Does Bismarck have a beetle in his box? The private lan-
guage argument in the Tractatus”, in: Crary & Read 2000, pp. 262–292
Drury, M. O’C. 1960/1967, “Assessments of the Man and the Philosopher”, in:
Fann 1967, pp. 67–71

265
Engelmann, P. 1967, “Letters from Ludwig Wittgenstein with a Memoir”, Oxford:
Basil Blackwell
Fann, K. T. 1967, “Ludwig Wittgenstein. The Man and His Philosophy”, New
York: A Delta Book
Fischer, M. 1993, “Wittgenstein as a Modernist Philosopher”, in: Philosophy and
Literature 17 (1993), vol. 2, pp. 279–285
Flanagan, O. 2011, “Wittgenstein’s Ethical Nonnaturalism: An Interpretation of
the Tractatus 6.41–6.47 and the Lecture on Ethics”, in: American Philosophical
Quarterly 48 (2011), no. 2, pp. 185–198
Fodor, J. A. 2008, “LOT 2. Language of Thought Revisited”, Oxford: Clarendon
Press
Fogelin, R.J. 1996, “Wittgenstein”, London: Routlegde
Frascolla, P.  2007, “Understanding Wittgenstein’s Tractatus”, London & New
York: Routledge
Frege, G. F. L. 1884, “Grundlage der Arithmetik”, Breslau: Wilhelm Koebner
Frege, G. F. L. 1892/1984a, “On Sense and Meaning”, in: Frege 1984, pp. 157–177.
Frege, G. F. L. 1892/1984b, “On Concept and Object”, in: Frege 1984, pp. 182–194.
Frege, G. F. L. 1984, “Collected Papers on Mathematics, Logic, and Philosophy”,
B. McGuinness (ed.), Oxford-New York: Basil Blackwell
Fumerton, R. 2010, “Phenomenalism”, in: J. Dancy, E. Sosa, M. Steup (ed.), A
Companion to Epistemology, Oxford: Wiley-Blackwell, pp. 586–590.
Gasser G. (ed.) 2007, “How Successful is Naturalism? Publications of the Austri-
an Ludwig Wittgenstein Society. New Series. Vol. 4”, Frankfurt: Ontos Verlag
Gellner, E. A. 2004, “Language and Solitude. Wittgenstein, Malinowski and the
Habsburg Dilemma”, Cambridge: Cambridge University Press
Glock, H.-J. 1996, “A Wittgenstein Dictionary”, Oxford: Blackwell
Glock, H.-J. 1997, “Kant and Wittgenstein: Philosophy, necessity and repre-
sentation”, in: International Journal of Philosophical Studies 5 (1997), no. 2,
pp. 285–305
Grasshof, G. 1997, “Hertzian Objects in Wittgenstein’s Tractatus”, in: British
Journal for the History of Philosophy 5 (1997), no. 1, pp. 87–120
Griffin, J. P. 1964, “Wittgenstein’s Logical Atomism”, Oxford: Clarendon Press
Hacker, P.M.S. 1971, “Wittgenstein’s Doctrine of the Soul in the Tractatus”, in:
Kantstudien 62 (1971), pp. 162–171
Hacker, P.M.S. 1972, “Insight and Illusion”, Oxford: Clarendon Press
Hacker, P.M.S. 1986a, “Laying the Ghost of the Tractatus”, in: Shanker 1986,
pp. 76–91

266
Hacker, P.M.S. 1986b, “The Rise and Fall of the Picture Theory, in: Shanker
1986, pp. 116–135
Hacker, P.M.S. 2000, “Was He Trying to Whistle it?”, in: Crary & Read 2000,
pp. 353–388
Haller, R. 1989, “Wittgenstein and Physicalism”, in: Critica 21 (1989), no. 63,
pp. 17–32
Hamilton, K. A. 2001, “Wittgenstein and the Mind’s Eye”, in: J. C. Klagge (ed.),
„Wittgenstein. Biography and Philosophy“, Cambridge: Cambridge Univer-
sity Press, pp. 53–97.
Hanks, P. W. 2012, “Early Wittgenstein on Judgement”, in: Zalabardo 2012,
pp. 37–63
Harré, H. R. 2001, “Wittgenstein: Science and Religion”, in: Philosophy 76
(2001), vol. 2, pp. 211–237
Harré, H. R. 2008, “Positivist thought in the nineteenth century”: in: Baldwin
2008, pp. 11–26
Herling-Grudziński, G. 2011, “Dziennik pisany nocą [The Journal Written at
Night] 1971–1981”, Kraków: Wydawnictwo Literackie
Hertz, H. R. 1894, “Die Prinzipien der Mechanik”, Leipzig: Johann Ambrosius
Barth
Hintikka, M. B. & Hintikka, J. 1990, „Untersuchungen zu Wittgenstein“, trans.
by J. Schulte, Frankfurt am Main: Suhrkamp Verlag
Hohwy, J. & Kallestrup, J. (eds.) 2008, “Being Reduced. New Essays on Reduc-
tion, Explanation, and Causation”, Oxford: Oxford University Press,
Hookway, Ch. 2008, “Pragmatism”, in: Baldwin 2008, pp. 74–89
Horwich, P. 2012, “Wittgenstein’s Metaphilosophy”, Oxford: Clarendon Press
Hume, D. 1739/2001, “A Treatise of Human Nature”, D.F. Norton & M. J. Norton
(eds.), Oxford: Oxford University Press
Hustwit, R. E. 2011, “Wittgenstein on Modernism and the Causal Point of
View”, American Catholic Philosophical Quarterly 85 (2011), pp. 565–576
Hyder, D. 2002, “The Mechanics of Meaning. Propositional Content and the
Logical Space of Wittgenstein’s Tractatus”, Berlin: Walter de Gruyter
Ishiguro, H. 1969, “Use and Reference of Names”, in: Winch 1969, pp. 20–50
Ishiguro, H. 1981, “Thought and Will in Wittgenstein’s Tractatus”, in: E. Morscher
& R. Stranzinger (eds), “Ethics. Foundations, Problems, and Applications. Pro-
ceedings of the Fifth International Wittgenstein Symposium”, Vienna: Hölder –
Pichler – Tempsky, pp. 455–463
Jacquette, D. 1992/1993, “Wittgenstein’s Critique of Propositional Attitudes and
Russell’s Theory of Judgement”, in: Brentano Studien 4 (1992/1993), pp. 193–220

267
Janik, A. S. 2001, “Wittgenstein’s Vienna Revisited”, New Brunswick: Transac-
tion Publishers
Johnston, C. 2012, “Russell, Wittgenstein, and Synthesis in Thought” in: Zalab-
ardo 2012, pp. 15–36
Jubien, M. E. 2007, “Analyzing Modality”, in: D. W. Zimmerman (ed.), “Oxford
Studies in Metaphysics”, Volume 3, Oxford-New York: Oxford University
Press, pp. 99–139
Kahane G. & Kanterian, E. & Kuusela, O. (eds.) 2007, “Wittgenstein and His In-
terpreters. Essays in Memory of Gordon Baker”, Oxford: Blackwell Publishing
Kannisto, H. A. 1986, “Thoughts and Their Subject. A Study of Wittgenstein’s
Tractatus”, Acta Philosophia Fennica 40 (1986)
Kelly, J. C. 1995, “Wittgenstein, The Self, and Ethics”, in: The Review of Meta-
physics 48 (1995), no. 3, pp. 567–590
Kenny, A. J. P. 1981, “Wittgenstein’s Early Philosophy of Mind” in: I. Block (ed.),
Oxford: Basil Blackwell, pp. 140–147
Kenny, A. J. P. 1986, “The Ghost of the Tractatus”, in: Shanker 1986, pp. 65–75
Keyt, D. 1965, “A New Interpretation of the Tractatus Examined”, in: The Philo-
sophical Review 74 (1965), pp. 229–239
Keyt, D. 1973, “Wittgenstein’s Notion of an Object”, in: Beard & Copi 1973,
pp. 289–303
Kim, J. 2008, “Reduction and Reductive Explanation: Is One Possible Without
the Other?”, in: Hohwy & Kallestrup 2008, pp. 93–114
Klagge, J. C. 1989, “Wittgenstein and Neuroscience”, in: Synthese 78 (1989), no. 3,
pp. 319–343
Klemke, E. D. (ed.) 1971, “Essays on Wittgenstein”, Chicago: University of Il-
linois Press
Klemke, E. D. 1971a, “The Ontology of Wittgenstein’s Tractatus”, in: Klemke
1971, pp. 104–119
Kremer, M. J. 2004, “To What Extent is Solipsism a Truth?”, in: B. Stocker (ed.),
“Post-analytic Tractatus”, Aldershot: Ashgate Publishing, pp. 59–84
Lampert, T. 2000, “Wittgenstein’s Physikalismum. Die Sinnesdatenanalyse des
Tractatus Logico-Philosophicus in ihrem historischen Kontext, Padeborn:
Mentis
Lampert, T. 2003, “Psychophysical and Tractarian Analysis”, in: Perspectives on
Science 11 (2003), no. 1, pp. 285–317
Levi, A. 1964/1967, “Wittgenstein as Dialectician”, in: Fann 1967, pp. 366–379
Locke, J. 1690/1999, “An Essay Concerning Human Understanding”, Hazleton:
Pennsylvania State University

268
Loewer, B. M. 2008, “Why There is Anything Except Physics?”, in: Hohwy &
Kallestrup 2008, pp. 149–163
Malcolm, N. 1984, “Ludwig Wittgenstein: A Memoir”, Oxford-New York: Ox-
ford University Press
Malcolm, N. 1986, “Nothing is Hidden. Wittgenstein’s Criticism of his Early
Thought”, Oxford: Basil Blackwell
Martin, C.B. 1980/1999, “Substance Substantiated”, in: M. Tooley (ed.), “Ana-
lytical Metaphysics. A Collection of Essays. Volume 4: Particulars, Actuality,
and Identity over Time”, New York: Garland Publishing, pp. 37–44
McGinn, M. 2007, “Simples and the Idea of Analysis in the Tractatus, in: Kahane
2007, pp. 200–220
McGuinness, B. F. 2002, “Approaches To Wittgenstein. Collected Papers”, Lon-
don: Routledge
McGuinness, B. F. 2012, “Two Cheers for the ‘New’ Wittgenstein?”, in: Zalab-
ardo 2012, pp. 260–272
Meyer, M. 1986, “From Truth-Tables to Ordinary Language: The Implications of
Generalised Analycity”, in: Shanker 1986, pp. 296–322
Midgley, M. 1992, “Science as Salvation”, London – New York: Routledge
Monk, R. 1991, “Ludwig Wittgenstein. The Duty of Genius”, London: Vintage
Moore, A. W. 2007, “Wittgenstein and Transcendental Idealism”, in: Kahane
2007, pp. 174–199
Morris, M. 2008, “Wittgenstein and the Tractatus”, London: Routledge
Morris, M. & Dodd, J. 2009, “Mysticism and Nonsense in the Tractatus”, in:
European Journal of Philosophy 17 (2009), vol. 2, pp. 247–276
Neurath, O. K. W. 1931a/1983, “Physicalism. The Philosophy of the Viennese
Circle”, in: Neurath 1983, pp. 48–51
Neurath, O. K. W. 1932/1983, “Protocol Statements”, in: Neurath 1983, pp. 91–99.
Neurath, O. K. W. 1959, “Sociology and Physicalism”, transl. by M. Magnus & R.
Raico, in: Ayer 1959, pp. 282–317
Neurath, O. K. W. 1983, “Philosophical Papers 1913–1946”, ed. and trans. by R.
S. Cohen & M. Neurath, Dordrecht: D. Reidel Publishing Company
Nordmann, A. 2002, “Another New Wittgenstein: The Scientific and Engineer-
ing Background of the Tractatus”, in: Perspectives on Science 10 (2002), no. 3,
pp. 356–384
Ostrow, M. B. 2002, “Wittgenstein’s Tractatus. A Dialectical Interpretation”,
Cambridge: Cambridge University Press
Papineau, D. 2008, “Must a Physicalist be a Microphysicalist?”, in: Hohwy &
Kallestrup 2008, pp. 126–148

269
Pears, D. F. 1970, “Ludwig Wittgenstein”, New York: The Viking Press
Pears, D. F. 1977, “The Relation Between Wittgenstein’s Picture Theory of Prop-
ositions and Russell’s Theories of Judgment”, The Philosophical Review 86
(1977), no. 2, pp. 177–196
Pears, D. F. 1987, “The False Prison. Vol. I”, Oxford: Oxford University Press
Pears, D. F. 1993, “The Ego and the Eye: Wittgenstein’s Use of an Analogy”, Graz-
er Philosophische Studien, 44 (1993), pp. 59–68
Potter, M. K. 2006, “Russell’s Ethics”, London: Continuum
Proops, I. 2001, “The New Wittgenstein. A Critique” in: European Journal of
Philosophy 9 (2001), vol. 3, pp. 375–404
Puhl, K. 1999, “Subjekt und Körper. Untersuchungen zur Subjektkritik bei Witt-
genstein und zur Theorie der Subjektivität“, Paderborn: Mentis
Putnam, H. W. 1986/1994, “How Old Is the Mind?”, in: H. Putnam, “Words and
Life”, J. Conant (ed.), Cambridge MA: Harvard University Press, pp. 3–21
Putnam, H. W. 1992, “Renewing Philosophy”, Cambridge MA – London: Har-
vard University Press
Putnam, H. W. 2008, “Wittgenstein and Realism”, International Journal of Philo-
sophical Studies 16 (2008), no. 1, pp. 3–16
Putnam, H. W. 2010, “Science and Philosophy”, in: H. Putnam, “Philosophy in an
Age of Science”, M. DeCaro, D. Macarthur (eds.), Cambridge MA – London:
Harvard University Press, pp. 39–50
Ramsey, F. P. 1923/1986, “Critical Notice of the Tractatus”, Mind 23, pp. 465–478,
reprinted in: Shanker 1986, pp. 34–46
Rhees, R. 1965, “Some Developments in Wittgenstein’s View of Ethics”, The Phil-
osophical Review 74 (1965), no. 1, pp. 17–26
Rhees, R. 1969/1970, “ ‘Ontology’ and Identity in the Tractatus à propos of
Black’s Companion” in: Rhees 1970, pp. 23–36
Rhees, R. 1970, “Discussions of Wittgenstein”, London: Routledge & Kegan Paul
Ricketts, T. G. 1996, “Pictures, logic, and the limits of sense in Wittgenstein’s
Tractatus”, in: D.G. Stern & H.D. Sluga (eds.), “The Cambridge Compan-
ion to Wittgenstein”, Cambridge – New York: Cambridge University Press,
pp. 59–99
Ricketts, T. G. 2010, “Concepts, objects, and the Context Principle”, in: M. Potter
& T. Ricketts (eds.), “The Cambridge Companion to Frege”, Cambridge: Cam-
bridge University Press, pp. 149–219
Rosenberg, J. F. 1968, “Intentionality and Self in the Tractatus”, Nous 2 (1968),
no. 4, pp. 351–358

270
Russell, B. A. W. 1913/1984, “Theory of Knowledge. The 1913 Manuscript” in:
“The Collected Papers of Bertrand Russell. Vol. 7”, ed. by E. R. Eames & K.
Blackwell, London: George Allen & Unwin
Russell, B. A. W. 1914a, “The Relation of Sense-Data to Physics”, in: Russell 1986,
pp. 5–26
Russell, B. A. W. 1914b, “On Scientific Method in Philosophy”, in: Russell 1986,
pp. 57–73
Russell, B. A. W. 1918, “The Philosophy of Logical Atomism” in: Russell 1971,
pp. 177–281
Russell, B. A. W. 1922, “Introduction”, in: TLP, pp. ix–xxv (TLP, trans. D.F.Pears,
B.F.McGuinness, Taylor & Francis e-Library 2002)
Russell, B. A. W. 1924, “Logical Atomism” in: Russell 1971, pp. 323–343
Russell, B. A. W. 1971, “Logic and Knowledge. Essays 1901–1950”, ed. by R.C.
Marsh, London: Gordon Allen & Unwin
Russell, B. A. W. 1986, “The Philosophy of Logical Atomism and Other Essays
1914–1919. The Collected Papers of Bertrand Russell. Vol. 8”, ed. by J.G. Slater,
London: Gordon Allen & Unwin
Ryle, G. 1951/1967, “Ludwig Wittgenstein”, in: Fann 1967, pp. 116–124
Schopenhauer, A. 1844/1966a, “The World as Will and Representation”, vol. I,
transl. from German by E. F. J. Payne, New York: Dover Publications; German
text: “Die Welt als Wille und Vorstellung”, 1. Band, Frankfurt am Main, Haf-
fmans Verlag 2006
Schopenhauer, A. 1844/1966b, “The World as Will and Representation”, vol. II,
transl. from German by E. F. J. Payne, New York: Dover Publications; Ger-
man text: “Die Welt als Wille und Vorstellung”, 2. Band, Frankfurt am Main,
Haffmans Verlag 2006
Shanker, S.G. (ed.) 1986, “Ludwig Wittgenstein. Critical Assessments. Volume
One. From the Notebooks to Philosophical Grammar. The Construction and
Dismantling of the Tractatus”, London: Routledge
Shanker, S.G. 1986a, “Introduction: The Philosophical Significance of the Trac-
tatus”, in: Shanker 1986, pp. 16–33
Shier, D. 1997, “How Can Pictures be Propositions?”, Ratio 10 (1997), no. 2,
pp. 65–75
Sluga, H. D. 1983, “Subjectivity in the Tractatus”, in: Synthese 56 (1983), no. 2,
pp. 123–139
Sluga, H. D. 1996, “ ‘Whose House is that?’ Wittgenstein on the self ”, in: H. Sluga
& D.G. Stern, “The Cambridge Companion to Wittgenstein”, New York: Cam-
bridge University Press, pp. 320–353

271
Sluga, H. D. 2012, “Simple Objects: Complex Questions”, in: Zalabardo 2012,
pp. 99–118
Soin, M. 2001, “Gramatyka i metafizyka. Problem Wittgensteina”, Wrocław: Funna
Sosa, E. 1997, “Subject Among Other Things”, in: M. C. Rea (ed.), “Material Con-
stitution. A Reader”, Lanham: Rowman & Littlefield Publishers, pp. 63–89
Stenius, E. G. 1960/1996, “Wittgenstein’s Tractatus. A Critical Exposition of the
Main Lines of Thought”, Bristol: Thoemmes Press
Stern, D. G. 1995, “Wittgenstein on Mind and Language”, New York-Oxford:
Oxford University Press
Stern, D. G. 2007, “Wittgenstein, the Vienna Circle, and Physicalism: A Reas-
sessment” in: “The Cambridge Companion to Logical Empiricism”, A. Rich-
ardson & Th. Uebel (eds.), Cambridge – New York: Cambridge University
Press, pp. 305–331.
Stevenson, Ch. L. 1963/1975, “Facts and Values. Studies in Ethical Analysis”,
Westport (Connecticut): Greenwood Press
Stokhof, M. J. B. 2002, “World and Life as One. Ethics and Ontology in Wittgen-
stein’s Early Thought”, Stanford CA: Stanford University Press
Sukopp, Th. 2007, “How Successful is Naturalism? Talking about Achievements
beyond Science”, in: Gasser 2007, pp. 77–102
Sullivan, P. M. 2004, “What is the Tractatus about?”, in: M. Kölbel & B. Weiss
(eds.), “Wittgenstein’s Lasting Significance”, London-New York: Routledge
Taylor, Ch. M. 1989, “Sources of the Self. The Making of the Modern Identity”,
Cambridge MA: Harvard University Press
Tolstoy, L. 1896, “The Gospel in Brief ”, trans. by I. S. Hapgood, New York:
Thomas Y. Crowell & Company
Walker, R. C. S. 1993, “Transcendental Arguments against Physicalism”, in:
H. Robinson (ed.), “Objections to Physicalism”, Oxford: Clarendon Press,
pp. 61–80
Waugh, A. 2010, “Das Haus Wittgenstein. Die Geschichte einer ungewöhnli-
chen Familie”, transl. from English by S. Röckel, Frankfurt am Main: Fischer
Verlag
Williams, B. A. O. 2002, “The Human Prejudice”, Walter E. Edge Lectures (giv-
en at Princeton University on 15 October 2002), www.princeton.edu/Web-
Media/lectures
Winch, P. G. 1969a, “The Unity of Wittgenstein’s Philosophy”, in: Winch 1969,
pp. 1–19.
Winch, P. G. (ed.) 1969, “Studies in the Philosophy of Wittgenstein”, London:
Routledge & Kegan Paul

272
van Fraasen, B. C. 2005, “Transcendence of the Ego (The Non-Existent Knight)”,
in: G. Strawson (ed.), “The Self?”, Malden MA – Oxford: Blackwell, pp. 87–110
Vollmer, G. 2007, “Can Everything Be Rationally Explained Everywhere in the
World?”, in: Gasser 2007, pp. 25–48.
von Wright, G.H. 1955/1967, “A Biographical Sketch”, in: Fann 1967, pp. 13–29
Zalabardo, J. L. (ed.) 2012, “Wittgenstein’s Early Philosophy”, Oxford: Oxford
University Press
Zalabardo, J. L. 2012a, “Reference, Simplicity, and Necessary Existence in the
Tractatus”, in: Zalabardo 2012, pp. 119–150.
Zemach, E. M. 1973, “Wittgenstein’s Philosophy of the Mystical” in: Beard &
Copi 1973, pp. 359–375
Ziemiński, I. 2007, “Death Is Not an Event in Life: Ludwig Wittgenstein as a
Transcendental Idealist” in: Idealistic Studies 37 (2007), no. 1, pp. 51–66

273