Professional Documents
Culture Documents
)
Wittgenstein: Issues and Debates
APORIA
Apori/a
ADVISORY BOARD
Pavo Barišić (University of Split)
Michel Le Du (Université de Strasbourg)
Guillermo Hurtado (National Autonomous University of Mexico)
Lorenzo Peña (Spanish National Research Council)
Nuno Venturinha (New University of Lisbon)
Nicanor Ursua Lezaun (University of the Basque Country)
Pablo Quintanilla (Pontifical Catholic University of Peru)
Aporia is a new series devoted to studies in the field of philosophy. Aporia (Aπορία)
means philosophical puzzle and the aim of the series is to present contributions by
authors who systematically investigate current problems. Aporia (Aπορία) puts special
emphasis on the publication of concise arguments on the topics studied. The
publication has to contribute to the explanation of current philosophical problem, using
a systematic or a historic approach. Contributions should concern relevant
philosophical topics and should reflect the ongoing progress of scientific development.
Band 3 / Volume 3
Eric Lemaire, Jesús Padilla Gálvez (Eds.)
Wittgenstein:
Issues and Debates
Bibliographic information published by the Deutsche Nationalbibliothek
ISBN 978-3-86838-083-5
2010
MICHEL LE DU 11
Tacit Knowledge and Action
SABINE PLAUD 31
Synoptic Views vs. Primal Phenomena: Wittgenstein
on Goethe’s Morphology
ERIC LEMAIRE 47
Critical Remarks on Anti-Metaphyscial Readings of Wittgenstein
AYùEGÜL ÇAKAL 65
What Does the Repudiation of Private Language Mean
in Wittgenstein’s Philosophy?
Abbreviations 149
Wittgenstein: Issues and Debates
Eric LEMAIRE and Jesús PADILLA GÁLVEZ
The title of the book refers to the philosopher Wittgenstein, and on issues
on which he was working. Through the publication of his writings of the
middle period, we have gained a new perspective of both, the Tractatus
and the Philosophical Investigations. In summary, one could say that
Wittgenstein has corrected some aspects that he had later considered as
mistakes in the Tractatus. He was very receptive towards the contemporary
discussions and addressed all the relevant issues in philosophy.
Consequently, his arguments became more and more sophisticated. The
book attempts to present a new view of Wittgenstein’s works.
Philosophy can give rise to a certain malaise. On the one hand, in
philosophy we ask important or fundamental questions about the nature of
human beings, the existence or inexistence of God, values we should
follow in our life, the limits of our knowledge, and so on. We want that
these questions do not remained unanswered while, on the other hand, as P.
van Inwagen notices, that they are no established facts or theories in
philosophy, no normal philosophy. Philosophical investigations do not give
rise to wide and long-standing consensus among philosophers, contrary to
what usually happens in natural sciences or mathematics. Why
philosophers never attain definitive solutions to philosophical problems?
During the history of philosophy, this fact has given to philosophers (like
Hume or Kant or the logical positivists for example) the occasion to raise
doubts on our abilities to pursue philosophical knowledge and has received
several explanations. One could say that philosophical problems are too
difficult for us, that our epistemic equipment is not suitable to solve them.
One could as well think that we could attain knowledge if we reform
philosophy, its methods, its ambitions, and its object. One could also
believe that we should be more patient while others assume this is a brutal
characteristic of philosophy itself. Wittgenstein’s answer, which is the
foundation of his conception of philosophy, is probably the most
disconcerting aspect of his works and maybe the most radical of all
answers ever given. According to Wittgenstein, philosophers should not try
to attain established theories because philosophical problems are only
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ). Ontos
Verlag, Frankfurt a. M., 2010, 7-10.
8 Wittgenstein: Issues and Debates
Introduction
In this paper, I try to estimate the importance and relevance of a
wittgensteinian legacy (themes, problems and perhaps solutions) in recent
philosophy of social sciences. The general idea underlying my talk is that
this legacy is never as visible as when the concept of understanding is
beeing questioned. In other words, my purpose here is to examine the
grammar of this concept ; however, Wittgenstein’s remarks on this topic
are scattered in many different places and difficult to catch in one single
grasp and that’s the reason why I shall try to apprehend them through their
consequences for social sciences.
During the last decades, a shared point of view (foreign to
Wittgenstein’s influence and resulting in fact from debates internal to the
social sciences) has consisted in saying that the structures govern the
actions and are, at the same time, generated by the actions. Different
versions of this point of view can be seen in the works of Pierre Bourdieu,
Anthony Giddens and many others.1 I wish to label this theoretical schema
the double status of structures and to question its relevance. The main
difficulty, in so doing, comes from the fact that the concept of structure has
been intrepreted, with the passing times, in very different ways. So, one
need first to clarify its use. As I suscribe to a form of methodological
separatism between social an natural sciences, one way to undertake this
clarification is to make explicit the differences in the way the concept is
used in the two domains. When one speak of a physical structure (the
structure of a water molecule, for instance), one has normally in mind a
system of objective relations which can’t be seen by the lay man: to get to
know such a structure is one and the same thing as to get to know the
molecule. In the social sciences, on the other hand, the concept of structure
is applied to a universe filled with agents who have elaborated their own
understanding of it and act in accordance with this understanding. In
addition, the understanding of his environment by the social agent, apart
1
See, for instance: Bourdieu, 1979; Bourdieu, 1987; Giddens, 1979; Giddens, 1986.
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ). Ontos
Verlag, Frankfurt a. M., 2010, 11-30.
12 Wittgenstein: Issues and Debates
from the fact it has its own legitimacy, is very well exercised without the
help of the stuctural explanation of the same topic a sociologist can supply.
That’s why it’s simply impossible to identify the knowledge of the social
world with the knowledge of such an explanation: on the contrary, the
concept of stucture, when applied to the social world, has to be combined
with an account of the agents’ spontaneous understanding and of the rules
they follow.
Moreover, in the natural sciences one might legitimitely waver
between a realistic interpretation of structures, conceiving them as spatio-
temporal realities, as the framework (endowed with causal powers) of the
observed phenomenas and an anti-realistic interpretation, seeing them as
abstract characteristics. One of the aspects of the thesis defended in this
paper consists in saying that, in the social sciences, it’s almost impossible
to substantiate the realistic interpretation because it doesn’t fit well the idea
that agents have intentions and understand their own actions. In other
words, I do agree with Giddens when he insists that structures are abstract
features. Let us take a linguistic example. The present state of a language
(English, French) can be refered to as a structure: such a characterization
implies that, in the description of the linguistic material, one doesn’t take
into account the idiosyncrasies, the tendency of a few speakers to set
themself apart or even to secede, in order to concentrate on the
convergences and the big macroscopic regularities that appear as the
balances of a language’s life. The speaker’s largely implicit linguisitic
capacity can then be described as a rather confused knowledge of those
regularities and his practice as their continuation. Such a continuation
supposes the speaker to have, through his tacit linguistic knowledge, an
access to the normativity of the linguistic institution.
So this paper will be largely dedicated to the examination of this
access to social normativity and that’s why I’ll not be able to save myself a
grammatical examination of the concept of understanding: it’s both
important and illuminating to specify what is meant by phrases saying that
an agent understands a rule or a social norm. This is also the point where
our argumentation will separate from Giddens’, both because his exegesis
of Wittgenstein’s thought on rule is inaccurate and because it inclines him
to a misleading interpretation of what an implicit rule is supposed to be.
The general idea I’ll defend on that topic will be that one needs to free
oneself from the belief that an implicit rule is a rule which is subject to an
implicit grasp: normally, a rule is implicit precisely because it doesn’t have
to be grasped anymore. The regularities we alluded to need, in order to be
Tacit knowledge and Action 13
1. Giddens on rules
In a chapter of his book Social theory and modern sociology, Giddens
notices that the concept of structure has elicited very few comments from
the english speaking sociologists and philosophers (in this respect, the fate
of the concept of function has been quite different). He comments upon
that point in the following way:
The reason is probably that most English-speaking social scientists have a clear
idea of how the concept of structure should be understood. When they talk of
structure, or of the “structural properties of institutions”, they have in mind a sort
of visual analogy. They see the structural properties of institutions as like the
girders of a building, or the anatomy of a body.2
In such a perspective, the structures appear as patterns or types of
relations one can oberve in different social contexts and the concept of
structure seems to be naturally connected with what has been called the
“objectivist” approach in social sciences. As Giddens emphasizes, the
difficulty raised by such a notion is that “the structure then appears as a
constraint which is ‘external’ to action.” 3 Giddens’ own conception of
structures is set out as an alternative to the interpretation in terms of
constraint. But precisely because this is his aim, his argument also has to
include the idea that, unlike the patterns we have just been speaking of, the
structures are abstract features with no location in space and time.
So, according to this last interpretation (1) societies, institutions have
abstract structural proprerties resulting from the ongoing of human action
(2) agents are able to proceed in their everyday business in virtue of their
capability to instantiate such structural properties. In other words, those
properties have no action on the agents (unlike the wind who has an action
on the tree to the point of bending it): on the contrary, they are abstract
descriptions of the result of convergent “upstream” actions by numerous
agents. But they are perpetuated only because those very same agents act
spontaneously in a rule-guided way. This is why the agents, seen
2
See: Giddens, 1984, pp. 52-72 and especially p. 60.
3
Giddens, 1984, p. 61.
14 Wittgenstein: Issues and Debates
4
As one knows, Peter Winch’s controversial essay (Winch, 1958) has become a
wittgensteinian locus classicus on the erklären/verstehen distinction. Winch defends a
form of methodological separatism partly on the basis of arguments inspired by the
author of the Philosophical Investigations (Winch’s other main source of inspiration,
as one can guess from the title of his book, is Collingwood’s Idea of history). Another
important essay by Winch on kindred topics is “Understanding a primitive society” in:
Winch, 1964.
Tacit knowledge and Action 15
5
See: Schatzki, 1992, p. 280-295.
6
Giddens, 1984, p. 65.
7
Giddens, 1984, p. 66.
16 Wittgenstein: Issues and Debates
falling apple in the same way ; it belongs rather to the physicist’s explanation
of the apple behavior.”8
This famous passage is both convincing and unsatisfactory. The
unsatisfactory point is that it seems to suggest that one must be able to
apply the concept of war to oneself in order to participate in a war. It’s
perfectly clear, indeed, that a soldier wearing a uniform would necessarily
justify some, at least, of his actions by relating them to this concept, but a
situation might easily be imagined that one would feel inclined to describe
“from the outside” as a war situation even if the belligerent agents showed
reluctance to describe themselves as warriors. So, the exact idea seems to
be that when a relation does exist between the concept of a member of a
belligerent country and the behavior of a combatant, then this relation is
internal (in an event such as an apple fall, on the contrary, we are
confronted to an external relation between the fall and the concept of
gravitation).9 Winch has such an internal relation in mind when he makes
use of the phrase behavior governed by a concept. What his argument
eventually shows, is that if you are a member of a belligerent country and
have the corresponding concept at your disposal, then this concept
essentially participates in your behavior.
Giddens draws from the previous considerations the consequence
that the notion of a mutual knowledge should be extended to the relation
8
Cf.: Winch, 1958, pp. 127-128.
9
The notion of internal relation is of an utmost importance here and requires
explanation.Let us take an elementary psychological example. Wittgenstein’s idea
consists (unlike the mentalist’s) in saying that pain is not only a subjective event and
(unlike the behaviorist’s) in saying that it can’t be identified with the pain behavior:
normally, the pain behavior is part of the pain, in other words to have pain expressions
is part of what is meant by the phrase to be in pain. Accordingly, the pain behavior is a
criteria of pain and the relation between the two is internal. Of course, counter-
examples are always imaginable (a feigned pain played by a pretender ; a stoïcal guy
curbing all his pain expressions), but without such a criteria, the learning of the
concept of pain would simply be impossible. Winch invites us to establish a similar
connexion (though more complex) between x’s behavior as a soldier and x’s
possession of the concept of a soldier: x’s behavior as a soldier is the criteria that
leads us to attribute him not only the concept of a soldier but the fact that he applies to
himself this very same concept and justify himself with its help. This relation of
justification, in turn, leads us to the conclusion that the concept of a soldier is part of
the soldier’s behavior and, in more general terms, to the idea that having the concept of
a soldier is part of what it is to be a soldier.
Tacit knowledge and Action 17
between the agent and the observer, but the way he puts things remains
somewhat vague.
“... in order to generate valid descriptions of social life, the sociological
observer must employ the same elements of mutual knowledge used by the
participants ...”10
The difficulties are indeed concentrated in the phrase “the same
elements of mutual knowledge” which has to be taken cum grano salis. In
fact, neither Winch, nor Giddens really think that sociologists are
committed to limit themselves, in the course of doing their job, to concepts
that social agents employ on their own initiative. In order to figure this out,
the best thing to two is to pick out two examples from Winch. The first of
those two examples illustrates the situation in which the social actor and
the observer do share the very same concept ; the second one illustrates the
situation in which the actor’s behavior shows his mastery of concept(s)
connected to the one (or those) used by the observer. When confronted to
such a case, one is forced to specify the nature of this connection. Example
1: an observer says of agent N that he has voted for the Labor Party
because he thought it would be able to preserve the industrial peace. The
strength of such an explanation lies in the fact that the concepts it uses
have to be grasped by N as well.11 Example 2: an observer explains the
behavior of a group of businessmen by using the keynesian concept of
liquidity preference. Such a concept doesn’t belong to the set of cognitive
tools mastered by those businessmen ; however, Winch notices,
“... it is logically tied to concepts which do enter into business activity, for its
use by the economist presupposes his understanding of what it is to conduct a
business, which in turn involves an understanding of such business concepts as
money, profit, cost, risk etc. It is only the relation between his account and
these concepts which makes it an account of economic activity as opposed,
say, to a piece of theology.”12
Those two examples are set out in two different sections of the book
and illustrate in two different ways the idea according to which the agent
and the observer share the same elements of mutual knowledge. As far as I
know, very few commentators have paid attention to this difference. Its
significance will become clear if we start from the following remark: if the
explanation by the observer of the agent’s behavior with the help of a
concept is to be valid, it’s not enough for the agent to have the capacity to
10
Giddens, 1984, p. 66.
11
See: Winch, 1958, p. 46.
12
Winch, 1958, p. 89.
18 Wittgenstein: Issues and Debates
13
The liquidity preference, as one knows, is inversely proportional to the interest rate.
If interest rates are low, the economic agents, expecting better days, will keep their
money and, accordingly, the liquidity preference will be maximum ; on the contrary, if
interest rates becomes high, the liquidity preference diminishes because making
investments becomes the best thing to do with one’s money .
14
This situation is, of course, quite different from the case of a discussion partner
helping a speaker to find a word he has on the tip of his tongue. In such a case, the
speaker does have the capability to use the word he is looking for, but this capability is
temporarily impeded by an external factor (tiredness, absent-mindedness etc.).
15
Some of the most important reflexions on the notion of awoval in the context of
Wittgenstein’s thought can be found in his Bemerkungen über die Philosophie der
Psychologie, see for instance Wittgenstein, 1984, vol. 1, § 836 and vol. 2, § 63. His
idea, roughly speaking, is that we are not, as far as our own mental states are
concerned, in the position of an very privileged witness: we do not get an information
Tacit knowledge and Action 19
justified to sum up, with the help of a single concept, what the agents
express by using a plurality of concepts. It’s also worth noticing that the
wittgensteinian concept of internal relation plays indeed to different roles
in the previous observations. First, it occurs between the agent’s behavior
and the concept said to be “essential” to it ; secondly it occurs between the
concept used by the agent and the concept used by the observer. In this
second case, we are confronted to the agent’s propension to employ such a
concept (when asked to jusfify his demeanour) rather than to his capability
to do so. In fact, the word “propension” is appropriate here because
without an external help, the agent remains unable to achieve a full
mastery of a concept which happens to be used, in other respects, to
describe his deeds.
Obviously, a speaker who only has the propension to adopt a way of
speaking doesn’t necessarily have its full mastery. He might need, to say
the least, a little help to gain this full mastery, and such an help might well
never occur. The small conversational dramas Pierre Bourdieu is prone to
describe very often involve a divorce between the agent’s own capability
to speak in a certain way and his propension (in specific social situations)
to speak in another way he takes to be more gratifying.16 Unfortunately,
Bourdieu has obscured this interesting observation by labelling habitus
sometimes the speaker’s competence to talk in a certain way, sometimes
his propension to adopt norms of expression he cannot fit and sometimes
the two things simultaneously. So far, the conclusion of the story seems to
be that the assertion according to which a sociological explanation has to
entertain an internal relation with the one the agent could give doesn’t
correspond to an unique kind of situation because the modal verb “can” in
the present context, covers different kinds of situations. The previous
conclusion is a good reason to give a maximum extension to the concept of
(which could happen to be false) about our own mind’s content. This is where his idea
that what he calls psychological verbs (to belief, to think, to understand etc.) in the
singular, first person, present tense are expressions (Ausserung, Ausdrück) comes
from. He also thinks that those very same verbs have a descriptive use as soon as they
are employed at other tenses and / or other persons. In other words, the avowal, by an
agent, of his reason to act in a certain way or to say a certain thing is not the result of a
cognitive contact he entertains with his own mind, but an expression. Becoming aware
that one’s own acts proceed from a certain reason has nothing to do with forming an
hypothese. The concept of awoval is, of course, also a central feature in Wittgenstein’s
reflections on Freud. See Wittgenstein, 1972.
16
See: Bourdieu, 1980. For a full examination of the relations between capacity,
capability and propensity, see; Scheffler, 1983.
20 Wittgenstein: Issues and Debates
17
Sociologists interpret a world which already contains interpretations.
18
Most probably, the reconciliation with common sense wished by the thinkers we
have been discussing cannot be summarized by the idea that social scientists have to
get back to concepts the agents master, or whose explanatory use they might approve.
A vast range of investigations in recent cognitive psychology is dedicated to the task
of explaining the efficiency and rapidity of common sense thought. In the course of
doing so, those investigations tap notions such as mental model or prototype,
sometimes echoing Wittgenstein’s own reflections on the blurredness of numerous
everyday concepts which prove to be unfit for definitions in terms of necessary and
sufficient conditions (one might think, for instance, of Eleanor Rosch’s seminal work
on categorization). It’s clear that prototype is anhypothetical concept (= a prototype is
attributed to a speaker on the basis of tests he is subjected to). When A explains B ‘s
demeanour by saying that a prototype structures its thought, A doesn’t normally expect
B’s approval (unlike the economist in the liquidity preference example). B might
eventually say that A hypothesis is insightfull, but such an approval is not an avowal:
first person authority is relevant in the businessmen case, not in B’s case. The moral of
this story seems to be that one needs to distinguish the descriptions of the common
world mobilizing elements from our mutual knowledge and descriptions of the
common world which are, indeed, theoretical descriptions of this knowledge.
Obviously, the concept of prototype has to do with the second kind of undertaking.
19
Cf.: Wittgenstein, PU, 1984, § 206.
Tacit knowledge and Action 21
and its application in particular cases. The agent receives the rule just like
he receives an order, at least if he isn’t already in possession of the text of
the rule, and is supposed to “translate” this text into a determinate action.
The idea, cherished by Wittgenstein, that following a rule is a practice
contrasts with the hypothesis of an subjective act of interpretation, thought
as responsible for the transition between the pronouncement of the rule and
its application, and supposed to be the place where the criterium separating
correct application from infringement is located. If someone suscribes to
such a subjectivist mythology, he finds himself deprived of all means to
distinguish an effective application of the rule from the mere thought that
he is following the rule. The second, and less noticed idea on this very
same topic consists in the recognition of the fact that, in most cases, the
rules are implicit in our practice: they remain unformulated. In this respect,
the problem is not the one raised by the transition between the general rule
and a particular action, but the question of the explicitation. In other words,
how does the transition occurs between a rule inscribed in the agent’s tacit
knowledge, of which he doesn’t currently think (and of which he might
even never have effectively thought of) and the very same rule verbalized
(or even interpreted) ?
Giddens is prone to assert that most rules are unformulated, but he
also maintains that every formulation of the rule is an interpretation. 20
When he says that social agents “draw on rules”, he certainly doesn’t mean
that they follow, consciously or unconsciously, already formulated rules.21
He wants to say that their actions are governed by a “tacit grasp” of
unformulated rules. The word “grasp” seems to denote an actuality, a
psychological act or event -and this should alert us. Should not we say, on
the contrary, that an implicit rule is not subject to any grasp instead of
saying it’s subject to an implicit one ? The distinctive feature of a familiar
rule is that it does not requires from the applicator, when he proceeds to a
new application, an act of recognition anymore, though the way he acts
gives us evidence that he has knowledge of the rule. The object of my
argumentation in the rest of this paper can be deduced from the content of
the previous remarks: I shall try to show that it is impossible to give any
substantial content to the idea of a “tacit grasp” as it is used by Giddens.
20
See: Giddens, 1986, chap. 1.
21
See on this point Schatzki’s remarks (Schatzki, 1992, p. 291).
22 Wittgenstein: Issues and Debates
order to act according to it. The first of those two interpretations is chronic
(= understanding is an act or an event entertaining a temporal relation with
action) ; the second interpretation is grammatical. The first interpretation
sees understanding as an event numerically distinct from the action. On the
contrary, the second interpretation sees understanding as part of the action.
This is eventually what was involved in the industrial peace example:
Winch’s idea was certainly not to say that this notion, essential to the
agent’s behavior, was to cross his mind at the moment he casts his vote.
One way to draw the consequences of the second interpretation is to notice
that the word “actualization” in the phrase actualization of a capacity
might in turn be taken in to differents ways. (1) A capacity (such as the
understanding a concept) can be actualized in the Erlebnis of the speaker
(as when one sees a halo of connexions between the word expressing the
concept and other elements of our linguistic universe and accordingly
comes to see this same word under an aspect). 26 (2) A capacity can be
actualized in the agent’s behavior, for instance through the fact that the
speaker uses an expression in an appropriate way. Obviously, the linguistic
behavior is, to some extent, part of the subject’s Erlebnis, but the important
point, in this last case, is that no conscious state, no mental event
intervenes, strictly speaking, in the understanding. And that’s why the only
thing to say is that the understanding is part of the action. It must be noted
that on a few occasions, Wittgenstein has said that the verb to understand
could be used to mean a mental event (as when one speaks of a sudden
insight), but this illustrates in the first place the vagueness of the concept.
Anyway, the extension of this concept, taken as a whole, has not to be
modelled on this specific portion. 27 The conclusion is that, most of the
time, the understanding of a proposition according to which one acts is not
an event concomitant with the action, and we eventually are to think of
form of understanding which is not a grasp and this is where the notion of
disposition enters the scene.28 In this perspective, the main criticism to be
adressed to Giddens is not the imprecision of his reading of Wittgenstein,
but the fact that he invents a syncretic mixture of mental tokens and
dispositions. But even the recognition of the dispositional nature of the
26
Cf.: Wittgenstein, BPP, 1984, vol. 1, § 192-202.
27
See: Wittgenstein, The Cambridge Lectures 32-35, 1979, p. 114.
28
One must remember that the word disposition is to be understood in two different
ways. Some dispositions are such as their actualization is automatic (like beeing
irritable). Others need to be exercized (our knowledge of elementary arithmetics, for
instance). Dispositions who need to be exercized are often called capacities.
24 Wittgenstein: Issues and Debates
rule either to the idea of a formula supposed to cover all the possible cases
or to the justification after the event. By so doing, he occults (1) that what
he calls “concept” has its place inside the spectrum of different things we
call “rules” (2) that the dispositional notion of implicit rule here mobilized
is the best safeguard one can find against the conceptual syncretism
illustrated by the notion of tacit grasp.
Blur-edged concepts are windows opened on what we try to grasp
through the notion of “implicit rule”. More precisely, if one pays attention
to everyday’s langage’s life, one can easily see that a concept can be used
by speakers who prove unable to interpret it (= to establish synonymy
relations in which this concept enters, to express what it “means”).32 The
speaker might well have the know-how allowing him to apply correctly a
concept and be deprived of the power to give the rules justifying this very
same application.33 The rule, in such a case, is implicit not in the (weak)
sense of being expressible though not expressed, but in the (strong) sense
of not being expressed because the speaker is deprived of the power to
express it. X can apply correctly the concept C 1 and the concept C 2, and
can be, at the same time, unable to give the rule R distinguishing the
contexts in which C 1 can be used and not C 2 and vice versa. Advancing
such a rule would be interpreting C 1. X’s cognitive performances here are
in relation with R, but does not proceed by an application of R. John Searle
has noticed that many institutional rules are like the one distinguishing C 1
from C 2. He explains that we are endowed with abilities which make us
sensitive to the normativity inherent in the institutions.34 However, those
abilities are not on an equal footing with conceptual capacities we have
been previously talked about. The main reason is that the practical
knowledge Searle is dealing with has to be figured out as functionally
equivalent to complete statements, permitting of forbiding things. In
32
I agree with Quine’s idea that in asserting what a term “means”, one doesn’t refer to
an entity: there are no substantive meanings. In fact, when I explain to someone what a
word or a phrase means, I reformulate it in terms I think to be more accessible to my
listener. Between the two expressions a sameness of meaning does exist, but this does
not mean, to repeat myself, that they express the same abstract entity or share the same
attribute. See: White, 2002, chap. 6 (for an illuminating state of the art on this topic)
and Quine, 1953, pp. 1-19 (especially p. 11).
33
The distinction between tun and können is recurrent in Wittgenstein’s remarks on
psychology. For an application of this distinction to calculation, see: Wittgenstein,
BPP, 1984, vol. 1 §§ 651, 655, 735. On being aware of one’s lie as a können, see §
263.
34
See: Searle, 1996, chap. 6.
26 Wittgenstein: Issues and Debates
between routine rules and the so-called strategical rules35 is much more
relevant than Searle’s pseudo-distinction between constitutive and
regulative rules.36 One could say of someone, if it came to it, that he is
good in his knowledge of chess rules or in buttoning up his uniform in the
army, but no one would say of an agent that he is excellent or brilliant in
such activities. Such predicates might, without hesitation, be applied to an
insightful players, equiped with high tactical capacities. Unlike strategical
rules, routine rules are acquired through a process of repetition (of the
same sequence of actions) ; the application of strategical rules involves the
consideration of the quality of the deeds, and their mastery requires, from a
player for instance, that he participates in a great number of games with
different players.
In The constitution of society, Giddens explains that it’s important to
distinguish, conceptually speaking, two different dimensions of the rules:
the one having to do with the constitution of meaning, on the one hand ;
the one having to do with the sanction of social conduct, on the other hand.
He seems to think that Winch has proved unable to make such a
distinction. Giddens has in mind, when he deals with this theme, the
difference between constitutive and regulative rules, and his goal is to
show that it can’t be understood as a difference between types of rules.
That’s why he insists that such a distinction must be thought as conceptual.
The rule defining the checkmate is clearly “constitutive” (by so doing it
contributes to the definition of what a chess game is) and the rule forcing
the employees of a company to clock in is regulative. Nevertheless, he
notices, the first one can also be seen, in some respect, as regulative (it
refers to an element of the game which is supposed to be respected and
involves the possibility of a sanction) and the second one as constitutive (it
doesn’t defines work as such but enters in the definition of more
circumscribed phenomenas). If the difference is conceptual indeed (and I
35
The predicate “strategical” is used here because the application of those rules
involves a judgment. See on this point: Scheffler, 1965, chap. 5. Scheffler distinction
concerns types of skills rather than types of rules, but it’s tempting to extend his
perspective.
36
See: Searle, 1969, p. 34-35. The so-called constitutive rules are supposed to define a
specific game. This means, among other things, that the physical appearance of a
checker doesn’t matter at all (it might be a trouser button), the essential thing being the
rules governing the moves. Breaking a constitutive rule is either committing an
infringement or practising a game other than the one apparently played. Violating a
regulative rule is normallly no infringement.
28 Wittgenstein: Issues and Debates
think Giddens is perfectly right on that point), the sinn he accuses Winch
of is a venial one. Anyway, Winch probably thought that rules might fulfil
numerous different roles, the previous distinction being between two of
those many roles. In addition to that, there is no sufficient textual evidence
that this distinction, no matter how it is understood, exists as such in
Wittgenstein’s writings.
The difference between routine rules and strategical rules, on the
contrary, is to be interpreted as a real one (no rules can be both routine and
strategical at the same time) and this assertion sheds light on what is meant
by the observation according to which some rules are such that the agent
doesn’t need to want their application when he follows them and on the
extension to be attached to the word to understand. This verb might denote
a posture involving the exercise of judgment but might also denote the
mastery of a routine. Following a routine rule to such an extend that it is so
to speak forgotten is not to be identified with an action deprived of any
comprehension. In that respect, the agent applying a protocol without any
judgment remains different from the thermostat turning up the heat when
the outside temperature drops.
Conclusions
What are we left with after this process of decantation? We have
propensions orienting the agent’s actions in our social furniture and
capabilities making him able to know what he is to do in contexts that can
well be unexpected. We have understood that some of those capabilities
are know-how while others have a more complex structure and include the
capacity to give one’s justifications for what one has done. In addition to
that, we have taken into account routines we have identified with a second
nature. In the ongoing of human life we see that is involved a mixture of
propensions and capacities. Understanding is on the side of capacities.
Many of those capacities are blur-edged (as we have seen through our
discussion of concepts) and the formulation of their content seems to root
them to the spot, except that the application of a formulated rule, as
Wittgenstein says, always leaves ways-out open. Even if rules are really
abstract features, it still makes sense to say that the agent’s actions
perpetuates them: this only means that a different stretching of actions
would have instantiated different abstract features. So the double status is
not really a problem.
Tacit knowledge and Action 29
Bibliography
Bourdieu, P., 1979: Le sens pratique, Minuit, Paris.
--- 1987: Choses dites, Minuit, Paris.
--- 1980: Ce que parler veut dire, Fayard, Paris.
Collingwood, R. G., 1956: Ithe Idea of history, Oxford University Press,
Oxford.
Giddens, A., 1979: Central problems in social theory, Macmillan, London.
--- 1984: The social sciences and philosophy”, in: Social theory and
modern sociology, Stanford University Press, Stanford.
--- 1986: The constitution of society, second edition, Polity Press,
Cambridge.
Quine, W.V.O., 1953: On what there is”, in: From a logical point of view,
Harvard University Press, Cambridge Mass., pp. 1-19.
Schatzki, Th., 1992: Do social structures govern actions?, Midwest studies
in philosophy, vol. 17, pp. 280-295.
Scheffler, I., 1965: Conditions of knowledge, Scott & Foresman, Glenview.
--- 1983: Of human potential, Routledge, London.
Searle, J.R., 1969: Speech acts, Cambridge University Press, Cambridge.
--- 1996: The construction of social reality, Penguin Books, London.
White, M., 2002: A philosophy of culture: the scope of holistic
pragmatism, Princeton University Press, Princeton.
30 Wittgenstein: Issues and Debates
Winch, P., 1958: The idea of a social science and its relation to
philosophy, Routledge, London.
--- 1964: Understanding a primitive society, American Philosophical
Quaterly, vol. 1, 307-324.
Wittgenstein, L., 1958: The blue book & the brown book, Blackwell,
Oxford.
--- 1972: Conversation on Freud, in: Lectures & Conversations,
University of California Press, Los Angeles.
--- 1979: The Cambridge Lectures 32-35, Blackwell, Oxford.
--- 1984: Philosophische Untersuchungen, Werkausgabe, Band 1,
Suhrkamp, Frankfurt a.M.
--- 1984: Philosophische Grammatik, Werkausgabe, Band 4, Suhrkamp,
Frankfurt a.M.
--- 1984: Bemerkungen über die Philosophie der Psychologie,
Werkausgabe, Band 7, Suhrkamp, Frankfurt a.M.
Synoptic Views vs. Primal Phenomena:
Wittgenstein on Goethe’s Morphology
Sabine PLAUD
1
Wittgenstein, 2001, PI, § 109.
2
Wittgenstein, 2001, PI, § 124.
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ).
Ontos Verlag, Frankfurt a. M., 2010, 31-46.
32 Wittgenstein: Issues and Debates
55
For further developments on Goethe’s influence on Wittgenstein, see: Nordmann,
2003, 91-110, and Genova, 1995.
34 Wittgenstein: Issues and Debates
Such a botanical project is thus not only a scientific project, but also
a romantic one: by its endeavor to bring back all the particular plants to a
single, primal one, its purpose is to exhibit the synthesis of unity and
variety displayed by the “beautiful garden of the universe”.
8
Goethe, 1790, §4.
9
Goethe, 1982, §175.
36 Wittgenstein: Issues and Debates
turn essentially connected with a third one: namely with the notion of a
“primordial picture” or “paradigm” (Urbild). This idea of a “primordial
picture” means in fact two things. On the one hand, the Urbild amounts to
the general pattern out of which plants and animals happen to be formed10;
in this sense, the notion of an Urbild is almost identical to Goethe’s
Urphänomen. However, in a broader sense, the Urbild can also be read as
the representation science is likely to give of such a pattern. In his
Writings on Morphology, Goethe claims for example that the main purpose
of natural science is to:
present this primordial picture, if not to senses themselves, at least to the mind;
and to elaborate our descriptions on its basis, as it were a norm.11
In other words, the very purpose of natural sciences is to set forth a
clear view (an Urbild) of the primal phenomenon that pertains to each
particular field of nature. Only thus can they set the natural phenomena in
order.
10
Cf. Goethe, 1962, “Hefte zur Morphologie”, Erster Band, p. 472.
11
Goethe, 1962, “Entwürfe zu einem osteologischen Typus”, p. 270.
12
Goethe, 1962, “Hefte zur Morphologie”, Erster Band, Introduction.
Synoptic Views vs. Primal Phenomena: Wittgenstein on Goethe’s Morphology 37
But when even such a primordial phenomenon is arrived at, the evil still is that
we refuse to recognize it as such, that we still aim at something beyond,
although it would become us to confess that we are arrived at the limits of
experimental knowledge. Let the observer of nature suffer the primordial
phenomenon to remain undisturbed in its beauty.13
Yet, Goethe’s morphological method is not devoid of ambiguity: for
despite this vindicated phenomenism, one may regard his very endeavor
towards origins and primacy as a mere transgression of the phenomenist
principle he assumes. I will now examine how such a reproach was
formulated by Wittgenstein.
13
Goethe, 1982, § 177.
14
Wittgenstein, 2001, PI, § 115.
38 Wittgenstein: Issues and Debates
15
Wittgenstein, 1980, CV, p. 14.
16
Wittgenstein, 1977, § 230.
Synoptic Views vs. Primal Phenomena: Wittgenstein on Goethe’s Morphology 39
17
Wittgenstein, 2001, PI, § 122.
18
Wittgenstein, 2001, PI, § 122.
19
Wittgenstein, 2001, PI, § 144.
20
Wittgenstein, 2001, PI, § 126.
40 Wittgenstein: Issues and Debates
Consider for example the proceedings that we call “games”. (…) – [I]f you
look at them, you will not see something that is common to all, but similarities,
relationships, and a whole series of them at that. To repeat: don’t think, but
look!21
There again, such an exhibition of the phenomena in their varieties could
certainly be described as a typically “morphological” one, and regarded as an
echo to Goethe’s search for connections among living organisms.22
4. Lastly, I have introduced above the idea that Goethe’s project was to
lead one to see unity in variety. Just the same goes with
Wittgenstein’s method, since his work itself is nothing like a
demonstrative speech, but consist of a series of sketches whose unity
has to be drawn afterwards:
After several unsuccessful attempts to weld my results together into such a
whole, I realized that I should never succeed. (…) – The remarks of this book
are, as it were, a number of sketches of landscapes which were made in the
course of these long and involved journeyings. (…) Thus this book is really
only an album.23
21
Wittgenstein, 2001, PI, § 66.
22
See also: “A perspicuous representation produces just that understanding which
consists in “seeing connections”. Hence the importance of finding and inventing
intermediate cases”. Wittgenstein, 2001, PI, § 66.
23
Wittgenstein, 2001, PI, Preface.
Synoptic Views vs. Primal Phenomena: Wittgenstein on Goethe’s Morphology 41
24
Goethe, 1992, Italian Journey, April 17th, 1787.
25
Goethe, 1992, Italian Journey, May 17th, 1787.
42 Wittgenstein: Issues and Debates
26
Wittgenstein, 1979, RFGB, p. 8.
27
Wittgenstein, 2003, Ch. 3, p. 311 (my emphasis).
44 Wittgenstein: Issues and Debates
From Goethe stems the conception of the ‘primal plant’; yet surely he saw in it
only an idea, not something real. (…) Goethe’s aphorism ‘All the organs of
plants are leaves transformed’ gives us a schema by which we group the organs
of plants according to their degrees of similarity, as it were around a central
case.28
In those conditions, what Wittgenstein does himself is indeed quite
similar to what Goethe does in the realm of botany, except that he does it
in the realm of language. Goethe’s purpose was in fact to view the leaf “in
its natural surroundings of forms” 29 ; Wittgenstein’s purpose is to view
social practices in their linguistic surroundings of forms:
And this is exactly what we are doing: we situate a linguistic form in its
surroundings, we see the grammar of our language against a background of
similar and related games, and that banishes disquiet.30
Such a linguistic turn of Goethe’s morphology will obviously be
achieved in his later philosophy with the so-called method of language-
games. The method of language-games is introduced by Wittgenstein in his
Blue Book31, and becomes a key device of the Philosophical Investigations.
In a broad acception, a language-game can in fact be defined as a
determinate set of linguistic and social practices. But in a more specific
acception, a language-game is also a theoretical and primitive
reconstruction of linguistic practices, whose function is to bring clarity
among the complexity of our real performances. This is what Wittgenstein
means when he claims that he “will sometimes speak of a primitive
language as a language-game” 32 . His language-games thus provide the
reader with primitive and over-simplified representations of linguistic
practices, as may be adopted, for example, by children or primal tribes33.
Yet, such a primacy has to be devoid of any psychological, historical or
anthropological purport; for this is the condition that, alone, may guarantee
the legitimacy of this search for linguistic primal phenomena:
Our mistake is to look for an explanation where we ought to look at what
happens as a ‘proto-phenomenon’. That is, where we ought to have said: this
language-game is played.34
28
Wittgenstein, 2003, Ch. 3, p. 311,
29
Wittgenstein, 2003, Ch. 3, p. 311.
30
Wittgenstein, 2003, Ch. 3, p. 311.
31
Cf. Wittgenstein, 1958, BBB, p. 18.
32
Wittgenstein, 2001, PI, § 7.
33
On these issues, see: Hilmy, 1987.
34
Wittgenstein, 2001, PI, § 656.
Synoptic Views vs. Primal Phenomena: Wittgenstein on Goethe’s Morphology 45
Conclusion
In a passage of his Farbenlehre, Goethe formulates the following judgment
on his own achievements:
We believe we merit thanks from the philosopher for having traced phenomena
to their origins (…). Further, it will be gratifying to him that we have arranged
the appearances in an easily surveyed order, even should he not altogether
approve of the arrangement itself.35
What I have tried to show in this paper is that, although Wittgenstein
was certainly a philosopher, he did not always give Goethe the thanks he
merited for having traced phenomena to their origins. Nevertheless, it was
certainly very gratifying to him to have had the appeareances arranged in
an easily surveyed order; so gratifying indeed that he made of this easily
surveyed order the key to his own method, and that he consequently tried
to reevaluate Goethe’s approach of primal phenomena so as to make it
compatible with his own conceptions.
Bibliography
Genova, J., 1995: Wittgenstein, A Way of Seeing. Routledge, New York.
Goethe, Johann Wolfgang von, 1790: Versuch die Metamorphose der
Pflanzen zu erklären. Carl Wilhelm Ettinger, Gotha.
Goethe, Johann Wolfgang von, 1962: Schriften zur Morphologie. “Hefte
zur Morphologie”, Erster Band, (Ed. D. Kuhn). Deutscher Klassiker
Verlag, Frankfurt a.M.
Goethe, Johann Wolfgang von, 1982: Theory of Colours. (Tras. Charles
Lock Eastlake). The M.I.T. Press, Cambridge, Massachusetts.
Goethe, Johann Wolfgang von, 1992: Italian Journey. (trans. W.H. Auden
and Elizabeth Mayer), Penguin, London.
Hilmy, S., 1987: The Later Wittgenstein : The Emergence of a New
Philosophical Method. Blackwell, London.
35
Goethe, 1982, Introduction.
46 Wittgenstein: Issues and Debates
Nordmann, A., 2003: “’I have changed his way of seeing’: Goethe,
Lichtenberg and Wittgenstein”, in: F. Breipthaut, R. Raatzsch And B.
Kremberg (ed.): Goethe and Wittgenstein. Seeing the World’s Unity
in its Variety, Wittgenstein-Studien, 5, Lang, Frankfurt a.M., 91-110.
Spengler, Oswald, 1991: The Decline of the West. (Ed. Arthur Helps, and
Helmut Werner. Tr. Charles F. Atkinson). Oxford UP, New York.
Wittgenstein, Ludwig, 1958: The Blue & Brown Books: Preliminary
Studies for the Philosophical Investigations. Blackwell, Oxford.
Wittgenstein, Ludwig, 1979: Remarks on Frazer’s Golden Bough. (Ed. R.
Rhees, trans. A. C. Miles). Brymill, Humanities Press, Atlantic
Highlands, NJ.
Wittgenstein, Ludwig, 1980: Culture and Value. (Ed. G. H. von Wright, tr.
P. Winch). Blackwell, London.
Wittgenstein, Ludwig, 1977: Remarks on Colour. (Ed. G.E.M. Anscombe,
trans. L. McAlister and M. Schätle). Blackwell, London.
Wittgenstein, L., 2001: Philosophical Investigations. (Trans. G.E.M.
Anscombe). Blackwell, London.
Wittgenstein, Ludwig, 2003: Dictation for Schlick, in The Voices of
Wittgenstein. (Ed. G. Baker, trans. G. Baker, M. Mackert, J.
Connolly and V. Politis). Routledge, London.
Critical Remarks on Anti-Metaphyscial
Readings of Wittgenstein
Eric LEMAIRE
Introduction
Has Wittgenstein destroyed Metaphysics? Wittgenstein scholars often hold
an affirmative answer as if it were obvious. If it is true, it has to be made
clear that, first, Wittgenstein produced a philosophy, which cannot be
qualified as Metaphysics; and second, he really established that
metaphysical propositions are nonsensical. There are several ways to
discuss these problems. The most direct consists in examining
Wittgenstein’s texts. Another one, the one that will be followed here,
consists in examining the readings proposed by his interpreters. These
strategies are not incompatible. In a longer study, it would be possible to
combine them. We shall raise mainly two questions here. In the first place,
is there an anti-metaphysical reading of the second Wittgenstein?
Secondly, assuming that the readings, which we are going to study are true
to the second Wittgenstein’s works, do they offer a plausible defense in
favor of the thesis according to which metaphysical propositions are
nonsense? Specifically, are they consistent? Following R. Fogelin, D. Stern
proposed a general classification of the readings of Wittgenstein:
“The principal fault line separating Wittgensteinians is over a question of
philosophical method: whether or not a radical philosophical change – putting
an end to philosophy – is possible. Robert Fogelin draws a helpful distinction
between ‘Pyrrhonian’ readings of the Investigations, which see the book as
informed by a quite general skepticism about philosophy and so as aiming at
bringing philosophy to an end, and ‘Non-Pyrrhonian’ readings, which construe
the book as a critique of certain traditional theories in order to do philosophy
better.[…] Another way of putting the distinction is to say that Pyrronhian
Wittgensteinians believe philosophy, properly conducted, should not result in
any kind of theory, while Non-Pyrhonian Wittgenstenians maintains that
Wittgenstein’s criticism of traditional Philosophy leads us to a better
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ).
Ontos Verlag, Frankfurt a. M., 2010, 47-64.
48 Wittgenstein: Issues and Debates
1. What is metaphysics?
In order to know if the second Wittgenstein has showed that metaphysics is
illegitimate, we have to ask first what metaphysics is. At least two types of
definition can be offered. V. Descombes distinguishes between dogmatic
and useful definitions of Metaphysics 2 . According to him, a useful
definition of “Metaphysics” tries to be theoretically neutral. In contrast
with this, a dogmatic definition favors certain philosophical theories. It
presupposes that certain theories are true. In a dogmatic definition, the
identity of Metaphysics depends on a set of metaphysical thesis but not on
the metaphysical problems themselves. On the contrary, when we look for
a useful definition, the identity of Metaphysics depends strongly on
problems, which are studied in Metaphysics. That a dogmatic definition be
not suitable to our purpose is clear. It would indeed be self-defeating to use
a dogmatic definition to undermine the possibility of Metaphysics. It
would be as mistaken as to think that physicists overcame physics since
they abandoned Aristotle’s physics.
This first requirement can be named the principle of neutrality.
However, there is a risk of emptying the concept “Metaphysics”.
Metaphysics is a very old discipline. It was developed in quite different
ways by several people and thus has received many definitions. We have a
second requirement: the principle of distinction and unity. A good
definition has to account for the intuitive fact that Metaphysics differs from
natural sciences, economy, religion, arts, etc. These two principles are
vague. They are as vague as, e.g., an article of the Bill of rights or the
French constitution. They can be interpreted in various ways. It happens
1
Stern (2004), pp. 34-6.
2
Descombes (1995), p. 113, also see: pp. 111-9.
Critical Remarks on Anti-Metaphyscial Readings of Wittgenstein 49
that they conflict with each other. And what they order us to do is not
always obvious. According to the French constitution, freedom and
intellectual property are two examples of inalienable rights. But the
constitution does not say explicitly how to solve the problem of illegal
downloading. According to the second principle, Metaphysics,
mathematics, physics, economy and many other disciplines differ, but
nothing is said about how much they do. The vagueness of a constitutional
right is not a problem in itself. On the contrary, it is necessary a
constitution be able to face unexpected situations. When such unexpected
situations happen, it is essential that we strike a balance between
conflicting rights. Our present situation is similar.
What is Metaphysics? According to us, the best definition is the
following: Metaphysics is the study of the most general and/or central
structure of reality or of our conceptual scheme, considered as a whole.
Certain characteristics of this definition need to be clarified in order to
avoid possible misunderstandings:
1. Metaphysics is vague and open.
2. Our definition has to account for the various forms of antirealism.
3. Our definition is neutral from the point of view of the nature of the survey,
its methodology and its ambitions.
Let us briefly examine these three points. To claim that Metaphysics
is vague means that there is no sharp limit between the concepts, which it
studies and those studied in other branches of philosophy (like philosophy
of language, of arts, of religion, etc.)3. Some concepts, like “table”, “star”,
“rock”, seem far too particular to be studied in metaphysics. Others, such
as “causality”, “identity”, “necessity”, “possibility”, “concrete particular”,
“abstract entity”, “property”, “change”, “event”, “fact”, “reality”, “time”,
“space”, “possible worlds”, etc. seem to satisfy our demand of generality
and/or of centrality. Other concepts, such as “God”, “propositions”,
“meaning”, appear to be borderline cases. But if what can be said about
Metaphysics can be generally said about Philosophy, what will the
difference be between Metaphysics and the other branches of Philosophy
such as epistemology or philosophy of language, etc.? If it is true that
Metaphysics is vague, then the distinction between Metaphysics and the
other branches of Philosophy is vague too. In fact, the frontiers of
Metaphysics seem to be somewhat arbitrary and conventional: it could be
3
We found the idea that “metaphysics” is a vague concept in Van Inwagen (1998).
50 Wittgenstein: Issues and Debates
2. Hacker’s interpretation
On P. Hacker’s view, Wittgenstein has showed that philosophy, correctly
understood, differs radically from Metaphysics as it was conceived. Let us
see first what he means by “Metaphysics”:
“In metaphysics, defenders of the metaphysical enterprise conceived of true
metaphysical propositions as descriptions of necessary, essential relations
between simple natures, or Platonic Ideas, or Universals. Metaphysics was
conceived as a super-physical investigation of the most general features of the
universe and its ultimate constituents, which would yield a description of the
necessary structural features not merely of the world but of any possible world.
Unlike truths of physics, metaphysical truths were held to be descriptions not of
contingent facts about reality, but of necessities (including transcendent
necessary truths about God and the immortality of the soul). Kant’s Copernican
5
See: Lowe (1998), chapter 1, (2002) (a), chapter 1 and (b), and (2007). See also
Armstrong (1978), chapter 1 and (1997), chapter 1. We regret that we cannot develop
these points, which would need more attention.
52 Wittgenstein: Issues and Debates
6
Hacker (1996), pp. 101-2.
7
Ibid.
8
Hacker (1996), p. 119: “All talk of essences is talk of conventions, and what seems to
be the depth of the essences is actually the depth of our need for the conventions.”
9
Hacker (1996), p. 113.
Critical Remarks on Anti-Metaphyscial Readings of Wittgenstein 53
10
Hacker (1996), p. 119.
11
Hacker (1996), p. 112 and 120.
12
Hacker (2001), pp. 361-2, for further details.
13
Hacker (1996), p. 110: “If one has to choose one a single fundamental insight from
the whole corpus of Wittgenstein’s later work, it might well be argued that it should be
the insight that philosophy contributes not to human knowledge, but to human
understanding; that there is not, and cannot be, a body of established philosophical
propositions, a corpus of philosophical truths to which successive generations may add
to constitute a body of ever growing philosophical knowledge on the model of
empirical sciences.” Hacker (1996), p. 114: “Philosophy can make no claims that are
testable in experience or subject to falsification by the discovery of new facts. New
facts may lead to the formation of new scientific theories, which may in turn involve
new conceptual articulations which give rise to philosophical puzzlement (e.g. in
quantum mechanics); that is grist for the philosopher’s mill, not a verification or
falsification of anything he may legitimately assert.”
14
Hacker (1996), p. 111: “On the one hand, philosophy is characterized as a quest for
a surveyable representation of the grammar of a given problematic domain, which
enable us to find our way around when we encounter philosophical difficulties. On the
other hand, philosophy is characterized as a cure for diseases of the understanding.
These different aspects correspond to the difference between connective analysis and
therapeutic analysis […].” Hacker (1996), p. 37: “Side by side with his demolition of
philosophical illusion in logic, mathematics, and philosophy of psychology, he gives
us numerous overviews of the logical grammar of problematic concepts, painstakingly
54 Wittgenstein: Issues and Debates
tracing conceptual connections that we are all prone to overlook.[…] Providing such a
perspicuous representation of some segment of our language, elucidating the
conceptual forms and structures of some domain of human thought that is
philosophically problematic, is a positive, constructive achievement that is
complementary to the critical and destructive task of shattering philosophical illusion,
destroying philosophical mythology, and dispelling conceptual confusion.”
15
Hacker (2001), p. 37, and Hacker (2000), pp. 18-9.
16
Hacker (1996), p. 125: “The critical destructive work is counterbalanced by the
constructive account. In place of the conception of language as a calculus of rules, we
Critical Remarks on Anti-Metaphyscial Readings of Wittgenstein 55
We shall now come back to our two initial questions. Is this an anti-
metaphysical reading of the second Wittgenstein? And does this plausibly
show that metaphysical propositions are nonsense? Both of these shall
receive a negative answer. First of all, P. Hacker attributes metaphysical
believes to Wittgenstein. He develops at length Wittgenstein’s anti-
scientism17. The core idea is that mankind occupies a particular place in
nature. In other words, human beings are a kind of being different from
stones, rivers, stars, tables, computers, etc. This non reductionist view is
the negation of a metaphysical thesis (which takes various forms) about the
nature of the human beings. At least, it means that human being are not
only physical or biological mechanisms. P. Hacker would reply that this is
not a metaphysical thesis. He could insist on the fact that this is neither a
description of reality as it is in itself nor of our conceptual scheme.
However, he does not seem perfectly coherent on this. This shows it:
“It is striking that all of these characterizations, most of which are true, are
dependent upon a more fundamental feature – namely, that mankind is unique
in nature in possessing a developed language. The languages of mankind enable
us to describe the world we experience, to identify and reidentify objects in a
spatio-temporal framework and to distinguish the objects we experience from
our experiences of those objects. Knowledge of truths of reason is knowledge
of the norms of representation, and of the propositions of logic correlative to the
inferential rules, of the conceptual scheme constituted by a language.”18
3. Pyrrhonian’s interpretation
According to the pyrrhionian readers, P. Hacker misunderstands
Wittgenstein’s anti-metaphysical conception of Philosophy. They think
P.Hacker is wrong to take seriously the constructive aspects of
Wittgenstein’s Philosophy. All theses, they say, are just examples of
philosophical nonsense. These constructive aspects are only illusions.
Wittgenstein has intended to reject them too. For Pyhrronians, nothing
should remain! Their main argument against P. Hacker is stated by J.
Conant. It extends our previous remarks. We can sum it up as follows19.
19
Conant (2002), pp. 74-5. His paper was published in French: “L’idée que le but de
Wittgenstein est de nous montrer qu’il y a quelque chose que nous ne pouvons pas
faire va de paire avec la tendance à le lire comme quelqu’un qui, à chaque étape de son
évolution, a cherché à indiquer à l’avance les conditions de tout discours pourvu de
signification – que l’on situe celles-ci dans la structure logique du langage, dans les
règles de grammaire, ou dans la forme de notre pratique linguistique. Chaque fois,
Wittgenstein aurait eu pour but de montrer au philosophe que ce qu’il dit est du non-
sens parce qu’il a violé certaines conditions. L’intégrité de sa méthode de critique
philosophique semble, au bout du compte, exiger de Wittgenstein qu’il soit en mesure,
en premier lieu, de spécifier quelles sont les conditions en question. Cette exigence, à
son tour, donne corps à l’idée qu’il est tenu de souscrire, à l’exigence d’un ensemble
de quasi-vérités – nommées principes de la syntaxe logique, règles de la grammaire,
Critical Remarks on Anti-Metaphyscial Readings of Wittgenstein 57
ou jugements constitutifs du cadre de notre pratique – qui marquent les limites du sens
et révèlent le point exact au-delà duquel le philosophe s’est égaré. Ces quasi-vérités
doivent se tenir hors du domaine des vérités empiriques, tout en se tenant juste en deçà
de la limite au-delà de laquelle commence le domaine du non-sens philosophiquement
prohibé. Mais sitôt qu’on soulève la question de savoir quelle sorte de vérité attribuer à
ces quasi-vérités, commencent verbiages et tergiversations. Car la conception du non-
sens qui émerge de la spécification d’un tel ensemble de conditions de possibilité de
tout discours pourvu de signification requiert toujours une distinction entre deux sortes
de propositions : les propositions empiriques ordinaires, et les propositions logiques,
grammaticales, ou appartenant au cadre – les conditions qui pèsent sur les premières,
et le problème étant que celles-ci finissent toujours par laisser transparaître, à leur
propre aune, leur échec à honorer les conditions de signification qu’elles cherchent
elles-mêmes à articuler.”
58 Wittgenstein: Issues and Debates
20
Stern (2004), pp. 13-15. Contrary to what he claims, his reading is a version of
pyrrhonianism: “In the Philosophical Investigations, one of the principal reasons for
Wittgenstein’s opposition to systematic philosophical theorizing is that our use of
language, our grasp of its meaning, depends on a background of common behaviour
and shared practices – not on agreement in opinions but in ‘form of life’ (§ 241). But
to say this so quickly is potentially misleading, for a great deal turns on how one
understands the ‘agreement’. Most readers take it to be a gesture towards a positive
theory of practice or the place of community in a theory of meaning. I shall be
proposing that we take Wittgenstein at his word when he tells us that the work of the
philosopher ‘consists in assembling reminders for a particular purpose’ (§ 127) – that
the remarks about common behaviour, shared practices, and agreement in opinions are
intended as reminders of what we ordinarily do, reminders assembled for the purpose
of helping his readers see the shortcomings of certain theories of knowledge, meaning,
and the like.”
21
Stern (2004), p. 14.
22
Ibid, p. 10.
Critical Remarks on Anti-Metaphyscial Readings of Wittgenstein 59
23
Ibid.
24
Ibid, p. 20.
60 Wittgenstein: Issues and Debates
28
Ibid, p. 100.
62 Wittgenstein: Issues and Debates
Conclusion
These two readings seem to be partially correct. But, as D. Stern writes,
they build Wittgenstein as more doctrinaire than he really is30. His works
seem to be inhabited by a tension between two strengths among which one
tends to give a definitive answer to the problems of Philosophy, whereas
the other tries to eliminate them. The second Wittgenstein seems to have
tried to minimize the constructive aspects of his Philosophy. How eloquent
is his use of the concept of family-likeness, in § 65 of the Investigations
and elsewhere 31 . He has acted as this concept exempted him from
constructing theories. It is used as a theory-deflation tool. But, as it is
shown by P. Hacker and others, we find constructive aspects, such as the
conception of meaning as use. This tension is not virtuous, contrary to D.
Stern’s opinion. It plausibly indicates Wittgenstein fell into a trap. Did he
show that Metaphysics is illegitimate? If what has been said is true, it is
not the case. None of these readings expounds anti-metaphysical
philosophy. None of these is really consistent. Of course, it does not follow
for a certainty that Wittgenstein was a metaphysician. It nevertheless
shows the necessity to minimize anti-metaphysical claims repeatedly made
by Wittgenstein. In other words, it is time to give up our picture of
Wittgenstein as the sworn enemy of Metaphysics. Moreover, metaphysical
researches are very active today; his writings could give us resources to
face its problems.
29
Baier (1967).
30
Stern (2004), pp. 36-7.
31
See the last page of the Blue Book.
Critical Remarks on Anti-Metaphyscial Readings of Wittgenstein 63
Bibliography
Armstrong, David, 1978: Nominalism and Realism, Universals and
Scientific Realism. Vol. I. Cambridge University Press, Melbourne.
--- 1997: A world of States of affairs. Cambridge University Press,
Melbourne.
--- 2004: Truth and Truthmakers. Cambridge University Press, London.
Baier, Anet, 1967: Nonsense. In: McMillan Encyclopedia of philosophy. 1,
Vol 5, pp. 520-522.
Conant, James, 2002: Le premier, le second et le dernier Wittgenstein. In:
Wittgenstein, Dernières pensées. ed by J. Bouveresse, S. Laugier, J.
J. Rosat. Agone, Paris, pp. 49-88.
Crary, A. and R. Read (eds.), 2000: The New Wittgenstein. Routledge,
London.
Descombes, Vincent, 1995: La denrée mentale. Les éditions de Minuit,
Paris.
Diamond, Cora, 2004: L’Esprit Réaliste, Wittgenstein la philosophie et
l’esprit. (french trans by E. Halais and J. Y. Mondon). PUF, Paris.
Hacker, P.M.S., 1972: Insight and Illusion, Wittgenstein on Philosophy
and the Metaphysics of Experience. Oxford University Press, New
York.
--- 1986: Laying the Ghost of the Tractatus. In: Wittgenstein Critical
Assessments. Vol. 1 (ed. by S. Shanker). Routledge, London, pp. 76-
92.
--- 1986: Wittgenstein Today. (with G. Baker) In: Wittgenstein Critical
Assessments. Vol. 2 (ed. by S. Shanker). Routledge, London, pp. 24-
36.
--- 1996: Wittgenstein’s place in the twentieth-century analytic
philosophy. Blackwell, Oxford.
--- 2003: Wittgenstein, Carnap, and the new American wittgensteinans.
The Philosophical Quarterly, 2003. Vol. 53, No. 210, pp. 1-23.
--- 2000a: Wittgenstein. (French trans by J. L. Fidel). Seuil, Paris.
--- 2000b: Wittgenstein, Mind and Will. Blackwell, Oxford.
--- 2001: Wittgenstein: connections and controversies. Oxford
University Press, Clarendon, Oxford.
64 Wittgenstein: Issues and Debates
Introduction
In a paper of mine – How to Make Opposite Ends Meet 2– I contrasted
Wittgenstein’s understanding of meaning in his two periods in several
respects. One respect was his position on language. I argued that
Wittgenstein has two opposite approaches to meaning in his first and
second periods: Meaning as representation (viz. picture theory of meaning)
and meaning as use. In his first period, where he adopted a theory of
meaning as representation, he considers language, the sentences of which
are pictures of reality, and which is the vehicle of thought, as independent
from reality, standing for the objects or states of affairs in the world.
However, in the second period, where the use conception of meaning is put
forward he conceives language as something dependent on the human
agents that employ it, hence he conceives it as something into which
actions are interwoven. In this sense, Wittgenstein, in his second period,
can be thought to consider language as an action in the world, thence
consider it as in and part of reality, when it comes to the semantic
connectedness between the two.3 In that sense, Wittgenstein can be thought
1
I am indebted to Prof. Peter Hacker for his very careful reading of my paper, his
invaluable remarks and suggestions for the refinement of the paper, and for his mind-
opening detailed explications at some points in my paper as well as on my peculiar
questions on the issue. I am also grateful to Prof. Ali Karatay who contributed on the
very emergence of the ideas in this paper through our long-lasting discussions and
through his useful comments on some key points. Without their respectable
contributions, this paper would not have appeared.
2
Cakal, 2006.
3
For in the latter view praxis as a part of reality is included in language; since
language and reality interwoven (as the speaking of language is part of an activity or
of a form of life –hence language is woven in a form of life). The meaning of linguistic
expressions is not independent of humans as it was in Tractatus. In the Investigations
it is the way they are used; it depends on how they are used (by humans –in a society).
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ),
Ontos Verlag, Frankfurt a. M., 2010, 65-81.
66 Wittgenstein: Issues and Debates
Language is inseparable from humans –human actions. While in the Tractatus the
statement that the proposition that p is made true by the fact that p is a meta-logical
connection between language and world, in the Investigations, it is an intra-
grammatical relation, viz., ‘the proposition that p’ is equal to ‘the proposition made
true by the fact that p’.
What Does the Repudiation of Private Language Mean in Wittgenstein’s 67
Philosophy?
whereas ‘mind’ is considered as something by which perceptual and
intellectual processes are activated. In this sense, mind is, in general, seen
as something that provides us with understanding through mental faculties
as well as that which perceives.4 According to Descartes, mind is a being
“which affirms, which denies, which wills, which rejects, which imagines
also, and which perceives”.5 So, the domain of sense perceptions is in a
sense under the control of mind [for a sense perception is a mode of
thought, the vehicle of which is language]. So, in this view, it is the mind
which sees, smells, feels, and intends to do things. Moreover, in Meditation
VI, Descartes argues that the mind, that is, a thinking thing, can exist apart
from its extended body. And therefore, the mind is a substance distinct
from the body, a substance whose essence is thought. The consequence of
this view is, apparently, to contrast the public character of the body with
the private status of the mind by drawing a distinction between ‘inner’ and
‘outer.’ In this view, while the operation of mind is considered as
‘internal’, the operation of body is considered as ‘external.’
Now, this dualistic view of the world, when taken to the extreme –
that is, to the idea that mental processes happen in isolated fields and
human beings are completely confined within the mental happenings of
their own so there is in fact no connection between happenings of the mind
and those of the other- surely brings about the problem of skepticism; and
this problem is labeled by Wittgenstein as something not irrefutable in
Tractatus-Logico Philosophicus (hereafter the Tractatus). 6 It seems,
Wittgenstein’s bugbear is the problem of skepticism. In the Tractatus,
thence, to tackle the problem, as Kenny also remarked, Wittgenstein might
have set out the programme of refuting skepticism, with the following
lines:
Skepticism is… 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.7
It is possible to read Wittgenstein’s programme then undermining
skepticism by showing that it is nonsensical. Now, if skepticism is
nonsensical -as Wittgenstein maintains- and raises doubts, then what must
4
Cf. Descartes, 1951, 23-33.
5
Descartes, 1951, 27.
6
Wittgenstein, 2004, TLP 6.51.
7
Kenny, 1986, 180.
68 Wittgenstein: Issues and Debates
be done, admittedly, is to abate that doubt. His programme that had already
been set out in the Tractatus, in this sense, must be to pick out the element
that is responsible for skepticism and to refine the representational theory
of meaning, which he put forward in Tractatus, in accordance with it. And
as the result of this refinement we’ll see there will emerge the use
conception of meaning, where there is no room for skeptical scenarios, and
which is the very opposite of the Tractarian representational theory of
meaning.
It can be thought that the doubtful connection between the two
components of the dualistic approach can be said to be the culprit for
skepticism as the connection between happenings of the mind and those of
the other can lead to skeptical problems when they are thought to be
separated and when taken to the extreme -as it is noted above. If so, then
Wittgenstein must either revise the semantic connection he established
between the two (language and reality) in his representational theory in
such a way that it is not doubtful anymore so that what he says would be
something can be said or in order not to encounter this problem at all, he
must eliminate this doubtful connection in his theory as a whole so that
there is no need to speculate on where something “cannot be said.” It
seems to me that Wittgenstein, in his later philosophy, follows the latter
way and eliminates the problem of skepticism by wiping out one
component of the duality, namely ‘(mere) internal’8 –or what belongs only
to mind. Note here that the internal part of dualism is associated with mind,
and mind was associated with language as a vehicle of thought. And I think
that Wittgenstein does this [wiping out one component of dualistic view
(which is the mere internal)] through his private language argument. In the
argument, by rendering bare ostensive definition unintelligible, he
apparently eliminates the possibility of having a meaningful and merely
internal language, a language the words of which can only be known to the
person speaking. So, in his later philosophy, it is not possible for a
linguistic expression to be known merely internally (through mere mental
faculties); hence, to have a meaning only internally –or by bare (internal)
8
As Descartes did not make the subtle distinction between mind and soul, but
Wittgenstein did, in order to provide a conflux and congruence in the terminologies
may be used or partly used for both philosophers, I will use the term ‘mere internal’ to
be standing for ‘private’ in the sense of incommunicability (not inalienability) in
Wittgenstein’s philosophy, and for ‘mind’ or ‘soul’ (which is not ‘matter’) in
Descartes’ philosophy.
What Does the Repudiation of Private Language Mean in Wittgenstein’s 69
Philosophy?
ostensive definition. As, for him, ‘mere internal’9 or what belongs only to
mind has no meaning. In this sense, his “programme,” can be said to avoid
skepticism by rendering the internal (or mental part of dualistic view)
meaningless by means of reducing it to what cannot be said, in Tractarian
way of approaching the issue, so that there will remain only the talk of
external where there is no room for skeptical scenarios. Note here that
what cannot be said, in the Tractatus, is considered as something
nonsensical or something that lacks meaning.10 So, what is merely internal
or private must be proved to be what cannot be said if it is meaningless.
And I think he does so through his private language argument by denying
the meaningfulness of linguistic expressions that have no criterion for their
correctness; and this will explain why one cannot give a(n) (ostensive)
definition of a private sensation by focusing on the sensation and on the
symbol: It has no meaning as it has no criterion for its correctness.
But then the question arises: How about thinking? As a “mental
activity” is it also meaningless? Is there no such thing as thinking as the
internal part of dualism? Wittgenstein claimed in Blue and Brown Books
that thinking is ‘the activity of operating with signs’. 11 As, there, he
maintains:
…thinking is performed by the hand, when we think by writing; by the
mouth and larynx, when we think by speaking; and if we think by
imagining signs or pictures, I can give you no agent that thinks.12
Although Wittgenstein later on leaves this strong behaviorism, the
idea will bring about, in the Investigations, the idea that words and actions
are interwoven. Language games, which are introduced in the
Investigations, in this sense, will be thought to be rule-governed activities.
Thus for Wittgenstein “teaching of language” will be considered as
training, not explanation. 13 Language will be considered as something
taught and learned by means of practices; not through one’s own mere
9
Caveat: the sensation names which we use are not ‘mere internal’, according to
Wittgenstein.
10
For my purpose, I will not consider the subtle difference between meaning and sense
Wittgenstein made in the Tractatus. I will rather use the term ‘meaning’ for both terms
in order to provide a congruity in terminology in overall paper.
11
Wittgenstein, 1958, BBB, p. 6 [my italics].
12
Wittgenstein, 1958, BBB, p. 6 [my italics].
13
Cf. Wittgenstein, 2001, PI § 5.
70 Wittgenstein: Issues and Debates
internal faculties –like focusing on the sensation and on the symbol –by
bare ostensive definition.
14
At this point one may object that since in Wittgenstein’s account it is a world of
objects and facts; it’s a world which lacks selves, values, cognitive relations (such as
belief), and God, the world Tractatus presents is monistic, not dualistic. Yet, I think
that it is dualistic. For:
(i) In the Tractatus, it is said “We picture facts to ourselves.” [Wittgenstein, 2004,
TLP 2.1, my italics] So, as Kemerling notes (Philosophy Pages,
http://www.philosophypages.com/hy/6s.htm, 24.04.2007). Human beings are
aware of the facts by virtue of our mental representations or thoughts, which are
most fruitfully understood as picturing the way things are.” Tully also touches
upon the same point and writes that Tractatus is not concerned with
determining the conditions of good representation or with deciding when a fact
must forfeit its claim to be a picture. For him, it is enough that there are many
non-failures; we do “picture facts to ourselves.”
(ii) As Tully writes in his article called Tractarian Dualism, “Wittgenstein seems to
have the activity of thought in mind although he elaborates very little beyond
remarking that “a propositional sign applied and thought out is a thought”
[Wittgenstein, 2004, TLP 3.5] and immediately after that “a thought is a
proposition with a sense”.” [Wittgenstein, 2004, TLP 4.0] For Tully, “Nothing
What Does the Repudiation of Private Language Mean in Wittgenstein’s 71
Philosophy?
As to Wittgenstein’s later views on meaning –especially in the
Investigations, for him, what linguistic expressions mean is not the
representation relation between world and language -not the objects or
states of affairs that correspond to reality, as it was in Tractatus, but it is
rather, the use of those linguistic expressions in a language game by
humans, and which is “the whole, consisting of language and the actions
into which it is woven.” 15 Therefore, according to Wittgenstein’s later
view, uttering something is considered as an action in the world; when one
uses a term in the language game it belongs, actions are also inherent in the
use of the term. Therefore, Wittgenstein does not view language as a fixed
and timeless framework but rather as a vulnerable and changeable aspect
of human life. In this sense, he denies the semantic connectedness between
words and world, which gave words their meanings. Thus, we cannot look
outside the linguistic practice to find, that which governs it; we cannot
“sublime the logic of our language”.16 What we have is after all language
and the forms of life, which arise from language and make it possible. It is
this idea that brings about the idea that words and actions are interwoven.
17
Wittgenstein, 2001, PI § 5.
18
Baker and Hacker, 1985, 176.
19
Wittgenstein, 2001, PI § 243.
What Does the Repudiation of Private Language Mean in Wittgenstein’s 73
Philosophy?
words refer to the speaker’s immediate private sensations, then one may
ask: how is it that words refer to sensations? As, for Wittgenstein, we talk
about sensations everyday and give them names, for him, this question
should be asked as “how the connection between the name and thing
named is set up?” This is according to him the same as asking the question:
“how does a human being learn the meaning of the names of sensations?”20
He believes, “...words are connected with the primitive, the natural,
expressions of the sensation and used in their place”.21 Sensation names are
thus the replacement of the natural behavioral expressions of the sensation
by the verbal expression. The sensation-word, however, does not describe
(but replaces) the natural expression of sensation; for, according to him,
language cannot be used to step in between sensation and its expression.22
In a private language, Wittgenstein argues, words that stand for
sensations cannot be “tied up with” the “natural expressions of sensation”
because the words of this language have no natural expression.23 As he
thinks that all sensations do have their natural expressions and they are
publicly observable, any sensation word which has its natural expression
for him is not private. For example, our (sensation) word ‘pain’ is not
“private” because it can be understood or known by other people through
its expression (as it is shared publicly).
A specimen: Suppose a person who has a kind of toothache, calls her
sensation as an ‘X’ secretly. When she feels that sensation, which she calls
“X”, for Wittgenstein if she shows ‘the natural expression’ of her
sensation, in our case, say, she holds her cheek in a sulk, “X” is not a word
in a private language. For it has its (publicly sharable) natural expression.
Now the question arises: What if one does have the sensation but does not
show any natural expression for it? In such a case, Wittgenstein avers, her
sensations remain unexpressed and no one else can know that she has the
sensation. However, from this one cannot infer that X would be
meaningless because it could be expressed and its expression would be a
criterion for ascribing X to that person. Then, one can further ask: what if
human beings do not show outward signs of sensations at all? If that were
the case, Wittgenstein asserts, by pointing out the connection between
sensation and the expression of sensation in human life, that if human
20
Wittgenstein, 2001, PI § 244.
21
Wittgenstein, 2001, PI § 244.
22
Wittgenstein, 2001, PI § 245.
23
Wittgenstein, 2001, PI § 256.
74 Wittgenstein: Issues and Debates
beings did not show outward signs of sensations, which amounts to saying
that if sensations were to lack expression, then it would be impossible to
teach the use of sensation-words.24
The remarks offered so far concerning private language as inventing
names are obviously not enough to show Wittgenstein’s claim that private
language is impossible. The impossibility claim is made in PI § 258. As a
private language is something impossible, from now on I will talk about it
as putative private language. Now, let us be clear about what is it that is
impossible: The thing which is impossible questioned in the argument by
Wittgenstein is whether one can name25 one’s own sensations and if so,
whether one could by oneself understand and use those names in one’s
private language. He investigates this through his celebrated diary example
in PI § 258: Imagine someone who wants to keep a diary about the
recurrence of a certain sensation, but in accordance with the putative
private language, viz., the words of which can only be known to the person
speaking. In order that one may do so, Wittgenstein thinks, one has to
“associate” one’s sensation with the sign ‘S’ and write this sign in a
calendar every time one has the sensation. Such a diary in fact cannot be
kept according to Wittgenstein due to lack of any criterion of correctness in
recollecting what ‘S’ means. So, in a (putative) private language, there is
no difference between ‘something’s being right’ and ‘thinking that it is
right’. In this sense, ‘S’ lacks criteria for its correct use. And as one cannot
know whether what one says is meaningful, one cannot know what one
means by it according to Wittgenstein. Therefore, one cannot name one’s
own (putative) private sensation.
Since, for Wittgenstein, one learns the meanings of words from
society through their use in a social environment, which determine their
senses, to understand a word is to know how to use it, the meaning of a
word is the use that we make of that word; the difference between correct
and incorrect use is manifest in publicly observable behavior. Then one
may ask: What is the criterion for a public language, which is not a private
one? The use of language, for Wittgenstein, is fundamentally rule
following and rule following is a practice, which is the result of custom and
training within a social milieu. A language is thus based on an agreement
for him. It is agreement in form of life; and it is what makes linguistic
24
Wittgenstein, 2001, PI § 257.
25
By ‘name’ he means ‘word whose meaning is learned by bare ostensive definition.
(cf., Wittgenstein, 1958, BBB, p. 82)
What Does the Repudiation of Private Language Mean in Wittgenstein’s 75
Philosophy?
expressions meaningful. As in PI § 242, Wittgenstein argues, in order for a
language to be a means of communication, there must be agreement not in
definitions but also in judgments.
The reason, in the diary case, for the impossibility of naming a
private sensation, for Wittgenstein, is that a definition for the private
sensation ‘S’ cannot be formulated (due to the lack of any criterion of
correctness in recollecting what ‘S’ means). Moreover, this claim is the
gateway to the arguments that private ostensive definition is unintelligible
and private language impossible.
A definition for the private sensation ‘S’ cannot be formulated
because one cannot know what one means by it. For, in order to mean
something, it must satisfy the condition for its meaningful use. In order for
it to be meaningful, it must have grammar –or rule- with which it can be
checked for correctness of its use. So in order to define the sign ‘S’, a
grammar or a ‘stage setting’ under which the sign can be used correctly or
incorrectly is needed. Now, since ‘S’ is the putative name of a private
sensation, which is impossible for another person to know or have, hence
which it has no grammar to check for its correct use, a criterion for its
correct use cannot be given. Therefore, it does not mean anything.
Then, the question arises: can I myself not determine the standard of
the correct use of the sign ‘S’? Or one may ask this question in another
way: Can I not give myself an ostensive definition –a private mental
ostensive definition- that would function as a criterion for the right or
wrong use of ‘S’? For Wittgenstein, it cannot be given either; because the
definition of a name serves to establish its meaning. That is, the function of
a definition is to provide the meaning of a sign and establish a criterion for
the correctness of its future use. And one cannot achieve this by simply
concentrating on the sensation as speaking or writing ‘S’ and thereby
memorizing the relation between the sign and the sensation. For one thing,
there is no criterion of correctness in recollecting what ‘S’ means, in other
words, one cannot remember whether ‘S’ means this –or the sensation one
has. In our (public) language, there is no such question; because we follow
rules in the course of the practices of everyday life. But in this case there is
no criterion (rule, grammar, or stage setting) in whether one is right or
wrong in recollecting what ‘S’ means.
Then again one may ask what if there is such a person who
remembers the connection by means of concentrating her attention on the
sensation through a mental pointing or “pointing inwardly”, i.e., a private
76 Wittgenstein: Issues and Debates
mental ostensive definition. Even then, Wittgenstein would reject it. His
rejoinder would be the criticism of the way of reasoning. He censures this
way of reasoning by labeling it as an “idle ceremony”.26 Remember that
early in the Investigations Wittgenstein argued against the idea that
ostensive definitions constitute the foundations of language. 27 He
considered the example of someone pointing to two nuts while saying
“This is called two” and asked: How does it come about that the listener
associates this with the number of items, rather than the type of nut, their
color, or even a compass direction? One conclusion of this, for
Wittgenstein, was that to participate in an ostensive definition presupposes
an understanding of the process and context involved, of the form of life.28
Another was: “an ostensive definition can be variously interpreted in every
case”. 29 Therefore, without a stage-setting –a determinate grammar– we
already cannot talk about an ostensive definition. We cannot, by ourselves,
connect a name and the thing named without an appropriate stage-setting.
In the diary case, there is no such setting, which will function as a
criterion for the correctness of the use of ‘S’; so, ‘S’ lacks meaning. And if
we grant that the definition of a name serves to establish its meaning, then
the putative private ostensive definition by reference to a putative memory
sample cannot function as a definition. That is to say, ‘S’ conceived as the
name of a “private” sensation, which is both privately known (or
incommunicable, viz., no one else can know it) and privately owned (or
inalienable, viz., no one else can have it) is unintelligible.
In this sense, it can be thought that Wittgenstein’s private language
argument eliminates the possibility of merely internal meaningful
language. Briefly, in a (putative) private language since the subject would
not have any criterion of correctness in recollecting what ‘S’ means, hence
to decide whether the sign ‘S’ is used correctly or not, it would lack
meaning. In this scene, as Kenny emphasizes, when next I call something
“S” I do not know what I mean by “S”, thence, according to commentators
of Wittgenstein, ‘I do not say anything’30 or such a use is ‘empty’.31
26
Wittgenstein, 2001, PI § 258.
27
Wittgenstein, 2001, PI §§ 23-28.
28
Wittgenstein, 2001, PI § 23.
29
Wittgenstein, 2001, PI § 28.
30
Ayer, 1971, p. 52; Rhees, 1970, p. 66.
31
Strawson, 1954, p. 28; Hacker, 1972, p. 234.
What Does the Repudiation of Private Language Mean in Wittgenstein’s 77
Philosophy?
3. What is not a private language?
Now, having seen in general what the private language argument is, let’s
see what is not a private language: From the above mentioned remarks we
can see that what Wittgenstein means is not a contingently private
language which no one else can understand, but rather, it is a purely private
language that it is logically impossible for any one to understand. The
words of such language refer to the speaker’s immediate private sensations
and inner experiences. So, the question for Wittgenstein does not simply
concern the possibility of someone’s inventing a name for her own
sensation. For this is something very well possible in Wittgenstein’s
philosophy, as it is something knowable by any other person by means of
its natural expression or it can be translated into an “ordinary” (public)
language. To open this up a bit, let’s consider a specimen: Suppose, now
and then, I feel a sensation of throbbing my head twice and then tingling
my ear for about 3 seconds right after that and suppose I call it, say,
“füútürük.” When it happens, I say, “this is ‘füútürük’ again,” but nobody
knows that I call it ‘füútürük.’ Now, according to Wittgenstein, is this, i.e.,
calling my sensation ‘füútürük’ secretly, a private language? The answer is
obviously not. For here, what is the case is only inventing a name for a
sensation that which is not unknowable by others; it can be expressed and
understood –therefore knowable- by any other person. Thus, it is neither a
private sensation nor a private language. Inventing a name for a sensation
is not something impossible in his philosophy. So, let’s not confuse it with
an invention of a name for a sensation.
5. Conclusive Remark
Wittgenstein’s “programme” of undermining the dichotomy of “inner” and
“outer” set in Tractatus seems to be undermined with his repudiation of
private language by means of reducing it to something meaningless (on the
grounds that there is no criteria for the correctness of it).
Now if there can be no such thing as private language whose words
acquire meaning simply by being linked to private experiences, then, it is
impossible for one to assign meaning to a linguistic expression by oneself –
neither by an act of meaning nor in a merely internal way. But in the
Tractatus, the latter was possible because in there, “we picture facts to
32
Incommunicability, as Kenny introduces the term, connotes that the language used
or the expression uttered is knowable by any other person.
80 Wittgenstein: Issues and Debates
Bibliography
Ayer, A. J., 1971: ‘Could language be invented by a Robinson Crusoe?’.
In: The Private Language Argument. (ed. O. R. Jones), Macmillan,
London.
Baker, G. P., Hacker, P. M. S., 1985: Wittgenstein, Rules, Grammar and
Necessity. Basil Blackwell, New York.
Baç, M., 2001: ‘Wittgenstein ve Anlamın Ortalıkta Olması’, Felsefe
Tartıúmaları, 28. Kitap, Bo÷aziçi Üniversitesi Yayınevi, Istanbul.
Cakal, A., 2006: ‘How to Make Opposite Ends Meet’. Papers of the 29th
International Wittgenstein Symposium. (Eds. G. Gasser, C. Kanzian,
E. Runggaldier), Vol. XIV, Kirchberg am Wechsel.
Descartes, R., 1951: Meditations on First Philosophy. Bobbs-Merill
Educational Publishing, Indianapolis.
Grandy, R. E., 1976: ‘The Private Language Argument.’ Mind, New
Series, Vol. 85, No. 338, pp. 246-250.
Hacker, P. M. S., 1972: Insight and Illusion. Wittgenstein on Phylosophy
and the Metaphysics of Experience. Oxford University Press,
Oxford.
Kenny, A., 1986: Wittgenstein. The Penguin Press, Allen Lane.
33
Wittgenstein, 2004, TLP 2.1, my italics.
What Does the Repudiation of Private Language Mean in Wittgenstein’s 81
Philosophy?
Pitcher, G., ed., 1966: Wittgenstein. Garden City, NY, Anchor Books,
Doubleday, pp. 22-64.
Rhees, R. 1970: Discussions of Wittgenstein. Routledge and Kegan Paul,
London.
Sidiropoulou, C., 2004: Sp. Top: Wittgenstein. Unpublished Lecture Notes,
Fall.
Soames, S., 2005: Philosophical Analysis in the Twentieth Century Vol. 2:
The Age of Meaning. Princeton University Press, Princeton.
Strawson, P., 1954: ‘Review of Philosophical Investigations’ Mind, 63:
reprinted in Tully, R. E. Tractarian Dualism. Retrieved over:
http://www.bu.edu/wcp/Papers/Meta/MetaTull.htm (28.02.2010).
Wittgenstein, L, 1958: Blue and Brown Books. Blackwell, Oxford.
--- 2001: Philosophical Investigations. (The German Text, with a
Revised English Translation, 3rd edition). Blackwell, Oxford.
--- 2004: Tractatus-Logico Philosophicus. (Trans. D.F. Pears, B.F. Mc
Guiness), Routhedge and Kegan Paul, London.
Wittgenstein and the
Myth of Hinge Propositions
Alejandro TOMASINI BASSOLS
I. Historical-Philosophical Considerations
My main goal in this paper is to examine critically, in order to put an end
to it once and for all together with all it has brought about, what from my
point of view is a very harmful myth, ascribed to Ludwig Wittgenstein by
several distinguished scholars, 1 viz., the myth of the so-called ‘hinge
propositions’. From the point of view of the history of ideas, the effects of
this myth just cannot be ignored. Indeed, based on this myth it now has
become a sort of unavoidable fashion to speak of a “third Wittgenstein”, a
thinker who apparently would have superseded or overcome his own
magnum opus, i.e., the Philosophical Investigations. According to those
who support this reading of Wittgenstein’s work, he would have left
behind his own phase of grammatical analysis in order to enter a new field
of philosophical speculation, a new kind of therapy (whose nature, by the
way, has never been made quite clear), much more similar to the sort of
rational adventure which is practiced in traditional metaphysics than to the
kind of analysis which Wittgenstein painfully managed to developed and
practice and which, last but not least, brought him so many good results.
Now in order to count with more elements to carry out our examination
it’ll be convenient to have at our disposal a simple but neat and well argued
panorama, a global but a convincing one too, of the philosophical
background in which Wittgenstein has recourse to the expression ‘hinge’,
which so many headaches it has already caused. We are thus forced to
exercise ourselves in a task of reconstruction of ideas. Our starting point,
therefore, can only be the following: if we contemplate his work in toto:
how many Wittgensteins is it reasonable to think there are? To answering
this question I’ll devote the first part of this essay.
1
Among the most important that should be mentioned are the following ones: Baker
2004; Harré, 2008 and 2009; Cook, 1994; Moyal-Sharrock, 2004 and Moyal-Sharrock
(Ed.), 2004; Stroll, 2002.
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ),
Ontos Verlag, Frankfurt a. M., 2010, 83-116.
84 Wittgenstein: Issues and Debates
are simply a “logical method” that shows the logical structure of reality
through its equations. Mathematics cannot be given any kind of foundation
for the simple reason that they do not express thoughts, since they do not
contain names. Mathematics are systems of identities which are worked
out by means of the method of substitution. But mathematics are in need of
foundations no more than logic is or, as Wittgenstein puts it, they take care
of themselves. The truths of mathematics can be established without
having to appeal to set theory. In the same vein, the idea of founding
language is totally alien to the Tractatus. That every possible language
divides itself into propositions which in some sense are the last residue of
analysis, that is, into their respective elementary propositions, is not a
matter of founding language on anything, but of the logic of language.
There are no elementary propositions which would be more fundamental
than others.
In the philosophy that Wittgenstein develops from 1929 onwards, the
foundationist approach is not only absent, but is openly attacked. In the
conception of language that Wittgenstein puts forward in the Investigations
there is no place for a pyramidal view. Of course, there are language-
games more primitive than others, more refined or sophisticated than
others, but the development of language is accomplished in relation with
practices and not as a purely symbolic extension, as when we expand a
calculus by demonstrating new theorems starting from some axioms by
means of a couple of rules of inference. Human practices need no
justification at all. We reach the end of explanations when we acknowledge
that a certain language-game is actually played and there is nothing else to
say. Since definitions and explanations end at some stage and cannot go on
ad infinitum, we reach the bottom when we are faced not with any kind of
special truths but with spontaneous human reactions, with human action,
i.e., with praxis, for what could be more fundamental than that? Now, since
there is no direct causal connection between spontaneous reactions and
language-games, since grammar is not conditioned by any reality, the limit
we speak about is not a foundational one, in the sense that starting from it
we could logically derive the rest, but simply a platform on which
language grows.
Naturally, it could be objected with respect to foundationism that I
am just begging the question and that I am simply denying that which
others have argued. However, and awaiting to say a bit more about it later
on, I’d like to point out that the only thing I’ve done so far is simply to
maintain in a purely abstract way that we don’t have the least reason to
86 Wittgenstein: Issues and Debates
2
It was G. H. von Wright who first used this term in relation with Wittgenstein’s
philosophy. As he makes it clear himself, he borrows the term from the great Polish
philosopher, T. KotarbiĔski.
3
Wittgenstein, PI, 1974, viii. (Emphasis mine. ATB).
Wittgenstein and the Myth of Hinge Propositions 89
c) Strategies and methods. From what has been said up to now it can
be inferred that the first and the second Wittgenstein’s philosophical
approaches and tactics had to be completely different. For the Wittgenstein
of the Tractatus, once the fundamental ideas had been outlined
philosophical work just boiled down to the searching of references for
names to determine whether or not a picture was meaningful and
eventually to ascribe to it a truth-value. The method in this case was the
method of logical-semantic analysis. For the second Wittgenstein, given
his rejection of the Augustinian conception of language, nothing could be
more futile and sterile than that. It is true that in both cases the ultimate
goal was to separate off sense from nonsense, meaningful assertions from
those which are so only apparently, but the method to achieve that had to
be completely different. Actually, in the case of the second Wittgenstein
we cannot speak of a single method, but rather of a whole variety of them.
In fact the list is an open one, since new methods can be devised at any
time and added to the list. Regarding this, section 133 of the Investigations
is particularly relevant. I quote it in extenso: “It is not our aim to refine or
complete the system of rules for the use of our words in unheard-of ways.
For the clarity we are aiming at is indeed complete clarity. But this
simply means that the philosophical problems should completely disappear.
The real discovery is the one that makes me capable of stopping
doing philosophy when I want to.– The one that gives philosophy peace, so
that it is no longer tormented by questions which bring itself in question.–
Instead, we now demonstrate a method, by examples; and the series of
examples can be broken off.– Problems are solved (difficulties eliminated),
not a single problem.
There is not a philosophical method, though there are indeed
methods, like different therapies”.5 It would be difficult to be more explicit
than that: the general goal is just to exhibit the hidden nonsense of
philosophical assertions, regardless of their content or of the branch of
philosophy to which they belong, while the methods Wittgenstein employs
are most diverse. Nowhere did he intend to give an exhaustive list of
methods. It is our task to get them from his work, discerning them out of
philosophical exercises. Here are some of them:
1) to replace questions like ‘what is x?’ by questions of the form
‘under what circumstances do we say that someone knows what “x”
5
Wittgenstein, PI, 1974, sec. 133.
92 Wittgenstein: Issues and Debates
6
Wittgenstein, TLP, 1978, 6.54 (b).
Wittgenstein and the Myth of Hinge Propositions 93
Tractatus had been refuted. However, for Wittgenstein it was obvious that
his task just couldn’t finish at that stage: the rejection of the Augustinian
conception of language and the construction of his new point of view in
terms of language-games and forms of life was purely propaedeutic. From
then onwards the goal was to demonstrate the superiority of the new
approach showing that, when properly applied, it enables us to dissolve
any philosophical puzzles one would analyze. In order to do that,
Wittgenstein went on designing, as I already said, a variety of methods and
tactics for philosophical debate thanks to which he was in a position to
dismantle, one after the other, the philosophical theses and doctrines he
decided to tackle. Classical philosophical enigmas like the mind-body
issue, the problem of universals, the puzzles concerning meaning, all the
philosophical difficulties around logical and mathematical truth, the
question of the existence of God, and many others, were dismantled (some
of us would have thought that once and for all, but this turned out to be an
illusion). Now for purely contingent reasons, the last problem Wittgenstein
came to face was a problem belonging to the theory of knowledge, that is,
the problem of scepticism. In other words, once his basic conception and
methods of work had been introduced, Wittgenstein passed to the next
phase, that is, the phase of application of what had been established in,
say, the first 135 or 140 sections of the Investigations, to concrete
philosophical problems, regardless of the area or branch of philosophy.
From the moment in which Wittgenstein approaches the issue of the nature
of understanding, to pass to those of reading, rule-following, private
languages, mental states and so on, what he does is to demonstrate in
practice the superiority of his new way of thinking. As a matter of fact, all
the problems he approached just vanished. So far as I am concerned, I
wholeheartedly acknowledge that I am unable to point to a theme or a
subject Wittgenstein dealt with which was not sufficiently elucidated by
him, at least to the extent that we don’t have to take care of it again.
If what I’ve so far stated is basically right, it follows that the most
absurd thing that could be done would be to suggest a proliferation of
Wittgensteins, specially if the only grounds to do that is the fact the on a
couple of occasions he employed some expression not previously
introduced. This is the case of the so called ‘hinge propositions’. Let’s
review this a bit more in detail.
Probably the first question to be asked is: Where does the idea of a
“hinge proposition” first appear, where does Wittgenstein actually use that
Wittgenstein and the Myth of Hinge Propositions 95
7
Wittgenstein, OC, 1969, sec. 341.
8
Wittgenstein, OC, 1969, sec. 343.
9
Wittgenstein, OC, 1969, sec. 655.
96 Wittgenstein: Issues and Debates
Now there are truths which lack this ephemeral or fleeting character, for
they describe or point to situations, phenomena, facts, etc., which simply
just don’t vanish but remain or endure. For instance, I can assert that Paris
is the capital of France and in saying this I both make a legitimate move in
some language-game (e.g., to answer a question during an exam, to inform
someone about it, etc.) and I also build a proposition which can serve me
as well as others in a variety of circumstances. It is crucially important,
therefore, to distinguish between propositional roles: there are sentences
for the moment and sentences which possess a recurrent utility. ‘I am
hungry’ exemplifies the former, ‘The Earth existed before I was born’ the
latter.
On the other hand, given the vital character of the Wittgensteinian
conception of language in terms of language-games, it had to be expected
that we should be given a dynamic view of assertions. Wittgenstein calls
our attention to this in an interesting remark in the Philosophical
Investigations where, after discussing Russell’s Theory of Descriptions he
says: “The fluctuation of scientific definitions: what to-day counts as an
observed concomitant of a phenomenon will to-morrow be used to define
it”. 11 In other words: we witness every day what could be called
‘propositional transitions’, that is, changes in propositions’ different statu.
Disposable propositions belong to the surface or, to employ Quine’s very
similar metaphor, to the shore, whereas there are propositions which are
rather fixed, the famous “bedrock propositions”, propositions that in
general experience leaves untouched. But what should not be passed
unnoticed is that there is nothing fixed once and for all: in principle,
propositions from the bottom may disappear and others, which at a certain
moment were on the surface may become more important, become
indispensable and locate themselves in the bed of the propositional river.
This is the way language functions, not to say ‘lives’.
Thus, sentences are tools, which may give rise to propositions of a
completely different nature. Needles to say that non well formed, elliptic,
etc., sentences may nevertheless give rise to genuine moves in the
language-games. So strictly speaking it could be the case that
communication among speakers does take place without our being forced
to talk about propositions at all, for sentences would be ill-formed. On the
other hand, there are circumstances in which a sentence like ‘I’ve never
traveled to Mars’ would allow me to make a legitimate move in the
11
Wittgenstein, PI, 1974, sec. 79.
Wittgenstein and the Myth of Hinge Propositions 99
12
See, for instance, his Moyal-Sharrock, 2003, where she argues in extenso in this
sense.
Wittgenstein and the Myth of Hinge Propositions 101
13
Just imagine, for instance, that someone asks Caesar or Napoleon or Stalin to have a
sit and that immediately after that he points out the any of them that the object he will
sit on is a physical object. Such a joke could have rather unpleasant consequences.
Mutatis mutandis, the same would happen in any other linguistic context (father-son,
teacher-pupil, doctor-patient, police man-citizen, etc.).
106 Wittgenstein: Issues and Debates
can’t be maintained that ‘I’ve never traveled to the center of the Earth’
doesn’t give any kind of knowledge, for to do that we should have to deny
that the proposition is true and we just can’t do that. What has to be done is
to employ the Wittgensteinian terminology and point out that although the
proposition is true, in fact with it we make no genuine move in the
language-game and therefore we convey no knowledge at all. Thus, the
concept of knowledge enables us to allude to quite different things. I’ll try
to make this a bit clearer.
First of all, we’ve got the paradigmatic, prototypical individual sort
of knowledge, that is, the knowledge I report: what I see, I remember, I
imagine, etc. This is immediate individual or personal knowledge. Now
this knowledge carries with it all those propositions which automatically
the speaker generates as soon as he says something. This is a derivative
sense of the first sense of ‘knowledge’. For instance, if someone says that
the lion is about to catch a zebra, he automatically generated the
proposition that both the lion and the zebra are animals, that the lion eats
meat, etc. The point is that these propositions, in the linguistic context in
which they are located, are completely useless for the speaker. They are
simply implied or presupposed or whatever is the epistemological relation
that holds between them and the actual linguistic move. It is important to
observe, on the other hand, that the same linguistic move made by another
speaker generates a different propositional system, although it is probable
that their respective systems will contain lots of common propositions. It is
highly plausible that from all the propositional systems that speakers
generate when they say something we could eventually form one which
would be “complete”, that is, it would contain all the propositions which
are common to all speakers’ propositional systems. That propositional
system would give all speakers objective certainty and in relation to it
doubt would be simply absurd. More or less the same holds for each
propositional system generated by each linguistic move by each particular
speaker.
Secondly, we have the knowledge that required observation,
measurements, calculations, specific methods of research, etc. This is
collective (scientific) knowledge. In this case we’ve got a variety of
propositions. Some of them may be put into question, but not all. Scientific
knowledge tends to form propositional systems which aim at total rigidity,
even if that ideal is never attained. Now what we assert is
epistemologically legitimate as long as it doesn’t conflict with the body of
scientific propositions. The problem, naturally, is that normal speakers are
Wittgenstein and the Myth of Hinge Propositions 107
not in contact with scientific propositional systems and that’s why very
often people make assertions which strictly speaking they don’t have the
epistemological right to make. For instance, someone can feel entitled to
assert that beings from other galaxies have visited us for he just ignores
that astrophysics has already discarded that possibility (because of the
number of years that they would have to travel, the speed concerned, for
biological reasons, etc.). But I must insist that the problems that might
arise are epistemological or cognitive, but not semantic.
What has so far been stated gives us, I think, a more or less
acceptable apercu of the panorama that, patiently through his grammatical
analyses, Wittgenstein outlines in On Certainty. Let’s quickly see now
what consequences or implications this panorama has for some classical
philosophical discussions.
knowledge. From his perspective, both Moore and the skeptic make the
same mistake: one says that he doesn’t know anything while the other
affirms that he knows a lot of things while both of them misuse the verb ‘to
know’. Their discussion is a typical pseudo-philosophical discussion. Once
we have seen this, we can finally pass to consider “hinge propositions”
themselves which, according to some, unambiguously point to a radical
change in the general outlook of the second Wittgenstein.
remains however that normally we, speakers, lack a clear idea of how we
actually see the world, of what we in fact hold about it, for in general
language has for us a purely practical utility. Thus the propositions which
stand fast for us, which are never modified, which practically never change
their status seem so obvious to us that we just never deign to consider
them. With respect to this I’d like to say a few words.
Since Wittgenstein is not concerned with language from a formal
point of view, the classifications of propositions that he draws are
classifications of roles. Therefore, traditional categories as applied to
propositions, like “analytic” and “synthetic” become in this framework
simply worthless. But then, apart from their roles: what features do
propositions like ‘space is real’ or ‘there are animals’ or ‘I’ve have a wide
range of experiences’ or ‘my life has a temporal dimension’ possess?
To begin with, from the point of view of their internal constitution
they certainly are synthetic. It is evident that they had to be so, for an
analytic proposition would make no contribution to my picture of reality,
since it says nothing, lacks content, is vacuous and results from a
stipulation. On the other hand, given the role they play and the way we
have access to them, there is a sense in which the most fundamental of
bedrock propositions are neither a priori nor not a priori. They are a priori
in the sense they are neither confirmed nor refuted by experience and they
are not a priori for in a sense they were discovered empirically. Thus such
categories simply don’t have a clear-cut application here. This is
understandable: in a sense, those propositions are not established by us. As
we saw, in so far as they simply underlie all our moves in the language-
games, it is language itself which automatically establishes them.14 What
we can do is trace them, although ‘trace’ here doesn’t mean just ‘deducing’
them. It is not an axiomatic system that we build when we get the
propositions from the bed of the propositional river. In fact we learn such
propositions, we get in touch with them, after we learnt how to speak,
much later in our development as speakers. That is, even though there is a
sense in which we can say that they are assumed or presupposed, in the
everyday practice of language they are not required at all. They simply
serve to complete a picture. It is in this sense that what we say
epistemologically depends on them. It is not the case that for what we say
to be meaningful they have to be true; it is rather that for what we say to be
understandable and not be a simple brushstroke on a canvass that they
14
About this I say below something else which I consider is important.
Wittgenstein and the Myth of Hinge Propositions 113
VII. Conclusions
I have argued against the idea of a “third Wittgenstein”, with all it implies.
I hope to have shown that nothing would be more damaging as to invent a
cut in the second Wittgenstein’s philosophical production. I must admit
that I left out lots of interesting subjects. For instance, I didn’t consider the
relations that hold between rules of grammar and “hinge” propositions.
There are those who have maintained that they are the same. In my view
that is a serious mistake. The subject is certainly worth investigating and
debating, for lots of questions would be solved depending upon how we
see the issue. Regardless of this, the reward we get from our discussion is
the idea of a new diagnostic for at least some philosophical puzzles: we are
now in a position to understand that they are not the result of a violation of
rules of grammar (in the Wittgensteinian sense, of course), but rather of the
confrontation between what is being said and the fundamental propositions
which constitute our conception of the world, i.e., our Weltanscahuung,
which underlies our discourse in all spheres of life.
116 Wittgenstein: Issues and Debates
Bibliography
Baker, G. P., 2004: Wittgenstein’s Method: Neglected Aspects. Essays on
Wittgenstein. Blackwell, Oxford.
Cook, John W., 1994: Wittgenstein’s Metaphysics, Cambridge University
Press, Cambridge.
Grice, P., 1989: Studies in the Way of Words, Harvard University Press,
Cambridge, Mas.
Harré, R. 2008: Grammatical Therapy and the Third Wittgenstein,
Metaphilosophy, 39, 4-5, pp. 484 – 491.
--- 2009: Wittgenstein’s Therapies: From Rules to Hinges, New Ideas in
Psychology, 27, 2, pp 118-132.
Hertz, H., 1956: The Principles of Mechanics, Dover Phoenix Editions,
Dover.
Moyal-Sharrock, D. 2003: Logic in Action: Wittgenstein’s Logical
Pragmatism and the Impotence of Scepticism, Philosophical
Investigations, 26, pp. 125-148.
--- 2004: Understanding Wittgenstein’s On Certainty, Palgrave
Macmillan, New York.
Moyal-Sharrock, D. (Ed.), 2004: The Third Wittgenstein. The Post-
Investigations Works, Ashgate Wittgensteinian Studies, Farnham,
Surrey.
Stroll, A., 2002: Wittgenstein, Oneworld, Oxford.
Wittgenstein, L., 1969: On Certainty, Basil Blackwell, Oxford.
--- 1974: Philosophical Investigations, Basil Blackwell, Oxford.
--- 1978: Tractatus Logico-Philosophicus, Routledge and Kegan Paul,
London.
--- 1994: Observaciones sobre los Colores. Translated into Spanish by
Alejandro Tomasini Bassols, Instituto de Investigaciones
Filosóficas/Paidós, Barcelona.
Hacker on Wittgenstein’s Ethnological
Approach
Lars HERTZBERG
1
Hacker, 2010.
2
Wittgenstein, 1998, 45e.
3
Hacker, 2010, 17.
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ),
Ontos Verlag, Frankfurt a. M., 2010, 117-126.
118 Wittgenstein: Issues and Debates
… it is the normative practices of the speech community that fix and hold firm
the internal relations between a word and its application, between explanation
of meaning and what counts, in the practice of using the word, as correct use, as
well as what is determined as following from its use in an utterance.4
This point, as I understand it, might also be made by saying that it is
through speakers’ responses to utterances that distinctions such as that
between correct and incorrect enter into language. The misunderstanding
we need to guard against is that we could base a description of correct use
on neutral observations of the linguistic behaviour of the members of a
speech community. That such an attempt could not achieve its purpose is
clear from the fact that someone who does not herself have command of
the language would not be able to tell what would be the relevant linguistic
features, indeed, would have no way of distinguishing speech from other
forms of behaviour. When Quine argues that the analytic/synthetic
distinction cannot be upheld 5 , the reason for this seems to be that he
considers observation of behaviour the sole source of judgments about
meaning. Observation gives no basis for distinguishing between assent
based on conceptual relations and assent based on agreement concerning
empirical fact. However, if neutral observation were all that is allowed, we
could not even get as far as Quine assumes. We need what Hacker calls an
internal point of view; if that is granted, however, Quine’s reason for
questioning the analytic/synthetic distinction dissolves.6 On the other hand,
as will be seen, I have some disagreements with the way Hacker construes
that point of view.
4
Hacker, 2010, 19.
5
Quine, 1963.
6
There are other reasons for questioning the distinction, but I shall not go into them here.
7
Hacker, 2010, 20.
Hacker on Wittgenstein’s Ethnological Approach 119
8
Wittgenstein, PPF, § 366. References to Wittgenstein’s “Philosophy of Psychology –
A Fragment” (previously known as Part II of Philosophical Investigations) will be
given with PPF and section number.
120 Wittgenstein: Issues and Debates
9
Hacker, 2010, 18.
10
Hacker, 2010, 18.
Hacker on Wittgenstein’s Ethnological Approach 121
13
Hacker, 2010, 19.
14
We should consider too that a normal speaker does not in a robot like fashion repeat
words or phrases she has heard, but utters them with intonations of pleasure, distress,
effort, concentration, etc, or accompanies them with the appropriate expressions.
15
As for the idea that our life with language is governed by rules or conventions (a
frequent misunderstanding of Wittgenstein), consider Wittgenstein, PI, § 83. What
Wittgenstein is suggesting there is that, in many situations, rather than rules
determining how we act, we may pretend to follow rules, make up rules for the
occasion, etc. Letting oneself be bound by rules or conventions is just one way of
relating to them.
Hacker on Wittgenstein’s Ethnological Approach 123
16
At the end of the passage, just quoted Hacker refers to Wittgenstein’s, 1978, 348,
where Wittgenstein speaks of the role of uniformity in learning to speak. But there is
no suggestion that Wittgenstein intends this as a general account of what it means to
learn to speak.
17
This is a central point of Segerdahl et al. 2005.
18
Hacker, 2010, 26.
124 Wittgenstein: Issues and Debates
tell… As the linguistic behavioural repertoire of the child grows, so too the
horizon of possible thought, feeling and volition expands. The child becomes
able to think things he could not conceivably have thought, to feel things he
could not possibly have felt, and to want things that no non-language using
animal could intelligibly be said to want.19
Nevertheless, in representing the learning of concepts as the learning
of a technique, in citing imitation, repetition and recognition as central to
what it means to become a speaker, Hacker conveys the impression that
language is a surface phenomenon, a mere set of conventions, something
that could be skimmed off life like a cream. Surely, this is to misrepresent
the place of language in our lives. This view of things may be an effect of
regarding language learning under the aspect of concept formation, a
perspective, which tempts us to regard the life we live with language in too
abstract terms. (It is true that we find a similar tendency in Wittgenstein’s
own work.). Next, I wish to argue that the emphasis on concepts has
consequences for the way Hacker thinks about the role of philosophical
clarification.
when we reflect on the use of words, not when we use them. This, anyway,
is Wittgenstein’s view. He quotes Augustine’s remark about time as an
expression of the predicament typical of someone in the grips of a
philosophical confusion: “quid est ergo tempus? si nemo ex me quaerat
scio; si quaerenti explicare velim, nescio”.23
Why does Hacker ignore the all-important distinction between use
and reflection? Here, of course, one can only speculate, but I would
suggest that his thinking on this score is a natural consequence of the
central role he gives to concepts. On the view I am attributing to him, the
type of insight the philosopher needs in order to map the conceptual
landscape is of a piece with the knowledge the child acquires in learning to
speak. The philosopher makes explicit the child’s implicit knowledge.
With this goes the idea that there might be a complete account of all the
concepts of our language provided through what Hacker calls explanations
of word meaning. Such an account, if we had it, would forestall the arising
of philosophical puzzlement in advance.
Against this, I would argue that the idea of a complete account of all
our concepts is a chimaera; not because providing such an account would
require an inordinate amount of time, but because the question of what the
account would include is indeterminate. When we explain a word to
someone, the form of our explanation will vary greatly depending on what
the learner already knows or is able to do. Similarly, the type of
clarification needed to resolve a philosophical puzzle will depend on the
nature of our interlocutor’s bewilderment. Thus, we may have to discover
what false analogies lead her thinking astray. As Wittgenstein puts it in PI:
… One might say: an explanation serves to remove or to prevent a
misunderstanding — one, that is, that would arise if not for the explanation, but
not every misunderstanding that I can imagine.24
“There is not almost enough sugar in the pudding” as opposed to “There isn’t nearly
enough sugar in the pudding”, Hacker, 2010, 18. But the inability Hacker is describing
here clearly belongs to the context of reflection, not of use. Besides, it is hard to
imagine the word pair “nearly” and “almost” giving rise to philosophical confusion. In
all, it is hard to see what Hacker’s example is supposed to illustrate.
23
“What then is time? If nobody asks me, I know; but if I want to explain it to
someone who asks me, I don’t know.” The quotation is in Wittgenstein, PI § 89.
Consider also Wittgenstein’s oft-quoted remarks about philosophical confusions
arising when language goes on holiday (Wittgenstein, PI § 38), or “when language is,
as it were, idling, not when it is doing work” (Wittgenstein, PI § 132).
24
Wittgenstein, PI, § 87.
126 Wittgenstein: Issues and Debates
Conclusion
To give a substantial account of language learning is not, of course, a task
for philosophy. Rather, philosophy’s concern with language learning is a
matter of forestalling misunderstandings of what it means to become a
speaker. This requires steering clear of both the Scylla of intellectualism
(treating mastery of words as constituted by knowing how or knowing that)
and the Kharybdis of mechanical conditioning. Both views fail to leave
room for the way in which speaking develops organically within the life of
the child. It is my sense that Hacker has not managed to avoid these risks.
Bibliography
Hacker, P.M.S., 2010: Wittgenstein’s Anthropological and Ethnological
Approach, in: Philosophical Anthropology. Wittgenstein’s Perspective,
(Ed. Jesús Padilla Gálvez), Ontos Verlag, Frankfurt a. M., 2010, 15-32.
Quine, W. V. O., 1963: “Two Dogmas of Empiricism”, reprinted in From a
Logical Point of View. Harper & Row, New York.
Segerdahl, Pär et al., 2005: Kanzi’s Primal Language: The Cultural Initiation of
Primates into Language, Palgrave Macmillan, Basingstoke.
Wittgenstein, Ludwig, 2009: Philosophical Investigations, PI, revised 4th ed.,
Wiley-Blackwell, Chichester.
---
1978; Remarks on the Foundations of Mathematics, revised 3rd ed., Basil
Blackwell, Oxford.
--- 1998: Culture and Value: A Selection from the Posthumous Remains,
German-English Edition, Ed. By G.H. von Wright. Blackwell, Oxford.
Wittgenstein’s Criticism against Gödel’s
Project of Metalogic
Jesús PADILLA GÁLVEZ
Introduction
Within the first decades of the last century, metamathematics had developed
rapidly. This advance took place during a time when there existed a whole
range of competing mathematical programs. Within this period of ongoing
scientific progress, mathematicians used opposing philosophical positions
within metamathematics. However, this provoked relatively few
discussions. For instance, some mathematicians used an axiomatic
approach, others applied an intuitionistic view, and certain questions could
be solved by a Platonist view. G. Cantor’s metaphysic results were taken as
an assumption and elements of intuitionism were used for argumentation.
During the 1930s, the diverse prerequisites of the different approaches to
metamathematics were rarely discussed. However, A. Tarski1, R. Carnap2, K.
Gödel 3 and others appreciated the importance of L. Wittgenstein’s views.
According to A. Tarski, R. Carnap and K. Gödel believed that L.
Wittgenstein’s proposals were fundamental for syntax, metalogic and
metamathematics. R. Carnap expresses this view in the following statement:
My “syntax” has two historic roots: 1. Wittgenstein, 2. Metamathematics
(Tarski, Gödel)4
In 1953, K. Gödel wrote a paper about mathematics in which he
summarized Wittgenstein’s position:
Around 1930 R. Carnap, H. Hahn, and M. Schlick, [...] largely under the
influence of L. Wittgenstein, developed a conception of the nature of
1
Tarski 1936, 11.
2
Carnap, Letter to Neurath of 23. 12. 1933. Hilman-Library, RC 29-03-06 A. Padilla-
Gálvez 1999, 169, footnote 8.
3
Gödel 1995, 171.
4
“Meine ,”Syntax” hat historisch zwei Wurzeln: 1. Wittgenstein, 2. Metamathematik
(Tarski, Gödel)”. Letter from R. Carnap to O. Neurath from 23 December 1933 in:
Archives of Scientific Philosophy, Hilman-Library, RC, Nr: 029-03-06, SI. Padilla-
Gálvez, 1998, 26.
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ),
Ontos Verlag, Frankfurt a. M., 2010, 127-148.
128 Wittgenstein: Issues and Debates
5
Gödel 1986 [1995], III, 334 and Gödel 1995, 171.
6
“Es gibt keine Metamathematik” Wittgenstein PG, Teil II. Über Logik und
Mathematik, 12 and Wittgenstein TS213, § 109, 539-541.
7
Many of the comments in his notebooks were summarized in 1932 in the form of a
typescript, which became an important work of reference (Wittgenstein, Typescript
based on 109-113 and the beginning of 114 (771 pp.), c 1932, Trinity College, 211).
This was a big collection of type-written material containing all his thoughts and ideas
which were prepared for publication. These were to be published later in his
Philosophical Investigations. Wittgenstein seemed to handle his scripts in a rather unusual
way by taking his typescript, cutting it into pieces according to topics of interest and
having these topics retyped. These were to form The Big Typescript? (Wittgenstein,
Typescript consisting of cut part from 208, 210 and 211 (3 boxes), c 1932-3, Trinity
College, 212). In the course of this procedure he took his topics of interests, put them in
several folders which he indexed by titles. One of these folders was labelled
“Metamathematics does not exist”. Both, the original typescript and The Big Typescript
contain parts of texts which were taken out of their context (Wittgenstein, The “Big
Typescript” (776 pp), c 1933, Trinity College, 213).
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 129
1. Gödel’s Result
D. Hilbert underlines that strict formalization of a theory requires a full
abstraction from the meaning.8 The result is a formal system or formalism.
Metamathematics involves the view of approaching the formal system as a
whole. Metamathematics includes the description or definition of formal
systems as well as the investigation of the properties of such formal systems.
One of the most important aims of metamathematics is to provide a
justification for the predicate calculus. It is rather based on mathematics and
logic rather than on ordinary language. Consequently, its task is to find a proof
for the internal consistency of a formal system. Any theory of calculus
should avoid contradictions. Therefore a theory of a formula A cannot
simultaneously prove A and non-A. The internal consistency of logic
corresponds to that of mathematics. 9 D. Hilbert expressed this thought as
8
Hilbert, 1923, 151-165. and Hilbert / Ackermann, 1972.
9
Wittgenstein summarized his discussions of 28 December 1930 as follows: The
problem of internal consistency of mathematics stems from two different sources: 1.
From the ideas of non-Euclidean geometry, which was dealing with the problem of
proving the parallel axiom after the principle of reductio ad absurdum. 2. From the
antinoms of Burali-Forti and from Russell. Also: “Das Problem der Widerspruchsfreiheit
der Mathematik stammt aus zwei Quellen: 1. Aus der Ideen der nicht-euklidischen
Geometrie, wo es sich darum gehandelt hat, nach dem gegebenen Vorbild einer reductio
ad absurdum das Parallelaxiom zu beweisen. 2. Aus den Antinomien von Burali-Forti
und von Russell.” (Wittgenstein, WWK, 121). The arguments can be outlined as
follows: The problem of internal consistency is derived from the antinomy. But
Wittgenstein thinks that these two elements don’t have anything in common, because the
antinomy never occurs in the calculus but it appears in the ordinary language. The reason
is that words are often used in an ambiguous way. This problem can be solved by using
words precisely and distinctly, which would lead to the disappearance of antinomy,
“but by analysis rather than by proof.” (Wittgenstein, WWK, 122) Then the types of
different proofs applied in mathematics were listed: 1. The proof, that gives evidence for a
certain formula. This formula occurs in the proof as its last element. 2. The inference proof.
What is striking is that the proposition that is supposed to be proved does not appear in the
proof. This means, that the inference is not a method that leads to a proposition, but it
shows us infinite possibilities, which is a relevant characteristic of the inference proof
(Wittgenstein, WWK, 135). Mathematical induction is a paradigm of a method for
proving generality propositions about the natural numbers. (When Poincaré defends
mathematical induction as an irreducible tool of mathematical reasoning, is also a
forerunner of the constructivist point of view (See: Poincaré 1902). A proof by
130 Wittgenstein: Issues and Debates
follows: “The absolute truths are those insights that were given by my theory
of proof and the internal consistency of those formula-systems.”10
In an article written in 1930, K. Gödel obtained his results by
metamathematic reasoning that has a formal structure and is considered a
system of objects. What is new about this view? He starts from the
assumption that objects of a formal system are formal symbols, formal
expressions and finite sequences of formal expressions. There is an
enumerable infinity of formal symbols given at the outset. Formal objects
form an enumerable class. By specifying a particular enumeration of them,
and letting, our metamathematical statements refer to the indices in the
enumeration instead of the objects enumerated. Metamathematics becomes
a branch of number theory. In this result, the use of G. Cantor’s diagonal
induction of the proposition for all n, P(n) shows that any given n would have to have
the property P, by reasoning which uses only the numbers from 0 up to n. Certainly,
for a particular proof by induction to be constructive, also the reasoning’s used within
its basis and induction step must be intuitionistic. An existence statement there exists a
natural number n having the property P, or briefly there exists an n such that P(n), has
its constructivist meaning as a partial communication of a statement giving a particular
example of a natural number n which has the property P, or at least giving a method by
which in principle one could find such an example. Therefore a constructive proof of
the proposition there exists an n such that P(n) must be constructive in the following
sense. The proof actually exhibits an example of an n such that P(n), or at least
indicates a method by which one could in principle find such an example. L.
Wittgenstein believes that the antinomy has encouraged him to deal with the problem of
internal consistency. His line of argument seems to be wrong in this aspect because the
antinomy does not relate to the internal consistency of mathematics. The antinomy
normally appears in the ordinary language rather man in the calculus, because words are
often used in an ambiguous way. The solution to this problem would be to replace
ambiguous meanings with precise ones. The antinomy disappears through analysis rather
than by proof. Thus, if ambiguity leads to contradictions in mathematics, it can never be
solved by a proof. And as a result of this, he states, that there can never be a proof for
internal consistency (comparing the contradictions in mathematics to those in predicate
calculus). On the basis of the discussion with Hilbert’s ‘New Foundation of
Mathematics’, Wittgenstein points out, that a proof for internal consistency must evolve
to understand the rules. These rules are usually described by the inference proof. The
inference suggests infinite possibilities. A contradiction cannot be considered as rule
because the grammar of the word “rule” is of a kind, that a contradiction is not a rule
(Wittgenstein, WWK, 194).
10
“Als die absoluten Wahrheiten sind vielmehr die Einsichten einzusehen, die durch
meine Beweistheorie hinsichtlich der Beweisbarkeit und der Widerspruchsfreiheit jener
Formelsysteme geliefert werden.” Hilbert, 1923, 153.
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 131
11
L. Löwenheim proved that for any set of propositions of standard predicate logic, if
there is an interpretation in which they are true in some domain, there is also an
interpretation that makes them true in a countable subset of the original domain
(Löwenheim, 1915, 447-470 and Skolem, 1929, 1 ff.). The result was called a paradox
since it was believed that certain sets of axioms characterised the real numbers, and
now Löwenheim’s result showed that the same axioms must hold in a countable subset
of the real numbers. It also seemed to attempt to axiomatic set theory somewhat of a
problem. Although it seems to contradict common sense (as do other results which
depend on the Axiom of Choice), there is no paradox. The result implies that no
uncountable mathematical system, such as those involved in analysis, geometry, and
set theory, can be characterised up to isomorphism using only first-order propositions.
If one examines the case of the real numbers more closely, then the axioms for an
ordered field are all first-order propositions. Löwenheim’s result then shows that the
real numbers contain a countable ordered field, which then cannot satisfy the least-
upper-bound axiom, which is a second-order proposition.
132 Wittgenstein: Issues and Debates
Wittgenstein was aware of the fact that many metamathematic results had used
Cantor’s diagonal method. Consequently, he was interested in the tacit
presuppositions on which Cantor’s proof was based. In this paper, we will
analyse this second objection.
15
Cantor, 1890-1, 75-78 and Cantor, 1980, 278-281. L. Wittgenstein challenged the
belief according to which the rules of classic logic have absolute validity, irrespective of the
content to which it is applied. His discussion of two obvious paradigms reveals that
principles that are valid for thinking about finite sets cannot be transferred to infinite sets.
One is the principle that the whole is greater than any proper part, when applied to a 1-1
correspondence among sets. Another is that a set of natural numbers contains a greatest
number. A principle of classical logic, valid in reasoning about finite sets, which
Wittgenstein does not accept for infinite sets, is the law of the excluded middle. In its
general form the law assumes for every proposition A, either A or non-A. If A is the
proposition there exists a member of the set S that have the property P. Then non-A is
equivalent to every member of S that does not have the property P, or in other words
every member of S has the property non-P. The law, applied to this A, hence gives either
there exists a member of S having the property P, or every member of S has the property
non-P. For definiteness, let us specify P to be a property such that, for any given member
of S, we can determine whether that member has the property P or does not. Now
suppose S is a finite set (Now suppose the set S is finite). Then we could examine every
member of (the set) S in turn, and thus either find a member having the property P, or
verify that all members have the property not-P. There might be practical difficulties,
e.g. when S is a very large set having say a billion members, or even for a small (set) S
when the determination whether or not a given member has the property P may be
tedious. But the possibility of completing the search is possible. It is this possibility
(choice) which for the constructive point of view makes the law of the excluded middle a
134 Wittgenstein: Issues and Debates
valid principle for reasoning with finite sets S and properties P of the kind specified. For
an infinite set, the situation is basically different. It is no longer possible in principle to
search through the entire set. Moreover, in a situation the law is not saved for the
constructivist by substituting, for the impossible search through all the members of the
infinite set, a mathematical solution of the problem posed. We may in some cases, i.e. for
some sets S and properties P, succeed in finding a member of S having the property P;
and in other cases, succeed in showing by mathematical reasoning that every member of
S has the property not-P, e.g. by deducing a contradiction from the assumption that an
arbitrary member of S has the property P. However, we have no ground for affirming the
possibility of obtaining either one or the other of these kinds of solutions in every case.
L. Wittgenstein differs essentially in their view of the infinite from the theories of
Weierstrass, Dedekind and Cantor. From the constructive point of view, the infinite is
treated as potential or constructive. In the classical position, the infinite is treated as
actual or completed. An infinite set is regarded as existing as a completed totality. L.
Wittgenstein’s criticism of the Gödel’s metalogic as applied to an infinite set S arises
from this perspective respecting infinity.
16
In mathematics, an uncountable set is an infinite set, which is too big to be
countable. The uncountability of a set is closely related to its cardinal number; a set is
uncountable if its cardinal number is larger than that of the natural numbers. The
related term nondenumerable some authors use set as a synonym, for “uncountable
set” while other authors define a set to be nondenumerable if it is not an infinite
countable set. There are many equivalent characterizations of uncountability. A set X
is uncountable if and only if any of the following conditions holds: 1. there is no
injective function from X to the set of natural numbers. 2. X is nonempty and any Ȧ-
sequence of elements of X fails to include at least one element of X. That is, X is
nonempty and there is no surjective function from the natural numbers to X. 3. the
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 135
The result of the diagonal method is to reveal with simple means that no list
or sequence of real numbers can ever include all real numbers. Therefore,
the totality of real numbers is uncountable. We assume that there is an
infinite sequence within arithmetic language. Within this language, a
property of numbers is expressed through a proposition. This description
results in a contradiction.
G. Cantor asked if the set of real numbers ( ) is uncountable. He
analysed the open interval (0,1) and started by claiming that it is not
countable. In order to prove this contradiction we assume that (0,1) be
f
countable. This leads to the following relation: x ¦x
j 1
j 2
j
, whereby
xj{0,1} and xj0. If j then is xj=1 and jj0 is not allowed. We then
denote (xj)j as a sequence with a range of {0,1}, which means that
x: ĺ{0,1} is an arbitrary mapping. If the transformation from A to B
through mapping as Map(A,B) is (sufficiently) defined it results in the
following injective mapping: \:(0,1)
x(x(j))j Map( {0,1}), with
Map( {0,1})\\((0,1)) being countable. If (0,1) is countable the same is
true for \((0,1)), and consequently the Map(( {0,1}) is also countable.
That produces the following bijective mapping: Ȝ:
kx(k)Map( {0,1}).
In that case, we can apply G. Cantor’s diagonal method in the following
form:
1 2 3 4 5
4 x(4)(1) x(4)(2)
5 x(5)(1)
cardinality of X is neither finite nor equal to ʠ0. 4. the set X has cardinality strictly
greater than ʠ0. The best known example of an uncountable set is the set of all real
numbers; Cantor's diagonal argument shows that this set is uncountable. The
diagonalization proof technique can also be used to show that several other sets are
uncountable, such as the set of all infinite sequences of natural numbers and the set of
all subsets of the set of natural numbers. The Cantor set is an uncountable subset of .
136 Wittgenstein: Issues and Debates
3. Wittgenstein’s Alternative
L. Wittgenstein analysed the assumption according to which the diagonal
sequence d defines a b-adic fraction and therefore denotes a real number. In
this respect he doubted whether Cantor’s diagonal argument could actually
prove the uncountability of real numbers.18 Wittgenstein presupposed that
the diagonal method would not create an infinite amount of real numbers
countable by means of the finite method. He pointed out that the sequence of
numbers produced by the diagonal method directly depended on the
axiomatic system and not on the diagonal method. He said:
Wendet man meine Betrachtung auf das Cantorsche Diagonalverfahren
an so ergibt sich: Eine unendliche Menge von Dezimalbrüchen kann nur
ein Gesetz bedeuten nach dem Gesetze gebildet
0 a11a12 a13a14 .... werden und das heißt eigentlich eine Funktion von
0 a12 a22 a23a24 .... zwei veränderlischen. F(x,y) ist die funktion von
0 a31a32 a33a34
zwei Veränderlichen. F(x,n) ist der n-te von ihnen
und F(m,n) seine m-te Stelle. Der Dezimalbruch
nach der Diagonale genommen ist F(x,x) und
verändert lautet er etwa F(x,x)+1 (dazu müßte festgesetzt werden, daß
0+1=1, 1+1=2, ...., q+1=0 etc ist) Und nun zeigt ein Inductionbeweis
daß F(x,x)+1 eine andere Entwicklung hat als jedes beliebige F(x,y). Wo
aber ist hier das höhere Unendliche? (oder gar das “eigentlich
Unendliche”).19
17
Wittgenstein 1978, 121.
18
Wittgenstein 1974, BGM, II, §§ 1-8, 125 ff.
19
Wittgenstein 1999, WA, 2, 268.
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 137
Formula (4) shows some relevant aspects, such as that an existing limit
lim
(D ) does not automatically require a diagonal number that functions as
i i d n
n of (Vk ) k d 10 n
a limit of the sequence of its finite expansion. Neither does it require a range
20
Wittgenstein 1974, BGM, II, §§ 9-15, 127-129.
21
Cantor, 1890-1, 75-78 and Cantor, 1980, 278-281.
22
Redecker, 2006, 67.
23
Wittgenstein 1974, BGM, II, § 9, 127.
138 Wittgenstein: Issues and Debates
n
of all infinite decimal numbers. The last part such as lim 0 indicates that
nof 10 n
for every value İ>0 there is a natural number, so that for all n the formula
n
<İ is true.24 The farer the diagonal number is expanded; the smaller will
10 n
be the part of the decimal fraction of equal length that is included in the
diagonal number. Therefore, L. Wittgenstein held that if the finite expansion
of the diagonal numbers converges against the diagonal number, then part of
the list that is crossed by the diagonalnumber, converges towards cero.
Conversely, G. Cantor’s point of view as described in (3) is incompatible
with Wittgenstein’s view shown in (4). This incompatibility is reflected in
the differing interpretations of the diagonal method as part of Wittgenstein’s
proposal. In his remarks on the foundations of mathematics, he depicted the
following two different figures for the expansion of diagonal numbers:25
The above figure shows Cantor’s version whereas the one below
represents Wittgenstein’s view. Whereas Cantor’s version terminates in an
n-ary horizontal sequence, Wittgenstein expands it with the remark “ad. Inf”
with a n+x-ary horizontal sequence. According to Wittgenstein’s view we
are not confronted with a mapping, but rather with an allusion of infinity:
“...die Andeutung der Unendlichkeit”. 26 L. Wittgenstein argued that the
diagonal proof verifies neither that the diagonal number is a real number nor
that it is different from all numbers of the sequence of real numbers. He
maintained that every transformation of the diagonal row is mapped as an
element of the sequence, with which this transformation must be identical.27
He assumed that if a subsequence of the enumeration existed, it converges
24
Forster, 1992, 19 and Redecker, 2006, 66.
25
Wittgenstein 1974, BGM, II, § 11, 128.
26
Wittgenstein 1974, BGM, II, § 11, 128.
27
Wittgenstein 1974, BGM, II, § 9, 127 f.
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 139
against the diagonal number, there must always exist a value identical with
the diagonal number.
28
Haller 1981, 57 ff.
29
“Hilbert stellt Regeln eines bestimmten Kalküls als Regel / der / einer / Metamathematik
auf.” Wittgenstein 153a, 136 links.
30
“Zu der eigentlichen so formalisierten Mathematik kommt eine gewissermaßen neue
Mathematik, eine Metamathematik, die zur Sicherung jener notwendig ist, in der -im
Gegensatz zu den rein formalen Schlußbweisen der eigentlichen Mathematik- das inhaltliche
Schließen zur Anwendung kommt, aber lediglich zum Nachweis der Widerspruchsfreiheit
der Axiome. In dieser Metamathematik wird mit den Beweisen der eigentlichen Mathematik
operiert, und diese letzteren bilden selbst den Gegenstand der inhaltlichen Untersuchung.”
Hilbert 1923, 153.
31
“Wird speziell die Mathematik in formalisierter Gestalt, so heißt die zugehörige
Metasprache die Metamathematik.” Hilbert/Ackermann 1972, 162.
140 Wittgenstein: Issues and Debates
32
“Wir mischen uns nicht in das, was der Mathematiker tut, erst wenn er behauptet
Metamathematik zu treiben, dann kontrollieren wir ihn.” Wittgenstein 114, 30 v.
33
Hilbert 1923, 151-165.
34
Carnap 1935.
35
Heyting 1930, 42-56.
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 141
36
“[...] operieren nach festgelegten Regeln // d.h. als Vorgang nach festgesetzten
Regeln//” Wittgenstein 211, 40.
37
“Und sollen wir also nicht von Regeln im Allgemeinen reden, wie auch nicht von
Sprachen im Allgemeinen? Sondern nur von Regeln in besonderen Fallen.” Wittgenstein
211,41.
38
“Hinter die Regeln kann man nicht dringen, weil es kein Dahinter gibt.” Wittgenstein
211, 42.
39
Wittgenstein 211, 44.
40
“Das ist klar, daß die Frage “was ist ein Kalkül” von genau der gleichen Art ist wie die:
“was ist ein Spiel” oder wie die: “was ist eine Regel.” WA 4, 37 and 211,44.
41
Wittgenstein TS213, 539.
42
“Ich sagte oben “Kalkül ist kein mathematischer Begriff; dass heisst, das Wort,
‘Kalkül’ ist kein Schachstein der Mathematik.” WA4, 39;TS213, 539 and 211, 45.
142 Wittgenstein: Issues and Debates
43
“Es brauchte in der Mathematik nicht vorzukommen. - Und wenn es doch in einem
Kalkül gebraucht wird, so ist dieser nun kein Metakalkül. Vielmehr ist dann dieses
Wort wieder nur ein Schachstein wie alle andern.” WA 4,39; TS213, 539 and
TS211,45.
44
“(Hilbert stellt Regeln eines bestimmten Kalküls als Regeln einer /der/
Metamathematik auf.)” PG297 and TS213, TS539.
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 143
45
Wittgenstein 211, 46.
46
Wittgenstein 211, 60.
47
“Die Logik und die Mathematik ruht nicht auf Axiomen; so wenig eine Gruppe auf den
sie defínierenden Elementen und Operationen ruht. Hierin liegt der Fehler, das
Einleuchten, die Evidenz, der Grundgesetze als Kriterium der Richtigkeit in der Logik zu
betrachten. Ein Fundament, das auf nichts steht, ist ein schlechtes Fundament.” PG, 297 and
TS213,540 f.
144 Wittgenstein: Issues and Debates
Conclusion
In 1933, Ludwig Wittgenstein wrote a chapter in the Foundations of
Mathematics of the Big Typescript that had the provocative title:
“Metamathematics does not exist”. This chapter presents a criticism of
Gödel’s metalogic point of view. We examined the frame within which the
meta-mathematical discussion took place in Vienna at that time. Right from
the beginning, L. Wittgernstein was critical towards Gödel’s project of
metalogic. We showed Wittgenstein’s critical objections against Gödel’s
project of metalogic. In fact, it was revealed that the project has left several
problems unexplained. It was shown that the concepts of proposition and
provable remain inconclusive and that the application of Cantor’s diagonal
method generates problems. Wittgenstein’s arguments create a new point of
view in this field. L. Wittgenstein has shown that many equivocal
48
“Gödel<s> sagt in indirekter Weise aus, dass er nicht beweisbar ist.” - was sagt also das
Gogenteil von “dh. die Verneinung” von Gödels aus Satz aus? der verneinte Satz
Gödel<sche> Satz aus
Folgt aus “įp ist beweisb. “įp”? D.H.: folgt daraus, daß “p” die (interne) Eigenschaft der
Beweisbarkeit hat, daß es wahr ist? - Der Beweis für die Beweisbarkeit gilt allerdings als
Beweis von “įp”, aber das heißt nicht, daß man aus dem unbewiesenen Satz “įp ist
beweisbar” folgern darf.” Trinity College Library, ítem 121, Bd. XVII Philosophische
Bemerkungen, 82v.
49
“Ich möchte beinahe sagen: wofür immer Du Dich entscheidest, entscheide Dich nicht
aus dem Gödelschen Grund, denn das ist ein dummer Grund.” Trinity College Library,
ítem 121, Bd. XVII Philosophische Bemerkungen, 82r.
Wittgenstein’s Criticism against Gödel’s Project of Metalogic 145
Bibliography
Cantor, Georg, 1890-1: Über eine elementare Frage der
Mannigfaltigkeitslehre. Jahresbericht der deutsche Mathematiker-
Vereinigung, Vol. I, 75-78.
--- 1980: Gesammalte Abhandlungen mathematischen und
philosophischen Inhalts, Springer Verlag, Berlin, Heidelberg, New
York.
Carnap, Rudolf, 1935: Logische Syntax der Sprache, Springer, Wien / New
York.
--- 1995: Metalógica - Metalogik, Ed. J. Padilla Gálvez, Mathesis, 11,
137-192.
--- 2000: Untersuchungen zur allgemeinen Axiomatik, Thomas Bonk and
Jesús Mosterín, Eds., Wissenschaftliche Buchgesellschaft, Darmstadt.
Forster, Otto, 1992: Analysis 1, Vieweg, Braunschweig.
Gödel, Kurt, 1930: Einige metamathematische Resultate über
Entscheidungsdefinitheit und Widerspruchsfreiheit, Anzeiger der
Akademie der Wissenschaften in Wien, mathematisch-
naturwissenschaftliche Klasse, 67, 214-215.
--- 1931: Über formal unentscheidbare Satze der Principia Mathematica
und verwandter Systeme I, Monatshefte für Mathematik und Physik, 38,
173-198.
--- 1986ff.: Kurt Gödel. Collected Works, Vols. I / II / III. Solomon
Feferman et alii., Eds., Clarendon Press, Oxford.
--- 1995: Unpublished Philosophical Essays, Francisco A. Rodríguez
Consuegra, Ed., Birkhäuser, Basel, Boston, Berlin.
Haller, Rudolf, 1981; Die gemeinsame menschliche Handlungsweise, Sprache
und Erkenntnis als soziale Tatsache, R. Haller ed., Hölder-Pichler-
Tempsky, Wien, 57-68.
Heyting, Arend, 1930: Die formalen Regeln der intuitionistischen Logik,
Sitzungsberichte der Preussischen Akademie der Wissenschaften.
Physikalisch-Mathematische Klasse II, 42-56.
146 Wittgenstein: Issues and Debates
L. Wittgenstein
AWL : Ambrose, A., Wittgenstein’s Lectures. Cambridge, 1932-1935
BF : Bemerkungen über die Farben
BGM : Bemerkungen über die Grundlagen der Mathematik
BPP : Bemerkungen über die Philosophie der Psychologie
BT : The Big Typescript
BB : The Blue and Brown Books
CV : Culture and Value
LCA : Lectures and Conversations on Aesthetics, Psychology and
Religious Belief
LE : Lectures on Ethics
LFM : Lectures on the Foundations of Mathematics
LO : Letters to Ogden
LRKM : Letters to Russell, Keynes and Moore
LWPP : Last Writings on the Philosophy of Psychology
LWL : Lee, D., Wittgenstein’s Lectures. Cambridge, 1930-1932
MSS : Wittgenstein’s Nachlass. The Bergen Electronic Edition
MWL : Moore, G. E., Wittgenstein’s Lectures in 1930-33
NB : Notebooks 1914 – 1916
NL : Notes for Lectures on Private Experience and Sensedata
OC : On Certainty
PB : Philosophische Bemerkungen
PG : Philosophische Grammatik / Philosophical Grammar
PI : Philosophical Investigations
PO : Philosophical Occasions 1912 – 1951
PR : Philosophical Remarks
PU : Philosophische Untersuchungen
PT : Prototractatus
Wittgenstein: Issues and Debates, (Eds. Eric LEMAIRE and Jesús PADILLA GÁLVEZ),
Ontos Verlag, Frankfurt a. M., 2010, 149-150.
150 Wittgenstein: Issues and Debates
RC : Remarks on Color
RFM : Remarks on the Foundations of Mathematics
RPP : Remarks on the Philosophy of Psychology
RLF : Some Remarks on Logical Form
TB : Tagebücher 1914 – 1916
TLP : Tractatus Logico-Philosophicus
TSS : Wittgenstein’s Nachlass. The Bergen Electronic Edition
ÜG : Über Gewißheit
VB : Vermischte Bemerkungen / Culture and Value
VW : The Voices of Wittgenstein. The Vienna Circle
WL : Wittgenstein’s Lectures
WN : Wittgenstein’s Nachlass
WWK : Wittgenstein und der Wiener Kreis
Z : Zettel