On the Idea of Analysis in the Late Wittgenstein

Jakub Mácha


Wittgenstein is committed to the idea of analysis in the Tractatus. His logical analysis aims to
reveal a hidden form of the proposition in terms of ultimate elements and their mutual logical
relations. Wittgenstein abandoned this metaphysical commitment in his late philosophy. The aim of
this paper is to show that Wittgenstein put forward a restricted idea of analysis throughout his
philosophical writings. This idea relies on two theoretical distinctions. The first is that between
cause and reason and the second between external and internal relations. I shall argue that the
former distinction is a special case of the latter and so, for our purposes, we can take it as one
fundamental difference. This difference might be confused in the surface grammar. The aim of
Wittgenstein’s analysis is to reveal the difference. Finally, I will indicate some philosophical
problems that can be surmounted using this difference.


It is often argued – even by the most resolute interpreters[1] – that in the Tractatus,
Wittgenstein is committed to the idea of analysis. In this respect, he was inspired by Russell’s
treatment of propositions containing definite descriptions. For Russell an analysis is “the
discovery of the constituents and the manner of combination of a given complex” (Russell 1984,
119). This manner of combination of a complex proposition is its logical form, whose revelation is
the goal of logical analysis. The logical form is given in terms of ultimate elements and their
mutual logical relations. Wittgenstein adopted this idea with one important difference. While
Russell was convinced that ordinary language does not need to reflect the proper logical form, for
Wittgenstein ordinary language was perfectly in order. Ordinary language has to be, according to
Russell, transformed into a logically ideal language in order to reflect the proper logical form.
His logical analysis starts from ordinary language and proceeds to ideal language, from which the
logical form can be derived. Wittgenstein did not see any need for this intermediate step. His
logical analysis derives the logical form from ordinary language. Indeed, Wittgenstein maintained
that a “proposition has one and only one complete analysis” (TLP, 3.25).


Wittgenstein repudiated much of this picture in his later philosophy. The ultimate goal of a
philosophical investigation is to clear away misunderstandings that can be caused by false
analogies between different regions of language. In order to dismantle a false analogy, some
expressions can be removed or substituted for other expressions. This method may be called
“analysis” only in a derived sense (see PI §90). Such an analysis bears no metaphysical commitment
to ultimate logical elements and actually to the whole idea of one and only one correct logical
form. Wittgenstein wrote indeed that “nothing is hidden” (PI §435).

I would like to argue that false analogies and ipso facto philosophical misunderstandings may be
caused by ambiguous words and sentences and, further, that it was Wittgenstein’s intention to
tackle such ambiguities. To do so, first one has to detect an ambiguity and, second, there has to
be a rule to resolve the ambiguity. Sometimes it is enough to point out that an expression causing
problems is ambiguous. But even this presupposes a generic distinction that makes it possible to
detach separate meanings. In what follows I will examine two such distinctions in Wittgenstein’s
works.

Cause and reason

The distinction between cause and reason can be found in Wittgenstein’s remarks from the 1930s
onwards.[2] It has proved to be a very powerful tool to surmount philosophical problems. First,
take a look at expressions that could be ambiguous this way. The conjunction “because” may
introduce a statement of the cause or of the reason. The conjunction “why” has the same ambiguity.
Even the word “cause” itself shows the cause-reason ambiguity. Consider the following examples from
Wittgenstein’s Lectures and Conversations on Aesthetics:


‘Cause’ is used in very many different ways, e.g.

(1)   “What is the cause of unemployment?” “What is the cause of this expression?” [Experiment and
statistics]

(2)   “What was the cause of your jumping?” “That noise.” [Reason]

(3)   What was the cause of that wheel going round?” You trace a mechanism. [Mechanism] (LA 1966
II.12)


Wittgenstein wants to reserve the expression “cause” for a (relation of) mechanical causality
between two events. A cause in this sense can be found statistically or by tracing an underlying
mechanism. The relation of “being a reason for” connects two propositions. Apart from the reason,
Wittgenstein speaks sometimes of the motive or the ground.[3] Between a proposition and its reason
there must be a conceptual (logical or even mathematical, e.g. a calculation) relation. Let me
outline the differences between these two sorts of relations. One could say that a proposition
somehow contains its reason unlike a cause, which is not contained in its effect. The
cause-effect-relation is only probabilistic; a cause and its effect are only cases of concomitance.
This is true when we arrived at this relation statistically, but also when we know the mechanism. A
logical relation of having a reason is, on the other hand, rigid and strict. Here is an example:


"I write the number '16' here because it says 'x²' there."

It might appear that some causality was operating here, but that would be a confusion between
'reason' and 'cause'. (PG §61)


If I calculate the second power of four, the result will be 16, because this is the correct way of
performing this operation, i.e. this result is in accordance with the rule of squaring. I might,
however, make a mistake, because I did not pay much attention to the calculation. There is a
certain probability of making a mistake. The first “because” introduces a reason, the second one a
cause. That the second power of four is sixteen holds with certainty. And again, the mistake also
has a ground: The second power of four is not 17 because this violates the rule of squaring (see OC
§74).

The next dissimilarity between reason and cause, which can hardly be overstressed, is that not all
propositions need to have a reason as opposed to all events having a cause or causes.[4]
Wittgenstein writes: “A reason can only be given within a game. The links of the chain of reasons
come to an end, at the boundary of the game. (Reason and cause.)” (PG, §97) There are propositions
that cannot (or do not need to) be justified within a given language-game: “And this again joins on
to the confusion between cause and reason. We need have no reason to follow the rule as we do. The
chain of reasons has an end.” (BBB, 143) Rules are precisely propositions we do not need to give
reasons for – in a given language-game. This does not exclude the possibility that there could be a
reason required in another language-game. The chain of reasons reaches (sooner or later) the
boundary of a language-game. If this boundary were exceeded then it would be nonsense or at least
another language-game. This implies that there could be unexplainable facts.[5]

Wittgenstein uses this difference to cope with philosophical confusions. Let me mention some of
these: (1) The meaning of a word is for him given in the explanation of meaning. The same word
might be connected with a characteristic sensation. There is the cause-effect relation between the
uttering of a word and its characteristic sensation, but an explanation of meaning might give a
reason for the uttering of the word (PG, p. 60). (2) Pleasure is not a sensation, for pleasure is
always “in something”, which means it has a reason. Between a sensation and pleasure is only the
cause-effect relation (Z §507). (3) Wittgenstein sees no way for causal relations to achieve the
certainty of logical relations. This would presuppose a sort of super-mechanism – that is, a
mechanism that cannot break down. Our mind is, however, no such super-mechanism. Therefore,
mechanical processes in our brains cannot explain the rigidity of human reasoning (LA 1966
II.23ff).

Internal and external relations

The terminology of internal and external relations was common among Wittgenstein’s philosophical
colleagues in the early 20^th century. Most significantly, Russell and Moore argued against
Neo-Hegelians that all relations cannot be internal. The distinction plays a very central role in
Wittgenstein’s early Notebooks and in the Tractatus as well. Wittgenstein defines an internal
relation (and internal property) so that “it is unthinkable that its object [or objects] should not
possess it.” (TLP 4.123).  There lies an internal relation between language and the world in the
heart of the Tractatus. Internal relations persist, however, between (typically) two structures as
abstract objects, i.e. forms (4.122). Therefore, what an expression (a sign) shares with reality is
its logical form.[6]

There is elementary disagreement among scholars concerning this internal relation of depicting.
According to the traditional (metaphysical) readings of the Tractatus, this internal relation
connects the logical structure of language and the logical form of the world, which comprises how
elementary objects are combined in states of affairs. According to recent opposite approaches, the
relation of depicting persists between a sign and what it signifies (the anti-metaphysical
approach, McGinn) or the distinction between internal and external relations as such is
inconsistent (the resolute reading, Diamond).[7] Although I believe the textual evidence supports
rather the traditional metaphysical view, the metaphysical and anti-metaphysical views can be
reconciled here. Consider once again Wittgenstein’s remark quoted in footnote 6. A sign signifies
something due to its internal property (i.e. unary relation), which is identical (i.e. related by
the internal relation of identity) to the same property of a state of affairs.

Whatever the case may be, Wittgenstein in his late philosophy conceives internal relations within
language only. Already in the Tractatus (4.123), he gives an example of an internal relation
between two color shades. Much later in his last Remarks on Colours (§1),  he points out that the
same proposition may express an internal as well as an external relation. The proposition “X is
lighter than Y” could be a temporal proposition expressing an external relation between two objects
or an internal relation between two color shades. Wittgenstein varies this example in other places:
A similarity between two faces could be an external relation between two men or an internal
relation between two shapes (LW I, §155nn). The outward form of the proposition is the same. The
proposition is, however, ambiguous with respect to the expression of an internal or external
relation. This ambiguity occurs and causes confusion only if one treats such a proposition in
insolation. If a sufficient context of a language-game were provided, no such confusion could
arise.

Thus, a given proposition can express an internal relation in one language-game and an external
relation in another. For a relation to be internal, it must express internal properties of language
or of a language-game. Propositions expressing internal relations are, strictly speaking, not
genuine propositions at all. They are expressing rules of a language-game. To put it simply,
internal relations are directed to language, external relations to what language is about. To
reveal this, in conclusion, is the ultimate goal of logical or linguistic analysis, which
Wittgenstein always maintained.

An internal relation is a conceptual relation, since it holds between language forms, and there
must be operations transforming one form into another and revealing that these forms are, in fact,
identical.  Then the relation between a proposition and its reason must be an internal relation,
for it is conceptual and such a connection cannot be contingent. On the other hand, the
cause-effect relation holds between separate objects or events and thus must be external. The
cause-effect relation is actually an instance of external relation par excellence.

Like the reason-cause distinction, Wittgenstein used this more general and abstract distinction
between internal and external relations to make philosophical problems clear. Here are the most
important ones: (1) Russell’s causal account of intention is based on external relations. Hence,
intention and its object are, for him, distinct objects or events. Wittgenstein maintains, to the
contrary, that there must be an internal relation of identity: “For expecting that p will be the
case must be the same as expecting that this expectation will be fulfilled” (PB 1975, 25, p.
65).[8] (2) Mathematical calculations demonstrate internal properties of structures (RFM I, 99)
rather than external properties of objects. The relation between a mathematical proposition and its
proof is an internal one (RFM VII, 6ff). (3) The importance of the phenomenon of seeing-as lies in
the fact that some apparent descriptions of objects (i.e. ascribing external properties) are
actually expressions of internal relations. If one sees a picture of a rabbit as a duck, then it is
a case of perceiving an internal relation between this rabbit-shape and the concept of a duck – or
a duck-shape. (See PI II, p. 212.)


References

References to Wittgenstein’s works use the abbreviations introduced by Wittgenstein

Archives at the University of Bergen.


Conant, James and Diamond, Cora 2004 “On Reading the Tractatus Resolutely”, in: Max Kölbel and
Bernhard Weiss (eds.), Wittgenstein’s Lasting Significance. London: Routledge, 46-99.

McGinn, Marie 2010 “Wittgenstein and Internal Relations”, European Journal of Philosophy 18(4),
495-509.

Russell, Bertrand 1984 Theory of Knowledge: The 1913 Manuscript, London: George Allen and Unwin.
________________________________

[1] See Conant & Diamond 2004, 82.

[2] The identification or unification of cause and reason is embodied in the Latin expression
“causa sive ratio” of modern times which can be found in Spinoza or Leibniz. They used this
expression inter alia in formulating the Principle of Sufficient Reason, which states that
“everything must have reason or cause”. Schopenhauer, who might have inspired Wittgenstein here, in
his dissertation The Fourfold Root of the Principle of Sufficient Reason accuses the philosophical
tradition of confusing different kinds of reasons.

[3] “The difference between the grammars of ‘reason’ and ‘cause’ is quite similar to that between
the grammars of ‘motive’ and ‘cause’.” (BBB, 15) “Or is the previous experience the cause of my
certainty, not its ground?” (PI §325)

[4] Wittgenstein hereby denies the Principle of Sufficient Reason for propositions and facts while
he preserves the Principle for material objects and events.

[5] Cf. PI, §1: “Explanations come to an end somewhere.”

[6] “Since language stands in internal relations to the world, it and these relations determine the
logical possibility of facts. If we have a significant sign it must stand in a particular internal
relation to a structure. Sign and relation determine unambiguously the logical form of the thing
signified.” (TB 25.4.15, p. 42)

[7] For an overview see McGinn 2010.

[8] Cf. „It is in language that an expectation and its fulfillment make contact.“ (PI, §445)