Sense, Category, Questions: Reading

Deleuze with Ryle

Peter Kgler

University of Innsbruck

Abstract
Gilles Deleuzes notion of sense, as developed in Difference and
Repetition and The Logic of Sense, is meant to be a fourth dimension
of the proposition besides denotation, manifestation and signification.
While Deleuze explains signification in inferentialist terms, he ascribes to
sense some very unusual properties, making it hard to understand what
sense is. The aim of this paper is to improve this situation by confronting
Deleuzian sense with a more or less contemporary, but otherwise rather
distant philosophical conception: Gilbert Ryles theory of categories and
category mistakes. The leading idea is that to understand the sense of a
proposition regarding X is to know the category of the concept X, which
requires that one knows which questions may appropriately be asked
with regard to X. Thus, sense, category and questions are intimately
related to each other. Finally, it seems to be consistent with Deleuzes
views to assume that abstract signification is contextually generated by
concrete sense.
Keywords: Deleuze, sense, signification, Ryle, category, question

I. Denotation, Manifestation, Signification and Sense

What Frege called sense (in German, Sinn) other philosophers preferred to call meaning. What Frege called meaning (Bedeutung)
others preferred to call reference. Thus, meaning became distinguished
from reference and identified with sense. For Gilles Deleuze, however,
meaning and sense are two different things and both must be kept apart
Deleuze Studies 5.3 (2011): 324339
DOI: 10.3366/dls.2011.0024
Edinburgh University Press
www.eupjournals.com/dls

Sense, Category, Questions

325

from reference and manifestation. Following the English translations of

Deleuzes books Diffrence et rptition and Logique du sens, I will use
denotation (which renders the French dsignation) instead of reference, and signification instead of meaning. In this terminology, then,
we have to consider denotation, manifestation, signification and sense.
Denotation and manifestation, however, are of minor importance in
the present context, so it will suffice to look at them rather briefly.
Deleuzes account of these notions is in accordance with established
theories in linguistics and the philosophy of language. Denotation,
he writes, is the relation of the proposition to an external state
of affairs (datum). The state of affairs is individuated; it includes
particular bodies, mixtures of bodies, qualities, quantities, and relations.
Denotation functions through the associations of the words themselves
with particular images which ought to represent the state of affairs
(Deleuze 2004b: 16). This passage includes some controversial claims.
Not all philosophers would describe denotation in terms of images
and representations. Nor would they choose normative vocabulary like
the word ought (French doivent, from devoir). But nevertheless, what
Deleuze says about denotation lies within the range of existing views, or
at least is very close to some of these views.
The same goes for the second notion, manifestation: It concerns the
relation of the proposition to the person who speaks and expresses
himself. Manifestation therefore is presented as a statement of desires
and beliefs which correspond to the proposition (Deleuze 2004b: 17).
A person who asserts that p thereby expresses the belief that p, as well
as the intention (desire) to express this belief. This is not to say that the
person actually does believe that p. She could also express a belief that
she does not have. In that case the statement would be a lie. Yet even then
we would assume that the person intends to express this belief, for this
is what lying means: intentionally expressing a belief one does not have.
Also, as Deleuze adds to the above, expression/manifestation is a matter
of causal inference. If we have no reason to assume a lie, we may assume
that the assertion that p has been caused by the persons belief that p.1
Let me now turn to the third notion, signification. In the late 1960s,
when the above mentioned books by Deleuze were first published, inferential semantics was not as popular as it is today. It is therefore remarkable that Deleuzes account of signification is of an inferentialist kind:
From the standpoint of signification, we always consider the elements of
the proposition as signifying conceptual implications capable of referring
to other propositions, which serve as premises of the first. Signification

326 Peter Kgler

is defined by this order of conceptual implication where the proposition
under consideration intervenes only as an element of a demonstration,
in the most general sense of the word, that is, either as premise or
conclusion. . . . Demonstration must not be understood in a restricted,
syllogistic or mathematical sense, but also in the physical sense of
probabilities or in the moral sense of promises and commitments. In this last
case, the assertion of the conclusion is represented by the moment the promise
is effectively kept. (Deleuze 2004b: 18)

According to inferential semantics, to grasp the signification of a

proposition is to learn how to use it in the context of inferences, where
inference must be taken in a theoretical as well as in a practical sense.
Michael Dummett describes this as follows:
Learning to use a statement of a given form involves, then, learning two
things: the conditions under which one is justified in making the statement;
and what constitutes acceptance of it, i.e. the consequences of accepting it.
Here consequences must be taken to include both the inferential powers of
the statement and anything that counts as acting on the truth of the statement.
(Dummett 1981: 453)

On this view, the signification of a statement is composed of justification

conditions and of consequences, the latter including commitments and
entitlements to say or to do certain things.2 It may help to give a
(somewhat trivial) example to illustrate the basic idea. A person who
sees that a window is open is justified to assert that the window is open.
By making this assertion, she is also committed to agree that the room
has at least one window. And under appropriate circumstances she may
be entitled to close the window. If she has additionally promised to close
the window, she will even be committed to doing so.
As questions will be coming into focus later, it is important to note
that one part of the signification of a statement is the entitlement to ask
certain questions. A person who says that the window is open is thereby
entitled to ask, for example, who opened the window or for how long it
has been open. One must add, however, that other factors may prevent
the person from actually asking these questions. It may be impolite or
forbidden to do so in a particular situation, or inappropriate in some
other sense. After all, the entitlement conveyed by a statement is just one
of the forces that act in a social situation and it may be overruled by
others.
Since inferential semantics is usually presented as a rival theory to
truth-conditional semantics, it may be misleading that Deleuze defines
signification also as the condition of truth, the aggregate of conditions

Sense, Category, Questions

327

under which the proposition would be true (Deleuze 2004b: 18). But
the context makes clear that here Deleuze does not appeal to truthconditional semantics, that is, to the view that the signification of a
proposition is given by its truth conditions. Again we must think of
inferential relations, which can be understood as concerning possibly
true propositions. A deductively valid inference, in particular, is one
whose conclusion would be true if the premises were true. Neither the
premises nor the conclusion need to be true in reality. This is probably
why Deleuze mentions truth conditions at all, and why he refers to
signification as the possibility for the proposition to be true (Deleuze
2004b: 22).
As to sense, in Deleuzes theory, we get a first understanding of this
notion by considering some features that distinguish it from signification.
Deleuze, following the terminology of traditional syllogistics, explains
that sense is neutral as to the quality, quantity and modality of a
proposition. Quality refers to the distinction between affirmation and
negation; hence a proposition has the same sense as its negation.
Differences in quantity do not count either; hence a proposition about all
things of a kind does not change its sense when all is replaced by some.
And finally, p has the same sense as possibly p and necessarily p, as
these are just differences in modality.
Deleuze even goes so far as to say that sense is affected neither by the
order of subject and predicate nor by that of antecedent and consequent.
Referring to the work of Lewis Carroll, he gives examples such as cats
eat bats and bats eat cats, as well as I breathe when I sleep and I sleep
when I breathe. Another quite surprising claim is that inconsistent
propositions like squares are round have sense but no signification
(Deleuze 2004b: 41).3
In the face of these statements, it does not seem helpful to express the
sense of a proposition by means of another proposition. If you do not
know what the sense of cats eat bats is, you will not gain much if you
are told that it is identical to the sense of bats eat cats, cats do not eat
bats, or cats possibly eat bats. Deleuze therefore tries some alternative
approaches, indicating that the sense of a sentence could perhaps also be
expressed in infinitive, participial or interrogative form (Deleuze 2004a:
194). For example, the sense of The sky is blue might be rendered as
the being-blue of the sky or Is the sky blue?
It is tempting to assume that these ways of expressing sense add up
to an account of propositional content. The sky is blue arguably has
the same propositional content as the being-blue of the sky and Is the
sky blue?4 However, propositional content cannot be what Deleuze has

328 Peter Kgler

in mind. For the propositional content of an utterance is affected by the
difference between affirmation and negation, by differences in quantity
and modality, as well as by the exchange of subject and predicate
or of antecedent and consequent. If sense were propositional content,
the sense of p could not be identical to that of not-p. The same goes
for the other examples mentioned above: possibly. . . /necessarily. . . ,
all. . . /some. . . , cats eat bats/bats eat cats, I breathe when I sleep/
I sleep when I breathe. Each of these pairs combines two different
propositional contents.
So it seems that we cannot identify sense with propositional content.
This result correlates with Deleuzes assertion that it would be a
mistake to construe sense as a neutralized double of the proposition
(Deleuze 2004b: 38; see also 2004a: 195). Sense is not identical to
propositional content, nor can we refer to it by using phrases that
resemble propositions, phrases like the being-blue of the sky or that
the sky is blue. These, too, are just doubles of the proposition The
sky is blue. And this also holds for the corresponding question, Is the
sky blue? There is still too much resemblance between questions and
propositions. But although the question is just another double of the
proposition, Deleuze suggests that replacing the latter by the former is
a first step towards a better analysis of sense, as questions are closely
associated with problems:
the interrogative formula has at least one advantage: at the same time as it
invites us to consider the corresponding proposition as a response, it opens up
a new path for us. A proposition conceived as a response is always a particular
solution, a case considered for itself, abstractly and apart from the superior
synthesis which relates it, along with other cases, to a problem as problem.
Therefore interrogation, in turn, expresses the manner in which a problem is
dismembered, cashed out and revealed, in experience and for consciousness,
according to its diversely apprehended cases of solution. Even though it gives
us an insufficient idea, it thereby inspires in us the presentiment of that which
it dismembers.
Sense is located in the problem itself. Sense is constituted in the complex
theme, but the complex theme is that set of problems and questions in relation
to which the propositions serve as elements of response and cases of solution.
This definition, however, requires us to rid ourselves of an illusion which
belongs to the dogmatic image of thought: problems and questions must no
longer be traced from the corresponding propositions which serve, or can
serve, as responses. (Deleuze 2004a: 195f.; see also 2004b: 139)

Sense is located in the problem, constituted in the complex theme.

Problem and theme are just two of the terms Deleuze uses to refer

Sense, Category, Questions

329

to sense.5 Another one is idea, one of his examples being the idea of
physical atomism (Deleuze 2004a: 232f.), which is the idea that matter
consists of particles some of which are indivisible. Here we are using
a propositional clause, that matter consists of particles. But Deleuze
denies that the idea is itself a proposition or something akin to it. Rather,
the idea is a problem, and the problem is a set of questions. Strictly
speaking, The problem is a set of questions is a kind of slogan that we
may use for the sake of convenience. It would be more precise to say that
the problem is an entity in its own right whose various parts or aspects
can be grasped by asking appropriate questions.6
The problem of physical atomism includes questions like the
following: Does matter consist of atoms?, What types of atoms exist?,
What shapes do atoms have?, What masses do they have?, Do they
move?, In which ways do they move?, Why does a particular atom
move?, and also, What is an atom?7 Any one of these questions
could be answered by an appropriate proposition, and this proposition
would be a partial solution of the problem, solving that part of the
problem which is covered by the question. For example, the question
Do atoms move? is answered by the proposition Atoms move. But in
order to understand more of the problem, it is not enough to ask this
single question. We must also have an understanding of other questions
associated with the problem of atomism.8

II. Categories and Questions

In the preceding section, the notion of sense has been partly clarified by
distinguishing it from the related notions of denotation, manifestation
and signification. This clarification did not go far enough, of course. We
have learned more about what sense is not than what it is. In order to
supplement our negative concept with a positive one, we could take the
usual way of Deleuze interpretation and appeal to those philosophers
who Deleuze himself refers to in developing his position. But I will
take a path that, to my knowledge, has not been pursued before. It will
lead us to Gilbert Ryle, who is not mentioned by Deleuze, although his
best-known book, The Concept of Mind, was published twenty years
earlier than Logique du sens. If we had to sketch something like a map
of philosophical temperaments and opinions, we would probably have
to place these two philosophers on opposite poles, which makes this
comparison even more interesting.
What I want to show in the following is that the notion of category as
developed by Ryle is a close analogy to Deleuzian sense. My hypothesis is

330 Peter Kgler

this: to understand the sense of a proposition regarding X is to know the
category of the concept X. The latter, in turn, requires that one knows
which questions may appropriately be asked with regard to X. But first
let us see what Ryle has to say about categories:
To determine the logical geography of concepts is to reveal the logic of the
propositions in which they are wielded, that is to say, to show with what
other propositions they are consistent and inconsistent, what propositions
follow from them and from what propositions they follow. The logical type or
category to which a concept belongs is the set of ways in which it is logically
legitimate to operate with it. (Ryle 1949: 8)9

Here, Ryle is concerned with inferences that involve propositions

containing the respective concept. These inferences, he argues, determine
the category of the concept. This seems to lead us back to the concept
of signification, defined in inferentialist terms by Deleuze. Inferentialists
usually maintain that the meaning of a word is given by its contribution
to the meanings of the propositions in which it occurs. Referring
to Wilfrid Sellars, inferentialist Robert Brandom writes: To grasp or
understand a concept is, according to Sellars, to have practical mastery
over the inferences it is involved in to know, in the practical sense
of being able to distinguish, what follows from the applicability of a
concept, and what it follows from (Brandom 1994: 89). Against this
inferentialist background, we might be tempted to read Ryle in the
following way: Category is for concepts what signification is for
propositions. The first denotes the meaning of a concept, the second
that of a proposition. Both are to be analysed in inferentialist terms,
for the logical relations between propositions determine the meanings of
concepts as well as those of the propositions.
However, although this picture of the relation between signification
and category is theoretically appealing, it fails to address an important
point. The difficulty is that signification, construed inferentially, is an
abstract concept that covers a large number and variety of inferential
relations, including justification conditions as well as consequences (that
is, commitments and entitlements) of various kinds. Most of these will be
irrelevant in typical conversational situations. Furthermore, it is certainly
not possible for a single speaker to know all these inferences or to draw
all of them in practice. Thus, it is neither necessary nor possible to
know the complete signification of a proposition. This concept is a
construction of linguistic theory, an abstract entity, which is too big,
as it were, for ordinary speakers. What they need to know is which
inferences are actually relevant in a particular context.

Sense, Category, Questions

331

Consider again the example of the window from section one. The
proposition The window is open, in conjunction with the premise
that most windows are rectangular, implies that the open window is
probably rectangular, having a diagonal of a length that is the square
root of a2 + b2 , where a and b are the sides of the rectangle. This is a
remote geometrical consequence that has no relevance at all in typical
cases in which the proposition The window is open is used. It might
be important in a classroom, say, as part of a mathematics exercise.
But usually we would not assume that a person must know how to
calculate the length of the diagonal in order to know the meaning of
The window is open. The same goes for the word window. In order
to know its meaning, one need not be aware of the question as to how
the length of the diagonal is to be computed from the lengths of the
sides.
The inferences Ryle utilises in his book to explain the notion of
category form only a subset of all the inferences that make up the
complete signification of a proposition. How is this subset determined?
What makes some inferences relevant and others not? In the following,
I will argue that this subset of inferences is delineated by questions. Thus,
as far as categories are concerned, the asking and answering of questions
is the type of operation that counts. In fact, Ryle frequently exhibits the
category of concepts by discussing questions.
Before we go on with Ryle, however, let us see whether Deleuzes
philosophy of sense provides any indication of a theory of categories.
The place to look for it is in Deleuzes discussion of paradoxes which
draws on the work of Lewis Carroll. As in other areas of philosophy and
science, the investigation of pathological cases helps to illuminate the
normal ones. Nonsense, the absence of sense, can teach us something
about sense. In the preface of The Logic of Sense, Deleuze even purports
to present a series of paradoxes which form the theory of sense
(Deleuze 2004b: ix).
At least some of Carrolls violations of sense discussed by Deleuze can
be interpreted as category mistakes as envisaged by Ryle. The following
is an instructive example from Alice in Wonderland (quoted by Deleuze
2004b: 31):
. . . and even Stigand, the patriotic Archbishop of Canterbury, found it
advisable
Found what? said the Duck.
Found it, the Mouse replied rather crossly: of course you know what it
means.

332 Peter Kgler

I know what it means well enough, when I find a thing, said the Duck:
its generally a frog or a worm. The question is, what did the Archbishop
find?
The Mouse did not notice this question, but hurriedly went on, found
it advisable to go with Edgar Atheling to meet William and offer him the
crown. . . . (Carroll 1994: 32)

The wit of this little dialogue rests on the confusion of finding a thing and
finding something advisable. As these belong to different categories, it is
not appropriate to ask What did the archbishop find? instead of asking
What did he find advisable? A comparable example is given by Ryle:
She came home in a flood of tears and a sedan-chair is a well-known
joke based on the absurdity of conjoining terms of different types (Ryle
1949: 22). Classical rhetoric registers this figure of speech as zeugma or
syllepis. On Ryles account, the rhetorical effect is caused by a category
mistake.
Another paradoxical case from Alice in Wonderland mentioned by
Deleuze (2004b: 38; 2004a: 195) is the well-known grin of the Cheshire
Cat:
All right, said the Cat; and this time it vanished quite slowly, beginning with
the end of the tail, and ending with the grin, which remained some time after
the rest of it had gone.
Well! Ive often seen a cat without a grin, thought Alice; but a grin
without a cat! Its the most curious thing I ever saw in all my life! (Carroll
1994: 78)

Ontologically speaking, a grin is not a substance that can exist by itself.

It is rather an accident, a property of the substance (the latter being the
cat or its mouth). So if this is what is meant by a grin without a cat, we
are facing another category mistake.
Suppose a person places a word in a category to which it does not
belong. If that person does not intend to create a rhetorical, perhaps
humorous effect, we would probably conclude that he or she does not
understand the word as the confused duck does not understand the
phrase to find it advisable. Assuming that to understand a word is to
know its sense, we may conclude further that to know the sense of a
word requires that one know its category.10
When Ryle discusses examples of categories and category mistakes,
he also explains how one can recognise that two words belong to
different categories. In particular, words of different categories cannot
be connected by the word and, as in the above joke of the lady in the
sedan-chair, or as in he bought a left-hand glove, a right-hand glove and

Sense, Category, Questions

333

a pair of gloves. Put in positive form, Ryles criterion is the following:

When two terms belong to the same category, it is proper to construct
conjunctive propositions embodying them (Ryle 1949: 22).
However, Ryles primary method of determining categorical sameness
or difference is the asking of questions. He often notes that certain
questions are not appropriate with regard to concepts of a particular
category. It does not make sense to ask Where is the pair of gloves?
when you see the two gloves. This question is improper because it implies
that the pair is something apart from its components. It conflates the
whole with its parts: the word pair which denotes the whole belongs to
a different category than the word glove which denotes the parts.11

III. The Contextual Generation of Signification

The lesson to draw from Ryles treatment of categories is that knowing
the category of X amounts to knowing which questions it is appropriate
to ask about X. At this point, the connection to Deleuzes notion of
sense becomes visible. As already explained, Deleuze correlates sense
and problem, and the problem can be characterised as a set of questions.
Furthermore, although sense is said to be a dimension of the proposition (Deleuze 2004b: 22), it is also addressed as idea, problem and
theme. And the problem is not propositional, although it does not exist
outside of the propositions which express it. . . . The problem is neutral
with respect to every mode of the proposition (Deleuze 2004b: 140).
This applies to Ryles categories as well. Categories are subpropositional entities. We are talking of the category of a concept
window, glove, atom and so on not of that of a proposition. A
concept can appear in propositions of various kinds. This is the reason
why propositions that differ in quality, quantity or modality can be
assigned to, or subsumed under, the same category. While the category
itself is not propositional, it does not exist outside of the propositions
which express it.
Hence the discussion of sense shifts from propositions to words. We
may still say that sense is a dimension of the proposition, but only
because propositions include concepts like window, glove and atom,
and because we can ask to which categories these concepts belong.
The notion of sense primarily applies to constituents of propositions.
This is obscured by Deleuze when he describes sense as a dimension of
propositions, along with denotation, manifestation and signification. At
the end of the day, however, the fact that the two propositions The
window is open and The window is not open have the same sense

334 Peter Kgler

traces back to the fact that both contain the word window, and that
both are therefore associated with questions like What is a window?,
Is there a window?, Is it open?, Why is it open?, and so on.
However, with which questions a proposition is associated depends
on the context in which it is used. It is easy to imagine circumstances
in which the negative proposition The window is not open triggers the
question Is the window open? which, if answered positively, causes the
next question, Why is it open? This could happen when the questioner
does not trust the person who said that the window is not open. In
other cases where there is no reason to doubt the speakers claim, these
questions would not arise. There would be other questions or none at all.
The context in which a proposition is being used is defined, at least
to a considerable degree, by the questions that are taken to be relevant
by the persons involved. In a way, we do not know what that context
is unless we know which questions are being asked. This also suggests
a new view of the relationship between sense and signification. We have
seen that signification is an abstract concept comprising a potentially
infinite set of inferential relations. This concept is derived from concrete
cases of language-use by steps of extrapolation and generalisation.
Signification is supposed to consist of those inferences that are relevant
in actual cases or could be relevant in potential ones. In a particular
conversational situation, speakers operate only with small subsets of all
the relations that make up signification, and these subsets are determined
by sense, by a set of relevant questions.
In sum, then, signification is generated by pragmatic, contextual
conditions that are set by sense. Deleuze ascribes to sense a generative
role, which is to engender the logical proposition with its determinate
dimensions (denotation, manifestation, and signification) (Deleuze
2004b: 138). And in a text written in 1986, he characterises his
empiricism in the following way:
the abstract does not explain, but must itself be explained; and the aim is not
to rediscover the eternal or the universal, but to find the conditions under
which something new is produced (creativeness). In so-called rationalist
philosophies, the abstract is given the task of explaining, and it is the abstract
that is realized in the concrete. . . . Empiricism starts with a completely
different evaluation: analysing the states of things, in such a way that
non-preexistent concepts can be extracted from them. (Deleuze and Parnet
2006: vi)

Of course, as always, the quotation is taken out of its context. Deleuzes

notion of empiricism is of a special kind, but this is not our issue.

Sense, Category, Questions

335

What makes the quotation interesting for us is that it gives an idea of

the relation between sense and signification. First comes the analysis
of sense in contexts of language-use. This analysis yields a set of
questions, which enables the identification of the problem around
which the situation is organised. The notion of signification, in turn,
is derived from these cases. In Deleuzes terms, it is a non-preexistent
concept that is extracted from sense. Signification is the abstract
in comparison to which sense is the concrete. When this abstraction
is retrospectively applied to the concrete, then sense, as expressed by
questions, is conceived as part of signification, more exactly, as that part
which exists in the respective context.
The function of questions in this process is to make certain inferential
relations relevant. Roughly speaking, the relevant inferences are those
which connect the answers of the questions to a particular proposition.
To explain this, let us return to Ryles example of the two gloves.
Suppose that these gloves are lying on opposite ends of the couch in front
of you. There is also another person present to whom you say, There
are two gloves on the couch. The question whether the gloves belong
together need not arise in this context. If it does not arise, the relevant
part of the signification primarily concerns perceptual justification
conditions. Your perceiving the gloves justifies your statement There are
two gloves on the couch. By uttering this proposition when perceiving
the gloves, you prove your mastery of this part of the signification. It is
a rather basic part, to be sure. But as I have described the situation so
far, there is no reason to look beyond it.
Next, suppose the question arises as to whether the two gloves
belong together, whether they form a pair. It does not matter who
raises this question, but if it is raised, the question broadens our scope
of the signification of There are two gloves on the couch. Now it
becomes relevant whether you know which answers to the question
might be appropriate. This task is still quite simple, of course. You
are confronted with a yes/no-question: Do the two gloves on the couch
belong together? So you only need to know that the answer is Yes, they
do or No, they do not. Yet, however simple this task may be, by giving
one of these answers you would prove that you know, in a concrete,
practical sense, another part of the signification of the proposition There
are two gloves on the couch.
Additional questions might arise. Someone might want to know why
the gloves are lying on the couch or to whom they belong. Thus, other
parts of the signification would be generated, and the actors in the
situation would become aware of possible ways of answering these

336 Peter Kgler

questions. If they lacked the requisite information, their answers would
only be rather general. For example, they might say that the gloves are
lying on the couch because someone who has been in this room before
has put them there; that the gloves on the couch belong to some person,
probably an adult (as they are too big for a child), and so on. Those who
know more about the gloves could give more specific answers.
This interplay of assertions, questions and answers concerns
inferential relations and therefore signification. The questions alluded to
in the glove example are inferentially related, in a certain way, to There
are two gloves on the couch, which is the proposition that started the
conversation by raising the problem of the two gloves. It should be
noted, however, that questions are not directly related to propositions,
but only by their propositional contents. For instance, the content of the
yes/no-question Do the two gloves on the couch belong together? can
be expressed by the proposition The two gloves on the couch belong
together.12 In the case of questions with interrogative words, we obtain
the propositional content by replacing why, who, where, and so on,
by a suitable variable. To choose just two examples, this replacement
yields The gloves are lying on the couch because p as content of Why
are the gloves lying on the couch, as well as The two gloves on the
couch belong to X as content of To whom do the two gloves on the
couch belong?
Inferentialism has it that the signification of the proposition There
are two gloves on the couch is given by the inferential relations this
proposition is involved in. Among these relations are those which exist
between the proposition itself and the propositional contents of the
above questions. The proposition is inferentially related to The two
gloves on the couch belong together, The gloves are lying on the couch
because p and The gloves on the couch belong to X. This proves that
the propositional contents of the questions, and therefore indirectly the
questions themselves, bear inferential relations to the proposition we
are dealing with. And these relations make up a subset of the complete
signification of the proposition.
Following Ryle, I have chosen somewhat dry examples concerning
windows and gloves. These do have the advantage of familiarity, but
nevertheless we shall end with one of the more appealing cases. Lewis
Carrolls story of the Cheshire Cat involves a category mistake that has
been made on purpose. Its paradoxical nature owes to the fact that we
are prompted to imagine a grin without a something that is grinning
(which John Tenniel could not picture in his famous illustrations, so he
decided to draw the cats head fading). The question arises as to who

Sense, Category, Questions

337

it is that is grinning, and this question makes certain answers relevant.

The reader of Alice in Wonderland knows that from There is a grin it
follows that there is someone who is grinning. The context of the story
even suggests the more specific answer that it is the Cheshire Cat who is
(still) grinning. Yet since the cat is said to have vanished, the story turns
into a piece of nonsense.
We know why. The story defies a part of signification: the inference
from There is a grin to The Cheshire Cat is grinning, the latter being
our best answer to Who is grinning? This is one of the questions that
constitute the problem of the Cheshire Cats grin and the sense of the
inferentially related propositions.

Notes
1. The nature of this causation arguably depends on the ontological nature of
mental states, but we need not go into this here. This article is mainly concerned
with an interpretation of Deleuzes philosophy of language; ontological issues
will be touched upon only tangentially.
2. In the above quotation from The Logic of Sense, Deleuze just talks of theoretical
inferences (syllogistic and mathematical demonstration, physical probabilities)
and practical commitments, such as the commitment to keep a promise. He does
not mention entitlements, but these can safely be added to his account.
3. Deleuze does not substantiate the claim that inconsistent propositions lack
signification. Perhaps he relies on the fact that these propositions cannot enter
into inferences as possibly true statements. Classical logic, however, does ascribe
an inferential role to them, if only a rather trivial one: ex contradictione sequitur
quodlibet.
4. I am using propositional content roughly as in Chapter 2.4 of Searle 1969.
Although an utterance of the being-blue of the sky is at best an incomplete
speech act, it works as an indicator of propositional content, as Searle would
say. Searle, however, prefers that-clauses for indicating propositional content
(that the sky is blue). Note also that for him a proposition is the propositional
content of a speech act, whereas Deleuzes usage of that term is wider.
5. The term event belongs to this list, too, as sense is said to be an incorporeal,
complex, and irreducible entity, at the surface of things, a pure event which
inheres or subsists in the proposition (Deleuze 2004b: 22). Another surface
entity is the concept, which speaks the event (Deleuze and Guattari 1994: 21).
All concepts are connected to problems without which they would have no
meaning [sens] (16). Except that concepts are supposed to be solutions of
problems, they have much in common with the latter. In particular, they are
not propositional (22). To keep the discussion simpler, however, I refrain from
considering these notions of event and concept. I will reserve the word concept
for Ryle who uses it for a linguistic entity.
6. As indicated in the previous note, sense inheres or subsists, but does not exist.
Furthermore, it is a dual entity on the border between world and language: It
is rather the coexistence of two sides without thickness, such that we pass from
one to the other by following their length. Sense is both the expressible or the
expressed of the proposition, and the attribute of the state of affairs. It turns

338 Peter Kgler

7.

8.

9.

10.

11.

12.

one side toward things and one side toward propositions (Deleuze 2004b: 25).
Thus, sense is something very peculiar, to say the least.
A superficial reading might suggest that Deleuze rejects questions of the type
What is X? per se. But in fact he only rejects the Platonic-Socratic tradition in
which these questions are understood as aiming at essences. Apart from that,
Deleuze acknowledges their propaedeutic function in the Platonic discourse:
Once it is a question of determining the problem or the Idea as such, once it is a
question of setting the dialectic in motion, the question What is X? gives way
to other questions, otherwise powerful and efficacious, otherwise imperative:
How much, how and in what cases? The question What is X? animates
only the so-called aporetic dialogues in other words, those in which the very
form of the question gives rise to contradiction and leads to nihilism, no doubt
because they have only propaedeutic aims the aim of opening up the region of
the problem in general, leaving to other procedures the task of determining it as
a problem or as an Idea (Deleuze 2004a: 236f.).
It would lead us too far astray to discuss in detail how these questions are related
to each other. A possible point of departure could be what Sylvain Bromberger
calls a cluster of questions (Bromberger 1966: 602). According to Bromberger,
a cluster of questions is raised by some proposition, and a question belongs to
this cluster if it stands to the proposition in a relation of mutual implication.
That is, if the proposition is true, the question must have a correct answer, and
conversely.
In his article Categories, published about a decade before The Concept of Mind,
Ryle prefers to talk of concepts without the adjective logical (Ryle 1937/8).
I follow him in this. Others may choose to replace logical by ontological,
as differences between logical categories are arguably related to ontological
differences. This logical/ontological ambiguity of category seems to correspond
to the two-sidedness of sense mentioned in note 7 above. Compare Deleuzes
distinction between the logical and the ontological proposition in The Logic of
Sense (Deleuze 2004b: 138).
Here, to know does not necessarily refer to knowing-that, that is, knowledge
expressible in propositional form. There are cases in which it is possible to say
that a word means such-and-such, thus giving the meaning of the word by using
other words. But the knowledge in question is primarily a linguistic kind of
knowing-how. A person who knows the sense of a word knows how to use it in
the course of utterances, in accordance with its category.
The glove example suffices for our purposes, but in The Concept of Mind
cases not based on a part-whole relation are more salient. In particular, there
is the distinction between occurrences and dispositions which, in Ryles opinion,
provides a solution (or rather, dissolution) of the mind-body problem. There are
questions about occurrences that cannot be asked about dispositions.
Thus, we could analyse the question as consisting of two parts, the first one
expressing its content: The two gloves on the couch belong together. Yes or
no?

References
Brandom, Robert B. (1994) Making It Explicit: Reasoning, Representing, and
Discursive Commitment, Cambridge, MA: Harvard University Press.
Bromberger, Sylvain (1966) Questions, The Journal of Philosophy, 63,
pp. 597606.
Carroll, Lewis (1994) Alices Adventures in Wonderland, London: Penguin.

Sense, Category, Questions

339

Deleuze, Gilles (2004a) Difference and Repetition, trans. Paul Patton, London:
Continuum.
Deleuze, Gilles (2004b) The Logic of Sense, trans. Mark Lester with Charles Stivale,
ed. Constantin V. Boundas, London: Continuum.
Deleuze, Gilles and Flix Guattari (1994) What Is Philosophy? trans. Hugh
Tomlinson and Graham Burchell, New York: Columbia University Press.
Deleuze, Gilles and Claire Parnet (2006) Dialogues II, trans. Hugh Tomlinson and
Barbara Habberjam, London: Continuum.
Dummett, Michael (1981) Frege: Philosophy of Language, London: Duckworth.
Ryle, Gilbert (1937/8) Categories, Proceedings of the Aristotelian Society, 38,
pp. 189206.
Ryle, Gilbert (1949) The Concept of Mind, London: Hutchinson.
Searle, John R. (1969) Speech Acts: An Essay in the Philosophy of Language,
Cambridge: Cambridge University Press.