Derrida and Peirce on indeterminacy,

iteration, and replication

UWE WIRTH

Those who understand by ‘the crisis of representation’ a semiotic Paradise

Lost where objects are no longer objects and signs have lost their power
to represent are faced with two questions: (1) Is it really a characteristic
of the ‘postmodern hell’ that the signs have lost their power to represent
anything at all, or have the signs perhaps gained the power to represent
everything? (2) Can this situation be called a crisis of representation at
all, or is it instead the crisis of a certain semiotic model of representation?
Is it the crisis of a model which presupposes that signs represent objects
and thoughts in a well-defined, conventionally coded manner and which
determines both the scope of adequate ways of using the signs and the
scope of their adequate interpretation? It is such a model of representation
against which Derrida argues in Of Grammatology when he writes:

Peirce goes very far in the direction that I have called the deconstruction of the
transcendental signified, which, at one time or another, would place a reassuring
end to the reference from sign to sign. I have identified logocentrism and the
metaphysics of presence as the exigent, powerful, systematic, and irrepressible
desire for such a signified. Now, Peirce considers the indefiniteness of reference as
the criterion that allows us to recognize that we are indeed dealing with a system of
signs. What broaches the movement of signification is what makes its interruption
impossible. The thing itself is a sign. (Derrida 1976 [1967]: 49)

In his chapter on ‘Unlimited semiosis and drift’, Eco (1990: 36) raises the
question as to whether Peirce ‘would have been satisfied with Derrida’s
interpretation’. The answer to this question is, of course, negative since
‘deconstructive drift and unlimited semiosis cannot be equivalent con-
cepts’ (ibid.). Eco’s argumentative strategy is to admit, on the one hand,
the ‘openness’ of Peirce’s concept of unlimited semiosis ‘for further
determination’. On the other hand, Eco suggests that Derrida had to
refuse the idea of such a determination since the ‘deferral’ of diffe´rance
implies a fundamental ‘indeterminacy’ of the whole dynamics of inter-
pretation, not only in the past and present, but also in the future. ‘I am

Semiotica 143–1/4 (2003), 35–44 0037–1998/03/0143 – 0035

# Walter de Gruyter
36 U. Wirth

simply repeating with Peirce’, Eco (1990: 39) writes, ‘that ‘‘an endless
series of representations, each representing the one behind it’’ (and until
this point Derrida could not but agree with this formula), ‘‘may be
conceived to have an absolute object as its limit’’ (CP 1.339)’. Eco claims
that the idea of an ‘absolute object’ is incompatible with the decon-
structive framework, since it implies that ‘outside the immediate
interpretant, the emotional, the energetic, and the logical one — all
internal to the course of semiosis — there is a final logical interpretant,
that is, Habit’ (Ibid.). Let us disregard whether Peirce would be satisfied
with Eco’s interpretation of the final interpretant as something ‘outside
semiosis’. Instead, let us complete the passages from Peirce that Eco
wisely left out in his quotation:

[A sign] is a vehicle conveying into the mind something from without. That for
which it stands is called its object; that which it conveys, its meaning; and the idea
to which it gives rise, its interpretant. The object of representation can be nothing
but a representation of which the first representation is the interpretant. But
an endless series of representations, each representing the one behind it, may be
conceived to have an absolute object at its limit. (CP 1.339)

Of course, we understand why Eco left out the passage that ‘the object
of representation can be nothing but a representation of which the first
representation is the interpretant’. It sounds as if Derrida had formulated
it and, furthermore, it does not fit in with Eco’s argumentation. Since
the object of representation as well as its interpretant is ‘nothing but
a representation’, there cannot be, at least if we take Peirce seriously,
anything ‘outside semiosis’: neither an ‘absolute object’ nor a ‘final
interpretant’.
At this point, we are apparently confronted with a ‘critical situ-
ation’, which is a ‘crisis of interpretation’ rather than a ‘crisis of
representation’. As in any crisis,1 there is a ‘general uncertainty’ at this
point. We do not know whether it is the Peircean account of semiosis
that is equivocal and vague or whether it is Eco who is molding Peirce
in order to make his theories fit in with Eco’s own strategic purpose,
namely a critique of Derrida’s interpretation of the Peircean account
of semiosis.
In the following discussion, I will raise two points. Firstly, I will call
into question whether the so-called ‘Derridaean problem’ of indetermi-
nacy can be resolved by using the Peircean notion of a limiting ‘final
interpretant’ conceived of as ‘habit’. Secondly, I will discuss Derrida’s
notion of ‘iterability’ in the context of Peircean semiotics and relate it to
representation in general and to the type-token relation of replica signs in
particular.
 Derrida and Peirce 37

Semiosis and the problem of indeterminacy

The reason why semiosis has to be an ‘endless series of representations’

(CP 1.339) is that the series of successive interpretants must not come
to an end; otherwise, the sign would be thereby ‘rendered imperfect’
(CP 2.303). Hence, semiosis is the process in which a sign ‘determines
something else (its interpretant) to refer to an object to which itself refers
(its object) in the same way, this interpretant becoming in turn a sign,
and so on ad infinitum’ (CP 2.303). However, as soon as we ask for
its relation to habit and thought, this well-known notion of semiosis
turns out to be pretty puzzling. On the one hand, Peirce writes: ‘A belief-
habit in its development begins by being vague, special, and meager;
it becomes more precise, general, and full, without limit. The process of
this development _ is called thought’ (CP 3.160). On the other hand,
however, Peirce argues that ‘in thought, an absolutely determinate term
cannot be realized _ . We can only say, in a general way, that a term,
however determinate, may be made more determinate still, but not that
it can be made absolutely determinate’ (CP 3.93). This second quote
implies that every sign has to remain indeterminate and thus open to
further determination. The first quote is just another way of describing
semiosis as a process of transforming ‘vague’ belief habits into more
precise and general habits. Now, since the ‘tendency to form habits
or tendency to generalize is something which grows by its own action’
(CP 8.317), we can conclude that taking a habit or generalizing is the
way determination takes place in semiosis. The problem, however, is
that Peirce himself claims that generality — like vagueness — is a form
of indeterminacy: ‘a sign can only escape from being either vague or
general by not being indeterminate’ (CP 5.506). In the same passage,
Peirce writes that ‘no communication of one person to another can
be entirely definite, i.e., non-vague’ (CP 5.506). Nonetheless, there is
a difference between generality and vagueness. According to Peirce,
a sign that is indeterminate is general, ‘insofar as it extends to the
interpreter the privilege of carrying its determination further. Example:
‘‘Man is mortal’’. To the question, What man? the reply is that the
proposition explicitly leaves it to you to apply its assertion to what man
or men you will’ (CP 5.447). A sign that is indeterminate is vague
‘in so far as it reserves further determination to be made in some
other conceivable sign _ . Example: ‘A man whom I could mention
seems to be a little conceited’. The suggestion here is that the man in view
is the person addressed; but the utterer does not authorize such an
interpretation or any other application’ (CP 5.447). Apparently, leaving
the meaning of an utterance vague is a form of making an allusion,
38 U. Wirth

or, as Grice (1975) calls it, conveying information by means of a

‘conversational implicature’.
A sign is vague insofar as it leaves ‘its interpretation more or less
indeterminate’ and reserves ‘for some other possible sign or experience
the function of completing the determination’ (CP 5.505). Thus, a sign
remains indeterminate and vague unless it is ‘by a well-understood
convention rendered general. _ Honest people, when not joking,
intend to make the meaning of their words determinate, so that there
shall be no latitude of interpretation at all’ (CP 5.447). On the contrary,
whenever we are joking, we leave ‘the determination of the implication
to the interpreter’ (CP 5.447). To me, this sounds quite similar to
what Derrida claims in Limited Inc.: ‘The imposition of a meaning
supposes a certain play or latitude in its determination’ (Derrida 1988
[1977]: 145).
Let us now focus on three aspects of the vagueness and indeter-
minacy of signs. The first is that both ‘latitude of interpretation’ and
‘latitude of determination’ refer to vagueness by leaving the right and
the responsibility of ‘further determination’ to the interpreter. The second
is that, although both vagueness and generality are forms of indeter-
minacy, they are different forms of indeterminacy because ‘no sign can be
at once vague and general in the same respect’ (CP 5.506). Hence, semiosis
is the process of transforming vague concepts into general concepts,
leaving the possibility of further determination open.2 The third point
could be that ‘vagueness’ coincides with Koselleck’s (1999: 105) notion
of crisis, conceived as ‘the general uncertainty in a critical situation’. As
soon as the interpreter has to decide which implication of an utterance
is to be fixed and which is not, he or she is in a critical situation. Hence,
the ‘crisis of representation’ is a ‘crisis of interpretation’, which is caused
rather by problems of decision than by problems of indeterminacy.
Indeed, Derrida states in his afterword to Limited Inc. that deconstruction
is not so much interested in ‘indeterminacy’ as in ‘undecidability’:
‘Undecidability is always a determinate oscillation between possibili-
ties _ . These possibilities are themselves highly determined in strictly
defined situations’ (Derrida 1988 [1977]: 148). There is determination,
according to Derrida (ibid.: 136), since ‘one cannot do anything, least of
all speak, without determining _ a context’.

Iterability and the problem of habit

Although Eco’s critique of deconstructive indeterminacy has become

pointless with the insight that deconstruction is really interested in
 Derrida and Peirce 39

problems of undecidability, we should, nevertheless, ask whether Derrida

is right when he claims that the function of a sign depends on the fact
that it is iterable in different contexts and that a code is ‘the organon of
iterability’, as he puts it in ‘Signature event context’ (Derrida 1988
[1977]: 7–8). In the following discussion, I want to show that the
‘iterability’ of a sign is a process that takes place independently from the
application of a code, if such a code is conceived of as a general rule of
signification. Iterability concerns only the ‘law of sign replication’ but not
the rule by which it refers to its object. Hence, we should ask whether
iteration can have a direct impact on semiosis at all.
According to Derrida (1988 [1977]: 7–8), the applicability of signs
to different contexts depends on whether they can be repeated: ‘The
possibility of repeating and thus identifying the marks is implicit in every
code, making it into a network [une grille] that is communicable,
transmittable, decipherable, iterable for a third, and hence for every
possible user in general. To be what it is, all writing must, therefore, be
capable of functioning in the radical absence of every empirically
determined receiver in general’. Due to this iterability, every sign can
break ‘with every given context, engendering an infinity of new contexts
in a manner which is absolutely illimitable’ (ibid.: 12).
Evidently, the notion of iterability refers to the openness of the sign to
multiple use, which requires its indeterminacy. The question remains as
to whether iterability has any impact on the forming of a habit and thus
on establishing a final interpretant. As we have seen above, habit for-
mation pertains to the tendency towards generality. However, according
to Peirce (CP 5.491), a ‘deliberately formed _ habit’, i.e., a final inter-
pretant does not depend on repetition, since ‘an endless series of acts is
not a habit’ (CP 2.667 and fn.):

[It] is noticeable that the iteration of the action is often said to be indispensable
to the formation of a habit; but a very moderate exercise of observation suffices
to refute this error. A single reading yesterday of a casual statement that ‘shtar
chindis’ means in Romany ‘four shillings’ _ is likely to produce the habit of
thinking that ‘four’ in the Gypsy tongue is ‘shtar’, that will last for months, if not
for years, though I should never call it to mind in the interval. (CP 5.477)3

The reason why we need no repetition in order to establish the habit

of a final interpretant is that it ‘does not consist in the way in which any
mind does act but in the way in which every mind would act’ (CP 8.315).
A final interpretant consists of a conditional proposition of the kind
‘if so and so were to happen to any mind this sign would determine that
mind to such and such conduct’ (CP 8.315). Peirce gives the example of
the everyday chat about the weather: ‘What sort of a day is it?’
40 U. Wirth

The Immediate Interpretant is what the Question expresses. _ The Dynamical

Interpretant is the actual effect that it has upon me, the interpreter. But the
Significance of it, the Ultimate, or Final, Interpretant is her purpose in asking it,
what effect its answer will have as to her plans for the ensuing day. I reply, let us
suppose: ‘It is a stormy day’. (CP 8.314)

While repetition is apparently not necessary in order to form a habit or

to establish a final interpretant, Peirce nonetheless argues that all signs are
determined by the possibility of being repeated. Indeed, the following
passage sounds as if Derrida had written it:

The mode of being of a representamen is such that it is capable of repetition. Take,

for example, any proverb. ‘Evil communications corrupt good manners’. Every
time this is written or spoken in English, Greek, or any other language, and every
time it is thought of it is one and the same representamen. It is the same with
a diagram or picture. It is the same with a physical sign or symptom. If two
weathercocks are different signs, it is only insofar as they refer to different parts
of the air. A representamen which should have a unique embodiment, incapable of
repetition, would not be a representamen, but a part of the very fact represented.
This repetitory character of the representamen involves as a consequence that it
is essential to a representamen that it should contribute to the determination
of another representamen distinct from itself. (CP 5.138)

In a certain way, this quote links our first topic concerning the deter-
mination of the sign with our second topic concerning its iterability, but
if repetition is essential to representation, we are faced with the question,
what is it that gets repeated? According to Derrida (1988 [1977]: 7; see
previous discussion) there is a ‘repeating and thus identifying [of ] the
marks’. However, Searle (1993: 8) objects that ‘actual marks and signs,
that is, actual physical tokens, are precisely not iterable. It is, rather, the
type of mark that can have different instantiations. This is one way of
saying that it is types and not tokens that allow for repeated instances of
the same’.4

Iteration and the type-token relation

Since Searle refers explicitly to Peirce in the previous passage, it might

be interesting to examine the relation between Peircean semiotics and
Searle’s critique of Derrida’s theory of deconstruction, but a thorough
study of this topic is not possible in this paper. Nevertheless, I will try
to demonstrate how difficult it can be to answer the question of what is
it that gets iterated. According to Peirce, a legisign, is a general type and
 Derrida and Peirce 41

not a single object,5 although ‘[e]very legisign signifies through an instance

of its application, which may be termed a Replica of it’ (CP 2.246).6 A
replica is a token, i.e., an individual sign. The word ‘man’, is, for instance,

a general mode of succession of three sounds or representamens of sounds, which

becomes a sign only in the fact that a habit, or acquired law, will cause replicas of
it to be interpreted as meaning a man or men. The word and its meaning are both
general rules; but the word alone of the two prescribes the qualities of its replicas
in themselves. (CP 2.292)

In this passage, Peirce seems to refer to two different rules. The first
concerns the sign itself — I would like to call it the law of replication —
and the second concerns the relation between the sign and its meaning.
This ‘general rule of meaning’ is a rule of signification, i.e., a code. The
‘significative value of a symbol consists in a regularity of association’
(CP 4.500), which ‘depends either upon a convention, a habit, or a natural
disposition’ (CP 8.335). However, since symbols are always abstract and
general, because habits are general rules to which the organism has
become subjected’ (CP 3.360), they must be embodied in replicas, in order
to be informational. Peirce describes this act of application as ‘translat-
ing every abstractly expressed proposition into its precise meaning in
reference to an individual experience. _ A symbol is not an individual,
it is true. But any information about a symbol is information about every
replica of it; and a replica is strictly an individual’ (CP 2.315). The
interesting point about this definition of replicas is that ‘individuation’ is
a mode of determination. An individual is ‘absolutely determinate in all
respects’ (CP 5.299), since existence implies ‘entire determinateness, in the
sense that nothing remains undecided’ (CP 4.431).7
This raises the question as to whether a replica token is determined
by the fact that it is an existing individual or whether it is determined by
the fact that it is an application of a general law. The answer is not easy
because, according to Peirce, replica tokens are ‘not ordinary Sinsigns,
such as are peculiar occurrences that are regarded as significant’. A replica
is only significant because of a ‘law which renders it so’ (CP 2.246).8 Such
a law, however, is not the ‘general rule of meaning’ by which a symbol is
related to its object. It is, rather, a law of ‘replication’ that regulates how
an abstract type is transformed into an individual token. This ‘law of
replication’ is, so to speak, stored in every replica. In a sense, we could say
that every replica token is self-referential insofar as it indicates the type
of which it is a replica. Peirce argues ‘that the mass of ink on the sheet
by means of which a graph is said to be ‘scribed’ is not _ a symbol, but
only a replica of a symbol of the nature of an index’ (CP 4.500).
42 U. Wirth

Conclusion

The question of ‘what gets iterated’ could be answered by stating that

neither the type, nor the token, but only the type-token relation is iterable.
‘Iteration’ concerns exclusively the ‘law of replication’ but not the forming
of pragmatic habits or semantic codes. Hence, Derrida’s concept of
‘iterability’ as an endless ‘grafting’ on different contexts is incompatible
with the Peircean concept of semiosis as an endless series of representa-
tions that render meaning and interpretation more and more determinate
and precise. While Derrida holds that ‘iterability’ is the ‘organon’ even
for the semantic aspects of a code, Peirce argues that the ‘repetitory
character’ of representation concerns only the ‘sign in itself ’, i.e., the law
of replication, which governs the type-token relation.
As mentioned above, according to Peirce, a replica is the ‘application’
of a sign that translates its abstract type into an individual token.
However, ‘being abstract’ is not the same as ‘being general’. Abstraction
allows paying ‘attention to a part of an idea’ and neglecting the rest
(CP 2.428), and attention is ‘the part that thought plays as an index, _ a
certain modification of the contents of consciousness with reference to
a center’ (CP 2.428). Abstraction, conceived as an index referring
to a center, that is, directing our intention to a selected part of an idea,
is a cognitive operation making our concepts more precise. For this
reason, Peirce uses the terms ‘abstraction’ and ‘precision’ synonymously
(CP 1.549). Precision, however, is the opposite of vagueness. Hence, we
can conclude that abstraction is a form of determination. In this sense,
abstraction is opposed to generality since generality is a form of indeter-
minacy. A replica, then, is subject to a twofold determination: insofar as
it is a token, it is determined by its individuality; insofar as it is an abstract
type, it is determined by the power of attention.
Since attention, according to Peirce (CP 5.295), ‘strongly affects
memory’, due to the fact that it is ‘the power by which thought at one time
is connected with _ thought at another time’, further questions emerge
concerning central issues of both deconstruction and media theory. What
do we have in our mind, what do we have in our memory: abstract types,
or individual tokens? How do we store signs, as types or as tokens, and
what is the semiotic status of information stored in a computer? Should
we consider the marks and chains in a computer as ‘electronic tokens’?
According to the hypertext theoretician George Landow (1997 [1989]: 19),
electronic information storage has been a ‘major shift in information
technology after the development of the printed book’.
Now, since storing is a kind of re-presentation, the ‘crisis of
representation’ is equally a crisis of storing and putting into archives.
 Derrida and Peirce 43

Derrida (1995: 34), in Mal d’archive, concludes: ‘La structure technique

de l’archive archivante détermine aussi la structure du contenu archivable’.
A further symptom of the crisis of representation is that we have become
increasingly capable of storing just about everything. Thus, we are
required to decide what to store at the same time as we have become less
and less capable of paying attention to the relevant information.

Notes

1. According to Koselleck (1999: 105), the term ‘crisis’ designates a state of uncertainty
requiring some decision: ‘The general uncertainty in a critical situation is mingled with
the certainty that the critical state of affairs will have an end sometime — although it
is not determined when this will take place’.
2. Of course, we might also conclude that the ‘determination’ is a vague term in itself, since
it can be used to designate quite different things, such as logical determination, causal
determination, intentional determination, or determination by individuation.
3. Although habit taking needs no repetition, iteration plays an important role in ‘habit-
change’, since ‘repetitions of the actions that produce the changes increase the changes’
(CP 5.477). Iteration has a strengthening power.
4. Searle (1993: 1) criticizes Derrida’s account of representation, especially his position
that ‘meanings are ‘‘undecidable’’ and have ‘‘relative’’ indeterminacy _ . Instead of fully
determinate meanings, there is rather the free play of signifiers and the grafting of texts
onto texts’. Concerning the type-token distinction, Searle writes:
I believe the distinction between linguistic types and linguistic tokens was first
formulated by Charles Sanders Peirce. _ We need this distinction because identity
criteria for types and tokens are quite different. What makes something a case of
‘the same token’ will be different from what makes it ‘the same type’. _ The distinction
between types and tokens, by the way, is a consequence of the fact that language is rule-
governed or conventional, because the notion of a rule or of a convention implies
the possibility of repeated occurrences of the same phenomenon. (Ibid. 8)
5. ‘A Legisign is a law that is a Sign’, and ‘this law is usually established by men. Every
conventional sign is a legisign [but not conversely]’ (CP 2.246). We could add here that all
conventional signs are symbolic legisigns, although not every symbol is conventional.
6. However, an utterance is not just a replica token, it is, rather, a replica token that is
used intentionally. This definition leaves it open as to whether it is the utterer or the
receiver who applies intention to the token. In his essay, ‘Literal meaning’, Searle (1979:
119–120) states, concerning the distinction between ‘type’ and ‘token’, that an utterance
is more than just the replica of a sentence-type. One and the same utterance can
have many replicas, as, for instance, in the case of a book, where one and the same
utterance is distributed in thousands of copies. According to Searle, a token always has
the same meaning as its type. Hence, the difference between the meaning of a sentence
type and an utterance token is due to the speaker intention and not to the type-token
distinction.
7. On the other hand, of course, Peirce argues that ‘the absolute individual _ cannot exist’
and that all ‘that we perceive or think, or that exists, is general’ (CP 3.93). This is
especially true of a replica-token.
44 U. Wirth

8. According to Peirce, ‘a legisign has a definite identity, though usually admitting a great
variety of appearances. _ The qualisign, on the other hand, has no identity. _ Instead
of identity, it has great similarity’ (CP 8.334).

References

Derrida, Jacques (1976 [1967]). De la Grammatologie. Paris: Minuit. [English translation:

Of Grammatology, trans. by G. Chakravorty Spivak. Baltimore, MD: Johns Hopkins
University Press.]
— (1988 [1977]). Limited Inc abc. Glyph 2, 162–254. [English translation: Limited Inc., trans.
by S. Weber. Evanston, IL: Northwestern University Press.]
— (1995). Mal d’archive. Paris. Galilée.
Eco, Umberto (1990). Limits of Interpretation. Bloomington: Indiana University Press.
Grice, Paul (1975). Logic and conversation. In Syntax and Semantics, vol. 3, Speech Acts,
P. Cole and J. Morgan (eds.), 41–58. New York: Academic Press.
Koselleck, Reinhart (1999). Kritik und Krise. Frankfurt/Main: Suhrkamp.
Landow, George (1997 [1989]). Hypertext 2.0. Baltimore, MD: Johns Hopkins University
Press.
Peirce, Charles Sanders (1931–1966). Collected Papers of Charles Sanders Peirce, vols. 1–8,
Charles Hartshorne, Paul Weiss, and A. W. Burks (eds.). Cambridge, MA: Harvard
University Press. [Reference to Peirce’s papers will be designated CP.]
Searle, John (1979). Expression and Meaning. Cambridge: Cambridge University Press.
— (1993). Literary theory and its discontents. Unpublished paper delivered at a conference in
Berkeley, California.

Uwe Wirth (b. 1963) is Assistant Professor at Universität Frankfurt 5U.Wirth@

lingua.uni-frankfurt.de4. His research interests include Peircean semiotics, abductive
inference, Romantic literature, and hypertext theory. His major publications include
‘Abductive reasoning in Peirce’s and Davidson’s account of interpretation’ (1999), Diskursive
Dummheit. Abduktion und Komik als Grenzphänomene des Verstehens (1999), Die Welt als
Zeichen und Hypothese. Perspektiven der Peirceschen Zeichentheorie (2000), and Performanz.
Zwischen Sprachphilosophie und Kulturwissenschaften (2002).