The Classification of Peirce’s Interpretants

Brendan Lalor. Semiotica 114-1/2, 31-40, 1997.

Note: Thomas Short wrote a response to my article in the Transactions of the Charles Sanders Peirce Society (1996, “Interpreting Peirce’s Interpretant: A Response to Lalor, Liszka, and Meyers,” 32:4, pp. 488-541)

Abstract. After characterizing the role of the interpretant in semiosis, I consider two passages in which Peirce makes a threefold division of interpretants, one from 1906, one from 1909. Then I suggest that Thomas Short and others are wrong in holding that in the two passages, Peirce put forward two completely separate trichotomies. Instead, I argue that the 1906 trichotomy is in fact a special case of that put forward by Peirce in the 1909 passage, not a separate trichotomy. I then explain more specifically how we ought to conceive the relationship between these two classifications.

[N]o present actual thought (which is a mere feeling) has any meaning, any intellectual value; for this lies not in what is actually thought, but in what this thought may be connected with in representation by subsequent thoughts; so that the meaning of a thought is altogether something virtual. (Charles Sanders Peirce 1868: 5.289)[ 1 ]

The meaning of ‘interpretant’ in the context of semiosis

The concept ‘interpretant’ derives its meaning from its role in Peirce’s theory of semiosis. Semiotic is the study of relations between signs, which are triadic by definition. A sign stands for its object to an interpretant sign, at least potentially. Peirce distinguishes two objects of the sign: the immediate object within the sign, which is the object as mentally represented by the sign, and the mediate or dynamoid object without, which is the object in itself (Hardwick 1977: 83f., 32).

Suppose, for example, a visual image[ 2 ] of a mug stands for the mug to me. This situation is only describable by an irreducibly three-place relation. A perspective view of the mug is a sign (of both kinds of object), which determines an interpretant. This interpretant is either equivalent to the sign, or more developed. If it is equivalent, when it subsequently serves as a sign, it will convey to its interpretant just what the first sign conveyed. But if it is more developed — for example, if the mug is interpreted not just as a mug, but as a coffee mug — the interpretant, when it subsequently functions as a sign, will convey more information about the object than the first sign did, provided the further determination of the mug as a coffee mug was guided by a logically good habit of interpreting.[ 3 ] As more collateral information is acquired, the immediate object becomes more determinate and better able to represent the dynamoid object in its concrete richness, which nevertheless remains intractable. The information conveyed by successive interpretants also increases as new perspective views further determine the immediate object. But the immediate object is never completely determinate, so as to represent every property of the dynamoid object; for then it would be ‘determinate in each one of the more than millions of respects in which things may vary’ (MS 318: 39f., in Peirce 1907: 391). As representations, signs always have some vagueness. ‘No concept, not even those of mathematics, is absolutely precise’ (c. 1906: 6.496). Thus, no interpretant is informationally determinate in every respect.

Two Classifications of Interpretants

Having thus indicated the role of the interpretant in semiosis, let us turn to Peirce’s interest in further specifying the concept of ‘interpretant’ itself. Peirce conceived of pragmatism as a method of ascertaining the meanings of intellectual concepts. In 1906, he argued that the only way pragmatism can ascertain such meanings is through the study of interpretants, or ‘proper significate effects’ of signs (5.475). He then proceeded to elaborate a trichotomy of emotional, energetic, and logical interpretants. ‘The first proper significate effect of a sign is a feeling produced by it,’ hence the emotional interpretant (5.475). The energetic interpretant is any further effect a sign might produce; this will always involve a mental or muscular effort and will always be mediated through the emotional interpretant. Thus, any energetic interpretant will involve an emotional interpretant as its condition. To take a coarse example, when Jones’ comrade yells, ‘Look out! Duck!’ the energetic interpretant involves the muscular effort it takes Jones to bend his knees and crouch over, ducking. But this interpretant depends on an emotional interpretant — i.e. it depends on Jones’ feeling of recognition of the qualitative character of the auditory signal involved in the shouted words, ‘Look out! Duck!’

Since the meaning of an intellectual concept, qua concept, is general, and neither of these two classes of interpretant is general, neither could be the meaning of such a concept. Peirce designates the logical interpretant as the meaning of a concept, and further breaks this division down into logical and ultimate logical interpretants (5.476). While the logical interpretant may be an intellectual sign, such a sign itself has a logical interpretant, and thus is not the ultimate logical interpretant of a concept. The ultimate logical interpretant is the concept’s ‘living definition,’ an interpretant which itself has no further interpretant (cf. c. 1906: 5.491). This, Peirce concludes, must be a habit. A habit is a disposition to perform a certain operation, given some mental content. For example, an utterance of ‘cuello’ might trigger a hearer’s disposition to associate the Spanish word ‘cuello’ with the English word ‘neck’. ‘The habit alone, which though it may be a sign in some other way, is not a sign in that way in which that sign of which it is the logical interpretant is the sign’ (c. 1906: 5.491). The description of the habit — of both the action to which it gives rise and under what circumstances — is ‘the most perfect account of a concept that words can convey.’ All this is the 1906 trichotomy.

In 1909, too, Peirce presented a classification of interpretants — specifically: immediate, dynamical, and final interpretants.[ 4 ] These are, respectively, the total unanalyzed effect the sign first produces, the direct actual effect on the interpreter, and finally, ‘the effect the Sign would produce upon any mind upon which circumstances should permit it to work out its full effect.’

Two main views have been put forward as to the relation of the 1906 and 1909 terminologies, the first asserting their semantic uniformity, the second their semantic distinctness. In the first camp, some scholars have held that the 1909 trichotomy is coextensive with the one of 1906 — that Peirce was simply exploring various terminological possibilities. Others in this camp, such as J.J. Liszka (1990), hold that the terminologies are not merely synonymous, but complementary in the sense that they clarify one another.[ 5 ] Scholars in the second camp, most notably Thomas Short (1981: esp. 212f., 1982: esp. 286-288), have held that the 1909 classification is a distinct second trichotomy of interpretants. According to Short (1982), Peirce ‘based these trichotomies on different principles which, if consistently followed, lead us to the view that the two trichotomies are not identical but intersect, yielding nine types of sign in all’ (306). That is, he holds that each of the immediate, dynamic, and final interpretants may be divided into emotional, energetic, and logical interpretants.

While it is not my purpose to provide a full analysis of Short’s position, a sample thereof will prove to be a useful specimen with which to contrast my view. For example, he argues that since ‘feelings do actually occur and habits and certain other laws do come to be formed, modified, or destroyed,’ thus, ‘a dynamic interpretant can be emotional or logical and need not in every case be energetic’ (1981: 213). So, since the alteration of a given habit is an actual event, thus, a dynamic interpretant, which Peirce also says is an actual event, can without contradiction also be said to be a logical interpretant. However, although habits do come to be altered, this fact goes no distance toward showing that the dynamic interpretant can be a logical interpretant. Even in a case in which a dynamic interpretant is an event affecting the alteration of a habit, Peirce would say that the logical interpretant is the result of the habit change, not just the event of its change. While he held that the dynamic interpretant is a single actual event, it was also his view that the logical interpretant is, as we have pointed out, a general concept. The idea that somehow a general concept is reducible to a singular event would have seemed absurd to Peirce — and this is just what is implied by the view that a dynamic interpretant can be a logical interpretant. As Peirce said of the emotional interpretant: ‘It never can be the meaning of an intellectual concept, since it is a single act, [while] such a concept is of a general nature’ (c. 1906: 5. 475). Prima facie, then, this is a serious difficulty for Short’s view.

I will offer an interpretation which alleviates the need for such maneuvers as the ones that might be offered to explain this puzzle. My view is both simpler and more consistent with Peirce’s general approach. While I agree with Short that the two trichotomies are not identical in an important sense (contra the semantic uniformity view), I think both views are mistaken.

Not only did Peirce regularly refer to the interpretant-trichotomy, he presented at least four other interpretant-trichotomy terminologies.[ 6 ] How can these facts be reconciled? The answer to this question will become clear in the next section. The question for the view purporting the semantic distinctness of the classifications is, Why should these not also be candidates for still more distinct and intersecting trichotomies? For instance, in 1906, Peirce introduced an intentional/effectual/communicational classification of interpretants to explain the nature of verbal communication.[ 7 ] Since this classification of interpretants is not obviously identical with either the 1906 or 1909 trichotomies, it would seem that, to be consistent, a scheme such as Short’s ought to count this as yet a third interpretant-trichotomy, whose intersection with the first two yields 27 types of interpretant. On the other hand, as we shall see, this fits nicely with my view.

How we Ought to Conceive the Relation of the Two Classifications

My thesis is that the emotional/energetic/logical classification is a special case of the immediate/dynamical/final one. More specifically, the 1906 trichotomy reflects the concrete human case, the human experience of semiosis, while the 1909 trichotomy is more abstract and lends itself to a characterization of semiosis generally. This relation is analogous (but only analogous) to the relation of the phenomenological to the metaphysical categories. That is, for example, just as a quality of red which exists intentionally in a feeling is how we experience quality (i.e. firstness), so also, an emotional interpretant (i.e. a feeling produced by a sign), is our version of the immediate interpretant of a sign (i.e. ‘the total unanalyzed effect the sign … might be expected to produce’(Hardwick 1977:110)). So, the relation of Peirce’s references to the interpretant-trichotomy, and his references to other interpretant-trichotomy terminologies might be said to be that of genus to species.

In 1906, Peirce confessed that the only way we comprehend the idea of ‘interpretant’ — taken as a general concept — is by assuming that it is sufficiently analogous to our experience of modifications of consciousness.

Although the definition does not require the logical interpretant (or, for that matter, either of the other two interpretants) to be a modification of consciousness, yet our lack of experience of any semiosis in which this is not the case, leaves us no alternative to beginning our general inquiry into its general nature with a provisional assumption that the interpretant is, at least, in all cases, a sufficiently close analogue of a modification of consciousness to keep our conclusion pretty near to the general truth. (c. 1906: 5.485, my emphasis)

Peirce hoped to be able to generalize the conclusions reached from our human point of view, to characterize semiosis universally.[ 8 ] He recognized that to do so unprovisionally would be to base a proposition of logic on a proposition of psychology. Thus, he stressed that the soundness of the generalization is conditional on the truth of the assumed analogy between modifications of human consciousness and interpretants in general.

In this light, it should be no surprise that his 1906 classification of the interpretants as emotional, energetic, and logical, reflects an anthropomorphic way of looking at semiosis. The 1909 trichotomy lays down a general structural pattern which Peirce believed can be found in all kinds of semiosis. The 1906 taxonomy applies specifically to the way in which human semiosis manifests that structure. To illustrate how we ought to reconcile the two trichotomies, I will briefly analyze the relationship between the emotional and immediate interpretants, which will suffice to demonstrate the strategy for relating the others.

In the 1906 passage, the method Peirce had in mind, by which the emotional interpretant or first proper significate effect would be individuated, was phenomenological introspection of modifications of consciousness. Consider the mug example. When we open our eyes and look toward the mug, suppose that roughly, our visual system operates on the visual input data so as to produce a sign by the interpretation of which we ‘see through’ to the intentional phenomenon, i.e. the immediate object, which is the mug. In terms of the underlying semiotic process, the emotional interpretant is the feeling ‘which we come to interpret as evidence that we comprehend the proper effect of the sign’ (c. 1906: 5.475). It is the minimal phenomenologically noticeable modification of consciousness. This feeling, if it could be adequately verbalized, would be something like, ‘familiar visual array X (is experienced).’ Perhaps even this description of the emotional interpretant, the feeling of sign-recognition, is too determinate.

But notice that even the smallest modification of consciousness is an experience, which in turn is a non-instantaneous event.[ 9 ] As Peirce said in 1868, ‘the striking in of a new experience is never an instantaneous affair, but [comes] to pass by a continuous process’ (5.284). Thus, by saying the ‘emotional interpretant’ is the ‘first effect of a sign,’ Peirce did not mean it is the sign in a related series which appears the instant after the previous one. For Peirce thought it reasonable to suppose that there are signs of a finer grain than those individuated in terms of modifications of consciousness.

It may possibly be, for example, that I am taking too narrow a conception of the sign in general in saying that its initial effect must be of the nature of feeling, since it may be that there are agencies that ought to be classed along with signs and yet that at first begin to act quite unconsciously. (MS 318: 43, in Peirce (1907: 392))

So by the phrase, ‘first effect of a sign,’ in this case, Peirce need not have meant to pick out a supposed instantaneous event, or even the narrowest definable event possible. Since the emotional interpretant thus conceived would be too fine grained for us to be conscious of it, this conception of it would commit Peirce to the view that experiencing it is beyond human capacities, and therefore we can have no experience of semiosis. Thus, we ought to take ‘first effect’ here as a more or less vaguely delimited ‘segment’ of continuous sign-interpretation, beginning with the unanalyzed sign and ending with a feeling of recognition of the sign of the mug. This segment of continuous sign activity, of various operations, comprises a single emotional interpretant. The interpretant supervenes on the lower-level activity — the details of which are too fine to be detected by human introspective capacities. Therefore, in his 1906 presentation of the theory of interpretants, when Peirce referred to the ‘first effect’ as the ‘emotional interpretant,’ he was taxonomizing this interpretant in terms of the minimum effect noticeable by humans.[ 10 ] He nevertheless explicitly recognized that the need for generalization.

The human conceptions of all three interpretants are based on the experience of modifications of consciousness. I suggest that Peirce’s 1909 trichotomy is the result of his generalization of these human conceptions in an attempt to characterize semiosis universally. On this view, the immediate, dynamical, and final interpretants are Peirce’s meta-theoretical generic place-holders for interpretants which play a role in semiosis taking place at any and all levels of reality — even levels of concreteness too low, or levels of abstraction too high, for humans to notice without the aid of instruments or theoretical speculation. Emotional, energetic, and logical interpretants are the theoretical terms for a species of interpretants with which humans are intimately acquainted.

The sign activity on which the emotional interpretant supervenes, which I mentioned above, is a fine-grained pre-conscious species of semiosis to which we do not have introspective access. As an example of such semiosis, take visual sense inputs as signs: the mind, by some pre-conscious process of interpreting them, gains a three-dimensional understanding of its world. Evidently enough, the interpretants involved in this kind of semiosis would be individuated by some principle other than that based on modifications of human consciousness. For we are not conscious of any such ‘construction.’ The underlying principles which guided Peirce’s generalization were familiar to him early on. For example, in ‘Some Consequences of Four Incapacities’ (1868), he pointed out that perception involves mental operations in which simple predicates are substituted for complex ones. The simpler sign is the interpretant of a pre-conscious complex of sensations.

To point to a few implications, the distinction between the emotional and immediate interpretants is not one of kind, then, but one of level of analysis. The 1909 trichotomy can be used to characterize semiosis which is finer-grained or coarser-grained than that to which the 1906 trichotomy applies. It allows the individuation of interpretants to be indefinitely narrower (as may suit theorizing about pre-conscious mental activity), or indefinitely wider (as may suit theorizing about public communication,[ 11 ] or economics). Thus, it is more general in applicability, since unlike the emotional interpretant, the immediate interpretant does not specify one perspective or principle of individuation. It only characterizes the structural pattern to be found. Also, my view explicitly allows for the supervenience of one type of semiosis on another.

One might argue that even if my view is right, Short’s view, that the two trichotomies of interpretant intersect yielding nine types in all, could be right as well, in the following sense. Perhaps what at one level of analysis is an immediate interpretant, could turn out to supervene on what at a lower level of analysis are emotional, energetic, and logical interpretants — and likewise in the case of dynamic and final interpretants. In this way, for example, perhaps a dynamic interpretant could in a sense also be said to be a logical interpretant. However, Short is committed to the conceptual clarity of the proposition, ‘this dynamic interpretant is a logical interpretant’. This is quite different from what my view asserts as conceptually clear: i.e. that ‘this dynamic interpretant in part supervenes on a logical interpretant’ — not that it is one. I will not make a judgment here about the prospects for working out some unnoticed way of showing that something like Short’s view is conceptually clear after all. If such a partial vindication is possible, however, I fail to see how it can be made apart from exploiting the notion of coarser- and finer-grained levels of semiotic analysis.

While I have not analyzed the other two kinds of interpretant, I want to comment on the last kind, the ‘final interpretant’ of 1909. By defining it as ‘the one Interpretive result to which every Interpreter is destined to come if the sign is sufficiently considered’ (Hardwick 1977: 111), Peirce’s general 1909 presentation of the theory provides a context for discourse about the truth of an interpretant. Such an interpretant would be a true, precise representation of the dynamical object mentioned above. Even though we have pointed out in the first section of this paper that no interpretant is informationally determinate in every respect, the human version of the final interpretant is for us an ideal. It would result from an indefinite series of interpretations of signs, perhaps by sign processing beings with fewer ‘incapacities’ than human beings. To say that the final interpretant is within our possible reach is the expression of a hope. The 1906 presentation, on the other hand, specifies the context as that of human semiosis, in which discourse about the ultimate logical interpretant is about meaning, not necessarily truth. The hope of science is that eventually the ultimate logical interpretant — that Homo sapien version of the final interpretant — will perfectly correspond to the final interpretant itself. Then we will have carved the world at its joints.[ 12 ] This is, as Pape (1991) put it, ‘the intellectual hope that the sequence of interpretations — perhaps there are infinitely many of them and we are connecting one infinity with another — will ultimately represent reality’ (174).

References

Hardwick, Charles, ed. (1977). Semiotic and Significs: The correspondence between Charles S. Peirce and Victoria Lady Welby. Bloomington: Indiana University Press.

Liszka, J.J. (1990). Peirce’s Interpretant. Transactions of the Charles Sanders Peirce Society 26, 17-62.

Meyers, Robert (1992). Peirce’s New Way of Signs. Transactions of the Charles Sanders Peirce Society 28, 505-521.

Pape, Helmut, ed. (1990). Charles S. Peirce on Objects of Thought and Representation. No?s 24, 375-396. Pape introduces this edited version of MS 318.

_____ (1991). Not Every Object of a Sign has Being. Transactions of the Charles Sanders Peirce Society 27, 141-177.

Peirce, C.S. (1931-35). Collected Papers of Charles Sanders Peirce. Vols. I-VI, eds. Hartshorne and Weiss. Cambridge, Mass.: Harvard University Press. Cited as ‘[year of original penning]: [volume].[paragraph].’

_____ (1907). MS 318. In Pape (1990).

Short, T.L. (1981). Semeiosis and Intentionality. Transactions of the Charles Sanders Peirce Society 17, 197-223.

_____ (1982). Life Among the Legisigns. Transactions of the Charles Sanders Peirce Society 18, 285-310.

___________________________

Notes

[ 1 ] References of the form ‘[year]: [volume].[paragraph]‘ refer to Peirce (1931-35).

[ 2 ] I use the term ‘image’ here as a kind of free variable to refer to whatever is the visual percept.

[ 3 ] We may be wrong in extrapolating from a sign to a certain interpretant, as in any hypothesis. Robert Meyers (1992) points out that what are today customary associations of sign and interpretant, were in very many cases originally related by abductions (517). A similar relationship obtains between our sensations and their stimuli. Sensation ‘is a simple predicate taken in place of a complex predicate; in other words, it fulfills the function of an hypothesis’ (1868: 5.291).

[ 4 ] Hardwick (1977: 109-111). Note: He had introduced the 1909 classification earlier, in 1906 in fact, but I refer to this as ‘the 1909 trichotomy’ for both the convenience of talking in terms of distinct years, and the clarity of the trichotomy as he offered it in 1909.

[ 5 ] See e.g. Liszka (1990: esp. 20-26). His position may be closest to my own. He holds that just as the 1906 and 1909 classifications complement one another, this

appears to be the case for the classification of his categories. There is the logical division of monad, dyad, triad; the metaphysical distinction of quality, fact, law; the phenomenological distinction of firstness, secondness and thirdness. Each set complements and clarifies the other. Thus, for example, signs are explained in terms of each set of divisions: a sign is a first and monadic; the interpretant is triadic and law-like, etc.’ (1990: 26).

As will become clear, however, on my hypothesis the interpretant classifications do not simply complement one another; the later classification is a generalization of the earlier.

[ 6 ] E.g. see Peirce’s letters of 1904 and 1908 in Hardwick (1977: 32, 85); cf. c. 1906: 5.475. For instance, he mentions ‘my gropings after the three kinds of interpretant‘ (Hardwick 1977: 109 — letter of 1909). While he did think the classes of interpretants were further divisible (cf. 85 — letter of 1908), he nevertheless conceived of these as the three basic classes. And neither of the further divisions seems to resemble those Short has in mind. Peirce rarely mentions the two trichotomies together, and never says they are distinct, so far as I know; when he mentions them together in MS 318, rather than distinguish them, he ostensibly identifies them.

The four other interpretant-trichotomy terminologies I have in mind were mentioned (i) in 1908, i.e. the destinate, effective, and explicit interpretants (Hardwick 1977: 85); (ii) in 1904, i.e. the immediate (the sign’s interpretant as represented or meant to be understood), the dynamic (the interpretant as it is produced), and without naming the third, he described it as a sign’s interpretant ‘in itself’ (32, 34f.); (iii) in 1893 (see 2.294), and (iv) in 1906, i.e. the classification introduced to explain the nature of verbal communication (Hardwick 1977: 196).

[ 7 ] In the intentional/effectual/communicational classification of interpretants, the first kind of interpretant ‘is a determination of the mind of the utterer,’ the next ‘a determination of the mind of the interpreter,’ and the last, ‘a determination of that mind into which the minds of utterer and interpreter have to be fused in order that any communication should take place’ (Hardwick 1977: 196).

[ 8 ] In a 1908 letter, Peirce told Lady Welby that although he defines interpretant as an effect ‘upon a person,’ he does so because ‘I despair of making my broader conception understood’ (Hardwick 1977: 81).

[ 9 ] Since there is no sign-interpretation in an instant, any thought, considered from the standpoint of being ‘immediately present’ to consciousness, is a mere unanalyzed feeling, void of cognitive significance (1868: 5.289). Insofar as any temporal span of experience is significant or meaningful — however short a span you please — its meaningfulness depends on the interpretation of signs.

[ 10 ] Analogously, we represent ideal ‘extensionless points’ by using points with some visible extension.

[ 11 ] Cf. note 7. This fits nicely with my view in the sense that the present classification provides useful categories for taxonomizing sign interpretation from the perspective of verbal communication as such.

[ 12 ] Cf. e.g. where Peirce mentions the discovery of ‘an idea of a leaf which includes every part of the flower, and an idea of a vertebra which includes the skull,’ in ‘The Doctrine of Chances’ (1878: 2.646).