Virtual Centre for Peirce Studies at the University of Helsinki


Paper originally published in the Transactions of the Charles S. Peirce Society: A Quarterly Journal in American Philosophy XXXVI, No. 2 (Spring 2000), pp. 225-254. Published on the Commens site with the kind permission of the editors of the Transactions journal.

The page numbers of the Transactions article are marked with {red}.

Mats Bergman
The University of Helsinki

1. Introduction: The Problem of Communication

While studies in the philosophy of Charles S. Peirce have established the seminal role of the semiotic perspective in his thought, there are still many aspects and possibilities of his sign-theoretical output that have received only slight attention from philosophers and other scholars. In general, investigations of Peirce's semeiotic[1] have tended to focus on the branches of grammar and critic, i.e., the studies of the general conditions of signhood and of the truth-conditional relation between sign and object. The third branch of semeiotic inquiry, which Peirce usually calls rhetoric or methodeutic, has attracted far less interest (Santaella Braga 1999: 379 f.). This is understandable, as his semeiotic writings appear to be primarily concerned with questions of grammar and critic, such as the classification of sign-types and the problems of formal logic. In comparison, Peirce's discussions of explicitly rhetorical matters are few and sketchy, and he never quite manages to pin down the precise character of rhetorical investigation (Liszka 1996: 78 f.). Yet, Peirce implies that rhetoric may, in the end, be the most important semeiotic endeavour of them all (see CP 2.333 [c. 1895]; EP 2:256 [1903]).[2] Hence, investigations of a rhetorical character ought to broaden our understanding of Peirce's theory of signs and perhaps even open up new possibilities for semiotic and philosophical inquiry. Of course, before this can be achieved it is necessary to establish, in a preliminary fashion at least, what rhetoric does or should contain.
One natural candidate for a place within semeiotic rhetoric is the philosophical study of communication. Several of Peirce's characterisations of rhetoric point in this direction, as he states that rhetoric is concerned with the study of the transmission of meaning by signs, and the ways in which one sign gives life to another (CP 1.444 [c. 1896]; CP 2.229 [c.1897]; NEM 4:331 [1898]). Peirce frequently speaks of the 'dialogic' character of thought (a prime example of semiosis, the action of signs), and even defines the sign as a 'medium of communication' in his mature semeiotic. Such expressions suggest that communicative considerations are an intrinsic part of semeiotic analysis. On the other hand, Peirce's explicit references to 'communication' are rare occurrences {226} (Habermas 1995: 243). It is uncertain whether his scattered remarks on the topic entail a consistent theory of communicative phenomena;[3] one may even question whether semeiotic can be plausibly developed in this direction at all. Peirce's most formal accounts of the sign relation, given in strictly unpsychologistic terms without references to human utterers and interpreters, indicate that pure semeiotic is after all only concerned with the abstract conditions of representation and truth, and that communication is a non-philosophical problem-area that is best left to the special sciences.[4]
Furthermore, Jürgen Habermas (1995) and Richard Parmentier (1985; 1994) have noted certain difficulties that cast doubts upon the possibilities of developing a thoroughgoing communicative conception of semeiotic. In rough outlines, Habermas's criticism is based on the argument that Peirce's evolutionary metaphysics, with its wide-reaching cosmological speculations, undermines his earlier social theory of logic and inquiry. The principal question here concerns intersubjectivity, and the fact that Peirce's peculiar combination of objective idealism and naturalistic realism seems to eliminate agency from his semiotic picture of the world (Habermas 1995: 261). However, although some questions of relevance for this problem will be touched upon in section 4 of this essay, I shall not address Habermas's challenge here, but choose to focus on some difficulties identified in or implied by Parmentier's account instead.[5] His criticism, which is also directed at certain aspects of Peirce's later thought, is of special interest because it targets the semeiotic definition of the sign in terms of communicative mediation. According to Parmentier (1985: 32), this very conception, which involves an ideal of transparency, restricts the utility of semeiotic to truth-functional epistemology and mathematical logic, and thus renders Peirce's theory of signs practically useless for other types of inquiry, such as studies of culture and social communications. The irony is obvious: precisely those passages that appear to give the strongest support to a communicative reading of semeiotic may in fact be directly damaging to such an interpretation, and even to the attempts to find a place for the study of communication within rhetoric.
The aim of this paper is to argue that Peirce's later characterisations of the sign-relation as communicative mediation do not necessarily block the road of semiotic inquiry. On the contrary, I wish to show that these sketchy definitions imply certain possibilities, which can open up paths that have been rarely, if ever, explored. However, I feel that this endeavour requires both a liberal interpretation of Peirce's philosophical project and the support of a number of reconstructive measures. Parmentier's criticism touches on important problems, and I do not think that it can be simply rejected by pointing out interpretive inaccuracies, even if such mistakes could be detected. I fear that we are dealing with such a fragmentary and indeterminate portion of Peirce's philosophy here, with considerable latitude of interpretation, that no final truth can be found merely by minute examination of his statements. In other words, I am not {227} primarily concerned with the textual accuracy of Parmentier's reading of Peirce, but rather with the question of what would be a productive interpretation of the problematic passages in view of future inquiries. Pace Parmentier, I shall contend that the definition of the sign-relation in terms of communicative mediation is both philosophically illuminating and open-ended enough to allow for various kinds of theoretical and practical investigations, as long as it is complemented by certain additional semeiotic and pragmaticistic considerations. For this reason, I shall also state the case for a broad conception of semeiotic rhetoric, which is more extensive in scope than scientific methodology in a strict sense.
I shall proceed through the following steps. First, I will consider Parmentier's criticism, and reconstruct its main line of argument. In the third section of the paper, I intend to take a closer look at some of the problematic concepts involved in the definitions that Parmentier criticises. Without denying the presence of certain difficulties in these notions, I shall argue that they can be meaningfully interpreted in pragmaticistic terms. Next, I will consider the problem in a slightly wider perspective, and make a number of observations on how the basic communicative sign definition may be augmented.[6] Finally, I shall consider the possible implications of such communicative considerations for the outlook of rhetoric, and also show that semeiotic involves implicit critical potential, which could be actualised if rhetoric were to be developed in a communicative direction.

2. Transparent Transmissions: Parmentier's Criticism of Peirce's Later Semeiotic

In general, it can be argued that the study of Peirce's semeiotic has reached a point where certain central findings, such as the triadic character of the sign and its reliance on Peirce's categories of Firstness, Secondness, and Thirdness, have been established firmly enough to speak of an interpretive consensus. At the same time, slightly less obvious discoveries and ideas, such as the important insights about the teleological nature of the semiotic process,[7] have been gaining wider acceptance. Still, we should not close our eyes to the fact that there are possible problems at the very centre of Peirce's theory of signs – difficulties that have not been considered quite as extensively as they deserve. Indeed, one of the virtues of Parmentier's critical account is that it draws our attention to one such widely ignored question, namely the tension between Peirce's characterisations of the sign relation in terms of mediate representation and his definition of the sign as a medium of communication. For contemporary studies of semeiotic, where emphasis is often (and perhaps somewhat uncritically) placed on the processual character of Peirce's conception of the sign, Parmentier presents a challenging hypothesis, as he holds that Peirce's earlier focus on the representative function of the sign is in many senses a more adequate starting-point for semiotic inquiry than his later ideas of communicative determination.[8]
In order to understand Parmentier's rejection of Peirce's definition of the {228} sign as a medium of communication, it is necessary to consider the general drift of his argument, and reconstruct the way he conceives of the central semiotic operations of representation, determination, communication, and mediation in various ways, this set of concepts figures in Peirce's numerous definitions of the sign, the common denominator being that the concepts can be said to indicate or signify the behaviour of signs, i.e., semiosis. Although there is no firm evidence that Peirce wishes to signal major modifications in his project by this changing terminology, the terms employed are certainly not straightforwardly synonymous. They carry differing connotations, which in turn may affect our perception of the scope and character of semiosis as an object of study. Hence, it is not trivial how we choose to characterise the process. One way of getting a firmer grasp of Peirce's many sketches of semiosis is to analyse the relationship between the semiotic operations, and perhaps even to develop these further, if necessary.
In spite of the often confusing plurality of terms, it is reasonably clear that mediation, which is almost equivalent to Peirce's third category (see, in particular, CP 1.328 [c. 1894]; CP 2.88 [1902]; PPM 193 f. [1903]; NEM 4:308), is the most generic way of characterising semiosis. That is, whatever else semiotic representation, determination, and communication may turn out to be, they can plausibly be construed as modes or aspects of mediation. In semeiotic, mediation is intimately associated with the fundamental semiotic relationship, taken as a triadic whole; the mediating sign is something that brings two other semiotic subjects into a certain kind of irreducible relation. Or, expressed differently, the sign mediates between the object and the interpretant (EP 2:410 [1907]).

In its genuine form, Thirdness is the triadic relation existing between a sign, its object, and the interpreting thought, itself a sign, considered as constituting the mode of being of a sign. A sign mediates between the interpretant sign and its object. Taking sign in its broadest sense, its interpretant is not necessarily a sign. [—] A sign [...] is an object which is in relation to its object on the one hand and to an interpretant on the other, in such a way as to bring the interpretant into a relation to the object, corresponding to its own relation to the object. I might say 'similar to its own' for a correspondence consists in a similarity; but perhaps correspondence is narrower. (SS 31 f. [1904].)

Hence, mediation can be characterised in terms of semiotic connection, a distinctive way of binding together three elements. However, this general formulation immediately provokes the question of how this mediating connection is produced, or how it might be further analysed. Parmentier claims that we can distinguish two basic 'vectors' in the sign relation; the vector of {229} representation, which points from the sign and the interpretant toward the object, and the vector of determination, which points from the object toward the sign and the interpretant.[9] According to Parmentier (1985: 27), determination "is the causal process in which qualities of one element are specified, transferred, or predicated by the action of another element". In semiosis, it is a process in which the object acts upon the sign, which in turn acts upon the interpretant. In this way, the sign brings about a mediated determination of the interpretant by the object. Representation, in Parmentier's account, works in the opposite direction. It is characterised as "the act or relation in which one thing stands for something else to the degree that it is taken to be, for certain purposes, that second thing by some subject or interpreting mind" (ibid.). In both of these vectors, the sign or representamen[10] occupies a mediating position between object and interpretant, and the object functions as a constraining factor of semiosis; yet, the types of semiotic action involved are different. In the case of determination, the action of the object must pass through the sign in order to reach the interpretant and determine some effect, such as cognition. In the second vector, the representation formed by the interpretant is limited by the fact that there must be some kind of prior 'standing for' relation between sign and object, in order for representative semiosis to be possible at all (ibid.: 28 f.). Peirce certainly acknowledges the relevance of this connection between sign and object, and offers an analysis of it in his well-known trichotomy of icon, index, and symbol.[11]
Further, Parmentier claims that the vectors of representation and determination interlock in a manner that accounts for the characteristic processuality of semiosis. He notes that there are two infinite series involved in semiosis – back toward the object and forward toward the interpretant – and implicitly affirms the idea that has become known as 'unlimited semiosis'.[12] This reading of semeiotic is based on certain passages, in which Peirce lets us understand that not only the interpretant (which he often characterises as a further sign produced by the original sign) but also the object involves endless representation (see, e.g., NEM 4:309; cf. CP 2.92 [1902]; CP 2.203 [1902]). Parmentier's crucial claim, however, is that the two semiotic vectors operate on different levels of semiosis, and are thus not symmetric (Parmentier 1985: 29). This asymmetry is due to the fact that the sign can stand for the object in three different ways. The iconic connection is based on some quality shared by sign and object, while the indexical relation is existential; but it is the third, symbolic mode of representation that accounts for the asymmetry. According to Parmentier, determination operates at the same level of semiosis in all of its phases. Hence, it could be characterised as linear influence. In representation, however, a metasemiotic level is introduced, as the interpretant of the symbol represents its object by forming a conception of the prior relation between sign and object. That is, we move to this second level of semiosis when the symbol's interpretant-sign represents the relation between sign and object as a new semiotic entity. Parmentier calls it object2, and claims that it is identical to Peirce's {230} ground (ibid.: 30).[13] Perhaps we can say that this semiotic accomplishment is due to the self-referential character of the symbol, which represents itself to be represented (EP 2:323 [c. 1904]).
Although Parmentier does not explicate the nature and relevance of the metasemiotic level in detail, it is obviously of utmost significance for both his critical interpretation of Peirce and his own conception of semiotic inquiry. Namely, it is implied that the process of symbolic representation, in which habits and conventions are involved, accounts for both the growth of meaning and the essential interconnectedness of signs. To the extent that it is symbolic, semiosis tends to produce more and more extensive objects or grounds, which not only serve to explicate the original object in a series of representations, but also form complex semiotic webs of signs and objects that would otherwise not be so connected (cf. Parmentier 1985: 31). In this manner, representations of second intention, in which the objects are semiotic entities, are formed (ibid.: 30). Parmentier's approach seems plausible; it would certainly help explain Peirce's often cryptic remarks to the effect that the object is also a sign (see, e.g., CP 1.538 [1903]; EP 2:328 [1904]; EP 2:380 [1906]). The first object can be construed as an ideal limit, which can be comprehended only within a representational process; it is never given as such, free from all representation.[14] Moreover, the postulation of a metasemiotic level also serves to highlight the creative role of the interpretant as a synthesising force, thus making it clear that semiosis is not mere determination; it involves a complication in the form of interpretation, which in fact is a necessary requirement for true semiotic development. The process of symbolic representation could be viewed as a highly abstract account of the basic dynamics of understanding and culture. Hence, the vectorial analysis of semiosis, coupled with a recognition of the importance of representation as a creative force, implies the applicability of semeiotic to a wide range of phenomena and inquiries.
To recapitulate: Parmentier recognises two principal processes involved in the more inclusive process of semiosis, the causal process of determination and the synthesising process of representation. In general, Parmentier finds this to be the acceptable upshot of Peirce's analyses; but he also wants to argue that Peirce's mature account of semiosis shows an unfortunate tendency to prioritise the determinative aspect. To be more specific, Parmentier feels that Peirce unwisely shifts his emphasis from representation to determination when the sign is defined as a medium of communication (Parmentier 1985: 32). In particular, Parmentier finds the following passage to be problematic:

[A] Sign may be defined as a Medium for the communication of a Form. It is not logically necessary that anything possessing consciousness, that is, feeling of the peculiar common quality of all our feeling should be concerned. But it is necessary that there should be two, if not three, quasi-minds, meaning things capable {231} of varied determination as to the forms communicated. As a medium, the Sign is essentially in a triadic relation, to its Object which determines it, and to its Interpretant which it determines [...] That which is communicated from the Object through the Sign to the Interpretant is a Form; that is to say, it is nothing like an existent, but is a power, is the fact that something would happen under certain conditions. This Form is really embodied in the object, meaning that the conditional relation which constitutes the form is true of the form as it is in the Object. In the Sign it is embodied only in a representative sense, meaning that whether by virtue of some real modification of the Sign, or otherwise, the Sign becomes endowed with the power of communicating it to an interpretant. (MS 793:1-3 [c.1905]; in Parmentier 1985: 42 f.; cf. EP 2:544 [c. 1905]; SS 197 [1906].)

Parmentier claims that Peirce's later semeiotic is marked by a tendency to see communication as an essential feature of all sign action. However, in this context communication does not necessarily refer to communication as it is ordinarily understood; according to Parmentier (1985: 42), Peirce associates communication with the transfer of truth in an object-interpretant continuum, which is endless in both directions. Furthermore, Parmentier states that truth and communication are perfectly isomorphic for Peirce, because communication is not construed in terms of social sharing of sign-knowledge, but rather as an argumentative dialogue between moments of one mind that realises the unity of semiosis (ibid.; cf. NEM 3:886 [1908]). Yet, this does not mean that Peirce would reduce his analysis of semiosis into a classical epistemological concern involving only the knowing subject and known object; that would certainly not be a viable option for an anti-Cartesian, who explicitly rejects such dualistic accounts of inference and cognition (see, e.g., SS 69 [1908]). Obviously, Peirce continues to affirm the need for mediation in his communicative definitions, as the sign is characterised precisely as a medium; but this perspective may nevertheless entail certain reductive consequences because of its causal linearity.
In Parmentier's interpretation, semiosis as communication is primarily a specification of semiosis as determination. At least, Peirce seems be concerned with the same vector of semiosis when he speaks of communication and determination in his definitions. The sign, which is determined by the object, determines the interpretant, so that a 'form' is communicated from object to interpretant. This transfer would take place on the same causal level of semiosis; no metasemiotic level of second intentions appears to be involved. Still, Peirce acknowledges the need for concrete expression; the form must be embodied in some manner, if it is to be conveyed from object to interpretant. In other words, some kind of vehicle of expression, which embodies the form representatively in the process, is required. This conveyance is the role assigned to the sign; its {232} mediating character is that of a carrier capable of determination. The form, which in Parmentier's reading is more or less equivalent to cognitive truth, cannot exist as such; if a thought is to have any active mode of being, it must be embodied in a sign (SS 195 [1906]; cf. NEM 3:406 [1903]; CP 4.7 [c. 1906]).
At the same time, Peirce appears to expound an ideal of semiotic transparency (Parmentier 1985: 43 f.). As anyone familiar with Peirce's logic knows, he is careful to distinguish the proposition from its accidental expression. That is, the proposition is not supposed to be affected by the material shape it is given for the purposes of outward or inward communication, but remains the same proposition whether it is asserted or denied, stated in English or Finnish, etc. (see, e.g., EP 2:312 [c. 1904]; CP 8.313 [1905]). Such a perspective implies that there is a central core to any thought that can be communicated, in spite of the variety of ways that can be used to express it. This, in turn, would seem to accord with Peirce's view that the embodiment of the form in the sign is 'merely' representative. According to Parmentier (ibid.: 43), the expressive vehicle that is required for communicative transmission does not contribute to the significant determination of the interpretant. That is, the function of the sign is to convey the form without affecting it (cf. EP 2:391 [1906]). However, we all know that the vehicles we use tend to influence semiosis; for instance, different languages offer different means of expression, enabling certain communications while limiting our capacities in other regards. Peirce is certainly aware of these restrictions on our semiotic capacities, and of the power of sign systems (see, e.g., W 1:494 [1866]; EP 2:10 [c. 1894]). Still, he never ceases to look for an adequate logical notation, in which the deficiencies of ordinary language could be overcome – that is, a mode of expression that would allow us to focus on the significant relations of communicated thought without being distracted or muddled by inessential extras. Parmentier (1985: 43 f.) argues that it is precisely for this purpose that Peirce creates his iconic system of existential graphs.
We are now in a position to see why Parmentier claims that Peirce's communicative definition of the sign limits the scope of semiosis and semiotic inquiry. By combining the requirement of expression with the ideal of transparency, Peirce ends up giving us a picture of semiosis as mere delivery of form. Contrary to the earlier account of symbolic representation, the sign is reduced to a kind of necessary evil, needed as a temporary vehicle, but of no relevance for the constitution of meaning. At the same time, the focus of semeiotic changes; its primary concern is now to conceive of a mode of expression that would be as transparent as possible. The representative function of the human mind is all but eliminated from the process, as the optimal sign is similar to a perfect translating machine (ibid.: 45). In other words, there is no growth of meaning by symbolic representation. Hence, Parmentier concludes that "Peirce in the end reduced the role of signs to being blind vehicles for communication of meanings that they do not influence" (ibid.); but it would perhaps be more illuminating to say that the sign only influences the process in a {233} detrimental way, as its expressive limitations make perfect transparency impossible. Semiosis is a process of transmission, in which the form transmitted ought not be affected by the media, but some disturbance is inevitable due to the necessity of embodiment.[15]
If Parmentier is right, then Peirce's conception of communication is both limited and limiting. As a strict logical ideal, it is not adequate for inquiries of social phenomena, such as everyday communication. In other words, the communicative definition of the sign relation reduces semeiotic's scope to truth-functional epistemology and logic (cf. Parmentier 1985: 25, 44). Consequently, it is inferior to the earlier idea of mediate representation, and almost of no use for attempts to investigate various modes of communication semiotically.
Reconstructed and slightly modified, the problems noted or implied by Parmentier can be summarised as follows:
1. The communicative definition places the emphasis on causal determination rather than mediate representation. Therefore, Peirce's later account of semiosis lacks a metasemiotic level, which could account for semiotic growth.
2. In communicative determination, the sign is reduced to a vehicle carrying form or truth from the object to the interpretant. It still has a vital function as a mediator; but it does not contribute anything positive to the process; it is simply a means to an end (cf. EP 2:5 [c. 1894]).
3. By simultaneously affirming the necessity of expression and the ideal of transparency, Peirce restricts his semiotic interests to logic and the effective expression of thought; other modes of semiosis are ignored or devalued as scientifically deficient. Consequently, Peirce ends up limiting the scope of semeiotic in a way that makes it an inappropriate starting-point for more comprehensive studies of communicative and cultural phenomena.

3. Form and Meaning

Parmentier's critical rejection of the communicative definition of the sign raises a number of questions concerning the adequacy of Peirce's later semeiotic. There is no denying that Parmentier makes many accurate observations about the focus and limitations of Peirce's conception. In general, however, Parmentier's claims threaten to invalidate attempts to incorporate Peirce's sparse comments into a philosophical or semiotic theory of communication (such as is presented by Johansen [1993a] and proposed by Colapietro [1995], for instance), with the exception of a strictly logical study of the conveyance of truth. In particular, the ideal of transparent media seems to transform Peirce's idea of communication into a naive transmission model, which is simply not viable after a century of linguistic philosophy and semiotics. Yet, one can note a number of omissions in Parmentier's account. In particular, he appears to ignore certain possibilities of connecting the communicative definition of semiosis to other Peircean concerns, such as those of his pragmaticism. Of course, it may be argued that these are {234} secondary matters in the truly scientific theory of signs; and indeed they are, if we choose to restrict semeiotic to grammar and critic. However, the potential value of adopting a broader perspective is that it can open up our eyes to certain suggestive hints in Peirce's writings, which may give us indications about how his admittedly sketchy ideas might be developed into a fuller communicative perspective, which would not limit the scope of his theory of signs. Moreover, this endeavour can, to a surprisingly large extent, be undertaken within the parameters of Peirce's semeiotic – although that requires a quite liberal interpretation of many of his central conceptions.
In order for such a project to be successful, it may be necessary to expand the basic conception of the sign relation. Although Peirce certainly strives for a general definition of the sign, which in its most abstract versions involves only object, sign, and interpretant in a triadic relation (e.g. CP 1.541 [1903]), many of his definitions involve additional conceptual ingredients. They may be reducible and perhaps even superfluous for strict semeiotic grammar; but for our purposes these auxiliary factors can be most illuminating, as they tend to crop up precisely in connection with the communicative definitions that Parmentier criticises.
One such ancillary component is the 'form' that is supposedly conveyed or transmitted in communicative semiosis.[16] For Parmentier, this form seems to be more or less equivalent to truth or a truthful idea, although he does not discuss this interpretation at great length. However, he suggests that the triad of mediation should not really be construed as consisting of sign, object, and interpretant, but rather of object, interpretant, and meaning (Parmentier 1985: 37). While it is not possible to examine this controversial claim in detail here, it leads us toward a possibility that Parmentier appears to ignore – that is, that the concept of form may be approached from the perspective of meaning rather than truth. Still, this move may amount to opening a veritable can of worms, as Peirce's use of 'meaning' in semeiotic settings is by no means free from ambiguities. Hence, caution and some distinctions are called for here.
The theme of meaning is obviously important for Peirce, as it is constantly present in his pragmaticistic discussions. Yet, his semiotic treatment of the subject seems to leave much to be desired; in particular, his mentions of 'meaning' or 'signification' in connection with his sign definitions are apt to cause bewilderment. It never becomes quite clear how meaning is related to the central semiotic concepts of sign, object, and interpretant; in various contexts Peirce suggests that it may be identified or associated with the object (e.g. MS 11 [1903]; SS 196 [1906]), one of the objects (e.g. EP 2:274 [1903]), the interpretant (e.g. EP 2:496 f. [1909]), one of the interpretants (e.g. PPM 232 [1903]; CP 4.536 [1906]; EP 2:499 [1909]), the sign relation taken as a whole (e.g. EP 2:429 [1907]), or something distinct from all of these (e.g. NEM 4:309; EP 2:305 [c. 1904]). Put simply, the problem of Peirce's conception of semiotic meaning concerns the connection between object and meaning on the {235} one hand, and between interpretant and meaning on the other. There does not seem to be a simple and straightforward answer to be found in his writings, but there are a number of statements and suggestions that can help us move forward in our examination. In particular, Peirce gives us good reasons to reject the association of object and meaning, as well as the unqualified identification of interpretant and meaning.
Taking the case of the object first, it may be noted that Peirce clearly distinguishes between the object of the sign and the meaning of the sign, defining the former as "the thing or occasion, however indefinite, to which [the sign] is to be applied" and the latter as "the idea which [the sign] attaches to that object" (CP 5.6 [1905]). This may be taken as a tolerably conclusive statement to the effect that the meaning of the sign cannot be identical with its object (although it is perhaps possible to specify circumstances in which such an identification would be reasonable). From a commonsensical communicative perspective this distinction is certainly plausible; it seems only natural to say that what is discussed is one thing, and what is said about that thing or topic something else. In traditional logical terms, it is simply the distinction between denotation and signification.
Of course, rejecting the object as a candidate for semiotic meaning strengthens the case for the interpretant; and this is one plausible result of the combination of Peirce's semeiotic and pragmaticistic writings. However, the matter is not quite that simple. Although the concepts of meaning and interpretant intersect at many points in semeiotic, it is possible to argue that the interpretant and the meaning are not equivalent; in fact, Peirce even warns us of such a 'confusion' (EP 2:305 [c. 1904]). In general, the interpretant can be characterised as the 'product' or the 'proper significate outcome' of the sign (CP 4.536 [1906]; CP 5.474 [1907]; EP 2:493 [1909]).[17] There may be interpretants that are not meanings in a full sense of the word, but rather emotions or reactions. Moreover, we also find Peirce speaking of meaning as something that is communicated in semiosis (NEM 4:309), or that is shared by all three correlates of the sign (MS 717; in Deledalle 1995: 17). This would suggest that, for certain purposes at least, meaning is best held distinct from the interpretant, as the interpretant is in some sense the receiver of meaning. The idea that meaning is something that is conveyed in semiosis would associate meaning with the communicated form, which is a determination of all of the correlates of the sign relation.
In order to make some progress in our analysis of semiotic meaning, it may be necessary to recognise that there is a certain vagueness about Peirce's usage of the concept. That is, the denotation of 'meaning' seems to partly depend on the matter being discussed; it is not a rigidly defined technical term of semeiotic (Short 1982: 308). This vagueness is by no means automatically incriminating; it is perhaps a quite accurate and unavoidable reflexion of the complexity of meaning. In a sense, Peirce acknowledges the sundry character of the concept as {236} he identifies several grades or aspects of meaning (e.g. W 3:258 [1878]; SS 159 [1903]; EP 2:496 f. [1909]), which are often connected to his analysis of the divisions of the interpretant.[18] The crucial question here, however, concerns the possible connection between form and meaning – or, to be more precise, how Peirce's pragmaticistic analysis of meaning can help us reach a somewhat more tangible notion of the communicated form.
As Parmentier observes, Peirce's communicative definition of semiosis is a feature of his later semeiotic; its appearance can be dated to the middle of the first decade of the 20th century. What Parmentier does not consider, however, is that this shift in perspective occurs during approximately the same period that Peirce establishes an explicit and illuminating connection between his pragmaticism and his extended theory of signs, which includes a growing interest in the interpretant and its various aspects. Now in order to grasp the significance of this turn we must first note how Peirce characterises the communicated form; it is not a singular thing, but possesses the being of a predicate (EP 2:544 [c. 1905]). It can in principle be formulated as a conditional proposition that states that certain things would happen under certain circumstances. In other words, it is a power, which can be understood as a kind of disposition or real potential (cf. EP 2:388 [1906]). Yet another Peircean term for the form so interpreted would be habit (cf. Colapietro 1997: 276); and here we find the vital connection to the pragmaticistic analysis of the meaning of signs. According to Peirce, the ultimate logical interpretant, which is the proper meaning of an 'intellectual sign', is a habit, or perhaps more specifically a habit-change (EP 2:412 [1907]; CP 5.476 [1907]).
Interpreting the communicated form in terms of habitualness moves our analysis toward a more concrete and practical level, as it establishes an association between the enigmatic form and the 'would-acts' and 'would-dos' of habitual behaviour – that is, the general habits of action of conscious beings or inanimate objects (CP 5.467 [1907]). Yet, it is reasonable to ask why, if that indeed is the case, Peirce still chooses to employ the term 'form' instead of 'habit' – a concept that he is certainly not afraid of using even in surprising contexts. This problem may be stated in terms of Peirce's categories; form, as it is often characterised by Peirce (see, e.g., NEM 4: 293 ff. [c. 1903]), is primarily a matter of Firstness, while habit may be associated with Thirdness.[19] The proposed interpretation would then appear to lead to an unfortunate clash on the categorial level. However, the difficulty is perhaps surmountable. Peirce's approach to the categories is relational, and permits us to apply categories to categories. Consequently, we may tentatively suggest that the communicated form can be characterised as a First of a Third.[20] Obviously, this is a rough hypothesis that is in desperate need of examination, but for now it is sufficient to note that it has the virtue of moderating the pragmaticistic conception of meaning as habit; it is in fact too crude to say that what is communicated in semiosis is habit, plain and simple. We can certainly understand others without adopting their habits, or the {237} habits embodied in the signs that they utter; but in order to comprehend the meaning it is necessary to grasp the forms of the habits involved, which in turn leads to the development of similar or related habits – that is, habits that possess the same form in some respect. In parenthesis, it may be added that this would also be applicable to our comprehending the modes of action of natural objects.
It should be noted that the close affiliation between form, habit, and meaning does not entail perfect equivalence. Moreover, meaning as the form of habit is not the whole truth about meaning in semeiotic. Hence, it is appropriate to note certain qualifications regarding the pragmaticistic analysis. To be precise, it concerns intellectual signs (concepts), and may be of limited applicability to other types of signs. Yet, habitual meaning cannot be communicated as such, without the aid of less sophisticated signs. In fact, Peirce claims that a proposition that has no connection to experience, actual or potential, is devoid of meaning (EP 2:1 [1893]). No communication is possible without some kind of indices that establish an experiential reference or contact (EP 2:7 [c. 1894]), and icons are needed to present objects and relations in their qualitative aspect. However, as such icons and indices assert nothing (EP 2:16 [1895]); it is only as they are involved in symbolic representation that they can be said to be able to communicate some meaning. Thus, these three kinds of signs cooperate in communication; without such a collaboration, there can be no communicative determination. True, in some modes of communication one of the mentioned sign-types dominates the proceedings, as in the case of the existential graphs, which are predominantly iconic. Their 'transparency' is due to the fact that they can present or represent the relevant qualities of the object directly (or at least relatively directly); but as the graphs are interpreted and understood, some indices and symbols are bound to become involved, and the graphs fall short of perfect transparency. Furthermore, their power of representation and communication is limited; they are applicable to diagrammatic reasoning with propositions and arguments (cf. SS 197 [1906]). The graphs have a certain capacity reminiscent of assertion, but do not appeal to the emotions, ask questions, shout out warnings, etc. In other words, the existential graphs cannot be used to represent the communicative process in full.
With regard to the criticism presented in section 2 of this paper, there are three important consequences of the proposed pragmaticistic reading of the communicated form. Firstly, it shows that there may be more to the communicative definition of semiosis than meets the eye. Without denying that it is one of Peirce's most cherished goals to construct an optimal system of logical expression, and that the communicative definition may even have been presented with this purpose in mind, it can be argued that this characterisation of sign action also points in a another direction – that is, toward the domain of habits and practices. Of course, it would be an exaggeration to say that we find a full-scale theory of praxis hidden in the depths of Peirce's communicative definitions; but there are good reasons to assume that his conception of communicative {238} semiosis is compatible with such pursuits. Therefore, the communicative definition of semiosis seems to be applicable to a far broader range of studies of communication than Parmentier's critical account would admit.
Secondly, as a result of the association of form with habit, the one-dimensional linearity of semiotic determination is broken, albeit in a somewhat different manner than in Parmentier's conception of metasemiotic representation. Parmentier's interpretation of semeiotic appears to largely rest on the idea of unlimited semiosis, based on Peirce's remarks about the continuous character of developed signs. As a result, Parmentier tends to view communicative determination as a straightforward transmission of truth. However, the pragmaticistic analysis suggests that there are meaningful ruptures in the process; signs cannot be endlessly translated into other signs, if they are to fulfill their function as signs – that is, to contribute to the evolution of concrete reasonableness by rendering inefficient relations efficient (SS 31 [1904]). Peirce denies that semiosis would be nothing but "an endless viaduct for the transmission of idea-potentiality" (EP 2:388 [1906]). Still, this does not mean that we have to abandon the synechistic drift of Peirce's semeiotic; it is possible to hold on to the thought that the process of semiosis is continuous on a larger scale, while maintaining that any specified significant process possesses beginnings and ends (cf. Johansen 1993b: 287).
Thirdly, the perceived connection between Peirce's communicative conception of semiosis and his pragmaticism opens the door for a number of further considerations. In particular, it points toward critical common-sensism, the idea of self-control, and other normative concerns. The conception of communication involved in the later semeiotic need not be strictly restricted to mathematical logic, if we acknowledge the adequacy of this more flexible perspective. Moreover, adopting a common-sensist position, we can admit as a fact that we are always in semiosis – that we are born into a universe full of meaning – without thereby reducing our semiotic role to nothing but that of a passive receiver of a ready-made world. Our habits can be made the objects of critical reflection, thus making possible a kind of hypostatic or metasemiotic level; we can review and reform our semiotic habits of communication, although only by making abstractions and temporal projections within the process itself.[21] Therefore, Peirce's conception of communicative semiosis is not necessarily a linear course of determination in the sense that Parmentier implies; instead, it can be construed as a complex process that involves the possibility of reflective understanding, self-control, and inwardness (cf. Colapietro 1989).

4. Common Signs and Communicative Teleology

One aspect of Parmentier's criticism that has not yet been dealt with satisfactorily is the actual role of the sign in communicative semiosis. The difficulty can be stated as follows: in the communicative conception, the sign seems to be something that stands between the object and the interpretant, but {239} not in the sense of genuine mediation; the sign is an entity that may temporarily carry a determination. In other words, the sign is reduced from a genuine mediator to a vehicle for form or meaning. Now, there is no denying that Peirce at times seems to describe the sign in this manner (e.g. NEM 4:309); however, such a conception of the role of the sign would clash with Peirce's insistence upon the triadic character of the sign relation. In other words, characterising the sign as a vehicle for the transmission of form appears to turn the sign into a mere receptacle, and would also imply that communication, in its most basic mode, is nothing but a combination of two dyadic processes of determination.
This problem may be put into perspective by examining one of Peirce's most curious examples of a sign process: a line of bricks standing on end, so arranged that if the first (or last) in the series is tipped over in the direction of the others, then this action will cause a chain-reaction and all of the bricks will fall over (see NEM 4:313 f. [c. 1906]). Now, to simplify matters we may assume that there are only three bricks in the line: brick A acts upon brick B, which in turn acts upon brick C. In a sense, brick B acts as a mediator, since it conveys the effect of A to C. In other words, B would be a sign, as it transmits the action (or the form of the action) of A to C; each brick is a sign to the succeeding bricks of the original effect (NEM 4:314 [c. 1906]). In our simplified case, this would seem to be easily reducible to two separate dyadic actions; brick A acts upon brick B and brick B acts upon brick C. The process is then one of efficient causation, in which mechanical energy or something similar is transferred from one point to another. Yet, Peirce chooses to characterise the dynamical reaction in the line as a sign relation – and what is even more damaging for the general argument of this paper, as a semiosis involving communication.
In order to get to grips with this dilemma, it may become necessary to propose some distinctions and modifications of Peirce's terminology. First, however, it is important to note what kind of sign is involved in the case of the bricks; it is not a symbol, but an index (ibid.). An index can be defined as a sign which is fit to act as such because of an existential relation between sign and object – that is, the kind of relation that can be found in the action of one brick upon another. As such, it does not function as a sign – for that an interpretant is required – but it possesses the grounding relation that nevertheless makes it an indexical sign apart from any possible relation to an interpretant. Now, the true difficulty of Peirce's brick example is not the status of the bricks as indices (or perhaps one should say as potential indices), but the fact that he seems to take the line to involve an interpretant as well, namely the third brick. We would then have a semiosis in the transmitted dynamical effect, independent from any actual or potential interpretation of it. That is, we would be faced with a conception of communicative semiosis, in which communication could consist of nothing but mechanical determination. What seems to be totally lacking from this picture is any kind of teleology or purposiveness – that which ought to distinguish semiosis from mere dynamical action (see, in particular, EP 2:411 [1907]; cf. CP 2.86 {240} [1902]; CP 1.532 [1903]). It may be suggested that the problem can be solved by taking the cause of the original effect into account; the first brick is perhaps tipped over in order to bring about a certain effect, and the intention to produce such a reaction would introduce an element of purposive mind or final causation into the proceedings. However, Peirce's formulation does not really support such a reading. Even if the first effect is purely accidental – someone may kick over the first brick by mistake – it would still appear to be communicated in the series of falling bricks, if the crucial condition for signhood is that of mediate determination.[22]
The communicative relation, which is involved in the example of the bricks, seems to consist of a compound of dyadic relations. However, this makes it less than triadic in Peirce's sense, as he emphasises the irreducibility of Thirdness and genuine triads (see, e.g., SS 28 ff. [1904]), and characterises mediation as a prime example of Thirdness (see section 2). Hence, the communicative process, as it is presented in the brick example, would appear to fall short of genuine mediation – which in turn would indicate certain defects in Peirce's communicative definitions of the sign relation, as mediation is supposedly of the character of a sign (CP 2.92 [1902]). Consequently, the process of communication described does not meet the criteria of mediation and true signhood.
The difficulty we have noted stems from interpreting the sign as a vehicle, thus reducing the sign relation into three separate components and two or more dyadic relations. As a matter of fact, we cannot grasp a sign relation that would be less than triadic; a sign does not function as such without an interpretant. Prescissive abstractions make it possible to speak of the relation between sign and object as such (as iconic, indexical, or symbolic) or the sign as it is as such (as qualisign, sinsign, or legisign); but the comprehended sign always involves three correlates. Now the function of the sign is not merely to transmit a determinate form from one subject to another; it also operates so as to bring these subjects into connection with each other. In a sense, its function is similar to the gift in the relation of giving, one of Peirce's favourite examples of a genuine triadic relation (see, e.g., CP 1.363 [c. 1890]; SS 29 f. [1904]). The object given is not a gift apart from the relation, but it is capable of putting the giver and receiver into a certain intentional relation to each other. This fact cannot be reduced to dyadic relations; the gift-object does not simply take on the intention in one act, and then transfer it in a completely distinct act. In other words, the giving is performed in view of three correlates. The gift is not a mere vehicle, but an integral part of the relationship; without the gift, there would be neither giver nor receiver in a proper sense. Similarly, the sign also functions as a unifying or synthesising power, in which case it is a Third (cf. CP 1.378 [c. 1890]; CP 6.32 [1891]; NEM 4:295 [c. 1903]). In abstract semeiotic terms, it brings object and interpretant together; but the sign is also a potential merger of communicating subjects, such as human beings in conversation (cf. EP 2:389 [1906]). In this {241} unification or 'welding of minds', the sign establishes a real social identity or community, which is not simply reducible to a collection of personal identities (cf. NEM 4:ix; EP 2:338 [1905]; CP 4.551 [1906]; Liszka 1996: 84).
Naturally, there are certain limitations to the analogy between the sign and the gift (although the gift is also a sign). Giving, as it is ordinarily understood, involves a giver parting with a gift. This is not so in communicative semiosis, since form or meaning does not cease to be in the source just because it is somehow communicated to a receiver; indeed, in one of his definitions (SS 196 [1906]), Peirce describes communication as extension. More importantly, however, we may note that the mode of causation of giving and that of semiosis is the same. It is, ideally at least, final causation – which does not mean that efficient causation would or could be wholly absent (cf. EP 2:121 [1902]). Neither giving nor genuine semiosis simply happens; both are directed and purposive processes. Yet signs must produce some physical effects; there is no other way of actually communicating ideas (EP 2:326 [1904]).
In view of Peirce's marked emphasis on triadicity and final causation, it seems unlikely that he could hold that any genuine case of semiosis would be a mechanical transfer of effects or energies merely. Consequently, it seems reasonable to conclude that his problematic example of the communicating bricks is misleading; in fact, it may even be irreconcilable with the general thrust of his semeiotic. In particular, the conception of communication that the illustration implies falls short of genuine mediation or semiosis, because it involves no true teleology. It can be argued that communication is precisely a sign process in which the purposive character is prominent, as its ideal goal is understanding – perhaps even mutual understanding. Such an argument seems to receive support from the conceptions of commind and communicational interpretant, which appear in connection with the communicative definition of the sign relation. The commind, which is defined as "that mind into which the minds of the utterer and interpreter have to be fused in order that any communication should take place" (SS 196 f. [1906]), implies that communicative semiosis requires a degree of common understanding or common ground, which may serve as a starting-point for the exchange. That is, the process does not take place in a semiotic vacuum, but against a background that involves previous experiences and established practices. At the very least, some common ground is necessary, if utterer and interpreter are to establish a reference to an object in a certain universe of discourse, which cannot be symbolically described in the signs themselves, but must be previously familiar through collateral experience (cf. CP 2.536 [1902]; CP 3.621 [1911]).[23] As a result of the communicative exchange, the minds of utterer and interpreter receive certain distinct determinations (which Peirce calls intentional interpretant and effectual interpretant, respectively). However, the process is directed toward the communicational interpretant, which can be construed as an ideal limit of communication, namely optimal understanding with regard to a specific subject-{242}matter, i.e. to the object of the semiosis in question. Ideally, the goal of the process would be to achieve an accurate determination of the commind as to the true meaning of the object. According to Peirce, such an ideal is best pursued in scientific inquiry, an inherently social mode of life.
It is important to emphasise the normative character of the above outline of the communicative process. In a sense, the communicational interpretant may be taken to be the discursive equivalent of the better-known Peircean idea of the final opinion, in which reality is established or revealed. Ideal communicative understanding, which would make further questioning pointless and obsolete, is never factually reached. In actual communicative interaction, there is always some latitude of interpretation, as no sign is completely determinate (CP 4.543 [1906]). Moreover, the experiential and habitual configurations of utterer and interpreter are a result of their lives, which always include interpretive aspects (CP 7.538; cf. CP 4.172 [1897]). Hence, no actual communication between human beings can be completely determinate in every respect (CP 5.506 [1905]; cf. Johansen 1985: 254). Now this would point in the direction of communicative skepticism; there is no guarantee that human beings in communicative interaction understand each other, but misunderstanding between the parties seems to be inevitable (cf. Johansen 1993b). How, then, can we speak of successful communication? Or, to put the problem in the terms of Peirce's communicative definitions of semiosis: is it not an unjustified idealisation to say that forms are extended as a result of the process, if communication does not ensue in shared cognitions and other common habits after all? A possible answer is: the definition does involve an idealisation, as it indicates that communication results in identical determinations at both ends of the process, but it can be at least partly justified by pointing out that that is the ideal tendency of our best communicative practices; we certainly strive to understand each other and the world. Moreover, it is possible to temper the strong normativity of the definitions, so that they are more readily applicable to everyday communicative situations. Falling short of the ideal of total understanding in every regard, we can talk about accomplished communication if the interaction establishes habits that would enable us to function socially or to cooperate in the pursuit of some goal within a common universe of action (cf. Johansen 1985: 261). In a certain sense, meaning is relative to purpose (cf. EP 2:393 [1906]; EP 2:420 [1907]). That is, we may speak of successful communication, although there is not a precise match of meanings in every respect. Indeed, here we may note a strength of the concept of form; we can say that two human beings possess or embody the same form of habitual meaning, without thereby claiming that they have adopted or developed exactly the same meanings. However, this does not require us to abandon the higher normative perspective; while perfect understanding is not absolutely achieved in actual life, it is still a hope that can motivate rational debate and social inquiry.[24]
Returning to the difficulty of the sign-vehicle, we observe that when the {243} question is thus approached from a different perspective, it becomes sufficiently clear that the medium of communication cannot be a mere vehicle.[25] A broader consideration of the communicative role of the sign indicates that its modus operandi is teleological and social. Yet, even if we take these additional factors into consideration, we are still faced with the problem of Peirce's rather unfortunate choice of words; in his example, he speaks of the falling of the bricks as involving 'communication' from the first effect to the subsequent bricks (NEM 4:314 [c. 1906]). Now, it is a well-known fact that Peirce strives for a comprehensive theory of semiosis, which would not only cover the social exchange of signs and meanings, but also the interpretation of natural signs and even semiotic phenomena in non-human nature. However, in the case of the bricks, he seems to be stretching his conception beyond plausibility. Of course, one can say that the relation between the bricks involves a potential indexical sign, which may receive or produce an interpretant in the future; but that is a rather different proposition from claiming that the middle brick acts as a sign for the effect to the last brick. Still more problematic is his use of the term 'communication'; identifying bricks falling on each other as a communicative process threatens to eliminate the teleological and social connotations of the concept. It is even somewhat awkward to say that natural signs communicate something to us as we interpret them, as the situation lacks the reciprocal development that is characteristic of ideal social communication (cf. Colapietro 1995: 36). Moreover, it would be odd to claim that we understand nature as involving an intention to communicate something to us (as we do when we understand mass-mediated communication, for example).[26]
In order to clarify our conceptual arsenal, it may be sensible to follow Liszka's (1996: 137 f.) recent suggestion, and distinguish between transmission and genuine communication. According to this proposal, we can say that a dyadic process is turned into transmission when it is interpreted; the triadic character of the semiosis is therefore attributable to the activity of the interpreter. Transmission, in turn, is transformed into communication when at least two subjects capable of triadic action are involved.[27] This distinction would possess the virtue of avoiding the occasionally counter-commonsensical demeanor of Peirce's remarks; however, three qualifications can be added. Firstly, it should be noted that the ground of the transmission – that is, the dyadic process – is not dependent on any actual interpretation. Secondly, it may be somewhat incautious to say that transmission is 'transformed' into communication, as if communication would be constituted by transmission plus an utterer. What needs to be emphasised is the irreducibility of genuinely triadic communication to transmission; the latter takes on a triadic character only as a result of interpretation, and is thus a degenerate sign process. Thirdly, a semiotic flow initiated by a transmission does not necessarily stop with the interpretation of the transmission; it may trigger a genuinely communicative process in our thought, in the shape of an internal dialogue in which the self takes on the roles of utterer {244} and interpreter.
According to the view presented here, communication is a central mode of sign action, but it is not all of semiosis. Hence, Peirce's communicative definitions of the sign relation may be less general than certain other characterisations. The definitions are valuable, however, as they serve to highlight the importance of the communicative and transmissive functions of signs. Genuine communication is an ideal standard, to which other kinds of semiosis may be compared; but communication is also a social fact. Introducing suitable modifications and complementations, so that the sign is not construed as a mere vehicle, it should be possible to employ Peirce's communicative sign definitions as starting-points for analyses of social communications. The path of semeiotic is not blocked; but it still remains to be seen if it will lead us to new and productive studies of communication.

5. Conclusion: Toward a Critical Rhetoric

Although Peirce's explicit comments on communication are infrequent and fragmentary, his semeiotic is still rich in implications and insights about communicative processes and practices. In this essay, I have only scratched the surface, as the purpose of my discussion has been to show that Peirce's later definitions of the sign in terms of communicative mediation do not necessarily involve a fatal reduction of the semeiotic project. By taking Parmentier's criticism as a starting-point for my investigation, I have argued that these often neglected aspects of Peirce's philosophical theory of signs are not isolated oddities or fatal blunders; they can be clarified by relating them to other facets of his thought, such as his pragmaticism. Moreover, I have tried to clear the way for coming endeavours by showing that it is possible to interpret Peirce's semeiotic in a manner that could be conducive to studies of various kinds of communicative phenomena. In particular, I have wanted to show that Peirce's normative-logical and teleological outlook does not restrict the scope of semiotic inquiry to iconic or mathematical logic; his definitions leave room for various lines of development, in spite of their often abstract rigour.
Throughout this paper, I have both explicitly and implicitly implied that the communicative considerations that I have discussed are semeiotically relevant matters. Yet, I have not placed them within Peirce's system of semeiotic inquiry. Instead, I have rather liberally moved between the abstract level of sign definition and the more concrete domain of pragmaticism, occasionally even dipping into areas that may fall completely outside of a strictly defined realm of semeiotic. My justification for such a strategy is that it enables us to get a firmer grasp on Peirce's sparse comments; however, some concluding comments on the role of the study of communication in semeiotic are needed.
In the introduction, I suggested that the most natural position for philosophical studies of communication in Peirce's system would be within rhetoric, the third and least investigated branch of semeiotic. Yet, there are {245} certain reasons for doubting this placement. Semeiotic is supposedly a strictly philosophical mode of inquiry, while communication research seems to entail more empirical investigations of social facts. Thus, it would involve psychological and sociological considerations – the kind of elements that Peirce banned from his semiotic logic. Indeed, Peirce specifically states that rhetoric is not a matter of psychology (CP 4.116 [1893]). However, keeping in mind Peirce's conception of philosophy as a heuristic science based on common experience (see, e.g., CP 1.241 [1902]; CP 1.577 [1902-3]; PPM 207 f. [1903]), there would still seem to be room for a philosophical study of communication, as long as we accept that there are communicative phenomena among our persistent everyday experiences; and in my opinion there certainly are. In fact, it may be one of semeiotic's most important functions to analyse such communicative and social aspects of our seemingly private experiences – traits which are so omnipresent that they would otherwise escape our attention.[28]
If we acknowledge the possibility of a philosophical study of communication and wish to retain Peirce's division of semeiotic, then rhetoric is the most natural home for such an inquiry (cf. Liszka 1996: 88 ff.). Still, some qualifications are needed, as not all rhetorical questions are necessarily pursued in philosophy. According to Peirce, there is, as a matter of fact, a universal art of rhetoric, which is concerned with "the general secret of rendering signs effective" (EP 2:326 [1904]). From this art, which ought to include such practical concerns as the teaching of eloquence and the improvement of organisational communications, one may abstract a science of rhetoric, which should investigate the principles of everything that the art covers or could cover. This theoretical science would examine "the essential conditions under which a sign may determine an interpretant sign of itself and of whatever it signifies, or may, as a sign, bring about a physical result" (ibid.). This characterisation is certainly broad enough to incorporate studies of the theoretical and conceptual basics of communication. However, it is not certain that Peircean rhetoric is concerned with all modes of communication; there are certain strong arguments in favour of limiting its scope to that of scientific communications and methodology. In fact, that appears to be the direction in which Peirce is heading.
According to Lucia Santaella-Braga (1999: 380), Peirce's rhetoric or methodeutic develops from a narrow to a broad sense. In the narrow phase, it is presented as rhetoric, which deals with the reception, life, and futurity of signs. In contrast, the broad conception is identical with methodeutic – that is, the general theory of the methods of science (ibid.: 389). Now in view of the actual evolution of Peirce's third branch of semeiotic, Santaella-Braga certainly has a point. Late in his career, Peirce explicitly states that the name 'methodeutic' is to be preferred over 'rhetoric' (CP 4.9 [1906]). Yet, the reasons for calling the earlier perspective 'narrow' and the later 'broad' are far from obvious; if anything, it ought to be the other way around (cf. Liszka 1996: 79). The study of rhetoric should, in view of Peirce's admittedly vague characterisations, include {246} methodological concerns, but also much more; while methodeutic would imply a more precisely limited area of study. In this sense, at least, methodeutic is narrower than rhetoric; but at the same time its scope and functions are better defined than those of rhetoric.
Now there is really no doubt that Peirce considered methodology or methodeutic to be the most central aspect of rhetoric. Also, it is quite plausible to claim that his preference for 'methodeutic' signals a more scientific approach to what he previously called 'rhetoric'. However, without denying the preeminence of methodeutic, we may still ask ourselves whether it is sensible to simply limit philosophical rhetoric to methodeutic. As methodeutic, rhetoric would only be concerned with discovering and developing such methods of inquiry and communication that would be conducive to finding out general truths. In other words, methodeutic should focus on the needs of heuristic inquiry, and ignore other forms of communicative phenomena. Yet, in one of his most suggestive sketches of rhetoric (the very text, in fact, which according to Santaella-Braga [1999: 391] signals the shift from rhetoric to methodeutic) Peirce recognises a rhetoric of fine art and a rhetoric of practical persuasion in addition to the rhetoric of science (EP 2:329 [1904]). Furthermore, his conception of rhetoric includes examinations of the medium of communication used (including speech and languages) and of the signs into which the interpretation is to take place. At the very least, this proposal extends the range of rhetoric beyond that of methodeutic in a narrow sense, which nevertheless possesses a key position as the rhetoric of science, arguably the most important specialisation of the branch.
While I do not wish to denounce the value of specifying the domain of rhetoric in various ways, I think it would be wise to follow Peirce in keeping the study more flexible than the other branches of semeiotic. In fact, Peirce even indicates that it is acceptable to take some psychological facts into consideration in rhetoric (CP 2.107 [1902]); and we could perhaps broaden its scope further by allowing a limited number of sociological insights to enter the proceedings. Of course, such an augmentation must be pursued with great care, if we want to stay true to Peirce's spirit; semeiotic rhetoric is, after all, a theoretical mode of inquiry, and not a tool for social engineering or coercion – although it can be so used. Indeed, at this point one advantage of placing methodeutic at the heart of rhetoric becomes evident; it highlights the distinctively engaged character of rhetoric. Since scientific inquiry is directed toward an ideal social goal of truth and reasonableness, the rhetoric of science cannot be absolutely value-neutral. Its objects of study are inherently purposive – as are those of the rhetoric of art and the rhetoric of persuasion. Moreover, Peirce claims that methodeutic – which is concerned with the methods that ought to be pursued in investigation, exposition, and application of truth (EP 2:260 [1903]; cf. CP 4.240 [1902]) – should be based on a more general doctrine of the nature of teleological action (CP 2.108 [1902]). Although Peirce is here probably referring to his general logic of {247} evolution, this statement suggests a sense in which the study of communicative action can contribute to methodeutic; for certainly, it is relevant for a sophisticated methodology to be aware of various kinds of teleological or quasi-teleological communication, of destructive as well as productive forms, so that it may be possible to identify the types of communication best suited for the ideal purposes. Furthermore, it would be wise to take human limitations and frailties into consideration. After all, inquiry is a social mode of conduct.
In sum, the philosophical study of communication ought to be an integral part of rhetoric. On the one hand, it is of relevance for methodeutic, understood as the rhetoric of science; but on the other hand, it may also entail a broadening of the scope of semeiotic to involve other modes of rhetorical behaviour. Peirce undeniably characterises his theory of signs as a scientific undertaking, but that does not mean that semeiotic would study nothing but science; Peirce himself tends to view practically everything that can in any sense be investigated in semiotic terms (see SS 85 f. [1908]). Hence, I would like to advocate a broad conception of rhetoric, in which general studies of communicative phenomena can be pursued.[29] This would not only serve to assemble Peirce's fragmentary reflections on communication, but might also reveal untapped critical potential in his semeiotic. Namely, his normative conception of communication, suitably complemented, could find new applications in critical examinations of our communicative practices (cf. Colapietro 1995). Of course, that would mean moving beyond semeiotic rhetoric into the arena of social criticism; but rhetoric might provide the theoretical basis for such an undertaking. In particular, semeiotic analysis could serve to show that certain communicative habits that we take for granted – that have become transparent for us – are in fact complex modes of sign action that can be criticised and perhaps even intelligently reformed.[30] However, such amelioration ought to be carefully considered and gradual; there is simply no feasible way of performing a total revolution of our established practices. Neither should we take communicative transparency to be an evil that must be fought to the bitter end; if our modes of conduct would not be highly habitualised, they would require too much effort to be able to function properly. Our communicative habits, no matter how they come into being, display a tendency toward transparency. Hence, it is imperative that we recognise the limited but real possibility of self-control in view of such social ideals as reasonableness, truth, and communal understanding. Toward this end, semeiotic may serve us in our lives.



Apel, Karl-Otto.
1980. Towards a Transformation of Philosophy. London, Boston & Henley: Routledge & Kegan Paul.
1998. From a Transcendental-Semiotic Point of View. Manchester & New York: Manchester University Press.
Cherry, Colin.
1957. On Human Communication: A Review, a Survey, and a Criticism. Cambridge, Mass.: The M.I.T. Press.
Colapietro, Vincent.
1989. Peirce's Approach to the Self: A Semiotic Perspective on Human Subjectivity. Albany: State University of New York Press.
1995. "Immediacy, Opposition, and Mediation: Peirce on Irreducible Aspects of the Communicative Process." Recovering Pragmatism's Voice: The Classical Tradition, Rorty, and the Philosophy of Communication. Ed. Langsdorf, Lenore & Smith, Andrew R. Albany: State University of New York Press.
1997. "The Dynamical Object and the Deliberative Subject." The Rule of Reason: The Philosophy of Charles Sanders Peirce. Ed. Brunning, Jacqueline & Forster, Paul. Toronto, Buffalo & London: University of Toronto Press.
Deledalle, Gérard.
1995. "Introduction to Peirce's Semiotic." Semiosis 79/80.
Eco, Umberto.
1977. A Theory of Semiotics. London & Basingstoke: The Macmillan Press.
Fisch, Max H.
1986. Peirce, Semeiotic, and Pragmatism. Bloomington: Indiana University Press.
Fitzgerald, John J.
1966. Peirce's Theory of Signs as Foundation for Pragmatism. The Hague & Paris: Mouton.
Habermas, Jürgen.
1995. "Peirce and Communication." Peirce and Contemporary Thought. Ed. Ketner, Kenneth L. New York: Fordham University Press.
Hausman, Carl R.
1997. "Charles Peirce and the Origin of Interpretation." The Rule of Reason: The Philosophy of Charles Sanders Peirce. Ed. Brunning, Jacqueline & Forster, Paul. Toronto, Buffalo & London: University of Toronto Press.
Hookway, Christopher.
1995. "Metaphysics, Science, and Self-Control: A Response to Apel." Peirce and Contemporary Thought. Ed. Ketner, Kenneth L. New York: Fordham University Press.
Hulswit, Menno.
1998. "A Guess at the Riddle of Semeiotic Causation." Transactions of the Charles S. Peirce Society 34 (3).
Jensen, Klaus B.
1995. The Social Semiotics of Mass Communication. London, Thousand Oaks & New Delhi: Sage Publications.
Johansen, Jørgen D.
1985. "Prolegomena to a Semiotic Theory of Text Interpretation." Semiotica 57 (3/4).
1993a. Dialogic Semiosis: An Essay on Signs and Meaning. Bloomington & Indianapolis: Indiana University Press.
1993b. "Let Sleeping Signs Lie: On Signs, Objects, and Communication." Semiotica 97 (3/4).
Keeler, Mary A.
1990. "Investigating Transparency in the Conditions of Mediation from a Semeiotic View." Semiotica 82 (1/2).
1998. "Iconic Indeterminacy and Human Creativity in C. S. Peirce's Manuscripts." The Iconic Page in Manuscript, Print, and Digital Culture. Ed. Bornstein, George & Tinkle, Theresa. Ann Arbor: The University of Michigan Press.
Lalor, Brendan.
1997. "The Classification of Peirce's Interpretants." Semiotica 114 (1/2).
Liszka, James J.
1996. A General Introduction to the Semeiotic of Charles Sanders Peirce. Bloomington & Indianapolis: Indiana University Press.
Oehler, Klaus.
1995. "A Response to Habermas." Peirce and Contemporary Thought. Ed. Ketner, Kenneth L. New York: Fordham University Press.
Pape, Helmut.
1996. "Object and Final Cause in Peirce's Semeiotic." Peirce's Doctrine of Signs: Theory, Applications, and Connections. Ed. Colapietro, Vincent M. & Olshewsky, Thomas M. Berlin & New York: Mouton de Gruyter.
Parmentier, Richard J.
1985. "Signs' Place in Medias Res: Peirce's Concept of Semiotic Mediation." Semiotic Mediation: Sociocultural and Psychological Perspectives. Ed. Mertz, Elizabeth & Parmentier, Richard J. Orlando, San Diego, New York, London, Toronto, Montreal, Sydney & Tokyo: Academic Press.
1994. Signs in Society: Studies in Semiotic Anthropology. Bloomington & Indianapolis: Indiana University Press.
Percy, Walker.
1976. The Message in the Bottle: How Queer Man Is, How Queer Language Is, and What One Has to Do with the Other. New York: The Noonday Press.
Pihlström, Sami.
1998. "Peircean Scholastic Realism and Transcendental Arguments." Transactions of the Charles S. Peirce Society 34 (2).
Ransdell, Joseph.
1977. "Some Leading Ideas of Peirce's Semiotic." Semiotica 19 (3/4).
1981. "Semiotic Causation: A Partial Explication." Proceedings of the C. S. Peirce Bicentennial International Congress. Ed. Ketner, Kenneth L. et al. Lubbock: Texas Tech Press.
Rescher, Nicholas.
1998. Communicative Pragmatism and Other Philosophical Essays on Language. Lanham, Boulder, New York & Oxford: Rowman & Littlefield Publishers.
Royce, Josiah.
1968. The Problem of Christianity [first published in 1913]. Chicago & London: The University of Chicago Press.
1998. Metaphysics [lectures delivered 1915-6]. Albany: State University of New York Press.
Santaella-Braga, Lucia.
1994. "Peirce's Broad Concept of Mind." S: European Journal for Semiotic Studies 6 (3/4).
1999. "Methodeutics, the Liveliest Branch of Semiotics." Semiotica 124 (3/4).
Seager, William E.
1988. "Peirce's Teleological Signs." Semiotica 69 (3/4).
Shannon, Claude E. & Weaver, Warren.
1949. The Mathematical Theory of Communication. Urbana, Chicago & London: The University of Illinois Press.
Short, T. L.
1981. "Semeiosis and Intentionality." Transactions of the Charles S. Peirce Society 17 (3).
1982. "Life Among the Legisigns." Transactions of the Charles S. Peirce Society 18 (4).
1986. "What They Said in Amsterdam: Peirce's Semiotic Today." Semiotica 60 (1/2).
1996. "Interpreting Peirce's Interpretant: A Response to Lalor, Liszka, and Meyers." Transactions of the Charles S. Peirce Society 23 (4).
1998. "What's the Use?" Semiotica 122 (1/2).
Tejera, Victorino.
1996. "Has Habermas Understood Peirce?" Transactions of the Charles S. Peirce Society 32 (1).


[1] In this paper, I refer to Peirce's theory of signs as 'semeiotic', by which I understand an approach within the field of 'semiotics'. However, I wish to avoid the appearance of carving the subject-matter of sign-theoretical study into Peircean and non-Peircean parts. Hence, when I talk about significative phenomena that can be studied and interpreted from various semiotic perspectives, I will use 'semiotic' and 'semiosis' instead of 'semeiotic' and 'semeiosis'. In other words, I reserve the term 'semeiotic' for Peirce's theory, and drop the 'e' when discussing the facts and processes that this theory studies.

[2] In accordance with the customs of Peirce scholarship, I will refer to Peirce's texts using abbreviations. CP x.y refers to The Collected Papers of Charles Sanders Peirce; v indicates volume number, p paragraph number. EP v:p refers to The Essential Peirce: Selected Philosophical Writings; v indicates volume number, p page number. MS m refers to an original manuscript; m indicates manuscript number. NEM v:p refers to The New Elements of Mathematics by Charles S. Peirce; v indicates volume number, p page number. PPM p refers to Pragmatism as a Principle and Method of Right Thinking: The 1903 Harvard Lectures on Pragmatism; p indicates page number. SS p refers to Semiotics and Significs: The Correspondence between Charles S. Peirce and Victoria Lady Welby; p {251} indicates page number. W v:p refers to Writings of Charles S. Peirce: A Chronological Edition; v indicates volume number, p page number. Where known, the year of writing or publication will be given.

[3] Although Peirce does not offer even an outline of a philosophical theory of communication, his ideas have served as a source of inspiration for important investigations. Of these, Jørgen Dines Johansen's (1985; 1993a) studies of human semiosis synthesise Peirce's scattered remarks in an elaborate one-to-one communication model. A similar approach, but with more emphasis placed on the communal character of semiosis, can be found in James Jakób Liszka's (1996: 78 ff.) account of universal rhetoric. From a different angle, Vincent Colapietro (1995) makes some illuminating observations on the dual grounding of a possible Peircean theory of communication in the doctrine of categories and critical common-sensism. More liberal appropriations are presented by Walker Percy (1976) and Nicholas Rescher (1998), who in varying ways draw on Peircean ideas in their accounts of linguistic communication. Karl-Otto Apel's (1980; 1998) transcendental-semiotic arguments for a communicative and ethical community are to a large extent inspired by Peirce and Josiah Royce, arguably the first philosopher to have interpreted Peirce's thoughts in a communicative manner (see Royce 1968; 1998). Worthy of note are also the few appearances (Cherry 1957; Jensen 1995) that Peircean ideas have made in media and communication studies.

[4] That is, to physics and psychics, as distinguished from mathematics and philosophy, the other heuristic sciences in Peirce's classification (see, e.g., EP 2:258 ff. [1903]). If we wish to follow Peirce's model, actual applications of semeiotic, such as endeavours to render organisational communications efficient, are probably best placed within the practical sciences.

[5] Habermas draws our attention to important problems that certainly deserve consideration; but as involving assessments of Peirce's general philosophical outlook during different periods, his criticism tends to fall outside of the scope of this essay. Furthermore, Habermas has received some able replies (Oehler 1995; Tejera 1996), while Parmentier's claims have (as far as I know) neither been examined nor challenged.

[6] I cannot claim any major originality for these observations, however, as several penetrating accounts of previously ignored communicative aspects Peirce's philosophy have been presented in recent years. See, in particular, Johansen 1985; 1993a; Colapietro 1995; Liszka 1996.

[7] For important reflections on the teleology of signs see, e.g., Ransdell, 1977; 1981; Short 1981; 1982; Seager 1988; Santaella-Braga 1994; Liszka 1996; Pape 1996; Hulswit 1998; Short 1998.

[8] In this case, 'Peirce's later semeiotic' refers approximately to the period 1902-12 (Parmentier 1985: 42). This rough temporal division is obviously a simplification, as anyone familiar with the complexity and sheer volume of Peirce's definitions of the sign will immediately recognise. Peirce certainly does not cease to speak of 'representation' in his final years. Yet one may discern a striving in Peirce's thought for ever more general characterisations of the sign relation, which according to Parmentier culminates in the communicative definitions.

[9] In a similar fashion, Max H. Fisch (1986: 329) also recognises two principal modes of semiosis, which he identifies as sign-action and interpretation.

[10] Parmentier tends to speak of 'representamen' when referring to the first correlate of the sign relation, and claims that Peirce often confuses the 'marked' (sign as representamen or vehicle) and 'unmarked' (sign as complete sign relation) senses of the {252} term (Parmentier 1985: 47). However, this does not appear to be Peirce's intended usage; he defines the sign as a representamen with a mental interpretant (EP 2:273 [1903]; cf. EP 2:291 [1903]; CP 1.540 [1903]). Moreover, Peirce abandons this distinction toward the end of his career (SS 193 [1905]), and I will not employ it in this paper. Instead, I shall indicate some reasons for retaining the ambiguity of the term (see section 4).

[11] For illuminating presentations of the icon-index-symbol trichotomy, see e.g. EP 2:13 ff. (1895); CP 2.304 (1902); PPM 170 f. (1903); EP 2:291 f. (1903); EP 2:306 ff. [c. 1904]; SS 33 (1904); NEM 3:887 (1908); EP 2:460 f. (1909).

[12] The term 'unlimited semiosis' is not found as such in Peirce's writings. It was introduced by Umberto Eco, who conceives of it as the thesis that "semiosis explains itself by itself" (1977: 71).

[13] 'Ground' is another problematic concept in semeiotic. It is an early conception, the relevance of which has been the object of much discussion. In my opinion, T. L. Short's (1986: 107 f.) arguments to the effect that the function of the ground is later taken over by the icon-index-symbol trichotomy are plausible. Parmentier's use of the concept is different, as it involves the implication that only symbols would possess or produce grounds.

[14] "[T]o try to peel off signs & get down to the real thing is like trying to peel an onion and get down to onion itself, the onion per se, the onion an sich" (from a letter to Russell [1905]; in Hausman 1997: 188; cf. EP 2:392 [1906]).

[15] This is also one of the central themes of the well-known mathematical model of communication (Shannon & Weaver 1949), in which the problem concerns the trustworthy and efficient conveyance of information in a noisy environment. Of course, Peirce's semeiotic operates on a different level than the theory associated with Claude Shannon and Warren Weaver, which is primarily concerned with the accurate transfer of information from a concrete source to a concrete destination (via mechanical or organic senders and receivers), and not with the philosophical questions of mediation. Yet, one may detect an uncanny similarity between their linear communication model and Peirce's conception of communicative determination, not least since both parties seem to conceive of 'form' or 'information' as the objective basis of communicated meaning. In fact, it has been suggested (e.g. Ransdell 1977) that there is a certain affinity between Peirce's semeiotic and modern cybernetics, which to a large extent builds upon the theory of information.

[16] Peirce talks about 'form' in a number of different contexts. As his definition in Baldwin's Dictionary of Philosophy and Psychology shows, he recognises many different uses of 'form' (see CP 6.360 ff. [1902]); unfortunately, he does not always specify in what sense he employs the concept in his own writings. I choose to pursue only the clues given in the communicative definitions here, as a more comprehensive examination of Peircean form is too vast a task for this paper.

[17] Other characterisations of the interpretant include 'more developed sign' (CP 2.228 [c. 1897]), 'mental effect' (CP 1.564 [c. 1899]), 'proper significate mental effect' (EP 2:429 [1907]), 'effect on the interpreter' (EP 2:500 [1909]), and 'signification' or 'interpretation' (EP 2:496 [1909]).

[18] Neither the divisions of the interpretant nor the grades of meaning can be discussed comprehensively and in detail in this paper. The relationship between Peirce's two principal trichotomies of the interpretant – that is, the immediate-dynamic-final and the emotional-energetic-logical division – is one of the most intricate problems of his semeiotic, a question that has been discussed intensively in Peirce scholarship (for various {253} positions see, e.g., Fitzgerald 1966; Short 1981; Johansen 1993a; Short 1996; Liszka 1996; Lalor 1997; Santaella-Braga 1999). Peirce's discussion of the three grades of meaning springs from his early writings on pragmatism, but it receives a slightly different appearance in his later semeiotic, where it is also associated with the division of the interpretant. Without arguing my case here, I would suggest that we can distinguish at least five different aspects of meaning in these rather sparse comments: (1) meaning as recognition or immediate clarity (cf. W 3:258 [1878]; SS 159 [1903]; EP 2:256 [1903]; EP 2:496 [1909]); (2) meaning as definition or distinction (cf. W 3:258; SS 159); (3) meaning as intention (cf. EP 2:256); (4) pragmatic or ultimate logical meaning (cf. SS 159); and (5) final meaning (cf. EP 2:256).

[19] To be more precise, it is habit-taking that is a prime example of Thirdness (see, e.g., CP 1.409 [c. 1890]). As an established fact, a habit can be said to belong to the second category; however, as a living, law-like, and future-oriented conception, habit is Third.

[20] Peirce actually speaks of the Firstness of Thirdness, "the peculiar flavor or color of mediation", and characterises it as 'mentality' (CP 1.533 [1903]). The 'First of a Third' introduced here could be said to operate on a lower level of generality, as it refers to a certain type of Firstness (form) that is typical for a certain manifestation of Thirdness (habitual conduct) (but see Liszka 1996:91 for an interpretation of the communicated form in terms of 'ground').

[21] This accords with Colapietro's (1995: 32) claim that semeiotic, which is abstracted from everyday experience (see CP 2.227 [c. 1897]), involves a critical examination of our common communicative practices.

[22] Parmentier (1985: 42) notes a somewhat similar example in one of Peirce's unpublished manuscripts (MS 12 [1912]), where the air through which a stone is thrown acts as a medium for the communication of a form (friction, in this case). See also EP 2:392 (1906).

[23] It may be objected that Peirce's use of utterer and interpreter is merely another 'sop to Cerberus' (see SS 80 f. [1908]), an attempt to make his abstract conceptions more tangible. According to Peirce, the general definition of the sign can be given without utterers and interpreters, as their essential functions are represented by the object and the interpretant (see EP 2:404 ff. [1907]). However, Peirce also acknowledges that the utterer and interpreter are characteristic of the sign, and often mentions the need for quasi-utterers and quasi-interpreters, 'theatres of consciousness' in which the drama of semiosis is acted out (see CP 4.551 [1906]; cf. EP 2:544 [c. 1905]; EP 2:403 [1907]). Hence, we may speak of the utterer and interpreter as semiotic roles, which are not absolutely necessary for signhood as such, but which nevertheless are in some sense present in almost all forms of semiosis (cf. Johansen 1985: 252 f.).

[24] In his early writings, Peirce tends to speak of the final opinion as a fated fact, but he subtly changes that to a hope for ultimate agreement and truth in later discussions (see, e.g., NEM 4:xiii [1913]). Whether these perspectives involve a transcendental grounding in terms of the necessary presuppositions for communicative understanding, as Apel (1980; 1998) has argued, is a difficult question (see also Liszka 1996: 88 ff.). In his semeiotic, Peirce often reasons in a manner strongly reminiscent of transcendental argumentation, but also openly denounces that type of justification (see Hookway 1995). From the point of view of critical common-sensism, an absolute grounding would seem to be out of the question, since philosophical conceptions are based on fallible common-sense practices and certainties (cf. CP 2.227 [c. 1897]). A {254} compromise may be possible; we could interpret certain 'quasi-necessary' semeiotic insights as 'quasi-transcendental' arguments, or perhaps speak of 'naturalised transcendental arguments', as Sami Pihlström (1998) has suggested.

[25] See also EP 2:391 (1906), where Peirce makes an explicit distinction between vehicle and medium.

[26] Since Peirce holds that nature reasons (see, e.g., NEM 4:344 [1898]), he would perhaps not balk from saying that nature is an utterer (cf. Ransdell 1977: 173; but see SS 111 [1909]). This idea can be rendered more intelligible by saying that nature performs the role of an utterer in our interaction with it. Also, it is plausible to acknowledge the presence of teleological semiosis in the animal kingdom, possibly even in the whole biological domain (see Short 1982: 296 ff.; 1998: 44 ff.). However, the claim that natural quasi-utterances (not including certain signs uttered by more advanced animals) are meant to be understood – directed to us, in a sense – seems to require the postulation of conscious cosmic purposes or a belief in a personal God. I think that we can develop semeiotic without involving such conceptions.

[27] Another, perhaps more Peircean way of expressing this distinction would be to associate communication with the purposive replication of legisigns, which is typical for living beings (cf. Short 1982). Further, we may distinguish peculiarly human communication from other modes of communication by its complex operations of control and reciprocity (cf. Johansen 1985: 252 ff.).

[28] Peirce even suggests that philosophy should begin with an investigation of human beings and their ordinary conversation (CP 8.112 [c. 1900]).

[29] The suggested perspective is similar to Liszka's (1996: 10 f.) conception, according to which rhetoric studies the conditions under which signs can be developed, understood, and mediated in a community – it being understood that 'community' should not be restricted to strictly scientific communities (cf. Short 1986: 113). In addition to actual social formations, the concept should be broad enough to include various ideal communities and dialogic thought in general (cf. Royce 1968).

[30] This possibility was suggested to me by Mary Keeler's (1990; 1998) studies.

| Top |