Article body

Between the years 1903 and 1910, Charles S. Peirce spent considerable energy trying to expand the tenfold classification of signs he published in the Syllabus for his 1903 Lowell Lectures (see CP 2.233-72; EP 2.289-99).[1] Albert Atkin characterizes the semiotic theory emerging from these late explorations as “speculative, rambling, and incomplete” (2010 : para. 2); and T. L. Short comments that “Peirce’s later taxonomy ... is sketchy, tentative, and, as best as I can make out, incoherent” (2007 : 259-60). Contrary to these assessments by leading scholars of Peirce’s semiotic theory, Tony Jappy purports in this book to show that one iteration of this late system, mentioned only in a single 1908 letter to Lady Welby, is a “coherent” (3), “fully functional, organically organized and autonomous system” (177). Jappy fails to defend either the coherence or the importance of the 28-sign classification, and that which is of genuine scholarly interest in this book is obscured by Jappy’s pervasive misinterpretations of Peirce.

1. Summary of Peirce’s Twenty-Eight Classes of Signs

Before developing these criticisms, let me begin by summarizing the content of Jappy’s book, which consists of five chapters. Chapter 1, “The Philosophy of Representation”, provides a mixture of historical background on semiotic theory, focusing primarily upon John Locke’s semiotic theory, and background on important elements of Peirce’s philosophy like his architectonic classification of inquiries, his phenomenology, and his categories. Jappy’s discussion of these aspects of Peirce’s philosophy is brief, uneven, and it fails to illuminate the real importance of these topics for understanding Peirce’s semiotic theory. The chapter concludes by reviewing the 10-sign classification published in Peirce’s 1903 Lowell Syllabus.

Chapter 2, “The Transition” traces the development of Peirce’s semiotic theory between the 1903 Syllabus and the 28-sign classification developed in 1908. In the course of this chapter, Jappy reproduces, in clear and carefully constructed tables, four of Peirce’s late semiotic classifications. Three of these four (Table 2.2. is missing) and five additional typologies from this period are included in an appendix (180-188). The collection in one place of these diverse typologies, which are scattered throughout Peirce’s unpublished manuscripts and letters, is the most important contribution of Jappy’s book. Scholars researching the development of Peirce’s semiotic theory during this transition period will find Jappy’s typological tables quite helpful. With that said, Jappy’s explication of these transitional typologies is not very instructive. He is usually content to trace terminological transitions, regarding every new term for an old concept as an important “theoretical advance” (65). When he does offer interpretations of the transitions he is documenting, Jappy errs almost unfailingly, reaching such implausible conclusions as : Peirce abandoned his triadic conception of semiosis for a hexadic conception[2] (51-2); Peirce’s 1906 definition of a sign as a medium for the communication of a form from an object to an interpretant is “radically different” than prior definitions (55); by 1906 speculative rhetoric became “redundant” because signs are determined exclusively by their dynamic objects, with their utterers contributing nothing (60-1); and intellectual concepts lack objects (71-2).

The third chapter, “The Sign-Systems of 1908” introduces the 28-sign classification. This typology proposes six triadic divisions of signs, with each division being defined according to Peirce’s universal categories of firstness, secondness, and thirdness. Concerning the category that the sign’s dynamic object belongs to, a sign is either an abstractive, concretive, or collective, respectively; concerning the category of the sign’s immediate object, the divisions are descriptive, designative, and copulate; concerning the category of the sign itself, Peirce’s terms are mark, token, and type; concerning the category of the sign’s immediate interpretant – hypothetical, categorical, or relative; concerning the sign’s dynamic interpretant – sympathetic, percussive, or usual; and concerning the sign’s final interpretant – gratific, action-producing, and self-control-producing (see Table 3.2 : 86). How, one might be wondering, do these eighteen terms generate twenty-eight classes of signs? The first thing to note is that every sign is supposed to be classifiable with one term in each of these six divisions. Thus, emphasizing thirdness across the board, a sign might be a collective, copulate, type, relative, usual, and self-control-producing sign. Because 36 = 729, we might expect this division to generate 729 classes of signs. However, Peirce insists that “It is evident that a possible can determine nothing but a possible, it is equally so that a Necessitant can be determined by nothing but a Necessitant” (SS 84). Jappy refers to this as Peirce’s “universe hierarchy principle” (85). Combining this principle with Peirce’s claim that “the Dynamoid Object determines the Immediate Object, Which determines the Sign itself, which determines the Destinate Interpretant, Which determines the Effective Interpretant, which determines the Explicit Interpretant” it follows that “the six trichotomies, instead of determining 729 classes of sign ... only yield twenty-eight classes” (SS 84-5). So, for example, on this scheme it is impossible to have a sign whose dynamic object is characterized by firstness – i.e an abstractive sign – but whose immediate object is characterized by either secondness or thirdness – i.e a designative or copulant sign. This universe hierarchy principle drastically reduces the number of possible signs from 729 to 28 (see Irwin C. Lieb’s concise explanation in SS 160-6). Having introduced this classification, Jappy’s final two chapters apply the classification to actual signs, mostly paintings and photographs.

“Rhetorical Concerns”, chapter four, begins with a discussion of Peirce’s three “hypoicons” – images, diagrams, metaphors – and includes three diagrams that helpfully illustrate the difference between these three types of iconic signs (110-117). The rest of the chapter is dedicated to applying the 28-sign classification to a variety of different signs (a few lines from a poem, one painting, and five photographs), attempting along the way to show that this classification differs importantly from the 10-sign classification presented in the 1903 Lowell Syllabus. The main difference Jappy highlights is that, whereas the 1903 system is only indirectly concerned with the object, marking only how the sign is capable of representing its object – either through resemblance (icon), physical connection (index), or convention (symbol) – the 1908 system attends to the object directly, making a distinction between signs based upon what category or universe their dynamic and immediate objects belong to. As argued below, I do not think Jappy satisfyingly defends the value of these object-focused semiotic divisions. Nevertheless, he deserves credit for highlighting this important difference between the 10-sign classification and the 28- and 66-sign classifications. Additional scholarship is required to critically assess the importance and fruitfulness of these object-focused divisions.

The fifth chapter, “Interpretation, Worldviews and the Object”, is as disjointed as its title. The first half of the chapter is focused on Peirce’s concept of the object and reveals Jappy at his most confused. Through inscrutable paths, Jappy brings himself to the incredible conclusion that :

... the artist, a version of Peirce’s utterer, is outside the process. He executes the painting, but in doing so he is simply the vector of the artistic trends and public and private ideologies of the age, and, if there was one, the desires of the patron paying for the work : according to Peirce’s conception of the sign as medium in 1906, it is the dynamic object which structures the representation on the sign, not the artist.

152

The second half of the chapter provides analyses of signs similar to those offered in chapter four, now highlighting three paintings conveying the idea of manifest destiny, two paintings portraying the cultural oppression of women, and one photograph showing a protest movement. For Jappy, the significance of these signs is that what is represented in each sign is not merely the individual entities depicted, but something general like the idea of manifest destiny or of female oppression. Jappy suggests that, because the 1908 classification provides a term for signs whose dynamic objects are general, namely, collective signs, it provides a theoretical advance over the 1903 classification. Jappy’s examples show that the 28-sign classification can be applied to actual signs, but the mere fact that one can apply the terms of a proposed classification to the objects classified does not constitute evidence that the classification divides those objects in a way that is important, theoretically fruitful, or “natural” – meaning a classification that, in Plato’s gruesome phrase, “divides nature at its joints” (see EP 2.115-132, esp. 126-7). Taking the set of all known signs, what divisions most illuminatingly and most economically divide that set into distinct classes? Any set of objects can be classified in countless ways. For example, we might classify signs based upon the first letter of the most common English word for the object represented by that sign. Thus, a painting of an apple would be an A-sign; a sentence about a painting would be a P-sign, etc. This classification would be relatively easy to apply yet completely unilluminating. I am not suggesting that Peirce’s 28-sign classification is as pointless as the preceding, but only that the justification of a classification system requires something more than merely demonstrating the bare applicability of the system. What does the division based upon dynamic objects help us to understand about the signs it classifies? How does it illuminate processes of semiosis in which signs belonging to that class are being interpreted? What confusions does this classification help us avoid? What further inquiries does it prompt? Jappy’s defense of the 28-sign classification never broaches these basic justificatory questions.

2. Critique of Peirce’s Twenty-Eight Classes of Signs

Perhaps the most serious problem with this book is that the project undertaken in it lacks motivation. Why is the 28-sign classification important or interesting? Jappy’s most clearly stated answer to this question borders on paradox : “A further reason for exploring the 28-class typology is that the letter to Lady Welby in which it is advanced is apparently the only reference Peirce ever makes to it, his ambition being, no doubt, to prove and exploit the more complex 66-class system” (75). But why does Peirce’s neglect of it constitute a reason to be interested in this classification? In fact, saying that Peirce neglected the 28-sign system is an understatement : Peirce moves on from the 28-sign system to his more ambitious 66-sign system not the following year, month, or even day, but in the same sentence in which the 28-sign system is announced!

... the six trichotomies, instead of determining 729 classes of signs, as they would if they were independent, only yield 28 classes; and if, as I strongly opine (not to say almost prove) there are four other trichotomies of signs of the same order of importance, instead of making 59049 classes, these will only come to 66.

SS 84

If this is the first and last time the 28-sign classification is mentioned by Peirce, why should we care about it or think it is worthy of in-depth study? It seems to me that anyone who closely examines the proliferation of inconsistent classifications developed between 1903-1910 must at least sympathize with Atkins’ and Short’s characterization of this flurry of late ‘systems’ as “speculative, rambling ... incomplete”; “sketchy, tentative, and ... incoherent” (Atkins 2010 : § 2; Short 2007 : 259-60). In my opinion, all of Peirce’s a priori, category-driven classifications – the 66-sign, 28-sign, and 10-sign classifications – should be regarded as on probation until they receive more convincing a posteriori justifications. Peirce says as much himself : “The principles and analogies of Phenomenology enable us to describe, in a distant way, what the divisions of triadic relations must be. But until we have met with the different kinds a posteriori, and have in that way been led to recognize their importance, the a priori descriptions mean little; – not nothing at all, but little” (EP 2.289). At a bare minimum, we should be able, for each proposed class of sign, to : (a) find uncontroversial examples of actual signs belonging to that class; and (b) show how the application of the 28-sign classification helps to better understand these signs and the processes of semiosis in which they are involved.

To Jappy’s credit, he applies the 28-sign classification to an analysis of actual signs, namely, the six photographs, six paintings, and one poem mentioned in my summary. However, Jappy’s classifications are not uncontroversial, nor has he supplied examples of each type of sign. Jappy describes his classifications as “necessarily speculative”, (129) and “a matter of informed guesswork” (119). Regarding almost every sign he discusses, it is easy to imagine arguments supporting an alternative classification. Moreover, concerning the last six signs discussed, Jappy bluntly asserts that “What their respective interpretants are, or have been, is, of course, impossible to determine” (170). All told, of the thirteen signs that Jappy categorizes using the 28-sign classification, only four signs are given a complete classification with a label corresponding to each of the six divisions;[3] one sign is classified in five of six divisions, one in four of six divisions, and seven signs are only classified in three of six divisions. If we include the nine incompletely classified signs, four of these could, given their specified divisions, belong to one of the four completely classified types, and the other five signs require only two additional classes. Thus, on the most generous reading, Jappy provides examples of six of the twenty-eight classes of signs. Indeed, Jappy’s thirteen example signs – twelve of which are photographs or paintings – fail even to illustrate the six triadic divisions that define the 28-sign classification :

  1. Dynamic Object : 0 abstractive, 3 concretive, 10 collective;

  2. Immediate Object : 0 descriptive, 5 designative, 6 copulant;

  3. Sign : 0 mark, 12 token, 1 type;

  4. Immediate Interpretant : 0 hypothetical, 4 categorical, 1 relative;

  5. Dynamic Interpretant : 1 sympathetic, 5 percussive, 0 usual;

  6. Final Interpretant : 5 gratific, 1 producing action, 0 producing self-control.

As the preceding list reveals, for each of the six triadic divisions of signs, one of the three types is not exemplified by any of Jappy’s narrowly-chosen examples.

Jappy’s application of the 28-sign classification is both dubious and incomplete, but the more serious problem is that it is unilluminating. To be clear, Jappy has many insightful things to say about the photographs and paintings he discusses, but these insights derive, not from applying the 28-sign classification, but from Jappy’s background knowledge about the artworks and the artists who made them, as well as his understanding of aesthetics, literary criticism, and philosophy. Though Jappy is capable of using the 28-sign system to partially categorize each sign he discusses, the application of the 28-sign classification does not seem to do any explanatory work or to generate any other theoretical fruits.

By way of contrast, let me provide two examples of semiotic classifications that I think bear positive theoretical fruits. A genuine index is a sign that : (a) is capable of representing its object due to being causally effected by the object; and (b) represents its object both indexically and iconically, not only pointing toward its object, but also communicating positive information about that object (see EP 2.7-8, 2.14-16, 2.163, 2.171-2, 2.274, 2.306-7). Examples of genuine indices include a weathervane which represents the direction of the wind; a bear track that communicates information about the size, speed, and trajectory of the bear; and light arriving from a distant star, which carries, in its frequency and other physical properties, information about the chemical composition, temperature, and relative distance of that star. When we examine the natural signs that appear to ground our perceptual capacities – signs like lightwaves and soundwaves – we find that these signs are easily classified as genuine indices. If this classification is correct, then these signs, in and of themselves – i.e prior to and regardless of their interpretation in animal perception – represent their objects iconically. This, as Peirce hints in several places (see EP 2.2, 2.193-4; CP 1.311), might help to explain how qualitative experiences of likeness and contrast – blueness and orangeness, high-pitched and low-pitched – emerge as the ultimate upshot of perceptual interpretation. Perhaps perception is an icon in → icon out interpretive process, such that the ultimate explanation of the qualities or likenesses that are immediately presented in percepts requires an appeal to the iconicity of the natural signs grounding perception. Though only an interesting guess, this example illustrates how the mere classification of signs can suggest explanatory hypotheses to pressing philosophical problems.

As a second example, one of Peirce’s classifications of interpretants divides them into emotional, energetic, and logical interpretants, thereby suggesting that human beings are capable of interpreting nature through three interpretive modalities : experience, behavior, and symbolic thought, respectively. To begin with, this classification raises important questions about how these interpretive capacities are grounded in human nature. Crudely stated, my guess is that experience is fundamentally grounded in the sensory nervous system (with memory and imagination being amplifications of more basic sensory capacities), behavior is fundamentally grounded in the motor nervous system, and symbolic thought is fundamentally grounded in symbol systems or languages that evolve continuously across millennia while the specialized brains needed to interpret these languages dissolve after death. Also, if this classification is natural and truly divides human interpretive capacities into experience, behavior, and symbolic thought, then this helps to explain why our normative inquiries naturally divide into aesthetics, ethics, and logic. We must experience nature through our senses, so we ought to cultivate beautiful rather than ugly experiences and environments. We must act in a world with others, so we ought to cultivate good rather than evil actions and habits. We must represent the world with symbols, so we ought to cultivate true rather than false thoughts and theories.

Though only briefly stated, these examples illustrate the sorts of theoretical fruits we desire semiotic classifications to provide. The basic point is that classification is not an end in itself. Proposed classifications need to be justified in terms of both (a) clear examples of each proposed class and (b) theoretical fruits generated from applying the classification. The fruitfulness of the 28-sign classification might be justified by future inquiries, but it is not significantly supported by Jappy’s book.

A further problem with Jappy’s book is a pervasive failure to critically examine the terms and method of the 28-sign classification. Given that the book is so filled with technical terms, its sparsity of explicit definitions is stunning. In addition to failing to define key terms, Jappy ignores basic questions about the a priori methodology underlying the 28-sign classification. Crucial to the internal logic of this classification is Peirce’s claim that “the Dynamoid Object determines the Immediate Object, Which determines the Sign itself, which determines the Destinate Interpretant, Which determines the Effective Interpretant, which determines the Explicit Interpretant” (SS 84). Before presenting my concerns about this proposed order of determination, an aside on the meaning of the last three terms is necessary. Jappy interprets the rare terms “destinate”, “effective”, and “explicit” interpretants to be alternative labels for what Peirce more commonly labels the “immediate”, “dynamic”, and “final” interpretants, respectively (88-89). The immediate interpretant of a sign is the sign’s capacity for being interpreted by a possible interpreter, or the “interpretability” of the sign prior to any actual interpretation. As Peirce says in another letter to Lady Welby, “My Immediate Interpretant is implied in the fact that each Sign must have its peculiar Interpretability before it gets any Interpreter ... The Immediate Interpretant is an abstraction, consisting in a Possibility” (SS 111). While the immediate interpretant is merely the abstract “interpretability” of a sign, the dynamic interpretant is the interpretant that is actually generated when some interpreter actually interprets the sign. Lastly, the final interpretant is the semiotic expression of Peirce’s emphasis upon the long run of inquiry (see EP 1.109-141) : the final interpretant is the meaning the sign would have if the final, fated opinion concerning the object were reached by some community of intellectual inquiry. In Peirce’s words, “the Final Interpretant is the one Interpretative result to which every Interpreter is destined to come if the Sign is sufficiently considered” (SS 111). Thus, for example, the final interpretant of the term “atom” is not the vague Greek conception of something uncuttable or indivisible, not Rutherford’s model, not Bohr’s model, and probably not the current quantum chromodynamic model; it is rather that model of the atom that would be represented in the final opinion, were the final opinion on the question reached by some community of intellectual inquiry. Given these definitions, it seems to me all but obvious that, contra Jappy, David Savan’s interpretation of these terms must be on the right track : the “destinate” interpretant should be identified with the final interpretant – it is the interpretant that is destined or fated to be established in the long run of inquiry; while the “explicit” interpretant should be identified with the immediate interpretant – it is the potential interpretability that belongs to the sign itself, even prior to any actual interpretation (1988 : 52; cited Jappy 2017 : 89).

Whether one agrees with Jappy or Savan on this peripheral issue, I think both the final interpretant and the immediate object are problematic given the order of determination posited by Peirce. Beginning with the final interpretant, assume first that Jappy is correct and that the order of determination – where (X → Y) means “X determines Y” – is : (sign → immediate/destinate interpretant → dynamic/effective interpretant → final/explicit interpretant). How does the dynamic interpretant determine the final interpretant? Undoubtedly, actual or dynamic interpretation is essential to the process of learning through which the working theory of a community of intellectual inquiry comes to approximate the final interpretant. Peirce saw clearly that the growth of a symbol’s meaning requires translation or interpretation of that symbol into embodied, habitual, and pragmatic engagements of the object(s) represented by the symbol. According to his thought experiment, a translation machine that perpetually translated symbols into further symbols without generating any practical, embodied interpretations could never contribute to the “growth in idea-potentiality” or significance of the symbols being translated (EP 2.387-8). In order to learn, one must venture forth on the basis of the sign, engaging the dynamic object in a pragmatic, embodied way. In response to such embodied engagements, the dynamic object often surprises us, showing up as secondness, resistance, and feedback. Via this corrective feedback, nature is revealed to be different than it was represented to be, and learning is made possible. Expressing the same point with respect to scientific inquiry, in order to test a scientific theory, one must design an experiment whose most essential ingredient is “an experimenter of flesh and blood” who can “act” on the basis of the theory, thereby allowing for “the subsequent reaction of the world upon the experimenter in a perception” followed by “the recognition of the teaching of the experiment” (EP 2.339-40). Granting the importance of actual, dynamic interpretation for reaching the final interpretant, does it make sense to say that the dynamic interpretant determines the final interpretant? Is it not, rather, the dynamic object that resists dynamic interpretants, surprises expectation, corrects prior theory, and ultimately determines the final interpretant? In the young Peirce’s words, the final opinion is determined “by some external permanency – by something upon which our thinking has no effect” (EP 1.120), by “a force outside themselves” which carries even antagonistic investigators “to one and the same conclusion” (EP 1.138).

The final interpretant remains problematic even if we adopt Savan’s more plausible interpretation of the destinate, effective, and explicit interpretants. If we follow Savan, the order of determination Peirce posits is (sign → final/destinate interpretant → dynamic/effective interpretant → immediate/explicit interpretant). Is it true that the sign determines its final interpretant? Clearly without some signs to interpret – no matter how incorrectly or incompletely those signs represent their dynamic objects – we could not reach the final opinion about anything. Nevertheless, in the course of inquiry most proposed signs will be abandoned as “experience ... gradually, and by a sort of fractionation ... precipitate[s] and filter[s] off the false ideas, eliminating them and letting the truth pour on in its mighty current” (EP 2.154). Once again, it seems better to say that the final interpretant is determined, not by the sign, but by the dynamic object. What about the claim that the final interpretant determines the dynamic interpretant? This would be true only in the special case where the final opinion concerning the meaning of the sign has already been reached. In this case, actual dynamic interpretants, whether purely theoretical or practical and technological, could be organized on the basis of this final interpretant. But this situation must surely be the exception rather than the rule. More typically, our dynamic interpretants are generated on the basis of immature signs incompletely understood; and, with respect to these unfinished signs and fallible dynamic interpretations, the final interpretant is only a dimly imagined possibility. To say that the final interpretant always determines the dynamic interpretant is to suggest that all actual interpretation is informed by the final, settled opinion concerning the objects being interpreted. But the prominence of error, surprise, and vulnerability in human life renders that claim incredible.

In addition to the final interpretant, it seems to me that the role of the immediate object in Peirce’s determination sequence is also highly suspect. Roughly speaking, the immediate object is the object as it is represented by the sign, while the dynamic object is the object as it really is, independently of how it is represented by the sign. Peirce describes the immediate object as being “within the sign” (EP 2.480), “the object as the sign represents it” (EP 2.482), and “the Object as cognized in the sign” (EP 2.495; see also EP 2.403-409, 477, 498). As T. L. Short notes, the distinction between immediate and dynamic objects is needed to register the contrast between true and false signs as well as interpretive “success and failure” (2007 : 191). If I say that, “The pine tree in my backyard is 200 meters tall”, the immediate object of that sentence is a pine tree, 200 meters tall, located in my backyard. But if you actually visit my backyard, you will see that the pine tree located there is closer to 15 meters tall. The actual 15-meter tree – which is as tall as it is regardless of how tall I or anybody else represents it to be – is the dynamic object.

Given this distinction, what sense does it make to assert that : (dynamic object → immediate object → sign)? If the claim were only that (dynamic object → sign), that would be relatively intelligible. Concerning natural signs, a bear track is determined by the bear that actually made the track, just as the level of mercury or alcohol in a thermometer is determined by the actual temperature of the environment. Concerning human symbols, one would have to tell a more complex, evolutionary story in order to explain how the content of common symbols like “fire”, “cat”, and “plant” have been determined by the dynamic objects represented by those signs; and the process involved in that explanation would only be a more temporally extended and less intentional instance of the same process whereby the content of scientific terms like “heat”, “atom”, or “gene”, is determined by the objective realities represented by these terms. At least as a tendency of inquiry, Peirce clearly thought that the dynamic object tends to determine the content of our symbols, especially those symbols employed in scientific inquiry. Indeed, to say that the dynamic object determines the signs interpreted in intellectual inquiry is only a different, object-centered way of describing the tendency for the destined final interpretant to emerge during the course of intellectual inquiry. Though not uncontroversial, I think the claim that (dynamic object → sign) is defensible. But what sense does it make to say that (dynamic object → immediate object) or that (immediate object → sign)? It is only true that (dynamic object → immediate object) in cases where the sign’s representation of the dynamic object is more or less true. In the special case where the final interpretant is already established, the claim that (dynamic object → immediate object) would be completely true : presumably due to prior experience, the sign represents that object as the object really is. But in the case of false signs, it is not the case that (dynamic object → immediate object). Rather, a false sign’s representation of the object is determined by something non-objective, arbitrary, or biased. Thus, it is not generally true that (dynamic object → immediate object). Indeed, to say that the immediate object is always determined by the dynamic object is to undermine the very distinction between truth and falsity that makes the distinction between these two objects necessary in the first place.

While there may be some limited cases where it is true that (dynamic object → immediate object), the claim that (immediate object → sign) is simply bizarre. Given that the immediate object is something “within the sign”, namely, “the Object as cognized in the sign”; what sense does it make to say that the immediate object determines the sign? Jappy never raises or answers this question. Rather, he accepts it as true that (immediate object → sign) and then distorts his understanding of the immediate object and the sign as required to accommodate this claim. Jappy describes the relation of the immediate object to the sign as follows :

the immediate object can’t interpret the dynamic; it receives form from it, which it communicates to the sign

151

material from the 1906 definition of the sign as medium had the effect of diminishing the importance of the sign in semiosis, and at the same time it gave the immediate object a specific representative status as a sort of filter, communicating parts of the form or structure of the dynamic object to the sign

153

what we see when we look at an image of any sort, or what we hear when we process an utterance of any sort, or what we read in a text of any sort is, of course, what their immediate object has filtered through to them from the object they represent

155

the immediate object, functioning as a sort of filter ... determines what form or forms emanating from the dynamic object find expression in the sign.

176

In his effort to sharply distinguish the immediate object from the sign, which is necessary in order to affirm that the immediate object determines the sign, Jappy seems to construe the form or content of a sign as being the immediate object and the sign itself as merely the matter or substrate that bears this form. Thus, the immediate object of a painting is the image painted on the canvas, and the sign is merely the canvas; the meaningful content of a written sentence is the immediate object and the sign is merely the piece of paper (see 113, 152, 167). In discussing an artist’s sketch of a woman’s face, Jappy says, “The immediate object functions as a ‘filter’ and is the determinant of the incomplete representation of the model’s face being sketched on the sheet of paper, while the sign, as Peirce described it in 1906, is the particular sheet from the sketch-pad on which the artist is working” (55). Based on these quotations, I have no idea what sort of entity Jappy imagines an immediate object to be, nor how this semiotic “filtration” process is supposed to occur. What does seem clear to me is that Jappy’s radical separation between the immediate object and the sign lacks any basis in Peirce’s writings. Taking Peirce’s writings about the immediate object as a whole, the claim that (immediate object → sign) makes no sense. Though Jappy ignores these problems concerning the methodology of the 28-sign classification, other Peirce scholars interested in defending the coherence and importance of the 28-sign and/or 66-sign classifications should address them in the future.

My final criticism of Jappy’s book is that he – apparently in an effort to make the 28-sign classification seem more interesting – overstates the differences between this system and Peirce’s previous work in semiotics. He does this in two ways. First, he lists as one of the “innovative features” of the 28-sign classification the “disappearance of Peirce’s first and most fundamental trichotomy identifying icon, index, and symbol” (103). This “first and most fundamental” (EP 2.273) trichotomy may have “disappeared” from Jappy’s reconstructed classification, but it is clearly articulated in the letter in which Peirce announces and surpasses the 28-sign system. Of the four additional divisions Peirce suggests in that letter, the last three are introduced tentatively with phrases like “I am pretty confident”, “I think”, and “I suppose”; but the first division into icons, indices, and symbols is described as “undoubtedly” needed (SS 84-85). Second only to the fundamental triadic relation between sign, object, and interpretant – which Jappy incorrectly claims Peirce abandoned for a “hexadic” semiotic relation (51-2) – Peirce’s distinction between iconic, indexical, and symbolic signs is the most stable feature of his semiotic theory. Contra Jappy, there is no reason to think Peirce rejected this distinction.

The second way Jappy overstates the difference between the 28-sign system and Peirce’s prior work in semiotics is by drawing a sharp distinction between Peirce’s three universal categories and the “three Universes” of “possibles”, “existents”, and “necessitants” (see SS 81-2). Jappy describes this “very significant difference” as follows :

the 10-class typology was defined within a phenomenological framework since it used three categories – predicates bearing on whatever can be present to the mind – as the criteria by means of which to subdivide the sign and the two sign-correlate relations. The second typology, on the other hand, was defined within an ontological framework, and employed three universes – receptacles of what there is, embracing possible, existent, and general objects – to define the subdivisions of the six correlates of semiosis which, when properly combined, generate twenty-eight very different classes.

175

Jappy is surely correct to see Peirce’s three universes as ontological, rather than phenomenological. Just as surely, he is wrong to suggest that the ontological universes are somehow radically different from Peirce’s categories. As T. L. Short notes, the mature Peirce gave “a formal, a phenomenological, and an ontological account of each of the three [categories], Firstness, Secondness, and Thirdness” (1974 : 184). These three accounts of the categories are well illustrated by lectures 2, 3, and 4 from Peirce’s Harvard Lectures on Pragmatism : lecture 2, “On Phenomenology”, provides a phenomenological account; lecture 3, “The Categories Defended”, provides a formal or mathematical account; and lecture 4, “The Seven Systems of Metaphysics”, provides an ontological account of the categories (EP 2.145-195; on Peirce’s categories, see also CP 1.300-353; EP 2.267-272, 360-370; RLT 146-150, 258-262). Jappy’s claim that Peirce made a sharp distinction between the three categories and the three universes is utterly implausible, which is why, as Jappy notes, “most if not all of the authorities ... consider that they are equivalent” (78). Rather than being sharply contrasted with the three universal categories, the three universes are the categories construed realistically as describing reality’s three modal divisions : pure possibility, brute actuality, and lawlike generality (see EP 2.434-5). If the three universes are just the three categories construed modally, then, contra Jappy, the fact that the 28-sign classification is constructed on the basis of the three universes does not distinguish that classification from Peirce’s prior category-driven sign classifications.

In summary, I have provided five general criticisms of Jappy’s book : he fails to properly motivate interest in the 28-sign classification, he fails to provide clear examples of each class of sign, he fails to show that application of the classification generates any theoretical fruits, he fails to critically examine the terms and method of the 28-sign classification, and he misleadingly portrays the 28-sign system as sharply discontinuous with Peirce’s prior work in semiotics.

3. Implications for Ongoing Efforts to Develop Peirce’s 66-Sign Classification

My evaluation of Jappy’s book has been quite negative, but this does not mean that I reject the general project of reconstructing and assessing Peirce’s late sign classifications. Because the 28-sign system Jappy discusses is only an arbitrarily truncated version of the 66-sign classification, I do not think the former warrants an independent inquiry. However, the 66-sign classification is a more stable feature of Peirce’s late explorations, and it remains poorly understood. While I am doubtful about the value of the 66-sign classification, I think that reconstructing and assessing this late system remains an important theoretical goal for the community of semiotic inquiry. As a concluding reflection, let me summarize the significance of the preceding review for this ongoing effort to reconstruct and assess the 66-sign system.

The preceding review offers four criteria that those attempting to reconstruct and defend the 66-sign classification should try to meet. First and foremost, they should offer a convincing defense of the methodology underlying that system, especially regarding the sequence of determination purportedly holding among the elements of the process of semiosis. In the preceding review, I raised questions about the roles in this determination sequence of the immediate object and the final interpretant. Similar questions could also be raised about the immediate interpretant, which, like the immediate object, seems to be an abstraction from the sign : whereas the immediate object is the way the object is represented in the sign, the immediate interpretant is the interpretability the sign has for some possible interpreter. If the immediate object is an abstraction that represents the content of the sign as it stands with respect to the dynamic object; then the immediate interpretant is an abstraction that represents the content of the sign as it stands with respect to some possible dynamic interpretant. As abstractions from the sign, neither is the sort of thing that can be dynamically determining anything else. Though I will not belabor the point there, I think that the placement in the sequence of determination of the four additional divisions proposed in the 66-sign classification (see SS 85) are also problematic. Unless advocates of this 66-sign system can defend Peirce’s dubious claims about the determination relations holding between the entities involved in semiosis, then the argument on behalf of this classification system is a non-starter.[4]

Secondly, defenders of this system should define, as clearly as possible, both the terms of each triadic division and the resulting 66 classes of signs. With these a priori definitions in hand, the third step of the defense must be to identify actual signs that can be uncontroversially classified as belonging to each class. Presumably, this will require compiling a set of signs much more diverse than the paintings and photographs Jappy focuses upon. Fourth, and finally, having provided clear examples of each class of sign, defenders of the 66-sign classification ought to show that classifying signs according to this system generates significant theoretical fruits. These fruits should include, minimally : explanatory power, meaning that the classification helps us to understand the signs classified and the processes of semiosis in which these signs are involved; and suggestiveness for future research, meaning that the classification opens up new avenues for semiotic research, including generating practically and empirically relevant predictions by which the classification can be tested. Other theoretical virtues like internal coherence, empirical adequacy, and simplicity will also be important in determining whether this classification system is preferable to other semiotic classifications, such as the 10-sign classification of 1903.