Semiosis & Sign Exchange

design for a subjective situationism, including conceptual grounds of business information modeling

Pieter Wisse

chapter 5

Critical synopsis of Eco’s semiotics

 

 

 

Especially in Part i of this treatise, the individual sign user holds the privileged position for explanation. His unique existence is taken as axiomatic. Then why do I include at this stage a synopsis of Umberto Eco’s A Theory of Semiotics (1976)? Doesn’t Eco assume that meaning is a social convention, rather than grounded in personal “conduct” or - a concept I consider equivalent to conduct - behavior?

It is precisely because Eco attempts to maintain a social, or cultural, explanation of semiotics¹ that a discussion promotes an understanding of subjective situationism. For at least regarding its assumptions, his semiotic theory provides the benefit of a clear contrast.

Eco starts from the idea that language - any signification system, for that matter - strictly is a social system. This leaves many aspects of meaning unexplained, though. I particularly mean the sequences of interpretants occurring in the intelligence of the individual sign user during his unique process instances of sign use. Also fundamentally missing from accounts such as Eco’s is the individual creativity in sign engineering. Exactly for these reasons the intentionally demeaning title of § 2.4 already refers to semantics as the poor cousin of pragmatics. For some research purposes it is of course fruitful to assume a separate “symbolic order.” But then, next, concepts need to be related as for example K. Silverman suggests (1983, p 282):

[S]ignification cannot be studied apart from discourse, discourse from subjectivity, or subjectivity from the symbolic order.

My shift from society annex culture to individual sign user permits clarification of where Eco’s overall theory shows contradictions. Especially as far as the concept of sharing identical meaning between different sign users, or even its possibility, is concerned, his explanation simply equals his original classification.

In other words, the variables of Eco’s theory lack requisite variety. For a theory needs to recognize necessary and sufficient explanatory variables that are different from what they explain. This is what R.A. Wicklund calls the inquiry into the background perspective. Instead, Eco already turns the possible result of theorizing into his perspective. Thus he fails to observe the problematic nature of shared meaning. As Wicklund remarks in Zero-Variable Theories and the Psychology of the Explainer on creating a zero-variable theory (1990, p 22):

The beginning point of the analysis - the grouping of behavioral instances into a seemingly meaningful whole - is also the stopping point of the theoretical analysis. But what else? No one would try to argue that the mere identification of a cluster of behaviors is a theory!

Short-circuiting “stopping point” with “beginning point” of course secures theoretical ‘proof’ but only as an immediate tautology. In particular shared meaning is left unexplained (also read: is not build up as a system of variables) in A Theory of Semiotics because Eco effectively starts from the concept of shared meaning, or mutual knowledge. It acts as his axiom, as I show below in more detail. Critizing what they call code models of communication, D. Sperber and D. Wilson pinpoint the faulty reasoning (1986, p 19):

There is a paradox here. Since the assumption of mutual knowledge may always be mistaken, the mutual-knowledge hypothesis cannot deliver the guarantees it was set up to provide.

They continue (p 21):

We see the mutual-knowledge hypothesis as untenable. We conclude, therefore, that the code theory must be wrong, and that we had better worry about possible alternatives.

In defense of Eco, I repeat his own stated objective. It is not directly oriented at shared meaning, but he wants (1976, p 3)

to explain every case of sign-function in terms of underlying systems of elements correlated by one or more codes.

Aiming at “every case” includes, I suppose, the Peircean dynamics within the intelligence of the individual sign user. Does it for Eco? The background perspective of the individual sign user, which Part i of this treatise introduces in force, suggests that Eco overstates his goal. The following critique shows that approaches of such closed semantics don’t contribute to the background perspective required for explaining processes of sharing meaning or, actually, the lack thereof.

 

 

5.1 society as independent actor

Contrary to Eco’s proclamations about “every case of sign-function,” A Theory of Semiotics fits within a strictly linguistic tradition. A pioneer of linguistics as a scientific discipline is Ferdinand de Saussure (1857-1913). Some of his students have actually contributed greatly to his subsequent fame. After De Saussure dies, they compile their lecture notes into a coherent book titled Course in General Linguistics (1916). One of the ideas² thus attributed to De Saussure is that signs, and all they entail, do not stand on their own. One sign’s particular meaning is always dependent on other signs and their meanings. Thus, De Saussure provides an early articulation of what later becomes known as structuralism, applied in a wide range of disciplines. In § 4.4 I have already cited the prominent structuralist Lévi-Strauss.

Another idea that De Saussure presents in Course in General Linguistics, and one implicitly present throughout Eco’s work, concerns the concepts of langue and parole. Their specialized meanings are usually underlined when these French terms are used in English texts, too. Langue stands for the whole of the language system, whatever that may be. Parole, then, corresponds to what this treatise considers as instances of - processes of - sign use.

Eco assumes that langue controls parole. Basically he says that persons communicate and thereby share meaning. And for him, langue is the repository of meanings which are shared by definition through a priori society, or culture. So, Eco does not explain how persons do their sharing. Instead, he merely assumes they do.³ His own terminology for langue is code or, rather, a system of codes. He writes, for example (1976, p 56):

[c]odes provide the conditions for a complex interplay of sign-functions.

The conditional nature of codes explains why Eco gives a “theory of codes” precedence over a “theory of sign production.”⁴ Although Eco himself writes that his (p 4)

distinction between a theory of codes and a theory of sign production does not correspond to the ones between langue and parole, competence and performance, syntactics (and semantics) and pragmatics[,]

my critique shows that he does not succeed

to overcome these distinctions [… by] proposing a semantics which solves within its own framework many problems of the so-called pragmatics.

There it is clearly Eco who is demeaning about pragmatics. Regretfully he fails to acknowledge that nothing can be explained within its own framework. A coherent interpretation A Theory of Semiotics is possible only when it is taken as a collection of attempts to keep his fundamentally limited semantic framework intact. As I already indicated, Eco assumes at the start of his theorizing (p 8) “an underlying system of significations.” Such a system is what I have called a repository, above. Eco accords to it an independent existence (p 9):

A signification system is an autonomous semiotic construct that has an abstract mode of existence independent of any possible communicative act it makes possible.

He associates it firmly with society, as when he explains that (p 49)

a sign is always an element of an expression plane conventionally correlated to one (or several) elements of a content plane.

The key term to notice in the sentence above refers to convention. For Eco continues that

[e]very time there is a correlation of this kind, recognized by a human society, there is a sign.

Earlier, he says (p17):

There is a sign every time a human group decides to use and to recognize something as the vehicle for something else.

Now my bicycle is a vehicle. When I carry a bag on my bicycle, does that make the bicycle the sign for the bag? Or, even, does the bike stand for me. Don’t I use it to carry me, especially so?

It is of course not what Eco means by vehicle. The emphasis to be read from his sentence is that a sign ‘originates’ from “a human group,” i.e., that it is socially determined.

Nowhere, however, does he explain how society performs such acts of determination and recognition. He must assume that society does, that is all. Eco presumes (p 61) “a cultural order, which is the way in which a society thinks.”

Throughout he awards to society such anthropomorphic nature. Making sense of A Theory of Semiotics requires a recognition of Eco’s concept of society as an independent, human-like actor.

For a synopsis of Eco’s semiotics, it must first of all be realized that he actually does not theorize about semiotics. His theme is semantics, treating it in a grammar-like fashion, at that. Secondly, it facilitates appreciation of his assumptions to make a minute caricature out of his axiom of society being the repository of meaning. In my sketch, Society is a person, too. She acts as the essential broker of meaning. Mrs Langue is the name of this privileged actor. She is called Connie - from convention, of course - by all her friends. In fact, everybody is supposed to be her friend. For, at one time or another, everyone will communicate, if only by interior monologue. The funny thing about that exemplary society is that all friends, which means everybody except Connie Langue herself, share the same last name. It is Parole. So there is John Parole, Mary Parole, and little Suzan Parole, etcetera, etcetera.

 

 

5.2 centralized communication

Embodying Society, Connie Langue controls all communication between ‘her’ members, that is, between all the Parole citizens. There is, however, something special about all the instances of communication. An individual Parole who originates a message (also read: sign) does not realize that Connie scans it. For example John simply experiences that he is talking directly to Mary. But in the background Connie always intervenes. Only when she understands the message or sign, i.e., when she experiences an already known content, does she pass it on. The Parole on the receiving end is equally unaware of her coordinating efforts.

My parody highlights what Eco’s main objective is with his theory of codes. Given an occurrence on the expression plane, he seeks to describe what Connie Langue recognizes on the content plane. An expression is a sign vehicle. It ‘carries’ a sememe. Such content is described by semantic markers (pp 84-85):

[A] network of mutually opposed features […] rule[s] the difference between two sememes. Thus, to say that a sign-vehicle conveys a given position within a semantic field constitutes a shorthand definition. […] As a matter of fact one must assume that a sign-vehicle may refer (i) to a network of positions within the same semantic system, (ii) to a network of positions within different semantic systems. These positions constitute the semantic markers of a given sememe.

Eco continues to distinguish between denotative and connotative markers (p 85):

[D]enotative [are] markers […] whose sum (or hierarchy) constitutes and isolates the cultural unit which first corresponds to the sign-vehicle and upon which rely all the other connotations.

This implies that

a connotation must rely on a preceding denotation.

So first of all, the ‘knowledge’ of Connie Langue may be described as a collection of trees. A single sign vehicle constitutes the root of every tree. Its branches are each composed of one or more denotations. Zero, one, or more leaves appear at the end of every branch as connotations. Trees may be interconnected between branches, between branch and leave, vice versa, and between leaves. It amounts to the concept that Connie Langue’s knowledge (also read: repository of content starting with form) ultimately is a network, rather than a collection of hierarchies.

In the one but last quotation, above, the (conceptual) content is referred to as a “cultural unit.” It is an important concept for Eco’s theory, and understanding what he means by it especially explains why his theory falls short of the pragmatics of Peirce. This is treated in the next paragraph. For now I continue with the way Eco models sememes.

Figure 5.2.1.
An abstraction of Eco’s model of meaning(s) of a sign vehicle.

 

Figure 5.2.1 presents Eco’s basic idea about the knowledge of Society (here: Connie Langue). The abstraction of only a single tree is shown.

An even more condensed schema exploits the recurrent nature of elements. So, Figure 5.2.2 is presented as the equivalent of Figure 5.2.1.

Figure 5.2.2.
A more compact model of meanings that might originate from a sign vehicle.

 

Figure 5.2.2 indicates that a major problem of semantics is homonymy. A single sign vehicle may have a large number of meanings, independent of their categorizations into denotations and connotations. According to Eco’s theory of codes, the task of Connie Langue is not to choose any particular meaning out of all possible meanings for a sign vehicle. Rather, it is to only ‘significantly’ check the sign vehicle. It passes her inspection when she considers it at all meaningful, i.e., when it triggers any meaning (also read: content) in her knowledge.⁵

For the sake of my own conceptual development I assume that Connie does like to understand what content her friends want to communicate to each other. Therefore, suppose she is not satisfied with merely passing on a sign vehicle when she recognizes it as a tree root. How does she tell the difference from one denotation/connotation to another? Eco suggests a mechanism of contextual and circumstantial selections to (p 105)

distinguish the different readings of the sememe as encyclopedia item and determine the assignment of many denotations and connotations [… with such selections also being] pieces of coded information, in other words semantic units just like the others except that they perform a switching function.

In his view a single tree, growing from a particular sign vehicle, has a structure in which permutations of denotation, contextual selection, and circumstantial selection occur. Please note that, like the previous figures, Figure 5.2.3 exemplifies an abstraction on my part. Eco does not systematize his own theory of codes to this extent.

Figure 5.2.3.
All of Eco’s semantic markers accounted for.

 

As with the difference between denotation and connotation, Eco does not offer much guidance for distinguishing between - his concepts of - context and circumstances.⁶ For an explanation he argues that (p 106)

[c]ontextual selections record other sememes (or groups of sememes) usually associated with the sememe in question; circumstantial selections record other sign-vehicles (or groups of sign-vehicles) […] usually occurring along with the sign-vehicle corresponding to the sememe in question.

The italics Eco places make it once again clear that his statement is about codes, that is, about Socially accepted (also read: shared) meanings. Now what about Connie Langue’s desire to unambiguously understand what one Parole might be telling another Parole in Society? It remains unfulfilled as far as contextual selections are concerned. Whatever Eco means by them, what they do is to draw attention to the fundamental interconnectedness of the semantic forest. No tree stands isolated, and I agree. But a contextual selection as Eco proposes cannot serve as a switch. For Connie Langue already needs directions upon entering the forest. After she is lost inside, it is too late. Indeed being dropped in the middle, so to say, Connie still does not know which occurrence of all possible meanings is relevant for the particular occurrence of the sign vehicle. What does help her selection is to broaden her perception to the configuration of sign vehicles in which the one under inspection appears. This is what the circumstantial selections are supposed to do. Following Eco’s theory, it is only after Connie Langue has thus chosen a particular tree that she is in a position for contextual selections to point out other trees of possible interest. It is a link, rather than a switch.

 

 

5.3 culturalized reality

In Chapter 2 I have shown that Peirce explains sign use on the basis of an irreducible triad consisting of sign, object and interpretant. He does not claim that an interpretant provides the ‘truth’ about an object. Instead, interpretants are beliefs and doubts in all degrees imaginable. They guide - the conduct of - the individual sign user.

The dynamics of sign use are driven by a belief, too. It resides in the referential nature of signs. Vaihinger (1911) writes about fictions, arguing that an individual person experiences his world as if it exists in such and such a way. His concept of fiction is similar to what Peirce calls belief.

The only way to make sense of signs is to believe they stand for objects. Because he does not want any obstacles for dynamics of sign use, Peirce proposes that “anything goes” as far as objects are concerned.⁷ Eco agrees with the referential nature of the sign for (p 7)

[a] sign is everything which can be taken as significantly substituting for something else.

His phrasing is somewhat unfortunate. Why does he add “significantly”? Does he mean “clearly,” perhaps? Or does he want to say that it is a sign that acts as the substitute? But doesn’t he want to describe what a sign is, in the first place?⁸ Further on, Eco is less confusing about his concept of sign (p 16):

I propose to define as a sign everything that, on the grounds of a previously established convention, can be taken as something standing for something else.

His emphasis on convention is precisely why his theory cannot explain, in support or in criticism, something like shared meaning. Again, his theory effectively starts with it.⁹ In terms of Peirce’s triad, Eco only calls something a sign when it stands for a conventional object. He needs this limitation to avoid extending his theory beyond semantics into pragmatics (p 58):

The problem in question is that of the referent, in other words the problem of the possible states of the world corresponding to the content of the sign-function.

Why it is a problem, he doesn’t make clear. Eco applies a reversal by stating that (p 58) “[a] theory of codes must study everything that can be used in order to lie.” Does he mean to say that the concept of referent has a “malignant influence” on a theory of codes because a sign function cannot guarantee the truth about the state of the world? Is that the reason why the “referential fallacy” must be avoided? His cure, however, is worse than the disease. It is even lethal, at least when I take Eco seriously with his insistence on convention. How can something outside convention ever be expressed? Anything new? If not by a sign, then by what?¹⁰

I happily grant that Eco probably does not mean any referent. For he, too, describes a sign as something that stands for something else. Apparently, he has a special kind of referent in mind. Perhaps it is an indication that - a theory of - codes governs the conditions for communication. Therefore, codes do not refer to objects. They only provide rules for eligible signs (p 59):

A theory of codes is concerned with intensional semantics.

But then, when indeed codes abstract from extensions - where extensions must probably be seen as generated instances of signs -, why does Eco include descriptions of meaning in his theory of codes? Why doesn’t he limit himself to purely grammatical specifications. I agree when the answer is that some sort of referential information is needed for communication rules to function properly. The linguistic compartments from phonetics through to pragmatics block comprehensive explanations. But Eco writes (p 60),

[i]t must be absolutely clear that [a theory of codes] has nothing to do with […] an extensional semantics. […A]n expression does not, in principle, designate any object, but on the contrary conveys a cultural content.

From a wider perspective it is hard to follow Eco’s argument. He sets up constraints that I find counterproductive. For example, he doesn’t want the sign to stand for an (p 61) “actual object” as that reflects “a distinctively naive attitude:”

[T]he codes, insofar as they are accepted by a society, set up a ‘cultural’ world which is neither actual nor possible in the ontological sense.

It is amazing to what lengths Eco goes in his attempts to maintain “a semantics […] within its own framework.”¹¹ He creates an opposition, without actually mentioning it, between world (also read: reality) and culture. His logic is that as it is possible to lie about the world, only culture can be trusted. It is trivial that it can when, as Eco assumes, culture is equivalent to convention, i.e., culture consists of meanings about which exist a priori consent. With ‘cultural truth’ secure by definition, I admit that Eco really does not need an explicit referent anymore. Of course it remains open how “cultural content” relates to, say, world content. Juggling words fail as a credible explanation (p 62):

The semiotic object of a semantics is the content, not the referent, and the content has to be defined as a cultural unit.

But when

the theory of codes is only interested in sign-functions and the rules of their possible combination[,]

why is meaning as cultural unit implied in codes at the semantic level? Why, at all, are codes at the semantic level relevant for eco? I recognize a consistent desire to treat successful communication from a strictly semantic perspective. Different persons, so he reasons with his theory of codes, always agree when they communicate. For the communicative instrument they employ preconditions their agreement. A priori shared meaning rules.

 

 

5.4 comparing interpretations

Eco’s semiotic theory is unproductive for explaining shared meaning, let alone for criticizing it. I hypothesize the main reason is that he fails to understand the essential nature of Peirce’s triad. Eco misses how Peirce establishes the fiction of the object. Instead, Eco mistakes the concept of object for a claim to absolute truth.

Closely reading both Peirce and eco, it strikes me that my interpretations of Peirce’s work differ so widely from how Eco understands exactly the same texts.¹² It looks as if Eco believes his own theory is already final and beyond improvements before he embarks on a study of Peirce. He therefore projects Peircean concepts onto his closed semantic framework. Eco enlists them, overwhelms the new signs with the weight of his previous cognitive mass which remains relatively unchanged through semiosis. But reinforcing semantics with recourse to Peirce involves an injustice. For Peirce doesn’t describe how a priori successful sign exchange proceeds. He concentrates on semiosis, sign action in particular as it relates to conduct. Peirce’s framework is definitely pragmatic. It is reduced to a strictly semantic framework at the cost of severe loss of ‘meaning.’ Reduction is precisely Eco’s procedure. An extreme example is his suggestion that (p 144)

one must radicalize Peirce’s approach and insert the notion of the interpretant into the framework of a non-referential and structural theory of codes and of semantic systems.

What remains is pure grammar, of course. This I find unacceptable to imagine as a concept that Peirce allows to be derived from his related theories of pragmatics and signs.

Eco actually depersonalizes society by insisting that individual persons can only successfully exchange meanings as cultural units. His semantics abstract completely from overall behavior by an individual. My idea is that any such theory is sterile. Peirce is also right, I believe, to view sign use at the service of personal conduct.

With fundamental differences in outlook between Peirce and eco, is it surprising that Eco consistently refers to Peirce? I give some examples of my inventory of misinterpretations on the part of eco.¹³

On p 15, Eco introduces Peirce by repeating what the latter considers a sign. Eco adds that

a sign can stand for something else to somebody only because this ‘standing-for’ relation is mediated by an interpretant.

I don’t agree. My interpretation of Peirce is that the interpretant is not the mediating element. It is the sign. This explains figure 2.4.2 where the sign is at the top of the triangle, and a broken line is drawn between interpretant and object indicating that only an indirect connection exists. Eco doesn’t notice the essential difference between what I called, in § 2.4, the classical semantic triangle (see Figure 2.4.1) on the one hand, and Peirce’s pragmatic triad on the other hand (see Figure 2.4.2). His own two figures on page 59 of A Theory of Semiotics confirm this. I consider it a significant departure from the focus on instances of sign use. Instead, Eco starts from what he takes is the necessary precondition of signs, that is, from codes as a property of a society.

My (own) major (pre)occupation in Chapter 2 has been the development starting from Peirce’s concept of ground. There I took my cue from what at first looks almost like an ornamental addition in a sentence in which Peirce summarizes his concept of sign. It turns out Eco reads those words very differently (p 16):

I suppose it is in this sense that one must take Peirce’s definition of the ‘standing-for’ power of the sign “in some respect or capacity.”

I find Eco’s reference to a particular “sense” difficult to fathom. Does he mean that “in some respect or capacity” equals a property of an object which qualifies it as a sign. If that is indeed what he means, why does he mention it? For what does it add to Peirce’s description?

An additional problem with understanding Eco on this point is that he is following C.W. Morris (1901-1979). The latter also doesn’t express himself too clearly, at least not in the quotation that Eco offers (p 16). There, Morris can be read as making a statement about, not an object as sign, but an object which is stood for by a sign. This interpretation is more in line with the original Peirce. But Eco chooses to differ from Morris:

The only modification that I would introduce into Morris’s definition is that the interpretation by an interpreter, which would seem to characterize a sign, must be understood as a possible interpretation by a possible interpreter.

I don’t engage in a discussion on how closely Morris follows Peirce, or not.¹⁴ Eco’s “modification” underlines that he ignores peirce‘s attention to actual process instances of sign use. Eco’s theme is that of rules for processes of sign use. It is exactly why he is interested in possibilities, rather than instances. What Eco calls “[t]he only modification” is, in fact, a major shift of focus, i.e., from a philosophy of cognitive and social psychology to, say, a naive sociology.

Another example of Eco’s preoccupation with conventions for sign use is how he abstracts interpretants from the sign user. He suggests (p 68) that Peirce occasionally confuses the interpreter for the interpretant. But according to eco

[t]he interpretant is that which guarantees the validity of the sign, even in the absence of the interpreter.

This can only be understood as a statement about a repository of sign possibilities. What Eco seeks to establish is

the foundation of a semiotic system capable of checking itself entirely by its own means. Language would then be an auto-clarificatory system, or rather one which is clarified by successive systems of conventions that explain each other.

I am puzzled why Eco insists on recruiting Peirce into his service, even blaming him that he (p 199) “does not abandon the reference to objects.” What Peirce in fact proposes is that sign use is a process of a dynamic sequence of irreducible triads at the service of facilitating conduct. His fundamental insight remains unchanged even though in this treatise I have extended triad to hexad (see Chapter 2) and subsequently to ennead (see Chapter 4). That insight implies the conducting subject, of course. That subject may also be called sign user, interpreter, or whatever. But it may certainly not be removed from the equation.

When a subject’s conduct, or behavior, is guided to a large extent by his “scientific intelligence,” it may indeed be characterized to an equally large extent as the collection of his interpretants. In this way, it surely makes metaphorical sense to state that an interpreter is his interpretants. But then, subjects don’t count in Eco’s theory of codes. Suppose Connie Langue couldn’t care less about what Mary Parole might want to tell her daughter Suzan Parole. With Connie’s knowledge switched off, Mary and Suzan would be unable to communicate.

I have established a clear pattern of preoccupation in eco. In § 5.6 I add my comments with an emphasis on Eco’s concept of sign production. He continues to define semiotics in terms of a priori agreement on meaning. Or, rather, that is how he defines his semiotic assumptions. See also Eco (1968). In fairness to Eco, I acknowledge that his actual “explorations in the semiotics of texts” with their emphasis on interpretative “openness” (1959-1977) often contradict “the theory of codes.” My criticism is therefore especially valid at the axiomatic level where premature contradictions (see also § 9.1) are sown.

I conclude this paragraph of comparing interpretations with a sentence by Eco that is typical of how he usurps concepts for his own “theory” (p 70):

The idea of the interpretant makes a theory of signification a rigorous science of cultural phenomena, while detaching it from the metaphysics of the referent.

Isn’t Peirce already very clear on metaphysical truth and falsity? See also § 2.3, above.

 

 

5.5 application of the ennead

My critique of A Theory of Semiotics, in particular of Eco’s underlying assumptions for his theory of codes, draws suspicion to the quality of his proposal for modeling a sign vehicle’s meanings. As shown in Figure 5.2.3, he employs the categories of [1] denotation, [2] contextual selection, [3] circumstantial selection, and [4] connotation. Can the metapattern (see Chapter 4) be applied for the same purpose?

The essential difference might be that the metapattern takes objects seriously. But then, so does Eco although he doesn’t want to admit it for his theory of codes. He favors the terminology of “cultural unit.” Like object it is a fiction in Vaihinger’s sense, all the same. In fact, taking objects seriously amounts to recognizing their fictional nature. It makes cultural unit and object equivalent for all practical and theoretical purposes. So, upon closer inspection, the metapattern is not disqualified because it cannot handle cultural units. It can.

Another difference, this one more superficial, concerns the roots of the semantic trees. As Eco conceptualizes them, every tree has a sign vehicle for its root. Metapattern-based models all have a root, too. That root reflects recognition of the - very practical - boundary of objectified reality. The chosen sign for that boundary is a thick, unbroken line.

Especially relevant for the purposes of modeling meaning of signs is the situation in which objects function as signs. That specific situation may be called signship, or something similar. Then by definition, objects in the situation of signship all behave as signs. In Figure 5.5.1, the transition to short-hand indicators is shown, too.

Figure 5.5.1.
A separate situation of signship.

 

One way to proceed is to consider a sign’s meaning(s) as its direct properties. That would be the sign’s intext. Why not model it using Eco’s categories? This makes Figure 5.5.2 to a large extent the equivalent of Figure 5.2.3.

Figure 5.5.2.
Eco’s semantic markers in a metapattern-based model.

 

Figure 5.5.2 no doubt makes perfect sense from a limited semantic perspective. However, a shift to pragmatic recognition of objects, i.e., working from a belief in a subject’s objectified reality, immediately goes beyond Eco’s categories of meaning. The idea is to reverse the dynamics, from observation to sign engineering. The modeler’s task is then no longer to describe the meanings of a particular sign. Rather, he describes an object’s situational behavior. And he actively makes signs to do so. An example is Figure 4.6.2 which is repeated here as Figure 5.5.3.

Figure 5.5.3.
Revisiting the fictional case study of the previous chapter.

 

In Figure 5.5.3 (some) - positions for - names are added. As suggested in Figure 4.5.6, names are not used as signatures. That is, a name is not a whole object - that is, not the particular object that the attention is currently focused on - but only stands for it at some distance. It is ‘only’ considered a property of that object.

Eco pursues a closed semantics and tries to keep objects out of his framework. My view is that semantics should strongly relate to pragmatics. Apologizing for my banal terminology, surmounting unproductive linguistic compartments is only possible with an open semantics. It is open because it integrates the perspectives on both objects and signs. As explained before, my procedure is to model a sign that stands for an object as that same object’s property (or attribute, or characteristic). The integration of Figures 5.5.2 and 5.5.3 in Figure 5.5.4 shows how a predominantly sign-oriented situation coexists with situations in which (other) behavior of the ‘stood-for’ objects is considered relevant.

Figure 5.5.4.
Integrated model of sign behavior and name behavior(s).

 

Figure 5.5.4 combines the situation of signship and all other situations of non-signship into an integrated model. It must be clear that ultimately every detail shown in a model draws upon an object that, at the minimum, exists in the situation of signship as its precondition for naming elsewhere. First of all, the establishment of the separate situation of signship effectively puts a halt to infinite regression of signs … describing signs, etcetera.

Secondly, names - please note, in their broadest sense - are usually either commonplace or particular. Both ways, modeling signs in their own right for them is often not deemed necessary. My goal here is different. It is to show how the metapattern supports modeling of semantic networks such as Eco proposes. It requires that signship explicitly appears as a separate situation.

Given such an example of a - model of a - semantic network, the obvious question is: What happens to Eco’s categories? To make it easier to follow the argument, Figure 5.5.5 presents a model with instances, not types. As an example of a sign I take “Semiotics.” Further, I assume that “Semiotics” has several known meanings, for example in science, in publishing, in identifying persons, dito organizations. In all those fields (also read: situations) other than that of signs themselves, I regard this sign as what names an object. Why indeed not a mrs or mr Semiotics?

Figure 5.5.5.
Different, i.e., situational, name uses of the sign of “Semiotics.”

 

From the model of Figure 5.5.5 it is straightforward to ‘read’ the meanings of the sign “Semiotics.” Just follow any dotted line from the sign itself. These relationships solve, what Eco puts forward as (p 121)

the real problem […] that every semantic unit used in order to analyze a sememe is in its turn a sememe to be analyzed.

Earlier in A Theory of Semiotics his question is (p 112):

Is it possible to establish componential trees that take into account all coded contexts and circumstances?

Eco persists in hierarchies, only later (p 121) referring to the work of R.M. Quillian (1968). My answer is that the metapattern is suited to model an approximation. Metapattern-based models in fact don’t show independent trees. Trees may be interconnected at many nodes. As in a forest, the life of any one tree is influenced by its neighbors in a myriad of ways. And because every tree has neighbors, the whole forest is a system of interdependencies. This model structure is quite isomorphous to Quillian’s approach about which Eco recounts that (p 122)

[t]he configuration of the meaning of the lexeme is given by the multiplicity of its links with various tokens, each of which, however, becomes […] the patriarch of a new configuration.

What is different about the metapattern is that connections are possible from any node of one tree to any node of any other tree. It allows for increased variety and confirms De Saussure’s fundamental remark that (1916, p 107)

[t]he mechanism of a language turns entirely on identities and differences.

Then, to continue with De Saussure (p 115),

what we find, instead of ideas given in advance, are values emanating from a linguistic system.

De Saussure introduces the concept of value to emphasize the essence of both the arbitrary relationship between sign and meaning and the relational nature of meaning, i.e., that meaning is not so much independent but, rather, interdependent. Ultimately, meanings are systemic properties. A systematic model is therefore required for approximating the anatomy of meaning. For, as De Saussure remarks (p 118),

[i]n a linguistic state[,…] everything depends on relations. […] (p 80) [A] language is a system of pure values, determined by nothing else apart from the temporary state of its constituent elements. […] (p 81) Nowhere else [but in language] do we find comparable precision of values, or such a large number and diversity of terms involved, or such a strict mutual dependence between them.

For example take in Figure 5.5.5 the immediate neighbor in the model of the ‘original’ sign of “Semiotics.” Then as Figure 5.5.6 shows, step 1 leads to “Semiotics” as a name. Step 2 leads to the situational object to which “Semiotics” lends itself as name. Of course, step 2 does not lead to that object itself, only to what stands for it in the model. That representative of the object’s identity is the signature as a specific node in the network model. A particular signature (re)focuses the sign user, who then may reach an inference about a corresponding situational object. In this case, the object is the scientific discipline of semiotics.

Figure 5.5.6.
Stepwise focusing for establishing meaning(s).

 

In general, an open semantics such as the metapattern supports with its underlying ontology and characteristic visualization technique makes it possible to state meaning directly in terms of objects. Please note that “in terms of” should be read as “in sign standing for.”

There is really no problem with any truth, or with any falsity for that matter. All objects are essentially fictional, anyway. It simply follows from letting signs stand for them. An interpretant - which by definition arises from a sign - can therefore never directly correspond to an object. Because the conditional nature of objects is pervasive, it is no longer necessary to specify conditions. With sufficient awareness of their fictional character secured, objects may once again be told about in the manner of naive realism. Eco, for one, fails to grasp the axiomatic quality of Peirce’s fiction of the object. He doesn’t recognize the daily practice of naive realism against the philosophical background of pragmatism.

When Peirce is properly appreciated and the metapattern properly applied, any difference between denotation and connotation disappears. The intext of a signature describes properties as they are modeled as relevant for an object in a particular situation; the signature’s context stands for that situation. The unambiguous distinction between the concepts of situation and context implies that any difference between contextual selections and circumstantial selections is also overcome. In metapattern terms it is the context which represents the circumstances. When circumstances are taken as synonymous with situation, that is.

An example of a sign function to which Eco often returns is a measurement system informing a remote operator on the water level in a reservoir (p 32 and on). The operator reads off measurements. An action on his part may result to control the water level. Eco calls the measurement of the water level the denotation of the sign; its connotation is the required action. In general, he remarks that (p 86)

[d]enotation is the content of an expression, connotation the content of a sign-function.

The difference between these confusing categories of meaning disappears when behavior is differentiated. The model is then conceived as showing, through corresponding contexts, the situations in which the signal emitted by the measurement system performs. It constitutes a series of derivations as summarized in Figure 5.5.7.

Figure 5.5.7.
Situational differentiation preempts any need for (sub)categories of meaning.

 

 

5.6 a theory of repairs

Do I ridicule Eco’s theory of codes? I see no other way to present a serious synopsis without getting overwhelmed by his contradictions. On the basis of The Name of the Rose I even consider it a possibility that Eco means it all as a joke. Is he applying for example his own maxim on lying? Does my criticism finally find him out? Suppose he tries, indeed, to parody semiotics. Then I gladly acknowledge his success equals that of his great novel.

After the bias of closed semantics is sufficiently digested there is in fact much to enjoy in A Theory of Semiotics. For example, Eco is not at all blind to changes in the conventions he assumes to rule sign production. Throughout, he also writes about creative activities of individual sign users. Consistently, however, he places such phenomena outside semiotics.

I find Eco an excellent columnist. He taxes logical argument, though, every time he fits one of his small theories into his narrow preoccupation with overall conventions about the proper production of signs. Running into inconsistencies he irons those out by suggesting more theory, etcetera. What he needs is less, not more.¹⁵ He seeks to repair his theory of codes, not by rethinking its axioms, but by a baroque collection of additional hypotheses.¹⁶ In this paragraph I offer examples from his argument where he introduces - what I call - the individual sign user.

Where Eco places the boundary of semiotics is addressed in the following statement (p159):

Let us call semiotic a judgment which predicates of a given content (one or more cultural units) the semantic markers already attributed to it by a previous code.

The general problem with statements like these is that the label for the whole field reappears as the label for one of its categories. Are not all judgments within semiotics by definition … semiotic? Where is the background perspective necessary for rational explanation?

On p 163 Eco develops the view that

the use of an expression means that the semantic analysis of a given sememe establishes a list of semantic properties that should correspond to the supposedly extra-semiotic properties of an object.

He cannot untie this knot. The simple reason is, as I explained in Chapter 3 on ontology, that nothing can be untied “within its own framework.” Detachment (p 70) “from the metaphysics of the referent” only makes the knot all that tighter. The knot must be cut through from the outside, by altogether forgetting about the metaphysics (also read: axioms) of Eco’s closed semantics.

As Peirce postulates, nothing about signs is extrasemiotic. When dealing with signs, by definition they are irreducibly connected to objects and interpretants. It is impossible to remove any one of these fundamental concepts, and still practice semiotics.

peirce does not have a problem determining what sign-related phenomena belong to semiotics. They simply all belong. Eco faces the inevitable problem of explaining phenomena that have a mixed character as a consequence of his categories. He acknowledges (p 129) “the mobility of semantic space.” But how does an individual sign user get informed about the changed conditions for communicative success? Eco’s answer is that such uncertainty

imposes on the activity of sign production and text interpretation […] the necessity of a continuous extra-coding.

It is inefficient to consider something an exception when there exists a continuous requirement for it. So, if necessity is enduring, why is it extra rather than standard? But, first, what does Eco mean by it? He writes that

[t]he interpreter of a text is at the same time obliged to both challenge the existing codes and to advance interpretive hypotheses that work as a more comprehensive, tentative and prospective form of codification.

Then, it can happen

that the interpreter is obliged to recognize that the message does not rely on previous codes and yet that it must be understandable.

For Eco it constitutes a “border-line situation.” His label is apt when only what is conventional is accepted as intrasemiotic. Next, Eco proclaims as his profound discovery that

the very activity of sign production and interpretation nourishes and enriches the universe of codes.

I agree completely that individual sign users change conventions. As a continuous possibility, however, I consider change the rule, not the exception. A convention is merely the result of absence of change, and only for as long as it lasts. Eco keeps change from his assumptions. For him, therefore (p 249),

[t]he main problem arises when trying to determine how it is possible to map onto an expression continuum the properties of something which […] is not yet culturally known.

When assumptions deny the ordinary it then becomes a problem. But change ordinarily happens, with Eco himself providing the example of a painting. But that, he argues (p 250),

is not a semiotic phenomenon, because there is neither pre-established expression nor pre-established content. […] (p 252) [… T]he process is not an easy one; sometimes addressees refuse to collaborate, and consequently the convention fails to establish itself.

As usual Eco applies a label for both showing his recognition, and keeping such a phenomenon outside his limited semantic framework. Invention is how he calls the extrasemiotic forces that change the semantic field. Or aesthetic activities. For Peirce nothing is more essentially semiotic than precisely what Eco excludes. The latter is actually confusing semiosis in general and successful communication in particular, when concerning - his concept of - invention he writes (p 254):

The sender gambles on the possibility of semiosis, and loses.

What the sender, by definition of sending, i.e., by engineering a sign, cannot lose is semiosis. Fundamentally, sign engineering is semiosis. However, there may not be any receipt. Or the semiosis of the receiver may not yield the result as desired by the sign’s engineer.

And then, any sign observation is semiosis, too.¹⁷ Communication across sign users involves their - please note the plural - semioses. In successful communication, what happens might be modeled as some sort of correspondence between otherwise separate semioses, i.e., various processes of sign use. That is the theme of the anatomy of meaning (see Part ii). Again, change must be considered the rule, resulting inevitably in a time series of states.¹⁸

De Saussure is already very clear about the necessity of theorizing along two axes (1916, p 80). They are simultaneity and succession, respectively. He refers to them as synchrony and diachrony. De Saussure is adamant they must not be confused. Even (p 83),

[t]he opposition between these two orders[, i.e., of synchrony and diachrony, respectively] must be grasped in order to draw out the consequences it implies.

Eco does not master the application of these axes. He tries to explain diachronic events from a synchronic perspective. As synchrony is convention, to him anything diachronic is unconventional. And because he limits semiotics to convention, whatever changes convention is extrasemiotic. De Saussure explicitly includes both axes.¹⁹ Peirce points to the sign user’s experience of uncertainty. It all amounts to acceptance of change as the rule, rather than the awkward exception it is with Eco. Change is fundamental for semiotics for every time a sign user establishes a belief - or a doubt - in response to a sign, his intelligence changes by definition. The more uncertainty a sign user experiences, the more changes occur (and the more change occurs).

Returning to Eco’s treatment of change, he actually places it outside his concern because (p 130)

in some other cases there are (besides the possible contexts which can be foreseen but cannot be coded) possible circumstances which are either unforeseeable or excessively complex and which make up a cluster of different extra-semiotic factors. In all these cases one is entitled to speak of extra-semiotic and uncoded determinants of the interpretation.

Once again I agree with a part of what Eco argues. Yes, often a sign lacks a priori possibilities of interpretation. Recognizing this need for creativity on the part of the sign user, however, is definitely not the important achievement of his theory of codes which Eco claims for it (p 129):

The theory of codes explains how one possesses rules of competence that permits one to disambiguate or to overambiguate, to form and to interpret given messages or texts.

My first question is how one can form something that is given. When it is a slip of the pen, the overall claim is clearly overstated. Eco’s theory explains nothing of the sort. All he does for it are repairs by somewhat loosening the rein on his postulate of intrasemiotics.²⁰ Extracoding - of which he distinguishes the forms of overcoding and undercoding - is Eco’s attempt to restore sufficient freedom to the individual sign user for change to conventions to happen. A theory of codes cannot be repaired in this half-hearted way so that a theory of semiotics may result. Semiotics needs a radical orientation at the freedom of the individual sign user.

Surely, freedom is often severely obstructed in the sign user’s daily life (see also Chapters 7 and 8). Understanding obstructions is nevertheless vital for the background perspective required for any encompassing anatomy of meaning. Here, I am particularly concerned about the claim that - preferences for - conditions or even prescriptions for successful communication are adequate assumptions for scientific explanation (see also Chapter 9 through 12). What Eco’s assumptions lead him to is to remark that (p 188)

we have a paradoxical situation, in which expression must be established according to a content model which does not yet exist as such.

It is a paradox of his own making because he apparently fears what he believes is the paradox of (p 178)

the presence of the referent as a discriminant parameter, a situation which is not permitted by the theory of codes proposed in [my] book.

What somebody proposes to permit, and what not, might very well be the outcome of a serious, elegantly argued theory. However, it can never be both a serious start to theorizing, and its result. Eco is therefore not in the position to be taken seriously with his advice on any permission. Actually, he has a glimmer of awareness of his zero-variable enterprise because (p 139)

the message (or the text) appear[s] as an empty form to which can be attributed various possible senses.

This places the individual sign user at the start of theory about meaning. Regarding meaning shared between individuals, my critique of A Theory of Semiotics shows that it cannot be explained properly on the basis of semantic conventions alone.

 

 

5.7 an anticipatory critique

If there is, neutrally speaking, one thing this chapter makes clear it is that assumptions vary widely. The assumptions from which Eco theorizes guide mainstream²¹ semiotics and linguistics as it has been ‘academized’ by Morris.²²

In Marxism and the Philosophy of Language (1929) Voloshinov devotes much space to criticizing what he considers the current (p 47) “two basic trends [... i]n the philosophy of language and in the related methodological sectors of general linguistics:”

[T]he problem of language [...] is to bring this whole multifarious system of features and relations, of processes and artifacts [of the organized social milieu into which we have included our complex and the immediate social communicative situation], to one common denominator: all its various lines must be channeled to one center-to the focal point of the language process.

According to Voloshinov (p 48), the main trends in futile pursuit of a solution are [1] individualistic subjectivism and [2] abstract objectivism. He summarizes (p 52):

If, for the first trend, language is an ever-flowing stream of speech acts in which nothing remains fixed and identical to itself, then, for the second trend, language is the stationary rainbow arched over that stream.

Especially relevant here is Voloshinov’s criticism of abstract objectivism which he considers the linguistic equivalent of logical positivism. It is relevant because Eco’s semiotic theory, especially his theory of codes, perfectly fits abstract objectivism’s mold. The criticism by Voloshinov, even though he makes it decades earlier, is therefore also valid for A Theory of Semiotics. Some quotations from Voloshinov (1929), selected here especially for the purpose of my critical synopsis of Eco’s semiotics, illuminate axiomatic differences. Voloshinov first describes and subsequently criticizes abstract objectivism:

[p 53] Language stands before the individual as an inviolable, incontestable norm which the individual, for his part, can only accept.
[p 54 T]here is only one linguistic criterion: correct versus incorrect, wherein linguistically correct is understood to mean only the correspondence of a given form to the normative system of language.
[p 56] The present state of a language and the history of a language do not enter into and are incapable of entering into mutual comprehensibility.
[p 57] What interests the mathematically [p 58] minded rationalists is not the relationship of the sign to the actual reality it reflects nor to the individual who is its originator, but the relationship of sign to sign within a closed system already accepted and authorized. In other words, they are interested only in the inner logic of the system of signs itself, taken, as in algebra, completely independently of the ideological meanings that give the signs their content. Rationalists are not averse to taking the understander’s viewpoint into account, but are least of all inclined to consider that of the speaker, as the subject expressing his own inner life.
[p 58] It should be noted [...] that the precedence of the understander’s viewpoint over the speaker’s has remained a constant feature of [abstract objectivism]. This means that on the basis of this trend, there is no access to the problem of expression nor, consequently, to the problem of the verbal generation of thought and the subjective psyche (one of the fundamental problems of [individualistic subjectivism]).
[p 67] Most representatives of abstract objectivism are inclined to assert the unmediated reality, the unmediated objectivity of language as a system of normatively identical forms.
[p 82 Abstract objectivism] leads us away from the living, dynamic reality of language and its social functions, notwithstanding the fact that adherents of abstract objectivism claim sociological significance for their point of view.

Having outlined abstract objectivism, Voloshinov contrasts it with his own preferences. Again, I select some quotations that strongly support my argument (and that show Voloshinov precedes my theoretical development on several aspects). How my fundamental position is different from Voloshinov’s is, as I indicated earlier, taken up in the next chapter.

[p 67 I]s the mode of being of language in the subjective speech consciousness really what abstract objectivism says it is? We must answer this question in the negative. [...] For [the speaker], the center of gravity lies not in [p 68] the identity of the form but in that new and concrete meaning it acquires in the particular context. What the speaker values is not that aspect of the form which is invariably identical in all instances of its usage, despite the nature of those instances, but that aspect of the linguistic form because of which it can figure in the given, concrete context, because of which it becomes a sign adequate to the conditions of the given, concrete situation.
[p 68] The basic task of understanding does not at all [...] amount to recognizing the form used, but rather to understanding it in a particular, concrete context, to understanding its meaning in a particular utterance, i.e., it amounts to understanding its novelty and not to recognizing its identity.
[p 69 T]hus the constituent factor for the linguistic form, as for the sign, is not at all its self-identity as signal but its specific variability[.]
[p 70] The linguistic consciousness of the speaker and of the listener-understander, in the practical business of living speech, is not at all concerned with the abstract system of normatively identical forms of language, but with language-speech in the sense of the aggregate of possible contexts of usage for a particular linguistic form. [... A] member of a language community does not normally feel himself under the pressure of incontestable linguistic norms. [...] Words are always filled with content and meaning drawn from behavior or ideology. That is the way we understand words, and we can respond only to words that engage us behaviorally or ideologically. [... p 71] The divorce of language from its ideological impletion is one of abstract objectivism’s most serious errors.

Voloshinov is so surprised by abstract objectivism’s popularity that he seeks to explain why it has risen to such prominence. In fact, his explanation resembles De Saussure’s criticism of earlier generations of linguists.²³ Voloshinov writes:

[p 71] At the basis of the modes of linguistic thought that lead to the postulation of language as a system of normatively identical forms lies a practical and theoretical focus of attention on the study of defunct, alien languages preserved in written monuments. This philological orientation [of ...] European linguistic thought formed and matured over concern with the cadavers of written languages; almost all its basic categories, its basic approaches and techniques were worked out in the process of reviving these cadavers. Philologism [...] lacked the range necessary for mastering living speech as actually and continuously generated.
[p 78] Linguistics [...] is oriented toward the isolated, monologic utterance. [...] Research is wholly taken up in study of immanent connections on the inside territory of the utterance.

Eco’s semiotic theory provides a clear example of how trying to remain “on the inside territory” leads to irrepairable contradictions. He certainly recognizes that multiplicity of meanings occurs. However, he does not switch to an essentially behavioral theory (Peirce: pragmatism, Voloshinov: ideology) for - more - inclusive explanations. For Voloshinov multiplicity is not an awkward exception, but the phenomenon he needs to address first and foremost:

[p 80] For abstract objectivism, the unity factor of a word solidifies, as it were, and breaks away from the fundamental multiplicity of its meanings. This multiplicity is perceived as the occasional overtones of a single hard-and-fast meaning.
This process of isolating a word and fixing its meaning outside any context [... is] further complicated by the fact that [the linguist] creates the fiction of a single and actual object corresponding to the given word. This object, being single and self-identical, is just what ensures the unity of meaning.

Eco does not exclude a concept such as context. He views homonymy as a linguistic puzzle, though, rather than as a phenomenon that requires behavioral criteria to eliminate ambiguity. Voloshinov (p 80) comments as if anticipating A Theory of Semiotics:

Another grave error on the part of abstract objectivism is [that t]he various contexts of usage for any particular word are conceived of as all lying on the same plane. These contexts are thought of as forming a series of circumscribed, self-contained utterances all pointed in the same direction. In actual fact, this is far from true: contexts of usage for one and the same word often contrast with one another. The classical instance of such contrasting contexts of usage for one and the same word is found in dialogue.

My contribution to increased clarity is that the multiplicity of meanings of context is somewhat ordered. Extending first the triad and next the hexad, the ennead differentiates between situation and context. I read Voloshinov as writing about both situations and contexts in their integrated enneadic sense. A context as a circumscription clearly is meant along the sign dimension. But a context for usage is a situation, i.e., a concept along the reality dimension. I propose that context can often more productively be read as situation in Voloshinov. Take, for example (p 80):

Of course, dialogue is only the most graphic and obvious instance of varidirectional contexts. Actually, any real utterance, in one way or another or to one degree or another, makes a statement of agreement with or a negation of something. Contexts do not stand side by side in a row, as if unaware of one another, but are in a state of constant tension, or incessant interaction and conflict.
[p 81 I]t is precisely a word’s multiaccentuality that makes it a living thing. The problem of multiaccentuality ought to be closely associated with the problem of multiplicity of meanings.

At the core of an understanding of multiplicity is therefore recognition of situational variety. As Voloshinov already underlines, the attribution of variety should not be limited to the observer of the sign. The sign engineer equally experiences “varidirectional [situations].”

How, from a rational perspective, any sign only imperfectly stands for situational behavior of objects is discussed in the next chapter. There I introduce Schopenhauer’s concept of the will. His insight is that the intellect, including the faculty of reason, is ‘only’ an instrument of the will. Applied as a boundary concept, just like Schopenhauer does for his own purposes, the concept of the will completes the picture that Part i sketches of the individual sign user.

Part ii proceeds with the anatomy of meaning. With engineer and observer united in an instance of sign exchange, it is radically dialogical. It calls for several more quotations from the work of Voloshinov (1929).

 

 

5.8 fear of idealism

In A Theory of Semiotics Eco emphatically renounces the label idealist for his work and, by implication, for himself. “Semiotics fully avoids any risk of idealism” (1976, p 317), he writes. Why? Because it is limited to “the social existence of the universe of signification.” All that lies “beyond the semiotic threshold” Eco leaves for its (p 316) “verification […] to other types of approach.”

Eco fails to admit that he is merely passing on - what he considers - the problem of an idealist position. He does not solve it, at all. It is obvious that, next, such necessarily complementary approaches must come to terms with the “threshold” he has preset for his private brand of semiotics. So, he does not own up to the conclusion that such boundaries are not at all absolute but, at the minimum, correlate with idealist conjectures of (p 315) “individual material subjects” made from a broader perspective. Indeed, by implication, his theory of codes - which is what his semiotics-as-theory actually amounts to - is associated with strong idealist assumptions (also read for assumptions: conceptual grounds).

Taking as another example a thinker who has already been introduced in this treatise, such dependency is not different with Peirce. The label that he is most often fitted with is that of pragmatist. What happens is that usually only a part of his work is highlighted, in this case his philosophy on the ground of conduct. What the label of pragmatism misses, then, are related parts of Peirce’s work. I especially mean his semiotic. When considering only that other part of his work, i.e., his semiotic theory, it would actually be quite logical to call Peirce an transcendental idealist.

As I have shown in Chapter 2, Peirce’s pragmatism, and his semiotic, respectively, can only be properly understood when one is seen against the background of the other. It is appropriate to state that at least²⁴ for pragmatism and semiotic Peirce designs an encompassing system of concepts. Merging labels, his system may be called idealist pragmatism, or pragmatic idealism. On his conceptual development, R. Collins remarks (1998, p 676):

Peirce absorbed metaphysics into logic, producing his own semiotic Idealism.

However, I am not concerned here with improving the label for Peirce and his work. For my theoretical development I emphasize, as I already did in Chapter 3, that idealism is a respectable approach for philosophical explanation, and for explanation in general. But any idealism should always be balanced by a corresponding realism, and the other way around. That is precisely what Peirce’s triadic conceptual scheme prescribes: both idealism and realism are transcended. As it were, they hinge on signs and are thereby irreducibly integrated.

 

 

notes

1. I use “semiotic” and “semiotics” intermittingly in this treatise. As semiotic is Peirce’s preferred term, I use it when dealing directly with his ideas.

2. My presentation of these ideas of De Saussure’s is extremely succinct. I recommend his book Course in General Linguistics (1916). The reader will recognize many themes that have inspired a wide variety of intellectual/scientific developments.

3. Sharing meaning is guaranteed by sharing - the same - language. Of course, that could be taken as the explaining factor. But, then, the question is how it could be ascertained that, indeed, different persons ‘use’ exactly the same language.

4. Eco shows the habit of labeling something as a “theory” when he does not go into specifying the “background perspective” (Wicklund, 1990). For example, only mentions appear of “a theory of referents” (p viii), “a theory of mentions” (p 58), “a theory of contextual and circumstantial selection” (p 110), “a theory of settings” (p 110), “a theory of code-changing” (p 152), “a theory of the extensional verification of correspondences between propositions and states of the world” (p 156), “a theory of the relationship sender-addressee (p 314), and “a theory of text-creativity” (p 315). He confuses raising the question with providing the answer.

5. De Saussure starts his book Course in General Linguistics with (1916, p 1) “a brief survey of the history of linguistics.” Eco’s approach seems very similar to what De Saussure mentions as the first phase: “Grammar aims solely at providing rules which distinguish between correct and incorrect forms. It is a prescriptive discipline, far removed from any concern with impartial observation, and its outlook is invariably a narrow one.” Eco does not maintain such a narrow prescription. But he sees codes-as-grammar as the rule, and anything that does not conform to them as exceptions. A similar orientation - and, like Eco, not really in the usually relevant conclusions he draws, but in his assumptions - demonstrates W. Nöth in Dynamik semiotischer Systeme (1977). Nöth argues that (p 3) the accepted concept of semiotics is that of a general science of sign systems, or (p 1) codes.
At many places, De Saussure already shows his appreciation of the vital role of the language user. He writes, for example (p 7): “In the final analysis, where languages are concerned everything has its psychological aspect.” And (p 9): “Language has an individual aspect and a social aspect. One is not conceivable without the other. […] The structure of a language is a social product of our language faculty. At the same time, it is also a body of necessary conventions adopted by a society to enable members of society to use their language faculty.” He asks (p 13): “What is the origin of this social crystallization?” The answer is that “[t]he individual’s receptive and coordinating faculties build up a stock of imprints which turn out to be for all practical purposes the same as the next person’s.” Like Eco after him, De Saussure does not yet concern himself with the psychological nature of such same imprints (also read: interpretants), and with questions how they become possible or what keeps them from occurring, and what is an optimal level of sameness, anyway? On p 89 he writes that static linguistics, or “[s]ynchrony has only one perspective, that of the language users; and its whole method consists of collecting evidence from them.” And (p 110): “The characteristic role of a language […] is to act as intermediary between thought and sound.” That is of course the thought of the sign user.

6. On p 105 Eco writes down instructions. Certainly, this is how Connie Langue would use her own knowledge network. But how does one selection come to be classified as contextual, and another one as circumstantial?

7. The quotation is from Feyerabend, of course. See § 3.4. And see note 7 in Chapter 2 for a quotation from Peirce on the amplitude of his concept of object.

8. On p 7 Eco adds that “[t]his something else does not necessarily have to exist or to actually be somewhere at the moment in which a sign stands in for it.” Then follows, as a conclusion, the sentence which contains the characteristic that Eco himself often presents to summarize his theory: “Thus semiotics is in principle the discipline studying everything which can be used in order to lie.” The problem, once more, is that Eco is confusing categories. A lie is only possible within a framework that - also - recognizes truth. But nowhere, and I agree with that, does Eco make provisions for truth statements. It simply means that he also cannot use the concept of lying to promote his theory.

9. H. Putnam commits the same reduction in Representation and Reality where he assumes that (1988, p 25) “reference is socially fixed and not determined by conditions or objects in individual brains/minds. Looking inside the brain for the reference of our words is [...] just looking in the wrong place.”
So-called rationalists don’t recognize anything about an individual person that lies beyond the mind. For the faculty of human reason occupies the most elevated position in their concept of personal order.
Chapter 6 of this treatise establishes the view of Schopenhauer. He argues that the intellect (also read: mind) is subservient to the will. Reference is therefore not “socially fixed” but hinges on the individual person who, as Schopenhauer suggests, is a unique objectification of the will.

10. A dilemma for Eco is that either his book A Theory of Semiotics is not new because he could at all write it, or that his theory is flawed precisely because he has written an original book.

11. In contrast, De Saussure offers a refreshingly accessible account of fundamental arguments for delimiting distinct areas of scientific activity. See also my recommendation in note 2, above.

12. So, where is the code?

13. I also believe that Eco doesn’t do justice to De Saussure where he treats his work. Again, our readings seem to differ greatly.
With respect to Peirce, I have encountered another influential interpretation that does not seem to do him justice, not from my perspective, anyway. In Signs: An Introduction to Semiotics (1994), Thomas A. Sebeok (1920-2001) has collected several essays. Please note that Sebeok is considered “the major force in international semiotics” during the second half of the twentieth century (Cobley and Jansz, 1997, p 119-120): “It is largely by dint of [an] administrative profile set up by Sebeok that the term ‘semiotics’ has superseded ‘semiology’ on both sides of the Atlantic.” That he should have worked at that terminology shift I find quite surprising, for all I can make out is that his own theory is essentially grounded in the dualistic Saussurean concept of the static sign as constituted by signifier and signified (1994, p 17). Sebeok does not apply his earlier introduction (p 5) of Peirce’s concept of dynamic semiosis. Or when he does, it is in a manner that I do not consider faithful to Peirce. For Sebeok immediately reduces the Peircean object inside the triadic dynamics of semiosis (p 12), “so that the initial twofold distinction [between object and sign] is resolved to one between two sorts of signs.” I really don’t believe that such a reduction is what Peirce has in mind, at all. But Sebeok’s turn allows him to stress the importance of his field of semiology/semiotics (p 14): “If objects are signs, in indefinite regression to a supposititious logos [logos is a concept that Sebeok credits to Heraclitus as meaning ‘a reality behind signs’ (p 12); supposititious would mean that the concept of logos is even fraudulently inserted], and if interpretants are signs marching in progression toward the ultimate disintegration of mind, what is there left that is not a sign?” At least Sebeok openly admits that such a conclusion labels him a pure idealist in philosophical terms (p 14). Now Peirce certainly believes that signs are pervasive but they are nevertheless only a means to represent beyond them. It is in the nature of human knowledge faculties, including perception, that by definition the ‘beyond,’ cannot be known directly. As object, reality however deserves a separate conceptual position, i.e., apart from sign (and interpretant). I agree with Peirce’s transcendental idealism.
What, as a secondary area of interest, comes out clearly from Sebeok’s work is the confusion arising from taking different sign types too seriously. For this emphasis in modern semiotics Peirce is no doubt largely to blame. Sebeok tries to control the worst misunderstandings by arguing (p 21) “that it is not signs that are actually being classified, but more precisely, aspects of signs.” The classification is of course without added value when every sign is subsequently discovered to show more or less all aspects. My approach is to forget about such a general classification and concentrate, instead, on actual sign use. The anatomy of meaning developed later in Chapters 7 and 8 of this treatise may fruitfully be deployed to combat the confusion caused by Peircean sign types. The fundamental contribution by Peirce is his exposition of the irreducibly triadic, and dynamic, nature of semiosis.
However, attempts at reduction have been persistent. In Peirce’s Concept of Sign (1973) D. Greenlee produces the objectivist turn that seems to originate with Morris (see note 13, below). He summarizes (p 9): “As I understand the sign, the sign-interpretant relation becomes more important than the relation Peirce makes the most of, the sign-object relation.” Greenlee effectively contradicts the realist emphasis that Peirce includes in his triad. Much like Eco, it reads: “The concept of the sign which I have arrived at [...] is the concept of something that is interpreted according to a rule or a convention of interpretation, rather than the concept of something which stands for something else.” I find it peculiar, embarrassing even, to discover how such authors still claim Peirce’s heritage.

14. Morris exerts a strong influence on - establishing - semiotics as an academic discipline (W. Nöth, 1990). Peirce is an academic outsider during his own lifetime. Morris (who also edits Self, Mind, & Society by G.H. Mead; see Chapter 11) helps to popularize Peirce’s writings but in his own work essentially reduces the triadic unity into separate branches: syntax, semantics, and pragmatics. This reduction simplifies study but it comes at the expense of contradictions. It is now increasingly recognized that syntax, semantics, and pragmatics cannot be independently understood. That simply follows from the irreducibility of the Peircean triad.
Authors referring to Peirce are often actually more directly influenced by Morris. Examples are Sebeok (see note 12, above), Eco (see this chapter), and Nauta (see note 3 in Chapter 13). An analysis of publications by Morris himself has been kept outside the scope of this treatise. He outlines his ideas on “a comprehensive and fruitful science of signs” in Signs, Language and Behavior (1946, p v) which carries the subtitle An Original, Important Contribution to Semantics. He expands his treatment in Signification & Significance, a study of the relations of signs and values (1964). There, Morris remarks that (p 15) “[t]he distinguishing feature of work in semiotic in recent years has been the extension of interest into the diversity of dimensions of signification and into the variety of uses which sign performs. Earlier in the century, philosophers were concerned mainly with the designative and formative dimensions of signification as they occurred in science and mathematics. This concern remains, but it has been supplemented by a growing interest in the place that signs have in the manipulatory and consummatory phases of action.”
In Elements of Semiology, R. Barthes (1915-1980) indicates the development of the concept of value in De Saussure (1964, p54): “[H]e increasingly concentrated on it, and value became an essential concept for him, and eventually more important than that of signification.” As Barthes remarks: “We must [...] tackle the sign, no longer by way of its ‘composition’, but of its ‘setting’: this is the problem of value.” My anatomy of meaning radically subsumes (see Morris) “the diversity of dimensions of signification” under a single purpose, i.e., to request compliance with interests (see Chapters 7 and 8).

15 Again, I express my conviction that De Saussure provides an excellent example of a balanced, explicitly stated axiomatic system. But the problem of his current celebrity status in linguistic circles probably is that nobody actually studies his original work.

16. Eco continues to exhibit his remarkable theorizing flexibility in Semiotics and the Philosophy of Language (1984). At first it seems he reaches an opposite perspective for he writes (p 1): “The concept of sign must be disentangled from its trivial identification of coded equivalence and identity; the semiotic process of interpretation is present at the very core of the concept of sign.” It does not turn out, however, as the radical departure from A Theory of Semiotics that it suggests. A sentence earlier in Semiotics and the Philosophy of Language, Eco already maintains that “there is no opposition between the ‘nomadism’ of semiosis (and of interpretive activity) and the alleged stiffness and immobility of the sign.” As usual, he only pretends to honor a different perspective. He borrows its reputation, and especially its author’s, for his own, indeed opposite, cause which he pursues without noticeable deflection. Just mentioning that no opposition exists, of course does not make it disappear. His obscure formulations to that extent should not be mistaken for tight arguments.
I myself am all in favor of poetic license in scientific texts. However, its goal should always be to support arguments, not hide or even avoid them. A both convincing and humorous critique of obscurity in reputedly scientific works is given by A. Sokal and J. Bricmont in Intellectual Impostures (1997).
Though concentrating my criticism on A Theory of Semiotics, I wish to acknowledge that Eco subsequently changes his ideas somewhat. Rather than commenting on his complete, extensive œuvre, I quote from Kant and the Platypus (1997) where he explains why he never has revised his earlier book (p 4): “[E]very time I thought of putting my hand to A Theory of Semiotics again, I wondered if I shouldn’t have restructured it starting from the second part.” I agree, especially when that would mean omitting the first part altogether. But then he would probably still put too much emphasis on his favorite subject of interpretation, i.e., on sign observation. What I propose instead is a process view of sign exchange, with both engineer and observer as participants. This anatomy of meaning is presented here in Chapters 7 and 8.

17. The distinction between engineer and observer, in this treatise developed from Peirce’s mention of experimentalist (see § 3.1, above), is already made by De Saussure (1916, p 13): “[I]n the psychological part localized in the brain, one may call everything which is active ‘executive’ […], and everything which is passive ‘receptive’ […].” What he means is that executive behavior transforms a concept into a sign, and that receptive behavior works in the reverse direction, that is, it transforms a sign into a concept.

18. The subtitle of my book Metapattern (2001) reads: context and time in information models. I choose to highlight time, too. For time-factoring information at the most finely grained level of meaning to relevant sign users (also read: stakeholders) makes it possible to consistently record each and every event. With - information on - all events throughout time available, it is then possible to derive - information on - the state at any point in time. As such, the metapattern is a pure application of De Saussure’s distinction between synchrony and diachrony. It provides for the explicitly manageable opportunity to shift attention from one perspective to the other.

19. De Saussure does not use the word semiotics. He coins semiology for (1916, p 15) “a science which studies the role of signs as part of social life.” He adds that “[i]t would form part of social psychology, and hence of general psychology.” Peirce does not arrive at his semiotic from - a new paradigm for - linguistics but from philosophy. And because his central concept is that of conduct, he was in fact already practicing psychology, both social and cognitive.

20. Eco does not use the term intra-semiotics. It is my invention, intended to highlight all he places outside of his narrowly semantic concept of semiotics.

21. It has been, and still is, mainstream academic science to such an extent that Wold feels compelled to include the term alternative in the title of the book The Dialogical Alternative (1993, editor).

22. See note 13, above, for some remarks on the influence of Morris.

23. One point where I therefore don’t agree with Voloshinov is precisely that he classifies De Saussure as an abstract objectivist. As I try to demonstrate in this chapter, De Saussure’s view on the psychological nature of language is more balanced than he is often credited with.

24. Commentators such as T.A. Goudge (1950) also point out discrepancies in Peirce’s thought.

 

 

2002, web edition 2005 © Pieter Wisse

 

 

table of contents; previous chapter; corresponding prelude; next chapter.