The thread of subjectivity

Innis, Robert E. 1988. The thread of subjectivity: philosophical remarks on Bühler's language theory. In: Eschbach, Achim (ed.), Karl Bühler's theory of languages: Proceedings of the Conferences Held at Kirchberg, August 26, 1984 and Essen, November 21-24, 1984. Amsterdam: John Benjamins Publishing, 77-106.

Making very few claims to originality or uniqueness, Bühler continually defined his own position, as figure, over against a vast set of other positions, as ground. Early on in Sprachtheorie he insisted that many of his main themes and pivotal categories were derived from the philosophical, as well as the linguistic, tradition. But his own contribution was to weld them together into a unity by primarily supplying an ordering matrix and an interpretative framework. (Innis 1988: 77)

I imagine that's what good writers/thinkers do.

Perhaps the most attractive part of Bühler's work for me has been the peculiar fusion he managed to make between the objective (essentially social) and the subjective factors in language theory specifically and in the genesis of sense and meaning quite generally. (Innis 1988: 78)

Bühler did use philosophical terms such as subject for the sender, addresser or communicator and object for the referent.

The picture of language that informs Bühler's work is the well-known one of language as an instrument or organon for the effecting of semiotic exchange, that is, for the exchange of signs and meaning in social life. The representational function of language, its capacity to articulate, objectify, and transmit information about objects and states of affirs between a speaker and a hearer is rooted in a social, public, and communicative matrix. This is a point Bühler, with an admittedly rather backhanded reference to Plato's Cratylus, put right at the beginning of his 'Axiomatix der Sprachwissenschaften' (1933; see Innis, 1982). The communicatively oriented tool-character of language goes directly against the standard philosophical concern with language as a primarily individual and subjcetively oriented 'representational device.' Language is, to be sure, a representational device (hence the subtitle of Sprachtheorie), but is use by any linguistic subject for purely cognitional ends is a derivative from an original communicative goal pursued in a social field of shared actions, tasks, goals, and existential commitments. In a way paralleling Vološinov's treatment of the primary unit of a philosophy of language, Bühler saw 'verbal interaction' as the originative context for the investigation of language forms and structures. (Innis 1988: 79)

"Language, in its primitive function, to be regarded as a mode of action, rather than as a countersign of thougt." (Malinowski 1946[1923]: 296) And: "Statements about language as a tool, instrument, vehicle, etc., can be found in any textbook, but, strange as it seems, the apparently self-evident interference from this truism was not drawn in the linguistic tradition of the last century." (Jakobson 1971[1962b]: 523) And more: "When speaking of language as a communicative tool, one must remember that its primary role, interpersonal communication, which bridges space, is supplemented by a no less important function which may be characterized as intrapersonal communication." (Jakobson 1985[1973]: 98) In this last iteration it would appear that autocommunication (intrapersonal communication) is identified with the using language as a representational device. In the Bühler-Jakobson combo model code (language) is indeed between the subject (addresser) and object (referent) and the other components (factors) such as the message, the receiver and the channel seem superfluous in this connection. I think I just figured out how to tie Jakobson's communication model with the theory of autocommunication (especially with Peirce's version of it).

While Bühler would certainly admit, with Cassirer, that language is a 'symbolic form,' part of that vast set of artificial prostheses that mediate between and constitute meanings for human beings, it is not for him first of all a form oriented toward a non-praxical and speculative 'mirroring' of the world, as philosophers like to talk about it, but rather toward communicating some existentially significant surplus of perception that one member of the group might be in possession of in the context of shared actions embedded in shared times and spaces. (Innis 1988: 79)

This definition of symolic forms (as artificial prostheses) can be read as a definition of sign systems. Similarly, the context of shared actions embedded in shared times and spaces could be read as a definition of the channel. In this last connection it is possible - like Jakobson actually did in talking about bridging either time of space - to talk about channels that bridge space or shared space such as face-to-face interaction and channels that bridge time or shared time such as instant messaging.

With his fundamentally Aristotelian notion of language as action (Handlung, Aristotelian praxis) Bühler wanted to put language right back into social life and he thematized, in relatively formal and abstract terms, to be sure, its coordinating and 'steering function.' This 'cybernetic orientation' has been pointed out by Gerold Ungeheuer (1967) and in fact remained a dominent theme in Bühler's work until the end of his life (see Bühler, 1969). (Innis 1988: 80)

Silvi Salupere is often so ready to jump the gun and refer things back to Plato or Aristotle. Correctly so, it would seem. Referene to this interesting connection with cybernetics: Ungeheuer, Gerold (1967): Die kybernetische Grundlage der Sprachtheorie von Karl Bühler. In: To Honor Roman Jakobson. Essays on the Occasion of his Seventieth Birthday 11 of October 1966. 3 vols. The Hague. Paris: vol. 3:2067-2086. Reprinted in G. Ungeheuer, Sprache und Kommunikation. Hamburg, 1972.

Although I cannot go into it here, the closest major analogue to Bühler's position on the social matrices of sense and meaning in the modern philosophical realm is Mead's (see Günther, 1968) and Morris's symbolic behaviorism and the cognate work of Vygotsky (1934) and Vološinov (1929) with their insistence on the sociality of subjectivity and subjectivity of sociality. There is also a clear affinity here (and numerous other places, too) with the later work of Wittgenstein. (Innis 1988: 80)

Good man. This is the first I've seen of even a suggestion of possible affinity between Bühler and Morris.

It is in this sense that Bühler can say that phonemes as types are actually closer to conceptual elements than to sense qualities (Bühler, 1934: 289) for they are abstract structures, Gebilde, upon which the individual minds or consciousnesses in a community depend and to which they are subject. Bühler schematized this relationship by means of a famous diagram in Sprachtheorie (227):

The figure on left represents the total flatus vocis. The shaded part within it represents its relevant phonological moments, which compose the differential aspects of the linguistic sign. The figure on the right represents the 'object-domain' on which the linguistic sign bears. The shaded part within it represents only its relevant moments, which, Buhler's extrapolation, are apprehended by process akin to those by which we grasp the phonemic structure of linguistic signs. (Innis 1988: 85)

I dislike the whole principle of abstractive relevance (especially in de Saussure because of its enormous influence on philosophers who obfuscate everything) because I'm not so sure that this is as important for general semiotics as some (like Massimo Leone) would argue. That is, I'm not sure if or how one would proceed with this principle in studying nonverbal communication (e.g. the case of Ray Birdwhistell's kinesics, which got quite tangled up with this principle). Though I find this scheme much more to my liking than Saussure's signifier/signified scheme which lacks this aspect "relevant phonological moments" (as far as I know - having avoided de Saussure I have no say in the matter).

Now language for Bühler is not just a tool for the intersubjective exchange of information and meanings but also a tool for bringing experience under control, a theme also dear to Vygotsky's an Vološinov's hearts. (Innis 1988: 89)

And also in a way dear to my heart. I would subsume this aspect under the notion of "regulation".

A slightly different position has been put forward by Gestalt theory, as in the following text from Wolfgang Köhler's Gestalt Psychology:

Gestalt psychology holds [that] sensory units have acquired names, have become richly symbolic, and are now known to have certain practical uses, while nevertheless tehy have existed as units before any of these further facts were added. Gestalt psychology claims that it is precisely the original segregation of circumscribed wholes which makes it possible for the sensory world to appear so utterly imbued with meaning to the adult; for in its gradual entrance into the sensory field, meaning follows the lines drawn by natural organization; it usually enters into segregated wholes. (Köhler, 1959: 82).

However, we will often see the segregated unit differently because the lines drawn in or superimposed upon the social field of perception by the linguistic sign and its concomitant system of semantic markers functioning like a coordinate system are projected and not really determined by the lines of the 'natural' object itself. 'Knuckle-fat' and 'elbow-grease' both fit the segregated unit but what that unit is, even perceptually, is not given independently of some perceptual category. In such a case as this, and, of course, many others could be cited, systems of expression guide and inform the perceptual segmeting acts themselves. (Innis 1988: 91)

I cannot help but to read concourse into this: 'natural' movement of the human body and its parts is segmented into meaningful units often by means of linguistic signs. What is the difference between a smile and a grin?

In fact, language cannot be a prison-house for Bühler as it has become for certain strands of French thought. Indeed, one of his main theses was the constant and ineluctable modulation and guiding of the articulation and application process by the Sachwissen of the linguistic subject, paralleling in some ways Gadamer's insistence on 'application' as a hermeneutical necessity (Gadamer, 1960: especially part II) and Gardiner's concentration on the 'thing-meant' (Gardiner, 1932). (Innis 1988: 92)

My interest in Alan H. Gardiner's work only grows. Also, this: "So I wrote a song. I hope you guys dig it. It’s a song about people and sasquatches... other French science stuff. That’s French science." (Reggie Watts)

A concept, as a schematic abstraction, is not the 'object' of our knowledge but an instrument of our knowledge. It's reality, as the Scholastic philosophers perspicuously showed, is the reality of a relation, its being directed toward something which is not itself. (Innis 1988: 93)

Taking this premise and running with it, I now see that the "semantic functions of language structures" as sign-functions (sensu renvoi - in the sense of referring to something other than itself) can be illustrated as semiotic relations between the sign and: (a) the experience of the subject [emotive]; (b) the extrasemiotic object [referential sui generis]; (c) the intended interpreter of the given sign [conative]; (d) the sign itself as a node in the sign system [metalingual]; (e) the sign itself as a message, a text, a word of art, etc. [poetic/aesthetic]; and (f) ... - See, this is where I get in trouble, because at the moment I cannot think of any reasonable, strictly semiotic, interpretation of the phatic function. It is a relation between the sign and what? At best I could say something like the relationship between the communicator and the communicatee, but aside from finding Morris's terms more suitable I'm not so sure it's wise to conflate matters of signification and communication so easily.
One line of thought that didn't lead to anywhere in particular was to add something to the Bühler-Jakobson combo model by conflating Bühler's three types of signs with Peirce's three semantic relations. That is, something like: symptom-icon, signal-index, and symbol-symbol. This would be the conventional way to go about it. Taking into account the "arrow" formations in the combo model, I'd rather fit together: (1) Bühler's symptom with Peirce's icon; (2) Bühler's symbol with Peirce's index; and (3) Bühler's signal and Peirce's symbol. I'm not sure I can rationalize these connections. Emotions and icons are both in the category of Firstness; the object/symbol in Bühler's model is opposite of contact in the combo model... well, here it would fit rather with the index because Secondness is Otherness... huh. So the first iteration (symptom-icon, signal-index, and symbol-symbol) is more correct. It should be noted, then, that in Bühler's model the subject is First, the addressee is Second and the object is Third. But all this is just speculation.

Esse ad defines its 'reality.' But the conceptual content borne by the linguistic sign is also controlled and defined by the 'intentional object' and its characteristics, by what the concept makes known, and the linguistic subject constantly appraises the 'fit' between the content and the object-domain into which it gears, a point often made by Peirce and also Wittgenstein. (Innis 1988: 93)

This is why understanding the exact nature of the language functions is so difficult: language in its functions as they are outlined by both Bühler and Jakobson don't exactly "make known" the concept of the component it is tied with. The functions are more like "undercurrets" or "orientations" of sign-functions. In the end the functions are there for the benefit of the speech analyst - in actual speech these functions are quite irrelevant. "Function" would seem to fit the notion of "object-domain", but that would relegate the sender, receiver, message, code and contact to "objects". This is unproblematic only in the case of the referential function, in which case it is pertinent to discuss objects. Could it be said that the functions "objectify" these componets? Does the metalingual operation make code the object of speech? I think it does. But what kind of "object" is apparent in the phatic function? Is contact itself an object? The relationship between the sender and receiver, on the other hand, could be an "object" in this sense... I don't really know anymore.

As is well-known to students of Bühler's work, Bühler built his language model around a fundamental distinction between the index field [das Zeigfeld] and the symbol field [das Symbolfeld]. This distinction adverts to the autonomous and separate functions of intuitive pointing and presenting and the abstract, conceptual grasping of the world. These two fields are "the two sources out of which in every case the precise interpretation of linguistic expressions is nourished" (Bühler, 1934: 149), namely, the situation, which as intuitive is determined through deixis in all forms, thus generating the Zeigfeld, and the context, the syntactic and semantic matrix in which symbols are to besituated, the Symbolfeld. (Innis 1988: 99)

This is a pretty cogent account of context and situation and accords to my own intuitive distinction between the pairs verbal context and nonverbal situation (in the sense that nonverbal communication is more about signals and verbal communication more about symbols, although this distinction is not absolute). It would be very complicated to draw a distinction between context and contact in Jakobson's model on the basis of these fields.

First, there is Bühler's insistence on the primacy of the social matrix of sense and meaning, rooted to be sure in his organon-model of language, but also in his appropriation of the Aristotelian notion of praxis, of Handlung, and of the objective orientation of language theory deriving from Hobbes and Mill. It intesects clearly with the work of Mead, Morris, Wittgenstein and many others. (Innis 1988: 100)

Too bad that the intersection with Morris wasn't elaborated in any significant way. That would be an interesting read, I think.