Peirce and Polanyi: Perceptual Consciousness and the Structures of Meaning

Robert E. Innis

Department of Philosophy

University of Massachusetts Lowell

Lowell, MA 01854

I.

Peircean sign theory contends that semiosis, or sign-action and sign-interpretation, encompasses all of mental life, even at the lower threshold of perception, where the embodied subject first encounters the world. Peirce’s central idea, formulated at the very beginning of his intellectual journey and remaining active in his thought at its very end, is that “the content of consciousness, the entire phenomenal manifestation of mind, is a sign resulting from inference” (Peirce 1868, 53). A fortiori, then, perceptual processes, culminating in the ‘uttering’ of a perceptual judgment, already are sign processes, giving rise to and conditioning those further ‘meanings’ or diversified and multi-leveled streams of ‘interpretant signs’ that make up the course of life. Ernst Cassirer, writing sixty years later in the masterful third volume of his Philosophy of Symbolic Forms, remarked, independently of Peirce though for cognate reasons, that “the ‘sign’ is never a merely accidental and outward garment of the thought, but...the use of the sign represents a basic tendency and form of thought itself” (1929, 57), including the primary stratum of perception. Cassirer charts in great detail and with systematic vigor the ‘semiotic grammars’ of expression, representation, and signification as permanent dimensions of signitive consciousness, each dimension defined by a progressive ‘dematerialization’ of signs and sign-systems and a distancing from the perceptual and intuitive world and culminating in the world of pure relations characteristic of mathematics and mathematical physics. Nevertheless, for Cassirer, as for Peirce, all access to the world is through signs and ‘meaning’ and there is no ‘reality’ accessible outside the play of signs, even on the perceptual level. Indeed, what Elmar Holenstein wrote about the significance of structuralism, which has theoretical roots in a branch of general linguistics, applies quite generally to a semiotically oriented epistemology and theory of perception within which any language theory must be situated: they “draw our attention to the root-like attachment of the world’s subjective constitution to sign systems” (Holenstein 1976, 5). Although we might, on reading such texts, also think of Jacques Derrida’s post-structuralist project or of Josef Simon’s parallel ‘philosophy of the sign’ (Simon 1989) with their critiques of all pretenses to get ‘outside’ of signs, Peirce’s sign theory, as developed over a lifetime, is wedded to a realist epistemology, although it is a realism of a very special sort, with key elements derived from Medieval scholasticism.

In spite of a quite different starting point and conceptual apparatus, Michael Polanyi has given under the rubric of ‘tacit knowing’ a complementary ‘realist’ and ‘inferential’ account of perception and of the structure and genesis of perceptual meaning that intersects at key points with Peircean sign theory. Polanyi’s account adds new features to the Peircean model and unfolds some of its implications in novel and more explicit ways. Polanyi has definitively shown, in my opinion, that perception, as a tacit, acritical process, both models and exemplifies fundamental aspects of the field of consciousness as such within which language and all sign systems function. It is one of the deepest common insights, albeit differently formulated, of the Peircean and Polanyian positions that ‘perception’ is both an instance and exemplar of semiosis or meaning-making. In one sense, for both of them perception--or the perceptual stratum--is the matrix and condition of all ‘later’ or ‘higher’ signitive events such as language and art, which drive the expanding spiral of semioses and the construction of those webs of signs by means of which we ‘articulate’ both ourselves and our worlds and are enabled to double back on ourselves and control and evaluate our conduct. Peirce and Polanyi offer us powerful analytical tools for ‘placing’ the emergence of meaning and for understanding, in Merleau-Ponty’s words, “the contingent arrangement by which materials begin to have meaning in our presence, intelligibility in the nascent state” (1942, 206-207).

Polanyi’s account of perception is embedded in his comprehensive attempt to develop an account of the ‘tacit logic of consciousness,’ based on a logic of tacit inference. This thoroughgoing inferential focus he shares with Peirce. While admitting a certain ‘primacy of perception,’ Polanyi also seemingly paradoxically affirms the centrality or pivotal nature of language and other meaning-carrying systems as vehicles of the distinctively human ‘world-building’ process of ‘articulation.’

All human thought comes into existence by grasping the meaning and mastering the use of language. Little of our mind lives in our natural body; a truly human intellect dwells in us only when our lips shape words and our eyes read print. (1962 , 160).

Nevertheless, “the logic of language itself--the way language is used--remains tacit. Indeed, it is easy to see that the structure of tacit knowing contains a general theory of meaning which applies also to language” (1964, 145). It is precisely in the essentially tension-filled cooperation between the ‘tacit’ and the ‘explicit’ that mental growth and the cognitive mapping of the world is effected. While, to be sure, language and other formal systems involve a kind of ‘break’ with perception, they nevertheless do not constitute an autonomous ‘layer’ of meaning or sense applied like a veneer to a perceptual field that otherwise remained the same. In fact, for Polanyi, the structures discerned in perception are both extended into the systems of expressions and define the basic parameters of our dwelling in and use of them. Polanyi is thus able to affirm a kind of continuity from ‘perception,’ broadly understood, to the highest reaches of formalization. His last work, appropriately entitled Meaning (1973), tried to show this in detail, though many of its main theses were already foreshadowed in his indispensable classic, Personal Knowledge (1958) and many essays. Not only does speech have “the fundamental structure of all meaningful uses of consciousness in animals and men” (Polanyi 1967, 181), but this structure is perspicuously and exemplarily present in perception. Rather than ‘reading’ perception in terms of language, Polanyi ‘reads’ language in terms of perception which is itself ‘read’ in terms of meaning. We thus have a kind of Polanyian analogue to the Peircean notion of semiotic closure.

In the space of a single paper I cannot undertake a comprehensive and detailed comparison of the work of Peirce and Polanyi. I will focus, as already indicated, primarily on the paradigmatic role perceptual consciousness and perceptual meaning plays in their thought and will try to indicate, schematically and allusively, some of the profound implications that result from tracing the birth of meaning--including linguistic meaning--to its perceptual roots. What can we learn from Peirce and Polanyi about perceptual meaning and its structures? Rather than seeing ‘perception’ as a first and relatively impoverished step in cognitional processes that is ‘surpassed’ by ‘later’ steps, we will see that a reflection on the structures and processes of perception and the generation of perceptual meaning reveal permanent and eminently accessible truths about the most fundamental features of the production and appropriation of meaning quite generally and upon the work specific to language and other expression systems.

II.

Christopher Hookway has rightly noted that “the theory of perception ... occupies a fundamental role in Peirce’s epistemology” (1985, 151) and that such a theory must be “phenomenologically plausible” (155). What, then, are the essential descriptive features of Peirce’s theory and what types of categories, observations, and distinctions does he adduce?

The pivot of Peirce’s account of perception, and, as we shall see, a major point of connection between his work and Polanyi’s, is formulated in the following text where perception is assimilated to abductive or ampliative inference, that is, the introduction of novelty in a chain of reasoning, which normally is aided, indeed supported, by a formalism.

Abductive inference shades into perceptual judgment without any sharp line of demarcation between them; or, in other words, our first premisses, the perceptual judgments, are to be regarded as an extreme case of abductive inferences, from which they differ in being absolutely beyond criticism. The abductive suggestion comes to us in a flash. It is an act of insight, although of an extremely fallible insight. It is true that the different elements of the hypothesis were in our minds before; but it is the idea of putting together what we had never before dreamed of putting together which flashes the new suggestion before our contemplation. (CP 5.181)

What induces for Peirce the abductive inference or perceptual judgment? We find ourselves inserted into a world through our bodies and accessing the world through a dynamically oriented sensory system which is impinged upon in multiple ways, ‘interrupted’ and imbalanced by altersense and set into ‘intentional motion’ by reason of our being put into a ‘situation of perplexity,’ to speak in Deweyan language. But what we are, on Peircean principles, first and foremost conscious of is not a somatically mediated and atomistically presented sensory array as such but the patterns, orders, and structures in it. In this sense Peirce develops a ‘holistic’ approach to perception which restores to us, with great sophistication, the common sense world of perception and which avoids the psychologist’s fallacy to which talk of ‘sense-data’ and ‘primitive givens’ is subject. As a thinker (Martin Heidegger) seemingly far from Peirce (who, nevertheless, was reading Peirce in German shortly before his death) has laconically noted:

We never really first perceive a throng of sensations, e.g., tones and noises, in the appearances of things....Rather we hear the storm whistling in the chimney, we hear the three-motored plane, we hear the Mercedes in immediate distinction from the Volkswagen. Much closer to us than all sensations are the things themselves. We hear the door shut in the house and never hear acoustical sensations or even mere sounds. In order to hear a bare sound we have to listen away from things, divert our ear from them, i.e., listen abstractly. (1950, 26)

In her helpful and provocative Charles Peirce’s Pragmatic Pluralism Sandra Rosenthal remarks that “what we ordinarily perceive, what instigates action in the ongoing course of experience, are not appearances but appearing objects” (1994, 55). These objects, patterns, and orders are the ‘percepts,’ indeed the “parish of percepts,” which are the starting points of inquiry as it unfolds in its articulate phases: linguistically, mathematically, scientifically, aesthetically.

Peirce accepted from James the notion that the inquiring, perplexed, and self-moving organism needs to disaggregate the global manifold, marked by continuity, that it finds itself not so much over against as within. At the same time, in being appropriated to or caught up in its surrounding language and other expression systems, it finds itself living in an already linguistically pre-structured world where ‘cuts’ have already been made in the ‘sea of indeterminacy’ within which we live. On the Peircean position, as human beings we begin and develop our cognitive engagement with the world normally from within, and by learning to avail ourselves of, a system of already accomplished cuts and their relational contexts. Within the ‘cut world’ we find it necessary both to ‘realize’ for ourselves, or grasp the ‘fittingness’ of, the traditional cuts and to make further modified cuts, and this process, whether exemplified in the recognition of types and qualitative unities or in the application of a term, resembles, indeed models, the more explicit and differentiated process of hypothesis formation or abductive inference. Perceptual judgments are first cuts that are stabilized, crystallized, potentiated, and induced, by language and other expression systems.

Peirce’s account of perception oscillates, with rather different effect, between the image of synthesis, of the production of a novel unity and focus within experience, and the image of segmentation. The reliance on the category of synthesis stems from his Kantian background while the emphasis on segmentation arises from multiple sources, especially James but also from Peirce’s deep-seated synechism or commitment to the metaphysical ultimacy of continuity. Perception for Peirce is the work of synthesis because the perceptual field is characterized by the appearing of unities or ordered wholes which are themselves complexes, with internal structures. Unity-in-diversity or wholes composed of ‘parts’ must be ‘held together’ or must have been ‘brought together’ by some ‘act’ or ‘process.’ Now, ‘percepts’ on the Peircean position are the interpretant-signs in and by means of which physical objects are known. The recognition of the objectivity of the percept, its veridical character and its power to reveal its object, is manifested in both the continuity and lawful sequence of the modes in which the object, as conceived, manifests itself in our future experience, but also in our intrinsically hazardous ‘practical’ commitment of ourselves to comport ourselves toward it in confident and lawlike ways. For Peirce, the ‘logical status’ of a perceptual judgment is defined by these two experiential or pragmatic marks.

Let us take a simple illustration.

While writing at my desk I ‘see’ my pencil with the green eraser or at least that part of it that is not covered by a book. It is a distinct unit in the global perceptual field, separated off, that is, segmented, from the field where it has its own ‘place.’ In Jamesian terms it is a thematic unity within a field which itself fades off into an indefinite margin. I am not, however, explicitly conscious of synthesizing the features of the perceived pencil into a unity: its linearity, its hexagonal shape, the metal band holding the green eraser, its tapering writing point. But the ability to hold this ‘de-fined’ unit together is, on Peircean terms, a synthesis that proceeds by distinguishing the object from all other objects in the field such that I can ‘re-cognize’ it. I determine my percept as veridical, however, by being able to reach out and take up the pencil and to write with it, adding to the visual features now tactile and motor elements, since I can determine by feel that the pencil is made of wood, covered with glossy paint, and that the implementation of its balance by its being long enough to fit in one’s hand is matched by the ‘expected’ weight of the pencil when it is taken up and inserted into the writing task at hand. Thus, Peirce writes that “to predicate a concept of a real or imaginary object is equivalent to declaring that a certain operation, corresponding to the concept, if performed upon that object would ... be followed by a definite general description” (CP 6.132; see Rosenthal 1994, 29). Perception, effected in the perceptual judgment, contains implicitly or commits us to a description.

‘Being a pencil’ is rooted in the felt unity of the percept and ‘it is a pencil’ is our linguistic expression of the perceptual judgment. The percept, as making known or as the appearing of the perceptual object, is interpretant sign of the object, hence a third, and is defined by its distinctive ‘quality’ or firstness that marks it as a kind of thing or type and that opposes us, as a secondness, in any attempts to ‘think it away’ or to misuse it. The case of perceiving a ‘pencil,’ to be sure, involves reliance on pregiven perceptual categories which are carried over into the percept, a kind of poneception, and an anticipation that the percept will continue to reveal the perceptual object in the future, a kind of anteception. There is a dialectic of remembered percept and anticipated percept within the formation of the present percept. As Peirce puts it, percipuum, antecipuum, and ponecipuum are “the direct and uncontrollable interpretations of percept, antecept, and ponecept” (CP 7.648). Of course, the habitual and acritical nature of perceptual processes cloaks the genesis of original sense when a novel unity is constituted in, or coalesces in, the perceptual field. In such cases we experience a shift over which we have no control, a sense of things coming together which we then ‘recognize’ as having happened. Perception, for Peirce, in this sense is ‘operative,’ but not ‘thematic,’ resembling in many ways Husserl’s account of ‘passive genesis,’ the formation of unities in the life-world without the cognitive elaborations and explicit positings of a thematically objectifying ego. Perception is for Peirce something that is first and foremost ‘undergone.’ But, paradoxically, what is undergone is the continuous process of encountering already synthesized complexes or internally differentiated unities in the continuous flow of experiencings. We find ourselves caught up in a series of ‘events’ or ‘outcomes’ that involve no explicit conscious control but that bear witness to a functioning spontaneity.

For Peirce the percept is the interpretant of a preceding set of perceptual ‘signs’ which may or not be consciously accessible, a point that Polanyi will also make, though not in such terms. The percept is, in this sense, the meaning of the preceding signs as well as the conclusion of a set of possibly and often inaccessible premisses. Rosenthal writes that the

perceptual meaning is an organization of characters by which one intends the meaning of an object as that to which essential properties must apply and to which nonessential properties may or may not apply, and these two types of applicability are built into the very sense of, or the meaning of, the concepts by which we delineate a world of perceptual objects. This meaning must be prior to the very possibility of denotable instances. (Rosenthal 1994, 36)

It is an ‘open’ meaning, indeeCOURIERformulation of a ‘type’ that allows us to recognize ‘tokens.’ Now we find out what the premisses leading to the abductively arrived at perceptual meaning are or must have been by attending to what we de facto perceive. The process of eruption of meaning in the perceptual field Peirce claims is a “subconscious process” not subject to logical analysis (CP 5.185). Indeed, “this process of forming the perceptual judgment, because it is subconscious and so not amenable to logical criticism, does not have to make separate acts of inference, but performs its act in one continuous process” (CP 5.185). The percept wherein the perceived object becomes known, moreover, arises from the “time-binding” operation that is the very course of life itself. Although there is a certain amount of “arbitrary spontaneity” (1892, 329) in mental action, we nevertheless experience a fundamentally stable world, for the division of the experiential manifold into qualitative unities also establishes law-governed reactive relations between the perceiver and the perceptual objects themselves. Thus arises “self-analyzing” (Rosenthal 1994, 47) habits of all sorts, the ultimate logical interpretants, as the rational purport of the interpreting sign which is the configured unity of the percept-perceptual judgment structure, making up the percipuum.

When Peirce speaks of the acritical nature of the inferential process of perception he does not mean that the result of the process cannot be criticized. This would make nonsense of his critical commonsensism or pragmatic realism that Skagestad (1981) discusses with such insight and acumen. It may be that the perceptual judgment is “a judgment absolutely forced upon my acceptance, and that by a process which I am utterly unable to control and consequently am unable to criticize” (CP 5.157), but, once made, the judgment can, and must, be reflected on or criticized, which it, at any rate, always is by reason of the presence of secondness or the ‘outward clash’ in all cases of veridical perception. Hookway makes this notion the guiding idea in his discussion of Peirce’s theory of perception. This reflection is demanded by the very norms of coherent or consistent action or conduct. The inferential process involves typification and a kind of abductively effected ‘migration of properties’ from object to object in a continuous process that we are forced, Peirce thinks, to call interpretation, involving as it does the judgment of the ‘fit’ between a concept and an individual instance. Polanyi will argue that this process involves at every step the personal and tacit appraisal of the knower. Peirce writes that the very abductive nature of the perceptual judgment confers on it “characters that are proper to interpretations” (CP 5.184) and “the fact is that it is not necessary to go beyond ordinary observations of common life to find a variety of different ways in which perception is interpretative” (CP 5.184). Here is point of deep affinity between Peirce and Nietzsche.

In the context of discussing James’s analysis of perception as unconscious but ampliative inference in his Principles of Psychology Peirce schematizes the form of perceptual abduction as follows:

A well-recognized kind of object, M, has for its ordinary predicates P1, P2, P3, etc., indistinctly recognized.

The suggesting object, S, has the same predicates, P1, P2, P3, etc.

Hence, S is of the same kind M

The first premise, Peirce contends, is in our minds habitually but this, he thinks, would not of itself make the inference unconscious. What makes it unconscious is that “it is not recognized as an inference; the conclusion is accepted without our knowing how” (CP 8.64-65). Polanyi draws attention to this characteristic when he notes that “the conditions in which discovery usually occurs and the general way of its happening certainly show it to be a process of emergence rather than a feat of operative action” (SFS 33). And Rosenthal notes that the “shading of scientific abductions into everyday perceptual claims is a continuity not of content organized but of method of organization” (1994, 148n56).

Rosenthal has importantly drawn attention to the fact that “this primitive synthesis” (1994, 60) effected in the perceptual judgment is a defining element in all cognition. “All cognition for Peirce involves the perceptual in the sense that it logically involves an iconic presentation of the cognized object” (1994, 46), for, as Peirce has noted, “icons have to be used in all thinking” (Peirce 1976, 4.21). If, as Rosenthal contends, icons, functioning as schemata in cognitional processes, involve “elements of firstness, secondness, and thirdness, or image, activity, and rule” (1994, 136n29) we can see why a reflection upon “the logic of perceptual awareness” (1994, 51) can reveal the contours of consciousness as such and allow us to affirm an essential continuity between ‘perceptual consciousness’ and the so-called higher forms. Perceptual consciousness avails itself of schemata because no percept is absolutely precise or identical with the perceptual object qua tale. A schema allows us to grasp or represent the organizing structures of the object without affirming a coincidence between the appearing and the appeared and it allows us to recognize and even construct future instances of a concept. Peirce notes that the mathematician uses the schematizing power of “diagrammatical reasoning” to introduce novelty into the deductive process. This is accomplished “through the formation in the imagination of some sort of diagrammatic, that is, iconic representation ... as skeletonized as possible” (CP 2.778), though it also involves visual and muscular imagery, a point also made by Einstein and many others. The key point, however, is that schematic structures ultimately have to be understood, as Rosenthal puts it, not as a “generalization of imagined instances but as a product of a predictive rule” (1994, 24). Theorematic reasoning, which characterizes mathematics, depends upon experimentation with individual schemata, which are “specially constructed” for the purpose.

The mathematician’s use of schemata is paradigmatic for knowing quite generally and for perception in particular. In one sense the percept is a diagram of its object, without being for all that a picture or copy. As a construct it allows us to freely vary the modes in which the object appears, giving us, as Rosenthal says, “a predictive rule generative of the action-image matrix of a schematic structure” (1994, 24) and facilitating, and even making necessary, by means of its variability and flexibility the perception of new relationships. In his deeply unsettling article, “From the Icon to the Symbol,” René Thom (1973) put the matter in the following way:

While exploring a new theory, while juggling with this new material, the mathematician sometimes sees an expression, or a relation, turning up again and again with embarassing insistence. He will then be tempted to introduce a new symbol to condense this expression into a single form and so continue the work on a new basis. This simple procedure may sometimes lead to success. More often he will be struck by the idea of new expressions to condense, new figures to construct and name through suspecting a priori their properties. To introduce a new symbol, that is, injecting a new letter on to the paper, promotes a kind of tearing away, with the establishing of a new semantic field which will be the support of the new actant and so free the mental movement from the obsessional presences which impede it (in Innis 1985, 290).

This ability to “suspect a priori,” an essential property of expression systems, is embodied in schematic structures upon which we can experiment and is present at the very heart of perceptual consciousness.

The diagrammatic reasoning of the mathematician, which relies on external systems of expressions, functions as both model and extension of the perceptual process itself. While mathematical reasoning depends upon the thematic invention of an appropriate formalism upon which we rely and which functions as a necessary support and scaffolding for the reasoning process, enabling it to derive novel theorems, perception likewise must be understood as embodied in and relying on equivalent formal structures which are in no way ‘inner’ in any Cartesian or Lockean sense. They are formed in the abductive process of making sense of experience by the development of habits and by the essential ‘openness’ of habit as ultimate logical interpretant. Series of percepts, that ‘make present’ aspectually the perceived objects in the utterings of perceptual judgments, are ‘linked’ by shared schematic structures that mediate between the conceptually articulated percept and the perceptual object or domain. Rosenthal (1994, 26) cites a passage (MS 31 293, p.14) which bears upon the problem of how meaning is contructed in perception. Peirce writes:

The Diagram remains in the field of perception or ‘imagination’ and so the Iconic Diagram and its initial Symbolic interpretant taken together constitute what we shall not too much wrench Kant’s term in calling a schema, which is on the one side an object capable of being observed, while on the other side it is a general.

Peirce’s notion here anticipates the later work of Mark Johnson (1987) on image-schemata, which mediate between concepts and percepts and which function as essential conditions for the rise and development of linguistic meaning, including the vast and intricate systems of metaphors within which we articulate our fundamental relations to the world.

Moreover, Peirce held that in corollarial reasoning taking place in language, as distinct from theorematic reasoning taking place in a technical formalism, “the very words serve as schemata” (CP 4.233), that is, heuristic devices that further the introduction of novel patterns and relations. Giovanni Vailati, the most creative of the Italian pragmatists, developed this insight in great and illuminating detail. (See Vailati 1911 and Innis 1989.) Words indeed carry the “aspects” of things, but prior to words we have living habits and living dispositions to sort or order experience in certain ways. Rosenthal remarks that “the series of possible schemata for the application of a concept to experience is ‘fixed’ prior to the imposition of linguistic structure” (1994, 32). On this position linguistic structures would ‘build upon’ as well as depend upon and incorporate

...the dynamics of lived experience at its most rudimentary level, a dynamics that in turn reflects a semiotic structure operative at its most fundamental level. The significance of the logic of language lies in the fact that it grounds itself in those most rudimentary semiotic structures by which humans experience a world of appearing objects. Hence, an examination of such epistemic foundations should lay bare the basis for the logic of linguistic structure. (1994, 27)

This is what Rosenthal, paraphrasing Peirce, means when she writes that “meaning beneath the level of language” employs schemata, for, in a Peirce text which she cites, Peirce also asserts that “meaning enters into language by determining it” (MS 1105, p.4; Rosenthal 1994, 26). This is one of Polanyi’s central notions, which he will trace not to the operative force of schemata as such but to the ever present tacit component that underlies and is potentiated by articulation in all its forms. Mark Johnson, I noted earlier, traces it to the image-schemata rooted in the fundamental invariant features of our bodily existence, which are taken up into the ‘mind.’

If any of the above notions have “phenomenological plausibility,” we are forced to agree with Rosenthal that “the difference between the perceptual and the conceptual is not a difference in kind but a difference in the proportions of sensory content and relational structure” (1994, 46), not in the absolute presence or absence of either. Cassirer makes the same point abundantly (see, for example, his 1944, 130-136 for a compendious discussion of materials he developed at length elsewhere). At the perceptual pole we have a predominace of sensory content while at the conceptual pole we have a predominance of relational structure. But at no point are we free of mediating semiotic structures. Perceptual schemata exemplify the “universalizing aspect of sense” and “the indeterminateness of meaning” quite generally, a central and complicated Peircean position. Peirce had claimed that “no concepts, not even those of mathematics, are absolutely precise” (CP 6.496; see Rosenthal 1994, 35). Even at the farthest limits of abstraction, behavior, instantiated in living habits of interpretation, remains for Peirce the matrix of the patterns of relations that make up conceptual meaning (Rosenthal 1994, 27). Meaning exists within purpose and within purposeful behavior, a point that Dewey resolutely foregrounded in his organism-based theory of inquiry.

Rosenthal also rightly, and uncontroversially, notes that conceptual meaning “must include within itself the emotional, energetic, and logical interpretants,” that is, firstness, secondness, and thirdness: feeling core or sensuous content, pattern of reaction, structure. But it is clear that this threefold structure is also there, differently proportioned, in perceptual meaning and in the dispositional habits that constitute it. These dispositional habits are creative, ampliative, abductive. They do more than “unify three preexistent elements--sensory cues, acts, and resultant structure” (31). They generate structure by in fact synthesizing sensory cues and reactions by ‘making sense of them.’ But, first and foremost, this making sense is not a feat of operative action. It is an event that carries us away.

Peirce pinpoints this feature when he speaks of consciousness as a bottomless lake. It is an image that ‘exhibits’ the experienced quality of the flow of perceptual consciousness.

I think of consciousness as a bottomless lake, whose waters seem transparent, yet into which we can clearly see but a little way. But in this water there are countless objects at different depths; and certain influences will give certain kinds of those objects an upward impulse which may be intense enough and continue long enough to bring them into the upper visible layer. After the impulse ceases they commence to sink downwards. (CP 7.547)

Such a metaphorical characterization captures something essential about perceptual consciousness in particular, but I think that, in fact, it also captures something essential about the very flow of conscious life in general. Signitive happenings are experienced as ‘e-vents,’ as ‘out-comes’ wherein the interpreting subject is put into ‘intentional movement’ by processes over which he or she has no control. The experience of meaning quite generally is exemplified in the appearing of perceptual meaning, for perception is precisely the determining of those both accessible and inaccessible streams of signs which are only recognized as signs in the result. We can attempt, through abstraction and through systematic reflection, to isolate the determining signs that are included in the interpreting sign, and thus make them determinate, but they are not always retrievable in a satisfactory manner, though the mark of human life in signs is to bring more and more of the operative signs that control conduct into awareness.

Thinking, of which perception is a fundamental layer or stratum, Peirce writes, is “a kind of conduct, and is itself controllable, as everybody knows. Now the intellectual control of thinking takes place by thinking about thought. All thinking is by signs; and the brutes use signs. But perhaps they rarely think of them as signs. To do so is manifestly a second step in the use of language.” (CP 5.534) It would seem to follow that the advent of language effects a shift in perception, first of all by making it problematic on a different logical level and by allowing it to become self-critical in a new way by trying to fix and to make explicit the actual perceptual signs which we are interpreting and whose meaning is the perceptual object itself. These signs, as Dewey (1896) argued, are the total state of the inquiring organism and in this sense ‘perception’ is ineradicably wedded to bodily existence in all its modalities. Polanyi will also make much of this.

But even if we speak of all the ‘premisses’ of the abductive perceptual inference as being accessible to consciousness, the process of perception is still acritical in that it involves a performative commitment on the part of the perceiver. For while we might experience a kind of self-organizing activity within the perceptual flow, the to-be-organized, the organizing, and the organized must be distinguished. Take the following passages which also rely on a powerful metaphor to model our consciousness of an object. Speaking of “two sorts of elements of consciousness” Peirce distinguishes between separate notes in a piece of music and the air or melody, the “orderliness in the succession of sounds” (1878, 128). The orderliness is not ‘immediately given’ but is what results from a mediation. We experience both the succession and the melodic line. The melodic line is the mediate object, the succession of sounds the immediate object. “These two sorts of objects, what we are immediately conscious of and what we are mediately conscious of, are found in all consciousness” (1878, 128-29). The perceived melody then is a “congruence in the succession of sensations that flow through the mind” and, analogously, thought “is a thread of melody running through the succession of our sensations” (1878, 129).

Which melody? That is the melody as taken, for while in one sense the configuration of “sensations” organizes itself according to the laws of the association of ideas, which Peirce never repudiated (see Innis 1998b), the essential continuity of the perceptual continuum demands a ‘re-marking’ on the part of the perceiver, grounding, in effect, a polyphonic notion of perceptual experience and the ‘pragmatic pluralism’ argued for by Sandra Rosenthal. The melodic threads of experience are multiple. Peirce writes that “just as a piece of music may be written in parts, each part having its own air, so various systems of relationship of succession subsist together between the same sensations” (1878, 129). By “the same sensations” Peirce does not mean discrete sense-impressions or sense-data but the flow of experiencing, the flow of experienced vectors, out of which coalesce objects, regularities, patterns, order. The Jamesian ‘free water of consciousness’ flows around objects as ‘ob-stacles’ and in flowing around them ‘de-fines’ them as what they are. Indeed, I think that Peirce’s model of perceptual consciousness is in full agreement with James’s model of consciousness as being related to experience as a ‘sculptor’ is related to a block of stone. The point is to ‘free the form,’ to perform, in short, what seems like an oxymoron: an analytical synthesis.

Speaking of such a phenomenon as the Schroeder stair, which is analogous to the famous ‘Necker’ cube so beloved of cognitive psychologists, Peirce illustrates another feature of perceptual processes: a kind of experienced shift that ‘changes the look’ of the perceived object and which can, so Peirce thinks, be brought under conscious control, although Peirce’s description perhaps reduces the degree of automatism in the process a bit too much. I would like to note a peculiarity of Peirce’s choice of descriptive language in such a case. Peirce writes that “the perceptive judgment, and the percept itself, seems to keep shifting from one general aspect to the other and back again.” (CP 5.183) This “general aspect” refers to a phenomenal quality which defines the object, how it “looks.” A percept, then, gives us the look of things by binding the configured elements together, resolving the ‘puzzling look of things’ by forming an interpretation that makes sense of them. This forming of a classification of the object as [x] is contained in the perceptual judgment itself, thus connecting abductions and perceptions. “If the percept or perceptual judgment were of a nature entirely unrelated to abduction, one would expect that the percept would be entirely free from any characters that are proper to interpretations” (5.184). But, in fact, the percept is thoroughly imbued with interpretation, emerging as it does in the field of consciousnss as something with a distinctive quality.

Perception, and a fortiori perceptual abduction, both exemplifies the fundamental dimensions of semiosis and is clarified by advertence to the principal division of signs that lie at the heart of Peircean semiotics, that is, the division of signs into icon, indexes, and symbols. The perceptual field itself, with its systems of objects-in-relation, is rightly modelled along these lines, as I have discussed in a broad comparative context elsewhere (Innis 1994). But I think it best to speak of iconic, indexical, and symbolic aspects or elements of the percept. Nevertheless, as qualitative unity the percept, as making known the perceptual object, has an iconic or qualitatively defined core.

The sensory core of the percept is a qualitative type, as Rosenthal rightly notes, calling this “feeling core” which is “there” in experience “the logically or epistemically final basis and ultimate referent for all cognitive activity” (1994, 32). It has, or rather is, a distinctive ‘feel’ and is known in and by this feeling, which ‘interprets’ it. Any schema must have a feeling element if concepts are to be applicable to experience. Hookway, for his part, comments: “Differences in the subject matter of the percept are ... reflected in systematic differences in its qualitative character” (1985, 158). It is, in every case, the qualitative character of “structured complexity” (159). Further, as structured, the complex must have parts. How are the parts related to the whole and to the perceiver ‘uttering’ the perceptual judgment? The presence of parts in a configuration ‘indicate’ the configuration, not by being separable from it but by individualizing the perceptual object. When I say, “That is a yellow pencil,” I am referring to something standing over against me that can ‘react’ with me and to me and that is characterized by just these properties, dimensions, locations, which I can point out and which, in being pointed out, show themselves to be pointing something out. Indexicality is intrinsic to the perceptual judgment, giving it and its object a peculiar ‘thereness,’ grounding the ‘outward clash’ and in so doing introducing what Hookway calls a “brute unintelligible element into our experience” (170), a notion that perhaps must be taken with a grain of salt. The reason is that the properties criterially displayed by the percept must themselves be apprehended as continuously instantiated by the perceived object, for objects are defined in terms of “continuity of reactions.” Indices are “marks” that, experienced as continuous, allow us to identify and to re-identify objects in the flow of experiencing and to bind them together in the unified manifold which has its distinctive feel. The unity of feeling is a feeling of unity. The binding is the result of the perceiver’s work of mediation, of synthesis, of sym-ballein. But thirdness for Peirce is ‘in’ the percipiuum, not imposed on it ‘from the outside,’ for in his view “thirdness pours in on us through every avenue of sense” (CP 5.158). It is discovered by the veridical ‘cuttings’ of the continuum of experience that ‘replicate,’ in Thomian fashion, the self-generating cuttings of ‘forms’ and morphological structures of greater or lesser stability out of nature itself. Already at the level of perception, then, Peircean sign theory, implementing a semiotic realism, takes direct aim at nominalism and all its implications.

I want now to indicate where central elements in Polanyi’s work intersect with, confirm, amplify, and perhaps even correct these Peircean analyses and emphases.

III.

Polanyi’s mature epistemological model is based on an expansion and transformation of some central observations of Gestalt theory, which he thought had startling implications not only for the analysis of scientific knowing but for knowing quite generally (see Innis 1992). Gestalt theory, represented for example by Wolfgang Köhler, denied, in full consonance with Peircean pragmatism, that the world was primarily given to us as an indifferent mosaic or an indifferent continuum. “It exhibits,” Köhler wrote, “definite segregated units or contexts in all degrees of complexity, articulation and clearness.” Moreover, these units “show properties belonging to them as contexts or systems [and] the parts of such units or contexts exhibit dependent properties in the sense that, given the place of a part in the context, its dependent properties are determined by this position.” (Köhler 1938, 85)

In his classic presentation Gestalt Psychology Köhler formulates his main point in a manner that bears directly upon the present topic of discussion.

Gestalt psychology holds [that] sensory units have acquired names, have become richly symbolic, and are now known to have certain practical uses, while nevertheless they have existed as units before any of these further acts 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 the gradual entrance into the sensory field, meaning follows the lines drawn by natural organization; it usually enters into segregated wholes (1959, 82).

Polanyi conceived of his project as a transformation of a classic theme of Gestalt psychology: the particulars of a pattern or a musical tune have to be jointly apprehended. If these particulars are observed separately they form no pattern or tune (1958, 55-57), a point gestured at by Peirce in his melodic analogy. This joint apprehension gives rise to the ‘segregated wholes’ that Köhler referred to in the preceding passage. Polanyi’s goal was to delineate the structures and implications of this joint apprehension by establishing and developing a pivotal distinction that marks one of his most fertile contributions to philosophy: that between focal and subsidiary awareness. Polanyi saw that in such variegated instances as the development of motoric skills, the grasp of visual patterns, the constructions of a medical diagnosis, the formulation of an hypothesis, and so forth, we see a common pattern: we attend from (subsidiary awareness) a field of movements, visual particulars, symptoms, articulate clues while we attend to (focal awareness) what they ‘mean.’

Take the development, first, of motoric skills, which play a central role in Polanyi’s thought and which help to ground his “structural analogy between knowledge and skill” (Polanyi 1961, 130). It is a well-known fact of everyday experience that a motoric achievement such as walking a tight-rope or performing on the parallel bars is a feat of coordination. What are being coordinated are sets of proprioceptively apprehended actions and movements. Now the feat of coordination involves bringing these movements to bear upon the performance that one has in mind and that lies at the focus of our attention. This bringing to bear is, Polanyi noted, a process of ‘integration’ wherein the particular movements, accessible kinesthetically and (to speak in phenomenological terms) prethematically, are brought into a unity: the completed performance itself as a ‘comprehensive entity’ or Buchlerian ‘natural complex.’ We know from experience that in the performance of motoric skills we can paralyze the action as a whole if we concentrate on the constituent actions in themselves, keeping them, rather than the task at hand, at the center of one’s attention, a process called ‘destructive analysis’ by Polanyi. We have to rely on the actions and feelings, use them in an instrumental manner, commit ourselves ‘acritically’ to them. As Polanyi put it in terms of conscious functions, we must ‘attend from’ these particulars while ‘attending to’ what we are doing. This incipient ‘from-to’ structure, espied in numerous other instances, will become the structural key to Polanyi’s whole epistemology and the source of its immense heuristic fertility. Since to learn a skill is a cognitive achievement, to think of skills as paradigmatic outcomes of cognitive strivings is already to shift the focus of the theory of knowledge. Skills control conduct. Knowing-how precedes and grounds knowing-that.

The recognition of a physiognomy--a face or its moods, for instance, but also the ‘facies’ of diseases and of species of insects and flowers--manifests, on Polanyi’s view, a similar structure. In Polanyi’s reckoning the various features of the face, symptoms of the disease, characteristics of the insect or flower function as a not completely specifiable and complex set of clues on which we have to rely in order to recognize--in an act isomorphic with Peircean abduction--the face, mood, disease, insect, or flower. We do not, according to Polanyi, go, in summative fashion, from one isolated item to another and from their collection, as necessary and sufficient conditions, ‘deduce’ the object by an act of explicit inference from explicitly formulated premisses. Rather, we attend, for example, ‘from’ the features ‘to’ the face or the mood. We do not focus on the features in themselves but rely on them, in the same fashion as we rely on our consciousness of our bodily movements for achieving a coherent action in the construction of a skill. Just as we can bring an action to a halt by turning our attention to its constituent particulars, so focussing our attention on the particulars of visual wholes such as those mentioned above, but easily extended to line drawings, geometrical figures, and so forth, will cause them to disintegrate as phenomenal unities.

A third paradigmatic example, exploited also by Merleau-Ponty, is the use of a probe by a blind man or, with appropriate modifications, by a surgeon or dentist. Here we have an illustration of another important aspects of this from-to structure. A probe is first of all an object external to our body and we can feel it as external to us by attending to the pressure of it on our hand. However, we do not use the probe in order to feel it, but to feel by means of it, using it as an instrument. It has, as Heidegger saw clearly and continued to emphasize, an um ... zu--an in-order-to--structure. When we no longer directly and objectively feel the probe in our hands and fingers but ourselves feel what the probe is itself touching so that we come to know this object both ‘directly’ and ‘mediately,’ we can then be said to be attending from the pressures and impacts made by the probe on our body to what these pressures and impacts mean. That is, we have to bring these pressures and impacts to bear on a focus, on an object or comprehensive entity, through an act of ‘integration.’ That is, we perform an abductive act and this act is the result of a distinctive fusion of perceiver and instrument, which Polanyi characterizes as indwelling or embodiment.

Now, in performing these integrative acts--and others like them--Polanyi thinks that we either rely on or come into possession of, or achieve, a kind of knowledge that we cannot put (fully) into words, a kind of knowledge that is, in short, ‘tacit,’ ‘unspecifiable,’ ‘inarticulate,’ ‘unformalizable.’ This tacit character is especially obvious in such cases as bicycle riding, swimming, walking, speaking, where we are not aware, except after long and difficult analyses of something that we do easily, of the rules or laws that we are obeying in performing an an action. Likewise, Polanyi notes, the actual, operative topographic knowledge of the human body possessed by a surgeon is itself inarticulate, although it may rely on detailed and complex articulate mapping of the human body. This knowledge, Polanyi contends, is the result of a massive preconceptual act of integration, a feat of imagination built up over the course of long dealing with the three-dimensionsal internal structure of the human body. No sum of direct, explicit knowledge or awareness of the discrete parts will generate the three-dimensional--relational--Gestalt which is the image of the human body. Rather, we must say that the surgeon must attend from these discrete items to their integrating center. Otherwise, he will have no ‘praxical’ grasp of the structure itself. Therefore, both the bicycle rider and the surgeon are in possession of tacit knowledge, and what we can say about this knowledge is in itself inadequate to transmit it, which can only be done by practice and initiation. The differentiation between ‘theoretical’ and ‘practical’ surgery that characterized the Medieval medical faculty is an epistemological monstrosity or at least curiosity, a preferring of the ‘seminary mind’ over the ‘laboratory mind’ that Peirce was committed to opposing. Articulate formal statements--which may indeed be possible and certainly desirable--are meant only to guide us into the realm itself, to function, as Polanyi put it, as maxims which have to be applied in the concrete consciousness of the knower.

Upon the basis of considerations such as the foregoing Polanyi distinguished four ways of schematizing the from-to relation of parts to wholes, the operative core of his cognitional model: a functional, a phenomenal, a semantic, and an ontological. I can only gesture, in the briefest of ways, at the rich implications of this schema, which is implicit in the Peirce texts cited earlier concerning our apprehension of a melody, but which Polanyi makes thematic.

First, the functional aspect emphasizes quite generally the specific role all subsidiary (subsidiarily attended from) particulars play as vectors, as ‘indicators’ of a focus upon which they bear. They are the proximal terms from which we attend to the distal terms into which they are integrated or to which they ‘point.’ Their role is distinctively ‘instrumental’ in a universal sense of that term in being defined by their relation toward something else. Epistemologically their role is to guide us, to move us, to ‘bias’ us toward a term or focus. They are not restricted to any domain.

Second, while attending from them in attending to their focal unity we find that the subsidiary particulars undergo a change of appearance. Indeed, Polanyi asserts that the subsidiary clues upon which we rely are known in the appearing of a whole. For example, “the clues offered by processes within our body, of which we become aware in terms of things perceived outside, may be completely unconscious” (1958, 44), so that they are actually only known by our becoming aware of a perceptual object. Again, visual illusions such as the Necker cube, the duck-rabbit that so exercised Wittgenstein, and other multi-stable phenomena illustrate the shift of appearance attendant upon different integrations of particulars. What these particulars are is known in the configurations and, indeed, from the phenomenal point of view they are different particulars in each case, while remaining physically identical. This phenomenal character corresponds to Peirce’s notion that a perceptual whole has a qualitative character that gives it its distinctive unity and feel.

Third, Polanyi combines the functional and phenomenal structures into a general semantic structure. “When something is seen as subsidiary to a whole, this implies that it participates in sustaining the whole, and we may now regard this function as its meaning, within the whole” (1958, 58). For example, the ‘meaning’ of the impacts on our hands when using a probe or cane is the term of an interpretative act, a ‘reading,’ just as is the deciphering of a script, the construing of a poem or letter, the perceiving of a painting. In Personal Knowledge, while admitting that wholes are meanings, Polanyi distinguished between two kinds of wholes and two kinds of meaning: an explicit sign/object whole and the kinds of wholes exemplified by physiognomies, tunes, and patterns.

The distinction between two kinds of awarenesses allows us to readily acknowledge these two kinds of wholes and the two kinds of meaning. Remembering the various uses of a stick, for pointing, for exploring or for hitting, we can easily see that anything that functions effectively within an accredited context has a meaning in that context and that any such context will itself be appreciated as meaningful. We may describe the kind of meaning which a context possesses in itself as existential, to distinguish it especially from denotative or, more generally, representative meaning. In this sense pure mathematics has an existential meaning, while a mathematical theory in physics has a denotative meaning. The meaning of music is mainly existential, that of a portrait more or less representative, and so on. All kinds of order, whether contrived or natural, have existential meaning; but contrived order usually also conveys a message (1958, 58).

In other writings Polanyi calls these two types of meanings the ‘physiognostic’ and the ‘teleognostic,’ of which the ‘denotative’ is a specification of the latter. (See, for example, Polanyi 1961, 129-130) So, in general, where there is order there is meaning. Perceptual objects are existential or physiognostic meanings, since they are ordered contexts. In this Polanyi and Peirce are also in full agreement.

Fourth, the notion that an object is an ordered context implies for Polanyi that objects quite generally exemplify ‘emergence,’ that is, that objects have ontologically and conceptually distinct levels. Objects are not their constituent particulars. They are the ‘meanings’ of these particulars.

Since tacit knowing establishes a meaningful relation between two terms, we may identify it with the understanding of the comprehensive entity which these two terms jointly constitute. Thus, the proximal term represents the particulars of this entity, and we can say, accordingly, that we comprehend the entity by relying on our awareness of its particulars for attending to their joint meaning (1966, 13).

Here again, as with Peirce, is a specific correlation of being and meaning, an assertion of meaning in the most basic stratum of consciousness and a widening of the very notion of an object, derived from a ‘logic of consciousness.’

By relying on the distinction between focal and subsidiary awareness and classifying meanings on the basis of the kinds of subsidiaries integrated into a whole, we can develop a powerful and differentiated model of meaning. Different types of subsidiaries give rise to different types of meanings or ordered contexts such as perceptual, motoric, affective, conceptual, aesthetic meanings. Perceptual meanings arise from the integration of sensory cues. Motoric meanings arise from the integration of bodily movements and actions, encompassing bodily and practical skills of all sorts. Affective meanings arise from the integration of feelings. Conceptual meanings arise in the great feat of ‘articulation’ achieved by ‘language’ in the broad sense of that term that includes all ‘external’ systems of representation. Aesthetic meanings arise from the ‘thickening’ of the expressive means or sign-configurations.

The fundamental structure of consciousness is, for Polanyi, composed of focal and subsidiary awareness and tacit integrating acts of consciousness. How do these elements in Polanyi’s epistemology apply to the problem of the continuities between perceptual and linguistic meaning?

Polanyi, complementing Peirce, first and foremost traces “the strange fact that language means something” to the “exercises of an integrative power” (1967, 193). In itself this integrative power is inarticulate, even though the medium within which it is operating is maximally articulate, formal, and symbolic. As Polanyi puts it in Personal Knowledge, “a symbolic formalism is itself but an embodiment of our antecedent unformalized powers” (131). These unformalized powers constitute our “faculties for recognizing real entities, the designations of which form a rational vocabulary” (114). Polanyi shares with Peirce a thoroughly realist claim concerning the relation of words and conceptual schemes and accordingly an adamant rejection of conventionalism and nominalism. Representative meaning-systems both mediate to us tacitly apprehended meanings and rely upon a tacit component. “When we come to a deliberately chosen system of signs, constituting a language, we must admit that these have a denotative meaning which is not inherent in a fixed context of things or actions” (1958, 91). Indeed, in a comment that calls to mind Peirce’s notion of collateral experience, Polanyi writes that “words convey nothing except by a previously acquired meaning ... our knowledge of things denoted by words will have been largely acquired by experience....(That is) the words will have acquired their meaning by previously designating such experience” (1958, 92).

Polanyi claims that “every time we use a word for denoting something, we perform and accredit our performance of an act of generalization and ..., correspondingly, the use of such a word is taken to designate a class to which we attribute a substantial character” (1958, 80). On this position, with which Peircean semiotics is in general agreement, generals--general structures--are real and are ‘mapped’ in our language-systems. This “act of generalization” bears upon a really existing configuration, which, however, could not be known as such independently of the language-generated mediating system. As expressions, to take two homely examples, both “knuckle fat” and “elbow grease” bear upon or bring into focus a ‘center’ to which we attend, with language playing a ‘subsidiary,’ indeed Peircean ‘schematic,’ role, for “the focus of all articulation is conceptual, with language playing only a subsidiary part in this focus” (1958, 101). But while subsidiary, these linguistic schemata, which formulate a form of perceiving, are also constitutive of what is perceived. The focus is perceived as because it is conceived as and vice versa. Both expressions really ‘fit’ the experienced configurations. In line with Peirce’s notion that the perceptual judgment is aspectual, we can see that on Polanyian principles the perceiver attends from rather different “allegedly recurrent features” (1958, 112) to the whole upon which the subsidiaries bear. “Knuckle fat” in fact belongs not to the conceptual system carried by English but by (at least) the Danish conceptual system. It brings into focus rather differently weighted, but really existing, features of the experiential unit, focussing on pressure and the wearing down of materials rather than the reciprocating motion of a joint. ‘Articulation’ and ‘perception’ are joined here in an inseparable fashion. It seems to me that we can rightly say that “knuckle fat” and “elbow grease” schematize experience and that we attend from the schematizing linguistic expressions to the object-meant. Linguistic schemata are subsidiarily apprehended and, like probes, become assimilated to us and become parts of our intentional existence.

Polanyi is also well aware, although he does not use the Peircean terminology but refers instead to his own notion of a ‘tacit triad,’ of the triadic nature of meaning and of the sign-object-interpretant relation. When he speaks of “our power for comprehending a text and the things to which the text refers, within a conception which is the meaning of the text” (1958, 100) he does so, not to foreground the triad, but to foreground a feature of the feat of articulation intrinsic to his theory of tacit knowing. Dwelling in a text, that is, in the sign-bearing and meaning-bearing marks, our focal attention is on its meaning or sense, which is not something tangible, but is “the conception evoked by the text....The conception in question is the focus of our attention, in terms of which we attend subsidiarily both to the text and to objects indicated by the text” (1958, 92). Just as there is a logical gap between perceptual clues and what they mean, so there is a logical gap between a text and the objects ‘indicated’ by it. The text is an ordered context that generates other ordered contexts upon which it ‘really’ bears. Language is a heuristic aid in stabilizing these contexts in the flux of perception and increases our mental power over experience. In this sense Polanyi has formulated in a different way the role of both corollarial and theorematic thinking in the work of ‘articulation,’ which he richly elaborates in his materful chapter on ‘Articulation’ in his (1958), with extensive discussion, relying on, among others, the work of Polya, of heuristics and the logic of problem solving.

Polanyi’s and Peirce’s realistic theories of perception constrain any attempt to look upon language as ‘merely’ conventional. The realist thrust of language is rooted in the realist nature of perception. But the realism is not a naive realism. Peirce would agree with the following passage from Polanyi:

A particular vocabulary of nouns, adjectives, verbs, and adverbs, thus appears to constitute a theory of all subjects that can be talked about, in the sense of postulating that these subjects are all constituted of comparatively few recurrent features, to which the nouns, adjectives, verbs and adverbs refer (1958, 80).

Intra-systematically, the constellation of grammatical and semantic relations constitutes a closed universe, which will be more closed to the degree that the domains referred to by the vocabulary are formalized or to the degree to which the language is no longer in common use. Extra-systematically, conceptual innovations do take place when the vocabulary is either enlarged or internally modified to make way for new concepts. As Dewey remarked, meaning is self-moving from case to case (1925, 148). In the latter case, verbal speculation and verbal confrontations can reveal unexpected affinities between disparate realms that, without the linguistic sifting and rearranging, would have been obscured, which is precisely Peirce’s and Thom’s point and a deep lesson to learn from George Polya’s work on heuristics. Since the linguistic speculation is, as significant, fundamentally conceptual, it “may therefore reveal an inexhaustible fund of true knowledge and new substantial problems, just as it may also produce pieces of mere sophistry” (1958, 95). The criterion, distinguishing substance from sophistry, is ourselves and our sense of rightness and correctness. All we can do is, in the last analysis, draw out our evidence and commitment ourselves on its basis. There is no external, automatic procedure.

The intrinsic selectivity of language, its grasp of and foregrounding of pertinent features, is based upon our veridical powers, rooted in perception, albeit aided by linguistic probes, to grasp the “recurrent features” spoken of in the preceding passage. In other words, for Polanyi and for Peirce, classification is rooted in mental powers which we acknowledge and accredit by using our language confidently. This is the fiduciary component in Polanyi’s theory of knowledge which is his analogue to Peirce’s affirmation of the life-enhancing role of instinct.

Further, and in the same fundamental vein, Polanyi foregrounds the acritical nature of our very appropriation of a language, which parallels the acritical nature of perception and perceptual commitment as such.

Our most deeply ingrained convictions are determined by the idiom in which we interpret our experience and in terms of which we erect our articulate systems. Our formally declared beliefs can be held to be true in the last resort only because of our logically anterior acceptance of a particular set of terms, from which all our references to reality are constructed (1958, 287).

Initially, of course, our acceptance of these terms is not just a-critical but uncritical. Indeed, whether we are concerned with learning a natural language or being initiated into a technical specialized language, our first movement is to pour ourselves into it and use it, just as we pour ourselves into any probe or instrument or set of movements. Later it is possible, as Peirce also noted, because of the very capacity of language to turn back on itself and examine its own content, to re-examine the terms, submit them to critical review and, perhaps, modification, organize their implications, and decide whether we still want to describe ourselves and our world according to the idiom’s exigencies. But in no case can we get a look at the “world” as it really is apart from our means of construing it: that is, talking about it and affirming it to be so and so.

Polanyi argues for a certain form of the linguistic relativity thesis, without the cognate conclusion that languages are fully closed or hermetically sealed. Indeed, languages are systems of hypotheses that ‘bind’ the experiential world, amplifying the primary articulation of the world on the perceptual level.

Different languages are alternative conclusions, arrived at by the secular gropings of different groups of people at different periods of history. They sustain alternative conceptual frameworks, interpreting all things that can be talked about in terms of somewhat different allegedly recurrent features. (1958, 112).

If such is the case, then denotation is an art or a skill.

To classify things in terms of features for which we have names, as we do in talking about things, requires the same kind of connoisseurship as the naturalist must have for identifying specimens of plants or animals. Thus the art of speaking precisely, by applying a rich vocabulary exactly, resembles the delicate discrimination practised by the expert taxonomist. (1958, 81).

The use of an articulate instrument, therefore, is identical in structure with perceptual processes, though it obviously cannot be reduced to perception as such. On Polanyian terms we are embodied in our languages and symbolic systems just as really as we are embodied in external tools and instruments, exemplified in the fusion of perceiver and probe or rider and bicycle. The very material character of the probe defines what can be mediated by means of it and we select appropriate materials for the probe in light of the task at hand. A stainless steel probe does a very different type of work than a cast iron probe. A hardrock maple probe may be preferred to a pine probe, but only by reason of the lack of other suitable materials. In general, to affirm the ‘tool character of language,’ as both Polanyi and Bühler, for example, do, is not to fall under the objection levelled by Gadamer that we are not related to language the way we are related to tools, which can be taken up and put down at will. (See Innis 1988b for a Polanyian reading of Bühler.) Gadamer is right in affirming that we cannot shed language, once learned, and that it is out there ahead of us, predefining for us access to the world, but Polanyi’s great tool analogy is not, strictly speaking, an ‘instrumental’ position. Language for Polanyi is constitutive, even while it is dependent upon its tacit underpinnings and supports. It supports the tacit dimension, potentiating it, even while the tacit dimension supports it.

Polanyi writes of the “curiously insubstantial character of the joint meaning ascribed to a group of objects by a general term....Compared with optical illusions or stereoscopic images, general conceptions are abstract, featureless. The focus in terms of which we are aware of the members of a class appears vague and almost empty” (1962, 168). Likewise for Peirce, all concepts have an essential (appropriately understood) vagueness, as do all images. Vagueness for Peirce is not a sign of weakness, but of richness. The vagueness of any concept, Potter points out in many places (Potter 1966, chaps. 10-12; see also Liszka 1966, 93-98) brings the personal judgment of the language user into play and guarantees a term’s semantic plenitude, as exemplified in Peirce’a analysis of theological language and his development of a ‘theosemiotic.’ (See also on this Orange 1984, Raposa 1989, and Corrington 1993.) Vagueness for Peirce is connected with, but is not identical to, the virtuality of meaning. If all interpretant signs were precise, knowledge could not grow. The very temporal nature of knowledge and the fact that no state of affairs repeats itself exactly and hence that its individualizing traits enter into our concrete knowledge of it leave open the possibility of modifying the term or applying it to a new instance. We are, in fact, simultaneously aware of both the recurrent features and of the qualitative distinctiveness of the individual object. For Peirce, however, not only is language intrinsically vague, but so are percepts and images. The vagueness of theological language, foregrounded by Potter and others, is rooted in the vagueness of perception and of imagination which are as ‘aspectual’ and ‘open’ as language itself is.

IV.

It is clear that for Peirce and Polanyi perception, while exemplifying semiosis and functioning as its primary stratum, is embedded in wider semiosic happenings which I have only been able to hint at in the course of this paper. Perception for both of them takes place in signs and by means of signs and in one sense all signs and sign systems are signs embedded in perception. But it is clear that for neither of them are all signitive happenings perceivings. Nevertheless, the evolution of external sign systems necessitates that they be perceptually accessible in order to be used and it is here that Polanyi’s analytical model, with its distinction between focal and subsidiary awareness and its notions of indwelling and embodiment, is extremely helpful. Since, on Peircean principles, each sign type must have its own qualitative distinctness and through its stream of interpretants makes its object present in a way proper to itself, Polanyi’s notion of indwelling helps us see the ‘biasing’ of perception by our embodiment in different Peircean speculative instruments. These instruments obtain a kind of ‘transparency’ by being made extensions of our embodied sensory systems. They have, as Don Ihde pointed out, an ‘echo’ effect. This echo effect is defined by the tacit background or field of subsidiarily apprehended sign configurations that make the object known in a variety of ways. When Polanyi speaks of the ‘probal’ nature of language in The Tacit Dimension (1966, 7) he is pinpointing the fact that just as we project ourselves out to the end of the probe, passing through it to the object, so we project ourselves into language and through it to the world. But the probe, and language, is an essential condition of accessing the object to begin with, which is not known or even knowable without it. In that sense to embody ourselves in a language--or in fact in any sign system--is to interiorize it and make it part of our mental existence.

Cassirer has a remarkable passage bearing on these issues.

By learning to name things a child does not simply add a list of artificial signs to his previous knowledge of ready-made empirical objects. He learns rather to form the concepts of those objects, to come to terms with the objective world. Henceforth the child stands on firmer ground. His vague, uncertain, fluctuating perceptions and his dim feelings begin to assume a new shape. They may be said to crystallize around the name as a fixed center, a focus of thought. Without the help of the name every new advance made in the process of objectification would always run the risk of being lost again in the next moment. The first names of which a child makes conscious use may be compared to a stick by the aid of which a blind man gropes his way. And language, taken as a whole, becomes the gateway to a new world. (1944, 132)

Indeed, speaking of the shaping and molding of the objective world by speech activity, Cassirer asserts that “Our perceptions, intuitions, and concepts have coalesced with the terms and speech forms of our mother tongue.” (1944, 133) In his (1868) Peirce, for his part writes that “since man can think only by means of words or other external symbols...in fact, therefore, men and words reciprocally educate each other; each increase of a man’s information involves, and is involved by, a corresponding increase of a word’s information” (54). This is a fateful process, involving deep existential and cognitive commitments. Since for Peirce “experience can only mean the total cognitive result of living” (CP 7.538) and since “pragmaticism makes thinking to consist in the living inferential metaboly of symbols whose purport lies in conditional general resolutions to act” (CP 5.402,n.3), the experience of thinking, and its actional matrix, is dependent upon its ‘symbolic’ (semiotic) carriers in which it is embodied. As Polanyi pointed out in his The Tacit Dimension:

All thought contains components of which we are subsidiarily aware in the focal content of our thinking, and all thought dwells in its subsidiaries, as if they were parts of our body. Hence thinking is not only necessarily intentional, as Brentano has taught: it is also necessarily fraught with the roots it embodies. It has a from-to structure. (1966, x)

Not only, then, is perception embodied in language; language is embodied in perception.

List of Works Cited

Bühler, Karl, 1934. Theory of Language. Trans. Donald A. Goodwin. Amsterdam: John Benjamins, 1990.

Bühler, Karl, 1927. Die Krise der Psychologie. Introd. Herbert Rohracher. Frankfurt: Ullstein, 1978 reprint.

Cassirer, Ernst, 1944. An Essay on Man. New Haven: Yale University Press.

Cassirer, Ernst, 1929[1957]. The Philosophy of Symbolic Forms. Trans. Ralph Manheim. Introd. by Charles W. Hendel. Vol. 3, The Phenomenology of Knowledge. New Haven: Yale University Press.

Corrington, Robert S., 1993. An Introduction to C.S. Peirce. Lanham, MD: Rowman and Littlefield.

Dewey, John, 1925. Experience and Nature. Carbondale: Southern Illinois University Press, 1988.

Dewey, John, 1896. “The Unit of Behavior.” In Thayer (1982).

Fisch, Max H., 1986. Peirce, Semeiotic, and Pragmatism. Ed. Kenneth Laine Ketner and Christian J.W. Kloesel. Bloomington: Indiana University Press.

Holenstein, Elmar, 1976. Roman Jakobson’s Approach to Language: Phenomenological Structuralism. Bloomington: Indiana University Press.

Hookway, Christopher. 1985. Peirce. London: Routledge.

Houser, Nathan, and Christian Kloesel, eds. 1992. The Essential Peirce: Selected Philosophical Writings. Bloomington: Indiana University Press.

Innis, Robert E., 1994. Consciousness and the Play of Signs. Bloomington: Indiana University Press.

Innis, Robert E., 1992. “Tacit Knowing, Gestalt Theory, and the Model of Perceptual Consciousness.” Danish Yearbook of Philosophy, vol. 27, 22-43.

Innis, Robert E., 1989. “Giovanni Vailati: Pragmatism and the Analysis of Meaning.” Differentia 3-4: 177-198.

Innis, Robert E., 1988a. “Die Überwindung der Assoziationstheorie durch zeichentheoretische Analyse: James, Peirce, Husserl, und Bühler.” Zeitschrift für Semiotik 10 (4): 149-73.

Innis, Robert E., 1988b. “The Thread of Subjectivity: Philosophical Remarks on Bühler’s Language Theory.” In Achim Eschbach (ed.), Karl Bühler’s Theory of Language. Amsterdam: John Benjamins: 77-106.

Innis, Robert E., 1985. Semiotics: An Introductory Anthology. Bloomington: Indiana University Press.

Innis, Robert E., 1984. “Bühler und Gardiner: Von der Indikation zur Prädikation.” In Achim Eschbach (ed.), Bühler-Studien, vol. 2. Frankfurt: Suhrkamp, 117-55.

Innis, Robert E., 1982. Karl Bühler: Semiotic Foundations of Language Theory. New York; Plenum.

Johnson, Mark, 1987. The Body in the Mind. Chicago: University of Chicago Press.

Köhler, Wolfgang, 1947. Gestalt Psychology: An Introduction to New Concepts in Modern Psychology. New York: New American Library.

Köhler, Wolfgang, 1938. The Place of Values in a World of Facts. New York: New American Library.

Liszka, James Jakób, 1996. A General Introduction to the Semeiotic of Charles Sanders Peirce. Bloomington: Indiana University Press.

Merleau-Ponty, Maurice, 1942[1983]. The Structure of Behavior. Trans. Alden L. Fisher. Foreword by John Wild. Pittsburgh: Duquesne University Press.

Orange, Donna M., 1984. Peirce’s Conception of God. Lubbock, Texas: Institute for Studies in Pragmaticism.

Peirce, C.S., 1976. The New Elements of Mathematics. 4 vols., Ed. Carolyn Eisele. The Hague: Mouton Press.

Peirce, C.S., 1958. Collected Papers of Charles Sanders Peirce. Vols. 7 and 8. Ed. Arthur Burks. Cambridge: Belknap Press of Harvard University.

Peirce, C.S., 1931-35. Collected Papers of Charles Sanders Peirce. Vols. 1-6. Ed. Charles Hartshorne and Paul Weiss. Cambridge: Belknap Press of Harvard University.

Peirce, C.S., 1892. “The Law of Mind.” In Houser and Kloesel (1992).

Peirce, C.S., 1878. “How to Make our Ideas Clear.” In Houser and Kloesel (1992).

Peirce, C.S., 1868. “Some Consequences of Four Incapacities.” In Houser and Kloesel (1992).

Peirce, C.S. The Microfilm Edition of the Peirce Papers. Houghton Library, Harvard University (cited as MS).

Potter, Vincent G., 1996. Peirce’s Philosophical Perspectives. Ed. Vincent M. Colapietro. New York: Fordham University Press.

Potter, Vincent G., 1967. Charles S. Peirce: On Norms and Ideals. Reprinted with a new introduction by Stanley M. Harrison. New York: Fordham University Press, 1997.

Polanyi, Michael and Harry Prosch, 1973. Meaning. Chicago: University of Chicago Press.

Polanyi, Michael, 1969. Knowing and Being. Ed. Marjorie Grene. Chicago: University of Chicago Press.

Polanyi, Michael, 1967. “Sense-Giving and Sense-Reading.” In Polanyi (1969).

Polanyi, Michael, 1966. The Tacit Dimension. New Haven: Yale University Press.

Polanyi, Michael, 1964. “The Logic of Tacit Inference.” In Polanyi (1969).

Polanyi, Michael, 1962. “Tacit Knowing: Its Bearing on Some Problems of Philosophy.” In Polanyi (1969).

Polanyi, Michael, 1961. “Knowing and Being.” In Polanyi (1969).

Polanyi, Michael, 1958. Personal Knowledge. London: Routledge and Kegan Paul.

Polanyi, Michael, 1946. Science, Faith, and Society. With a new introduction. Chicago: University of Chicago Press, 1964.

Raposa, Michael, 1989. Peirce’s Philosophy of Religion. Bloomington: Indiana University Press.

Simon, Josef, 1989. Philosophy of the Sign. Trans. George Heffernan. Albany: SUNY Press, 1995.

Skagestad, Peter, 1981. The Road of Inquiry. New York: Columbia University Press.

Thayer, H.S., 1982. Pragmatism: The Classic Writings. Indianapolis: Hackett.

Thom, René, 1973. “From the Icon to the Symbol.” In Innis (1985).

Vailati, Giovanni, 1911. Scritti. Eds. Mario Calderoni, Umberto Ricci, and Giovanni Vacca. Leipzig: Barth Verlagsbuchhandlung and Florence: Seeber.