MIND FACTORY: FROM ARTIFICE TO INTELLIGENCE
Mind denotes a whole system of meanings as they are embodied in the workings of organic life ... Mind is a constant luminosity; consciousness is intermittent, a series of flashes of different intensities. --John Dewey, Experience and Nature, 303.
In the Tractatus, Ludwig Wittgenstein poses a dilemma for the concept of universal reason in terms of the incommensurability of different experiential or language systems. For Wittgenstein, this incommensurability is not restricted to the limits of our knowledge about other forms of experience, but is foundational to experience itself, and consequently to the conception of language and mind. Hence, "if a lion could speak," Wittgenstein argues, "we could not understand it." And whilst we may invent analogies to account for such a speech as a lion's, there can be no analogical basis upon which any actual comprehension of that speech could be founded: what we comprehend will always remain an approximation according to our own conception of language. In other words, we are confronted with a limit-experience of translation and interpretation. The same may equally apply to all other forms of experience, something which for Wittgenstein is exemplified by the very condition of literacy; of being with language. "The limits of my language," says Wittgenstein, "are the limits of my world."1
A great deal has been postulated over the years about the technological conditions of something referred to as "artificial intelligence," while analogous models of intelligence have abound, at least since the time of Descartes and the Enlightenment. The transposition from the body-machine idea to the mind-machine idea is one that has accompanied the history of the Industrial Revolution and the advent of computing engines and modern computers, and it is this "transposition" upon which the concept of artificial intelligence has come to devolve--a conception that seeks, above all, to account for a material basis of "mind." Or more precisely, of how the interaction of matter can give rise to a situation of what is called intelligence, beyond mere computability. And yet the question remains, to paraphrase Wittgenstein: if so-called artificial intelligence were possible, would we actually be able to comprehend it or even recognise it?
It is a notable feature of cybernetics, cognitive science and particle physics, that, at key moments in the recent evolution of their disciplines, a number of theoreticians and practitioners have found cause to turn to the work of one particular "experimentalist" writer who, within the institution of literature, is not usually credited with having significant import outside the "humanities" or "applied arts"--and indeed it is a rare enough thing generally for the work of a writer to possess any such importance, regardless of literary fame or merit. And yet this is the case with the two major texts of James Joyce--Ulysses and Finnegans Wake--from which have been derived: the name for the fundamental constituent of the nucleon (Murray Gell-Mann's quark); a model of cognition (Daniel Dennett's Joycean machine); a cybernetic conception of language (Jacques Derrida's Joyceware); a psychoanalytical paradigm (Jacques Lacan's sinthome) and the basis of post-War media theory (Marshall McLuhan's Gutenburg Galaxy, originally conceived as The Road to Finnegans Wake). And yet, within the various philologies, the importance of Joyce's major works--in particular Finnegans Wake--remains largely at a descriptive, analogical level.
This may have something to do with the problem Wittgenstein poses, such that: if, for example, Finnegans Wake is indeed a text, how are we able to recognise it as such? how are we able to understand it? how are we able even to begin to read it?
The history of Joyce studies reveals that this problem has in no way been overcome. Indeed, it is rather a truism that, in the absence of overwhelming interpretation, mediation or translation, Finnegans Wake remains essentially unreadable. That is, unless it is rendered as somehow novelistic, an oedipal drama, a human comedy, or according to some other more or less familiar analogical mode. But this tells us little about the nature of reading and much more about the institution of modern philology--an institution preoccupied with its own discourses to an almost narcissistic degree, such that it is not a question of whether or not Finnegans Wake is unreadable, but whether or not it can be read. And this leads us to the matter of accounting for the Wake's significance beyond the philologies--of its very significant readability outside the field of literary competence. For in this context, it is the astonishing lucidity of Finnegans Wake, and not its obscurity, that most demands accounting for.
The accumulated wisdom of philology, semiotics and the philosophy of language, has not succeeded in accounting for how Finnegans Wake may be intelligible in its own terms. Indeed, how it may be seen to affect by means of signifying-structures something like an intelligence. That Finnegans Wake is not a mimesis of thought processes, but affective of them, in ways that are largely un-remarked--and this is because they pose an apparently insurmountable obstacle to any "reasonable" view of literature, if not of the relation between language and cognition. And yet if Joyce's text is to be considered readable at all, then something like a general condition of literacy must apply, even if this condition is itself unrecognisable. Or, to proceed further, if a condition of literacy, then also one of intelligibility. But what would be a "condition" of intelligibility, if not something which would also be commensurate with a condition of intelligence? This is an important question, and it is one that without which the concept of "artificial intelligence" must remain ungraspable--for it is indeed a matter of art, of artifice, of a techne of language, of the possibility of signification, of what we might call a literate technology.
It is arguable that Finnegans Wake does not in fact require us to break with existing reading habits or to invent new semiologies--rather it requires that we scrutinise the unexamined assumptions belonging to these existing habits and seek to come to a greater understanding of what we call ordinary language and "commonsense": their function, their machinery, as it were, and the possibilities available to them. For Joyce does not pose a language or thought of exceptions in the writing of Finnegans Wake, but rather a general condition of language as such--and this condition must be generalisable across the whole field of existing intellectual habits for it to be possible at all.
The fact that we "read" Finnegans Wake does not, however, relieve us of the responsibility of accounting for what reading, literacy and comprehension actually entail. Above all, if these questions appear to arise from an historical crisis in the understanding of texts--dating from the 1950s and 1960s--or even of a particular text, such as the Wake itself, this should alert us to the nature of the problem as having to do with both the most profound and yet trivial aspects of experience. Profound because they entail the very basis of understanding, of knowledge, of reason. Trivial because they appear to assume the form of an abstraction, a removal from the condition of the ordinary life-world--that is, as something indulgent and "theoretical." But it is precisely this trivial aspect that exercises what is most fundamental to human experience, and that is the possibility of reflective thought. And it is in the nature of reflexivity that so-called abstraction is always inherent to its object, and not a deviation from it: the trivial inheres in the profound, just as language and thought inhere in the materiality--or factography--of the life-world.
Nevertheless, it seems that we are accustomed to making a distinction between apparent facts--such as "linguistic facts"--and our mental attitudes towards them, and consequently we tend to regard Finnegans Wake as firstly a test of attitudes, rather than as constitutive of any kind of fact--largely because of a presupposition that almost nothing about the Wake is verifiable in the first instance. The question immediately obtains as to whether there is such a thing as semantic verifiability, or whether there are merely linguistic structures defined by habitual use or misuse. In any case, it is assumed that the language of Finnegans Wake does not possess any primary "factuality," while such factuality is taken as a mark of so-called "ordinary" language.
But if Finnegans Wake constitutes an exemption from linguistic factuality, how is this so, and what does it reveal about the "nature of language" as such? Moreover, how does thought--acting independently upon this non-factual material--shape the latter into results which are held to be valid, that is, objective (and for which we may assume as examples the primary body of Wakean philology and textual genetics)? More precisely, if Joyce's text is non-factual in this sense--or if it is ambiguous with regard to any "linguistic factuality"--then how can we begin to assume to read Finnegans Wake at all? What is there in its "nature" that permits us to continue treating it as a text? And what does this imply for the belief in linguistic factuality as such; as in that specious category of "ordinary language"?
For if we accept Wittgenstein's contention that there is no such thing as "poetic" language--outside "ordinary" language--then we must also accept that there is no such thing as "ordinary" language: there is only language, which is determined not by a prior recourse to factuality, but by usage. And usage, as distinct from utility, has to do with a measure of the possible, where possibility defines the sole limit of semantic inference.
This brings us, by a vaguely circular route, to the question of intelligibility from which we set out. Intelligibility, it has been argued, devolves upon a general condition of language, of literacy, such that this condition is bound up with materiality as the antecedent of thought. It remains to determine to what extent we may also view materiality as the medium of thought--that is, in the structural relation of linguistic or proto-linguistic elements. Such a consideration evokes the conflict between "mere" coincidence and coherence, a conflict which may be seen as underwriting Finnegans Wake with its countless acrostics, acronyms, mnemonics, codes, neologisms, puns, portmanteaux, sigla, footnotes, marginalia, hand-drawings and so on--and which can be illustrated by reference to any arbitrarily selected passage from that book. This is perhaps most clearly exemplified in the "figures" of HCE and ALP, whose apparent stability and centrality within much of Wakean criticism belies the fact that, in Joyce's text, these tripartite figures are rendered extremely unstable through the seeming limitless permutation and combination of the letters H C E A L and P, and of their supposed cognates or referents--rendering anything like a structuring identity in terms of them fortuitous at best (something Joyce himself constantly draws our attention to).
In this way Joyce also demonstrates--to borrow a felicitous passage from John Dewey's Essays in Experimental Logic--that (linguistic) events "do not cohere; at the most, certain sets of them happen more or less frequently than other sets; the only intelligible difference is one of frequency of coincidence."2 Likewise, Joyce's text indicates that it is the "matter," and not simply the "meaning or content of ideas that is associated." And this extends beyond the probabilistic arrangement of individual lexemes to a generalised typography or tropography, extending through the entire fabric of the "text" and instigating a form of communication or congruence between schema and elementa. In effect, Finnegans Wake enacts a radical parataxis in which each aspect of signification is revealed in the possibility of open communication with every other, and it is on the basis of this radical possibility--seemingly so alien to "ordinary language"--that the Wake affects to demonstrate a general condition of signifiability. That is, of language as such.
Moreover, Joyce's text demonstrates that it is not linguistic factuality but rather linguistic situations that give rise to intelligibility. This is consistent with Dewey's observations from 1903, and it is inviting to speculate that Joyce himself may have been familiar with the Studies in Logical Theory which Dewey edited in that year. According to Dewey: "there is neither coincidence nor coherence in terms of the [individual] elements in any couple or pair of ideas taken by itself. It is only when they are co-factors in a situation or function which includes more than either the 'coincident' or the 'coherent' and more than the arithmetical sum of the two, that thought's activity can be evoked."3
We might say, then, that it is by way of a type of Joycean parataxis that such linguistic situations arise as the antecedent of a general mode of thought. Importantly, it is in the incommensurability of the various factors contributing to a "situation" that the possibility of thought, of intelligibility, resides. What at first presents itself as anomalous--as an aporia or crisis of thought--thus reveals itself as an "engine of possibility": the very basis of thought. Above all, there is nothing "fictive" about this situation. The situation as such, as Dewey insists, "is clearly 'objective.' It is there; it is there as a whole; the various parts are there; and their active incompatibility with one another is there. Nothing is conveyed at this point by asserting that any particular part of the situation is illusory or subjective, or mere appearance; or that any other is truly real." Consequently, the conflicting nature of this active incompatibility "is not only objective in a de facto sense (that is, really existent), but is objective in a logical sense as well; it is just this conflict which effects a transition into the thought situation--this, in turn, being only a constant movement towards defined equilibrium."4
The material situation defined by Joyce's paratactic writing can otherwise be regarded as an effect of what we might call a literate technology, wherein a certain mechanics of thought finds its antecedent. And if we accept the congruence of terms such as artifice and techne, we may be tempted to posit within this parataxis the conditions of an "artificial intelligence" as antecedent to intelligence as such. But that would be to lose sight of the implications of Joyce's text.
In the Pola Notebook of 1905, Joyce himself offers a critique of the Aristotelean notion of techne as merely prosthetic or artificial, arguing instead that Aristotle's phrase "e tekhne mimeitai ten physin" had been falsely rendered as "Art is an imitation of Nature." According to Joyce: "Aristotle does not here define art, he says only 'Art imitates nature' and meant the artistic process is like the natural process."5 Likewise Finnegans Wake demonstrates the inherent relatedness of techne and any form of literate praxis or poiesis. It is this latter aspect of the Wake that indicates the way in which techne is always linked to a certain possibility of invention, and above all at that point at which "ordinary" intelligibility appears to break down. "The limits of my language," to return to Wittgenstein's phrase, describe a boundary beyond which intelligibility enters into a crisis--but only in the sense that this crisis itself is a necessary techne of invention, of possibility, as the antecedent of thought. This is one of the implications of Dewey's insistence that:
There is always an antecedent to thought in experience of subject-matter of the physical or social world, or the previously organised intellectual world, whose parts are actively at war with each other--so much so that they threaten to disrupt the situation, which accordingly for its own maintenance requires deliberative redefinition and re-relation of its tensional parts. This redefining and re-relating is the constructive process termed thinking ..."6
Between thought and intelligibility, then, exists a technological movement of redefinition and re-relation, in whose dynamic we might say intelligence itself is somehow constituted. This "ambi-dual" dynamic, as Joyce says, comprises the situation we might call Joyce's "mind factory" (FW 282.R1), and it is on this basis that Finnegans Wake assumes its singular importance in the step away from analogical thought, towards the material foundations of an experiential or experimental thought.
© Louis Armand, 2005
|volume 6, issue 1, 2005|