The Discourse of Modernism

1 On Method, Discursive Logics, and Epistemology

What a piece of work is man! how noble in reason! how infinite in faculty! in form, in moving, how express and admirable! in action how like an angel! in apprehension how like a god! the beauty of the world! the paragon of animals! And yet, to me, what is this quintessence of dust? man delights not me.

—William Shakespeare, Hamlet

Mark all mathematical heads, which be only and wholly bent to those sciences, how solitary they be themselves, how unfit to live with others, and how unapt to serve in the world.

—Roger Ascham, The Scholemaster

For almost two centuries, the European mind has put forward an unprecedented effort to explain the world, so as to conquer and transform it.¹ Only recently has this European mind—at least commonly—become aware that something may be amiss in this desire for conquest, dominion, and possession. Yielding in part before the growing evidence of its own impotence, we are beginning to realize that the expression and implementation of this desire is the mark of a particular epistemological inflection that is far from the only one possible, or even available. Forms of thought and the desires accompanying them are open to study and to the clarification of what they may seek to hide. They can be changed. Generally speaking, whatever may be the modalities of action with which we endow it, we place the moment of our particular inflection somewhere within the period in the West termed the Renaissance. Anchoring itself principally, though not only, in the close consideration of a literary corpus, the following study will be concerned with that moment when two separate modes of conceptualization become visible. It will seek to follow the production of what I will call one class of discourse from within another. It will attempt to describe the gradual domination of the earlier by the later, to show how the separation itself becomes functional in discourse, how it is elaborated, and how the domination occurs. It will try to reveal some of the necessary ramifications and consequences of that domination.

Such a project is clearly a highly motivated one. I will be tending to reduce the episteme preceding that with whose rise to dominance I am chiefly concerned to the rule of a single class of discourse. I will not try to argue that such a move is anything but a heuristic tactic, and one that conduces, I confess, to telescoping a thousand years of complex development and to flattening out the enormously rich variety expressed especially by the thinkers of the eleventh-thirteenth centuries (though this variety itself, I argue, occurs under the sway of a single discursive dominance). It is indeed perfectly possible that several classes of discourse coexist in the Middle Ages: difficulties and slowness of communication, the separation of the educated clerical stratum from that of the politically dominant feudal aristocracy, and of both from the popular or village culture, seem to ensure that such would be the case. But that there may be several classes of discourse vying, so to speak, for power is not the main point here. Such struggle will doubtless always exist, but what the present essay aims to show is the accession to dominance for our modernity of a single discursive class. That dominance replaces another, and the fact that other discursive relationships may blur the transition must remain for my purposes a secondary consideration, as must also the possibility that a prior break of some kind may have occurred around the eleventh century—though both matters will be discussed briefly and my choice justified in Chapter 2.²

I will argue that a discourse of ‘resemblance’ produces from within itself a certain kind of analytical system. When this becomes a conscious process viewed as promising a ‘truly objective’ knowledge of the ‘real order’ of things, then it is seen as superior to the earlier structure that bore it. An initial displacement is therefore speedily followed, as an essential part of the same process, by a complete replacement. What I will call an analytico-referential class of discourse becomes the single dominant structure and the necessary form taken by thought, by knowledge, by cultural and social practices of all kinds. This dominance still exists today, and I would suggest that it is only by revealing the occultations, the repressions, the willingly accepted ‘traps’ of the newly dominant discourse that we can hope to understand fully its consequences for ourselves. Only by studying the specific discursive response to what was viewed (by such as Galileo, Bacon, Descartes, and Hobbes) as a particular crisis of discourse can we ourselves hope to respond to what is perhaps a similar crisis in our time.

Three events of considerable symbolic significance lie at the doorway of what Alfred North Whitehead called the first century of modern science. The first, in 1600, was the burning of Giordano Bruno, the thinker who in so many ways represents an effort to combine two different modes of thought, two classes of discourse: In his execution there was an unconscious symbolism: for the subsequent tone of scientific thought has contained distrust of his type of general speculativeness.³ By the time he was sent to the stake, Bruno was a figure known throughout Europe, both by his writings and by his extensive travels.

During the same year occurred another event of great importance, though it lacked the striking impact of Bruno’s death or the massive symbolic significance of the third event. Sometime in the course of 1600 William Gilbert published in London his De magnete. P. Fleury Mottelay writes that the work created a powerful impression at the time, though less in England than on the Continent, where Galileo among others seems to have greeted its appearance with enthusiasm. Richard Foster Jones has somewhat modified that estimation by remarking that though Gilbert’s reputation was high with those who were qualified to appreciate his work, it was otherwise never widely noticed.⁴ Be that as it may, the work explained an instrument, the compass, that not only possessed some symbolic value by virtue of its obvious connection with the voyages of discovery (with which it was often credited), but that was also one of the three ‘inventions,’ along with gunpowder and the printing press, which writers of the age constantly urged as evidence of superiority against those who would disparage the moderns. Indeed, the immediate significance of the book lay more in the strong stance it took against the ancients and in its concrete practical demonstration of an efficacious experimentalism than it did in the discoveries in electricity and magnetism recorded there. Certainly the De magnete had been preceded years earlier by the experimental anatomical work of Vesalius, and was very soon to be joined by that of Harvey. Gilbert, however, happened to publish at a particularly auspicious conjuncture.

Still, whatever the importance of the Englishman’s work from this point of view, it was not to produce the same shock as the third event: the publication, just a decade later, of an even more celebrated work, as much polemical pamphlet as record of ‘scientific’ observations. In the latter half of 1609 modern technological thinking was provided with its most eloquent metaphor, as Galileo interposed the distance of the telescope between the human mind and the material world before it, the object of its attentive gaze. It goes without saying that the interposition and the space are both simultaneous creations of the metaphor: the one presupposes the other. They will become the transparent instrumentality of a supposedly ‘neutral’ and ‘objective’ scientific discourse. The metaphor is itself the mark therefore of a particular epistemology created by the discursive activity into which it is inserted. The metaphor was to become distinctly literary and haunt, directly or indirectly, much of the imaginative writing of the following century once the scientist had written it down and published it as the Sidereus nuncius in 1610.

The reaction to the appearance of this text was, as Gérard Simon has recently remarked, a general one of curiosity, incredulity and stupefaction.⁵ Yet Galileo’s telescope only confirms a visual distancing whose entrance into consciousness is especially marked by this metaphor. At least since the considerations of Nicolaus Cusanus and others upon perspective, the abstractions of a continuous interpretation of the material signs (‘signs’ and ‘objects’ being taken as ontologically equal) were reduced by a certain visualization. What Father Ong has been able to write of Ramus’s ‘reduction’ of the linguistic order to visual image, Jean Paris and Michel Beaujour could claim as already the case in Rabelais’s writing.⁶ Nancy Struever can show how Cusanus opposes, in his very writing, the figurai to the abstract as the radical to the traditional. At first this corresponds to a replacement of a kind of potential ‘wholeness’ of knowledge (concluded in the divine Word) by a perspectival and incomplete process of knowing. It opposes an infinite essence to a finite semiotic process: an opposition that is the mark of an incommensurability viewed by Cusa as the proper foundation of all inquiry: ‘finiti ad infinitum nulla est proportio.’⁷ We will see this opposition at work in precise manner in More’s Utopia.

Cusanus, Rabelais, and others represent, as we will also see shortly, a critical moment of passage. But this passage is soon countered and replaced by a new certainty. Once the telescope metaphor becomes generalized, it increasingly comes to preempt other possible forms of sign-functioning. A passage in Frege concerning conception and meaning, in which the telescope aimed at the moon is used as an extended illustrative simile, helps to confirm the importance and implications of the metaphor for the modern episteme and as a symbol for the functioning of signs in its discourse. For the metaphor is thereby demystified, and at a moment (the late nineteenth century) perhaps as critical for the by then long-dominant discourse of analysis and reference it symbolized as the sixteenth century had been for its inception. At almost exactly the same moment, as we will see especially in my concluding chapter, Freud hypostasizes and ‘internalizes’ the metaphor as a true description of the psyche’s relation with the external world and as a means of understanding real mental functioning. In a sense, Galileo, Frege, and Freud symbolize the limits a quo and ad quem of analytico-referential discourse: a leitmotiv that will recur in this study and be considered more fully, as I say, in Chapter 12. The first represents the moment when what had been the scattered elements of crisis congeal unmistakably into a potentially precise discursive direction. The second two mark a moment when the dominance of that discourse had been cast into ineradicable doubt, even though it maintains to this day a considerable vivacity.⁸

The telescope may therefore be taken as a fair representation of what happened to the linguistic sign itself, increasingly able to be defined as an arbitrarily selected transparent instrument placed between concept and object. This ‘fact’ is of particular interest to the present study because it is chiefly concerned with ‘literary’ texts, and because it is widely assumed that literary discourse, in some sense and degree, emphasizes its own status as a linguistic operation. It is indeed precisely in writing a baroque rhetoric/poetic that Emanuale Tesauro, a mere fifty years after the Sidereus nuncius, reverses Galileo’s metaphor, though quite without any historical effect.

The conceit of the very title of Tesauro’s treatise, Il cannocchiale aristotelico (1654), is that metaphor is itself a telescope (cannocchiale). The presence of ingegno, writes Tesauro, is the mark of all superior literary texts. Its constituents are clever conceits and subtle metaphors: which are really all forms of metaphor in general. And this is to be divided into two aspects: perspicacia and versabilità. Perspicacity, he affirms, penetrates the farthest and most minute circumstances of every subject. At the same time, versatility [but also reversibility] speedily compares all these circumstances among themselves and with the subject: it connects and separates them; increases or diminishes them; derives one from the other; outlines the one with the other or with marvellous dexterity puts one in place of the other as the juggler does his stones.⁹

Tesauro emphasizes that what is called ‘truth’ is therefore in its essence a ‘lie’: for what is ‘known’ of any object is but the consequence of these activities of mind. The telescope is nothing but metaphor. An object may be unambiguous inasmuch as it is but an object to be observed, but as soon as it becomes a suggetto of understanding it is, precisely, subject not object. As Ezio Raimondi has remarked, the meaning given to the object of such activities is entirely contingent upon the whim of the operator and the power of the metaphor: "The result of such a view of things becomes, like the human body of which the first pages of the Cannocchiale speak, ‘a page always ready to receive new characters and to erase them’: a great encyclopedia of images and ideas [nozioni; also, meanings], which, embellished with the inventions of art, arouses ‘wit’s lust.’"¹⁰

Only fifty years later, then, the metaphor of the telescope was powerful enough to withstand being stood on its head. Tesauro’s treatise is a commentary on a discourse which, I claim, has by that time become hegemonous, and it is rapidly relegated to the ranks of the unread. One of the (minor) indications of that hegemony is the fact that until very recently indeed Tesauro’s seven-hundred-page treatise was considered worthy of little more than the passing footnote, if that, even in the most extensive literary histories published in Italy. Its existence and the silence which rapidly enveloped it are both indications that it is the mark of a certain passage: a passage from one discursive space to another.

In what follows, the name discourse will refer to a rather large and somewhat ill-delimited definitional field taken over, at least partially, from the studies of Michel Foucault. Firstly, discourse—here—is a coherent set of linguistic facts organized by some enunciating entity.

Such a statement elicits at least two comments. In the first place it is clear that by the use of the term discourse must be meant any semiotic system as practiced, not necessarily a simply ‘linguistic’ one in any narrow sense (as the term has been generally used). I assert that such a broader definition is essential: language, in other words, is but one of the possible materials through and in which discursive order manifests itself. The narrower statement simply refers to the corpus studied here. It enables me to avail myself of certain simplifications—for it is clear that other signifying material and processes cannot necessarily be analyzed in the same way. In the second place the term entity may seem an odd way of talking about the production of sense and in need of some explanation.

I dare not write some such phrase as ‘enunciating subject’ because that immediately implies a particular class of discourse: a class based on the assumption of the identity of an I, the discursive order I will be calling the analytico-referential. It has yet to be shown that such a class of discourse is universal, and what follows disputes any such a priori assumption. The term entity should be relatively ‘safe,’ in part because of its very strangeness. It does not, of course, mean to indicate any incontrovertible origin of discourse. It merely supposes that the practice of discourse, any discourse, appears to depend on a supposition of ens, of ‘being,’ even though such ‘being’ is of necessity itself produced by discourse. Entity is taken here as a meaning produced by all discourse, a meaning which is at the same time the producer of discourse. It is, if you will, an empty metaphor marking the production of discourse (at once produced and producing). The telescope, for example, is one particular way of filling that ‘emptiness.’ Once given specificity (as an I, for example, in the case of the telescope metaphor), the mark becomes a mask, sign of an occultation as well as the form the occultation takes: the 7 of a cogito for which the mark of the seizure of knowledge (concept; concipere; begriff) becomes a neutral, transparent sign. It becomes the index of a discursive class that conceals the necessity, as Foucault puts it, of conceiving discourse as a violence that we do to things, or, at all events, as a practice we impose upon them: a practice, he continues, in which the events of discourse find the principle of their regularity.¹¹

In all discourse not only are the linguistic (semiotic) facts, the signifying elements, organized by some entity, but we may assume that they are not aimless. They show some goal. Ernst Cassirer would have it that they are always a mediation between 7 and the world, but such a view takes the separation of 7 and the world as an a priori and reduces all discourse to the same intentionality. Such a view is therefore entirely indicative of the particular discursive hegemony which the following study will seek to elucidate.

Secondly, then, if discourse speaks of phenomena (no matter what precise inflection one may wish to give to the word ‘phenomena’), it orders them. By this is meant simply that discourse and the material in which it is manifest are never the elements of what might be taken as a neutral mediation. That is so even assuming that discourse may, according to its class, serve a primary function as mediation of things, for example, as opposed to mediation between enunciators. Foucault remarks in this connection that it is not a matter of "treating discourses as groups of signs (signifying elements referring to contents of representations) but as practices that systematically form the objects of which they speak. Of course, discourses are composed of signs; but what they do is more than use these signs to designate things. It is this more that renders them irreducible to the language [langue] and to speech."¹²

Third, elaborated discourse therefore always has a ‘reference’ of some kind, even if it may only be ‘apparent’ to a ‘reader’ accustomed to a different discursive class. Such would necessarily be the case for a discourse whose own elaboration is its reference. (Is this what is being sought by a Jacques Derrida? or by a Roland Barthes—in his last writings?) In one sense, because reference is always the creation of discourse, such is always the case. But the analytico-referential at any rate assumes an exterior and marks that assumption in its own elaboration. Indeed, this reference is always in some sense grasped by discourse. It is—and I insist on the ambiguity of the word because it enables us to use it of discourse in general—it is the relation of discourse. This relation, what Foucault calls the more of language, will have to be given a status ‘beyond’ the language which is but one of its possible materials. This relation is the way in which semiotic systems are used, organized (though it goes without saying that such systems can only be studied in that use and organization: they do not preexist).

The relation of which I am speaking varies, and that variation makes it possible to speak of different classes of discourse. Two such variations immediately come to mind, partly because both have been studied at some length and partly because they appear to correspond to the epistemic change this study aims to describe in some detail. I am thinking of a relation of narration, assuming some commented exterior whose existence as a knowable reality is taken as prior to that of discourse (the discourse of analysis and reference, of historicism, of experimentalism), and of a relation of the ‘formation of patterns’ (the placing of things by their resemblances, by their similarities, for example—though such notions themselves need to be made precise). Lévi-Strauss has suggested that the discourse of the modern natural sciences (a ‘type’ within my analytico-referential) is an ordering of the world by the mind, while what he calls mythical thought is an ordering of the mind ‘by’ the world (bricolage).¹³

The first of these two, narration, is a ‘telling’ (relating) of exterior things and events. It is thus that in the De augmentis scientiarum (especially book II, chapters 1–3) Sir Francis Bacon calls the individual sciences histories (and we still speak, of course, following his practice, of natural history). The second of Lévi-Strauss’s two discourses is a kind of patterning which refuses the ontological and epistemological distinction made in the first between an interior and an exterior, which assumes therefore that all ‘objects’ are ‘signs,’ all signs objects (though the distinction is obviously not one it can make). It assumes that discourse is a part of the ‘world’ and not distinct from it. It gives no special privilege either to the enunciator of discourse or to the act of enunciation.

Gérard Simon, in his recent study of the conceptual foundations of Kepler’s thought (as he terms it), remarks that what gave rise to his researches was the striking similarity between the episteme of resemblance (what I will be calling the discourse of patterning) and what Lévi-Strauss calls the savage mind. He comments of astrology that it operates, once it has been constituted as a form of knowledge, as a system of transformations whose function is to guarantee an ideal convertibility between the celestial and the terrestrial (meteorology), the universal and the individual (genethlialogy), nature and history (universal apotelesmatics).¹⁴ I am struck by the same similarity: indeed, my own textual analyses and the fundamental theoretical schema to be set forward here were completed well before Simon’s researches were available. My argument then will be that there is a gradual disappearance of a class of discursive activity, a passage from what one might call a discursive exchange within the world to the expression of knowledge as a reasoning practice upon the world. This last initially emerges simply as a new and perhaps unfamiliar element within a still dominant class of discourse. Over a period of time other elements gather around this ‘first’ one and come to form a discursive order capable of producing meaningful constructs that are quite new. Such consolidation eventually results in the dominance of what has now become a different discursive class.

Analytico-referential discourse, in the sense I use the phrase here, seems largely the creation of the European sixteenth century. (Chapter 2 will deal with the question of whether such a discursive class existed earlier.) Its name is composed from what may be seen as the fundamental scheme of its functioning, the basic process through which it enables thought and action to occur. During the period of which I will be speaking, a discursive order is achieved on the premise that the ‘syntactic’ order of semiotic systems (particularly language) is coincident both with the logical ordering of ‘reason’ and with the structural organization of a world given as exterior to both these orders. This relation is not taken to be simply one of analogy, but one of identity. Its exemplary formal statement is cogito—ergo— sum (reason—semiotic mediating system—world), but it is to be found no less in the new discursive instauration worked out by Bacon (to be considered in Chapter 6). Its principal metaphors will be those of the telescope (eye—instrument—world) and of the voyage of discovery (self-possessed port of departure—sea journey— country claimed as legitimate possession of the discoverer).

Simultaneously with this claim of logical identity, various devices are elaborated enabling a claim for the adequacy of concepts to represent objects in the world and for that of words to represent those concepts. The outcome is that the properly organized sentence (a concern dominant among grammarians of the second half of the century) provides in its very syntax a correct analysis of both the rational and material orders, using elements that refer adequately through concepts to the true, objective nature of the world. Such is the basic ordering process of analytico-referential discourse.

What appears to have been initially important is that the awareness of the sign’s arbitrary nature grows to a point where that arbitrariness becomes conceptually ‘usable.’ It will be possible to assign a ‘signification’ independent of the object but which, in its denotative precision, allows a belief in its referential adequacy—or, rather, allows a claim for such adequacy to be developed. When the word was felt to inhere in some way in the thing, it was possible to gloss it endlessly. Indeed, no other kind of knowledge was possible. To accumulate bits and pieces of meaning, a variety of nozioni, was to approach understanding of the thing, of its place and that of the glosser in some divine plan—or world soul, as it became in the early European Renaissance. The greater the accumulation of such meanings, the nearer the approach to a wisdom conceived as knowing participation in a totality (as we will see with respect to Paracelsus and Kepler himself).

Between the ‘new’ word and the object to which it is applied there is now no ‘give’—indeed, no give and take. The thing is frozen in its name. The very arbitrariness of the word permits the assimilation of the phenomenon it denotes to a mental order, to a discursive system in which it can be assigned a position and within whose symbolizations and relationships it may be known.¹⁵ It allows the world of phenomena and of concepts (any phenomena and concepts supposed as outside the discourse being used to denote them) to be serialized into a grammar, and to be analyzed by virtue of the signification given to each element in that grammar. Such an operation is impossible when noun (name) and object are perceived as essentially inseparable (as is the case, we will see, in the discourse of patterning, of resemblance). Paracelsus will be able to say that it is by the inherent signs (signatures) that one may know another—what there is in him. One must understand that all things have true and genuine names.¹⁶

In such a discourse name and object are themselves part of an order of which the enunciator is also a part. Paracelsus will call such an enunciator the signator, of whom there are three different levels: Man, Archeus, and the Stars. The names they ‘give’ are not in the least bit arbitrary, and he writes, for example: it should be remarked that the signs signed by man carry with them perfect knowledge and judgment of occult things, as well as acquaintance with their powers and hidden faculties. It is Adam himself, the Protoplast, who was the originator of all skill in the science of signatures.¹⁷ Such a class of discourse places the enunciator within the same structure as englobes name and object as well. Its elements may thus be available to a continuous interpretation, but they cannot be grasped as a whole from within and thereby known in the same sense as they may be by a discourse based on a practice of difference and alterity. The patterns of such a discourse suggest an essence that escapes its enunciator as a whole must its parts. That may explain why, at the very moment of the changing discursive practice, Ramus criticizes Platonic ‘Forms,’ themselves suggestive of just such an essence, with the remark that by Idea nothing else was meant in Plato but logic.¹⁸ The essence in question becomes but the product of discourse. The remark indicates an attempt to reduce one class of discourse to the norms of another.

Ramus, indeed, views Platonic discourse as one of resemblance or of patterning. He criticizes Plato precisely for his assumptions concerning the identity of things, forms, and the words which relate them. It is in such terms that he opposes Plato to Aristotle: Aristotle, says that the mind of man has not made present in our bodies, as Socrates has sometimes argued, a knowledge of all things but rather the faculty and power of knowing them, just as our eyes do not bring with them from our mother’s womb the species of color but only the power of seeing them.¹⁹ Interpretation, which is alone possible in the class of discourse Ramus intends here to criticize, cannot be equated with what we now term ‘knowledge’ or, more precisely, ‘scientific knowledge.’ That is indicated, for example, by the four stages of medieval criticism, terminating as they do in the anagogic, pertaining to the unknowable Truth, Essence, Idea, Logos, or, as it becomes for such as Paracelsus or Kepler, World Soul. This question of medieval patterns will, as I say, be discussed at greater length in Chapter 2. For the present I want to deal more particularly with the emergence of a new dominant discourse and with what appears to have immediately preceded it during the European Renaissance.

I have suggested that the reductive metaphor of the telescope was of considerable importance: it stood for the emergence of a new kind of conceptualizing practice. Nonetheless, like Tesauro later, Galileo himself always recognized that signs (whether mathematical or linguistic) fall in between what is taken as the conceptualizing mind and a world of objects. They fall, to keep the metaphor, in the space of the telescope itself. They can be identified neither with the mind nor with the world, but they are subject to the organization of the former. Such a division of elements is essential to our modern episteme. It is the order of the cogito—ergo—sum: mind—signs— world. But Galileo himself always emphasizes that knowledge is a sign-manipulating activity. This is why he argues that a star seen through the telescope is not the same object as the star seen with the naked eye, or that changing the length of the telescope gives us a different instrument and therefore yet again a different object.²⁰ The terms used by Galileo to describe the scientist’s acquisition of knowledge are always those of violence.²¹

For Galileo discourse represents not the object itself but the distance between the object and the mind perceiving and then conceiving it. For a similar point of view, though Bacon and Descartes both remain ambiguous on the matter, we have to jump (once again) to the late nineteenth century and the discussions of Helmholtz and Hertz—in the context of a crisis in the very discourse Galileo’s work helps to install. The grave problems of representation that such an idea of knowledge poses is squarely faced by Galileo: for we cannot, he affirms, translate into discourse a conception that is an individual, and therefore unusable, representation of a thing perceived as though it were the thing, as Bacon calls it, in itself. That assumes as a consequence that the scientific knowledge of objects is nought but the result of sign-manipulation, and that their ‘truth’ is merely their utility for the betterment of men’s lives. One might say that this awareness within discourse of the individual’s ‘enunciative responsibility’ is an indication that the analytico-referential discourse is as yet but emergent, and still far from dominance. It will in fact take Locke’s discussions about private and public concepts and private and public language before the ‘objectivity’ of such discourse can be finally and formally ‘justified,’ before the presence in discourse of enunciative responsibility can be dispensed with.

In Bacon the solution to the epistemological problem thus posed (for a science wishing to be ‘objective’) is sought in what he terms a gradual and unbroken ascent—in effect, an endless chain of reasoning. Experimental discourse is not at first an attempt to describe the thing itself so much as a description of the human sighting of the thing, of a particular discursive relationship (Bacon) between an object clearly defined in space and time and the mind perceiving it under the same restrictions. There is always the tacit implication that these restrictions are themselves a product of the discourse (mind) that inscribes the description in question. Discourse stands for distance, and it is just that distance that poses the problem. It is largely ‘overcome’ by Descartes.

There is no question but that Descartes, from the Regulae on, is far more radical than this. He chooses to reject ocular evidence altogether, or we should rather say, perhaps, ocular propositions. He prefers to reconstruct them from ‘first principles,’ as he does in the Monde and in the Traité de l’homme. The effect ultimately is to impose an intellectual structure upon the perceived world. Like Bacon, Galileo, and Hobbes, Descartes’s avowed aim is possession and utility. That this is a stage in the development suggested as already underway in Rabelais’s writing seems clear. There, for instance, the Pic-rocholine war, started by a minor quarrel between buyers and sellers, expands by a kind of discursive naming process into an immense conquest of empire. Past, present, and future merge in the deliberations of Picrochole and his lieutenants into a vast nomination of possessions—the entire world: after France, Spain, North Africa, and the Mediterranean basin, it will be better for you first to conquer Asia Minor, Caria, Lycia, Pamphilia, Cicilia, Lydia, Phrygia, Mysia, Bythinia, Charazia, Adalia, Samagaria, Castamena, Luga, and Se-baste, all as far as the Euphrates. As the conversation proceeds the invasion and conquest of the rest of Europe and Asia Minor will be, is being, has already been, achieved. The troops are now waiting at Constantinople. Jean Paris remarks with justice: To name is the wondrous equivalent of possession. To enumerate provokes a kind of hypnosis whereby the distance between the real and the fictional evaporates.²²

The systematic discourse that names and enumerates becomes, replaces, the order of the world that it is taken as representing. In Rabelais’s text, however, the discursive practice remains ambiguous, because the enumeration is a simple accumulation of adjectives, of substantives, of verbs and adverbs: and that simple accumulation prevents the use of a syntax as the analysis of that to which it would refer. In a sense the ‘real’ here becomes a fiction. And whether Rabelais is here parodying a certain discourse or not makes no difference: such a parody would still indicate the existence of what is being parodied. Faced with this kind of discursive practice, our own is likely to find itself at a loss. This syntactically ‘unrestrained’ accumulation is quite foreign to us. Terence Cave has thus recently commented that Rabelais’s writing confronts us with a narrative so plural that no commentary can control it.²³ ‘Pieces’ of the real, so to speak, are seized within a process whose discursivity is not only not concealed but is indeed emphasized as a system of transformations. The similarity with Paracelsus’s signatures is evident. One would be tempted to argue on the other hand that in later Cartesianism, if not in Descartes himself, a ‘fictional’ system becomes the real (that is indeed what happens in the Monde). And if this opposition between the two modes corresponds to that made by Lévi-Strauss between mythical thought and scientific thought it is obviously not accidental.

The Galilean trinity of mind/discourse/phenomena, which opens up various distances within the order of which Rabelais’s text is still to some extent representative, is reduced by later Cartesianism to a dichotomy. Language reveals thought, and inasmuch as that thought can be taken as referring directly to objects (in perception) language can operate as a perfect stand-in for both. To be sure, it is not the object itself, but it is a sufficiently accurate representation for the purposes of discourse, into whose analytical process it can be inserted. When the Port-Royal theories of language and thought argue that the verb is an affirmation of things rather than a simple statement of perception, they are arguing that grammatically speaking (and for them, epistemologically) the affirmation of things contains the statement of perception and subsumes it. This is because the signified (concept/percept) is seen as congruent, if not in fact identical, with the referent (percept/thing). Port-Royal has already traveled some way from the still ambiguous practice of Descartes himself to the certainty of Cartesianism. Word and thing are brought to coincide in the sense that the former is a completely adequate and transparent representation of the latter. Possession by and in the discourse of analysis and referentiality is made possible.²⁴

Instead of an ‘ideal’ exchange between the scientist’s encoding of nature and his perception of it, which would call for a constant readapting of the codifying practice, of discourse (such as Galileo appears to seek), the Cartesians, and the empiricists likewise, will assert their possession of the domain to which any given discourse refers. This was achieved thanks to the establishment of a discursive class (to which such given ‘types’ of discourse belong) determined as true, objective, and the permanent manifestation of universal common sense. This marks a denial, an occultation, of the acknowledgment that the human view of the world is necessarily a ‘perspectival’ one. It marks the assertion of such a view as absolute. To be sure, the problems raised are not immediately simple of resolution. The silence of Cyrano’s narrator at the end of the Voyage dans le soleil might well be interpreted as a recognition of his inability to resolve the manifold questions he has raised (see Chapter 9). But by the beginning of the eighteenth century things will have solidified to such a degree that even a critique of the now-dominant discourse will fall within the same limits.

It is only at the end of the nineteenth century that the analytico-referential discourse of assertion and possession, of permanent and universal human reason, and of absolute objective truth comes to be opposed by another. Michel Foucault is able to remark, for example, that in the analysis of finitude then undertaken, discourse will once again be marked by its own discursive nature. ‘Knowledge’ will be the process of enunciation itself, not the object of that enunciation: To be finite, then, would simply be to be trapped in the laws of a perspective which, while allowing a certain apprehension—of the type of perception or understanding—prevents it from ever being universal and definitive intellection.²⁵

The accession to dominance of a single discursive order is naturally visible in all the discursive practices of the age, though some, no doubt, are more exemplary than others. A good case is that of the development of probability theory. As Ian Hacking has observed, this is an important example of such effects as those being discussed.²⁶ It assumes the possibility of the direct application of an abstract mathematical calculus to concrete phenomena in the world, and it assumes at the same time that a single structure of order is common to an unlimited number of such phenomena. In the literary discursive development to be followed in this volume it is by no means indifferent, therefore, that such a ‘theory’ should become an important element in Cyrano’s novels. (Though I will only discuss it with regard to the Voyage dans la lune, probability seems to be behind the changing form of the ‘tree people,’ for example, in the Voyage dans le soleil.) It is quite possible, indeed, that its appearance in Cyrano’s novels preceded its more formal elaboration by Hobbes, Pascal, and the later seventeenth century. The possibility of elaborating a theory of probability is axiomatic in the analytico-referential discourse of representation. Though in a nonmathematical form, it will have become central to epistemology and to the very concept of the human by the time of Locke’s Essay Concerning Human Understanding in 1690: a probabilistic theory of knowledge and human action based upon the statistical reliability of the accumulation of particulars.

Such, too, is the case for the linear narration of causality, the very model, no doubt, of analysis for this discursive class. Its most simple and schematic literary form will become that of the modern detective novel: as I will suggest later with regard to Cyrano and Defoe, this is really but a simplification of earlier novel forms as it is also an almost programmatic fulfillment of the structure of experimental-ism. Literary ‘effects’ of this kind are by no means marginal or mere digressions. They are exemplary of the elements fundamental to the institution of the dominant discourse of analysis and reference. Their presence in ‘literary discourse,’ however, is especially significant: it is worth reminding ourselves that the very concept of ‘literature’ in the sense it has today is not simply a growth of the new order, but is also one of its ideological foundations.²⁷ In this connection, one example of the effect of this discursive elaboration is of particular interest, and I will glance at it briefly: that of literary criticism.

The very priorities I have been underscoring seem apparent in the ‘Malherbian’ systematization of a literary code (for example) that seeks to provide an adequate representation of a (human) nature claimed as permanently one with itself, the same everywhere, and objectively knowable. That systematization was to be carried out with a vengeance in the second half of the seventeenth century by such as La Mesnardière, d’Aubignac, de Pure, Rapin, Dacier, Valincour, and Boileau in France, by such as Dryden, Rymer, Dennis, Pope and Addison in England. Yet it is perhaps of more interest to know that it was simultaneously with the scientific and philosophical work of Galileo, Bacon, and Descartes that the need was felt in France for a word that would express the ‘inexpressible’ as an objective, describ-able something. The very fact of discourse, which for Galileo expresses the exchange between mind and matter (once that division has been asserted), should make such a need irrelevant. Yet E. B. O. Borgerhoff was able to note a first substantival use of the phrase je ne sais quoi in 1628.²⁸ This will be consecrated in the second half of the century as the object of discussions on the sublime, which will become in turn the foundation of discussions on taste and of the new science of aesthetics in the eighteenth century.

All discourse is now a descriptive instrument capable of being regularized into geometrical, algebraic, and even, perhaps, précieux forms. Literature, it will be argued, is at once an order identical to that of the world (both human and natural) and an exact representation of that world. Hence the attraction of critics to literary rules (initially undermined somewhat by the concept of the sublime, which was itself however to become the object of a scientific discourse) whose possible analogy with an experimental discourse proves so attractive to the eighteenth century. This is clear from Hobbes, and becomes even more so in a critic of the late seventeenth century such as Thomas Rymer, who can write of Aristotle:

The truth is, what Aristotle writes on this Subject [tragedy], are not the dictates of his own magisterial will, or dry deductions of his Metaphys-icks: But the poets were his Masters, and what was their practice, he reduced to principles. Nor would the modern Poets blindly resign to this practice of the Ancients, were not the Reasons convincing and clear as any demonstration in Mathematicks. ‘Tis only needful that we understand them, for our consent to the truth of them.²⁹

This position will be taken to its logical extreme in the course of the following century: as an object for critical study, literary discourse will be equated rigorously with the natural (and ethical) order insofar as its status regarding veracity, actuality, and reality is concerned. The metalanguages that deal with it (grammar, poetics, rhetoric, aesthetics) will thus become the sciences which progressively reveal the knowledge contained in the (discursive) order of literature. In this way, just as Newton was able to trace out a series of truths bound each to the next and thus to establish a true physical science by the observation of nature, so too for every other metadiscourse:

Just as good grammars and good poetics were composed only after there had been good prose and verse writers, so it happened that the art of reasoning was known only in proportion as we had good minds which reasoned well in diverse manners. You may suppose by this that this art has made its greatest progress during the 17th and 18th centuries. Indeed, the true method is due to these two centuries. It was first known in the sciences, where ideas are formed naturally, and determined almost without difficulty. Mathematics are the proof. ... If the Tartars wanted to form a poetics, you may be sure it would be a bad one, because they do not have any good poets. The same is true of logics composed before the seventeenth century. There was then only one way of learning to reason: that was to consider the sciences in their origin and progress.³⁰

Scientific discourse was destined to remain the model and exemplar of all discourses of truth—of all knowledge—with few doubts until the last third of the nineteenth century and even, though with increasing attacks, to the present.

Hobbes sought to apply Galileo’s method to moral and political philosophy and to criticism. The Cartesian cast of Rymer’s affirmation is apparent and quite typical of such efforts to demonstrate the identity of the functioning of the laws of nature and of these fundamental Rules and Laws of literature.³¹ This, too, is the significance of the appearance of the so-called normative grammars toward the end of the seventeenth century. Leonard Bloomfield dismisses the grammarians responsible as contradicting linguistic principles,³² but it is clear that any correspondence between the discourse and an order taken as outside discourse could have no meaning if the linguistic side of the equation could not be fixed. That fact would seem to be behind, for example, Boileau’s criticism of pastoral on the grounds that it used words in such a way as to multiply their meanings and so to make them incomprehensible—a practice which Tesauro had seen as essential to all great literature: for Tesauro the danger was not incomprehensibility but the belief in objective representation. As Foucault puts it: It is thus part of the very nature of grammar to be prescriptive, not by any means because it is an attempt to impose the norms of a beautiful language obedient to the rules of taste, but because it refers the radical possibility of speech to the ordering system of representation.³³

A practice of discourse that uses words as though they are in some way essential and inherent in the object can do with a relatively loose verbal structure. It can undermine its own syntactic forms. It can play with grammar. This kind of discourse is manifest in Rabelais’s text. Utopia, we will see, strives in vain to reach a practice of this sort through rather different techniques. Things have a structure of their own: a material manifestation of the ‘divine soul of the world,’ they may be subject to the interrogation of human discourse, but they cannot be organized by it. Rather, they ‘organize’ it. The new practice requires a tight discursive system because things otherwise become incoherent. For the mind that works in this way, writes Gaston Bachelard (from the vantage point of analytico-referential discourse), "every phenomenon is a moment in the theoretical mind, a stage of discursive thinking, a prepared-for result."³⁴ Bachelard calls it the formal imagination. He opposes it to the material imagination, a mode of thinking that uses the immediate images of things as its code: those very experiences that Galileo scorns as utterly useless. It is what Lévi-Strauss refers to as mythical thought, what I will be calling the discourse of patterning.

Analytico-referential discourse assumes that the world, as it can be and is to be known, represents a fixed object of analysis quite separate from the forms of discourse by which men speak of it and by which they represent their thoughts. This is the case whether the difficulties of analysis then be posed in terms of the world which is to be seized (idealism) or in those of its representation (empiricism). This assumption is basic to all the discourses of neoclassicism. Equally basic is the assumption that the proper use of language will not only give us this object in a gradual accumulation of detail (referen-tiality), but will also analyze it in the very form of its syntactic organization. Moreover that syntax not only analyzes the world but also presents (performs, even, according to Port-Royal) the mental judgment taken to be coincident with that process of analysis in syntax. The assumption of this coincidence of universal reason and general grammar was essential.

After its installation neoclassical discourse does not include within its functioning, within its practice, what we may call the responsibility of enunciation. The assumption of objectivity and the consequent exclusion of whatever cannot be brought to fit its order are necessarily accompanied by the occultation of the enunciating subject as discursive activity and, therefore, of its responsibility for the status of the objects of which it speaks: Galileo’s I becomes Descartes’s we and the objectivity of common speech from Locke to Lavoisier. It is not simply the formal ordering of language, obviously common to all discursive practice, that is in question. It is the fact that after the Renaissance such ordering is taken as isomorphic with the (unique) order of phenomena, whether natural, social, or ethical, and so able to provide (not invent or create) an adequate and even precise knowledge of that order.

It may be that this analytical discourse is only a particular twist in a dialectical discursive search (as Lévi-Strauss calls it) for some form of pragmatic knowledge that does go back to ancient Greece. Husserl argues this way when he affirms that what he calls Western thinking since the Aufklürung is an aberration.³⁵ The present study will question such an interpretation and such a claim. The evidence seems convincing that before the modern advent of analytico-refer-ential discourse some other class of discourse was dominant, that its mode of functioning was quite different from our own, and that it was from within that functioning that the discourse of analysis and reference was produced.

When the linguistic sign was felt to inhere in the thing or to be coextensive with it, it was at once the subject and the object of an interpretative reading of the signatura rerum. Paracelsus (1493–1541) can claim, for example, to be teaching a science in which there are two kinds of operation, one produced by nature itself, in which there is a selected man through which nature works and transmits her influence for good or evil, and one in which she works through other things, as in pictures, stones, herbs, words, or when she makes comets, similitudes, halos or other unnatural products of the heavens.³⁶ Here man is at once a part of ‘nature’ and an intermediary for the effecting of nature’s ‘influence.’ Words and all other signs are in just the same situation: they can be read, while at the same time they produce the reading. Form and essence, writes Paracelsus, are one thing: Whatever anything is useful for, to that it is assumed and adapted. So if Nature makes a man, it adapts him to its design. And here our foundation is laid. For everything that is duly signed its own place should properly be left; for nature adapts everything to its duty. And this is how the signature is provided, which has to do with the signs to be taken into consideration, whereby one may know another—what there is in him. There is nothing hidden which Nature has not revealed and plainly put forward.³⁷ As Ian Hacking observes of the series of signs quoted just previously, the members of the set mentioned by Paracelsus are no longer for us a ‘natural kind,’ namely a collection between which there are manifest family resemblances. The resemblances between words and stones, herbs and comets, are now lost to us.³⁸

The case is exactly similar for our and Kepler’s different conceptions of the planet Mars, which nonetheless remains for each of us unambiguously the same object inasmuch as it is ‘something’ we can observe in the sky. Gérard Simon notes that "what has changed is the manner of classifying it." Indeed, he remarks later that for Kepler the medical theories advanced by Paracelsus were at the same level of progress as the reformed religion and Copernican astronomy. And Kepler viewed that last as inseparable from astrology.³⁹ The scientist has no doubt whatsoever that the earth is endowed with a soul whose existence is proved by the generation of metals, the conservation of terrestrial heat, the exuding of vapors intended to give birth to rivers, rain, and other meteorological effects. All these things prove that its form is not simply conservative, as is the case for the stars, but well and truly vegetative.⁴⁰ He therefore argues that all things participate in a whole (the earth itself being part of a greater vital harmony: hence his faith in astrology) and that all things are signs of the same totality. All reasoning from cause to effect is, under such circumstances, only a special case of analogical reasoning, as metonymy can be read as a special case of metaphor. Given this participation in a whole, reasoning can only be the search for resemblances, the accumulation of patterns: thus the need for men who have had experience in the art of signature and who can discover the genuine names and natures of things.⁴¹

In Paracelsus, as elsewhere in what Hacking terms the low sciences of the age (astrology, medicine, alchemy, mining), no distinction is to be made between a verbal sign and any other kind of sign in nature. All are equally valid manifestations of the relation between man and nature, between the universal and the individual, between the natural and the supernatural: their organization into patterns of resemblance allows these aspects to be transformed into one another. Paracelsus is able to affirm that Nature indicates the age of a stag by the ends of his antlers and it indicates the influence of the stars by their names. Once again Hacking remarks, "in our