publisher colophon


The present work examines aspects of the emergence and development, of the consolidation and growth to dominance, of modern Western discourse—or, as I will be calling it, the “analytico-referential.” The book sets up a model to describe how one dominant discourse gives way to another. In particular, it shows the creation and development of the various elements fundamental to analytico-referential discourse, and it demonstrates at the same time the necessary occultation of other elements whose visible presence in discourse would subvert its overt aims (though such occultations are not “intentional”). The basic claim is that the kind of epistemic development seen in the early seventeenth century, preceded by at least a century of crisis and followed by a half century of consolidation, is being echoed in our own time and is leading to a similarly significant and complete conceptual change. I also claim that the change of our own time can be understood only in the light of the earlier one, and that it is urgent to understand just what kind of arguments and assertions were necessary to the establishment of the “discursive class” now being placed in question. The principal corpus I use in establishing these claims is composed of science fictions and utopias drawn from the critical historical moment of the European Renaissance and Neoclassicism. These brief assertions require some immediate explanation and clarification.

The term “discourse” refers to the way in which the material embodying sign processes is organized. Discourse can thus be characterized as the visible and describable praxis of what is called “thinking.” For thinking is nothing but the organization of signs as an ongoing process. Signs themselves may be ‘defined’ provisionally as the nondiscrete ‘elements’ composing the process toward meaningfulness that itself is both defined by and defining of what signs are. So cumbersome a turn of phrase is necessary for the moment, in order to avoid such simplifying definitions as: signs are mediators between concepts and things, or concepts and other concepts; or signs are discrete units of signifying systems (usually taking natural languages as exemplary cases) that in combination with other signs of the same system refer to other such units or to nonsignificant objects outside the system; or, more simply, a sign is a unit of meaning (undefined) that stands for some other meaningful or nonmeaningful unit.

All such definitions as these last refer the idea of the sign to a particular order of conceptualization, to what used to be called a particular Weltanschauung. The excessively cumbersome ‘definition’ offered above implies that the only evidence whatsoever not only for thinking but for all human knowing and doing, without exception, is our common use of signs. We may therefore say that such use is human action, and that it is all human action of whatever kind. No doubt such action (process, exchange, production, or whatever) is not definable only in such terms, but no matter in what additional terms one may wish to define it, it is specifically human action only to the extent that we can make it meaningful. That is the sense in which such action may be considered first and foremost a matter of the production and movement of signs.

The definition is cumbersome in particular because I wish to indicate that the use and definition of signs themselves change as the discursive process moves along. Precisely the why, how, and wherefore of such change are what this book seeks to explore in the context of a specific historical period in the West: such change, too, is the reason for the provisional nature of the ‘definition.’ By the end of the volume it will be found impossible to give any universally valid definition of the sign, of meaningfulness, or of their associated terms, because such concepts will be clearly seen as inseparably bound up in the constant development of discourses. With this proviso, one might well parody the beginning of Wittgenstein’s Tractatus and affirm that our use of signs is all that is the case. A similar view is expressed by C. S. Peirce and M. M. Bakhtin: all human action, all human mental life, and indeed the universe as a whole, insofar as it relates to things human, are a matter of the production, interpretation, and interrelating of signs.

Signs are not disembodied ideas, therefore. They are available only in specific material and in specific processes. Such material and such processes are themselves caught within a network of contextual relations, within a definable if exceedingly complex environment, from which they are inseparable. They are the stuff of history, of society in movement, of conceptual processes in flux, of economic and political forces in motion, of developing artistic forms, and so on. Actual processes of material production, for example, are as much systems of meaning as are natural languages—and only so are they available to us as knowledge and action (though the processes through which they come to be meaningful are obviously not those of language). What I will be calling “discursive classes”—the particular network of relations ordered by given practices of signifying systems (to some degree what Michel Foucault has called epistemes)— are therefore specific to a time, a place, and a society, though it is such discursive class in turn that permits these three to be defined.

I use the singular “class” here because I will be arguing that, generally speaking, one discursive class is dominant at any given time and place—barring specifiable moments of transition. That is to say that one such class provides the conceptual tools that make the majority of human practices meaningful: meaningful in the sense that they may be analyzed into their manner, nature, and purpose, and may be related to one another as defining the ‘human.’ Yet where one class is dominant, there may well be others that are contemporaneous with it. Indeed, I will suggest that the dominant theoretical model is apparently invariably accompanied by a dominant occulted practice. This is composed of widespread activities (though the phrase is awkward) that escape analysis by the dominant model, that do not acquire ‘meaningfulness’ in its terms, that are therefore in the strictest sense unthinkable.

When elements from such occulted practice start to become tools for analysis, then the previously dominant model is gradually rendered inoperative, and there is a passage to new dominances, making use both of quite new elements and of already emergent ones, and accompanied by residual elements that only slowly pass out of sight and mind. Such a passage occurs when internal contradictions of the dominant model begin to prevent its effective functioning, when it begins to produce strong alternative elements of discourse (as we will see in Kepler, for example), and when the dominant occulted practice begins to become conceptually useful. This assertion implies, of course, that what was just said about signs as being the whole case applies absolutely only insofar as such practices are grasped in their historicity: that is, as already meaningful for us looking ‘back’ at them. Only as past activities can all human practices be considered to be discursive. This argument is akin to Peirce’s constant assumption that something can be done about a set of facts only if that set conforms to a (previously known) generalization. The practice “occulted” for us in our time escapes signification by very definition. Nonetheless, it may be hoped that this way of conceptualizing what I am calling discursive practices can enable us to grasp the nature of such practice, its place, and its manner of functioning.

All this is extremely abstract, and the first two chapters will make the matter somewhat more concrete. The Middle Ages may thus be characterized as ruled by a dominant theocratico-theological model, accompanied by an occulted feudal practice (“occulted” in the simple sense that it is not generally used to explain and analyze most human—and nonhuman—practices). Notably in Machiavelli certain of these feudal relations begin to provide elements for what will become a “capitalist” analytic, leading directly to the dominance of what I call analytico-referential discourse—a term clarified in Chapter 1 but whose development is the object of this entire volume. It is accompanied, I suggest, by occulted relations of socioeconomic production. In the nineteenth century Marx will play the role Machiavelli played in the sixteenth, and while this book is mainly concerned with the emergence and growth to dominance of analytico-referential discourse, it will also pay some attention, as I say, to the crisis and limits ad quem of that discourse (brought on in part by the accession to meaningfulness of the previously occulted socioeconomic relations of production).

Such a theoretical model allows for both continuity and change, for moments of stability and passages of discontinuity. It also allows for some ‘explanation’ of how and why such moments and passages alternate. In a single volume one cannot hope to cover all discursive domains, all kinds of discourse, for it is clear that in order to show the working of such a model and such processes of change and development some close analysis of punctual examples of the use of sign systems is essential: the corpus must be precise, relatively homogeneous, and sufficiently complex to avoid oversimplification; it must make use of an unsimplified signifying material (natural languages being clearly exemplary); and it must be open enough to the overall discursive environment to make contextualization fairly easy. For an initial study, therefore, the corpus of utopias and science fictions already indicated seems almost ideal. It provides precision and homogeneity. It admits of facility neither in its elaboration nor in its signifying material. It furnishes a clear meeting ground for political and scientific theory, for economic and philosophical elaboration, not to mention for explorations in ‘linguistics’ and ‘psychology,’ literature and aesthetics.

The claims of the book therefore exceed rather considerably what might seem to be suggested by the apparently narrow corpus explored in Chapters 3 to 11. The book tries to show the emergence and development of elements fundamental to the discourse variously called “positivist,” “capitalist,” “experimentalist,” “historicist,” “modern,” and so on—depending on which particular type of discourse is the object of the description. The critical method and many of its assumptions are derived principally from the work of Michel Foucault, though not uncritically: certain of their aspects are further developed here, the notion of epistemic “rupture” (recently pasted over by Foucault himself) is replaced with the attempt, already explained in brief, to show how one episteme (or “discursive class,” as I call it) in fact develops out of another, and other aspects are provided with the added precision afforded by the use of a rather narrowly defined textual corpus—even if I do not always stick to it.

This theoretical basis, its present application, and its epistemological ramifications and implications are explored in the first two chapters. Here is justified the hypothesis that the European seventeenth century saw the rise to dominance of a new class of discourse, the analytico-referential, and the gradual occultation of others that might have been available as alternative modes for human thought and action. The chief of these for the period immediately preceding is that of “patterning,” a class of discourse whose functioning is explored briefly by reference to such as Paracelsus, Rabelais, Cusanus, Agricola, Bruno, and to the contemporary work of Claude Lévi-Strauss, Gérard Simon, and Ian Hacking, among others. This is also opposed in Chapter 2 both to the Greek experience and to that of the Middle Ages. The following nine chapters are composed of close analyses of the texts of my literary corpus, these texts being constantly referred to parallel developments in philosophy, political theory, and science, as illustrated most notably by Kepler himself, Bruno, Descartes, Galileo, Bacon, Hobbes. The analyses are at once an investigation of the ordering principles of the texts studied and a demonstration of the general epistemic development that is the volume’s hypothesis.

The conclusions explored in Chapter 12 essentially suggest (1) that these developments are indeed general discursive ones, and situated quite precisely in a specific historical moment, and (2) that such notions as those of truth and valid experiment (in science), of referential language and representation (in all types of discourse), of possessive individualism (in political and economic theory), of contract (in sociopolitical and legal theory), of taste (in aesthetic theory), of common sense and the corresponding notion of concept (in philosophy) are, in fact, hypostatizations of a particular discursive system. It is also suggested throughout the book, suggestion that culminates in a brief proposal for an analysis of the controlling discursive constraints of Freudian theory and practice, that the series of necessary occultations and traps accompanying this development have now been brought back to the surface of discourse and are responsible for a contemporary crisis—at the same time that they propose suggestions for its solution.

The debts I have incurred over the years are numerous. I thank first of all the students who participated in seminars at Yale University and at the universities of Montreal, Toronto, and British Columbia—some concerned with this material, all with the underlying theoretical matrix. Without the stimulation they provided, this volume might not have seen the light of day, certainly not in its present form. I owe thanks as well to Jacques Neefs and the students in his seminar at the Université de Paris VIII (Vincennes) in 1971–72, who, by inviting me to present the ideas in what are now the three chapters centered on Cyrano, enabled me to profit from their responses and questions.

It is impossible to name all the colleagues, friends, and acquaintances who have commented upon versions of parts of the volume, and all the others who, in one way or another, have provided encouragement and occasions for fruitful exchanges of ideas. Nonetheless, it is a pleasure to record those to whom I feel most indebted: Pierre Beaudry, Michel De Certeau, Robert Elbaz, Françoise Gaillard, Pierre Gravel, Claude Imbert, Wladimir Krysinski, Louis Marin, Georges May, Christie V. McDonald, Walter Moser, Patricia Parker, Chantal Saint-Jarre, Michel Serres, Paul Zumthor. The general debt to Michel Foucault is self-evident. I also thank Nancy S. Struever, William J. Kennedy, and especially Peter Haidu, whose admirably attentive readings of earlier versions of the complete text enabled me to make essential corrections and improvements. Eugene Vance gave most generously of his time and knowledge of the Middle Ages in reading the penultimate version of Chapter 2, allowing me at once to avoid egregious errors and to avail myself of texts that provide more thorough support for what, in so short a space, must still remain little better than a series of rather contentious assertions.

One further acknowledgment of this nature is a necessary pleasure here. Only subsequent to his reading of Chapter 2 did I learn from my colleague Gene Vance of the existence of a remarkable series of unpublished papers by F. Edward Cranz of Connecticut College, dealing with the Middle Ages and early Renaissance and with their intellectual relations both with Antiquity and with our own time. Upon reading them, I was both astonished by the proximity of our positions (and sometimes their expression) concerning the probable impossibility of our ever fully ‘understanding’ Antiquity, and disconcerted by the fact that he, along with others, placed the conceptual break with which I am here concerned with Anselm and Abelard around 1100 a.d., whereas I insist it comes much later. My Chapter 2 argues that while ‘emergent’ discursive elements may be found long before the sixteenth century, it is not possible to maintain that anything like our modern episteme begins to develop prior to that time. On the other hand, I do think it is possible to allow for the development of a new episteme around 1100 that is itself replaced during the sixteenth century: the disagreement is thus less considerable than I had at first thought. The fact remains (I argue) that at least one vital element of our episteme is lacking until very much later than the twelfth century. In any case, the existence of these important papers needed mention, and I am grateful to Gene Vance for making the first of them available to me and to F. Edward Cranz for sending additional ones and permitting me to use them in my text and bibliography. I would also like to acknowledge the care taken by Bernhard Kendler and Kay Scheuer in shepherding this book to publication, and the excellent copy editing achieved by Jane Reverand.

A few paragraphs of Chapter 1 and much of Chapter 5, though in very different form, appeared in Yale French Studies, no. 49 (1973). Chapter 3, again in a quite different version, appeared in Sub-Stance no. 8. (Winter 1974). I thank the editors of these journals for permitting me to use much of this material again. The research of which this book is one result (another being my recently published Tragedy and Truth, written, so to speak, ‘inside’ the present work, and now forming something of a footnote to it: the demonstration of a special case) was begun in 1971–72, during my tenure of a Morse Fellowship from Yale University. I owe a deep debt of gratitude to the previous year’s Morse Fellowship Committee of Yale College, and I am happy to record it. I am also grateful to the administration of the Université de Montréal, who, by granting me a sabbatical leave in 1977–78, made it possible to complete the research. A Canada Council Leave Fellowship in 1978, though granted for the writing of Tragedy and Truth, also allowed me the possibility of working in libraries of the first rank, from which this book has greatly benefited.

Our children have had to put up with a lot of absence, unavailability, and preoccupation, and I am grateful to them for not complaining overmuch. But finally, the main gratitude is due Jean Reiss, whose forbearance is beyond words, marital pact, or the reckoning of amity.



Note to the 1985 printing

While it has been possible to correct here misprints and at least one glaring mistake, material considerations have entirely prevented any elaborate revisions or additions. I can, however, make use of the occasion to indicate five texts whose appearance is especially apposite. Too late for me to have used either of them in the 1982 publication of this work, there have appeared two new English translations of Campanella’s Città del sole (p. 169, n. 2). The first is a bilingual edition: The City of the Sun: A Poetical Dialogue, tr. Daniel J. Donno (Berkeley and Los Angeles, 1981) ; the second is The City of the Sun, tr. A. M. Elliott and R. Millner (London, 1982). Even more recently, a French translation of Kepler’s Dream (p. 143, n. 8) has been published: Le Songe . . . , tr. Michèle Ducos (Nancy, 1984). More substantially, the arguments proposed on pp. 58 ff., concerning the concept of “will” and the network of concepts related to it, have received important confirmation in a volume by Albrecht Dihle, The Theory of Will in Classical Antiquity (Berkeley, Los Angeles, and London, 1982). Dihle’s Sather Lectures, here published, discuss the concept in Antiquity and down to the neo-Platonists and early Church Fathers. His arguments, thoroughly grounded in a wealth of primary material, are a strong buttress for my own. Finally, the discussion of psychoanalysis on pp. 363–76 has been considerably extended in my “Science des rêves, rêves de la science,” Etudes Françaises, 19, no. 2 (Autumn 1983), pp. 27–61. This text is itself, I hope, the precursor to a yet longer, monographic, treatment of the subject.

T.J. R.


Additional Information

Related ISBN
MARC Record
Launched on MUSE
Open Access
Creative Commons
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.