A final post this series on language concerns the mutation of consciousness triggered by the shift away from the word as signature of the thing in the world, to word as signfier. I again use Foucault's Order of Things as reference, (even though it like all other text has its critics).

According to Foucault's history this shift from words as signature of the thing, or as pure representation occurs through the advent of “General Grammar” in the 17^th and early 18^th century. He traces the shift through the work of such Grammarians as Bopp and Rask

Foucault's method here is different than most historians who trace history through the linearity of ideas or the impact of individuals by establishing historical continuity. Rather he follows a method he calls an Archaeology. The study of Archaeology proceeds with the analysis not of individuals or specific events, but rather through the systems of thought and apprehension present in a given era which determine its discourse. He calls these eras epistemes. Moreover, Foucault says : “Archaeology is much more willing than the history of ideas to speak of discontinuities, ruptures, gaps, entirely new forms of positivity, and of sudden redistribution  (Archaeology of Knowledge)

Foucault does not treat the history of grammar - later to become the science of philology and linguistics - in isolation but rather excavates it in the episteme of the Classical Age. He connects the history of Grammar with the study of Natural History -which will later become Biology - and the History of Wealth -which will later become the science of Economics-. Its probably not to much of a stretch to see this threefold study of history in comparative terms to Sri Aurobindo's method of analysis as follows: biology (physical) economics (vital) linguistics (mental)

Here Foucault describes the interrelatedness of these three domains:

The history of grammar is not the projection into the field of language and its problems of history that is generally that of a reason or of a particular mentality , a history in any case that it shares with medicine, mechanical services, or theology but that it involves a type of history -a form of dispersion in time, a mode of succession of speed of deployment or location,- that belongs to it alone even it is not unrelated to other types of history, (Archaeology of Knowledge)

and here is how he traces back the origins of the study of general grammar:

.. the practice of the history of comparative grammar was to rediscover -beyond Bopp and Rask- earlier research into the filiation and kinship of language it was determined how much Anquuetil-Duperron contributed towards the composition of the Indo-European domain it was to uncover the first comparison of Sanskrit and Latin Conugations it may even lead back to Harris or Ramus (Archaeology of Knowledge)

One must first understand how Foucault perceives the episteme of the Renaissance which precedes the Classical Age in which General Grammar emerges to fully appreciate the mutation of consciousness that occurs. Knowledge in the Renaissance was concerned with knowledge by Resemblance and Similitude, while the Classical age is annunciated through an knowledge of  Difference,

( It may be worthwhile to explore the similarities and differences between knowledge by Resemblance and Similitude in the European tradition and Sri Aurobindo's metaphors, tropes, discourse, and the Indic sources he draws on, in speaking of a knowledge by Identity) :

The divergence of epistemes of the Renaissance and the Classical Era is explored below. Its annunciating figure is Don Quixote who reads the world through a book.

“The four modes of resemblance are pretty straightforward: 1) convenience = spatial proximity, which relies upon and breeds resemblance; 2) emulation = resemblance at a distance; 3) analogy = resemblance of relation; man is center of world; 4) sympathy = resemblance provoking spatial and qualitative change.

The thesis goes something like this. The episteme of the 16th century was founded on similitude. All phenomena and designative modes were based on a manifold mirroring and interplay of analogies and affinities. The Renaissance world was a kind of weave, folding upon itself, forming a chain of vital resemblances through which alone individual facts or objects could find a meaningful location. This principle of analogy made of the eye both a receptor and source of light, almost tangibly threaded to the object contemplated. It was thought that language works first because it is a system of autonomous signs and second because it is a kind of "organic mirror" in which every named or inferred thing has its exact counterpart. A perfectly comparable system of emblematic similitudes obtained in Renaissance biology and in the 16th century view of the inherent, one might well say magical, worth and singularity of gold.

The episteme of the 17th-18th century Classical period is radically different. It involved "an immense reorganization of culture," a literal re-orientation of the space in which Western consciousness perceived subject and object, reality and dream. The old kinships between knowledge and divination, the mirroring reciprocities of language and fact, break off. Now, instead of similitude, the crucial instrumentality is representation. Foucault seems to mean by this that words are now entirely transparent and arbitrary counters. Thus, to say things, to name them, is to put them in a kind of necessary order. The "necessity" seems to derive from the fact that Classical man now sees objects in a logical space or framework.

The language of the Classical age is caught in the grid of thought, woven into the very fabric of its unrolling. It is not an exterior effect of thought, but thought itself." In other terms, knowing and speaking are interwoven. Every speech act, every mental proposition "down to the least of its molecules" becomes an exact way of naming. Grammar is a kind of tracing-paper laid across the ordered contours of the world. Hence the primary impulse of Classical thought and science toward taxonomy. The classificatory genius of the great botanist Linnaeus represents the true spirit of the age.

Signature is the being of the Renaissance sign, a resemblance that is sign of, that indicates, points the way toward, another resemblance. Obviously here we have an infinite task of chasing resemblances around, which yields abundant, yet empty knowledge: since everything has a hidden resemblance to everything else, you can find (know) resemblances everywhere, but what do you find everywhere? Just another resemblance. The 16th C "condemned itself to never knowing anything but the same thing, and to knowing that thing only at the unattainable end of an endless journey."

Since resemblances spoke through signatures that were parts of the world and were themselves resemblances, then science, magic, and erudition (reliance on ancient authorities), which are all forms of interpreting natural signs, are on an equal footing; this is not because of credulity, but because of the episteme.” Gary Gutting, Michel Foucault's Archaeology of Scientific Reason (Cambridge UP, 1989)

Now here is Foucault on the history of language as signature prior to the Classical age:

“In its original form, when it was given to men by God himself, language was an absolutely certain and transparent sign for things, because it re­sembled them. The names of things were lodged in the things they desig­nated, just as strength is written in the body of the lion, regality in the eye of the eagle, just as the influence of the planets is marked upon the brows of men: by the form of similitude. This transparency was des­troyed at Babel as a punishment for men. Languages became separated and incompatible with one another only in so far as they had previously lost this original resemblance to the things that had been the prime reason for the existence of language. All the languages known to us are now spoken only against the background of this lost similitude, and in the space that it left vacant. There is only one language that retains a memory of that similitude, because it derives in direct descent from that first vocabu­lary which is now forgotten; because God did not wish men to forget the punishment inflicted at Babel; because this language had to be used in order to recount God's ancient Alliance with his people; and lastly, because it was in this language that God addressed himself to those who listened to him. Hebrew therefore contains, as if in the form of fragments, the marks of that original name-giving. And those words pronounced by Adam as he imposed them upon the various animals have endured, in part at least, and still carry with them in their density, like an embedded fragment of silent knowledge, the unchanging properties of beings:

Thus the stork, so greatly lauded for its charity towards its father and its mother, is called in Hebrew Chasida, which is to say, meek, charit­able, endowed with pity . . . The horse is named Sus, thought to be from the verb Hasas, unless that verb is rather derived from the noun, and it signifies to rise up, for among all four-footed animals the horse is most proud and brave, as Job depicts it in Chapter 39 [29].

But these arc no more than fragmentary monuments; all other languages have lost these radical similitudes, which have been preserved in Hebrew only in order to show that it was once the common language of God, Adam, and the animals of the newly created earth.

But though language no longer bears an immediate resemblance to the things it names, this does not mean that it is separate from the world; it still continues, in another form, to be the locus of revelations and to be included in the area where truth is both manifested and expressed. True,it is no longer nature in its primal visibility, but neither is it a mysterious instrument with powers known only to a few privileged persons. It is rather the figuration of a world redeeming itself, lending its ear at last to the true word. This is why it was God's wish that Latin, the language of his Church, should spread over the whole of the terrestrial globe. And it is also why all the languages of the world, as it became possible to know them through this conquest, make up together the image of the truth. Their interlacing and the space in which they are deployed free the sign of the redeemed world, just as the arrangement of the first names bore a likeness to the things that God had given to Adam for his use. Claude Duret points out that the Hebrews, the Canaans, the Samaritans, the Chaldeans, the Syrians, the Egyptians, the Carthaginians, the Phoe­nicians, the Arabs, the Saracens, the Turks, the Moors, the Persians, and the Tartars all write from right to left, following 'the course and daily movement of the first heaven, which is most perfect, according to the opinion of the great Aristotle, tending towards unity'; the Greeks, the Georgians, the Maronites, the Serbians, the Jacobites, the Copts, the Poznanians, and of course the Romans and all Europeans write from left to right, following 'the course and movement of the second heaven, home of the seven planets'; the Indians, Cathayans, Chinese, and Japanese write from top to bottom, in conformity with the 'order of nature, which has given men heads at the tops of their bodies and feet at the bottom'; 'in opposition to the aforementioned', the Mexicans write either from bottom to top or else in 'spiral lines, such as those made by the sun in its annual journey through the Zodiac'. And thus 'by these five diverse sorts of writing the secrets and mysteries of the world's frame and the form of the cross, the unity of the heaven's rotundity and that of the earth, are properly denoted and expressed' [30]. The relation of languages to the world is one of analogy rather than of signification; or rather, their value as signs and their duplicating function are superimposed; they speak the heaven and the earth of which they are the image; they repro­duce in their most material architecture the cross whose coming they announce - that coming which establishes its existence in its own turn through the Scriptures and the "Word. Language possesses a symbolic function; but since the disaster at Babel we must no longer seek for it -with rare exceptions [31] - in the words themselves but rather in the very existence of language, in its total relation to the totality of the world, in the intersecting of its space with the loci and forms of the cosmos. Hence the form of the encyclopedic project as it appears at the end of the sixteenth century or in the first years of the seventeenth: not to reflect what one knows in the neutral element of language - the use of the alpha­bet as an arbitrary but efficacious encyclopaedic order does not appear until the second half of the seventeenth century [32] - but to reconstitute the very order of the universe by the way in which words are linked together and arranged in space. It is this project that we find in Gregoire's Syntaxeon artis mirabilis (1610), and in Alstedius's Encyclopaedia (1630); or again in the Tableau de tous les arts liberaux by Christophe de Savigny, who contrives to spatialize acquired knowledge both in accordance with the cosmic, unchanging, and perfect form of the circle and in accor­dance with the sublunary, perishable, multiple, and divided form of the tree; it is also to be found in the work of La Croix du Maine, who en­visages a space that would be at once an Encyclopaedia and a Library, and would permit the arrangement of written texts according to the forms of adjacency, kinship, analogy, and subordination prescribed by the world itself [3 3]. But in any case, such an interweaving of language and things, in a space common to both, presupposes an absolute privilege on the part of writing.” (Order of Things)

It is only by the emergence of General Grammar, which approx 150 years after it arose became philology (end of 19^th century), that the primacy of the representational values of words are reduced and they become merely a functioning placeholder in the being of language, which coheres through grammatical structure..

The condition of the possibility of modern thought. (philology in this instance) is not as an unchanging as transcendental structure of human consciousness  but time as a historically and discursively a priori. A word continues to represent but the functional or representative content of the word no longer accounts for the being of the word, its essential architecture. Instead the grammatical totality with its own internal history to which the word belongs constitutes the being of the word. The word is displaced in its role as a name, as a representation, and finds its function determined by it place larger linguistic coherence. It becomes an element in a larger grammatical structure. (Michael Mahon)

Foucault's Nietzschean Genealogy: Truth, Power, Subject p72/73

Now here is Foucault from the Order of Things on the rupture which followed the advent of General Grammar

“Language is the original form of all reflection, the primary theme of any critique. It is this ambiguous thing, as broad as knowledge, yet always interior to representation, that general grammar takes as its object. But a certain number of consequences must at once be drawn here.

1. The first is that it is easy to see how the sciences of language are divided up in the Classical period: on the one hand, rhetoric, which deals with figures and tropes, that is, with the manner in which language is spatialized in verbal signs; on the other, grammar, which deals with articulation and order, that is, with the manner in which the analysis of representation is arranged in accordance with a sequential series. Rhetoric defines the spatiality of representation as it comes into being with language; grammar defines in the case of each individual language the order that distributes that spatiality in time. This is why, as we shall see, grammar presupposes languages, even the most primitive and spontaneous ones, to be rhetorical in nature.

2. On the other hand, grammar, as reflection upon language in general, expresses the relation maintained by the latter with universality. This re­lation can take two forms, according to whether one takes into considera­tion the possibility of a universal language or that of a universal discourse. In the Classical period, what was denoted by the term universal language was not the primitive, pure, and unimpaired speech that would be able, if it were rediscovered beyond the punishment of oblivion, to restore the understanding that reigned before Babel. It refers to a tongue that would have the ability to provide every representation, and every element of every representation, with the sign by which it could be marked in a univocal manner; it would also be capable of indicating in what manner the elements in a representation are composed and how they are linked to one another; and since it would possess the necessary instruments with which to indicate all the possible relationships between the various seg­ments of representation, this language would also, by that very fact, be able to accommodate itself to all possible orders. At once characteristic and combinative, the universal language does not re-establish the order of days gone by: it invents signs, a syntax, and a grammar, in which all conceivable order must find its place. As for universal discourse, that too is by no means the unique text that preserves in the cipher of its secret the key to unlock all knowledge; it is rather the possibility of defining the natural and necessary progress of the mind from the simplest repre­sentations to the most refined analyses or the most complex combina­tions: this discourse is knowledge arranged in accordance with the unique order laid down for it by its origin. It traverses the whole field of know­ledge, though as it were in a subterranean manner, in order to reveal, on the basis of representation, the possibility of that knowledge, to reveal its origin, and its natural, linear, and universal link. This common de­nominator, this foundation underlying all knowledge, this origin expressed in a continuous discourse is Ideology, a language that duplicates the spon­taneous thread of knowledge along the whole of its length:

Man, by his nature, always tends towards the nearest and most pressing result. He thinks first of his needs, then of his pleasures. He occupies himself with agriculture, with medicine, with war, with practical politics, then with poetry and the arts, before turning his thoughts to philosophy; and when he turns back upon himself and begins to reflect, he prescribes rules for his judgement, which is logic, for his discourse, which is grammar, for his desires, which is ethics. He then believes himself to have reached the summit of theory...;

but he perceives that all these operations have 'a common source' and that 'this sole centre of all truths is the knowledge of his intellectual faculties'[11].

The universal characteristic and ideology stand in the same opposition to one another as do the universality of language in general (which arranges all possible orders in the simultaneity of a single fundamental table) and the universality of an exhaustive discourse (which reconstitutes the single genesis, common to the whole sequence of all possible branches of knowledge). But their aim and their common possibility reside in a power that the Classical age attributes to language: that of providing adequate signs for all representations, whatever they may be, and of establishing possible links between them. In so far as language can repre­sent all representations it is with good reason the element of the universal. There must exist within it at least the possibility of a language that will gather into itself, between its words, the totality of the world, and, inversely, the world, as the totality of what is representable, must be able to become, in its totality, an Encyclopaedia. And Charles Bonnet's great dream merges at this point with what language is in its connection and kinship with representation:

I delight in envisaging the innumerable multitude of Worlds as so many books which, when collected together, compose the immense Library of the Universe or the true Universal Encyclopaedia. I con­ceive that the marvellous gradation that exists between these different worlds facilitates in superior intelligences, to whom it has been given to traverse or rather to read them, the acquisition of truths of every kind, which it encompasses, and instils in their understanding that order and that concatenation which are its principal beauty. But these celestial Encyclopaedists do not all possess the Encyclopaedia of the Universe to the same degree; some possess only a few branches of it, others possess a greater number, others grasp even more still; but all. have eternity in which to increase and perfect their learning and develop all their faculties [12].

Against this background of an absolute Encyclopaedia, human beings constitute intermediary forms of a composite and limited universality: alphabetical encyclopaedias, which accommodate the greatest possible quantity of learning in the arbitrary order provided by letters; pasigraphies, which make it possible to transcribe all the languages of the world by means of a single system of figures [13]; polyvalent lexicons, which establish synonymies between a greater or lesser number of lan­guages; and, finally, rational encyclopaedias, which claim to 'exhibit as far as is possible the order and concatenation of human learning' by examining 'their genealogy and their filiation, the causes that must have given rise to them and the characteristics that distinguish them'[14]. Whatever the partial character of these projects, whatever the empirical circumstances of such undertakings, the foundation of their possibility in the Classical episteme is that, though language had been entirely reduced to its function within representation, representation, on the other hand, had no relation with the universal except through the intermediary of language.

3. Knowledge and language are rigorously interwoven. They share, in representation, the same origin and the same functional principle; they support one another, complement one another, and criticize one another incessantly. In their most general form, both knowing and speaking consist first of all in the simultaneous analysis of representation, in the discrimina­tion of its elements, in the establishing of the relations that combine those elements, and the possible sequences according to which they can be un­folded. It is in one and the same movement that the mind speaks and knows: 'It is by the same processes that one learns to speak and that one discovers either the principles of the world's system or those of the human mind's operations, that is, all that is sublime in our knowledge'[15]. But language is knowledge only in an unreflecting form; it imposes itself on individuals from the outside, guiding them, willy-nilly, towards notions that may be concrete or abstract, exact or with little foundation. Know­ledge, on the other hand, is like a language whose every word has been examined and every relation verified. To know is to speak correctly, and as the steady progress of the mind dictates; to speak is to know as far as one is able, and in accordance with the model imposed by those whose birth one shares. The sciences are well-made languages, just as languages are sciences lying fallow. All languages must therefore be renewed; in other words, explained and judged according to that analytic order which none of them now follows exactly; and readjusted if necessary so that the chain of knowledge may be made visible in all its clarity, without any shadows or lacunae. 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. Destutt de Tracy once observed that the best treatises on logic, in the eighteenth century, were written by grammarians: this is because the prescriptions of grammar at that time were of an analytic and not an aesthetic order.

And this link between language and knowledge opens up a whole historical field that had not existed in previous periods. Something like a history of knowledge becomes possible; because, if language is a spon­taneous science, obscure to itself and unpractised, this also means, in return, that it will be brought nearer to perfection by knowledge, which cannot lodge itself in the words it needs without leaving its imprint in them, and, as it were, the empty mould of its content. Languages, though imperfect knowledge themselves, are the faithful memory of the progress of knowledge towards perfection. They lead into error, but they record what has been learned. In their chaotic order, they give rise to false ideas; but true ideas leave in them the indelible mark of an order that chance on its own could never have created. "What civilizations and peoples leave us as the monuments of their thought is not so much their texts as their vocabularies, their syntaxes, the sounds of their languages rather than the words they spoke; not so much their discourse as the element that made it possible, the discursivity of their language.

The language of a people gives us its vocabulary, and its vocabulary is a sufficiently faithful and authoritative record of all the knowledge of that people; simply by comparing the different states of a nation's vocabulary at different times one could form an idea of its progress. Every science has its name, every notion within a science has its name too, everything known in nature is designated, as is everything in­vented in the arts, as well as phenomena, manual tasks, and tools[16].

Hence the possibility of writing a history of freedom and slavery based upon languages [17], or even a history of opinions, prejudices, super­stitions, and beliefs of all kinds, since what is written on these subjects is always of less value as evidence than are the words themselves [18]. Hence, too, the project of creating an encyclopaedia 'of the sciences and arts', which would not follow the connecting links of knowledge itself but would be accommodated in the form of the language, within the space opened up in words themselves; for that is where future ages would have to look to find what we have known or thought, since words, in their roughly hewn state, are distributed along that mid-way line that marks the adjacency of science to perception and of reflection to images. It is in them that what we imagine becomes what we know, and, on the other hand, that what we know becomes what we represent to ourselves every day. The old relation to the text, which was the Renaissance defini­tion of erudition, has now been transformed: it has become, in the Classical age, the relation to the pure element of the language.

Thus we see glowing into life the luminous clement in which language and learning, correct discourse and knowledge, universal language and analysis of thought, the history of mankind and the sciences of language freely communicate. Even when it was intended for publication, the knowledge of the Renaissance was arranged within an enclosed space. The 'Academy' was a closed circle which projected the essentially secret form of knowledge onto the surface of social configurations. For the primary task of that knowledge was to draw speech from mute signs: it had to recognize their forms, interpret them, and retranscribe them by means of other graphic signs which then had to be deciphered in their turn; so that even the discovery of the secret did not escape this array of obstacles, which had rendered it at once so difficult and yet so precious. In the Classical age, knowing and speaking are interwoven in the same fabric; in the case of both knowledge and language, it is a question of providing representation with the signs by means of which it can unfold itself in obedience to a necessary and visible order. Even when stated, knowledge in the sixteenth century was still a secret, albeit a shared one. Even when hidden, knowledge in the seventeenth and eighteenth centuries is discourse with a veil drawn over it. This is because it is of the very nature of science to enter into the system of verbal communications[19], and of the very nature of language to be knowledge from its very first word. Speaking, enlightening, and knowing are, in the strict sense of the term, of the same order. The interest shown by the Classical age in science, the publicity accorded to its controversies, its extremely exoteric character, its opening up to the uninitiated, Fontenelle's popularization of astronomy, Voltaire reading Newton, all this is doubtless nothing more than a sociological phenomenon. It did not provoke the slightest alteration in the history of thought, or modify the development of know­ledge one jot. It explains nothing, except of course on the doxographic level where it should be situated; but its condition of possibility is never­theless there, in that reciprocal kinship between knowledge and language. The nineteenth century was to dissolve that link, and to leave behind it, in confrontation, a knowledge closed in upon itself and a pure language that had become, in nature and function, enigmatic - something that has been called, since that time, Literature. Between the two, the intermediary languages - descendants of, or outcasts from, both knowledge and lan­guage - were to proliferate to infinity.

4. Because it had become analysis and order, language entered into relations with time unprecedented hitherto. The sixteenth century ac­cepted that languages succeeded one another in history and were capable of engendering one another. The oldest were the mother languages. The most archaic of all, since it was the tongue of the Eternal when he ad­dressed himself to men, was Hebrew, and Hebrew was thought to have given rise to Syriac and Arabic; then came Greek, from which both Coptic and Egyptian were derived; Latin was the common ancestor of Italian, Spanish, and French; lastly, 'Teutonic' had given rise to German, English, and Flemish [20]. In the seventeenth century, the relation of language to time is inverted: it is no longer time that allots languages their places, one by one, in world history; it is languages that unfold representations and words in a sequence of which they themselves define the laws. It is by means of this internal order, and the positions it allots to its words, that each language defines its specificity, and no longer by means of its place in a historical series. For language, time is its interior mode of ana­lysis, not its place of birth. Hence the paucity of interest shown by the Classical age in chronological filiation, to the point of denying, contrary to all the 'evidence' - our evidence, that is - the kinship of Italian or French with Latin[21]. The kinds of series that existed in the sixteenth century, and were to reappear in the nineteenth, were replaced by typologies, typologies of order. There is the group of languages that places the subject being dealt with first; next the action undertaken or under­gone by that subject; and last the object upon which it is exercised: as witness, French, English, Spanish. Opposed to these is the group of languages that places 'sometimes the action, sometimes the object, some­times the modification or circumstance first': for example Latin, or 'Slavonian', in which the function of words is indicated, not by their positions, but by their inflections. Finally, there is the third group made up of mixed languages (such as Greek or Teutonic), 'which have something of both the other groups, possessing an article as well as cases'[22]. But it must be understood that it is not the presence or absence of inflections that defines the possible or necessary order of the words in each language. It is order as analysis and a sequential alignment of representations that constitutes the preliminary form and prescribes the use of declensions or articles. Those languages that follow the order 'of imagination and interest' do not determine any constant position for words: they are obliged to emphasize them by means of inflections (these are the 'transpositive' languages). If, on the other hand, they follow the uniform order of re­flection, they need only indicate the number and gender of substantives by means of an article; position in the analytic ordering of the sentence has a functional value in itself: these are the 'analogical' languages [23]. Languages are related to and distinguished from one another according to a table of possible types of word order. The table shows them all simultaneously, but suggests which were the most ancient languages; it may be admitted, in fact, that the most spontaneous order (that of images and passions) must have preceded the most considered (that of logic); external dating is determined by the internal forms of analysis and order. Time has become interior to language.

The history of the various languages is no longer anything more than a question of erosion or accident, introduction, meetings, and the mingling of various elements; it has no law, no progress, no necessity proper to it. How, for instance, was the Greek language formed?

It was Phoenician merchants, adventurers from Phrygia, from Mace­donia and Illyria, Galatians, Scythians, and bands of exiles or fugitives who loaded the first stratum of the Greek language with so many kinds of innumerable particles and so many dialects [24].

French is made up of Latin and Gothic nouns, Gallic constructions, Arabic articles and numerals, words borrowed from the English and the Italians - as journeys, wars, or trade agreements dictated [25]. This is be­cause languages evolve in accordance with the effects of migrations, victories and defeats, fashions, and commerce; but not under the impul­sion of any historicity possessed by the languages themselves. They do not obey any internal principle of development; they simply unfold repre­sentations and their elements in a linear sequence. If there does exist a time for languages that is positive, then it must not be looked for outside them, in the sphere of history, but in the ordering of their words, in the form left by discourse.

It is now possible to circumscribe the epistemological field of general grammar, which appeared during the second half of the seventeenth century and faded away again during the last years of the following century. General grammar is not at all the same as comparative grammar: the comparisons it makes between different languages are not its object; they are merely employed as a method. This is because its generality does not consist in the discovery of peculiarly grammatical laws, common to all linguistic domains, which could then be used to display the structure of any possible language in an ideal and constricting unity; if it is indeed general, then it is so to the extent that it attempts to make visible, below the level of grammatical rules, but at the same level as their foundation, the representative function of discourse - whether it be the vertical function, which designates what is represented, or the horizontal function, which links what is represented to the same mode as thought. Since it makes language visible as a representation that is the articulation of another representation, it is indisputably 'general'; what it treats of is the interior duplication existing within representation. But since that articulation can be accomplished in many different ways, there must be, paradoxically, various general grammars: French, English, Latin, German, etc. [26]. General grammar does not attempt to define the laws of all languages, but to examine each particular language, in turn, as a mode of the articulation of thought upon itself. In every language, taken in isola­tion, representation provides itself with 'characters'. General grammar is intended to define the system of identities and differences that these spontaneous characters presuppose and employ. It must establish the taxonomy of each language. In other words, the basis, in each of them, for the possibility of discourse.

Hence the two directions that it necessarily takes. Since discourse links its parts together in the same way as representation does its elements, general grammar must study the representative function of words in relation to each other; which presupposes in the first place an analysis of the links that connect words together (theory of the proposition and b particular of the verb), then an analysis of the various types of words and of the way in which they pattern the representation and are distin­guished from each other (theory of articulation). However, since discourse is not simply a representative whole, but a duplicated representation that denotes another representation - the one that it is in fact representing -general grammar must also study the way in which words designate what they say, first of all in their primitive value (theory of origins and of the root), then in their permanent capacity for displacement, extension, and reorganization (theory of rhetoric and of derivation).(Order of Things p84-91)