SOURCE: Andresen, Julie T. “From Condillac to Condorcet: The Algebra of History.” In Progress in Linguistic Historiography: Papers From the International Conference on the History of the Language Sciences, edited by Konrad Koerner, pp. 187-97. Amsterdam: John Benjamins, 1980.

[In the essay which follows, Andresen highlights the connection between Condorcet's idea of progress and his beliefs about language, including the language of mathematics as the reduction of natural language to its purest form.]

In the 18th century the theory of language intersects with the theory of history. The epistemological framework that unifies these theories, the linguistic and the historic, reveals itself through the works of Condillac (1715-80), one of the most influential thinkers in the study of language, and Condorcet (1743-94), who, during the aftermath of the French revolution, composed the Enlightenment's ‘histoire générale et raisonnée’. In fact, Condorcet's Esquisse d'un tableau historique des progrès de l'esprit humain (1794) marks both the high point in the line of historical thinkers that begins with Fontenelle (1657-1757) and the end point as well (Gusdorf 1966:77). For these philosophes, the history of reason propels the history of culture, and the doctrine of progress evolves from the history of language. History has a specific direction: the progress of a civilization depends upon the implementation of that society's language, which, in turn, parallels the development of a writing system, whose perfection leads to future progress.

The interlocking of historical progress and language, more than merely underscoring the impulse of the Enlightenment towards the unification of knowledge, actually derives its inner structure from a common notion, namely, that of the tableau, a spatial configuration within which time has meaning. Condillac in his Grammaire (1775) stresses the function of language as the method of analyzing, or decomposing, the tableau. With the advent of analytical geometry, a new language, that of numbers, expresses our knowledge of space and spatial relations. Thus mathematics will serve as the model language in the analysis of the tableau. If Condorcet was critical of Condillac the mathematician in the latter's Logique (1780),¹ it is unfortunate that Condorcet never knew Condillac's unfinished work La Langue des Calculs, first published four years after Condorcet's death. In Condillac's last work, the concept of number occupies a central role in his conception of language with respect to the spatial tableau, in much the same way that number signifies for Condorcet the key to understanding of the temporal tableau.

The general historical consciousness that marks the Enlightenment manifests itself during the second half of the eighteenth century in language studies. Here the study of history divorces itself from the questions of genesis, and the theory of language becomes released from its spatial confines when the form of language is sought in historical derivation. With the rise of a historical orientation in the natural and physical sciences at the end of the century and the realignment of the science of language with these disciplines, the Romantics come in full force to invert the relationship of time and space.

The metaphor of the tableau, playing into the notion of tabula rasa, gained popularity with the empiricists of the eighteenth century, as evidenced by the permeation of the word in titles during that period.² As a visual concept, the tableau portrays the mental representation of a sense experience or a thought, which, like the perception of a painting, is instantaneous. Condillac describes this perception as elapsing literally in the ‘wink of an eye’:

Cependant si nous considérons qu'un peintre habile voit rapidement tout un tableau, et d'un clin d'oeil y démêle une multitude de détails qui nous échappent, nous jugerons que des hommes qui ne parlent encore le langage des idées simultanées, doivent se faire une habitude de voir, aussi d'un clin d'oeil, presque tout ce qu'une action leur présente à la fois.

(Condillac 1947-51 I, 430)

As a spatial concept, the tableau captures the unity of an entire thought or perception, again like a painting, containing within its framework a “multitude de détails” to be discerned and unraveled by the perceiver. The tableau thus describes the espace mental, the psychological reproduction of the external world which the senses perceive and strive to understand.

The goal of understanding the tableau predicates itself upon the rationalist conviction that behind the apparent chaos of the world and our perception of it lies a fundamental, rational order. Although belief in innate ideas came under serious attack from the sensationalists and awareness of relativity in the physical world replaced an abstract absolutism, the rationalist postulate of unity and order in the natural world played a dominant role in the 18th century. Reason, no longer the essence of innate ideas, carried the burden of unification and organization (Cassirer 1932:29). Moreover, reason, a constant, assures us that our knowledge that describes that which we perceive is also continuous and unified. The tableau can then extend its range of application from that of just a mental representation to embrace the unity of all knowledge, its globality. By virtue of being instantaneous, unified and organized, the tableau represents the plan de simultanéités, or, to use Condillac's terminology, expresses the “langage des idées simultanées”, the only natural language (I, 430).

Discursive language, which is an art, unravels the simultaneity of the tableau and renders it successive. Thus a second tableau emerges as the plan de successions,³ a dynamic representation of the physical world. For Condillac, language analyzes our thoughts by decomposing the ever changing discourse of the universe, by ordering the movement of the syntagm of nature. Language, as a method of analysis, contains a two-fold process: by returning to the origin of signs, language represents the ongoing ontological process of the mind constantly interacting with the physical world, recreating the paradigmatic tableau. And by making observations and distinguishing among things in order to express a thought in succession, the activity of analysis creates new ideas, or combinations of ideas.⁴ In fact, this activity looms so important that for Condillac there is no thought independent of that organized by language. Language, which for Port-Royal was only the mirror of thought, functions in the 18th century as the instrument or tool that organizes thought itself.

If the rationalists and empiricists disagree as to the function of language, they coincide in their assumption that the knowledge that language provides of the tableau is real, either in terms of its mirroring of rational thought or its representation of sense information. Discursive language for Condillac may be an art, but it is not an artificial reconstitution of the tableau: it is the authentic transcript of reality itself. The possibility of the knowing was not doubted precisely because the relationship between the knowing and the reality known was kept constant by the unifying faculty of reason. Thus the pre-discursive tableau, the one not yet mediated by language, should not differ from the tableau described by language. There is only one, unified reality, just as there is a unified knowledge which describes that reality. It follows, then, that the problem was to establish an exact language to best describe the tableau. For Condillac, the starting point of a rigorous language lies in the principle of the liaison between word-sensory perception.

Because of this belief in the ability of knowledge to reflect reality, the method of science, i.e., the language of science, enjoyed a privileged status in the thought of Condillac and Condorcet. The method of analysis—here used in its mathematical sense to refer to algebra, or calcul—had already proved its merits in the physical sciences and promised similar successes if applies to other would-be sciences. Since for Condillac “toute langue est une méthode analytique, et toute méthode analytique est une langue” (II, 419), the process of establishing the terminology of a science is identical with the analytical procedure of everyday language, where the data of sense experience reduces itself to its components. The combinatorial analysis of mathematics serves as the essential model for scientific reasoning, for the more perfect the language, the more perfect the reduction of the data, hence the more perfect the understanding of the tableau. Indeed, Condorcet associates the method of analysis, or algebra, with the concept of science itself: “cette grande découverte [les méthodes du calcul], en montrant pour la première fois ce dernier but des sciences, d'assujettir toutes les vérités à la rigueur du calcul, donnait l'espérance d'y atteindre, et en faisait entrevoir les moyens” (Condorcet 1970 [1794]:50).

The concept of calcul, however, became synonymous with science only to the extent that the analytical methods of mathematics were assimilated to the more general model of ordering signs through classification (Baker 1975:112). The mathematician operates first by distinguishing similar elements through their relative position, like the articulations of the body, and then by bringing these simple elements into different combinations. Our words, maintained Condillac, are to our ideas what numerals are to numbers, and the more we distinguish and form new classes the more ideas we have (I, 435). As algebra had made possible the ordering of simple phenomena through mathematization, so must a more perfect language order the world of complex phenomena through a philosophical calculus. Condorcet's ideal followed an identical path (1970 [1794]:50):

Une des premières bases de toute bonne philosophie est de former pour chaque science une langue exacte et précise, où chaque signe représente une idée bien déterminée, à bien circonscrire les idées par une analyse rigoureuse.

The perfection of a language paves a direct path to progress. A perfect language, or langue bien faite, differs from a less perfect one only in terms of the extent to which it analyzes, according to Condillac: “les langues ne se perfectionnent qu'autant qu'elles analysent” (I, 435). In other words, there exists no qualitative difference between normal, everyday language and so-called scientific language, only one of degree, the passage to perfection requiring, true to mathematical form, that simple ideas combine to form complex ideas. The perfect language, whose signs form a closed system of signification, opens up to the possibility of infinite combinations. This mathematical model finds a further application in the interpretation of history. From within the spatial orientation of language and its perfection derives the ideas of progress which develops across the temporal axis. Condorcet sketches his tableau historique on the model of the open-ended combinatorial possibilities of arithmetic. For Condorcet (p. 39), the people of the troisième époque make such progress possible with the invention of arithmetic scales:

C'est là le premier exemple de ces méthodes qui doublent les forces de l'esprit humain, et à l'aide desquelles il peut reculer indéfiniment ses limites, sans qu'on puisse fixer un terme où il lui soit interdit d'atteindre.

The complete historical picture apprehends an infinite variety of distinct tableaux, e.g., the tableau des moeurs, des sciences, des beaux-arts.

Progress does not have a temporal movement of its own but results from the relationship of language to the spatial tableau. The diachronic dimension, so to speak, as Condillac views it, is merely a succession of synchronic tableaux: “les tableaux peuvent se succéder, mais chaque tableau est un ensemble d'idées simultanées” (I, 430). Thus, as an outgrowth of its organization of the spatial tableau, language also orders the apparent disorder of time. Progress, in fact, predicates itself upon the perfection of language, which functions to link society in a given space and to span generations, thereby giving continuity to the rupture of time (cf. Foucault 1966:129). And Condorcet's concept of linear progress, modeled on the never-ending algebra of the combination of ideas, reinforces the primacy of the spatial tableau. Here the past has no intrinsic value, for it serves only as a preliminary to the future state of perfection (Gusdorf 1973:420). Like the idea of the perfect language that will usher in this state of perfection, the idea of the future becomes an ‘ideal’. The same method that analyzes the past forecasts the future; calculer equals prédire.

Language seconded by a writing system, specifically alphabetic writing, secures the continuity of the temporal within the spatial. Whereas Condillac suggests that “le langage dans ses progrès, a suivi le sort de l'écriture” (I, 96), Condorcet elaborates his entire theory of history around the invention of writing and the printing press. Alphabetic writing, whose parallel to mathematical symbols attracted Condorcet, proves to be the most efficient system because it uses to best advantage the principle of double articulation which characterizes human speech: “un petit nombre de signes suffit pour tout écrire, comme un petit nombre de sons suffisait pour tout dire. La langue écrite fut la même que la langue parlée” (6). This insight into the nature of everyday language anticipates the structuralists' observation that each language articulates its messages according to a particular code, and each unity of this code is, in turn, articulated in distinctive unities which form a particular system, such that “cette double articulation, qui paraît être une nécessité économique … est en quelque sorte le cadre qui préserve le caractère conventionnel des faits linguistiques” (Martinet 1955:157).⁵ Alphabetic writing specifically draws upon the human tendency towards economy which governs all of human behavior and is, there, the easiest system to learn. This simplicity appealed to various philosophes not only because it makes the valuable skills of reading and writing more democratic,⁶ but also because it makes the keeping of history—so essential to progress—easier.

Although Condorcet (p. 7) hastily claims that the step to alphabetic writing “assura pour jamais les progrès de l'espèce humaine”, he insists over a dozen times in the Esquisse that true progress could only come with the invention of the printing press. The ability of the printing press to “multiplie[r] indéfiniment, et à peu de frais, les exemplaires d'un même ouvrage” (116) facilitates the search for truth and brings together readily all the authorities on a given subject which enables us to judge differing testimony. Condorcet even attributes the decadence of the Greeks and Romans to their ignorance of the art of printing (88).

Beyond the printing press, the key to the future, implicit in Condillac and explicit in Condorcet, remains in the formulation of a universal calculus to analyze and classify the whole of human knowledge. Condorcet (p. 7) conceived of this algebra as the apotheosis of the written language:

Peut-être serait-il utile aujourd'hui d'instituer une langue écrite qui … n'étant employée que pour des raisonnements d'une rigueur logique … fût entendue par les hommes de tous les pays, et se traduisît dans tous leurs idiomes, sans pouvoir s'altérer comme eux, en passant dans l'usage commun.

Condorcet's ideal falls short, however, because of the difficulty of determining what the simple ideas exactly are and how they are combined (Knowlson 1975:153). More than that, this plan reduces natural language, the basis of our knowledge of the world, to a single articulation. Such a reduction is easy to imagine especially if one eliminates the phonetic substance, as Condorcet advocates, and continues only a perfect ideographic writing system. This single articulated language, or ‘dead’ language in Martinet's terms, can never be completely autonomous, as

le système envisagé ci-dussus d'une langue ‘morte’ à idéographie parfaite ne pourrait avoir aucune autonomie réelle, car ceux qui s'en serviraient seraient nécessairement amenés à faire coIncider au moins certains des idéogrammes avec les mots de leur langue au sens propre du terme

(Martinet 1965:18)

While everyday language by its very nature resists the kind of analysis ad infinitum that Condorcet practices, scientific language, other than mathematics, must ultimately have contact with the natural language it derives from.

Condorcet, as Condillac before him, obscures everyday language behind the concept of number. Both sought in numbers the smooth bridge across which natural language blended without rupture into scientific language. Numerical concepts, however, act not as the bridge that connects, but as the line that finally separates the two. Although the numerical concepts of mathematics originate in the numerical signs of everyday language, the former can emancipate themselves from our sensual perceptions and become a purely abstract language (Cassirer 1964 [1923]:186).

Similarly, Condorcet's mechanistic explanation of history, which in essence considers the past and the future as calculable in function of the present, is valid only for a closed system which reason artificially detaches from the whole. If reality is a given and unified whole, it is also eternal. Within this framework, with the correlative theory of progress, time has no specific function. It merely unfolds the consequences of an already established direction of development.

Condillac's view of scientific language falters in the plan for a universal language just as Condorcet's science of the histoire des progrès, a kind of ‘mythistoire de la raison’ (Gusdorf 1973:389), eroded in the tide of the positivist program of examining the facts first and then constructing the theory. Thus the linguistic and historic orientation of Condillac and Condorcet was to have no outlet into the 19th century. Already towards the end of the 18th century, such figures as Charles de Brosses (1709-77) and Antoine Court de Gébelin (1725-84) began to dislocate the theory of history from the theory of language, and, with a dialectic shift, the 19th century inverted the primacy of the spatial over the temporal. When linguists began to study language as a product of temporal development, the science of language was born. Seen from this new perspective, the forms and patterns of language do not merely recapitulate the process of thinking as it unfolds in the interior of the tableau analyzed by language, they are instead the result of historical derivation that exists through time. The problem of writing and writing systems, having lost the theoretical significance it held for the doctrine of progress, was transferred to the important task of interpreting the external data of written records.

Notes

1. Baker (1967 and 1975:114-16) discusses the intellectual and personal differences between Condillac and Condorcet.

2. See Furet and Fontana for a semantic analysis of the word tableau, among others, appearing in the titles of books in the eighteenth century.

3. The opposition plan de simultanéités and plan de successions suggests one aspect of the “pré-structuralisme embryonnaire” (Droixhe 1971:32) of the 18th century. Both Angenot (1971) and Droixhe (1971) examine other parallels between general grammars and structuralism, such as the arbitrariness of the linguistic sign, the notions of economy and valeur, and the principle of analogy.

4. Auroux underscores the importance of the complementary activity of synthesis that plays an equal role in the language theory of the Enlightenment: “La phrase est une synthèse autant qu'une analyse: elle est l'une parce que l'autre” (1973:31). While Beauzée specifies in detail the nature of synthesis in language and makes of it something other than analysis in reverse, Condillac conceives of the activity of synthesis as the mere combination of the results of analysis.

5. The principle of double articulation appears very early in the history of linguistics; Aristotle, for example, formulates the notion of the smallest significant unit, and the methods of Priscian resemble those of American structuralism. Double articulation in the 18th century, however, in relationship to 20th century structuralism derives its significance as yet another aspect of the ‘pre-structuralist’ tendencies of the grammairiens philosophes.

6. The Idéologues were particularly enthusiastic about the democratization of writing. Destutt de Tracy saw the need for an organized educational system: “Or c'est là un petit talent très-facile à acquérir, sur-tout si l'orthographe était régularisée; et tellement facile, qu'avec une bonne organisation sociale, au bout de très-peu d'années, il n'y aurait presque pas un individu dans une nation policée, qui fut privé de cet avantage (II, 285-86). Droixhe places Condorcet's work entirely within a social and political framework and stresses the social benefits of Condorcet's ideal of the universal language (1978:362-366).

References

Aarsleff, Hans. 1975. “The Eighteenth Century, including Leibniz”. Current Trends in Linguistics 13:383-479. The Hague: Mouton.

Angenot, Marc. 1971. “Condillac et le ‘Cours de linguistique générale’”. Dialectica: International review of philosophy of knowledge 25.119-130.

Auroux, Sylvain. 1973. L'Encyclopédie: ‘Grammaire’ et ‘Langue’ au XVIIIe siècle (= Coll. Repères, série bleue ‘Linguistique’ 3.) Paris: Mame.

Badareu, Dan. 1968. “Le ‘calcul’ logique de Condillac”. Revue philosophique de la France et de l'étranger 158.337-60.

Baker, Keith. 1967. “Un ‘Eloge’ officieux de Condorcet: Sa notice historique et critique sur Condillac”. Revue de synthèse 88.227-51.

———. 1975. Condorcet: From Natural Philosophy to Social Mathematics. Chicago & London: Univ. of Chicago Press.

Cassirer, Ernst (1874-1945). 1964 [1923]. Philosophie der symbolischen Formen I: Die Sprache. Darmstadt: Wissenschaftliche Buchgesellschaft.

———. 1932. Die Philosophie der Aufklärung. Tübingen: J. C. B. Mohr.

Condillac, Etienne Bonnot de (1715-80). 1947-51. Oeuvres philosophiques ed. by Georges LeRoy. Paris: Presses universitaires de France.

Condorcet, Marie Jean Antoine de Caritat (1743-94). 1970 [1794]. Esquisse d'un tableau historique des progrès de l'esprit humain. Text reviewed and presented by O. H. Prior, ed. by Yvon Belaval. Paris: Vrin.

Destutt de Tracy, Antoine Louis Claude (1754-1836). 1977 [1801-15]. Eléments d'Idéologie. Faksimilie-Neudruck. Stuttgart-Bad Cannstatt: Frommann-Holzboog.

Droixhe, Daniel. 1971. “L'orientation structurale de la linguistique au XVIIIe siècle”. Le Français Moderne 39:1.18-32.

———. 1978. La Linguistique et l'appel de l'histoire (1600-1800): Rationalisme et révolutions positivistes. Geneva: Droz.

Essar, Dennis. 1976. The Language Theory, Epistemology, and Aesthetics of Jean Lerond d'Alembert. (= Studies on Voltaire and the Eighteenth Century 159.) Oxford: Voltaire Foundation and the Taylor Institution.

Foucault, Michel. 1966. Les mots et les choses: Une archéologie des sciences humaines. Paris: Gallimard.

Furet, François, and A. Fontana. 1968. “Histoire et linguistique: les titres d'ouvrages au dix-huitième siècle”. Langages 11.112-38.

Granger, Gilles-Gaston. 1954. “Langue universelle et formalisation des sciences. Un fragment inédit de Condorcet”. Revue d'histoire des sciences 7:3.197-219.

Gusdorf, Georges. 1966. Les sciences humaines et la pensée occidentale. I: De l'histoire des sciences à l'histoire de la pensée. Paris: Payot.

———. 1969. Les sciences humaines et la pensée occidentale. III: La révolution galiléenne, II. Ibid.

———. 1973. Les sciences humaines et la pensée occidentale. VI: L'avènement des sciences humaines au siècle des lumières. Ibid.

Knight, Isabel. 1968. The Geometric Spirit: The Abbé de Condillac and the French Enlightenment. New Haven: Yale Univ. Press.

Knowlson, James. 1975. Universal Language Schemes in England and France. Toronto & Buffalo: Univ. of Toronto Press.

Leibniz, Gottfried Wilhelm (1646-1716). 1961 [1704]. Nouveaux essais sur l'entendement humain. Frankfurt am Main: Insel-Verlag.

Lovejoy, Arthur (1873-1962). 1957 [1936]. The Great Chain of Being. Cambridge, Mass.: Harvard Univ. Press.

Martinet, André. 1970 [1955]. Economie des changements phonétiques. Bern: Franke.

———. 1965. La linguistique synchronique. Paris: Presses universitaires de France.

Rosenfeld, L. 1972. “Condillac's Influence on French Scientific Thought”. The Triumph of Culture: 18th century perspectives ed. by Paul Fritz and David Williams, 157-68. Toronto: A. M. Hakkert

Saussure, Ferdinand de (1857-1913). 1949 [1916]. Cours de linguistique générale. 4th ed. Paris: Payot.

Turgot, Anne Robert Jacques (1727-81). 1750. “Tableau philosophique des progrès successifs de l'esprit humain”. Oeuvres ed. by Schelle (1913-1923) I, 298-323. Paris.

Warburton, William (1698-1779). 1811 [1751]. Divine Legation of Moses. London.