Language and Myth. Ernst Cassirer. (Translated by Suzanne K. Langer, New York and London, Harper & Bros., 1946.) Pp. x, 103.


Wilfrid S. Sellars
University of Minnesota

Published in Philosophy and Phenomenological Research 9 (1948-49): 326-29.

This small volume is a translation of Sprache und Mythos, written some twenty-two years ago at a time when the late Professor Cassirer was in the midst of his labors on his monumental Philosophie der Symbolischen Formen. It is an off-shoot of the latter work, and together with his recent Essay on Man, provides the English reader with a first hand account, in relative non-technical terms, of Cassirer's approach to philosophy.

The study opens with a critique of negativistic theories of myth, theories which explain the occurrence of myth in terms of error, and specifically that kind of error which is based on the deficiencies (e.g., ambiguity) of language. Misconceptions of this kind are traced by Cassirer to naive realism, the notion that nature confronts the mind "as something directly and unequivocally given" (p. 6). To obtain a correct understanding of the nature of myth, and indeed, of any phenomenon of the human spirit we must "accept in all seriousness what Kant calls his 'Copernican Revolution' " (p. 8). In Cassirer's argument, however, this Revolution is given a "nominalistic" twist. The forms which bind together the thinking which is the generation of world-for-mind, are no longer the fixed pure schematized categories of Kant, but are instead conceived of as essentially bound up with the symbolism that would ordinarily be said to express them, and, consequently, as sharing in the historicity of human utterance.

. . . the special symbolic forms are not imitations, but organs of reality, since it is solely by their agency that anything real becomes an object for intellectual apprehension, and as such is made visible to us (p. 8).

. . . the analysis of reality in terms of things and processes, permanent and transitory aspects, objects and actions (does) not precede language as a substratum of given fact . . . language itself is what initiates such articulations, and develops them in its own sphere (p. 12).

In characterizing his own brand of Critical Philosophy, Cassirer divides the symbolic forms into groups, each of which relates (in a transcendental sense) to a different mode of being.

From this point of view, myth, art, language and science appear as symbols. . . in the sense of forces each of which produces and posits a world of its own (p. 8).

A moment later he tells us that "language, myth, art, and science . . . function organically together in the constitution of spiritual reality," but that "each of these organs has an individual assignment" (p. 9). One regrets that no account of the nature of a symbolic form is given other than the mentioning of the transcendental function they perform. One also regrets that the specific differences of the four types of symbolic forms are nowhere clarified, for this means that in our attempt to understand them we are wholly thrown back on their supposed analogy with the usual categories of the Kantian type of immanent metaphysics; and, indeed, it is only these latter that are elaborated, to any degree, by Cassirer. One could also wish that the organic unity of the several posited worlds had been characterized a bit more fully. Surely the extreme initial implausibility of listing language, science and myth as coordinate (though organically inter-related) forms of spirit requires some greater attempt at explanation than we are given. If it would be a sufficient answer to a German reviewer of Sprache und Mythos to say that these questions are answered in Die Philosophie der Symbolischen Formen, this can hardly comfort the puzzled English reader. A judicious footnote or two would have been of great assistance.

Symbolic forms both exist in the world as the evolving frames of human experience, and produce or posit worlds-for-man. How is their evolution to be understood? Taking the special case of language, Cassirer finds that the question as to its genesis has proved a "veritable monkey puzzle" (pp. 23-31). On the one hand, it has become obvious that an adequate account of the genesis of language cannot begin with the symbolic forms characteristic of theoretical cognition. Even the most rudimentary distinction between name and class term requires a genetic account. On the other hand, such explanations as have been offered have been rooted in naive realism, in the notion that structure and classification and features of the world that need only be noticed and reflected in language. With the adoption of the critical standpoint, these explanations wither away.

It is in the conception of myth as the original symbolic form from which the separate forms as we know them have developed that Cassirer finds the clue to the understanding of the genesis of language.

It is in [the] intuitive creative form of myth, and not in the formation of our discursive theoretical concepts, that we must look for the key that may unlock the secrets of the original conceptions of language (p. 34).

A study of the genesis of myth will throw light on "that emancipation whereby a sound is transformed from an emotional utterance into a denotative one"(p. 35). A considerable proportion, of the volume, consequently, is occupied with the anthropological evidence relating to the evolution of religious myths, from the level of a sheer felt contrast between the "holy" and the "profane" to the sophisticated formulae of Christian theology. To the first stage corresponds "the most primitive level of interjections" (p. 71). The second stage of myth is that of the "momentary god." He quotes from Usener, to whom the term is due ". . . the individual phenomenon is deified, without the intervention of the most rudimentary class concept" (p. 33). However, though the momentary god is "the creation of a moment . . . he becomes an independent being, which henceforth lives on by a law of its own" (p. 35).

The same function which the image of the god performs, the same tendency to permanent existence, may be ascribed to the uttered sounds of language (p. 36).

But there is more here than just an analogy between the origins of language and myth.

Here we encounter a law that holds equally for all symbolic forms, and bears essentially on their evolution. None of them arise initially as separate independently recognizable forms, but every one of them must first be emancipated from the common matrix of myth. (p. 44).

Primitive words and "the image of the god" are the same sort of thing, a fact that finds expression in the gender character of language, name magic, and the role of the word-concept in theology.

Among the questions occasioned by, though not raised in, this volume and which would have to be answered before any final evaluation could be made of the kind of thing Cassirer is trying to do, the following stand out in my mind. Does not the conception of philosophy as a study of "symbolic forms" bring with it an obligation to distinguish between the philosophical and the non-philosophical study of symbolism? If no fundamental distinction can be drawn (other than, perhaps, in degree of generality), does epistemology become an inductive science? Does Kant's Copernican Revolution become a set of propositions in the inductive linguistics of sense-perception? If, on the other hand, a sharp distinction can be drawn, what light is thrown on the philosophical study of symbolism by historical investigations? Again, granting that myth, art, language, and science (why not also morality?) or some similar set can be regarded from the stand-point of the inductive study of symbolisms (psychology, sociology, history) as distinguishable but mutually interacting dimensions of symbol-using; may it not be necessary to conceive of philosophy as a non-inductive study which has no place for any other meaning of "meaning" than designates? or for any other symbolic forms than those analyzed in pure semantics and perhaps (as I have suggested in this journal*) pure pragmatics.

Whatever may be the philosophical value of the study of the origin and development of the syntax of human language, it is beyond doubt an important area for investigation. Surely, however, no claim concerning the genesis of human language can be formulated which is worth consideration, which can be more than metaphor, in the absence of psychological theory concerning the nature of linguistic phenomena, and their connection with learning in both man and animals. These are the topics that must be pursued. They are not only of far more genuine interest to the philosopher than the historical genesis of language; they alone make an understanding of the latter possible. Anthropological evidence as such does not go beneath the level at which man, however primitive, "experiences a world." This involves symbols and meaning, though no conventional signs may be involved. When Cassirer says that ". . . it is [the work of naming] which transforms the world of sense impressions, which animals also possess, into a mental world, a world of ideas and meaning." (p. 28), the implication that symbolic mechanisms come in with "naming" and do not occur at the animal level, he is mistaken. When he speaks of the "tendency to permanent existence of uttered sounds" which perform the function of "concentrating" experience "rather than relating and comparing," and where "the manner of this concentration always depends upon the direction of the subject's interest, and is determined not so much by the content of the experience, as by the teleological perspective from which it is viewed," (pp. 36-7), he is not only using metaphor (which is probably unavoidable in the present state of psychology), but is using metaphor which did not help me, at least, understand "that emancipation whereby a sound is transformed from an emotional utterance into a denotative one.


* "Realism and the New Way of Words," Philosophy and Phenomenological Research, June, 148.

Transcribed into hypertext by Andrew Chrucky, Oct. 7, 1997.