A perceptual model of the Whorfian thesis.
It may well be the case that Sapir's inclusion in the "Whorfian" context is erroneous. It is not simply that Sapir was more cautious in his speculation than Whorf: Sapir was vigorously speculative but at the same time far more circumspect than Whorf in his estimate of the rule of language in the formation of ideas.
For Sapir the relation between perception, thought, language, and speech appears to be as follows: upon reaching the mental level where the collation is possible, the data of sensation are classified into categories which are implicit in the language system. The process of classification is also the process of conceptualization; these conceptualizations are set into mutual relations at least in part by the structure of the language system. The single significant elements of speech are symbolic of the concepts; the flow of speech represents a record of the established mutual relations of the concepts. From this process, man moves to the formation of a worldview.
Sapir considered concepts to be abstractions from the world of experience: our real world is constructed from these concepts. If language classes either correspond to or symbolize concepts, then articulations within a language system are vocal descriptions of the real world. If our collection of concepts depends upon language for its existence, then we might conclude that for Sapir the real world is an abstraction from the experiential world, made possible by language; further, since languages differ, a corresponding difference is to be found in the real worlds of the participants of the various languages.
Whorf, on the other hand, hypothesized a radical, illogical, and ultimately untestable version of linguistic relativity. It is the purpose of this paper to examine the Whorfian thesis as it exists apart from the work of Sapir.
For Whorf, the existence of a mind before the existence of language is a simple and self-evident truth. However, he seems to look upon the products of the mind as of little value without language. The mind thus seems to be some sort of producer of a malleable substance (speaking metaphorically), which is shaped into the various elements of our world-view. Whorf seems to conceive of linguistic symbolism as the prime cause of the arrangement of sensory data. In 1936, he wrote:
... It is possible to have descriptions of the universe, all equally valid, that do not contain our familiar contrasts of time and space. The relativity viewpoint of modern physics is one such view, conceived in mathematical terms, and the Hopi Weltanschauung is another and quite different one, nonmathematical and linguistic. Thus, the Hopi language and culture conceals a METAPHYSICS [upper case letters are Whorfs], such as our so-called naive view of space and time does; yet it is a different metaphysics from either. (1)
Whorf appears to be saying: (1) any world-view conceals a metaphysics (perhaps is a product of a concealed metaphysics would be an accurate interpretation). (2) Both the metaphysics and the resultant world-view are linguistic in origin: they are a product of the classificatory action of language on the stream of sensory experience. Thus for Whorf an individual's metaphysics is made up of the basic conceptions which he possesses about the world of experience, and which he uses to build his world-view. Whorf concludes that the thoughts of individuals in different language groups are themselves different in the sense that they constitute different conceptions of the world about us.
Whorf's own statements contain a metaphysics of which he seemed not wholly aware. In his statements concerning the segmenting effect of language, Whorf appears to use "nature" as a name for a reality that exists beyond the senses; a reality which is diverse enough to be in some sort of experiential accord with all the world-views made possible by languages. In the Theosophist (Madras, India, January and April issues, 1942), Whorf contends:
This idea is one too drastic to be penned up in a catch phrase. I would rather leave it unnamed. It is the view that a noumenal world--a world of hyperspace, of higher dimensions - awaits discovery by all the sciences, which it will unite and unify, awaits discovery under its first aspect of a realm Of PATTERNED RELATIONS, inconceivably manifold and yet bearing a recognizable affinity to the rich and systematic organization of LANGUAGE, including au fond mathematics and music, which are ultimately of the same kindred as language. (2)
Thus, for Whorf, language both segments the emergent physical world into discrete entities and reflects the patterned relations of the noumenal world. His "idea ... too drastic to be penned up in a catch phrase" seems to be simply that a causal sequence exists between a noumenal world of (perhaps) ultimate reality, the physical world which emerges from it, and the world of our personal consciousness which is constructed out of the flux of impressions arising from the physical world and which is given a definite character by the patterns of language. Furthermore, the patterns of language, albeit in a "wavering and distorted, pale and substanceless manner," reflect the patterns not in the physical world but in the noumenal world, which Whorf supposes to exist beyond the physical. Although he does not mention Kant specifically, Whorf's use of "the term "noumenal" suggests a type of pseudo-Kantianism in which the noumenal world is the beginning point of a causal chain which ultimately results in the human world-view.
As described by J. W. Swanson (3) the Whorfian thesis appears to have two forms: first, a strong form which has to do with the metaphysical presuppositions of the hypothesis, and which provides a schema in which the problem of linguistic relativity arises; second, a weaker form which has to do with the supposed loss of meaning in translations. The strong form yields three inferences: (1) A world exists beyond the world of our perceptions, and is at the roots not only of experience but of the reality upon which the perceived world rests; (2) the principle of linguistic relativity rests on the strong form of the hypothesis; and (3) Whorf carried the principle of linguistic relativity to the point of becoming a form of linguistic determinism.
The weaker form yields two inferences which should be collated with those of the strong form: (1) The weaker form describes a property of languages such that the nuances of meaning and subtle connotations found in the expression of any particular language cannot be completely expressed in any other language without a loss in meaning; and (2) while not directly related to the problem of linguistic determinism, the weaker form ultimately seems to be derived from the stronger form.
While Sapir advanced a hypothesis similar to the one that was later offered by Whorf, the similarity is only superficial. (1) Sapir cautiously outlined a schema in which he described the parallel development of language and concepts. (2) Sapir hypothesized the influence of language on thought, but did not carry the idea to the point of linguistic determinism. (3) Although he was cautious in most of his statements concerning the relation between language and thought, Sapir in his last years acknowledged language as an influence "greater than most men believe." (4) Even in his most forceful statements, Sapir did not develop the metaphysics of language implicit in the contemporary interpretation of the Sapir-Whorf hypothesis.
On the other hand, Whorf's contribution to what has come to be called the Sapir-Whorf hypothesis does not reflect the caution of Sapir. In its broadest context, Whorf's thesis may be called a metaphysics of language. Within the text of Whorf's conception of the influence of language on thought, there exists a Kantian dichotomy between phenomena and noumena: Whorf's version, while apparently Kantian in its broadest structure, involves a misinterpretation of the Kantian metaphysics, using the patterned relations of language in much the same way that Kant used the forms of sensibility and the categories of the understanding. (4) Thus, while Sapir acknowledged the influence of language on perception, Whorf described language as the prime determinant of our conception of the world.
In order to reveal the difficulties inherent in Whorf's position, let us hypothesize a model of perception congruent with Whorf's view of linguistic relativity. First, there is the necessity of a noumenal realm, which contains patterns. If these patterns are actually reflected by language, then it would seem that some manner of patterning must exist between the elements or constituents of the noumenal realm. That Whorf could not possibly, without contradiction, support such a statement will be demonstrated shortly. If patterns exist in this noumenal realm, then it would seem that there must be more than one entity in the realm of independent reality, since patterns or patterned relations seem to connote a plurality. If patterned relations exist, then apparently they exist between discrete entities. If this is true, then there must be a manifold of noumenal realities, each related to the others in some way.
Briefly then, our conception of Whorf's noumenal realm is (1) it is plural-that is, it contains more than one entity; (2) these entities are related to each other; (3) it exists without dependence on anyone's knowledge of it; (4) the relations which exist between the independent realities are patterned in some way that is reflected in the patterns of language.
Whorf's use of the term "reflect" is far from precise; however, he considers the patterns of language to reflect or somehow represent a causal world which lies beyond the senses and which initiates the process which ends in knowledge. If language somehow reflects or represents the patterned relations of noumena, then, through linguistic patterning, man perceives at least one aspect of reality: the patterns of reality as reflected in the patterns of language. If, however, each language has slightly different patterns, it is difficult to see how they can all be accurate reflections of the noumenal patterns. Perhaps Whorf means that the patterns of the noumenal realm are inclusive of all possible reflecting language patterns.
If there is to be any relation between perception and independent realities, then there must be some part of the independent reality which presents itself to man. This property we could call the objective datum in the sense that it is accessible to all men in the same form. Such a datum would correspond to Whorf's "kaleidoscopic flux of impressions" and would be that part of noumena which, because of both the capacity and the limitations of man's sensory apparatus, incapable of evoking a response.
When the objective datum is received or ingested by the perceiver, it begins a process which results in the subjective datum. On one level of abstraction, we could call this result a certain configuration of static charges on the neurons of the brain. On another level of abstraction, we could say that the subjective datum is that datum which, after the operation of whatever forces come into play upon sense data, is presented to the individual consciousness. Whatever the process may be, it is in this last stage before consciousness that the action of language is thought to occur. Whorf's thesis seems to call for the action of grammatical categories such that the raw material of sensation is formed into a coherent perceptual whole, both by past experiences and by the grammatical categories.
Within this process, grammatical categories seem to occupy a place in the mind--a sieve through which all data must pass on their way to the consciousness.
Given the (as yet unfathomed) process that precedes the awareness of a perception, at length the individual comes into possession of an experience. We began the process with an independent reality; we end it with an object of perception. The objects of perception, existing within the mind as the end products of the perceptual process, whether they are books or candlesticks, ships or sealing wax, are inferred to be representations of concrete, external objects.
Further, the objects of perception are for Whorf a product not only of sensation, but of the action of grammatical categories as well; that of which the mind is ultimately aware would be the end product of the action of the grammatical categories, memories, and all associations which act upon the subjective datum.
IF MAN is thus aware only of the contents of his mind, then we must admit that Whorf has trapped his experiencer within his head. Further, if the language system has a role in the formation of these objects of perception, then Whorf appears to be lacking in any criterion with which he can establish the correspondence between the objects of perception and the external objects with which they are supposed to correspond: he provides the reflecting quality of language, but provides no procedure whereby the accuracy of this reflection can be demonstrated. If we say that our perceptions correspond to the things of which they are a copy (or if we say that the perceptions themselves, as a joint product of sensation and linguistic patterning, correspond to reality) because the patterning of language reflects the patterning among the external objects, then we are denied a workable criterion of truth.
This follows because Whorf would have it that there is no knowledge that is not in part a product of linguistic patterning; knowledge of external objects is therefore impossible, since external objects are not patterned by language. Further, even if language reflects the patterning that is supposed to exist between external objects (independent realities), there is no way in which we can demonstrate the commensurability of the reflection. Thus, not only is it impossible within the framework of the Whorfian schema to demonstrate that reflection occurs; it is a contradiction to assert that knowledge of the accuracy of the reflection is possible.
If we are correct in this analysis, then a more profound problem exists whose presence has been hinted earlier. Whorf's use of the term "noumena" and his affirmation of the existence of a higher realm are more than a hint of similarity between his views and those of Kant. From the outset, however, it seems a pseudo-Kantianism at best, for Whorf would have us believe that entities in the noumenal realm trigger the process that ends in perception. He can make this assertion, however, with no more justification than can a thorough-going Kantian. He has left man trapped within the confines of his perceptual consciousness; he can do no more than assume that objects in the noumenal realm are the ultimate (but partial) causes of the objects' perception, or that the patterned relations in language reflect the patterned relations in the noumenal realm. In fact, it would seem a contradiction to say "I know X and X is external to me," if the only X accessible to us (or demonstrable to us) is the X as an object of perception.
Further, Whorf's idea of a pragmatic or operational similarity in all human perception regardless of language differences raises another problem because of his conception of sense data. Whorf claims that reality is presented to man as a kaleidoscopic flux of impressions. Impressions, however, are descriptive of what happens to a perceiver. The kaleidoscopic flux, then, appears to be a description of the first stage of reaction: the first level of the subjective datum. We do not, however, experience the kaleidoscopic flux: by the time that an awareness of experience comes, the flux of impressions has been formed into a coherent perception by language patterning. If the only datum that is common to all men is a kaleidoscopic flux of impressions, then it is difficult to see how there is a pragmatic or operational similarity between the perceptions of all men. Further, if what we perceive is only what has been threaded through the categories of language, then it is contradictory to say that we could have knowledge of the kaleidoscopic flux of impressions.
WHORF CONTENDS, then, that the differences between language systems is such that each language both stimulates different thoughts (however slight the differences may be) on the part of the speaker, and exists as a separate and distinct system. The strong form of Whorf's thesis asserts that language acts upon the perceiving individual and causes the form of his conceptions; the weaker form of the thesis implies that in translation, there is a loss of the subtle nuances and connotations found in the original language. That is, it is impossible to translate all the nuances of meaning from one language to another language. In this form, the argument is not that translation is impossible: rather, it is that some translation is possible, but not all. This view is common among translators, as a perusal of any commentary on poetry translation will reveal.
However, within the context of the Whorfian thesis, we have no criterion with which to make the judgment that something is lost in the translation. We can know of discrepancies or content losses in translation only if we are able to compare the sentences in language X with the original sentences in language Y. Further, such a comparison would have to include a knowledge of the total meaning of the expression in both languages: if we do not know the total meaning of an expression in both languages, then while we could know of additions, we could not possibly judge precisely what is lost in the translation.
This is precisely what the relativists deny: that it is possible to know the nuances of meaning that have been lost. We cannot, however, judge that they have been lost if we have not identified them. If we are bilingual, we are in no better condition: all our judgments are colored by the fact that we perceive (to be consistent with Whorfs thesis) and conceive according to the effects of both languages. It is impossible that we can have knowledge of only one language and know what has been lost in translation; if we know both languages, it is impossible to judge the nuances of meaning in one language without having our judgment influenced by the other language. In any case, Whorf leaves us with no clear-cut criterion for untranslatability.
IF, as Swanson puts it, the strong form of the Whorfian thesis proposes the use of grammatical categories in much the same manner as Kant used the categories of the understanding, it appears that the strong form is assuredly a metaphysical hypothesis: it is an assertion about (1) the reality that lies behind the world of phenomena as it is experienced by man, (2) the process through which this higher realm causes the beginning of the perceptual process, (3) the correspondence between the patterned relations of language and the patterned relations which Whorf supposed to exist between entities in the noumenal realm.
The strong form is a metaphysical hypothesis because it concerns a reality that lies beyond the direct experience of man, and yet which is the physis upon which our experiential cosmos is constructed. As such, the strong form, within the context of Whorf's thesis, is beyond empirical verification. If we have no knowledge of what does not thread its way through linguistic categories, and if we as empiricists know only what our perceptual capabilities allow us to know, then an empirical verification of the Sapir-Whorf hypothesis seems logically impossible.
If we consider the weaker form within the context of the Sapir-Whorf hypothesis, then we can consider it, as is all of knowledge, dependent upon the relations and processes implicit in the strong form. If this is true, then empirical tests of the Whorfian thesis appear to be out of the question.
If the strong form is correct--that is, that experience is determined by the effects of language patterns on the perceptual process-then the weaker form is intelligible as a specific case produced by the processes described by the strong form. Within Whorf's schema, the argument for the weaker form derives its validity as a logical outcome of the strong form. If we accept this judgment as fact, then the two forms may hang together metaphysically, but they, and the thesis which they constitute, fall empirically.
This paper has outlined some of the reasons why the Whorfian thesis is not susceptible of empirical verification. It should be noted, however, that in no way has the conclusion been drawn that the Sapir-Whorf hypothesis is worthless. Neither has this study conclusively demonstrated that all tests relating to the hypothesis are either worthless or impossible.
The hypothesis is neither simple nor unambiguous. Possibly, in its comprehensiveness, too many hypotheses are included for it to be adequately tested with present research methods. The logical problems inherent in the Whorfian portion of the hypothesis seem to preclude the verification of large segments of its assertion. Perhaps the best course to be followed by philosophers and linguists would be to recognize the Sapir-Whorf hypothesis for what, at the end of this study, it seems to be: a vigorously speculative suggestion about a possible relation between language and the processes through which we form our thoughts about the world around us. The absence of conclusive empirical data should disturb only the philosophically unsophisticated.
(1.) 'Benjamin Lee Whorf, "An American Indian Model of the Universe," Language, Thought, and Reality, ed. John B. Carroll (Cambridge: M.I.T. Press, 1956), p. 58.
(2.) Ibid, pp. 247-248.
(3.) J. W. Swanson, "Linguistic Relativity and Translation," Journal of Philosophy and Phenomenological Research XXII, 185-192.
This article first appeared in ETC: A Review of General Semantics Vol. 27, No. I (March 1970).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||FROM THE VAULT|
|Author:||Carnes, Ralph L.|
|Publication:||ETC.: A Review of General Semantics|
|Date:||Jul 1, 2014|
|Previous Article:||Focal knowledge, medium bias, and metamedium.|
|Next Article:||Not since Babel.|