Commentary on Laura Weed’s "Experience as the Root of Language Development"

In Art as Experience, Dewey remarks that thinkers have typically taken artwork as their point of departure for theory building in esthetics.  He argues that this initial focus on the work of art has been unfortunate, since it has lead theorists to begin their investigations by analyzing the esthetic in one of its most complex manifestations.  It would be as if one were to start learning math by skipping basic arithmetic and moving directly to the study of algebra.  The resulting confusion in esthetic theory has been exactly what one would expect.  It is the merit of Prof. Weed’s paper to point out to us that a similar misstep has often occurred in the philosophy of language.  Linguistic meaning is a highly refined form of meaning that has as one of its necessary preconditions the experience had by an organism interacting with its surroundings.  No few difficulties in the philosophy of language stem from the implicit assumption that linguistic meaning encompasses all genuine types of meaning.  By taking language as the starting point for an explanation of meaning, the subject is unnecessarily obfuscated.  As in the biblical account, nature is made to appear wholly dumb, blind, and directionless until humans bestowed significance upon their environment through the act of naming.  The emergence of the spoken word is left seeming miraculous and mysterious, an event for which nature was completely unprepared.

Although I agree with the general spirit of Prof. Weed’s essay, I do wish to make a suggestion and raise one quibble, in that order.  Given the broad scope of her paper, I will limit my comments to a consideration of what I take to be her main argument.

It seems clear enough that language involves a system of symbols used for the purpose of communication, and perhaps this is a point upon which even analytical and Continental philosophers of language could agree.  It seems just as clear that in order for language to serve as an effective means of communication for human beings who are engaged in the business of living, it must include words that refer to nonlinguistic entities, such as "fire" or "cave bear."  However, it is just at this juncture that Prof. Weed identifies a breakdown of agreement among the theorists.  She argues that "language fundamentalists" from both the analytic and Continental camps deny that words are in any way tied to a nonlinguistic reality that lies behind—and, in a sense, beyond—all language systems or games.

On the analytic side of language theory, Prof. Weed criticizes Saul Kripke’s contention that linguistic meaning is created through an act of naming by an initial baptizer, a meaning that is then propagated and preserved by the "historical chain" constituted by succeeding generations of language users.  I don’t think that Kripke’s views on naming can serve to illustrate the stance of what Prof. Weed refers to as the "language fundamentalist," since he argues that the meaning of proper names and natural-kind terms directly depend upon the essential nature of that to which they refer.  Far from making meaning an affair that is wholly internal to language, Kripke’s theory of reference posits meaning as grounded in objective and scientifically ascertainable features of the external world.

Although the nature of Kripke’s relationship to analytic language fundamentalism as described by Prof. Weed is unclear, I think that she is right to identify his view as constituting a polar extreme to some hermeneutical philosophies of language.  Furthermore, I believe that Prof. Weed’s thesis that pragmatism offers a "middle way" between the extremes of analytic and hermeneutic theories of language is correct.  However, in order to successfully prosecute her case, I think that she will need to further develop some of the arguments that she advances in her paper.

Take, for example, Prof. Weed’s criticism of Kripke’s theory of reference.  She argues that his attempt to fix reference for proper names and natural-kind terms by declaring these words to be "rigid designators" fails since his theory cannot adequately account for the semantic aspect of language.  This shortcoming of Kripke’s account can be illustrated by the following thought experiment.  There exists an exceedingly rare form of color blindness which leaves the afflicted individual unable to perceive any colors whatsoever besides various shades of black, white, and gray.  Now imagine that every human being on the planet was so afflicted, except for one person whose color vision was perfectly intact.  Let us further suppose that this lone individual learns to take great esthetic delight in colors and their arrangement, and even develops different words to refer to them.  He or she takes up painting with watercolors, becomes quite fastidious about wearing color coordinated clothing, and develops an entire theory of interior decorating based on harmonizing color and form.  Fascinated by such apparently methodical yet inexplicable behavior, a team of scientists begins to study this initial baptizer, and is able to determine that his or her color vocabulary does in fact correspond to scientifically measurable differences in the spectrum of visible light.  Through observing and interviewing their subject, they learn to correctly refer to various colors as "pink," "brick red," etc.  They try to imagine what their subject means when he or she talks about different colors "clashing" or "complementing" one another by turning to the relations between different sounds and musical notes as an analogy.  Nevertheless, despite their best efforts, the scientists will forever find their subject’s watercolor paintings dull, and his or her enthusiasm for certain combinations of clothing and interior arrangements obscure.  They are congenitally unable to grasp the esthetic meaning of the words they have learned from the initial baptizer.

I take it that the result of our thought experiment supports Prof. Weed’s contentions about Kripke’s theory of reference.  The semantic aspect of language depends upon a shared backdrop of experience between speakers.  Dewey maintains that such a common situation or context constitutes a "universe of discourse" for whatever can be explicitly asserted in an argument (see the essay "Qualitative Thought" in John Dewey, Philosophy and Civilization (Gloucester, Massachusetts: Peter Smith, 1968) 97-8.).  He points out that just as a universe of discourse cannot appear as a member of discourse within itself, so a situation cannot appear as an element of the argument to which it forms the backdrop (p. 98).  Language is inherently social because it presupposes the shared nature of experience.  Prof. Weed’s discussion of Helen Keller’s acquisition of language represents a case in point.  Unable to interact with her pupil through sight and sound, Keller’s tutor naturally turned to the shared experience of touch.

It seems to me that Prof. Weed’s account of language is more or less in agreement with a view popular among pragmatists.  According to this standpoint, language originally emerged as a tool in the collective business of living, and the connotative and denotative meanings of words are still relative to groupal experiences and activities.  The open-ended character of experience should discourage us from seeking to permanently fix linguistic meanings either in terms of word-world relations or in terms of word-word relations.  From this perspective, Kripke’s theory is fundamentally flawed due to its failure to take into proper account the relative nature of the human condition, especially as regards knowledge.  Prof. Weed’s criticism of Kripke’s theory of naming is, I believe, correct, but does not go far enough.  In order to make a solid case against Kripke, she not only needs to bring to light the deficiencies of his theory, but also show how what it purportedly explains can be better accounted for by her approach.

In her article "Pragmatism and the Philosophy of Language," Danielle MacBeth (Philosophy and Phenomenological Research, Vol. LV, No. 3, September 1995: 501-523, and on the Internet at http://csmaclab-www.cs.uchicago.edu/philosophyProject/sellars/macbeth/macbeth.html) points out that the intuitive appeal of Kripke’s theory of reference is due at least in part to distinctions that are commonly drawn between matters of meaning and matters of fact.  Nonnatural or nominal-kind terms like "bachelor" do not seem to be open to revision through empirical discoveries in the manner of natural-kind terms like "water."  It would seem impossible that scientific investigation of bachelors could ever reveal that, in fact, not all bachelors are unmarried males.  However, it is always possible that our identification of water with H2O is mistaken, and might one day need to be revised in the light of new evidence.  The appeal of Kripke’s theory is that it supposedly explains how our descriptions of what is designated by natural-kind terms may change, and also why the terms’ denotation remains constant (pp. 3-4).

MacBeth suggests an interesting pragmatic solution to the problem of the relation between matters of meaning and matters of fact by arguing that instead of construing meaning and reference as external or nonintentional relationships, we should interpret them as kinds of epistemic commitments.  However, my purpose here is not to consider MacBeth’s views, but to underscore the need for Prof. Weed to formulate a more complete response to Kripke.  Although the issue is caste in the very technical terms of the philosophy of language, I think that underlying the debate is a topic of much more general philosophical significance.  For lack of a better phrase, I’ll call this the problem of working out the objective features of relativity.  To my mind, the strength of Prof. Weed’s paper lies in her opting for the middle path of pragmatism as an alternative to foundationalism on the one hand, and a free-floating relativism on the other.

My quibble concerns the problem of the objective features of relativity, and specifically involves Prof. Weed’s tendency to treat the Jamesian and Deweyan conceptions of experience as more or less identical.  Prof. Weed characterizes Dewey’s view of experience as rooted in the first person singular, and she equates it with the stream of consciousness.  Although Dewey’s conception of experience certainly takes into account the stream of consciousness, I think that it is ultimately mistaken to identify this phase or type of experience as defining his conception of experience.  In his unfinished re-introduction to Experience and Nature (John Dewey, The Later Works: 1925-1953, Vol. 1: 1925, ed. Jo Ann Boydston (Carbondale, Illinois: Southern Illinois University Press, 1981) 361-2.), Dewey writes that, due to all of the misunderstandings surrounding the word "experience," were he to write the book again he would entitle it Culture and Nature.  He insists that "experience" was meant to stand for both "what is experienced and the ways of experiencing it" (p. 362).  In its anthropological sense "culture" encompasses both the material and the ideal, and, Dewey points out, has the added advantage that it is at once both psychological and collective (pp. 363-4).  In short, I think that Prof. Weed’s argument could be strengthened if she would take more care in her choice of wording so as not to suggest, however inadvertently, that experience is primarily psychological and private.