Dewey's Muffled Call for a Larger Unit of Psychological Analysis.

Paul F. Ballantyne

*Presented at the 103rd meeting of the American Psychological Association, (Division 26) Toronto, Canada, August, 1996. [*Updated November, 2002].

This talk marks a hundred years of routine understatement and outright misrepresentation regarding the main message of Dewey's 1896 article called: "The Reflex Arc Concept in Psychology." No..., I'm not merely arguing that it WAS a critique rather than an acknowledgment of the "reflex arc" concept of that era. That fact is obvious to anyone who reads Dewey's article. Rather, the aim here is to point out that his critique went well beyond the "reflex circuit" view which has long been said to be his position (e.g., see the commentary on this "Mead Project" link). After presenting the cryptic opening quotation from his article, both the present misrepresentation of his views in history of psychology texts, and the original discipline-building motivations behind the muffling of his message by early functional psychologist such as Angell and Carr (1904-1930) are covered.

The related issue of the exact nature of his call for a larger unit of analysis is then address by way of contrasting both his actual (1896) argument (for an "organic circuit" view) and his subsequent (1920s era) elaborations with those of Edward Scripture and Walter Pillsbury. It is pointed out that 'getting Dewey's argument right' is not merely an amusing excursion into a bygone philosophical debate. Rather, what is at stake is not only the issue of the very definition of psychological subject matter but also the means by which both early and present-day experimental psychology defines its variables. The overall tone of the talk starts rather critically but is intended as a constructive engagement with the experimental psychology tradition, so I will end on an optimistic note by stating the progressive implications of Dewey's real position for ongoing integrative efforts in 21st century general psychology.

A Troublesome Quotation:

One hundred years ago, John Dewey wrote the following words in his (1896) Psychological Review article:

"That the greater demand for a unifying principle...in psychology should come at just the time when all generalizations and classifications are most questioned....is natural enough. It is the very cumulation of discrete facts...that also breaks down previous lines of classification. That material is too great in mass and too varied in style to fit into existing pigeon-holes...The idea of the reflex arc has...come nearer to meeting this demand for a general working hypothesis than any other single concept.....In criticizing this conception it is not intended to make a plea for the principles of explanation and classification which the reflex arc idea has replaced; but, on the contrary, to urge that they are not sufficiently displaced, and that in the idea of the sensori-motor circuit, conceptions of the nature of sensation and of action [derived from the nominally displaced association psychology] are still in control (Dewey, 1896, p. 357, emphasis added).

On the Cryptic nature of Dewey's position

So began John Dewey's 1896 article "The Reflex Arc Concept in Psychology" an exceedingly difficult, though conceptually coherent, piece of philosophical writing. Its historical import is said to be that it contains a powerful criticism of behaviorist psychology years before that school even become established. I think it does provide a powerful argument. But Dewey's argument is not the one portrayed in past psychology texts nor even in contemporary history of psychology textbooks.

Not only did contemporaneous figures such as E.B. Titchener (1898, 1899, 1921, 1922) and even fellow "functional" psychologist J. R. Angell have difficulty with the theoretical message contained in the article, but so have subsequent general psychologists and historians of psychology (such as Hilgard, 1987, or Benjafield, 1996).

The main culprit here, however, is not Dewey, but our own lack of imagination about what a non-associationist and non-reductive psychology would be like; and especially, what it would mean for the daily empirical practice of experimental psychologists.

Do you know what such a psychology would be like?

Don't worry; ...Whichever way you answered, you are not alone.

Current Oversimplification of his argument

The most common solution, when faced with this visionary gap has been to oversimplify Dewey's position into something we feel relatively comfortable with (i.e., the reflex circuit view). That is, to portray him as calling for a 'circuit' rather than an 'arc' view of reflexive acts (see figure 1).

Reflex Arc View (A)
Reflex Circuit View (B)

Figure 1: Diagrams such as this are intended to make a distinction between "two conceptions" of a reflex act: the reflex arc (A), and the reflex circuit view (B). That which Dewey was 'against' appears uppermost, and that which he is said to be 'for' appears bottom-most (from Benjafield, 1996) .

Indeed ever since the time of James (1890) the reflex arc concept has provided us with a convenient straw-man model of environmental contingencies imposing their will on a passive organism. It is obviously problematic because it implies a one-way causal chain of events where a sensation or stimulus in the environment "hits" the organism with an inevitable response being produced. In fact, if contemporary general psychology can be said to be "functionalist" at all, it is in this very recognition of the necessity of asserting some form of active or practical relationship between the nature of the organism and the contingencies of the environment.

The reflex circuit view, on the other hand, has long been suggested to be the viable functionalist alternative to this older arc view. In the reflex circuit, a two-way interaction between the environment and an organism involving some sort of active processes (like a sensory-motor feedback loop) is said to be involved. By placing the "organism" or a "mental process" in the middle we are supposed to have solved the problem of environmental immediacy.

But, another glance back at Dewey's (1896) opening statement that: "in the idea of the sensori-motor circuit, conceptions of the nature of sensation and of action [derived from the nominally displaced psychology] are still in control" should indicate that this (now standard claim) about his position is not entirely accurate.

In fact, this circuit portrayal of human acts is more in line with the views of early experimental functional psychologists (such as James Roland Angell or Harvey Carr) or with the later S-O-R/ S-M-R views of Robert S. Woodworth than it is in any way related with Dewey's views (who explicitly rejected it as too limiting).

When, How, and Why Dewey's call was "muffled"

J.R. Angell (1907) as the spokesman for the "province of functional psychology," was the first to explicitly portray Dewey's argument in the now standard arc vs. circuit manner. This seemingly incidental understatement had the immediate effect of smoothing over the differences with the contemporaneous experimental mainstream (of physiological and structural psychology) but it also relegated the finer points of Dewey's argument (against environmental immediacy in particular) to the province of antiquity.

The clearest and most Americanized portrayal of the circuit view I have found from this period appears in various versions of Howard Warren's aptly named Elements of Psychology between 1918-1930 (see fig 2).

Figure 2 Reflex circuit view of seeing and acting (from Warren, 1930).

It is interesting indeed to note how the use of this popularizing device by Warren is reminiscent of those used by Edward Scripture's 1890s era Americanization of the reaction time concept. And after all, what could be more ecologically valid for an American than baseball?

The similarities, however, go far further back than the immediate disciplinary need to popularize successive manifestations of the "new" (experimental psychology) to the textbook reading audience of American students. The long associationist/elementist (i.e., philosophical and physiological) traditions behind the reaction time concept are virtually identical to that of both the reflex arc and the reflex circuit view of acts -both of which Dewey (1896) rejected as too limiting on the new psychological science.

Warren on the other hand literally 'wrote the book' on the history of associationist psychology in 1921 and he never diverged significantly from that point of view in his portray of general psychology (though he did do so -occasionally- in his book on association psychology itself). Under the older (18th through early-19th century) philosophical associationist view, of course, our higher mental processes are more or less 'additive sums' of lower states. Likewise, under the late-19th century evolutionary manifestations of that elementist view animals have more 'reflexes or instincts' and less 'reason' than humans but differ in degree but not in kind (see my opening chapter on the additive 'continuity view' of mind for more on this point).

It is indeed ironic then that Warren's associationist characterization of the ecological validity of the circuit view of reflex acts (which discusses an organismic relationship with the environment without really explaining how it fits into a larger whole) is virtually identical to the portrayal found in the ostensibly functional psychologist texts of Angell and Carr.

And we all know what happened to the school of functional psychology.....

The fall of Functional Psychology as a school

It was first absorbed into general experimental psychology (via Angell, 1904, 1909, 1918; Woodworth, 1921, 1929, 1938) and then through the intermediary work of Harvey Carr (1925, 1927, 1930) into behaviorism itself (see Tolman, 1990).

While the former absorption was intentional the latter was not. Angell's efforts between 1904-1918 were aimed at explicitly outlining the functional psychology (implicitly stated in prior works by Dewey and James) in a manner that addressed the concerns of Titchener's structuralist approach to experimental psychology. The smoothing over of differences was his goal and in this respect his efforts can be assessed as fairly successful. Carr's Psychology: A study of mental activity (1925), on the other hand, attempted to explicitly outline the difference between functionalism and the emerging behaviorist movement, but the limitations of his understanding of the basic tenants of Dewey's (1896) article (i.e., regarding action, activity, and consciousness) as well as his close personal working relationship with J.B. Watson undermined his efforts in this respect (see also Carr, 1930).

Hence, in 1927, while attempting to resolve the debate between the defense of anthropomorphism contained in Washburn's Animal Mind and the behaviorist negation of the concept of consciousness Carr wrote: "I am somewhat of a behaviorist in the field of animal psychology, although I do not class myself as such so far as human psychology is concerned" (Carr, 1927, p. 104). [For more on the continuity view of mind as it was played out in the animal psychology of the era see chapter 2 of my 2002 work].

The dye was thus cast and by the time of Woodworth's Schools of Psychology (1931) the distinctions between the fundamental tenants of behaviorism and the ongoing practice of functional psychology were fuzzy indeed. In other words, no distinct school of functional psychology remained and the narrow "conceptions of the nature of sensation and of action" derived from the nominally displaced association/physiological psychology tradition were "still in control."

This is what I mean by the "muffling" of Dewey's call. The garbling of his message (against the reflex circuit view in particular) was an outcome of the disciplinary necessities of establishing functional psychology as a school (based at the University of Chicago) at a time when psychological science was still (for the most part) equated with physiological experimentation.

*Now, what about this "larger unit of analysis" mentioned in the title of this talk?

Dewey versus Scripture and Pillsbury on subject matter.

The task of outlining exactly what Dewey (1896) was 'getting at' requires that I give you a few sound-bites on what he actually wrote. Fair enough, I'll do that. But it also requires that we at least briefly contrast his ongoing views on action, activity, and consciousness with that of two of his contemporaries (Edward Scripture and Walter Pillsbury) who followed the physiological psychology tradition. As stated at the outset of this talk, I believe Dewey had a valuable message contained in his 1896 article, one that implied that a truly functional psychology must be both organismic and intentional. This part of his message was brought into the portrayal of early functional psychology by Angell and also Carr, but other parts (particularly the argument that in the case of human beings, this 'larger coordination' [i.e., unit of analysis] is not only biological and individual but also social (and by extension cultural) are only stated explicitly in Dewey's later (1920s era) works (i.e., after the initial flame of the functionalist school proper had fizzled and was beginning to die out).

***Stated in another way, in advocating neither 'sensationist' associationism (the "nominally displaced psychology"), nor 'material reduction' (reflex arc psychology), Dewey (1896) hinted at a third "emergent" psychological alternative. This would be a psychology in which the recognized "units" of psychological analysis would retain that which is important to the particular processes being discussed. The particular processes, in turn, would be understood not as additive elementary components but as functional aspects of a larger whole in which they operate.

His (1896) dialectical treatment

True to his Hegelian roots, Dewey (1896) first argues convincingly that the reflex arc account of action is inadequate because it contains an apparent logical paradox which is not resolved until the discourse of investigation moves upward to an intentional level of analysis. Dewey illustrates his case by carefully considering the deficiencies of the reflex arc account of a child's first and subsequent encounters with a candle-flame (see fig 3).

Figure 3: Child's first encounter with the candle flame left and second encounter right which entails an active avoidance of being burned (From James, 1890; Vol. I).

The logical paradox of the reflex arc approach lies in the fact that the whole initial child-candle encounter can be analytically described, either in terms of physical motion (responses), or in terms of the transformation of sensory stimulation. That is, the 'seeing and reaching' aspects of the child's encounter with the flame, when considered from the physical-process side of the dualism, can be described as responses. They are, as Dewey put it: "one uninterrupted, continuous redistribution of mass in motion" (p. 364). But the same encounter, when described from the psychical- process side, appears to be all sensation. As Dewey put it: "What we have [on this psychical account] is a certain visual-heat-pain-muscular [quality], transformed into another visual-touch-muscular [quality]" (p. 364). Dewey points out that the physical reflex arc account never bothers to resolve this logical contradiction of its discourse, and therefore, it leads a spurious and oversimplified double-life.

The major theoretical suggestion made by Dewey (1896), is that an accurate and explanatory account of the child's second encounter with the flame requires a larger unit of analysis capable of including some sort of "teleological" (goal-oriented) component. The candle can not be accurately described as a 'stimulus' until the child looks at the candle. In the same way, the movement of the hand toward the flame can not be described as a 'response' until we know that the child intends to touch the flame (in the first encounter) or avoid touching it (in the second encounter). In other words, the so-called, stimulus and response have "a special genesis or motivation and a special end or function [utility]" rather than a detached "preexistence" (Dewey, 1896, p. 370).

In Dewey's terminology, by the act of attending to some aspect of its environment, an organism "constitutes" that aspect as a stimulus. Similarly, by manipulating some aspect of its environment the organism "constitutes" that action as a response. Thus the above mentioned "genesis" of the stimulus or response is not to be sought outside but inside the act (as a larger intentional "coordination"). That is, these two smaller units do not have "detached preexistence". They can only be understood by reference to the larger intentional unit in which they reside. That intentional psychical unit, Dewey writes, is "transformed" after the child's finger is burned (while the underlying physiological circuit remains the same -and therefore is not explanatory in either case but especially in the second one).

In one line, Dewey's summary argument ran as follows: "Neither mere sensation, nor mere movement, can ever be either stimulus or response; only an act can be that...." (Dewey, 1896, p. 367; emphasis added). Further, with regard to their "development in the individual, or in the race, or from the standpoint of the analysis of the mature consciousness" (p. 360), the two acts (seeing and reaching) fall within a "larger coordination" in which the conflicting ends of each action (i.e., whether in the same, different, or successive situations) must be considered. These intentional properties are not there at the component reflex arc or even reflex circuit level of analysis.

As for the disciplinary implications of shifting our analysis away from the standard "sensori-motor" toward an "organic" (rather than reflex) circuit understanding of such larger (shifting) intentional coordinations he suggests that the point "is in its application" but those, he said must be "deferred to a more favorable opportunity" (p. 370). His (1896) analysis is therefore to be viewed as a prolegomenon to later possible applications which (as noted above) were indeed deferred (at least within experimental psychology).

But what, exactly, was the wider unit of analysis that he seems to be envisioning here for psychology? I view his article as a prelude to what would later formally be called emergent evolutionary or life-span developmental understanding of psychological subject matter. As portrayed by Dewey, in the case of individual development, an expanding series of psychological "coordinations" with inherently conflicting ranges and ends (i.e., goals, intentions, and motives) -which the constituent acts refer to and fulfill- implies a conception of the subject matter of analysis in which successive envelopes within envelopes of experience are encountered (and "constituted") by that individual (see fig 4).

Fig 4 "The Infant's Horizon Widens as He Matures" (from Prothro & Teska Psychology: A Biosocial Approach, 1950).

As I have already stated, this is the implied and not explicitly demonstrated "larger unit of analysis" referred to in the title of this talk. Further evidence for this suggested meaning, however, can be gained from contrasting Dewey's later (1910s onward era) work with that of Walter Pillsbury who both attempted to explicitly reject teleological analysis in psychology and to explicitly define the subject matter of psychological science accordingly. Again, as Dewey has already pointed out, the proof of these contrasting approaches is indicated in the "psychological adequacy" of their application.

Pillsbury's definition of Psychology

Pillsbury's, 1910, APA presidential address "The place of movement in consciousness" was intended to be inclusive of past motor and sensory views of "higher mental functions." It ends, however, on an exclusionary note by suggesting goal-oriented actions have "no place in science" (p. 99). This was a direct repudiation of his former Michigan colleague John Dewey, who embraced action-toward-an-end as a larger unit of analysis. In his How We Think (1910), Dewey points out that: "The problem [to be solved by a thinker] fixes the end of thought and the end controls the process of thinking" (p. 12). For example, the traveler who's goal it is to find the most beautiful path during a stroll will pick a different route than those wishing to find their way to the next town. Dewey's point seems to make perfect sense but it was Pillsbury and not Dewey who had the lasting impact on the way experimental psychology defines its subject matter.

Prior to Dewey and Pillsbury's difference of opinion, psychology was defined as the science of conscious processes (Wundt), or of conscious mental life (James). Pillsbury's Essentials of Psychology (1911) defined psychology as the "study of behavior" (p. 1). In 1913, the young J. B. Watson used this definition as a backdrop for his behaviorist school. Pillsbury's Fundamentals of Psychology (1916), then avoided an explicit definition but openly rejected behaviorism as too limiting. "Certain...questions which interest the psychologist concern mental states, and these have been answered by the older methods" (p. 6). Pillsbury's implied definition (i.e., mental states and behavior) has been adopted by the experimental mainstream and is still currently dominant under the guise of cognition and behavior.

Even though Pillsbury was trying to include both higher and lower aspects of psychological subject matter, it can be argued that this "wider" definition again leaves the old "conceptions of the nature of sensation and of action" derived from the nominally displaced psychology, still in control" because it explicitly rejects teleology (final cause) as the guiding aspect of its structure. This serious flaw showed up in the way Pillsbury (1928) applied it to the investigation of human language. That is, the additive understanding of psychological subject matter effects not only the conclusions drawn from experimental research, it effects the very form of experimental investigation which is carried out.

Pillsbury's Psychology of Language (1928), presents a contrived experimental situation for collecting physiological data on the pronunciation of single words. Meader and Shepard are identified as the two, and only, subjects. Their definition of language as the communication of a thought, idea, or emotion through expressive movement was equally dated. Dewey's Experience and Nature (1925), had already argued that the essence of language should not be sought in verbal expression, or syntax, but in the mutual assistance between human partners. Such an analysis is not only maturational and developmental but also socio-cultural in nature.

Concluding remarks

The implication (and message of this talk) is that we are all now faced with a very similar intellectual task or challenge as that of Dewey in 1896: To ponder the fate and content of our "new" century's psychology. The one import difference between Dewey's disciplinary situation and our own, however, is that the necessary basic empirical evidence (lacking in 1896) has been present in both general introductory and cognitive psychology texts for some years now.

The interdisciplinary presentation of this cumulative evidence in its proper emergently integrated form and the application of this knowledge in further empirical research seems to be the next logical move for general psychology! Careful appeal to a larger qualitatively distinct realm of individual or social actions (in animals); and of collective societal activities (in the case of human beings) is the only way to understand the origin and course of smaller constituent components (e.g., cortical operations and spino-muscular events).

The application of computer aided imaging of brain structures and functions (MRI and PET scans) are already showing promise in this respect. It is here too that both the functional systems approach to neuropsychology (Luria, 1970; see also Luria's The Working Brain, 1973; Goldberg, 1990), and the Activity Theory approach of A.N. Leontiev (which distinguished between operations, actions, and activities), might yet have their day in the disciplinary sunlight.

For the first time in the history of psychological practice we have a conceptual and empirical basis for both integrating various past forms of empirical data and for guiding the shape of future research situations -toward an explanatory and ecologically relevant account of societally formed higher mental processes. What will our new psychology be like? Hopefully it will be similar to that implied by Dewey way back in 1896.

References:

Angell, J.R. (1904). Psychology: An introductory study of the Structure and Function of human consciousness. New York: Holt.

Angell, J.R. (1907). The nature of Functional Psychology. Psychological Review, 14, 61-91.

Angell, J.R. (1909). The influence of Darwin on Psychology. Psychological Review, 16, 152-169.

Angell, J.R. (1918). An Introduction to Psychology. New York: Henry Holt and Company.

Benjafield, J.G. (1996). A history of psychology. Boston: Allyn and Bacon.

Carr, H. (1925). Psychology: A study of mental activity. New York: Longmans, Green.

Carr, H. (1927). The interpretation of the Animal Mind. Psychological Review, 34, 87-106.

Carr, H. (1930). Functionalism. In C. Murchison (Ed). Psychologies of 1930. (59-78). Worchester: Clark University Press.

Dallenbach, K.M. (1915). The history and derivation of the work 'Function' as a systematic term in psychology. Journal of Psychology. 26, 473-484.

Dewey, J. (1896). The Reflex Arc Concept in Psychology. Psychological Review, 3, 357-370.

Dewey, J. (1899). The School and Society. Chicago: University of Chicago Press.

Dewey, J. (1902). The Child and the Curriculum. Chicago: University of Chicago Press.

Dewey, J. (1910). How We Think. Boston: D. C. Heath.

Dewey, J., & Dewey, E. (1915). Schools of To-morrow. New York: E.P. Dutton.

Dewey, J. (1928). Progressive education and the science of education. In R. D. Archambault (Ed). John Dewey on Education. New York: The Modern Library.

Dewey, J. (1932). Democracy and Education. Westport: Greenwood.

Dewey, J. , Kilpatrick, W. H., et al., (Eds.). (1937). The Teacher and Society. New York: Appleton-Century.

Dewey, J. (1938). Experience and Education. New York: Collier.

Dewey, J. (1959). Dewey on Education. (S. Dworkin, Ed.). New York: Bureau of Publications Teachers College Columbia University.

Goldberg, E. (Ed.). (1990). Contemporary Neuropsychology and the Legacy of Luria. Hilsdale: LEA.

James, W. (1890). Principles of Psychology. New York: Holt.

Jordan, J. S. (1998). Recasting Dewey's critique of the reflex-arc concept via a theory of anticipatory consciousness: Implications for theories of perception. New Ideas in Psychology, 16(3), 165-187.

Leontiev, A.N. (1978). Activity, Consciousness, and Personality. Hillsdale: Prentice-Hall.

Leontyev, [Leontiev], A.N. (1981). Problems of the Development of the Mind. (Trans. M. Kopylova) Moscow: Progress Publishers.

Luria, A.R. (March 1970). The functional organization of the brain. Scientific American, 222(3), 66-78.

Luria, A. (1973) The Working Brain. London: Allen Lane.

Luria, A. (1976). Cognitive Development: Its Cultural and Social Foundations. (Trans. M. Lopez-Morillas & L. Solotaroff). Cambridge: Harvard University Press.

Luria, A. (1979). The Making of Mind. London: Harvard University Press.

Munn, N.L. (1971). The Evolution of the Human Mind. New York: Houghton Mifflin.

Rogoff, B. (1990). Apprenticeship in Thinking: Cognitive development in social context. New York: Oxford University Press.

Ruckmick, C.A. (1913). The use of the term function in English textbooks of psychology. Amer. J. of Psyc., 24, 99-123.

Scripture, E. (1895/1907). Thinking, Acting, and Doing. New York: Putnam's.

Scripture, E. (1898). The New Psychology. New York: Scribners.

Scripture, E. (1936). E.W. Scripture. In C. Murchison (Ed.). A history of psychology in Autobiography. Vol.III, 231-260.

Titchener, E.B. (1898). Postulates of a structural Psychology. Psychological Review. 7, 449-465.

Titchener, E.B. (1899). Structural and Functional Psychology. Philosophical Review. 8, 290-299.

Titchener, E.B. (1921). Functional psychology and the Psychology of Act, I. Amer. J. of Psyc., 32, 519-542.

Titchener, E.B. (1922). Functional psychology and the Psychology of Act, II. Amer. J. of Psyc., 32, 43-83.

Tolman, C.W. (1989). Pluralistic Monism: William James as Closet Heraclitean. Psychological Record. Vol. 39, pp. 177-194.

Tolman, C.W.; & Piekkola, B. (1989). John Dewey and Dialectical Materialism: Anticipations of Activity Theory in the Critique of the Reflex Arc Concept. Activity Theory, 1, Nr. 3/4, pp. 43-46. *[see My Re-Post of this article (P.B)].

Tolman, C.W. (1990). Continuity/discontinuity in the evolutionary thinking of the Chicago functionalists. Storia Della Psicologia, 2, 64-72.

Young, R.M. (1968). The functions of the brain: Gall to Ferrier (1808-1886). ISIS, 59, 251-268.

Warren, H. (1918-22). Elements of human psychology. Boston: Houghton.

Warren, H. (1930). Elements of Psychology. (Rev. ed.). NY: Harper.

Woodworth, (1921, 1929, 1934). Psychology. New York: Holt.

Woodworth, R.S. (1931). Contemporary Schools of Psychology. New York: Ronald Press.

Woodworth, R.S. (1938). Experimental Psychology. New York: Holt.