Functionalism
Harvey A. Carr
University of Chicago
What is a functional type of psychology, and who were the functional psychologists?
According to Boring (4), functionalism was a revolt of colonial psychologists against Germany. (Perhaps American would have been the better term to use.) The controversy between Titchener and Baldwin was a phase of the whole. Germany was the more philosophical and America the more practical. Chicago functionalism was the explicit movement, but I think it was symptomatic of what was quietly going on all over the country except at Cornell.
Titchener (11, 12) groups the various psychologies into two classes : (a) the structural or what is now termed the existential type of psychology represented by Wundt, Külpe, Ebbinghaus, and Titchener, and (b) the empirical type which attempts to portray mind as it is, i.e., as it works in dealing with the world about it. This empirical type of psychology goes back to Aristotle and Aquinas, and it forms the staple contents of most psychologies down to and including our twentieth-century textbooks.
Titchener farther subdivides the empirical group into two sub-classes —the act and the functional types of psychology. Brentano, Lipps, Witasek, Stumpf, Meinong, Messer, and Stout are referred to as act psychologists, while Ladd, Judd, Angell, James, Baldwin, and Dewey are referred to as functionalists. Titchener states that functionalism was primarily an American psychology, which traces its descent from Aristotle, but which was born of the enthusiasm of the post-Darwinian days when evolution seemed to answer all the riddles of the universe. Functionalism is further described as the dominant psychology of America which suddenly became conscious of itself, and which attempted to justify itself as a system with the introduction of existentialism.
According to Angell (2), functionalism was a movement that embraced a large number of psychologists who had certain principles in common, but who differed considerably in many other respects. He specifically states that functionalism is not to be identified with the Chicago type of psychology. Functionalism found its roots in Aristotle; its modern origin is traced to Spencer and Darwin, while the movement became self-conscious and first attempted to define and formulate itself as a protest and defense against the inroads and threatened dominance of
( 60) the existentialism of "Titchener and his disciples. Angell gives no list of functional psychologists as does Titchener.
These three writers agree that functionalism refers primarily to the dominant modern American type of psychology as contrasted with the structuralism or existentialism of Wundt and Titchener. I doubt if Angell would limit the term exclusively to American psychologists. I am inclined to think that he would classify Stout, for example, as a functionalist, while Titchener refers to him as an act psychologist. Perhaps the distinction between a functional and an act psychology is not as clear-cut and definite as Titchener assumes, or perhaps the two psychologies are not mutually exclusive and the same person may legitimately be assigned to both classes.
These minor differences will be ignored, and, for the present, we shall use the term functionalism to refer to the American empirical movement that rebelled against the proposed limitations of the structural or existential school of Titchener and his disciples. I shall adopt the caution of Angell and refrain from adding to Titchener's list of functional psychologists, as I fear that some might be rudely surprised, if not insulted, at being labelled a functionalist. Functional psychology is not to be identified with that of Angell or the Chicago group of psychologists. There .is no functional psychology; rather there are many functional psychologies. In speaking of functionalism; we are dealing with a group of psychologies which differ from each other in many particulars, but which exhibit certain common characteristics in virtue of which they are labelled functionalistic.
What are these common characteristics, and in what respects do the functional psychologies differ from the existentialism of Titchener? In answering these questions, we shall again refer to the writings of Titchener and Angell—the chief antagonists in this structural-functional controversy.
Before doing so, it may be well to note some points of agreement. At the time of which we write roughly the period from 1890 to 1910—practically all psychologists professed to be engaged in the study of consciousness. Structuralists and functionalists were alike then in that they defined their science as the study of the conscious processes as distinct from their organic conditions and correlates. The two schools differed somewhat as to the meaning of the term consciousness, and they might differ considerably as to the metaphysical implications of the dualistic distinction involved. Again, introspection was regarded as the chief, _ if not the only, method of psychological observation, although the two schools did not agree as to the connotation of this term.
Functionalism, according to Angell (2), differs from structuralism in three respects.
1) Structuralism deals with the what's or contents of consciousness, and it attempts to describe these in terms of their analytical elements. Functionalism does the same thing, but it refuses to confine itself to this limited program. It proposes to deal also with the whys and howls of
( 61) these contents, and to study them in their relation to the context of which they are a part.
2) This context in its widest and most inclusive sense is the biological process of adjustment. Functionalism regards mental processes as means by which the organism adapts itself to its environment so as to satisfy its biological :feeds. Mental events are thus studied from the standpoint of their relation to the environmental world and to the ensuing reaction of the organism to. that world. Functional psychology is thus practical and utilitarian in spirit and interest. Functionalism studies the uses and utilities of conscious processes, and it is naturally interested in developing the various applied fields—educational psychology, industrial psychology, abnormal psychology, mental hygiene, etc.
3) Functional psychology insistently attempts to translate mental process into physiological process and, conversely, it is interested in discovering and stating the organic concomitants and correlates of the conscious processes. Such a program is obviously incumbent upon any dualistic psychology which regards mental processes as means of adjustment to the environmental world. A functionalist can accept any one of the various conceptions of nature of the mind-body dualism with the single exception of that of epiphenomenalism.
Titchener (11) lists four characteristics of a functional type of psychology.
1) Functional psychologies distinguish between the activity or function of consciousness and its content or structure. They emphasize the study of function in preference to that of content.
2) Consciousness, especially in its active phase, has a value for organic survival. Consciousness is regarded as a solver of problems.
3) A functional psychology is teleological. The whole course of mental life is regarded teleologically.
4) Functional psychologies are written as a preface to philosophy or to some practical discipline. They psychologize as a means to some foreign end and not as an end per se. Their spirit is primarily that of an applied science rather than that of a pure science. Presumably existentialism is a representative of the pure scientific attitude.
These two writers agree that functionalism differs from existentialism in that it refuses to confine itself to the limitations of the existential program, but insists upon doing something more, viz., study functions. Both agree that this program will include a study of the uses or utilities of mind in practical situations, and of its biological or survival value. The reader is left in some doubt as to the extent of agreement in other details, and one still feels the need for a more precise and comprehensive definition of the term, function.
Ruckmick (6) canvassed fifteen modern American and English texts, and carefully studied the meaning of the term function whenever used. He found that all usages of the word could be grouped in two classes, and that the same author might use the term in both senses.
1) In the first usage the term function is equivalent to mental ac-
( 62) -tivity. All mental activities such as seeing, hearing, perceiving, conceiving, imagining, recalling, etc., are termed functions. Mental functions and mental acts are thus synonymous expressions.
2) The term function was also employed to denote service or use for some end, as when an author speaks of the function of a word when it is used as a symbol for an object.
Psychology, according to Titchener, borrowed the term from physiology, and psychologists use it, in my opinion, in the same way. Physiologists refer to breathing as a function, and they also speak of its function or use in furnishing oxygen to the blood and in the elimination of waste products. There is nothing peculiar in the psychological use of the term.
Critics of functionalism have frequently commented on this dual usage of the term. They point out that with such a dual usage — it is possible to speak of the "function of a function," or to say that a "function has a function." These writers apparently attempt to discredit the functionalistic movement by suggestive innuendo. Their remarks seem to suggest that such phrases are ridiculous, illogical, or absurd, and that the term function is evidently being used in two inconsistent ways. At least this has been my interpretation of their comments.
Without being contumacious in the matter, the writer is willing to defend the three following propositions:
1) The two usages mentioned by Ruckmick are not inconsistent.
2) They do not, in fact, represent two different meanings. The term function is used in exactly the same sense in both cases.
3) Finally, it is neither illogical nor absurd to speak of the function of a function.
With both usages mentioned by Ruckmick, the term function, in my opinion, is used in the same way as it is in mathematics. When a mathematician says that X is a function of Y, he is asserting that the term X stands in a contingent relation to Y without specifying as to the further nature of that relation. Psychologists, in my opinion, use the term function whenever they are dealing with a contingent relation irrespective of whether that relation is also one of act and structure, cause and effect, or means and end. A contingent relation and a functional relation are synonymous expressions.
The statement that the oxygenation of the blood is a function of breathing merely asserts that this end result is contingent upon the act of breathing. Likewise, when psychologists state that one of the functions of a vocal process is that of symbolizing objects, they are merely stating that the object of thought in this particular case is contingent upon the vocal process. Again the statements that breathing is a function of the lungs and that seeing is a function of the eyes obviously mean that these acts as acts are each contingent upon those respective structures.
Both physiologists and psychologists frequently refer to activities like breathing and seeing as functions without specifying the structures with which they are correlated even when they are known. In other words, they refer to these activities as functions without stating what they are
( 63) functions of. The nature of the correlated term —some structure in this case—is implied or taken for granted.
Psychologists also refer to various mental acts as functions when their organic correlates are somewhat hypothetical, or inadequately known. Reasoning, conceiving, feeling, and willing are cases in point. In labeling these activities functions, psychologists are asserting that these acts are not disembodied activities, but that each is contingent upon some distinctive set of organic conditions even though the exact nature of these may be largely unknown.
Whenever mental acts are referred to as functions, the term is invariably used, in my opinion, to indicate that these acts are not disembodied acts but are acts of an organism and that each is contingent upon some distinctive organic factor. Sometimes this organic correlate is specifically stated at the time, sometimes it is not stated though known, and often it is not stated because its nature is inadequately known.
In dealing with contingent or functional relations, we may define either term, on the basis of its relation to the other. For example, one function of a vocal act is that of representing an object, or we may say that the representation of an object is a function of the vocal activity. One of the functions of breathing is that of the oxygenation of the blood, and this latter may also be characterized as a function of breathing.
We may also note that a series of phenomena may be contingently related to each other as when A is a function of B, and B is a function C, and so on. To keep to our stock example, we may state that the oxygenation of the blood is a function of breathing which is itself a function of the lungs. In this case it is perfectly legitimate to speak of the function of a function, or to say that the function of breathing has a function, viz., the oxygenation of the blood.
Contingent or functional relations frequently exhibit a considerable degree of complexity. A given term may be contingent upon or a function of a number of factors. For example, the color of a negative after-image may at the same time be a function of the color and intensity of the stimulating object, the duration of exposure, the part of the retina affected, and the color of the background upon which the after-image is projected. Breathing may be said to subserve two functions—the oxygenation of the blood, and the elimination of carbon dioxide. Laryngeal activities may likewise be used as a means of communication or as a device for thinking.
Contingent or functional relations constitute a general class that is capable of further specification or particularization. As already noted, functional relations include the relation of activity to structure, and that of use or means to end. It also includes the relation of stimulus and response, cause and effect, the relation between two correlates that are both effects of a common cause, and the relation of present experience to the past experience of the subject. I am not concerned here with the problem of logical classification, but I merely wish to give the reader some sort of a preliminary notion of the wide variety of specific sorts of relation with which a functional psychology is concerned.
( 64) With this conception of the term function, we may now return to the distinction between the programs of an existential psychology and a psychology of function, and we shall contrast them on the basis of their treatment of a specific behavior situation.
I leave my laboratory to go home to lunch, come out of the building and encounter a cold and drizzly rain, spy on the other side of the street the parked automobile of a friend with whose habits I am acquainted, wait until he appears, and secure a ride home.
As we have noted, both an existential psychology and the functional psychologies of the period under consideration are couched in dualistic terms and will deal with the above situation in terms of the subject's experience with it.
In this experiential situation it is possible to distinguish between (a) the fact of awareness, (b) the various sensory contents, i.e., the sensory attributes of the objective situation, of the organism, and of the actions of the organism to that situation, (c) the various meanings of these contents, and (d) their intrinsic and extrinsic relations. For the sake of simplicity we shall ignore the possible presence of affective and imaginal contents and confine our treatment to the sensory aspects of the experience.
The program of existentialism may be stated as follows:
1) It proposes to limit itself to the study of these contents as bare existences, i.e., as abstracted from the fact of awareness, from their values and meanings, and from their functional relations.
2) Its problem is that of the description of these contents.
3) It assumes that these contents are to be described only in terms of their constituent elemental contents. It follows then that the existential psychologist first attempts to analyze the various contents into their elements, and these elements, be it noted, are themselves contents. With the descriptive technique thus obtained by analysis, the psychologist then describes these complex contents as a combination of the elemental contents involved.
As previously noted, the functional psychologist has no quarrel with the positive features of this program. Most functional psychologists are accustomed to incorporate a considerable amount of such material in their texts. They object to the proposed limitations of this program, and insist upon the inclusion of other data.
1) Functional psychology chooses mental acts, such as seeing, tasting, conceiving, and willing, as its objects of study, rather than bare contents.
2) It thus includes the phenomena of meaning and of functional relationships within its subject-matter.
3) Some functional psychologists, I am inclined to think, would object to limiting their scientific task to that of mere description.
4) Functional psychologists, in so far as they do describe, insist upon the necessity as well as upon the right of describing an object—be it a content or a mental act—in terms of its relations to other objects, as well as in terms of its analytical components.
( 65)
5) They have also continually insisted that a description even of contents in terms of their analytical constituents must embrace other components than elemental contents if the description is to be adequate and complete.
I have heard that this latter proposition has been lately rediscovered by the configurationists, and hence I shall add by way of illustration a quotation from an article (7) published in 1909.
"Is the nature of a mental compound accurately seized, after all, when we have told off its constituents, even in their right proportion? . . . And yet nothing, it seems to me, could well be farther from the truth. For the original mental fact which we would describe has, in most instances, what we might call architectural features, and its nature and quality consists not only in the character of its materials but in the manner of their union or arrangement.
"Any analysis that names merely the ingredients may therefore miss the full truth; it may note no difference in compounds that actually are different. The safe and reliable description of the more complex mental facts accordingly requires that our idea of analysis be revised to include an attention to the architectural features of such phenomena, including of course their manner of change. Or if we prefer to let analysis mean what it has ordinarily meant, then only when analysis is supplemented by an account of the form of the process or object is there any guarantee that the description will be faithful to all the fulness of the reality."
A science must first break up its world into convenient units or objects for separate study. As indicated, mental acts are the objects with which a functional psychology is concerned. In experiential terms, an act is a group or pattern of contents exhibiting a unity from the standpoint of its meaningful implications as to end result. An act thus involves the awareness of the adaptive meaning or significance of a pattern of contents, and different acts are to be distinguished on the basis of their end results as well as in terms of their constituent components. The first act in the above illustration is not merely a given pattern of visual and somaesthetic contents, but a pattern exhibiting various meanings. For one thing it is a leg activity, it is also an act of walking, and it is also an act of walking home to lunch. As an act, it_ cannot be adequately defined except in terms of it actual or potential end result. The act of perceiving the cold and drizzly rain is more than a spatial and temporal pattern of visual contents. These contents also exhibit a meaning and they involve a reaction on the part of the percipient subject. The act of perception involves an interpretation of these contents as to their particular objective significance that is relevant to the preceding act of walking home. Thus a functional psychology in the very choice of its objects necessarily deals with meanings and functional relations as well as with contents. It is also obvious that a study of meanings involves that of functional relations and vice versa, for there can be no meanings without such relations.
A functional psychology studies these acts in various ways. It is willing to analyze these acts into their simpler components of meaning, contents, and the relations involved in a pattern of contents. It is also willing to analyze these contents into their elemental contents. It is also interested in studying the various contingent
( 66) relations between the several components of an act, such as the contingent relation of meaning to content, the stimulus and response relations of the alternate leg motions in walking, the effect of the adjustive reaction on the sensory contents in perception, etc. It also studies the contingent relations between the various acts of the series, such as the contingency of the perception of the rain to the act of walking home, the effect of this perceptual activity on the act of walking, the effect of the resulting dilemma upon the discovery of the parked automobile, etc. It will also call attention to the contingency of this series of acts upon the preceding fact of hunger, and to the further fact that this series of acts was instrumental in allaying that condition. A functional psychology is also willing to note incidentally that this satiation of hunger entailed consequences of a physiological and biological character. A functional psychology will also study these acts from the standpoint of their genetic history and note the various features of these acts that are contingent upon the previous activity of the organism. Finally, it will correlate these acts with the structure and physiological features of the organism so far as it is possible to do so. A functional psychology is thus primarily interested in correlating these acts in all possible ways. It suffers from no taboos in this respect. It will attempt to correlate the various features of these acts with anything, provided hat= the correlations are of an observable and demonstrable character.
Functional psychology studies acts whose unity is a matter of reference. Existentialism studies complex contents; it speaks of blend fusions, combinations, and patterns of contents. What is the basis of the distinction between one complex or pattern and two? The same question may well be asked concerning gestalts and configurations. Are the somaesthetic contents involved in each leg movement separate patterns, or is the whole series of contents involved in walking home just one pattern? Are the unitary complexes qualitatively homogeneous spatial and temporal units? What is the criterion of unity involved? Is there any unity except in terms of meaning or reference? Titchener in his texts first develops his descriptive technique of elemental contents, and then proceeds to describe the group of contents involved in perception, ideas, emotions, moods, memory, imagination, and' action, and yet Titchener (13) has taken Wundt somewhat petulantly to task for his lack of insight in retaining a whole array of empirical terms such as perception, emotion, memory, and imagination. Are not the objects of existentialism indirectly differentiated on much the same basis as those of functional psychology, i. e., on the basis of meaning and reference?
Existentialism, as a matter of fact, does not discard all meanings and relations. The contents are named, compared, classified, analyzed into their constituent elements, and described in terms of these elements. Obviously these contents must have some meaning in order to be objects of a science, and obviously these objects are being manipulated on the basis of their relations of similarity and of part and whole, to say the least. The intent of these remarks is not critical. I merely wish to note by way of contrast that existentialism merely discards certain meanings and relations and re-
( 67) -tains others, for it studies these contents on the basis of certain meanings and relations which they bear to one another.
Existentialism does not even discard all contingent relations. Existentialists frequently study the psychophysical relation. Titchener in his Primer of Psychology (8) states that a science must explain, and that mental processes are explained by a statement of their bodily conditions, i. e., in terms of their bodily correlates. Weld (15, p. 65) asserts that the task of the psychologist includes also the correlation of mental and neural processes, but he adds that this correlation implies no causal connection. The writer has always been at a loss to decide whether these relations are studied in their own right, or whether they are utilized merely as a means of analyzing and classifying contents as in the distinction of visual, auditory, and gustatory sensations. If these two relations are studied in their own right, the question naturally arises whether their inclusion is inconsistent with the existential program of analytical description. If their inclusion is not inconsistent with this program, what is the distinctive principle that differentiates the two programs? The author will not attempt to answer these questions.
So far we have been primarily concerned with contrasting the two rival programs, without attempting to evaluate them. We shall now briefly review some of the more important arguments as to the legitimacy of the functional program.
It has been charged that the very term function has been used in a loose, vague, and perhaps inconsistent manner. Certainly the functionalists did not attempt to define the term in any precise way. Perhaps they assumed that the meaning of the term would be evident from the context. Ruckmick has shown that the functionalists did use the term in some consistent way inasmuch as all usages can be grouped under two well-defined categories, while I have indicated that the term as used is capable of a precise and definite formulation.
It has been said that meanings, values, and relations are not introspectable items of experience; only contents can be introspected. Inasmuch as it was generally admitted at this time that introspection is the only observational method of psychology, it follows that meanings, values, and relations are non-psychological data. One cannot introspect a mental act; one can only introspectively apprehend the contents involved in such acts. Much of the functionalistic program is thus non-psychological in character. Meanings, for example, are said to belong to the realm of logic. Functionalism is thus not a true psychology, or rather it is a psychology mixed with logic and other things, with psychology constituting but a small part of the conglomerate mixture.
Titchener (9) has developed his conception of the nature of introspection in a couple of articles. He asserts that we cannot introspect causal relations, physiological dependence, and genetic relations. Causation, dependence, and development are matters of inference and not data of introspection Introspection, we are told, cannot itself be introspected. Perceiving is an act or function, and acts and functions cannot be introspected;
( 68) they are logical abstractions, and we cannot (introspectively) observe any product of` logical .abstraction. We cannot (introspectively) observe relations, but we can observe content processes in relation. We cannot observe change, though we can observe changing content processes. We cannot observe causation, though we can observe content processes that are causally related. Introspection approaches mind from the special standpoint of descriptive psychology; it gives data with which to describe objects. The introspectively observable items of experience are content processes. Consciousness as a describable object is that which can be described in terms of elemental contents and their attributes. Mental data exhibit a host of real relations, and a competent experimenter will note these relations, but he will not use them for purposes of psychological description.
Verbal statements of meaning are informative, but they are not psychologically descriptive. Differences of import or value also transcend description, and psychology must limit itself to description'. Titchener is ere engage, in t e task of expounding and defining the term introspection as he is accustomed to use it, and it is well to note that all usages of terms are to some extent arbitrary. He defines introspection in both negative and positive terms. On the negative side, introspection cannot itself be introspected, i. e., it cannot be psychologically described on the basis of its analytical constituents. On the positive side, introspection is one of those mental acts or functions that is to be defined in terms of its object, and these objects of introspection are invariably contents and their attributes as abstracted from the context of relations, meanings, and values in which they always appear.
All this is quite clear and simple. If one assumes that introspection is the only psychological method of observation, and also accepts the Titchenerian definition of this term, it requires no great feat of logic to conclude that psychology is concerned only with contents, and that meanings, values, and relations are data of a non-psychological character.
Inasmuch as functionalists do concern themselves with these features of mental life, one must assume that their use of the term introspection differs somewhat from that of Titchener. The question at issue then is a matter of terminology and not one of fact.
There can be no dispute concerning the factual question whether one can give a valid observational report about meanings, values, and relations. According to Titchener, a competent experimenter will note and report these meanings and relations; he is merely forbidden to use them for purpose of psychological description. It is also obvious that if one cannot go ' beyond these contents and report what these contents mean or represent there can be no science of physics, chemistry, or biology. In fact, the only possible science would be that of existentialism. Questions of terminology should never be allowed to obscure questions of fact, and certainly the phenomena of meaning, value, and relations cannot be excluded from the realm of psychology on the grounds of their non-observability. According to Bentley (3, p. 401), structuralism has never justified its dogmatic assertion that first-hand observation of human experience was synonymous with structural observation.
( 69)
Several psychologists with functionalistic inclinations have proposed the addition of relational elements to the conventional list of sensory, imaginal, and affective elements, and a few have suggested the inclusion of a meaning element. The writer has sympathized with Titchener in this controversy. Certainly meanings and relations are not contents, and neither are they elements in the same sense of the term as are the conventional elements of existentialism. To refer to meanings and relations as elements that are to be classified as coordinate with the sensory and affective elements is not only illegitimate but confusing. But this fact does not entail their exclusion from all psychological consideration.
One of the most serious charges against functionalism, and in fact against the whole empirical movement, is that it lacks somewhat in respect to its scientific character. Sometimes we are led to infer that functionalism is not a true science, but rather a pseudo-science or a scientific pretender. Empirical psychologies—functional and act psychologies—belong to the realm of the applied sciences as contrasted with the purity of existentialism. Existentialism is a critical science, and empirical psychologies are non-critical or pre-scientific, and, finally, existentialism is referred to as the experimental type of psychology as contrasted with those that presumably are not experimental.
A few excerpts (12, pp. 79-81) may here be quoted to illustrate the general tenor of these criticisms.
"Functional psychology is a parasite, and the parasite of an organism doomed to extinction, whereas intentionalism is as durable as common sense."
"We have found that in both cases (functionalism and intentionalism) they are empirical, that is, technological: they begin and end with 'mind in use.' They represent what we may call an art of living as distinguished from a science of mental life—a general 'applied psychology' that is logically prior to the special 'applied psychologies' of education, vocation, law, medicine, industry."
"It (intentionalism) is thus, like common sense, an applied logic, though unlike common sense its interest lies more in the logic and less in the results of application."
"The one complete and positive reply to intentionalism is the existential system, the system that is partially and confusedly set forth in the works of Wundt and Külpe and Ebbinghaus. If we can build psychology upon a definition that is scientific as the word 'science' is to be understood in the light of the whole history of human thought; and if we can follow methods and achieve results that are not unique and apart but, on the contrary, of the same order as the methods and results of physics and biology; then, by sheer shock of difference, the act-systems will appear as exercises in applied logic, stamped with the personality of their authors. They will not, on that account, languish and die, because 'mind in use' will always have its fascination, but they will no longer venture to offer themselves as science."
It would seem from these and other comments that empiricism (functionalism and intentionalism) transgresses the spirit of a pure science i three respects: (a) It brazenly studies the uses or utilities of mental act singly and as a whole. (b) It has been avidly instrumental in exploring and developing the various special fields such as testing and educational, industrial, legal, and abnormal psychology. (c) And, finally, it has exhibited some pride in the social utility of its labors.
( 70)
There is no doubt that functionalism has done these three things, but the charge that its so doing is a violation of the spirit of a pure science is another question, and one concerning which there may be a legitimate difference of opinion.
What is the difference between a pure and an applied science, and why do we regard a pure science as the more valuable?
1) The two cannot be differentiated on the basis of the situation or the locality in which the work is done. Pure scientific research may be conducted in an industrial laboratory as well as in a university laboratory or in a secluded cloister. In fact, many exhibitions of pure scientific research are being furnished yearly by some of our better industrial laboratories and by some of our psychological clinics as well.
2) Neither can they be differentiated on the basis of the field or phenomenon investigated. In the field of educational psychology most of the studies on memory and learning have been conducted in the spirit of pure science. I know of some studies of the perceptual activities involved in reading that are models of a pure scientific attitude. A few of the studies in the field of mental tests are exhibitions of pure scientific procedure, and many studies of aberrant behavior have been conducted in the same spirit.
3) We are sometimes told that a pure science is one that has no concern for values, but it is concerned at least with scientific Values. Not all facts or attributes of a phenomenon are equally significant or valuable from the standpoint of science any more than they are from the standpoint of everyday behavior. One might study and compare and classify rocks on the basis of such superficial qualities as color or size and conduct the investigation in a pure scientific attitude, but such a study would hardly be considered a legitimate scientific undertaking. Such facts would lack any scientific value. Many of the early botanical classifications were scientifically futile, and we may refer to James's comment upon the status of the early studies of emotion. Science does not study anything and everything even within its own field. Not all scientific facts are equally valuable even from the standpoint of science. Science does have some sort of a concern for values. What is the criterion of the scientific value of a fact? I raise the question, but shall not attempt to answer it.
4) According to one definition, a pure science is one that is solely interested in an adequate understanding of the phenomena under consideration, but one that has no concern for the social or. practical value of its
findings. A pure science merely wants to know and is wholly unconcerned as to whether the knowledge it obtains can or cannot be usefully applied to the guidance of conduct.
This unconcern as to the utility of scientific knowledge needs a word of comment. A pure scientist can exhibit no aversion to the discovery of useful knowledge. He will neither intentionally nor inadvertently arrange his investigations so as to avoid the possibility of obtaining useful data. Neither will he refrain from studying certain problems and investigating certain fields for fear he may discover something useful. A pure scientist will welcome both useful and useless knowledge with equal gusto. It is
( 71) related that a noted mathematician concluded his demonstration of a new mathematical formula with the statement, that he was specially proud of the fact that the formula could never be turned to any practical use. Such an attitude is not consonant with that of pure science.
5) Finally, there is the pragmatic point of view that science must ultimately justify itself on the basis of the social value of its findings, but that the pure science attitude of seeking to understand without any concern as to immediate values is the best method of ultimately achieving socially useful knowledge. A scientist thus hopes and expects that his labors will ultimately be socially fruitful, but he recognizes that the best way to achieve this result is to adopt an attitude of unconcern as to the immediate value of his experiments. With this attitude of mind, a scientist may deliberately choose, if he wishes, to enter those fields where the probabilities are greatest of discovering socially significant results. This point, of view is, perhaps, a reflection of our national temperament.
We may now return to the three charges lodged against functionalism and empiricism in general. The fact that functionalism exhibits some pride in the social value of its achievements is no violation of the spirit of pure science. A pure scientist welcomes both useful and useless knowledge with equal acclaim. We may note that chemistry, physics, geology, and even mathematics are also accustomed to point with considerable pride. As already indicated, the development of the various special fields does not necessarily involve a transgression of the strict letter of the law, for a pure science is not to be characterized on the basis of what it studies. What better exhibition of the pure scientific attitude can be found than that of Spearman in the field of mental tests? Finally, the uses or utilities of mind can be studied with purity of scientific attitude. There is considerable difference between being concerned with studying the use of mind and being concerned with the uses of what we find out from that study. Theoretically it is possible to secure wholly useless knowledge about the uses of mind.
We may now raise the question whether existentialism is entirely free from taint in this respect. Do the existentialists exhibit an attitude of strict unconcern and indifference? Do they not show some slight concern lest they find something useful? Why all this aversion to anything that is tinged with use? Why the emotional complex against the special fields? Why the fear of contamination? Why the horror against the useful? Is this the proper attitude of a pure and critical science, or is their attitude somewhat hypercritical? I suspect that the existentialists, like the mathematician referred to, have been leaning over backwards in their attempt to preserve a spotless purity.
Functional psychologies, according to Titchener (11), are teleological, and, teleology is essentially non-scientific. Functional psychology was born of the enthusiasm of the post-Darwinian days, when evolution seemed to answer all the riddles of the universe; it has been nourished on analogies drawn from a loose and popular biology. Not only psychology but biology is suffering from an unbridled license of teleological interpretation. Tele-
( 72) -ology came down to the functional psychologist from the older empiricism. It is guaranteed by philosophy and technology, and it is justified by biological example. Small wonder then that he should step easily, even heedlessly, into the teleological attitude.
Titchener's charge that teleological interpretations have been overdone in both the fields of psychology and biology, in my opinion, is true. Starting with the doctrine that the direction of evolution is a result of natural selection and that natural conditions operate by eliminating the most unfit and selecting those that are fit, many early writers assumed that each and every evolutionary product must have a survival value. If no value is apparent, they must discover and assign one irrespective of the facts. Since emotional reactions, for example, are presumed to be evolutionary products, each emotion and each characteristic of these emotions must have a survival value, and it is the business of the psychologist to assign these even though he can do little better than make a wild guess as to their nature. This attitude is the resultant of several illicit assumptions as to the logical implications of the theory of natural selection.
As careful thinkers early pointed out, evolutionary products need have a survival value only under those circumstances in which they were selected. After they have been selected, they may be perpetuated and continue to exist when the conditions have so changed that they have no survival value. In other words, biologically useful characters may become useless with a pronounced change in the conditions of life.
An organism may be regarded as a unitary group of hereditary characters—structural and behavioristic. It is often tacitly assumed that natural selection operates directly upon the individual characters themselves, and that it eliminates and preserves these characters each according to its own individual merit. Natural selection, however, operates upon the organism, i. e., it selects a complex group of characters. It is the organism that either survives or goes to the wall in the struggle for existence. Not all of the characters of the surviving organisms thus need to be useful. Characters may appear and persist that are neither useful nor detrimental to survival. Organisms with a number of biologically neutral or indifferent traits may survive if they have a sufficient number of useful ones. As a matter of sheer theory, organisms with a detrimental characteristic may continue to exist if this defect is sufficiently compensated for by useful traits. There is thus no need to assume that each and every biological character has a survival value
The very term natural selection erroneously suggests that natural forces directly select the fit organisms. The natural forces, however, operate to eliminate the unfit, and the selection of the fit is incidental to the process of elimination. Moreover, the degree or extent to which the unfit are eliminated is a function of the degree of competition in the struggle for life, and this latter varies with circumstances. Only the most unfit are eliminated, and the least unfit survive. Again not all of the characters of the surviving organisms need be useful, and furthermore the organisms that survive do not need to be perfectly adapted to their environment, i. e.,
( 73) 100% fit. According to the theory of natural selection, they need only to be more fit than those that were eliminated.
It may be well at this point to note the distinction between biological utility and other modes of usefulness. Trees are useful to man for their lumber, but this is not a biological utility. The theory of natural selection does not pretend to account for the evolution of this characteristic of trees on the basis of such a use. The theory accounts for the evolutionary development of a character only on the basis of its utility to the organism that possesses it, viz., the tree, and not on the basis of its usefulness to some other organism like man. Again some characters of an organism may be selected and preserved because of their survival value, and then be utilized for other purposes at a later time. A person might employ his toes for purposes of writing, but this use in no way accounts for the evolutionary development of these organs. Society is accustomed to use the fear reaction to attain certain social ends, but this does not necessarily represent its biological or survival value; in fact, this social value does not even justify the assumption of a biological value for this trait.
In respect to teleological explanations, we may note that the process of natural selection on the basis of survival value accounts merely for the preservation of traits and not for their origin. The process of natural selection does not purport to explain the origin of mutants, but given mutants it accounts for the direction of evolutionary development. Biological needs and utilities select but do not create. The existence of a need does not guarantee the development of an organ to supply that need.
We have admitted that psychologists have been guilty of some weird teleological interpretations, but psychologists have not been the only sinners. Even Titchener is not entirely free from guilt in this respect, for some of his criticisms involve certain of the erroneous assumptions that have just been mentioned.
He takes the functionalist to task for his inconsistency in not giving a teleological interpretation to every mental item. The psychologist may answer any number of whys, but he is still faced by unanswerable why-nots that throw doubt upon his positive explanations. How has the development of red-green vision aided man in the struggle for existence, or what has man gained by the "unique compromise process"which gives rise to the purple sensation? These and like questions are not touched, we are told. Is not Titchener here assuming that all evolutionary products must have a survival value?
He refers to Judd's statement concerning the lack of an electric sense in man and the utility of such a sense-organ equipment, and then makes the following comment :
"Granted that the facts are as stated and granted that this furtherance of knowledge is useful, why have we not the special organ?—for it is surely evident that biological conditions, which have produced the `electric fishes,' are also competent to produce an electrical sense-organ in man" (11, p. 539).
Does Titchener assume that the theory of evolution by natural selection involves the doctrine that biological needs create the means of their attainment?
( 74)
As a part of his criticisms, he caustically comments upon the fact that the human eye is far from perfect inasmuch as its native usefulness has been immeasurably improved by the microscope and the telescope. Does this criticism not involve the assumption that the theory of evolution implies a 100% fitness?
While teleological interpretations have been overdone, it does not follow that teleology is essentially non-scientific and that all teleological interpretations should therefore be discarded. One might as well argue that science should cease theorizing and making hypotheses and conclusions because it has made so many mistakes in these respects in times past.
We must recognize that the place of teleology in science is a moot question concerning which there are differences of opinion among biologists, psychologists, and philosophers. Titchener's attitude that teleology is non-scientific finds its supporters among biologists, but it is also well to note that many biologists as well as psychologists have not discarded all telic conceptions.
What is teleology, and in what respects is it legitimate and when is it illegitimate? I would say that telic conceptions are involved in all statements concerning use, utility, adaptation, purpose, and means and ' ends, and all of these terms imply a certain kind of contingent relationship.
I see no objection to noting and stating these relations in so far as their factual character is observable and demonstrable. Such statements as the sense-organs are the means whereby we gain knowledge of the objective world, the muscles are devices for reacting to that world so as to satisfy organic needs, vocal activities are used in thinking, etc., are unobjectionable as mere statements of fact. One difficulty arises when one of the terms of the relation is supplied by a process of speculative inference, and these speculations masquerade under the guise of fact. But this type of difficulty is not peculiar to the study of telic relations.
The usual objection to such statements of telic relations—even factual ones—is that they imply an illegitimate type of explanation. It is sometimes charged that such statements imply the existence of some design, purpose, insight, or intelligence—some prior existential factor that is causally responsible for these telic relations. Again it is said that such statements tacitly assume that the end result operates as the cause of the prior process by which it was attained—an assumption which violates the temporal requirements of a cause-and-effect relation.
Can one make a statement concerning any of these telic relations as mere statements of fact without any explanatory implications whatever? The author is disposed to believe that these statements can be and are often made without such implications on the part of either the writer or the reader.
When implications are involved, the statements may not imply any particular kind of explanation—let alone an illegitimate one, such as that of design. The purposive psychologist does assume more or less explicitly the existence of innate conscious purposes to explain the origin of adaptive behavior, but in my opinion the great majority of functional psychologists
( 75) do not do so either implicitly or explicitly. With those functionalists with whom I am well acquainted, implications of design are foreign to their intent and to their unconscious biases as well. If design is suggested, is the fault to be found in the mode of statement or in the interpretative reaction of the reader?
However, there can be no objection to statements that are explanatorily suggestive, if these telic relations can be legitimately explained. The usual explanation of the adaptive character of our acquired reactions is that of the law of effect, which accounts for the selection and elimination of acts on the basis of their consequents. The law does not attempt to explain the origin of these acts, any more than does the theory of natural selection purport to account for the origin of mutants. The law merely accounts for the fixation of the adaptive acts and the elimination of the non-adaptive ones, and thus accounts for the direction of mental development. Neither does the law of effect violate the temporal requirements of a cause-and-effect relation, for many of the effective consequents occur during the performance of the act, and besides the law assumes that these consequents merely affect the subsequent performance of that act (5, pp. 95-96).
We would thus conclude that telic concepts can be legitimately retained in a science so long as it confines itself to factual statements of these relations and explains these facts in a legitimate manner.
Titchener's statements that science is concerned only with description and that objects can be described only in terms of their constituent elements deserves a few words of comment . What is description and why does science describe? Scientists necessarily report their findings, and in this sense of the term they "describe"not only their objects of study but their methods, procedures, hypotheses, and the knowledge they obtain of these objects as well. Description in this sense is only the final step of science, for obviously this description presupposes a considerable variety of prior activities. Moreover, this type of description cannot be limited to statements of the analytical composition of that which is described, for procedures, hypotheses, and analytical elements, as well as the objects analyzed are described. What, then, does Titchener mean by description? Perhaps the question may be clarified by ignoring the term description and defining Titchener's program in terms of the type of knowledge sought. In effect, the Titchener doctrine merely asserts that any legitimate scientific knowledge of psychological objects is limited to a knowledge of their constituent elements and the laws governing their combinations in those objects. Titchener's appeal to physics and physiology in support of this doctrine is hardly appropriate. The analogous program among the natural sciences is that of chemistry and histology, while the program of physics, physiology, geology, and biology is more akin to that of functionalism. Analytical knowledge of the constituent elements of objects is not the only scientific goal, and in this connection we may quote from Bentley (3, pp. 401-402) :
"Neither has it (structuralism) justified its contention that the main method of
(76) science was analysis. It is, as I think, not much less than a caricature of the sciences of nature to say that the physicist, the chemist, and the zoologist are always and only analyzing it has, for some time, been generally conceived to be a formal and logical—not a realistic—view of science which has brought into relief the typical chemist or physicist as forever breaking down his substances into constituent elements. Analysis, surely, but not simply analysis: and, for many problems, not analysis at all."
I would add to this quotation the further statement that there are other modes of analysis than that of the existential type.
The main defects of the functional psychologies of the period under consideration are, in my opinion, those that arose from their adoption of a dualistic position. Dualism involves no difficulties to an existentialist because he stays strictly within the confines of consciousness. When conscious activities, however, are conceived as a separate but effective part of the total biological process, the question of the mutual relations of these dual parts to each other immediately comes to the fore.
The existentialists have been caustic and trenchant in their criticisms. We may here refer to the much criticized and widely quoted statement of Angell (1, p. 59) :
"Let it be understood once and for all that wherever we speak, as occasionally we do, as though the mind might in a wholly unique manner step in and bring about changes in the action of the nervous system, we are employing a convenient abbreviation of expression ..."
Titchener has also voiced his objections to statements as to the origin of consciousness, when and where consciousness comes in, and its function as a solver of problems.
When the functionalist treats of the observed uses of particular acts like perception, he is on safe ground. When he deals with the biological origin of consciousness as a whole and its function in the biological process, he is entering the field of speculation where there is an opportunity for a legitimate difference of opinion. Moreover, speculative opinions are likely to be expressed as statements of fact. Neither should an empirical science of fact adopt a position which forces it to substitute circumlocutions for straightforward statements of fact.
What happened to this functionalistic movement? Did it evolve and disappear in the process of development, or does it still persist in a modified form? In my opinion, American empiricism has undergone two major developments since the time of which we write.
Dynamic psychology represents a further development of the implications of the biological point of view. Functionalism had assumed that mental acts grow out of and minister to the biological needs and impulses of the organism. According to this conception, the organic background of needs and desires operates to motivate and direct the whole course of mental development, but this fact was more or less taken for granted, or at least the influence of these factors was not sufficiently emphasized. In their emphasis upon drives and motivation, dynamic psychologists have been attempting to portray these factors in a manner that is more commensurate with their importance.
( 77)
Behaviorism, to a considerable extent at least, was an attempt to avoid the difficulties inherent in a dualistic position. The radical behaviorists solved the problem by either denying or ignoring the fact of consciousness, while the moderate behaviorists are prone to talk in monistic terms of the behavior of a psychophysical or a psychobiological organism.
The above fact has been well developed by Weiss (14). He notes that the functionalistic assumption that conscious activities influence behavior is inconsistent with its dualistic position. The further assumption of parallelism the functionalist fails to explain. The functionalist to be consistent must accept interactionism, and he is then confronted with the task of rationally conceiving of this process. The further possibilities are to study consciousness alone and omit its influence upon behavior, i. e., discard a large part of the functionalistic program, or to study behavior alone and neglect or disregard the fact of consciousness. Weiss then proceeds to develop and justify his particular program in which consciousness is disregarded.
Weiss apparently assumes that the dualism of the functionalist is necessarily ontological in character. Given this assumption, there is no escaping his conclusions. I doubt the truth of his assumption, however. Angell has said that a functionalist can accept any one of the various conceptions as to the nature of the mind-body dualism with the single exception of that of epiphenomenalism. I do not pretend to know the philosophical inclinations of most functionalists, but it has always been my impression that Angell's dualism was of the methodological variety. It has also been my opinion that dualism is a poor methodological device for a functionalism with strong biological leanings. I agree with Weiss that a functionalist is bound to adopt some sort of a monistic conception, but I think that there are other monistic positions possible than the two alternatives that he mentions.
The functionalistic movement has thus undergone considerable development. Did functionalism disappear with this development, or are these later developments functionalistic in character? The answer depends upon the definition of functionalism adopted. Functionalism and existentialism represent two opposing points of view toward the subject-matter of psychology, and this subject-matter, at the time of this controversy, was conscious processes dualistically conceived. If functionalism is to be defined in terms of point of view as well as in terms of subject-matter, i. e., as a study of the functions of conscious activities, then functionalism per se is on the wane. If functionalism, however, is to be defined solely in terms of its point of view without any regard to what it studies, then the various behaviorism's are functional psychologies. For example, one can study behavior in two ways: (a) One can assert that the object of psychology is to describe behavior, and that it can be described only in terms of its constituent elements, viz., reflexes. It is thus the business of psychology to analyze the various complex forms of behavior into their simplest reflex elements, and to study the laws governing the combinations of these elemental reflexes in behavior patterns. We have here a program essentially like that of the existentialist with simple reflexes substituted for his sen-
( 78) -sation elements. (b) On the other hand, one can adopt the functionalistic program of studying functional interrelations of the temporal parts of a complex act, its functional relation to organic needs, its dependence upon previous behavior, and its relation to the structural and physiological characteristics of the organism. How one shall answer the question thus depends upon the definition adopted. I shall let the reader answer the question as he sees fit.
What has been the outcome of this controversy? Some of the existentialists still maintain the faith, some have developed functionalistic inclinations, and a few have given signs of seeking refuge in configurationism. I know of no whole-hearted conversions to existentialism from the functionalistic ranks. The American empirical movement has maintained itself against attack and has gone on developing in accordance with its own particular genius. The controversy in acute form did not persist for long. A working truce of mutual respect was soon attained—a truce that has not been violated except for an occasional outburst on the part of some irrepressible spirit.
REFERENCES
1. ANGELL, J. R. Psychology. New York: Holt, 1904. Pp. vii+402.
2. ———. The province of functional psychology. Psychol. Rev., 1907, 14, 61-91.
3. BENTLEY, M. The work of the structuralists. Chap. 18 in Psychologies of 1925. Worcester, Mass.: Clark Univ. Press, 1926. Pp. 395-404.
4. BORING, E. G. The psychology of controversy. Psychol. Rev., 1929, 36, 97-121.
5. CARR, H. A. Psychology. New York: Longmans, Green, 1925. Pp. 226.
6. RUCKMICK, C. A. The use of the term function in English textbooks of psychology. Amer. J. Psychol., 1913, 24, 99-123.
7. STRATTON, G. M. Toward the correction of some rival methods in psychology. Psychol. Rev., 1909, 16, 67-84.
8. TITCHENER, E. B. A primer of psychology. New York: Macmillan, 1898. Pp. xvi+314.
9. ———. Prolegomena to a study of introspection. Amer. J. Psychol., 1912, 23, 427-448.
10. ———. The schema of introspection. Amer. J. Psychol., 1912, 23, 485-508.
11. ———. Functional psychology and the psychology of act, I. Amer. J. Psychol., 1921, 32, 519-542.
12. ———. Functional psychology and the psychology of act, IL Amer. J. Psychol., 1922, 33, 43-83.
13. ———. Experimental psychology: a retrospect. Amer. J. Psychol., 1925, 36, 313-323.
14. WEISS, A. P. Relation between functional and behavior psychology. Psychol. Rev., 1917, 24, 301-317.
15. WELD, H. P. Psychology as science. New York: Holt, 1928. Pp. vii+297.
