Theoretical Unification in Psychology

Tolman, C.W. (1988a). Theoretical unification in psychology: A materialist perspective. In W.J. Baker, et al. (Eds.). Recent Trends in Theoretical Psychology (pp. 29-36). New York: Springer-Verlag.

THEORETICAL UNIFICATION IN PSYCHOLOGY: A MATERIALIST PERSPECTIVE

Charles W. Tolman
University of Victoria, British Columbia, Canada

SUMMARY: The case for a materialist foundation for psychology is described in which theoretical unification emerges as an on-going historical process, not as a state. Theoretical indeterminacy is overcome through appeal to evolutionary and historical contexts which allow evaluation of theories in terms of essentiality and relevance.

The problems of theoretical indeterminacy, disunity and fragmentation in psychology will not be solved by the revival of any form of positivism (e.g., Staats, 1986). It is well known that the Humean epistemology that defined positivism made it impossible to find a satisfactory solution to the problem of objectivity, and that, as a result, the ultimate goal of a unified science remained chimerical. Many psychologists saw that down the positivist epistemological road lay the dreaded end of solipsism (e.g., Wertheimer, 1972, p. 126). The only reasonable alternative between the Scylla of solipsism and the Charybdis of dogmatism seemed to be some form of metaphysical pluralism. Some of us settled, in short, -however uncomfortably- for the indeterminacy and fragmentation that mark what others now see as a crisis in our discipline.

Despite all the critiques of positivism, we seem to be no further ahead. The reason is that most critiques have not penetrated to the skeptical-subjectivist epistemology that is the root cause of its problems. As a result, the tendency has been to reject the scientistic appearances of positivism, while preserving its essential defect, which, from the point of view of objectivity, and therefore also of theoretical determinacy and unification, is the inability to specify an independently existing, nonarbitrary ground to serve as a basis for resolution of theoretical differences.

Philosophical materialism [FN-1] claims to avoid this defect. If, as materialism claims, such a ground exists and is knowable, then the foundation of a unified conception of science is possible. I can only hope to [p. 30] sketch out this possibility here.

Materialism asserts that there is no limit, in principle, to our knowledge of the external (or internal), objective, mind-independent world. This does not mean, however, that knowledge cannot be partial or mistaken. It is almost always the former and to often the later. This is accounted for by two characteristics of knowledge. First, knowledge is developmental: it is a process of 'getting to know'. In this development we almost never come to know a thing completely because, as we come to know more about it, we also come to recognize that there is more to be known. It is reasonable to suppose that things are epistemologically inexhaustible. The result is that our knowledge of anything at any given time must be partial from one perspective or another.

Second, a distinction must be made between perceptual and conceptual knowledge. The blind men examining the elephant had correct, but inadequate, perceptual knowledge, and jumped to premature, and thus largely mistaken, conclusions about what it was they were touching. From the materialist point of view, the probability of true conceptual knowledge increases as the fund of perceptual knowledge increases. Continued, cooperative exploration of the elephant would eventually have led the blind men to its correct conceptualization.

It follows from this that truth is relative, not to some kind of arbitrary 'framework' as in skeptical/relativist theories of truth, but to the objective content of knowledge, i.e., the extent to which knowledge reflects not sense data, but the object itself.

The acceptability, indeed necessity, of theoretical pluralism is thus relative to the extent of objective knowledge about something. At an early stage when ignorance dominates, pluralism is necessary and facilitates getting-to-know. As the objective content of knowledge increases through practice, pluralism ought to decrease. Once the elephant has been adequately explored, there will tend to be agreement that it is an elephant, and anyone who continues to insist that it is an umbrella will be regarded as intransigent and disruptive. Thus pluralism represents a developmental stage of knowledge and not a permanent condition.

Materialists also insist that knowledge is social. No single individual does, or can, bear the burden of discovering everything by and for himself. That is why we have schools, books, computers, research organizations, and conferences. This means that individuals transcend whatever biological limits might exist for the various cognitive functions. Whoever has acquired the skill of reading and some simple rules of access 'possesses' the whole fund of historically accumulated knowledge contained in the library.

Opponents of a unitary conception of science often charge its advocates with dogmatism. The implication is that only skepticism/pluralism can save us from this end. The correct response is to show that skepticism, [p. 31] like pluralism, is not a position, but the absence of one, and that its proper historical role is to overcome itself. As exemplified in the work of Descartes, skepticism is a means, not an end. The metaphysical skeptics and pluralists simplistically assume there to be only two extreme possibilities: complete knowledge or none at all. But this is itself a form of dogmatism that materialism successfully avoids. Metaphysical pluralism is vulnerable to dogmatism precisely because it denies any independent rational criterion for the resolution of theoretical differences. The materialist is not suggesting that resolution be brought about by weight of traditional or other arbitrary authority, but rather on the basis of appeal to the thing itself, which is the surest way of avoiding dogmatism.

But what makes materialists believe they can appeal to the 'thing itself'? The answer is most easily given by another question: Why shouldn't they believe it? Why, when seeing a chair, should one claim to be seeing anything other than the chair itself? Those who deny access to things in themselves (as did Hume & Kant) generally do so because they hold to an indirect realist or representationalist theory of perception. But such a theory has been shown to be both incoherent (e.g., Wilcox & Katz, 1984) and unnecessary (e.g., Gibson, 1966). I shall not attempt here to make the case for a direct realist theory of perception; it is sufficient for present purposes to indicate that materialism ultimately relies upon such a theory.

Summarized to this point, my argument is that a materialist [direct realist] epistemology is plausible, and that only materialism -certainly not positivism in any form- can provide the necessary [ontological-methodological] foundation for a truly unified psychology. It must be emphasized, however, that the development of a unified and cumulative psychology is not a simple function of time and effort. Nor is it a simple function of materialist [or direct realist] assumptions. This is abundantly clear from the fact that, despite much positivist rhetoric, most experimental psychologists in this century have operated according to roughly materialist norms. Yet, the 'historical truth' is that "opposing theories have survival value, and that an appeal to the facts as a way of choosing between theories is a very complex process, not nearly as decisive in practice as we might expect it to be" (Hilgard & Bower, 1966, p. 9).

Materialism, time, and effort are necessary, but something more is needed to make a sufficient condition. According to Hilgard and Bower (1966):

Accumulation of knowledge means neither mere fact-gathering nor isolated hypothesis-testing, but thoughtful systematic approaches to meaningful questions leading to conclusive thinking... Some good high-level thinking is needed, keeping in mind criteria of relevancy, so that significant controversies are resolved on the basis of firmer knowledge... We need, along with contemporary model-building, some general psychologizing, that is, repeated reflection upon the larger questions of [p. 32] psychology to determine whether or not our approaches are indeed leading to their answer (p. 583).

The kind of program which Hilgard and Bower appear to be recommending has been underway at the Free University of Berlin (and elsewhere in West Germany, Denmark, and the Netherlands) under the rubric of 'Kritische Psychologie' [Critical Psychology] for almost 20 years. I shall describe the main points of the materialist critical-psychological formulation of the problem and its solution as it was summarized by Klaus Holzkamp in 1977 (1978).

Start with a typical psychological theory, say, a 'consistency' theory in social psychology. From the theory we deduce hypotheses which are then translated into independent and dependent variables. Experimental design and statistical assessment are used to ensure both necessity and sufficiency of the independent variable conditions with regard to the expected result. If the result occurs and is judged reliable, the hypothesis and theory are treated as having been confirmed.

In this procedure there are two sources of theoretical indeterminacy, one minor and one major. The minor one is what might be called 'weak confirmation'. Our experimental and statistical methods have come to be so highly refined that under existing rules of procedure almost any selected relationship has very favourable chances of being confirmed. Holzkamp has called this 'Bestatigungsfreundlichkeit' (confirmation-friendliness). Holzkamp shows how this is supported by our standard criteria for Null hypothesis testing. With the right sample size and 'alpha level', very little explained variance can count as a confirmation of the experimental hypothesis.

Holzkamp is not saying that this is wrong in itself. Nor is he saying that these methods should be relaxed or abandoned. That would surely only make matters worse. Rather, he is pointing out how our methods allow the confirmation of hypotheses, which, though correct, may be entirely irrelevant and inessential.

The major source of theoretical indeterminacy is a lack of broader perspective. While we may go on confirming our 'consistency' theory ad infinitum, we will never learn from our efforts whether 'consistency' is actually relevant or essential to human sociality. Another theorist, interpreting sociality in other, even opposing, terms, and adopting similar experimental procedures, may likewise be able to find infinite confirmation. The resolution of the differences between the theories is therefore not to be sought in further empirical confirmation or further refinement of experimental methods, but in a broader perspective on sociality that can allow us to judge the relevance or essentiality of particular theories.

What, then, is meant by relevant and essential? According to Holzkamp (1978): [p. 33]

Relevance is... a real mark of status within the dimensional structure, such that theories become scientifically more relevant, i.e., capable of containing more real knowledge, to the extent that they conceptually embrace more relevant dimensions: dimensions are... more essential than others to the extent that they are more fundamental, i.e., depend less on others than others on them, and consequently a variety of different superficial phenomena, the variability of which derives from less essential dimensions, are lawfully related to one another by virtue of relation to them (p. 157).

How is essentiality determined? From an evolutionary point of view, essential is what contributes to survival. It is essential that an organism respond appropriately to the demands placed upon it by its environmental context. Essentiality of a process can thus be determined by a study of the evolutionary or other developmental context that has shaped it.

Holzkamp calls this approach to essentiality 'functional historical' and it is clearly akin to the ecological approach of J.J. Gibson (1966). An important difference between the two is Holzkamp's insistence upon the centrality of the distinction between phylogenetic and societal-historical contexts. Human beings relate to the world both as evolved organisms and as societal-historically conditioned beings. The distinction is important for psychology because its subject matter is a more immediate reflection of the latter than of the former.

What this means at the organismic level is illustrated by Gibson's (1966) discussion of "The Environment as a Source of Stimulation" (Chapt. 1). "In considering the problems of perception in man and animals the first question to ask should be, what is there to be perceived?" [(p. 7)]. Gibson identifies the organism-relevant dimensions of the physical environment such as rigidity, gravity, and electromagnetic radiation, and shows how each contributes to the evolution of perceptual capacities. What emerge as essential to perception are those characteristics which contribute to the survival of individual organisms, that is, that make possible the identification of food, predators, features of the terrain for purposes of getting about, etc.

At the societal level a uniquely human quality emerges and transforms the organism-world relationship into a person-world relationship. Humans have come to live in a world which they have themselves produced through collective, planned intervention in Nature. Children may be considerably developed before they ever come into contact with Nature (if they ever do). Meanwhile, children have to deal with language, spoons, prepared foods, diapers, toys, strollers, etc., all parts of an environment that has been manufactured for them by their fellow human beings. [p. 34]

At least two significant contrasts with the organism-world relationship should be noted. First, whereas animal formations depend for their survival upon the survival of individuals, human individuals depend for their survival upon the survival of their group or society. Humans no longer live in Nature; they are evolutionarily suited only for survival in society. Second, while adaptation in the original biological sense may still play a role in human existence, it has been displaced in importance by a process of acquiring the skills necessary for the survival of society. This process has been called appropriation ("Aneignung," Holzkamp, 1973, p. 53).

It is important to note that historical nature of the societal environment that shapes personality and the special psychological processes. A child born into a traditional feudal society faces very different problems of adaptation and appropriation than a child born into an advanced industrial capitalist society.

What is being proposed here is that, just as we cannot gain an unequivocal understanding of an animal's perceptual capacities without having first identified the relevant dimensions of the environment in which they have developed, we cannot hope to understand human psychological processes unequivocally without having first identified the relevant societal dimensions in which they have developed.

It is useful here to be reminded how radically traditional psychology has evaded questions of developmental context. In a fairly well known passage, Lewin (1969, cited in Holzkamp, 1978, p. 168) advised that if we want to understand the behaviour of an unruly child who has been threatened by his mother with the police, it is not its actual relations with the police that matter, but rather only what the child believes about the police. In short, explanation of the behaviour comes from what is in the child's mind, not from the context that shaped its mind.

This kind of thinking has been very common in psychology. Zajonc (1968), cited in Holzkamp (1978), wrote that:

The physical and objective properties of social stimulation and incentives have always been assumed to be less significant for the analysis of social behavior than their subjective counterparts. There is less correspondence between the objective properties of social stimuli and that way in which they are perceived than there is in the case of other forms of stimulation, for example hue. Why this is true is not understood even now, but the lack of systematic dependence of social perception on objective properties of stimuli has prompted social psychologists to focus on the cognitive representations of social stimuli and incentives themselves (p. 170).

But if it is the objective world in which we are born, raised, and must work effectively in order to guarantee continued survival of the species, such a lack of correspondence ought to be of immense and urgent [p. 35] concern. Nothing will be learned about it if we continue to withdraw to the level of 'cognitive representations'. What's more, we won't be able to develop unequivocal theories about 'cognitive representations' and behaviour. The current theoretical confusion in social psychology provides ample proof of that.

Holzkamp recognizes at least three distinct ways in which theories can be irrelevant or inessential. First, they may be confused about conceptual distinctions and classification boundaries. This is evident in the distinction between learning and innate. Based on work by Lorenz and other ethologists, Holzkamp-Osterkamp (1975, pp. 112ff.) showed that the two are not exclusive, that learning is an innate strategy, and that the real distinction is one of adaptive strategies varying in potential flexibility with respect to external contingencies. At the societal level, Holzkamp (1973, pp. 173ff.) identified similar theoretical difficulties stemming from a false exclusion of subjective and objective factors in human perception.

Second, theories may be falsely extended from one level (e.g., organismic) to another (e.g., societal). a recent example is the early version of sociobiology. Genetics may be an entirely relevant and essential dimension for a theory of insect social behaviour. In a theory of human social behaviour, however, it can obscure what is really essential and relevant: the influence of the historical societal context of human development.

Third, theories can falsely universalize that which is true in particular instances. Ute Holzkamp-Osterkamp (1975) demonstrated that much of our understanding of classical and operant conditioning is in fact a misunderstanding of broader, evolutionarily determined flexible adaptive strategies. The theory of reinforcement, for example, evaporates in the broader view as an extreme instance of a particular strategy. None of this can be seen if attention is confined, as it usually is, to the abstract variables specified by operant theory. And from this restricted view it will remain forever impossible to understand how the grain of truth in operant theory can be integrated with the grains of truth in opposing theories. Only the broader evolutionary perspective makes that possible.

The advantage of any materialist theory is that differences of opinion and interpretations pertain to matters of fact. The resolution occurs through a direct appeal to those facts. It is on this basis that a materialist expects that competing theories will eventually merge into a single, unified theory and that there will be an accumulation of genuine knowledge. The levels of fact and hypothesis have never been particularly problematic. Serious problems have existed at the level of theory because we have not known to which objective facts we should appeal. We have now learned where to look, and that is at the phylogenetic and historical developmental contexts of the process under study. Until very recently psychologists have simply not done this in any systematic way. [p. 36]

It should be noted, finally, that unification cannot be understood as an absolute state, even potentially. Rather, it is a process. What can be expected is that relative unification will occur as theoretical differences regarding specific processes are resolved. This, however, will lead to the recognition of new problems and thus create a new relative 'disunity' marked by renewed theoretical pluralism.

What distinguishes this process from that which now exists is that psychologists will, in recognition of it, act with the aim of resolving differences and producing an increasing fund of real knowledge, instead of capitulating to metaphysical pluralism and producing a rag-bag of seemingly incommensurable bits of knowledge-pieces. This promise, it seems to me, makes materialism worth taking seriously.

References

Gibson, J.J. (1966). The Senses Considered as Perceptual Systems. Boston: Houghton Mifflin.

Hilgard, E.R. & Bower, G.H. (1966). Theories of Learning. (3rd ed.). New York: Appleton-Century-Crofts.

Holzkamp, K. (1973). Sinnliche Erkenntnis: Historischer Ursprung und gesellschaftliche Funktion der Wahrnehmung. Frankfurt: Athenaum Verlag.

Holzkamp, K. (1978). Die Uberwindung der wissenschaftlichen Beliebigkeit psychologischer Theorien durch die Kritische Psychologie. In Holzkamp, K. (Ed.), Gesellschaftlichkeit des Individuums. Koln: Pahl-Rugenstein..

Holzkamp, K. (1983). Grundlegung der Psychologie. Frankfurt: Campus.

Holzkamp-Osterkamp, U. (1975). Grundlagen der psychologischen Motivationsforschung, 1. Frankfurt: Campus.

Lewin, K. (1969). Grundzuge der topologischen Psychologie. Bern/Stuttgard/Wien: Huber.

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

Tolman, C.W. (1987). Dialectical Materialism as Psychological Metatheory. In H. Stam, T. Rogers, and K. Gergen (Eds.). The Analysis of Psychological Theories (pp. 211-229). New York: Hemisphere Publ. Co.

Vygotsky, L.S. (1978). Mind in Society. Cambridge, MA: Harvard University Press.

Wertheimer, M. (1972). Fundamental Issues in Psychology. New York: Holt, Rienhart & Winston.

Wilcox, S. & Katz, S. (1984). Can Indirect Realism be demonstrated in the psychological laboratory? Philosophy of the Social Sciences, 14, 149-57.

Zajonc, R.B. (1968). Cognitive theories in social psychology. In G. Lindzey & E. Aronson (Eds.). Handbook of social psychology (Vol. 1, 2nd ed., pp. 320-411). Reading, MA: Addison-wesley.

FN-1 The materialism represented here is dialectical materialism (e.g., Tolman, 1987). Psychologists are most likely to be acquainted with dialectical materialism in the works of Vygotsky (e.g., 1978) and Luria (e.g., 1979).