A Critique of Evolutionary

In this essay I will assess evolutionary psychology (E.P) as a scientific research programme. I will outline the concept of a scientific research programme and the utility of discussing E.P in this manner. I will also give an outline of the origins of E.P, discussing on the one hand its fundamental principles and their auxiliary hypotheses and on the other the earliest work done in this spirit. I will discuss certain ambiguities and weaknesses present in the programme, those concerning reverse engineering, modularity, adaptation and the Environment of Evolutionary Adaptedness (E.E.A). I will complete the discussion of Archer’s claim by assessing criticisms of E.P and their motivation, and providing some warning comments about progress and degeneration.

E.P as a Scientific Research Programme

Archer (2001) presents the hierarchical theoretic structure of E.P. as outlined by Buss (1990). His basic claim is that the fundamental principles of E.P are not being tested in empirical research; the role of these fundamental principles (the first and second level) is to generate ‘novel hypotheses’ (the second and third level) which are then empirically tested. To understand fully what is being claimed one needs at least a brief gloss on work done in Philosophy of Science in the latter part of the twentieth century.

In his Conjectures and Refutations Karl Popper made what many consider to be a key move in the justification of scientific method. His suggestion was that science follows a ‘hypothetico-deductive’ system. Rather than being an attempt to prove theories inductively from set of relevant observations, a more stable characterisation (which would also account for the success of scientific theories) would be as follows: hypotheses are deduced from theories; these hypotheses generate predictions which are submitted to rigorous tests; any observations which violate these predictions refute the theory via a retransmission of falsity under the logical law of modus tollens. Theories are conjectures which stand only until falsified; the wider the range of predictions and the more rigorous the test, the better the theory. (Popper, 1972)

One problem noted by one of Popper’s students, David Miller (in his Critical Rationalism), is the lack of any theory of verisimilitude. Given that there is no way of identifying a specific set of all the hypotheses a theory may generate in potential there is no way of knowing which un-refuted theory is objectively better, and thus closer to the truth, than its contemporaries. (Miller, 1994)

An alternative view presented by Kuhn in his Structure of Scientific Revolutions is that ‘normal science’ consists of ‘paradigms’, which are themselves dependent upon historical factors such as their socio-political context. Unfortunately Kuhn’s use of the ‘paradigm’ is fairly ambiguous and whilst it is useful for talking about dramatic changes in accepted scientific thought e.g. the ‘Williams revolution’ in evolutionary biology, it is unclear whether it can be specifically applied to E.P. Kuhn recognising this ambiguity changed the term to 'disciplinary matrix' or 'matrices' (Kuhn, 1970, 1977). However, some e.g. Buller & Hardcastle (2001) read ‘paradigm’ as a shared and unquestioned framework of theory, methodology, and exemplars (specific explanations that serve as models for further scientific research) and believe E.P may be represented in this way. I am unconvinced that they have wrestled with the ambiguities and will argue that a Lakatosian ‘scientific research programme’ (S.R.P) is closer to Archer’s representation of E.P as a “research agenda” and can better account for the diversity within the field (Archer 2001, p414).

Imre Lakatos, also a student and colleague of Popper (but sympathetic the idea that history is a key factor in considering the nature of scientific research), proposed the idea that an S.R.P is characterised by a continuity which ties together its members: the metaphysical ‘hard-core’. The hard-core is metaphysical in the Popperian sense that it is not subject to refutation by empirical testing. Within the hard-core are the central theoretic principles from which auxiliary hypotheses are deduced. In E.P. the hard-core, at the very least, certainly comprises the principle that species evolve through genetic inheritance, mutation and differential selection: Darwinian evolution. It would also include certain 'Neo-Darwinian' ethological principles such as Trivers’ theory of reciprocal altruism, Hamilton’s theory of kin selection and Maynard Smith’s application of ‘game theory’ to evolution. The inclusion of these is indicated by the words immediately following the title quote - “natural selection has made us this way” (Archer 2001, p414). Note that these theories are empirically testable, however what is important in the context of E.P as an S.R.P is that they are tested elsewhere (they are the auxiliary hypotheses of other S.R.P’s); the experiments conducted under the banner of E.P serve to test the auxiliary hypotheses which form a protective belt around the hard-core (Archer 2001, p415). This, I hope, will add some depth to Archer's comment that it is a “fundamental misconception” about E.P research to believe “that in each case 'the evolutionary view' is being tested” (Archer 2001, p416).

The auxiliary hypotheses are interpretations of the hard-core with specified empirically observable predictions from which experiments may be constructed. If the predictions are falsified by experimental observations the relevant auxiliary hypotheses will be discarded as false interpretations of the core. A degenerative S.R.P is one which is continually forced to discard auxiliary hypotheses, thus reducing the range of empirical prediction and continually reinterpreting the hard-core. A progressive S.R.P is one which continually increases the range of empirical prediction through the survival of auxiliary hypotheses and the generation of novel hypotheses from the former.

The 'Santa Barbara School' and the ...

This is a preview of the whole essay

The 'Santa Barbara School' and the early auxiliary hypotheses

In the light of this discussion we can more clearly discern what might be characteristic features of E.P and see whether it does what Archer proclaims. Here I will examine the claim that the early writings and mission statements of Leda Cosmides and John Tooby provides the “defining features of the field” (Laland & Brown (2002)). Together with Donald Symons, who brought them to the University of Santa Barbara, this group of researchers will be referred to as the ‘Santa Barbara School’ for the sake of convenience.

Cosmides & Tooby's views on the nature of E.P seem to be in the same spirit as Archer's rather general statement (serving as title for this paper), as shown the following quote:

“Evolutionary psychology is an approach to psychology, in which knowledge and principles from evolutionary biology are put to use in research on the structure of the human mind.” (Cosmides & Tooby, 1997)

One noteworthy disparity between the Santa Barbara School and Archer (and perhaps also Buss, see Laland & Brown (2002) p158) seems to be the extent to which they emphasise a 'computational theory of mind'. Archer uses the word 'mechanism' freely but without committing himself to any clear statement of his stance upon the functionality of the mind. Cosmides and Tooby on the other hand clearly emphasize that they take mind to be an information processing machine consisting of circuits operating programs which manifest adaptations:
“When evolutionary psychologists refer to "the mind", they mean the set of information-processing devices, embodied in the human brain...the programs comprising the human mind were designed by natural selection to solve the adaptive problems faced by our hunter-gatherer ancestors.” (Cosmides 2001)
Thus we should exhibit evolutionary adaptations in our behaviour and cognitive abilities.
An apparent thesis of the Santa Barbara School is what might be called hyper-modularity. Fodor originally introduced the idea of modularity to the Philosophy of Mind in arguing that the areas of mind involved in perception have certain pointed and discrete functional properties and are thus domain-specific whilst the rest of the mind may be a domain general problem solving machine (Fodor, 1983). A crucial feature of these modules is informational encapsulation; the module is only privy to the very specific information related to its function and does not access information processed in other parts of the brain. The Santa Barbara School adopt the idea of domain-specific modules whilst seemingly rejecting the idea that the mind may be domain-general. It is useful to understand this as a reaction to what they referred to as the Standard Social Science Model (SSSM), which retained what they considered a dogmatic view (inherited from the 17th and 18th century empiricists) of the mind as 'white paper' to be written upon by social cultural and physical experience and so must be domain general (as it begins with no information regarding the problems it may face). A quote from Symon illustrates their view:
“It is no more probable that some sort of general-purpose brain/mind mechanism could solve all the behavioural problems an organism faces (find food, choose a mate, select a habitat, etc.) than it is that some sort of general-purpose organ could perform all physiological functions (pump blood, digest food, nourish an embryo, etc.)” (Symons, 1997 as cited in Buller & Hardcastle, 2000)

To understand what problems these domain-specific modules have adapted to solve one must be familiar with the environment in which they evolved. Evolution takes a long time, only comprehensible in terms of thousands of generations. Cosmides and Tooby suggest, following Bowlby, that we have not significantly evolved since the time when our hunter-gather ancestors roamed the African Savannah in roughly the Pleistocene period; this would be the Environment of Evolutionary Adaptedness (E.E.A) and only in reference to this, rather than our post-agricultural/technological-revolution modern day world, may we understand present day adaptations. Thus we have a stone-age mind in a space age world, but “the past explains the present” , and we must 'reverse-engineer' to understand a particular trait or disposition (Cosmides & Tooby, 1990; Pinker, 1997 - cited in Archer, 2001).

Cosmides' (1989) doctoral dissertation provides an illustration of the methods proposed above. Following the principle of reciprocal altruism and the idea that its effective function requires an ability to detect 'cheats' to the rules of reciprocation (i.e. in solution to the 'free-rider' problem), Cosmides designed a series of elegant experiments using the Wason selection task to test the hypothesis that humans possess a 'cheater detection module' adapted to solve this problem. The data showed that even where subjects had training in formal logic 75% failed the task when in an abstract form, whereas 75% got it right when it was presented in a social form in which subjects had to detect cheats to a social rule (both tasks had the same logical form). This indicated that the subjects possessed a cheater detection module that was not party to the subjects' knowledge of logic in a separate stream of the brain.

In sum, we have a number of auxiliary hypotheses detailed above that have been interpreted from the hard-core which we established earlier. These are that
the mind functions as a computer with discrete domain-specific modules,
we will exhibit evolved adaptations in our behaviour and cognition
our adaptations were selected for in an environment vastly different from that which most of us currently live in
to understand these we must focus upon the E.E.A
I shall give brief criticisms of each of these.

Ambiguities, weaknesses, criticisms of E.P, degeneration and progress.

I shall not address the concept of a computational theory in general but rather the specific modularity thesis. There are two initial points: firstly a modular theory of mind does not necessitate a mind composed of entirely domain specific modules – as Fodor (1983) suggests, there may be informationally encapsulated modules which do not interact with the rest of a domain general mind; secondly whilst Cosmides' early conclusions at least require some domain-specific/informationally encapsulated modules there is little to suggest that this precludes them being a part of a domain-general mind. In fact there seems to be evidence that some psychological traits are domain-general, and when pushed upon the matter Cosmides and Tooby agree, pointing to their own recent work which indicates just this, making it somewhat ambiguous what their stance is.

Work done by Ramachandran and others has shown that hand-gesturing in babies comes before the use of language, gesticulation experiences are reported by phantom limb patients when communicating, blind persons use the same gestures in the same manner as sighted persons, sighted persons still gesticulate when talking to blind persons and subjects recall details of a story better when allowed to gesticulate. All of which indicates that gesticulation may not be a culturally learned behavioural trait but rather a tool for the speaker's language use. This is in turn would indicate that there is a dynamic interaction between brain areas involved in language and those involved in motor control; this is a direct criticism of Pinker's arguments that there are discrete language modules and of hyper-modularity in general (Badcock, 2000, pp23 &25).
Adaptations are a point of controversy within Darwinian theory. Darwin's own comments seem to suggest that an adaptation selected to solve a specific problem may be an illusion or very heavy interpretation of what is really just an amalgam of surviving parts (see quotation in Badcock pp 22 & 23). Also Williams warns against regarding something as an adaptation prematurely, stating that the criterion for regarding a feature of an organism as adaptive (and if so what it was originally adaptive to) are far too unclear, as such he warns that we should only regard something as adaptation when we have exhausted all other possibilities. There seems to be a problem here for the reverse engineering thesis that presumes we can glean semantic properties (the 'why') rather than just causal properties (the 'how') from our analysis of whether a particular behaviour or ability is adaptive (Badcock, 2000, pp18 &19).

Having anything interesting to say about the E.E.A may be precluded by the difficulty of being able to say with confidence that something is an adaptation rather than an epiphenomenon of other unnoticed adaptations. This aside it is not particularly clear what exactly the E.E.A is; denoting a general time-period and place does not get us far in having a working model. Cosmides and Tooby (1990) have attempted to provide something more definitive:

“...it is a statistical composite of the adaptation relevant properties of the ancestral environments encountered by members of ancestral populations, weighted by their frequency and their fitness consequences.”(as cited in Laland & Brown 2000 pp179)

Yet this does not denote a particular ancestral population or even a particular time period which would mark key stages in our evolution of certain adaptations (Laland & Brown, 2000) The idea of an E.E.A and focus on the general hunter-gatherer concept also gives a misleadingly static view of evolution in the surrounding time periods. In addition, to say that we are maladapted to our current environment is akin to saying that the hunter gatherers, being originally vegetarians, were maladapted to theirs. It seems a moot point to argue focus upon one environment rather than another.

A few things must be noted in conclusion. Firstly very few of the criticisms above are decisive or even arguments. To do this effectively one would have to level each against all the diverse work done in the field. But the line of attack is clear; if the auxiliary hypotheses cannot achieve the generality they purport or worse still their predictions are disconfirmed then they will have to be reinterpreted in order that the programme may survive. If reinterpretation persists then the field should be considered degenerative. But if it can field all the criticisms and make new predictions, progress must be noted.

One thing that is omitted above is a proper treatment of the diversity of research done in this field, as I have tried to look at the bigger picture. The picture is overall a highly deterministic one and rejecting this aspect is a common motivation for some of the more valid criticisms. Rose & Rose point out that there is no account taken of plasticity and the dynamic complex nature of 'living systems' (Archer, 2001 pp428 & 429). And as a final note, volitional action is still considered one of the things that makes us human, our behaviour seems under-determined by genetic mechanisms, something may be being left out.

Bibliography:

General note to reader: Where philosophical literature has been cited, exact page references have been omitted intentionally, in the belief that all references made are simply interpretations of their work. If my interpretation is deemed correct, the insight belongs to the referee, if deemed incorrect then I am simply mistaken.

Archer, J. et al. (2001) - Evolving Theories of Behaviour, in ‘The Psychologist’, 14(8): 414 - 431.

Badcock, C.R. (2000) - Evolutionary Psychology: a critical introduction; Blackwell, Polity Press

Buss, D.E. (1990) – Evolutionary Social Psychology: prospects and pitfalls; in ‘Motivation and Emotion’, 14: 265-286.

Buller & Hardcastle (2000) - Evolutionary Psychology, Meet Developmental Neurobiology: Against Promiscuous Modularity; in ‘Brain and Mind’ 1: 307–325: Netherlands, Kluwer Academic Publishers.

Cosmides & Tooby (1997) - Evolutionary Psychology: a primer: (http://www.psych.ucsb.edu/research/cep/primer.html) Copyright John Tooby and Leda Cosmides, Updated January 13, 1997

Cosmides, L. (1989) - The Logic of Social Exchange: has natural selection shaped how humans reason? Studies with the Wason selection task; in 'Cognition' 31: 187-276
(2001) - Interview with Alvaro Fischer and Roberto Araya for the Chilean newspaper 'El Mercurio' (http://www.psych.ucsb.edu/research/cep/ledainterview.htm); part of a project entitled "New Paradigms at the Beginning of the Third Millenium".

Fodor, J. (1983) - The Modularity of Mind: an essay on faculty psychology Cambridge, Mass. ; London : MIT Press

Hume, D (1975) - Enquiries Concerning Human Understanding and Concerning the Principles of Morals; L.A Selby-Bigge: Third Edition, Oxford, Clarendon Press.

Kuhn, T. S (1970) - The Structure of Scientific Revolutions; Second edition: Chicago, University of Chicago Press.
(1977) - The Essential Tension. Selected Studies in Scientific Tradition and Change. Chicago: University of Chicago Press
.
Lakatos, I. (1976) - Proofs and refutations: the logic of mathematical discovery; edited by John Worrall and Elie Zahar. Cambridge: Cambridge University Press.

Laland & Brown (2002) - Sense and Nonsense: evolutionary perspectives on human behaviour; Clarendon, Oxford University Press.

Miller, D. (1994) – Critical Rationalism: a restatement and defence; Chicago: Open court

Popper, K. (1972) - Conjectures and Refutations; Fourth Edition: Routledge & Kegan Paul Limited