['Animal Thought' © Stephen Walker 1983]
pdf file of this chapter
2 Darwinian continuity
The problem of the relation between human and animal minds clearly did not begin with the advent of the Darwinian theory of evolution, but all attempted answers to the problem after Darwin must be coloured by the existence of a theory which presupposes a common origin for human and animal existence. In one sense, Darwinian theory did remarkably little to alter the nature of the problem, since the concept of a scale of animal intelligence had been introduced by Aristotle. But Darwin’s account of how human peculiarities could be reconciled with his general theory of species development and descent, and the boost which this theory gave to the thesis of continuity between the human and animal worlds, can hardly be ignored. In many ways, Darwinian theory was the culmination of two centuries of investigation of comparative anatomy and physiology, which began with Descartes; the French tradition of natural history, represented by Buffon (1701—88), Lamarck (1744—1829) and Cuvier (1769—1832) was a powerful influence, not so much on Darwin himself, but on his immediate precursors and contemporaries. But one of the reasons for the central position of Darwin in the growth of modern biology was the intense and public controversy surrounding evolutionary theory in Victorian England. And perhaps not so much the controversy, but the fact that Darwin, with the help of Huxley, won.
Buffon, in Hume’s day, a century earlier, had produced a ten-volume Natural History, in which, Darwin admitted, ‘whole pages are laughably like mine’ (Himmelfarb, 1959, p. 143). Buffon held that species progressed and degenerated very gradually according to ecological pressures, and that thus, by descent, new species developed from old ones. In particular, all the component parts of the orangutang, including the brain and the vocal organs, closely resemble those
of man, and man and ape, like horse and ass, must therefore share a common ancestor. But Buffon was eventually obliged by the religious authorities to make a formal denial that there was any evidence in his work that would contradict the biblical story of creation. Darwin and Huxley, however, stood firm against theological pressures (although Darwin made many concessions to ill-advised scientific criticism), and succeeded, where previous protagonists had failed, in getting the evolutionary account accepted, if still sometimes uneasily, by most subsequent professional and intellectual opinion.
The factors which led to the eventual triumph of Darwin’s evidence over traditional prejudices are many, and include most clearly the accumulation of fossil findings and the early Victorian concern with theories of geology, as well as the wild and undisciplined speculations of Herbert Spencer. As an influence on the popular imagination, pride of place should probably go to Vestiges of Creation, an unequivocal and provocative pot-boiler which promoted a theory of progress in organic life under natural laws of development, with very small and modest changes allowing one species to develop into another. This was published in 1844, went through four editions in seven months, and became a cause célebre, parodied in Disraeli’s Tancred as Revelations of Chaos’. Its anonymous author was not one of the widely touted candidates such as Prince Albert or Thackeray, but Robert Chambers, a shrewd Edinburgh publisher whose firm still puts out the worthy Chambers Dictionary. Darwin later paid tribute to the excellent service performed by the Vestiges in calling attention to the subject and preparing the ground, in spite of its unsound zoology. And it was Chambers who persuaded Huxley to stay on at the British Association meeting at Oxford in 1860 to defend the Darwinian cause against the onslaught expected from Bishop Wilberforce, son of the antislavery Wilberforce and an extremely influential figure. The fact that Huxley was felt to have successfully countered Wilberforce’s attack— by saying that he would rather be descended from an ape on either his father’s or his mother’s side as an alternative to being related to someone who posed such a frivolous question—is held to have been an enormous public relations coup for the Darwinian side (e.g. Himmelfarb, 1959, pp. 236—41).
However, a more enduring reason for the success of Darwinian theory is the evidence in its favour, and the persuasiveness of Darwin’s argument. The Origin of Species, published in 1859, is remarkable for its detail, even though Darwin referred to it as an abstract, presented
prematurely because ill-health might prevent his completing it. Two points are especially relevant to the question of mental abilities. First, it was one of Darwin’s conclusions that his theory would eventually provide a new basis for psychology—’the necessary acquirement of each mental power and capacity by gradation’ (1968, p. 458). Second, a chapter of the Origin is devoted to ‘Instinct’, a term which Darwin used synonymously with ‘mental powers’, ‘mental qualities’, or ‘mental actions’. It is made explicit that these mental qualities are a product of evolution—they arise ‘as small consequences of one general law, leading to the advancement of all organic beings, namely, multiply, vary, let the strongest live and the weakest die’ (1968, p.263).
This last is perhaps the briefest encapsulation of the theory of evolution given by Darwin, and it is in some ways not very representative of the more general argument. The facts which Darwin wished to explain were not enormously different from those known to Aristotle, and certainly not at variance with those collected by Buffon. But Darwin put forward a new and powerful analogy: the process of breeding and change in domesticated species of animals became the model for change in the natural world. And Darwin stressed, for instincts and for anatomical features, an aspect of natural history previously ignored —anomalies and accidents in the adaptations of wild species to their environments.
The analogy between domestic breeding and the natural origins of species is the core of Darwin’s case, and it includes the consideration of instinctive mental tendencies: anomalous adaptations provide the support and ought to be remembered in the context of mental evolution. That particular species of animal are apparently designed for their own style of life is a starting point for all accounts of the origin of species alike, including those which appeal to a creator who designs species as a watchmaker designs a watch, those like Lamarck’s which assume that species design themselves, and Darwin’s theory which appears, as Dewey put it, to offer ‘design on the instalment plan’. Lamarck had supposed that animal forms were slowly changed by their needs, and this was interpreted by both Darwin and Schopenhauer as requiring a succession of acts of will on the part of individual animals (and plants). Schopenhauer, revising The Will in Nature only five years before the publication of Darwin’s Origin, took the apparent ‘universal fitness for ends’ as evidence for a universal process of design, and was familiar with the popular examples of
webbed feet in swimming birds, the teeth and claws of carnivores, and the long tongue of the anteater. But Schopenhauer was impatient with the slow and gradual willings proposed by Lamarck—if there was to have been any willing he would rather assume that a primary animal willed itself into existence in the form of the various species, fully equipped with all necessary organs, which would thus not have to go through aesthetically unsatisfying intermediary stages!
Darwin was equally dissatisfied with Lamarck’s explanation, but sought for natural laws which would explain gradual adaptations to needs. Although unable to specify genetic mechanisms, he said simply that whatever causes gradual change in artificial selection which produces the strains and varieties of domesticated species, this also caused the development of natural species. Deliberate selection of animals to produce for instance, better breeds of sheep, is an extremely ancient art, but Darwin was able to draw on systematic work catalogued by nineteenth century specialists in horticulture and husbandry. A great many examples of selective breeding were thus provided, but many of Darwin’s favourite cases were taken from the less utilitarian science of pigeon-fancying. Pigeon breeding was something he was able to take up himself and he was ‘permitted to join two of the London Pigeon Clubs’. A skilful breeder could claim that ‘he would produce any given feather in three years, but it would take him six years to obtain head and beak’ (1968, p. 90). It was clear that all domestic breeds were descended from the same ancestor, the Indian rock-pigeon (Columba livia). But according to Darwin, the established varieties such as the English carrier, the short-faced tumbler, the pouter and the fan-tail, were ‘so different in form and habit that an ornithologist judging them as wild birds would probably not even put them in the same genus, and as wild birds the domestic varieties would certainly be counted as separate species.
If deliberate selection by the intervention of human agency could bring about such changes in the humble rock- pigeon, in so short a time compared with the new geological scales, why could not a similar process account for gradual changes which result in the origin of new natural species? All that was required was an unseen hand, operating in nature as a substitute for human judgment, as a procedure of ‘natural selection’. Such an unseen hand had already begun to be suspiciously visible in the Malthusian struggle for existence. The processes of reproduction would allow for geometric increases in the populations of all forms of life—only a small proportion of each species
needs to survive in any generation for the species to keep up its numbers. One need look no further for a natural filter; indeed the struggle for survival is much more efficient than any human selection, since there can be a continuous assessment of minute differences in internal structure and the organisation of behaviour. The final piece in the puzzle is the presence of sufficient inherited variation in structure and function to provide something worth filtering for. The Origin has three chapters on variation—there is a good deal of wavering over the possible contributions of Lamarckian sorts of use and disuse and individual habits as sources of gradual change, but it is natural ‘selection acting on inherited variation which is the distinctive Darwinian doctrine.
In the chapter on instinct, it is acknowledged that the first appearance of some peculiar mental habits can only be due to accidents. But a range of inherited individual differences can be seen between and within different breeds of domesticated species, as for instance in retrievers, pointers and sheep-dogs. Pronounced variations in the observed expression of nest-building and migratory instincts occur in wild birds, partly but not wholly dependent on climate and other environmental circumstances. There is generally plenty of diversity in instinctive behaviours in individuals of the same species, and within the same individual there are ‘different instincts at different periods of life, or at different seasons of the year’. Any of these sources could provide a range of variation wide enough that ‘one or the other instinct might be preserved by natural selection’.
Natural selection is blind to all considerations of fitness for function except those of survival and reproduction. The fact that instincts are ‘liable to mistakes’ is used by Darwin, like cases of anomalous bodily ‘structures and vestigial organs, to counter theories of unique creation, and to ‘support the contention that aesthetically unfit and partially adapted specie’s can survive. The American ostrich (the emu) wastes ‘a surprising number of eggs.. . strewed over the plains’; on the same plains of La Plata there is a woodpecker very like the European ‘species which survives in the absence of trees; in nearby uplands there are geese with webbed feet which never go near water—while aquatic coot’s and grebes make do with only membrane-bordered toes, and the water- ouzel, whose body looks like that of any other thrush, subsists by diving into streams, grasping stone’s with its feet, and using its wings under water. Form does not always predict function, and function may belie form.
The Origin thus explained more of the facts than its predecessors, but does not in itself contain much by way of exciting speculations about human ancestry. The Descent of Man, first published in 1871, certainly did. It was immediately followed by The Expression of the Emotions in Man and Animals (1872), which was equally blunt about human evolution; after this Darwin concerned himself almost exclusively with plant life.
It is a measure of Darwin’s genius that he not only cut the Gordian knot of contemporary evolutionary theories, by giving, in natural selection, a simple and elegant account of how and why gradations and radiations of species could result in the observed diversity of organic life, but also, in the concept of sexual selection, he elaborated his theory by discussing factors whose importance was not fully appreciated until the 1960s (see O’Donald, 1980). The term natural selection was used by Darwin to refer to the filtering action of the ‘conditions of life’—the effect of climate, disease, predation, food supply etc. on the health and survival of individuals. But the successful negotiation of these trials, though necessary, is not a sufficient condition for Darwinian survival, since it is the inheritance of adaptations, and not their effects on individual longevity, which is crucial. The reproductive process is at the heart of natural selection, and in all but a handful of animal species this requires sexual interactions. The modern term ‘inclusive fitness’ emphasises the degree to which an individual’s genes are included in the next generation—from one point of view coping with the conditions of life is merely an incidental preliminary to genetic reduplication—but Darwin’s term ‘sexual selection’ is retained to cover aspects of structure and behaviour whose importance depends solely or mainly on their role in sexual reproduction. The simplest examples are the elaborate vocalisations and bodily decorations of male birds. These may be of no use in negotiating the conditions of life, but are ‘essential to individuals if they are to attract females. It goes without saying that special-purpose instincts are as necessary as bodily accoutrements for competition, courtship, mating and parental care. Darwin’s own treatment of sexual selection is exhaustive and complex, and the subject is now part of the domain of mathematical model’s in theoretical genetics.
One aspect of Darwin’s discussion of sexual selection is the bizarre and surprising conclusion that human language first began, not as an aid to abstract reasoning, but as a technique of courtship. He assumes, of course, that speech very soon became a vehicle for thought, and
indeed that the intelligent use of language promoted the large size of the human brain 1901, p. 134), but suggests that the initial emphasis on vocalisation might have arisen from its utility in charming and impressing sexual partners and rivals—’some early progenitor of man, probably first used his voice in producing true musical cadences, that is in singing,... this power would have been especially exerted during the courtship of the sexes’ (1901, p. 133). One may wonder if Darwin was led into strangely unsound speculation by the musical evenings and protracted engagements of middle- class Victorians, but his arguments are based on a characteristically thorough examination of the biological evidence. Sounds are used throughout the animal kingdom for emotional expression and social communication and ‘a strong case can be made out that the vocal organs were primarily used and perfected in relation to the propagation of the species’. The case applies pre-eminently to bird song, but there are many exceedingly vocal species of monkey, and several type’s of gibbon (‘lesser apes’) have a social organisation based on neighbouring nuclear families, and are strong singers. According to Darwin’s information some gibbons sing in semi- tones organised into octaves (but see Tenaza and Marler, 1977; Tenaza and Tilson, 1977).
Some sort of prelinguistic use of the voice in human ancestry can hardly be doubted, although whether at one stage ‘the males or females or both sexes, before acquiring the power of expressing their mutual love in articulate language, endeavoured to charm each other with musical notes and rhythm’ is another matter. It must be admitted that the breaking of the male voice at puberty is unlikely to be related to the intellectual uses of language, and singing and dancing, at least in their less refined forms, seem to have retained some connection with sexual display and attraction.
Darwin’s concern with the emotional and sexual functions of human language, and its biological continuity with animal vocalisation, is perhaps a sufficient indication of his general position on differences between the mental activities of man and beast: although these differences are immense, they developed by imperceptible degrees, by the processes of natural selection, sexual selection, and in some cases as a consequence of civilisation. Higher animals possess attention, memory and association, and a certain amount of reason and imagination. More complex faculties, such as self- consciousness and rational abstraction, could have evolved via the elaboration and combination of simpler ones (1901, p. 128). The scale of the differences
between man and ape should be compared to the difference between apes and lower vertebrates like fish.
Quantitative and qualitative intellectual differentiation
The 1901 edition of the Descent of Man is over 1,000 pages long, and remains an invaluable compendium of evidence and argument concerning the application of the doctrine of evolution to the human species. Moral, economic and linguistic development as factors over and above biological evolution are given their due. The question of animal mental abilities is covered mainly in the two chapters headed ‘Comparison of the mental powers of man and the lower animals’, and a concise and convenient summary of these is given at the end of Chapter 4, Chapter 21 being an overall summary of the whole work. The central ideas are that ‘the difference in mind between man and the higher animals, great as it is, certainly is one of degree and not of kind’ (p. 193) and the dimension of the quantitative difference is the learning of associations: ‘The lower animals differ from man solely in his almost infinitely larger power of associating together the most diversified sounds and ideas’ (p. 131).
A large part of the human intellect is thus attributed to the ‘half-art and half-instinct’ of language, since the first and simplest use of this ‘wonderful engine which affixes signs to all sorts of objects and qualities’ would have acted as a spur to brain expansion, and made available otherwise impossible trains of thought (p. 932). However, ‘the extremely complex and regular construction of many barbarous languages, is no proof that they owe their origin to a special act of creation’ (p. 140) since there is a curious and interesting parallel between the gradual changes and blendings which may alter languages, and the evolution of species.
Anecdotal evidence of animal thought—Darwin and Romanes (1848—94)
Darwin’s theorising about human evolution seems sophisticated and modern. A. R. Wallace, given credit for anticipating the idea of natural selection, held, by comparison, rather old-fashioned views. His heresy, as he called it, was that natural selection could not account for the ‘size of the human brain, since savage human races have human-
sized brains, but because of their primitive conditions of life they should have only ape-sized brains if the usual evolutionary forces were responsible. Darwin’s knowledge of the social organisation, linguistic complexity and skilled craftsmanship observed in what were then called the lowest savages prevented him from making this sort of blunder.
However, the type of evidence on which Darwin based his theorising about animal behaviour has just as much of an old-fashioned flavour as the most jingoistic Victorian anthropology. He relied on travellers’ tales and hunters’ stories, culled from contemporary sources such as The Naturalist in .Nicaragua, and on letters written to him by interested readers about their pets. For instance, the existence of reasoning in dogs is documented by the report of a Mr Colquhoun, in The Moor and the Loch, that his retriever, on an occasion when he had winged two wild ducks which fell on the far side of a stream, first tried to bring both back at once, but then: ‘though never before known to ruffle a feather, deliberately killed one, brought over the other, and returned for the dead bird’ (1901, p. 118). And the existence of a capacity for connecting specific sounds with definite ideas, in animals other than man, is assumed on the basis of the beliefs held by the owners of pet parrots. In particular, Admiral Sir B.J. Sullivan remembered that a parrot kept by his father invariably addressed members of the household by their names, and once said ‘you naughty Polly’ to another bird which had got out of its cage and was eating apples on the kitchen table (1901, p. 130).
The assembling of anecdotal reports in this way now seems quaint, if not slightly ridiculous, since we are used to elaborate statistical analysis of carefully controlled experiments and field observations, from which the subjective impressions of the observer ought to have been eliminated. Darwin’s attitude to anecdotal reports of animal behaviour, and to the behaviour of his own dogs and cats, is one of amiable credulity. Although he is concerned that his correspondents are ‘careful observers’, he does not believe in doubting ostensibly reliable reports, and indeed felt that too much scepticism of this kind would hold back the progress of science. His acceptance of rather weak evidence in favour of human-like mental attributes in animals seems to be at least partly due to a desire to support his contention that human intellectual abilities do not provide an exception to the theory of evolution. When animal behaviour seemed to provide evidence of abilities in excess of those to be expected on the basis of natural
selection, no one was better than Darwin at interpreting complexities in terms of simpler mechanisms—the analysis of the method by which bees construct architecturally elaborate honey combs by the action of simple and understandable behaviour patterns, in the ‘instinct’ chapter of the Origin, is a superb essay of this kind. Therefore the failure to consider whether the behaviour of domestic dogs and parrots might not be explained in terms of mental abilities of a very much simpler type than those involved in human thought can be counted as a flaw in the case made out in The Descent of Man. Certainly the conclusions reached by Darwin, that dogs exhibit a sense of humour, a sense of wonder, and beliefs in supernatural agencies, are difficult to defend on the basis of the anecdotal evidence that he quotes. .
This does not mean that Darwin was necessarily wrong. Laboratory testing of animal behaviour was virtually unknown, and methodologically Darwin may be said to have made the best use he could of the only evidence available to him. Although we cannot possibly accept the anecdotal evidence as satisfactory indications of the nature of mammalian cognition and feeling, this does not mean that Darwin’s conclusions can be rejected out of hand, merely that better tests are required. Undergraduates ever since Lloyd Morgan (1894) and Thorndike (1898) have been warned against the dangers of overestimating the ‘mental powers of the higher animals’, but Darwin made the point that it is possible to err on the side of underestimating these powers, if we too quickly assume that simple instinctive or associative mechanisms must provide the explanation for all nonhuman behaviour. The statistical unreliability of anecdotal reports (was the ‘intelligent’ action a rare accident?) and the fact that they often depend on the subjective judgment of the observer (what does it mean to report an animal’s action as ‘deliberate’?) leave the question of the explanation of animal behaviour open, not closed. I shall go into this further shortly, in discussing Lloyd Morgan.
The apotheosis of the Darwinian anecdotal method was reached in Romanes’s Animal Intelligence, first published in 1881. Romanes was Zoological Secretary of the Linnean Society, and a devoted disciple and friend of Darwin’s; and he was entrusted with a much expanded version of the ‘Instinct’ chapter of the Origin (included in Romanes’s Mental Evolution in Animals, 1883) and all of Darwin’s notes and files on animal behaviour. Animal Intelligence was based on this material and written with Darwin’s collaboration. The introduction includes a reasoned assessment of the virtues and pitfalls of the anecdotal method,
and expresses a resolve to give more weight to the reports of reliable observers, and facts which receive numerous confirmations. But it does not seem as though Romanes’s enthusiasm was often assailed by doubts during the compilation of the substance of the book. It would be churlish, indeed, to be suspicious of the abilities of the grey parrot kept by Venn, the Cambridge logician (p. 268), or even perhaps of those of Buffon’s parrot which, having been trained to offer its claw in response to a human request, is said to have acquired the habit of saying ‘Give me your claw’ to itself, and then complying with the request by placing one of its claws in its own beak. But the fact that stories of blind rats being led about by their seeing companions were very numerous was not, with the best will in the world, a very good reason for asserting that rats are highly intelligent (p. 360).
Therefore, although Romanes is given the credit for founding comparative psychology (Boring, 1950), his gullibility in Animal Intelligence tended to obscure the Darwinian case, and the occasional sensible things Romanes said in his later books, Mental Evolution in Animals (1883) and Mental Evolution in Man (1888). These works share a remarkable frontispiece which presents a curious and elaborate scale, divided into 50 levels, in which the ‘Psychogenesis of Man’, from the spermatozoa at level one to a 15-month-old child at level 28, is correlated with ‘Products of emotional development’ and ‘Products of intellectual development’ in various animal taxa. Consciousness appears to develop very gradually, with chordates at the same level as the human newborn, and insect larvae and annelid worms matching a 2-week-old baby. Romanes suggested that one might get some idea of the evolutionary progression of states of consciousness by experiencing them backwards as one loses awareness gradually under anaesthetic. Reason is bracketed with the recognition of persons, and association by similarity, and is identified with the higher Crustacea (lobsters) and a child of 14 weeks. Communication of ideas is seen in the social insects (ants and bees) ; understanding of words and dreaming both appear in birds (parrots having been reported to talk in their sleep); and the distinctiveness of mammals starts at the understanding of mechanisms (cows being particularly adept at unlocking gates). The highest level of intellectual development is rather oddly ‘indefinite morality’, observed in anthropoid apes and dogs. Romanes says in Mental Evolution in Animals that he sees this as the first approximation to a conscience, and that he will return to the topic in later work, but there is not very much more on it in Mental Evolution in Man. In that book, the emphasis has
shifted to the differences between man and animals, rather than the similarities. The similarities are not forgotten, since it is stressed that ii terms of emotions, instincts, volition and reason, the differences are minor. A stand is taken on abstraction, however, and Romanes quote long passages of Locke. According to the frontispiece diagram abstraction begins in mammals, but Romanes adopts the Lockean view that there are things called abstract concepts which humans have but even the best of animals do not. Nothing is added to Locke in this, but Romanes goes on to give a more extensive and searching examination of the character of human language.
It is first stressed that animals possess the germ of language in their faculty of making signs. Typical ‘gesture-signs’ are seen in the pointing of gun-dogs (and pigs can be trained to do the same thing). These are intentional acts of communication, inasmuch as the dogs expect their behaviour to influence human observers, and the natural behaviours may be artificially extended by training, as was apparently the case in Sir John Lubbock’s experiments on ‘teaching animals to converse’. Lubbock (1884) wrote words such as ‘bone’, ‘water’, and ‘pet me’ on cards and believed that his dogs became able to select cards according to their wishes. Dogs have difficulty in mimicking human vocalisation, but Romanes suggested that this is a matter of limitations of vocal organ anatomy, rather than psychological deficiencies. (Alexander Graham Bell wrote to Romanes that he had succeeded in training an English terrier to say ‘How are you grandmamma’ with great distinctness, but, in a rare outbreak of caution, Romanes refused to believe him.) Although the ability to make ‘gesture-signs’, and other aspects of animal intelligence, are taken as support for the theory that human mental abilities evolved gradually, Romanes is quite clear that human speech differs radically from animal sign-making. For one thing, ‘gesture language has no grammar’ (1888, p. 114). In the postChomskyan era, this is a familiar point. Romanes’s discussion of the abstractions underlying grammar may now seem rather primitive, but he should be credited with the attempt to pin down exactly what it is about human speech which is significantly different from animal gestures. In Romanes’s terms, speech is distinctive because it allows predication, and the making of propositions (i888, p. 163). The unique ability of man ‘consists in his being able to mean a proposition’ (p. 164, original italics).
In other words, the essential feature of human speech is the sentence, or ideas expressed as relations between words, most simply as connections between subjects, objects and verbs. In all of Romanes’s
anecdotes, there were none which suggested that dogs, apes, or parrots communicated in this way, although he quotes cases of dogs appearing to understand sentences such as ‘the sheep are in the potatoes’. Placing the line between man and beast at grammar would therefore have been supported by the evidence at hand, but Romanes followed Locke in going further back, to something in the naming of concepts, and added on, gratuitously in view of what he had written under Darwin’s supervision, self-consciousness and reflection.
Huxley’s materialism: mind and brain in ape and man
‘I wish to God there were more automata in the world like you’, Darwin wrote to T. H. Huxley. This was in March 1882 when Darwin had only a month to live, having suffered the final blow in a lifetime of poor health in the form of a heart- attack when he was ringing the doorbell of Romanes’s house near the Zoo. Darwin’s quip was a reference to Huxley’s latest irreverence, the extension of Cartesian automatism to man, given initially in the paper ‘On the hypothesis that animals are automata and its history’ (Method and Results, 1893). Huxley (I 825—95) was a brilliant physiologist and anatomist, elected to the Royal Society before he was twenty- six for his work with the dissecting scalpel and microscope. He was important as an essayist, educator and public figure, and a formidable champion of the Darwinian cause. He invented the term agnosticism, which enabled high-minded Victorian scientists to sidestep religion; he used Pasteur’s experiments to argue for the physical basis of life, and he put forward the doctrine of epiphenomenalism, in which mental experiences were dismissed as by-products of the physical operations of the brain. Human subjective experience was thus kept firmly within the confines of scientifically investigatable material reality. Conscious experience is merely the ‘smell above the factory’ — the real workings of the mind are the activities of the brain. Huxley was therefore about as much of a materialist, reductionist and determinist as it is possible to be, but somehow managed to succeed in always seeming to be more morally serious than his opponents. In 1878, Huxley wrote (in six weeks) a 300-page book on Hume, in which most of Hume’s scepticism was retained, but the subjective empiricism updated so that sense impressions are interpreted as states of the brain. (A rather similar version of Hume is given by Ayer, 1980.)
Huxley took on the job of explicitly bringing the human species into
the Darwinian theory of evolution before Darwin did this himself, and dealt in greater detail than Darwin did with man’s position within the order Primates (a Linnean term which Huxley preferred to Darwin’s ‘Quadrumana’). His first essays on these topics were in Evidence of Man’s Place in Nature (1863), and there is a short but forceful supplement by Huxley in Darwin’s Descent of Man entitled ‘Note on the resemblances and differences in the structure and development of the brain in man and apes’.
A good deal of Huxley’s time was spent on refuting the assertion of Sir Richard Owen (the ranking comparative anatomist of the time, who fought a rearguard action against human evolution) that the human brain contains bits not to be found in those of apes (‘the third lobe, the posterior horn of the lateral ventricle, and the hippocampus minor’). All these structures were thought to be found in the occipital (posterior) lobes of only the human cerebral hemispheres, which cover the cerebellum. In small mammals, such as the rat, the cerebral hemispheres of the brain do not cover the cerebellum. But the occipital lobes of the chimpanzee and the gorilla do cover the cerebellum, and are anatomically very similar to the corresponding parts of the human brain, as Huxley easily demonstrated. Ironically, the occipital lobes of the rhesus monkey brain, which, like the rear of the hemispheres in all mammals, receive projections from the visual pathways, are now taken as very useful approximations to the posterior lobes of man—it is the anterior (frontal) lobes, at the other end of the brain, that are now most often said to be the repository of especially human intellectual qualities. Huxley’s conclusion was that there are no anatomically identifiable components of the human brain that are missing from the brain of other large primates:.
So far as cerebral structure goes therefore, it is clear that man differs less from the Chimpanzee or the Orang, than these do even from the monkeys, and that the difference between the brains of the Chimpanzee and of Man is almost insignificant, when compared with that between the Chimpanzee brain and that of a Lemur. (Darwin, 1901, p. 312)
Even the greater size of the human brain, though it ‘doubtless will one day help to furnish an explanation of the great gulf which intervenes between the lowest man and the highest ape in intellectual power’, was not to be regarded as crucial, since ‘the difference in
weight of brain between the highest and lowest men is far greater, both relatively and absolutely, than that between the lowest man and the highest ape’. Huxley went into other obvious possibilities. The gyri and sulci of the human cerebral hemispheres tend to be more complicated than those of the chimpanzee, but ‘Every principal gyrus and sulcus of a chimpanzee’s brain is clearly represented in that of a man’. Human gyri and sulci seem to be asymmetrical when the left hemisphere is compared to the right, but ‘the degree of asymmetry of the convolutions of the two sides in the human brain is subject to much individual variation’ and ‘in some individuals of the chimpanzee, their complexity and asymmetry become notable’ (Darwin, 1901, p. 312). And therefore:
Regarded systematically the cerebral differences, of man and apes, are not of more than generic value—his Family distinction resting chiefly on his dentition, his pelvis and his lower limbs. (Huxley, 1906, p. 96)
If the mind is construed as brain activity, and the human and chimpanzee brains are so similar, whence arises the intellectual gulf? Huxley curtly gives the language argument in a footnote:
A man born dumb, notwithstanding his great cerebral mass and his inheritance of strong intellectual instincts, would be capable of few higher intellectual manifestations than an Orang or Chimpanzee, if he were confined to the society of dumb associates. (Huxley, 1906, p. 26)
Marx (1818—83) and Engels (1820—95) as post-Darwinians
Darwin was much more interested in plants and beetles than in human societies, and the brief section on ‘Natural Selection as affecting Civilized Nations’ in the Descent is largely defensive, with the aim of rebutting suggestions such as that of Galton, to the effect that, in civilised societies, the poor and reckless members outbreed the provident and virtuous, thus preventing any improvements by natural selection. Darwin countered that a high rate of mortality among the poor and reckless might do something to check ‘this downward tendency’ and in general seems ready to envisage natural selection operating in civilised nations in a rather ruthless way. He is against
intentionally neglecting the weak and helpless, but hopes that they will not marry so freely as the sound, since ‘excepting in the case of man himself, hardly anyone is so ignorant as to allow his worst animals to breed’ (190 1, p. 206). Because of our social and sympathetic instincts, we must tolerate ‘the undoubtedly bad effects of the weak surviving and propagating their kind’, although ‘as man suffers from the same physical evils as the lower animals, he has no right to expect an immunity from the evils consequent on the struggle for existence’ (1901, p. 206 and p. 2l9). Darwin was not quite the kindly old liberal he is sometimes portrayed as. But he had very little to do with social Darwinism in any of its forms, and turned down the chance of having the English edition of Das Kapital dedicated to him.
This is just as well for our purposes, since the Marxist position on the relation between human and animal mental life is not an obvious extension of the writings of Darwin, still less of those of Romanes or Huxley. That Darwin influenced early Marxist thought, but also that Engels, at least, felt free to make extravagant alterations to the biologically based accounts of mental evolution, can be illustrated by Engels’s paper ‘The Part Played by Labour in the Transition from Ape to Man’ (written about 1875; see Marx and Engels, 1968). In this, Darwin’s name is used, but little of his outlook. ‘Labour’, first as the use of flint tools, and then as the social organisation entailed by tool-using, necessitated, according to Engels, the creation of language and, with language, instigated the growth in the size of the brain in ancestral human species. And it is labour and society which produce human consciousness. Consciousness as we now experience it is therefore a product of history, rather than of biological evolution. In a phrase of Marx’s, ‘Man confirms his real social life in his species-consciousness and in his thought he merely repeats his real existence’ (1971, p. 151). The reality here is political, rather than physical. This is important because it gives rise to a split of almost Cartesian proportions between the mental qualities of ape and those of man. If the human mind consists of internal reflections of exterior economic conditions, perceived in the light of a sense of history, then any continuity with the mental activities of animals was lost with the division of labour. In part, this is a restatement of the uniqueness of human self- consciousness and reflection, but clearly there is included a distinct hypothesis that these depend upon a sequence of social and economic events, and a considerable degree of confusion results from the use of the term ‘consciousness’ to imply an all-embracing Hegelian absolute,
as well as to denote a specific result of the place of individuals in human societies.
It would be possible to agree that the acceptance of current social norms, or a knowledge of history, or an ideological commitment, might influence the way people think, without going on to deduce that a dog or a chimpanzee, lacking any of these influences, has no mental life at all. But in practice, accepting Marxist economic determinism tends to result in a theological kind of division between animal and human consciousness, and indeed between the mental states of members of the human species exposed to different sets of social institutions. Distinctions of this kind are found, for instance, in Steven Rose’s The Conscious Brain (1976). Rose suggests that there is a discontinuity between the consciousness of man and other mammals and that this is because of the greater size of the human brain; but political rather than physical variables are lurking in the background. ‘With the emergence of consciousness a qualitative evolutionary leap forward has occurred, making for the critical distinction between humans and other species’ (Rose, 1976, p. I 79). Consciousness is then defined as an interaction between the individual and the social environment, so that ‘as human relationships have become transformed during the evolution of human society, so human consciousness too has been transformed’. When forms of society change, we must expect consciousness to change also, and therefore modern human consciousness is a very different thing from that which occurred in earlier members of the same species, living under different social conditions (Rose, 1976, p. 179). Rose makes it clear that he is using the term consciousness in the Marxist sense, as something which occurs in a society rather than in a brain, and which depends on such factors as the mode of production of consumer goods. As he points out, although this encompasses the dependence of thought on culture, it gives rise to a number of paradoxes. In particular, the pre- emptive social definition of consciousness begs the questions of what sort of mental processes might occur in isolation from human social forces, whether in isolated human beings, or in groups of individuals of other species. There may be an important sense in which ‘a Robinson Crusoe figure, would scarcely be conscious at all’ (Rose, 1976, p. 34), but this does not mean that such an individual might not sleep and dream, and wake and wonder, even if the contents of his dreams and wonderings were as limited as his physical and social circumstances. Similarly, because a troop of chimpanzees has no sense of history, no mythology, and no
novels, there are obvious limitations on the thoughts of its members. But they will also sleep and dream, and behave as if they have hopes and fears when awake. Deciding whether these activities imply the existence of mental states may be difficult, but there is surely more to the question than merely defining consciousness in terms of human social institutions. .
Lloyd Morgan (1852-1936) — the beginnings of the behaviourist reaction
The amiable credulity which Darwin and Romanes applied to anecdotal evidence of mental ability in animals quickly came under fire. Morgan supplied a heavy, but inaccurate, piece of artillery, by bringing forward, in his Introduction to Comparative Psychology (1894), what is known as Lloyd Morgan’s Canon. This runs:
In no case may we interpret an action as the outcome of the
exercise of a higher psychical faculty, if it can be interpreted as
the exercise of one which stands lower on the psychological
Whether or not one adopts such a rule of thumb, and how it is applied, determines the answers one will give to many of the questions that may be asked about the extent of mental organisation in animals. Subsequent writers on comparative psychology frequently assume that the virtue of Lloyd Morgan’s Canon is self-evident, because it expresses a law of parsimony, which underlies the success of scientific method, and derives from a maxim of logic, ‘Occam’s razor’, accepted since the fourteenth century. The razor says ‘Entities are not to be multiplied without necessity’, or ‘It is vain to do with more what can be done with fewer’ (Russell, 1946).
There is no doubt that parsimony is generally a good idea. But it is debatable whether Lloyd Morgan’s Canon necessarily leads to parsimonious explanations, and also whether parsimonious explanations, especially of biological and psychological phenomena, are necessarily the best ones. Applied to human actions, the Canon would require us always to assume that people act from the most straightforward and least intellectually demanding motives. This might be a useful palliative against the urge to search for deep and dark
explanations every time someone forgets a name, or misses a bus, but it would surely be unwise to assume that the simplest explanation of a politician’s promise, or a child’s tears, is always true. Similarly, because all animals feed, and move, and fight, it does not follow that the mechanisms which control these actions in the lowest and simplest species are the only ones at work in all the others. ‘Nature is notoriously prodigal; why should we interpret it only parsimoniously?’ (Boring, 1950, p. 474). Clearly, because sticklebacks and chimpanzees both build nests, we should not be obliged to believe that the psychological processes available to the chimpanzee are the same as those utilised by the stickleback. Still less should we want to deduce that because the stickleback’s nest is more elaborate than that of the chimp, that the stickleback exercises a higher mental faculty. There is a very general, and a very difficult problem behind all this, which infects the roots of behaviourist systems of explanation of animal activities. It may be described as the ‘same behaviour—same mechanism’ fallacy.
If one makes any attempt at parsimony in the explanation of behaviours, it seems eminently reasonable to apply similar explanations to similar behaviours. If a cockroach lifts its leg away from the source of a painful electric shock, and we can draw a nice graph of the strength of the leg-withdrawal response, as a function of the intensity and number of previous shocks received, we may be able to construct an aesthetically pleasing set of mathematical equations which simulate very precisely the way that behaviour changes, and which may even predict what will happen when other factors, such as the interval between the shocks, are varied. Now suppose that the same kind of experiments are performed on dogs, and the dogs flex their legs away from the shocks in more or less the same way as the cockroaches. Perhaps the dogs will be a bit quicker (or a bit slower) than the cockroaches, and we can alter a couple of constants in the mathematical equations, neatly dealing with the slight differences of behaviour. Surely this would mean that we ought to say that the dogs and the cockroaches are controlled by the same mechanisms? It would certainly violate Lloyd Morgan’s Canon if we did not. But is it really parsimonious to assume that the 100-gram brain of the dog, which will playfully fetch sticks, and obey his master’s voice, is working in the same way as the microscopic ganglia of the cockroach? Nothing could be more absurd, yet this is precisely what the invocation of parsimony, Occam, and Lloyd Morgan invites us to do. Such invitations are sent out because the Canon contains a fallacy—the
fallacy that similar actions must necessarily be given similar explanations.
Therefore Lloyd Morgan’s Canon ought to be rejected, and spurious invitations to parsimony refused. We should interpret an action as the outcome of the most likely causes, not as the outcome of the most ridiculously simple mechanism we have ever come across which could accomplish that sort of thing. The most likely cause will be the one which takes into account the most evidence. Assembling the evidence may be difficult, but it is not hard to make informed guesses about the probable capacities of a given species, or about the dispositions and circumstances that may affect a given individual. Just because an action can be attributed to the exercise of minimal psychological capacities does not mean that such an attribution is necessarily correct, if there are good grounds for alternative conjectures. In general, it is possible that actions which have much in common at the level of behavioural description are the product of quite different psychological processes. The vast majority of theorists, before and after Lloyd Morgan, have been ready to assume, for instance, that insects and cold-blooded vertebrates are controlled by collections of prewired reflexes, whereas large and specialised mammals like the dog and the monkey have evolved additional kinds of psychological capacity. If someone were able to provide a description of a monkey eating a leaf, which could be applied without alteration to an insect eating a leaf, few would feel compelled to deduce that eating leaves in monkeys results from behaviour-controlling mechanisms which are shared with the insect, although some behaviourists have gone a considerable distance in that direction. Even if one is dealing with the same species, or with the same individual, similar actions may not always result from the same causes. It is not unusual for even the most hard-bitten observers of animal behaviour to say that scratching or grooming in a particular animal may be a product of fastidiousness on one occasion, but boredom, or nervousness, on another. (That is, grooming may be classified as ‘displacement activity’; e.g. Hinde, 1970.)
To propose that isolated incidents of animal behaviour should always be given the simplest conceivable explanation is therefore quite bizarre and unrealistic. In fairness to Lloyd Morgan and to William of Occam, it should be pointed out that neither of them held views of animal psychology as fatuously oversimplified as those their names are used to defend. Occam has been credited with the opinion that animal thought is a partial and restricted form of human thought: ‘Man thinks in two ways—first, in natural terms, which he shares with
animals, and second, in conventional terms, which are peculiar to man alone’ (quoted in McCulloch, 1965). Morgan wished to be more careful than Romanes in accepting the interpretations implied in anecdotal reports, but had studied under Huxley, and was committed to the concept of the gradual evolution, of mental abilities. He preferred to rely on his own experimental observations, and took a fairly conservative view of the psychological capacities of animals, by the standards of his day, but his theoretical agreements with Romanes now stand out more clearly than his methodological differences.
Lloyd Morgan’s concept of consciousness
One indication of Morgan’s concern with consciousness is the cross-quotation between his writings and those of Wundt (1832—1920), the founding father of the scientific study of conscious sensation. In Lectures on Human and Animal Psychology (1894) Wundt expounded the principle that the mental processes of animals must be estimated by comparison with our own introspections. This principle was repeated by Kulpe (1862—1915) before it was disputed by Thorndike and Watson. Even in the light of such estimations, though, Wundt recommended that we always ‘choose the simplest . explanation possible’, explicitly repudiated Romanes, and adopted the tack taken by Leibniz, that manifestations of animal intelligence should always be interpreted in terms of relatively simple mental associations and conditioned expectations (Wundt, 1894, pp. 345—6, 354 and 362). Wundt succeeded in training his pet poodle to shut the various doors to his study on command, but stressed that after being trained with the first door, the animal did not immediately respond to commands directed at the other doors. One of the doors opened outwards, so that after the dog had pushed it closed, it was left outside. It would usually then scratch at the door to be let in again, from which Wundt deduced that the poodle did not properly understand the final purpose behind the operation, which was to save Wundt from getting up.
Lloyd Morgan, in Habit and Instinct (1896) repeats the strictures against Romanes, but in chapters on ‘The relation of consciousness to instinctive behaviour’ and ‘Intelligence and the acquisition of habits’, he puts forward a carefully considered theory of the limits and functions of subjective awareness in animals. The essence of the theory is the assumption that instincts, and well-formed habits, are automatic, but the non-habitual utilisation of previous experience requires consciousness. This is a very old distinction, and Darwin had
specifically disagreed with the emphasis placed on it by Cuvier, but Lloyd Morgan gave it a systematic restatement. Habits depend on previous experience, and amount to acquired automatisms, after sufficient repetition. But, rather as the drilling of raw recruits requires the direction of a drill- sergeant, the organisation of activities which may become habitual requires the intervention of consciousness, Qrepresented in the flesh by the cerebral cortex’. Consciousness may be defined as whatever it is that ‘enables an animal to guide its actions in the light of previous experience’, something which ‘Looks on and makes a memorandum of what is going forward’: ‘when we say that conduct is modified in the light of experience we mean that the consciousness of what happened yesterday helps us avoid similar consequences today’ (Morgan, 1896, pp. 127, 131 and 145). In other words, in ‘the higher grades of animal life’ at least, even simple associations require mental awareness, rather than only unconscious and mechanical connections. As examples of this, Morgan quotes his own observations of the effects of prior experience on the behaviour of young birds. A chick when it first sees a cinnabar caterpillar will seize and eat it, but find it extremely unpalatable. The next time it sees such a caterpillar, it shrinks from it on sight. Morgan assumes that the first encounter with the caterpillar must have been conscious, for it to have any effect on future behaviour. This is the topic of another Canon, now forgotten, which states that present behaviour can only be guided by previous experiences which were conscious in the first place.
Similarly, when moorhen chicks, which Morgan fed by turning over soil with a spade to expose worms, learned to run towards him as soon as he picked up the spade, Morgan deduced that the chicks must have been conscious at some point of the conjunction of the spade with worms. When they ran to the man with the spade it was thus because they remembered the worms. If they did this often enough, running to the spade might become an automatic habit, but conscious direction of actions by remembered experience was an essential initial phase of habit formation in Lloyd Morgan’s theory. This is certainly not the simplest possible explanation of habit formation (although perhaps it was the simplest explanation that occurred to Lloyd Morgan). In any event, it should be noted that Lloyd Morgan’s own application of the rule of parsimony did not preclude the discussion of awareness and memory in chicks, as a reminder that there may be such a thing as an explanation of animal behaviour that is too simple to be sensible.