Alfred Russel Wallace, co-founder of the theory of evolution, argued that the exceptional ‘mental and moral qualities’ possessed by human beings exempt them from the normal processes of evolution. These qualities enable them to anticipate, annul, sidestep and generally defeat the normal processes of evolution. This paper examines how far current knowledge of natural intelligence (human and otherwise) supports this attractive but extremely dangerous hypothesis. It argues that the ‘objects’ in terms of which intelligence constructs its activity and its environment overcome the fundamental mutual indifference of variation and selection that powers evolution; that all forms of subjectivity – the internal structure of intelligence – tend to converge on a single invariant structure regardless of species, and so eliminate the possibility of variation at the most profound level of intelligence; and that the world intelligent beings jointly create – massively structured by culture and technology and driven by economic and political systems that bear little resemblance to any biological force – is sufficiently powerful and intentionally directed to make it a force to which evolution is subjected rather than vice versa. In brief, evolution does not know what it is doing, and intelligence does. Taken together, these factors explain how it is possible for the logic of variation and selection to be first superseded and then subordinated to that of intelligence itself. Indeed, once the specific nature of intelligence is recognised, this model also implies that intelligence’s independence of evolution is far more radical than even Wallace realised.
One of the best known (and perhaps most notorious) ideas hatched in the early days of Darwinism was Alfred Russel Wallace’s suggestion that human beings are exempt from the normal strictures of natural selection: unlike other animals, our ‘mental and moral qualities’ allow us to circumvent the selection pressures on which evolution relies.
So when a glacial epoch comes on, some animals must acquire warmer fur, or a covering of fat, or else die of cold. Those best clothed by nature are, therefore, preserved by natural selection. Man, under the same circumstances, will make himself warmer clothing, and build better houses; and the necessity of doing this will react upon his mental organisation and social condition – will advance them while his natural body remains naked as before…
Thus man… has taken away from nature that power of changing the external form and structure which she exercises over all other animals. As the competing races by which they are surrounded, the climate, the vegetation, or the animals which serve them for food, are slowly changing, they must undergo a corresponding change in their structure, habits, and constitution, to keep them in harmony with the new conditions – to enable them to live and maintain their numbers. But man does this by means of his intellect alone; which enables him with an unchanged body still to keep in harmony with the changing universe. (Wallace 1864: 163)
As a result, ‘We must… look back very far in the past to find man in that early condition in which his mind was not sufficiently developed to remove his body from the modifying influence of external conditions, and the cumulative action of “natural selection”’ (Wallace 1864: 167). Nor was Wallace alone in this belief.
[I]n the scale of life there is a gradual decline in physical variability, as the organism has gathered into itself resources for meeting the exigencies of changing external conditions; and that while in the mindless and motionless plant these resources are at a minimum, their maximum is reached in the mind of man, which, at length, rises to a level with the total order and powers of nature, and in its scientific comprehension of nature is a summary, an epitome of the world. (Chauncey Wright in Norton 1877: 104)
Unfortunately, neither Wallace nor Wright had a credible explanation for these remarkable ‘mental and moral qualities’. Wallace could only argue that ‘some different agency, analogous to that which first produced organic life, and then originated consciousness, came into play in order to develop the higher intellectual and spiritual nature of man’ (Wallace 1905: 17). This ‘explanation’ cannot be distinguished from divinely ordained creation, and Wallace’s position eventually descended into overt mysticism (Oppenheim 1985: 303-325; Richards 1987: passim; Robinson 2004: 324-327).
In fact, the capabilities that Wallace believed put on us a special position – abstraction, foresight, reasoning, calculation, concepts of symmetry and infinity, and so on (Richards 1987: 183) – remained largely unexplained by science until Piaget’s studies of child development (e.g., Piaget 1953a). However, not only is Piaget’s solution completely naturalistic but it is also logically opposed to Wallace’s. So, as I shall argue below, if Wallace should have the credit for defining the problem, it was Piaget who gave Wallace’s thesis its ultimate plausibility.
Meanwhile, Darwin’s reaction to Wallace’s argument was one of surprise and caution, but initially he could not resist its logic:
I have now read Wallace’s paper on Man, and think it most striking and original and forcible… I am not sure that I fully agree with his views about Man, but there is no doubt, in my opinion, on the remarkable genius shown by the paper. I agree, however, with the main new leading idea. (Quoted in Eiseley 1961: 305)
Nevertheless, whenever Darwin discussed human evolution (in his major works at least), he disregarded Wallace’s argument. Certainly The Descent of Man (Darwin 1901) is quite untouched by Wallace’s ‘new leading idea’.
But the idea was not quite dead. James Mark Baldwin (by whose theories Piaget’s own views were strongly conditioned – Piaget 1979, 1980a) argued that human beings might buffer environmental pressures (Baldwin 1895, 1896, 1902; Richards 1987) by cultural means, pending the evolution of appropriate organic adaptations. Much later still similar ideas were elaborated by Waddington (1957), and Deacon and others have since adopted Baldwin’s ideas as part of a more general account of the relationship between language, symbolism, culture and evolution (Deacon 1997).
However, Baldwin’s reasoning also creates the opposite possibility. In short, if a functional (e.g., behavioural) response can deflect a given selective pressure until an adequate organic adaptation is available, would this very functional success not serve to reduce the pressure for a structural (e.g., genetic or organic) solution of any kind? If so, then this seems to create an indefinitely large space not only for non-genic biological inheritance (as modelled by developmental systems theory – Oyama et al. 2001) but also for the kind of long-term structural stability Wallace hypothesised. Yet, although Baldwin’s ideas are now studied widely, Wallace’s more radical conclusions remain unconvincing to the modern ear. For example, the complexities and power of culture and technology, of social and economic systems, of science and politics, all make his emphasis on our ‘mental and moral qualities’ seem extraordinarily narrow. But perhaps this is a little churlish, given the radicalism of Wallace’s underlying idea and the relatively undeveloped nature of the human sciences in his day. But his practical examples are not particularly convincing either, and it is not difficult either to show that Wallace’s claim is overstated or to catalogue ways in which the human body has been modified by natural selection more recently than Wallace’s hypothesis, taken at face value, allows (e.g., Moran 2000; Harrison 1993).
Nevertheless, many of the best documented instances of human evolution are of physiological and anatomical modifications that arose before we had reached the cultural or technological level (i.e., acquired by historical means the ‘mental organisation and social condition’ to which Wallace also appealed) needed to pre-empt the relevant selective pressures. If the last few millennia have seen a convergence of European and African lactase tolerance (Tishkoff et al. 2006), or Andeans acquiring permanent physiological adaptations to high-altitude hypoxia, or the long-term inhabitants of tropical regions dealing with humid heat better than the rest of humanity, this is perhaps because milk, oxygen pressure and humidity are features of the environment to which human intelligence had no effective ‘mental or moral’ response until very recently. Now, of course, we have innumerable devices, from pressurised cabins to air conditioning to food supplements to disease eradication programmes, that would render any permanent organic adaptation quite redundant.
For example, we no longer seem to be as plagued by plague as we once were. Yet the human body has probably not evolved much since the plague pandemics of the sixth and fourteenth centuries, and has plague not gone away. On the contrary, it remains endemic across the planet (WHO 2005). Despite this, unlike much of the rest of the world, Europe and North America managed to avoid the plague pandemic that closed the nineteenth century. This is certainly because the changes to our mental organisation and social condition introduced since the previous outbreaks – through medical science and epidemiology, public hygiene and drainage systems, health education, the pharmaceutical industry, welfare systems, and so on – make it extremely unlikely that a modern outbreak would have the usual cataclysmic impact on industrial societies. Indeed, between 1947 and 1996 390 cases of plague were reported in the United States (Morbidity and Mortality Weekly Report 1997), but no epidemic. In short, we have ‘adapted’ extremely successfully to plague, yet without undergoing biological changes that could properly be described as ‘adaptations’.
But there have been other criticisms of Wallace’s hypothesis. From the present point of view, the most important is the argument that the greatest selective pressures on human beings now come from competition between social systems. Which is to say, far from selective pressure having ceased, it has been internalised into the human way of life. But this is not altogether convincing. First and foremost, this point was not lost on Wallace himself, who expressly included in his analysis ‘the competing races by which they are surrounded’ among the forces to which human beings respond. Secondly, it is doubtful how much pressure social forces really exert on either social systems or their members’ reproductive fitness (Hallpike 1984, 1988). Finally, our neighbours are precisely as intelligent as we are, so both the critical inter-societal pressures and our major responses to them – political, economic, cultural, technological, and so on – are actually only further expressions of our common intelligence – i.e., the very factors of ‘mental organisation and social condition’ to which Wallace appealed. Such pressures are created by competition between intelligent beings, take specifically intelligent forms, and are responded to in intelligent ways. These are neither created nor structured like biological variations and adaptations, and neither do they lead to noticeable biological changes.
Even as massively ‘selective’ an event as the Second World War was not even metaphorically a variation in that evolutionary sense, however much its architects may have imagined that they were carrying out the dictates of biology. More generally, the fact that major aspects of human existence (wars, large-scale economic systems, cultural shifts, and so on) are poorly understood and hard to control by those involved in them does not turn them into biological variations, not even in the sense of ‘extended phenotypes’ or meme theory (Dawkins 1976, 1982; Blackmore 1999). Conversely, once they are understood, such fluxes and irruptions always turn out to follow a plainly non-biological logic that the human sciences can analyse without recourse to biological concepts.
By contrast, the unstructured character of strictly biological variations is inherent in the process of variation. That is, they are essentially random, although not generally in the rather literal stochastic sense (essentially, that they come out of the blue). There is actually a great deal of structure to most biological variation, not only in its origins but also in its mechanisms, in the constraints and conservation processes that regulate its effects (Maynard Smith and Vida 1990) and in the layering of the process of inheritance itself, which is by no means limited to the traditional Mendelian genes (Jablonka in Oyama et al. 2001). The global biological context within which variation occurs also has an immense effect on its significance (e.g., Oyama 2000). All in all, not much variation is of the popular ‘cosmic-ray-flips-codon’ type.
On the other hand, there is more to randomness than, as has been suggested (e.g., McKinney 1998), simple lack of knowledge of the real causes. As Eble has said of stochastic and evolutionary randomness, ‘The former implies a combination of indiscriminate sampling and unpredictability due to multiple causes; the latter codifies independence from adaptation and the directionality imposed by natural selection’ (Eble 1999: 75). Or as Ghiselin put it, for Darwinvariation ‘is not random but fortuitous’ (Ghiselin 1969: 165; see 164-167 generally). In short, evolutionary variation is objectively random in the functional sense that there is no link between the causes of variation, the effects such causes have on the organism’s structure, the impact this change of structure has on the organism’s functioning, and the consequences of this behaviour for the organism’s reproductive fitness. Shorter still, there is no internal link between the changes in functioning a variation will engender and the criterion of functionality by which selection will ultimately ‘judge’ it. Variations are not ‘for’ anything at the point of variation, no matter the ‘use’ to which selection may later put them.
Exactly the opposite is true for intelligence: to act intelligently is (among much else – Robinson 2004, 2005) to design, execute, monitor and judge that act for the sake of and in terms of a definite goal, rule, value, and so on – which is to say, specifically in terms of the criterion by which it will be judged. The functioning of intelligence is not only connected to but specifically defined and driven by its intended functionality. We do things on purpose, and we can tell whether or not we have succeeded.
This break with the rules of evolution is fundamental to humanity’s extraordinary qualities: we can manage biological problems like the plague (and, hopefully, our current environmental crises) only because we no longer rely on biological ‘methods’ such as variation and selection to do so. For example, although medical research throws up a hundred failures for every successful remedy, not one of those failures is a ‘random variation’ in any sense. Not even the most poorly conceived drug development programme happens without a prior purpose, success criteria, a rationally structured methodology, explicit specifications, a development plan and all the other trappings of a rational project. Neither does a mass vaccination or sewer-building programme. Or if it does, we have no difficulty in detecting why it fails. They are all explicitly and more or less rationally connected to the nature of the problem, the goals of their agents and a large corpus of established knowledge and powers, solutions and methods. Even ‘blue sky’ research relies on the internal ability to recognise a desirable outcome when it appears.
Likewise, the systems, processes and mechanisms of human culture, history, technology, and so on, can hardly be said to rely on biological changes, if only because massive variation has occurred in them without their being accompanied by any detectable biological differences in ourselves. In the last ten millennia we have been everything from hunter-gatherers to industrialists and used every technology from sticks and stones to linear accelerators, without any apparent change in either our biology or the nature of our intelligence (though its level seems to have increased markedly – Hallpike 1979; Wynn 1979, 1981, 2002). This is not evolution, even though it represents far more change than any other species has ever undergone, not only in such a short time but perhaps at all.
This immediately raises two questions. Firstly, what are the implications of this unprecedented departure from the normal rules of evolution? And before that, what is the nature of the intelligence that is capable of such unprecedented results?
2. Theories of intelligence
Until the recent rise of adaptationist and modular explanations for everything, most scientists (biologists included) asked what it is about human beings that makes them so unusual would have answered, their intelligence. Despite this, science finds it difficult to say what this elusive term means: like ‘mind’, ‘consciousness’, ‘thought’ or ‘reason’, ‘intelligence’ resists definition even by those who have the most. But perhaps exact definition is too ambitious a goal: trying to define intelligence in our present state of knowledge is like trying to define matter or life two centuries ago. After all, in more than sixty books and hundreds of papers, even Piaget provided no convincing definition. This creates an extraordinary paradox that any self-aware being may experience: as Sir Thomas Browne put it in his Religio Medici in 1643, ‘Thus we are men, and we know not how’.
For present purposes, I suggest a simple (and perhaps annoyingly abstract) definition of intelligence. An activity is intelligent when it knows itself to be that activity and is that activity by virtue of that knowledge (Robinson 2005). Likewise for intelligent relationships and experience. That is, intelligent activity is more or less explicitly informed of and by its own structure, content and context. This in turn implies a capacity for insight into the nature (signs, properties, qualities, relationships, structures, rules, systems, and so on) of the things, situations and events that provide so much of that structure, content and context, in terms of which they may be created, explained and controlled. Hence the intimate link between intelligence and the rationality, for by this definition, an intelligent being can always be expected to give the ratio (reason, explanation and justification – Smith 1993 – plus supporting motives, goals, attitudes, and so on) for their activity. It also implies the ability to reflect on and criticise those structures for their validity, effectiveness, compliance with norms and values, and so on. Or if it doesn’t, it is hard to imagine what philosophy and then science have been doing for the last few millennia, or what you the reader are doing at this very moment.
This analysis places intelligence in an ambiguous position from an evolutionary point of view. Most of all, any appeal to the fantastic scale of the difference between human and non-human accomplishments and their impact of their respective environments can look suspiciously like one of those near-magical saltations that, like a vacuum, nature is supposed to abhor. On the other hand, if human beings are plainly intelligent by this standard, so are certain primates (Parker and McKinney 1999), and perhaps several cetaceans and birds. But the claims of less sophisticated mammals and birds to, say, some degree of insight into their own actions (Vauclair 1996) is often as difficult to judge as disputes as to whether a virus is alive. So it is often simpler for scientists to concentrate on the continuities between intelligent and non-intelligent forms of activity, or perhaps to ignore the intelligent nature of human activity, experience and organisation altogether. However, I do not believe that this is any more fatal to my definition than the existence of viruses is to the idea that life is fundamentally different from chemical reactions.
There is no lack of would-be scientific theories of intelligence. Historically speaking three have been especially prominent: neo-Darwinism, various forms of information processing theory and, trailing rather badly these days, Piaget’s ‘genetic epistemology’. Neo-Darwinian arguments are not only extremely widespread throughout psychology, philosophy and the human sciences in general, but a neo-Darwinian solution would create a simple and direct link to the natural sciences as a whole. Information processing theory is equally prominent, and widely used in conjunction with neo-Darwinism in areas such as ‘cognitive science’, and unlike neo-Darwinism its relationship to intelligence is at least superficially self-evident. Piaget’s influence, by contrast, seems to be rapidly waning, yet as I shall argue below, his account of intelligence offers much the most convincing solution.
Many neo-Darwinians have claimed to explain aspects of human intelligence in evolutionary terms, under such rubrics as ‘sociobiology’, ‘evolutionary epistemology’, and ‘evolutionary psychology’ (e.g., Barkow et al. 1992; Campbell 1974; Hahlweg and Hooker 1989; Radnitzky and Bartley 1987; Ruse 1986). They have generally argued that human cognition, perception, motivation, memory, and so on, are conventional biological adaptations, engendered by conventional variations, shaped by selection and ultimately driven by reproductive fitness. Species-specific characteristics are traced to species-specific contingencies, often (though not always – Tooby and DeVore 1987) in a putative ‘Environment of Evolutionary Adaptation’ (EEA). But in either case evolutionists assume that the nature of intelligence is adequately explained by its evolutionary origins, and that it will continue to be shaped by these same evolutionary forces. The only distinctions human ‘mental and moral qualities’ can claim are exceptional effectiveness and (perhaps) extraordinary methods. In particular, culture is often credited with an alternative, quasi-Lamarckian mode of inheritance, though this does not alter culture’s ultimate biological raison d’être, which remains reproductive fitness.
So how is it, generally speaking, that evolutionary methods might not apply to intelligence? In brief, the fact that intelligence is indisputably a product of evolution (Robinson 2005) does not imply that, once it exists, intelligence remains subject to further evolution, any more than the fact that life emerged from and, in a sense, consists solely of chemical reactions means that chemistry can adequately describe or explain the organism, evolution, heredity, and so on. If I take poison, how can I express this in chemical terms? The poison itself maybe purely chemical and the pathways through which it has its dire effect may be open to chemical assay, yet there are no strictly chemical criteria differentiating ‘alive’ and ‘dead’. Or rather, there are none that adequately express the radical biological result.
An intelligent organism is likewise still an organism and its intelligence has a massive and profound impact on its adaptiveness, fitness or biological performance generally. Nevertheless, one should not necessarily expect evolutionary theory either to explain the nature of intelligence or to define what an intelligent organism is doing when it is being intelligent – even when it is doing something that is obviously adaptive. As I have already argued, a great deal of human adaptiveness comes from our very ability to circumvent the basic logic of variation and selection. So if, from a neo-Darwinian perspective, intelligence is just another adaptation, that is precisely the problem. If neo-Darwinism assumes that intelligence not only was created by but also continues to be shaped by evolution, then the question of whether or not intelligence evolves cannot even be posed. Conversely, if intelligence is immune to further evolution, then there is nothing in neo-Darwinism’s intellectual armoury that can explain (or even imagine) this fact. As neo-Darwinism’s Humean approach to knowledge and activity is far from self-evident (Smithurst 1995), this is a fundamental obstacle. Hence the inappropriateness of neo-Darwinism from the present point of view: it assumes the very thing that needs to be proved.
Information processing theory is equally blind to the possibility of a qualitative break between the intelligent and the non-intelligent. Its basic definitions and models are assumed to be universally applicable. By analysing all forms of activity in terms of ‘encoding’, ‘transformation’, ‘processing’, ‘storage’, ‘coding’, ‘attention’, ‘access’, ‘retrieval’ and so on, it necessarily glosses over any qualitative differences that may exist between intelligent and non-intelligent organisms. Thus information processing theory too assumes the very thing the present essay calls into question. As with neo-Darwinism, its rejects the possibility that intelligence may not be subject to evolution not by proving it false but by rendering it unthinkable.
Indeed, although it is possible to imagine experimenting in a classically ‘intelligent’ area such as, say, causality or means-ends relationships in such terms, it is hard to see what it is about causality or means-end relationships that could be ‘explained’ within such the framework of information processing theory. I may be able to discern how causal relationships are encoded, stored and retrieved, but what this tells me about causality as such is not clear, given that means-end relationships are analysed – and everything else – in exactly the same terms. Rather, it seems as though all the knowledge of causality or means and ends is in fact injected by the experimenter.
Hence, perhaps, the rise of the all-purpose doctrine of ‘modules’. Being unable to specify what causality is when one is limited to the extremely generic logic of information processing theory, it becomes necessary to postulate the evolution of a special-purpose structure that accounts for the actual function (e.g., memory, face recognition, and so on), whose functional workings can then be analysed in standard information processing terms. But how was it that this module was capable of grasping causal relationships in the first place? Because it was adaptive, so when variation threw up a structure that was capable of grasping causality, it was selected for. Hence the ‘evolution’ of causal reasoning. The problem of explanation is thus delegated to omniscient, omnipotent, omnipresent Evolution. Or perhaps an Intelligent Designer put it there.
The common problem underlying all this is that, if the hypothesis that intelligence is immune to evolution is to be properly examined, this examination can only be based on a clear conception of the nature of intelligence. By this I mean an analysis of not only of its functions but also of its structure and its genesis. Only this can explain whether it is the kind of entity to which concepts such as evolution or information processing apply. There are after all plenty of structures to which they do not apply. Practically everything changes, of course, and many things develop – certainly the elements (in a sense), and perhaps even physical laws. However, such things do not change in the manner of biological variations (however understood) or by being subjected to any kind of exogenous selection process. There are after all no functional criteria analogous to ‘reproductive fitness’ among atoms and molecules, by which physical or chemical selection could be driven. As for the sub-atomic plane, quantum theory rejects the idea of intermediary atomic orbits or electron counts between the specific fixed alternatives dictated by its own a priori laws, so the very idea of variation is excluded a priori.
In fact in any but metaphorical terms, evolution seems to be absent from any plane of material organisation but the strictly biological. The question then arises as to whether exemptions from evolutionary logic apply only to physical and chemical structures operating below the biological level. For if there are also structures that operate above the level of the organism, may not they too be free of variation, selection and evolution generally, in the specifically Darwinian sense? I suggest that the answer to this question is Yes, and that intelligence, which has healed the rift between functioning and functionality evolution itself introduced, is such as structure. In the case of chemical and physical structures, this freedom from evolution follows from their lack of anything resembling function; intelligence, by contrast, is free from evolution because it realises a self-reference and closure, so to speak, that short-circuits the basic functional logic of evolution, and replaces it with an altogether more powerful logic of its own devising.
3. Piaget’s theory
If neither neo-Darwinism nor information processing theory can judge whether intelligence is immune to evolution, are there any alternatives? Few other conventional approaches seem any more appropriate. For example, intelligent awareness, knowledge and insight are substantially independent of perception, memory, communication and the other psychological functions we share with non-intelligent species, so these are not suitable substitutes (Langer 1998: 40-46). Although human beings are not especially strong performers in areas such as location memory or the direct perception of numerosity, as soon as they start to introduce strictly cognitive forms (e.g., true number systems, symbolic forms, technological support, etc.), their performance in all these domains is infinitely superior to that of any non-intelligent species. Nor is any model that assumes that the basic mode of interaction between intelligent being and their ‘environment’ is passive observation likely to suffice. Action actively disturbs, relates, creates and transforms objects, and so reveals previously unobserved features, non-empirical aspects of the object, inter- and trans-object relationships, the structure of the activity itself, and the overall logical properties of the situation in which it takes place. Without these, no knowledge of the world can be rational or even substantial, even if it can quite enough to support reproductive fitness.
My own view is that, of the theories that are able to address the evolvability of intelligence in such terms, Piaget’s is most capable of testing the present hypothesis fully. This may seem an odd choice: isn’t Piaget’s ‘genetic epistemology’ a little passé, if not downright obsolete? In fact, although regularly criticised and occasionally pronounced dead, Piaget’s theory remains intact (Lourenço and Machado 1996), and the evidence in its support continues to grow. Of course, his ideas are no longer focal even to child development or education, and his original findings have certainly been subjected to a great deal of refinement and correction. Yet the idea that a single structure informs intelligent action, experience and relationships, and that it develops according to a sequence of stages of the general kind Piaget originally described, and that both its activity and its development revolve around continual rational re-structuring, remains robust. In addition, because it analyses intelligence in structural as well as functional terms and is grounded in a very mature developmental model, Piaget’s theory provides a more promising starting point for evaluating the evolvability of intelligence than any of its competitors.
Furthermore, Piaget’s theory is one of the most wide-ranging in the whole of psychology, so the support it might offer – or deny – to the present hypothesis is considerably more telling than that offered by its competitors. Piaget’s intelligence infuses a universal (if not always homogeneous) structure into activity of all kinds, including the full gamut of psychological, social and symbolic relationships. Piaget’s more than sixty book-length publications cover not only the huge range of cognitive and educational topics for which he is best known but also perception (Piaget 1969) and memory (Piaget 1968b); social relationship, values and moral judgment (Piaget 1932b, 1995); affects and emotions (Piaget 1981); play, imitation, mental imagery and dreams (Piaget 1951, 1971c); judgement, reasoning and language (Piaget 1928, 1932a, 1951); the epistemology, history and philosophy of science (Piaget 1950, 1971b Piaget and Garcia 1989); the psychology and epistemology of logic and mathematics (including geometry, quantification, number and chance) (Beth and Piaget 1966; Piaget 1952, 1953b, 1972; Piaget and Garcia 1991); identity, contradiction and dialectics (Piaget 1968a, 1974, 1980b); both general biology and evolutionary theory (Piaget 1971a, 1979, 1980a); and consciousness (Piaget 1977, 1978).
But not even this monumental personal oeuvre has exhausted Piaget’s ideas. Areas to which his ideas have been extended include primatology (Antinucci 1989; Parker and McKinney 1999; and many others); the anthropology and archaeology of cognition (Damerow 1995, 1996; Hallpike 1979; Wynn 1979, 1981, 2002); the historical development of religion and moral values (Barnes 2000; Fowler 1981; Hallpike 2004); the transformation of medieval reasoning, theology, architecture and drama (LePan 1989; Radding 1985; Radding and Clark 1992); the history of art (Gablik 1977); the nature and logic of norms (Smith 1993); animal rights (Wise 2002); and even the most general structures of culture and technology and the large-scale structure of history itself (Robinson 2004).
Despite its enormous scope, Piaget’s theory is fundamentally simple. With only a few exceptions, such as his accounts of moral judgement and emotion, the whole of Piaget’s account is characterised by a profound unity. On the other hand, like any powerful simplicity, there are as many ways of looking at it as there are interests in its subject matter, but for present purposes only a few basic elements need be identified.
Firstly, notwithstanding the widespread misunderstanding of his theories even among psychologists, in the Piagetian model intelligence is not essentially either intellectual or even cognitive, at least insofar as the latter term is narrowly defined. Piaget’s intelligence is a universal structure informing all forms and aspects of activity, from cognition and memory to values and social relationships. In that respect, to be ‘intelligent’ is as comprehensive and profound an attribute as being ‘alive’. This unity allows Piagetians to conceive of problems such as space, play, scientific method, number, moral values or history as simply different facets of activity as such. As a result, they move seamlessly from the central abstract problem of how we (for example) experience objects in general or objects as such to innumerable more concrete aspects of objects (such as identity, topology or causation, or even religious belief and artistic expression) and back again; or from a narrow component of activity (such as grouping, seriation, reversibility or possibility), up to the logic of activity in the abstract (and especially its formal logico-mathematical structure) and back again; and to do both without losing track of the wider picture of intelligence as a whole.
Secondly, Piaget’s intelligence is inherently reflexive. Like any organism, intelligence experiences its environment by acting on it, and this activity creates an awareness that fuses the things on and through which intelligence acts with the patterns of the activity itself. In non-intelligent organisms this confusion is not resolved, but because, in intelligent activity, the product of this fusion takes the form of an object – which is to say, of an entity that is experienced as existing independently of both the agent and the activity through which it was constructed – successive constructions will progressively define three things in increasingly distinct and refined terms. Firstly, they will define the object itself – for its own sake, in its own right and on its own terms. Secondly, by allowing the same activity to be applied to a multiplicity of objects in a multiplicity of circumstances, they will progressively differentiate the activity from its effects, and so define the activity too. This will culminate in recognition (i.e., objectification) of both the activity ‘as such’ (e.g., ‘throwing’, dropping’, ‘hiding’, and so on) and the abstract relationships between various forms of activity (i.e., logic). Finally, the progressive differentiation of activity and object will also reveal (i.e., objectify) a further feature, namely the agent of that activity and the structure that unites all this activity with its objects, namely the subject.
Hence the third crucial feature of Piaget’s model of intelligence. This constant process of reflection on its objects drives experience in two directions: not only towards empirical abstraction of the features, properties and structures of things, situations and events, but also towards reflecting abstraction of the features, properties and structures of its own activity (Piaget 2001). This in turn provides the basis for intelligence’s continuous, open-ended development. The objects the infant can construct during its sensorimotor phase are recognised as existing independently of the infant but, from the infant’s own perspective, possess only sensory and motor properties. However, by distinguishing and reflecting on the actions through they are composed, arranged, transformed, and so on, the toddler constructs stable groupings, classes, series, and other structures, which are then reconstructed by the older child into yet higher structures such as number, which are themselves re-synthesised by the young adolescent into intellectualised structures furthermore formal mathematics. At the same time, practical action and experience are also progressively constructed and reconstructed and, like the logico-mathematical structures, used to inform yet higher cycles of activity.
If Piaget’s analysis of this process is correct, then intelligence’s developmental potential is logically unlimited. On the one hand, it does not require new abilities to reach each new developmental stage, because essentially the same cycle of objectification, reflection and synthesis recurs over and over again. It may be limited by general neural capacity, but this is a limitation that human evolution has consistently pushed back, not only because the rewards higher intelligence brings is always worth the additional investment in such ‘expensive’ tissues (Aiello and Wheeler 1995; Passingham 1975) but also because the creation of more and more cultural and technological tools has allowed the intelligent organism to embed its own intelligence in a world of artefacts and institutions that are capable of implementing intelligent activity and realising intelligent values in the absence of a natural intelligence. From speech sounds to global economic systems and deep-space astronomy, this world is not only accessible to every intelligence (including members of other cultures and even other intelligent species) but also capable of supporting new objects of any scope, depth or complexity, limited only by the objective laws of the universe.
But perhaps the most radical feature of Piaget’s theory is that none of these unprecedented features are built into the infant. There seems to be is nothing in a newborn baby that tells them how to develop or what stages they should pass through, any more than a stone contains instructions for rolling downhill or water for splashing. That such propensities are the results of universal structures of matter rather than peculiar to the objects in question is shown by the fact that the same principals of ‘splashing’ are shared by oil, custard, piles of fine dust, and the Haldane’s hypothetical falling horse encountering the unyielding bottom of a mineshaft. Instead of being determined by any positive feature of the objects in question, it seems to be the absence of either constraints or positive propensities that causes all material objects to fall in the same manner, all fluids to flow in the same way – and the development of all intelligences to follow the logic of the object as such and the logic of activity as such rather than any predefined method or template, biological or otherwise.
Although this point will be elaborated later, it may seem paradoxical, but in fact precisely as much material effect is caused by the lack of structure as by its presence. For example, the singularity of black holes, the colour of blue eyes, albinism, cystic fibrosis, and so on, all arise from the absence of specific ‘programming’. It is what that type and level of matter does if it is not ‘told’ to do anything else. I have argued elsewhere (Robinson 2005) that the origins of intelligence lie in driving adaptation to the limits of ‘adaptability’, but adaptability itself is at least partially a lack of predetermination (i.e., the absence of specific forms of ‘adaptedness’) as to which structures are applied to or how they be coordinated. In that respect Piaget was many decades ahead of recent ideas on emergent structure, complex systems, and so on. Conversely, built-in solutions, practically useful though they might be in the short run, would prevent intelligence from pursuing to its logical conclusion the nature of the objective reality by which it is confronted and of which it is composed. Built-in modules and genetic programmes may get the organism off to a flying start, but in the long run they can only hinder it from coming to the finish line set by reality. That is why we start out inferior to so many species in every arena we enter, yet quickly overtake every other contender. A century ago we could barely get off the ground, but I would challenge any other species to out-fly us now.
The final aspect of Piaget’s theory that is relevant here is that it is inherently social and symbolic. With the partial exception of mental objects (images, ideas, dreams, thoughts, and so on), objects exist for the most part in a public arena where they may be encountered by any other competent intelligence. A child’s ball, experimental evidence, a corporate bureaucracy, are all ‘out there’, even if what we make of them is not. By the same token, our cognition of the objects that are produced by other subjects (including specialised social objects such as language, facial expressions, gestures, writing, cultural artefacts, and so on) and our developing grasp of the objective relationships through which we relate to other intelligences (exchange, roles, property, rules, gifts, theft, obligation, institutions, shared values, negotiation, war, etc.) allow us to construe a good deal about, and establish relationships with, other individuals, and indeed to deliberately create systems, institutions and other structures that embody, manage and further such relationships on our behalves. Finally, if objectifying something is the consequence of construing the internal structure that ‘must’ be responsible for its detectable functioning, then there is no special problem in objectifying the internal structure of another intelligence – which is to say, in objectifying and understanding other subjects – and no special theoretical problem in understanding how one human being can understand another.
Hence the inescapably social character of the development of Piagetian intelligence – the very antithesis of the ‘lonely scholar in the pram’, as the Piagetian infant was once called. From the very first there is very little that intelligence creates that is not either produced with, directed at or mediated by other intelligences, starting with its caretakers, siblings and peers and culminating with massively abstract social systems such as (in our society) law, roles, employment, money, markets and commodity relations (Robinson 2004). In addition, these others’ actions and constructions are themselves mediated by their own expectations, goals, capabilities and relationships, and by wider social structures of various kinds. Even the radical individuality and individualism that psychologists tend to take for granted (and which alone allow them to proceed in such isolation from the other social sciences) are actually relatively recent and historically very specific social developments, having become the norm only with the widespread formation of capitalist societies (Robinson 2004). In other words, such individualism is itself possible only by virtue of intelligence’s inherent social dimension.
The object is also the basis for intelligence’s propensity for symbolism. Because intelligence is capable of constructing objects in terms of supra-empirical classifications, models, structures, and so on, it can refer from one object to another through any number of relationships – analogy, equivalence, simile, metaphor, metonymy, and so on. At the same time intelligence also acquires the especially telling capacity for completely arbitrary and conventional symbolism – a defining feature of so much of human language and culture. Because its development is equally driven by empirical and reflecting reflection, intelligence can increasingly reflect on the structure of its own actions without reference to any particular empirical content or context. This distinction is present from very early on (for certain purposes, throwing a feeding bottle out the pram is just as good as ejecting a toy), and eventually reaches the point where the empirical object is required only to instantiate the action, the ideal, the intention, and so on, and can be selected from all possible objects that are amenable to the action in question. From that point onwards, to the extent that a completely conventional object will do for the activity in question (i.e., not only a sound or written mark but also a team colour, a national flag, a gesture or stance, and so on), the symbol itself can be completely arbitrary even though what it symbolises is anything but.
It would be difficult to exaggerate the significance of intelligence’s social and symbolic dimensions. Although this is not the place for an extended analysis (but see Robinson 2004), the long chains and complex systems of subjects and objects to which social and symbolic relationships lend themselves allow intelligent activity to be constructed, transmitted and reconstructed over epochs and continents. This raises the possibility not only of culture and technology but also of history itself. When combined with the ability to construct objects, systems of objects and systems of systems to implement and support social and symbolic relationships, then the methods, tools and techniques available to an advanced intelligence quickly replace the more or less hand-to-mouth and individualistic strategies of even the most sophisticated non-human primate with true economies, and their small-scale and primarily face-to-face social systems with true political systems. It is hardly necessary to say what the impact of such changes has been, not only on human existence but also on the planet as a whole.
So much for the main dimensions of Piaget’s theory. But what is it about that intelligence that specifically allows it to rise above evolution? Answering this questions requires a more detailed analysis of three interrelated aspects of intelligence: object permanence; the internal structure of intelligence (which I will refer to as the ‘subject’); and the organisation of subjects and objects into the fundamentally new form of ‘environment’ we call ‘the world’.
4. Object permanence
The single best known and most widely researched Piagetian concept is object permanence. As a Kantian, Piaget assumes that it is impossible for us to know what things are ‘in themselves’. However, we are still capable of recognising that things exist in themselves, which is to say, independently of any experience we may have of them (Piaget 1953a, 1955). Indeed, intelligence is not capable of not recognising this fundamental property of objects. And as Piaget’s own analysis demonstrates, it is both one of intelligence’s earliest constructions and all-pervasive in its scope and consequences.
Furthermore, it is only insofar as we construct things and events as objects that we can know them intelligently. Even things that are normally treated as non-objective, such as mental images and opinions, exist for intelligence only to the extent that they are formulated as conceptual objects. Only when we thus detach ourselves from them can they be dealt with, or even properly recognised for what they are. Only as objects can specifically intelligent operations such as identification, analysis, evaluation, comparison, ordering, valorisation, planning, and so on be applied to emotions, bodily states, and other affects. Indeed, although Piaget himself did not address this issue clearly, my own view is that consciousness is object permanence, and vice versa.
More generally, for an organism with object permanence, ‘out of sight’ is definitely not ‘out of mind’. Without object permanence, it is impossible to recognise that hidden things are still ‘there’, that things ‘go somewhere’ when they disappear, that reality is not defined by appearances, that there are different points of view, that there is a past and future, and so on. Conversely, the implied existence of things despite their phenomenal absence makes a huge variety of human phenomena both literally and metaphorically thinkable, including not only individual consciousness and cognition but also cultural and technological artefacts and social systems more complex than face-to-face interactions (Robinson 2004). Object permanence also engenders various forms of strictly cognitive (as opposed to perceptual) conservation, implication and invariance. But permanence is also expressed in the recognition that objects possess a fundamental ‘otherness’, with autonomous powers. Hence the growing recognition even among babies that things can be used as instruments, tools, supports, screens, weapons, doorstops, and so on, or among adults that the correct lift-drag co-efficient will support powered flight. On the other hand, if objects harbour their own potential causality and effects, they need to be watched out for and even wary of, even if we don’t have any use for them ourselves.
So object permanence entails more than either external perception or internal representation, which would not be enough to differentiate intelligence from its non-intelligent cousins and ancestors. Or rather, intelligence’s internal representations represent things as existing independently of the organism that represents them, or of the representation itself. Images, thoughts, memories, may all be treated as objects – were that not the case, entire branches of philosophy and psychology would be impossible. Later they develop into both reflexive awareness and higher-level insights into – and designs for – still more abstract relationships and systems of relationships, which can be apprehended and worked on without reference to any concrete object. Hence the possibility of, amongst much else, logic, theory, philosophy and theology. From a strictly biological point of view, internal representation need never go so far, being quite plausibly limited to mediating and informing outward action and experience by more or less complex internal framing effects, models, mediating states, cognitive maps, and so on. These may subtly regulate the organism’s actions and experience without needing to extend as far as recognising the objective reality of things or their hidden structural and functional depths, let alone a reflexive awareness of either the self or of the very process of comprehension, with all that that would entail.
It is not hard to imagine selective pressures to which evolution is most unlikely to offer an effective response but which are amenable to object permanence. To take the single most famous event in the whole of evolution, the ability to recognise that a very large and rapidly approaching asteroid threatens one’s very existence and then do something about it assumes a number of abilities. Unfortunately none of them is intelligible in normal biological terms, but all of them come quite naturally once one has mastered object permanence.
Firstly, the entirely stochastic and pragmatic logic of natural selection makes it extremely unlikely that earthly organisms would ever adapt to a threat that typically arises only once every few tens of millions of years (French 1998: §2.3). So an organism that depended solely on natural selection for its defence against asteroids is unlikely to have picked up the knack for seeing such things as threats, or indeed seeing them at all. Even if (somewhat bizarrely) treated as a generalised version of a more earthly problem (perhaps as a ‘looming’ stimulus), it is incredible that natural selection could have refined such an understandable capability to the point where it encompassed the detection or avoidance of plummeting asteroids. Selection is driven not only by the maximisation of benefits offered by adaptations but also by the minimisation of the costs they incur, and it seems wildly disproportionate to invest any resources whatsoever in such extravagant talents – even if one could imagine a sequence of selections by which they could be created. Conversely, what are we to make of the human ability to spot not only incoming asteroids but also quanta and quasars, not to mention speculating on the fate of dinosaurs that have been dead for 65 million years – all without any evident addition to either our biological natures or our reproductive fitness? In fact we spend a good deal of our time speculating about such things and their even more speculative reasons and implications. Were it not so, ideology and science would be equally unthinkable, not to mention entire literary genres from Revelations to science fiction.
Secondly, assuming that one recognises the fact and significance of an approaching asteroid, how exactly is one supposed to deal with it? It is not a threat that can be dealt with by running faster, biting harder, growing a thicker carapace or developing a more robust digestive tract. The only practical solution is extremely advanced technology. Shooting down asteroids really is rocket science, and would certainly demand vast conceptual sophistication, immense chains of means-end linkages, innumerable integrated technologies, organisations that coordinate millions of individuals, and immense (though evidently not inconceivable) political, economic, ideological and scientific forces. Although object permanence is far from all that is required for such an enterprise, it would be quite impossible without it.
Of course, not all evolutionary scenarios are quite as theatrical as incoming asteroids, though many are almost as dramatic in their impact on those concerned. For example, a single pleiotropic gene may affect many different and perhaps unrelated aspects of the phenotype. This makes it extremely difficult for selection to ‘decide’ whether or not the gene in question should be expunged or promoted: its ‘bad’ effects, severe though they may be, may still be outweighed by the ‘good’. In sickle cell anaemia, the heightened immunity to malaria conferred by heterozygosity outweighs the risk of fatal genetic illness in homozygotes. But how much better it would have been for human beings, even from the point of view of reproductive fitness in its narrowest sense, if its victims had the one without the other. Yet it is unlikely that evolution will create any such solution.
It is important to recognise what the problem is here. In cases like this, extricating genetic causes from phenotypic effects is not simply complex or difficult. Natural selection cannot even ‘see’ sickle cell anaemia’s ‘bad’ effects in isolation from the ‘good’. The distinct phenotypic effects are effectively fused together by their genetic connection, and the fact that selection works only on complete organisms presents evolution with an undifferentiated, and on average perfectly acceptable, result. By contrast, although we don’t yet know how to re-engineer the human body to eliminate the anaemia while preserving the resistance, it is not impossible in principle, and the residual difficulty is certainly not attributable to intelligence not being able to recognise what the problem is, how it might be solved or indeed that the problem exists. Evolution, alas, is capable of none of these. And of course, we are already able to treat many of its symptoms (through surgery, infection control, analgesics, nutrition, hygiene, rest, support groups, and so on) in ways that probably enhance reproductive fitness – but no thanks to evolution.
Like all aspects of intelligence, object permanence develops through a succession of stages. The initial, sensorimotor object is limited to its global existence as an undifferentiated ‘thing’: it exists or not, has gross perceptual and behavioural properties (colour, weight, speed, and so on) rather than relationships, dimensions, and so on. As a result, sensorimotor object permanence probably contributes very little to the infant’s individual adaptiveness, which is assured primarily by its caretakers. However, during the subsequent ‘pre-operational’ period, the child’s activities progressively organise its objects into various kinds of grouping, classification, series and other patterns. As the great apes, African grey parrots and dolphins who also achieve this level of development show (Marino 2002; Parker and McKinney 1999; Pepperberg 1999), this is enough to permit individual survival in the right conditions. The next major stage, ‘concrete operations’ (Inhelder and Piaget 1964), when intelligence is capable of a level of insight, coordination and control that is unimaginable in non-intelligent organisms. These more or less discrete and isolated patterns are then expressed, objectified, and re-cognised in relation to one another until they are synthesised into still more highly structured systems – an outcome that initiates the stage of ‘formal operations’ (Inhelder and Piaget 1958). Impressive enough in individuals, taken collectively formally structured intelligence is capable of the advanced economic and political systems and global cultures and technologies most of the readers of these words inhabit, and without which these words would never have been written (Robinson 2004).
All these higher structures lead, on the one hand, to intelligence’s recognition that activity, objects and the world at large are informed by abstract structures, and on the other, to the ability to control these structures, notably by embodying them in tools, resources, institutions, systems, and so on. They also allow intelligence to impute rational force (necessity, authority, legitimacy, value, propriety, obligation, and so on) to rules, ideals, money, duties, contracts, institutions, orders, roles, justice, and any number of other abstract entities that have no biological counterparts, but have extremely broad and deep implications for human cultural and technological activity, and so for human ‘adaptation’.
As I have already implied, object permanence is not a peculiarly human capability. After all, we possess object permanence by virtue of our intelligence, not because we are members of the species Homo sapiens. Any intelligent species (or non-biological entity) would be capable of as much. Although the evidence is not wholly convincing in all cases, some degree of object permanence has been demonstrated in primates (Parker and McKinney 1999), perhaps even as far as marmosets (Mendes and Huber 2004), dolphins and a number of birds (Marino 2002; Pepperberg et al. 1997; Pollok et al. 2000; Zuca et al. 2007). However, there is no evidence for it in mammals and birds generally (Kirkpatrick 2001; Vauclair 1996; Young et al. 2001), and in no non-human species has it reached anything like the level found among human beings. So its impact remains minimal, and only human beings exhibit the cultural and technological efflorescence advanced object permanence naturally supports.
Plainly object permanence is not simply a minor curiosity of human infancy. It is possible that cultural and technological developments over the next few centuries will make it all but impossible for nature to throw up any form of variation or selection that could modify human nature in directions we were not willing to accept on strictly rational grounds. Indeed, we may soon be able to do a great deal to improve on nature’s rather poor designs. For all their undoubted wonders, random variation is entirely lacking in insight and natural selection is relentlessly pragmatic, eclectic and backward-looking, not to mention willing to make do with the least bad solution. It is certainly hard to imagine that intelligence would have spent three billion years creating such a flawed and vulnerable result as the human anatomy. Fortunately artifice has always been our second (if not first) nature: it is the logical corollary of our capacity for looking at the world in terms of what is (or could be) objectively there rather than a range of stimuli to which we happen to have been attuned by evolution.
Of course, even if we can transcend biology in this way, we may not escape evolution altogether. Whatever the originality of our means, the ends to which we apply our intelligence and the motivational systems that shape our activity may still be profoundly shaped by reproductive fitness. However, although it is likely that small residual pockets of this kind remain, it is doubtful how significant they could be. An intelligent being is after all no more simply a ‘standard’ organism plus a little cognition than organisms in general are chemical reactions glued together with a smattering of ‘life’. Precisely because in non-intelligent organisms motivation, action, experience and goals are tightly interwoven in the body, the dissociation and reconstruction of the ‘techniques’ of biological activity (Piaget 1971a; Robinson 2005) into the neonate’s sensorimotor reflexes, from which intelligence arises, must lead to complementary changes on its ‘affective’ dimension: goals, motivation, drives, feelings, and so on.
So, just as organisms still consist of, yet are neither dominated nor explained by their chemistries, so one must ask whether the emergence of intelligence leaves human beings still organisms, but not exclusively or even primarily so. Certainly the assumption that intelligence is dominated by biological drivers such as adaptation or fitness may be contested. It is not that we have overcome the need to reproduce, but our intelligence certainly throws into question just what we do once reproduction has been achieved. Very few of the more sophisticated aspects of human activity and experience – art, philosophy, music, religion, science, and so on – seem to have much to do with reproductive fitness. There is no shortage of suitably convoluted arguments as to how these tings may have remote reproductive consequences; however, these generally seem to be motivated by the need to find an contribution to reproductive fitness in everything – a need that falls away if intelligence really is as different from life in general as a chemical reaction. In any case, as soon as it is recognised that any innate urges and dispositions that may inscribed in our bodies and brains may themselves be open to intelligence’s natural propensity for objectification, analysis, evaluation and modification, it becomes difficult to exclude any area of human existence from future intervention.
If this seems to defy the basic logic of evolutionary theory, then so be it. Human beings have long since learned to see into the infinite, the infinitesimal, the vasty deeps and the invisible spectrum. It is hard to see such abilities either as determined by or limited to adaptive functions or as explicable in terms either of reproductive fitness or of any other narrowly biological driver. It is especially incredible to regard such a massive efflorescence as a by-product of other, more directly advantageous adaptations: evolution is surely not so beneficent as to allow such expensive but biologically useless systems to be created, let alone sustained and elaborated indefinitely. Given that it may well be hundreds of millennia since human life included ‘objective’ forms of activity (Wynn 1979, 1981), it is hard to see how evolutionary theory could possibly explain it. Or perhaps philosophers, Latin teachers, evolutionary theorists, radio astronomers and the rest are having more children than the rest of us, and just keeping quiet about it.
So object permanence is a good deal more than simply a superior adaptation. By recognising that objects exist in themselves, independently of our relationship to them, we can have a knowledge of and an interest in them that extends beyond any strictly biological functional interest in, say, feeding, shelter or social relationships. Object permanence is not even limited in the same way as a wide-ranging sense such as the visual system. As ‘blindsight’ and other neuropsychological puzzles demonstrate, the systems that underpin human vision are fragmented (Weiskrantz 1997) in a way one would have thought anathema to the integration implied by object permanence, and recent hypotheses regarding the modular structure of the human nervous system tend in a similar direction. But human vision is also subordinated to human intelligence, such that the scope and range of our vision is defined not by human biology but objectively, which there is to say, by what is to be seen. That is why we strive to extend our visual capacities through forms of technology (spyglasses, spectrometers, space-born telescopes, and so on) and culture (geometry, aesthetics, etc.) that are as indifferent to the specifics of human biology as they are irrelevant to reproductive fitness.
That is not to say that such things have no bearing on reproductive fitness. But biological variation and selection are certainly not the process through which they are actually constructed or incorporated into our activity. Conversely, it is only by virtue of our intelligence that human vision can be put at the disposal of any form of activity that may objectively be able to make use of it, rather than being limited to specific structures and functioning, as non-intelligent vision invariably is. So the notion of multiple fragmented visual systems, which is entirely appropriate to a certain level of human biology, is wholly irrelevant to understanding human vision considered as a function of our intelligence, and in particularly the cognition of object permanence through which the perception of various fragmented patterns is unified into a single point of view, indeed a single world-view.
The key issue here is the nature of the structures created by object permanence. The way a non-intelligent organism relates to its environment is inseparable from its biological interests, the nature of the organs through which it is engaged, the inner systems, dispositions, drives and biological functions that motivate it, the stimuli to which it is attuned, and so on. Given that these determinants are constantly changing – not least as the result of the organism’s own activity – these relationships are inherently non-permanent. Conversely, because the relationship between them is fluctuating, a non-intelligent predator may be capable of either interest or uninterest in its prey, but not of disinterest. Disinterest presupposes the ability to relate to something independently of the nature that relationship, of conditions in which the relationship arises and any specific functional interests one may have in it. This capability is the natural and logical corollary of the ability to treat things and events as objects, so even when generalised motives like curiosity or gratuitous exploration intrude, if they are not accompanied by object permanence, they remain internal drivers, and do not require even a rudimentary objectivity.
Conversely, although an immature or inexperienced intelligence’s relationship to its objects is always to some extent egocentric, one-sided and conditional, the relationship itself always implies that the object exists independently of our relationship to or interest in it. In other words, being capable of object permanence may make you intelligent, but that does not mean that you are very good at either. Conversely, the forms of experience we share with non-intelligent organisms – perception, belief, imagery, memory, stimulation, and so on – may be inherently fallible and so potentially distorting to reality according to the (intelligent or non-intelligent) organism’s state, disposition, perspective, and so on; however, when reshaped by object permanence they become, if not reliable in themselves, then certainly open to forms of rational insight, criticism, regulation and development that allow us to distinguish between what is reliable about them and what is not. Leaving aside the vexed question of whether we are actually capable of complete objectivity (or, a fortiori, truth), merely aspiring to objectivity sets intelligence apart. It is certainly on this aspiration that science rests. So even if Kant and Hume were right and we never manage to track the object to its ontological lair, the fact that we can dream of such things makes the dreamer a very different kind of creature from one that cannot. For example, only a creature capable of object permanence could dream up the theory of evolution, or question whether the latter applied to organisms in full possession of the former.
What general conclusions can be drawn from all these implications of object permanence? Firstly, Wallace was right: intelligence can pre-empt and even actively manage evolution (which, thanks to Wallace and Darwin, we can now construe as a theoretical object). Presumably not even the most ardent adaptationist would doubt that we might manage to deflect a threatening asteroid, even though this ability seems to resist any sensible explanation in evolutionary terms. But secondly, this difference represents more than simply a quantitative improvement in adaptation. Intelligence transforms the very logic of activity and development. In evolution, the functioning of a given organic structure is modified by (functionally) random variations, and then the functionality of the resulting structure is judged by selection for its impact on reproductive fitness. Leaving aside the complex interplay between selection at one level and variation at another (e.g., Maynard Smith and Vida 1990), the ways in which certain kinds of structure limit and channel the scope and opportunities for either (Kauffman 1993), and the somewhat simplistic quality of much neo-Darwinian reasoning (Dover 2000; Oyama 2000; Oyama et al., 2001; Rose and Rose 2000), it remains fundamental to biology that the two sides of this process are not only logically distinct but also materially decoupled.
For example, many forms of niche construction substantially alter how selection operates. Termite mounds radically restructured the termite environment to the point where one of their own creations became perhaps the single most powerful factor in their subsequent evolution. But the termites do not recognise this fact – the nest has no objective reality for them – and it is not a lesson from which they can profit further except through yet further biological adaptations. However triumphant this adaptation may have been from the point of view of the termites’ operational sophistication and reproductive fitness, they are completely blind to it, and cannot see (in any sense, no matter how metaphorical) how this change in their functioning has affected their functionality.
The effect of object permanence, by contrast, is to abolish this distinction between the functioning of a structure and its functionality. The ability to objectify the niche and environment they occupy, their own bodies, selves, methods, goals and values, the nature, processes, mechanisms and consequences of adaptation, and all the other factors involved in their activity and experience, allows intelligent organisms such as human beings to recognise the forces of evolution and redirect them to their own benefit. They can appreciate the difference between what things are and what they should be – the kernel of the difference between a structure’s functioning and its functionality – without every ‘should’ being defined in terms of reproductive fitness. This process of evaluation increasingly is increasingly performed by cultural and technological systems that actively and explicitly assume, apprehend, analyse, evaluate and intervene in the relationship between our functioning and its functionality, and so increasingly ensures that the normal forces of evolution cease to intrude in human affairs.
Of course, all adaptations are internalised responses to past selective pressures, and in that sense they too pre-empt future selection. A termite (or bird or fish) nest internalises an adaptive response to weather, gravity, predators, and so on (von Frisch 1974). But intelligence does much more than internalise particular selective pressures. Firstly, the ability to look at things objectively allows intelligence to determine the nature of that thing, and to identify the possibilities objectively inherent in it. This is a very different thing from generalising from empirical experience. For example, it means that intelligence can do something about the problems an objet may pose before they arise. Secondly, intelligence is not specific to any particular circumstance: it can apply the same general abilities to solve infinitely many similar problems. Finally, as I have already argued, there is nothing random (in any sense) about its solutions: they are specifically engineered to solve the specific problems by which they were originally inspired. Unlike either natural selection or the non-intelligent organism, intelligence knows what it is trying to do. Of course, it may misconstrue the problem and its solution may not work, but the non-intelligent organism is simply precluded from even trying to solve problems in this active, specific and universal manner. Perhaps that is why Heath Robinson and Rube Goldberg are so amusing: yes, their fabulous contraptions work, but who in their right mind would have made them that way? Well, evolution would have. But then evolution hasn’t got a mind, let alone a right one.
Finally, not content with making all forms of variation directly manageable, object permanence also engenders all sorts of new kind of ‘variation’. The most obvious from a biological point of view are medicine and genetic engineering, but we have been consciously applying artificial selection to ourselves and our environments for many millennia. The process of selection, insofar as it is taken to mean the exogenous sifting of results, has also long since been replaced by intelligence’s own preferences and purposes. Of course, we will, as the rules of reproductive fitness dictate, have to maintain ourselves in suitable numbers if we wish to carry out our own plans. But of what level of matter – biological or non-biological – is that not true? Our technology must also obey the laws of physics, but the laws of physics hardly explain why Lord Kelvin invented his eponymous scale. So in what sense does that the reproductive fitness of human existence imply that our lives are ‘ultimately’ regulated by the forces of evolution?
5. The development of the subject
Assuming that we do not foul our own nest irretrievably – a large, dangerous and increasingly improbable assumption in the present point in history – there seems to be no evolutionary force organisms armed with a high level of object permanence could not in principle deal with, especially when acting collectively. Quite probably the traditional view, that every species is doomed to extinction, applies to advanced intelligent species only in the sense that they might inadvertently commit collective suicide. But this is not a totally convincing argument for intelligence’s independence of evolution. There may be other biological constraints on intelligence that leave deeper aspects of evolution beyond even object permanence’s reach. They are hard to imagine, though not, as has often been argued (e.g., McGinn 1989), because the limitations imposed on (or by) our intelligence themselves prevent us from recognising them. As almost our entire culture and technology demonstrate, object permanence situates intelligence directly and massively in the world of things and events that exist independent of our awareness of them. Were fundamental biases, shortcomings and lacunae built into our intelligence, they could not be concealed for long. Instead, they would express themselves in conflicts and contradictions not only in our awareness of things and events but also in the objective reality we try to create. At the simplest level, our actions would fail and the things we make would not work. At the highest level, reality would seem to contradict itself (or, at the very least, contradict us). Far from intelligence’s intrinsic limitations concealing the fact and nature of our shortcomings, object permanence and its massively practical consequences would make them unavoidable.
But in any case, Piaget’s theory implies (though Piaget himself never stated) other proofs that evolution can no more modify intelligence than it can change the effect of pouring acid onto metal. However, these proofs relate to two other, quite distinct dimensions of intelligence. The way intelligence’s world transcends any possible biological environment will be addressed in the next section. In this section my concern is with the internal structure of intelligent activity, which, for want of an accepted scientific term (though very conscious of its implications), I shall call the subject.
In Piaget’s model, the internal structure of intelligence develops through the expression, objectification, alignment, coordination and re-internalisation of prior internal structures. As developmental stages succeed one another, the infant’s initial sensorimotor reflexes (such as visual tracking or auditory localisation) are articulated, embodied in objects, and then the combined patterns of activity to which this leads allows the original reflexes to be re-synthesised first into concepts and then into higher-level relationships, systems, and so on. These developments range from piecemeal and parochial extensions and corrections to a sequence of global stages, each signalling the emergence of a qualitatively new, systematically more powerful subject. Each new level of subjectivity not only combines the talents and knowledge provided by its predecessors but also abstracts from them a new level of capability, on the basis of which wholly new types of activity are made possible.
Thus, the young child constructs primitive empirical concepts (such as ‘dog’) by synthesising the reflexes (visual-motor tracking, auditory localisation, tactile-motor, and so on) that have been successfully applied to dogs into a single, multifaceted structure that permanently embodies the notion of dogginess. Once created, such a concept can operate quite independently of the child’s experience of actual dogs – they can talk about dogs, draw a picture of a dog, tell you some of the more superficial differences between a dog and a cat, and so on. Furthermore, being synthesised from the infant’s own activities, it is neither a direct summation nor a simple generalisation of empirical experience. On the contrary, empirical experience provides the opportunity to develop, validate and elaborate such structures, but it can never define them exhaustively. Still less can empirical or functional experience alone account for the ‘logico-mathematical’ concepts that synthesise the structures of activity as such (Piaget 2001).
The abstract character of concepts becomes clearer with a higher level empirical concept such as ‘mammal’. No one has ever observed a ‘mammal’ as such – or for that matter ‘gravity’ or ‘natural selection’. Conversely, because they flow from the activity whereby is or that type of object was constructed (as opposed to simply externally sensed – whatever that would mean), they internal connections between the different aspects of that individual’s activity naturally inform the subject – rightly or wrongly – that certain properties and relationships will necessarily inhere in any empirical exemplar of that concept, even when they are not – or indeed cannot be – observed directly; that radically dissimilar entities that to which the same structure of activity can be applied may share common properties regardless of their empirical differences; that empirically diverse appearances may disguise instances of the same thing; and so on. If these inferences are false, then the effect of applying this new organisation of activity will be to generate a contradiction between the expected and actual results of activity.
But this is not simply failure. Because intelligence operates solely in terms of independent objects, when it gets things wrong it cannot avoid being presented with the direct expression of its own failure – the object it actually constructs as opposed to the one it meant to construct. What is more, the specific nature of the failure, as expressed in the objective (but faulty) result, provides at least indirect evidence of the nature of the original misconstruction. That is, mistakes result not only in failure but also in contradiction, and contradiction cannot simply be written off (as failure may be by a non-intelligent organism) but rather ‘proves’ to subjects who suffer it that there is something amiss with them. They made this mistake. Hence the peculiarity of intelligence, that it will puzzle at a problem because it is a problem, rather than because its solution will reap any empirical or functional reward.
Once basic concepts have been established, still higher structures provide intelligence with still more powerful instruments (relationships, systems, methods, and logic and mathematics themselves) that are even more obviously irreducible to empirical or functional actions or experience. Yet they provide unprecedented power over the empirical and functional realms. However, there is more to the subject than a growing independence of activity, awareness and authority. As both the logic of its development and the empirical evidence (outlined below) indicate, none of the arguments being offered here relies on the specifics of human biology – not on our general biological make-up, not on our specific evolutionary history, not even on exactly which sensory or motor organs, reflexes, capabilities or appetites are present at the start of sensorimotor development. Rather, the key is the cycle of expression, objectification and internalisation, which relies on entirely general principles. A good deal (perhaps all) of the relationship between infant and its inner and outer milieux needs to be governed by sensorimotor reflexes for intelligence to develop, but that seems to be all. The subject’s developmental trajectory will be just the same, regardless of whether the infant in question is a human being living on Earth with eyes and ears and hands and a rigid skeleton or a species of slug that inhabits deep space, feeds on gamma rays, senses nothing but dark matter and moves by teleportation.
Furthermore, not only will the same kinds of structure (series, classifications, systems, and so on) be formed regardless of the specific biological substrate or sensorimotor structures from which they originated, but their development necessarily culminates in a single overall structure that synthesises all aspects of intelligence (Robinson 2004). In other words, because the logic of the developmental process described above is not context-, content- or even species-specific, subjects are always and everywhere the same. But if the development of the subject is thus a universal process that can have only one outcome, then variation and selection can neither cause nor affect the nature of subjectivity. Evolution may stunt or hinder its development, or even destroy it completely, but it cannot explain either its nature or its development. Yet both the development of such subject and its outcome are entirely material: there need be no recourse either to Wallace’s ‘Intelligence’ or to a more contemporary Intelligent Designer.
This interpretation of subjectivity is strongly supported by the available data. For example, grossly physically handicapped infants children lacking normal sensorimotor capabilities follow the same developmental process, sequence and outcomes as the non-handicapped (e.g., Bebko et al. 1992; Eagle 1985; Gottesman 1973; Gouin Décarie in Foss 1969; Kopp and Shaperman 1973). Nor do massive differences in culture, technology and social organisation affect intelligence’s nature or developmental stages, though they can influence how fast or in what empirical and functional directions it develops (Hallpike 1979; Robinson 2004). Perhaps most convincingly of all, different species appear to develop exactly the same kind of intelligence and develop it through the same stages. Although definitive evidence is lacking, if one takes into account related areas such as tool-making, true imitation and self-recognition, it seems that certain primates, dolphins and birds are intelligent in exactly the same sense and develop their abilities in exactly the same way (e.g., Parker and McKinney 1999; Pepperberg 1999; Bauer and Johnson 1994; Reiss and Marino 2001).
Evolutionary psychologists may be tempted to reply that the uniformity of the developmental process and stages Piaget identified are evidence for their control by species-wide genetic factors. But even the most rigid genetic straitjacket (whatever, in the face of developmental systems theory, that may mean – Oyama et al. 2001) leads to far more vagueness and imprecision than would ever be countenanced by a typical product of intelligence such as, say, mathematics or bureaucracy. This is not because intelligence is somehow perfect or ‘above’ matter; rather, it has functional and structural principles built into it that oblige it to act with increasing rigour and precision, clarity and exactness. Indeed, if this were not the case, then it would be impossible to articulate such criticisms with anything like the rigour on which science itself absolutely relies.
Conversely, it has been argued that Piaget’s developmental sequence is not a material process at all, but only the sequence that follows from analysing the demands of development process into a logical order. But far from being a criticism, it is this very fact that brings out the most striking feature of Piaget’s developing subject. There is no way of applying such a sequence to the development of, say, a butterfly. It too passes through a series of qualitatively distinct stages, but even the most superficial comparison with Piaget’s ‘logical’ sequence will prove the latter to be quite irrelevant. Yet the development of any intelligent structure – and this claim extends far beyond merely the individual (Robinson 2004) – corresponds to Piaget’s programme so exactly that its appropriateness as a model is self-evident. And this, I would argue, is because the liberation of the intelligent organism from biological constraints ensures that the only avenue left open to its development is indeed to follow the logical sequence for the organisation of activity as such.
Of course, any intelligent being may be more interested in kinds of object that reflect its biological character: an object that fascinated you or me might leave a bonobo, an Alpha Centaurian or an intelligent machine completely cold. But if one starts from the opposite point of view, from the abstract rules according to which any of these subjects constructs any and every object, then it is difficult to escape the conclusion that, at that level, subjects everywhere are increasingly equivalent, and finally identical. They all seem to construct (produce, act on, organise, define, analyse, transform, and so on) objects in terms of the same universal types of structure.
Not all intelligent species develop to the same extent, of course. As one might expect, as it reaches its species-specific limits, development tends to break down in species-specific ways (e.g., Antinucci 1989; Langer 1998, 2006; Parker and McKinney 1999). If human and non-human intelligences do eventually part company, this seems to be not because our respective intelligences are qualitatively different but because the limited developmental capacity of, say, a gorilla prevents it from developing beyond a certain level. Almost until the end, our paths are otherwise very similar. A chimpanzee has much greater developmental potential than a gorilla, so our respective developments parallel one another for much longer. Conversely, monkeys’ developmental trajectories diverge markedly from both, so it is no surprise that (as far as current evidence shows) they never achieve full object permanence. To the extent that we have any real insight into cognitive development in dolphins and birds, they too follow the ‘normal’ track for intelligence as such.
If true, this is extraordinary. Primates, birds and dolphins differ not only in their innate sensory and motor organs and capabilities but also at every level of their nervous systems from gross anatomy to detailed cytoarchitecture. Our latest common ancestor with dolphins died out almost a hundred million years ago, and almost certainly exhibited zero intelligence (Marino 2002). With birds, the evolutionary distance at least doubles. At first sight convergent evolution seems a plausible solution, but convergence normally assumes a variety of structures performing the same functions: bats, birds and bees all fly, but have radically different types of wing. However, both the uniformity of the stages through which all intelligent species develop and the consistent nature of the unforced errors committed by undeveloped individuals reinforce the idea that there is only one structure that can be called intelligence, and it pops up everywhere. It is as though leathery bat-wings – and only bats-wings – had evolved among insects and birds as well as mammals. Given the lack of a shared intelligent ancestor, the apparent structural identity of intelligence among dolphins, parrots and primates is inexplicable by any standard evolutionary argument (though see Robinson 2005 for a hypothesis). However, as I hope my preceding argument makes plain, on the ontogenetic level it is perfectly explicable: because the relationship between intelligence’s developmental stages is one of increasing abstraction from the organism’s initial structures and propensities, not accumulation of empirical or functional content and context, the nature of the higher stages is determined not by ancestry but by increasingly general principles of organisation that apes, humans, dolphins, birds, Alpha Centaurians and intelligent machines all share.
This can be illustrated with innumerable examples. As I have already argued, object permanence itself arises not out of empirical or functional experience but from the infant’s synthesis of its original sensorimotor reflexes, and is inherently independent of any particular empirical or functional content or context. Indeed, object permanence as such is not ‘experienced’ at all. Rather it expresses the mutual implication of the various forms of sensorimotor activity. Of course, the activity that led to this development had empirical and functional content and context, and object permanence has massive empirical and functional consequences. Nevertheless, object permanence as such relies not on specific-specific patterns of sensation or motricity, perception or behaviour, appetites or drives, but on the coordination, synthesis and abstraction of principles shared by whatever reflexes are present.
Later, the concept of number arises from a synthesis of the ability to classify objects with the ability to organise them into series (Piaget 1952). Again, it does not matter which specific classifications and series the subject starts from: what counts is the synthesis of classification and seriation as such. As a result, one would expect the same number concept to arise in any species that was able to develop its intelligence to the level where classification and seriation can be merged. Or again, the diverse history of number systems is well documented (Damerow 1995; Hallpike, 1979: 240ff; Ifrah 1998; Nicolopolou in Cole et al. 1997: 212, 219; Nissen et al. 1993), but the presence of innumerable pragmatic local variants does nothing to undermine the universal themes the same history also reveals (Robinson 2004).
This process of abstraction has precedents far back in the history of matter, and in a sense intelligence only recapitulates how each previous level of matter emerged from its predecessor. Chemical structures become possible only when the many conflicting physical forces were first synthesised into the ‘standing contradiction’ of protons, electrons and other particles we know as the atom. For atoms not only stabilise the relationships between these physical forces but their interactions release a new, specifically chemical level of activity. Organic structures then emerge when molecules are organised so as to subsume, and so negate, the normally mutually disruptive reactions of molecules through their transmutation into the standing contradiction between anatomy and physiology we recognise as life. And finally, the generally inconsistent, particularistic, fragmentary and rigid sensory and motor capabilities that typify all animal life are synthesised into the new standing contradiction of abstraction and concretion we call intelligence (Robinson 2005). In a classic dialectical supersession, intelligence achieves its universal capacity for positively comprehending its world by not only synthesising but also subordinating all its specific positive skills and experience.
In each such case, the new, higher form of matter achieves that higher form by abstraction – which is to say, by nullifying the conflicts inherent in the lower, and so achieving new levels of both internal integrity and external effectiveness that allow it to co-opt or suppress further changes instigated from the lower level. Of course, its power to do so is limited: atoms are smashed, life dies, reason dissolves into madness, senility and ideological rigidity. But in each case, each new level of structure reveals new forms of functioning that seem to inhere in the new level rather than in any particular concrete component that level may include. Thus, just as any physical entity falls under gravity without having any special organ or propensity for falling, so all life shares the same principles of adaptation, heredity and epigenesis, yet these are nowhere inscribed in the organism. Of course, a great deal is inscribed. But not everything, and especially not the most abstract, powerful and pervasive forces of all.
So abstractions can be material, for abstraction is nothing but the reorganisation of existing material structures and their functioning. Furthermore, because each abstraction is a synthesis of the normal functioning of lower level structures, there is no inherent contradiction between the higher and the lower. As a result, although intelligence’s own most abstract achievements (such as logic and mathematics) are so abstract that they often seem irrelevant to practical matters, they can be validly applied not only to any possible expression of intelligence but to all forms of pre-intelligent matter too. Hence also the succession of cognitive stages, each constructed by this same process of abstraction (Piaget 2001), without the intervention of any innate ‘programming’ or specification of which stages will arise or what particular kinds of activity or experience each new stage will unleash. Of course, intelligence always has a content and context of some kind – otherwise it would not be real. It is always doing something. But unlike non-intelligent organisms, there is no specific content or context to which many products of this process of abstraction (such as logic or mathematics) is limited or which it cannot transcend.
This in turn leaves us with an ever widening and deepening grasp of the general forms of activity, up to and including activity as such. These forms are then applicable to any possible content or context because they synthesise not only all their precursors’ specific forms but also (with a little familiarisation and exercise) all other forms of that general type. This is why human infants can easily assimilate non-human modes of sensory experience (Bower 1989), some great apes can understand functional equivalence (Russon et al. 1998), children can learn rules for conjugating French verbs, concepts such as democracy or equality before the law make any sense or the reader of this essay can evaluate its argument without having had the experiences from which such ideas were originally derived.
This must seem a very bold claim. If we know anything about organisms, it is surely that they are subject to any number of species-specific predispositions, constraints, quirks, appetites, closures and limitations. And, considered strictly as organisms, it must be accepted that this is equally true of intelligent species. In particular, intelligence is instantiated above all in the human central nervous system – an indisputably biological structure. So surely it too must be subject to evolutionary pressures? Certainly, yet intelligence’s freedom from innate, functional and experiential constraints and propensities that is conferred by the process of abstraction is equally fundamental.
Surely this is a flat contradiction? Surely our existences cannot be determined by both the vagaries of biology and the rigours of intelligence – or at least no in connection with the same aspects of that existence. What is more, the dilemma could hardly be more fatal to the present argument, given that it seems to imply that we must accept either that the structures Piaget describes are instantiated in the organic structures of the central nervous system or that they are purely theoretical abstractions devised to support laboratory experiments and school tests. If the former, then they are necessarily subject to evolution at some level (not to mention every other biological vicissitude), and if the latter, then they are only not subject to evolution in the trivial sense that no theoretical entity is subject to evolution – true but wholly irrelevant to the question of whether intelligence itself evolves.
It is tempting to reply that this is a kind of ‘category error’ – that my hypothetical critic is making essentially the same mistake as any reductionist who asserts that, as all organic processes are implemented exclusively in chemical pathways, they can all be reduced to chemical terms. The qualitative changes introduced by the formation of a higher-level organisation (on which the present argument also rests) are thereby disregarded. But as Hegel argued, to claim to reduce the higher to the lower is simply to attribute the properties of the higher to the former. So if there is a chemistry – and only a chemistry – of consciousness, does that mean that the chemical in question are conscious ($$)? Or is the chemist who makes this claim also claiming to have made it unconsciously?
Unfortunately, the qualitative novelty of intelligence is not so well established that such criticisms can be answered by analogy: there is a long, long queue of eminent authorities only too happy to deny that explaining intelligent functions in neural terms would be a reduction. However, there is an alternative that preserves the present argument.
The proposal is that significant aspects of intelligence are embodied in the brain and that the brain, like any biological structure, is susceptible to evolution; therefore must intelligence itself be. This proposal assumes that intelligence is the product of evolution. But this is not the case. It is true that intelligence’s precursors are produced by evolution, but the sensorimotor infant shows no signs of intelligence proper. Rather, the formation of intelligence itself occurs in the course of the individuals own development, is implemented solely by virtue of its own activity and it subjected to validation only in the course of that individual’s own actions. So both the creation and the continual ‘maintenance’ of intelligence lie outside the realm of evolution proper. What is more, the basis of the formation of intelligence is a process of abstraction that progressively divorces the structure of that intelligence from its specific beginnings. Not even Piaget’s detractors (recently Mandler 2004$$, Baillargeon$$ and others) claim that intelligence enters the world fully formed. The widespread (and generally unsubstantiated) enthusiasm for innate information processing faculties over Piaget’s sensorimotor reflexes notwithstanding, there remains little doubt of intelligence’s irreducibly ontogenetic and functional origins and character.
At the same time, intelligence’s self-validation – which is to say, the method it uses to check itself against its own ideal form – relies not solely on the brain but also on its own external objects. For example, by the time it reaches a truly formal level, intelligence is inherently systematic and explicit as to the validity of its own formal structure (Inhelder and Piaget 1958, Robinson 2004). By that stage, all its expressions (such as formal logic, bureaucratic systems, the rules of chess, theories of the evolution of intelligence, and so on) themselves embody basic models of formality, validity, systematicity, and so on. And it is against these models – these ‘objects’ – that formal operational activity is implicitly verified, for if intelligence contains errors, conflicts, omissions or any taint of irrationality, these shortcomings will be expressed in failures, gaps and contradictions in the relationship between the two sides. Of course, we frequently overlook the contradictions, but they are always available in principle, even if recognising them sometimes takes a lot of effort. That is why it is always possible to appeal to, say, conscience, ideals or just logical coherence. But no amount of effort will tell a sunflower why it is tracking the sun or a wolf why it is howling at the moon.
So it is true that, were the evolutionary precursors of intelligence to evolve to a form that did not support the formation of intelligence proper, then intelligence would be unable to develop. But intelligence itself would no have evolved into another form, any more than a runner who breaks their leg has developed a new method of running. On the other hand, given intelligence’s vast operational superiority over all possible rivals, it is hard to imagine selective forces that would prefer a non-intelligent outcome, and the more thoroughly we collectively surround ourselves with the products of our own intelligence itself (on which see below), the less non-intelligent forces impinge on any single individual. Perhaps most importantly of all, as the evidence already cited indicates, intelligence is capable of developing from any number of bases, some of them severely damaged by any normal standards, yet the intelligence that flows from even quite gross handicaps is the same.
In short, there are many things in evolution without which intelligence would be impossible, but a specification for intelligence itself is not one of them.
The process of abstraction through which intelligence is formed increasingly distinguishes the individual’s intelligence not only from its organic substrate and the individual’s capacities, sensitivities or propensities but also from its own actions and experiences. The highly abstract nature of the results explain some of intelligence’s most remarkable features. For example, it accounts for ‘the unreasonable effectiveness of mathematics’ (Wigner 1960), and why logic and mathematics can be used to construct any possible object and regulate any form of activity, be it physical, chemical, organic or intelligent. This could scarcely make intelligence more different from other organisms, as evolutionary has recognised from the first. As Konrad Lorenz observed:
There is not one organism in existence in which the mutual causal coherence of all parts is… complete…, because every organic system contains unchangeable structures which, though certainly causally influencing the form and the function of the whole, are not appreciably influenced by it in turn. (in Whyte 1968: 158.)
Julian Huxley concurred (Huxley 1947: 14-17), and in a similar vein Darwin devoted many pages on his Descent of Man to fragmentary ‘rudiments’, concluding that ‘It is, as I can now see, probable that all organic beings, including man, possess peculiarities of structure, which neither are now, nor were formerly of any service to them, and which, therefore, are of no physiological importance’ (Darwin 1901: 92-9). Haeckel likewise noted that the human body was decked with numerous ‘useless rudimentary structures’ (Haeckel 1913: 217). Many other inherent structural problems are also only too visible in the human anatomy. The unfortunate compromise between locomotion and reproduction that exposes the female pelvis to all manner of ills from vulnerable sacroiliac joints to life-threatening parturition is a case in point.
As I say, the universality, autonomy and integrity of logic and mathematics could scarcely be more different. We are capable of them not because such insights are built into our thoughts (be it via programs, modules, innate analysers, Platonic reminiscence, learning or divine spark) but because the objectification, alignment, coordination, integration and synthesis of various types of activity and object – the very process of abstraction – leads to two quite distinct yet fundamentally connected insights. Firstly, by reflection of its own activity, intelligence generates the internalised insight into universal principles of form we call logic and mathematics. Secondly, through reflection on the objects it constructs intelligence subordinates the positive knowledge each of the structures from which this process of abstraction began into the multi-dimensional awareness of substance that culminates in (among other things) science. With that, all pre-existing types of object and activity are combined, their mutual implications revealed, their mutual contradictions shorn away and yet higher syntheses initiated (Piaget 2001).
In practice, of course, even subjects nominally operating at a relatively high level of competence and performance make mistakes, misunderstand and generally fall short of both their potential and their own subjective standards. After all, the mechanisms by which all these syntheses and abstractions take place are biologically instantiated, rely on empirical knowledge and are exercised with partial knowledge in pursuit of functional goals under complicated and imperfectly understood conditions over even the most competent intelligence generally exercises limited power. We are also perfectly capable of denying the logic of a situation out of fear, vanity, self-doubt, arrogance and innumerable other psychological factors, or out of pride, interest, ideology or any number of social attachments and impulses. In fact complete rationality has often been postulated as an ideal (especially by warriors and mystics) but usually only at the cost of abandoning all inconvenient commitments. Hence the prominence in history of so many institutions from Zen monasteries to the Chicago school of economic theory. In practice, such highly formalised rationality is conspicuous only by its absence from most human existence. Yet we are always capable of applying rational criteria to our activities, however uncomfortable they may make us. In fact that is exactly what acting intelligently means.
And that is exactly the point. Whereas even the most highly adaptive but non-intelligent organism is completely incapable of guiding its actions in terms of their adaptiveness, any intelligent subject is at least capable of planning, directing and evaluating its own actions in terms of its current ideal structure, which is to say, its current level of abstraction. No non-intelligent organism is capable of acting adaptively on purpose, whereas absolutely any intelligent being can deliberately exercise the various forms of innovation, evaluation, circumspection, and so on that define these acts as intelligent. Beyond a certain quite low developmental level we are likewise perfectly capable of knowing that we are making a mistake, telling a lie, allowing self-interest to colour our perceptions, and so on. We can even (in principle) work out whether our actions are evolutionarily adaptive. Such abilities do not require any specific talent or circuitry or ‘module’ for spotting errors, deceptions, omissions and adaptations. Rather, they demand only the ability to apprehend the relationship between its own actions and their results and the rules by which it believes that it and its creations should operate – which is to say, between various kinds of object.
6. The world made by intelligence
The mass of objects the subject makes, inherits from, co-produces with and shares with other subjects is far more than simply a collection of discrete entities. The words I am writing now are created in a specific context not only of other cultural artefacts – theories, journals, disciplines – that give them a specific meaning but also of a wide range of electronic technology that makes the dissemination and debate of that meaning physically possible. Indeed, not only are all objects invariably created in specific relationships to other objects but their material independence of their producers once created and their embodiment of other physical, chemical, biological and intelligent structures – in a word, their very objectivity – mean that they spontaneously organise themselves into relationships and systems that not only develop, unite and oppose them to one another but frequently transcend any purpose their producers may have had in mind in creating them. By any account the world ‘our’ objects form is enormously more than the sum of its parts.
Living in such a world adds great range of new problems to the usual list faced by non-intelligent organisms, such as learning how to deal with random variation and natural selection more effectively. In fact, mere biological adaptation is a very secondary issue from intelligence’s own point of view – a prerequisite of and platform for building its own values, goals, systems and processes, and so for achieving intelligence’s proper goals, but no more able to explain them than the Earth’s gravitational field. Even as far as our strictly biological adaptation is concerned, our intelligence is no more limited to the individual, the brain or even groups than adaptation is to be found in the molecules that make up the body. In fact, the structure, content and context of intelligent activity, organisation and experience are increasingly determined by intelligence’s own objectified products and institutions. If the reader doubts this, I suggest that they estimate how long any significant aspect of their lives would continue without the non-organic artefacts by which they are surrounded – the buildings, the transport systems, the communications, not to mention the food and water systems by which they are fed. Here is a limited sense in which all such artefacts are adaptive, but whether their origins, operation or significance can be explained as adaptations or indeed in any biological terms at all is a much more doubtful question. The fact is, intelligence is less and less embodied in biology. One day it may have no biological basis at all.
The significance of this progression is clear. “The organisation of the higher unit does not simply interact with the external environment, it is also the agent of selection on the lower unit. To the extent that control over replication of the lower unit is required for effective interaction with the external environment, organisations must appear in the higher unit to limit the origin or expression of variation at the lower unit… In each case the lower unit may replicate, but only within bounds set by the influence of this replication on the higher unit’s effectiveness in its interactions with its external environment” (Buss 1987: 171-172). In other words, human intelligence’s progressive transcendence of human biology obliges the biological systems and processes that define us as organisms to respect, and eventually subordinate themselves to, requirements imposed by intelligence’s increasingly powerful and pervasive products rather than vice versa. As evolution has little power over most of these objective conditions and constraints, it is again excluded from many aspects of the development of intelligence. It is quite inadequate to describe this as an ‘extended phenotype’ (e.g., Dawkins 1982): the biological principles that define and determine phenotypes of all kinds are quite incapable of explaining the origins, nature, functions and operations of a bank, a nation state or the internet.
Of course, much of the structure needed to make intelligence work currently is embodied in our individual nervous systems (skills, intentions, and so on), as is a huge store of internalised experience and implications (though, as with the logico-mathematical structures referred to above, exactly where this store resides is another question). So from a strictly psychological perspective it is proper (if narrow) to describe the formation of intelligence in terms of the progressive reconstruction of the neonate’s sensorimotor reflexes, and through that the central nervous system as a whole. Yet not even subjectivity (i.e., the internal organisation of intelligence), can be regarded as bounded by cranium or psyche: structurally ‘internal’ is not the same as empirically ‘inside’. Rather, the internal organisation of intelligence – the internal structure of intelligence as such – increasingly resides in cultural and technological structures embedded in our collective world. If I were looking for the place where the most important decisions are made in most of my life, I would look not to individual brains but to great bureaucratic systems that embody the rules whereby corporations and governments come to their debatable conclusions. These systems and institutions organise activity, experience, relationships, knowledge, ideals, goals, resources, instruments, capabilities, force and much else on our behalf. To live in society we must know how to access and operate these structures, so to speak, but exactly how they do what they do may be quite hidden from view, quite beyond not only my grasp but beyond the grasp of any single subject.
In short, the world is intelligence’s substitute of the organic environment. But does it exceed the environment in the ways that subject and object transcend their counterparts?
One of the most remarkable corollaries of and subjectivity and permanent objects is the unlimited interest they give us in our ‘environment’. As the very practice of biological science constantly demonstrates, this interest is quite independent of any functional interest we may have in them. If we can appreciate the existence of things, situations and events independently of any biological (or any functional) interest we may have in them, we may (and must) also be concerned that they may have spontaneous, latent, potential and unplanned effects on our interests, quite independently of any biological relationship we may have with them. That is, in a sufficiently developed subject, objectivity confers an increasingly explicit grasp not only of the facticity of things and events by which a non-intelligent organism is confronted but also of necessity, possibility and contingency. This bring with it a potential interest in the nature of absolutely anything that may impinge on our existence in any way we deem significant, regardless of whether it actually does so right now, or may do in any foreseeable future. It also engenders a whole range of existential concerns that arise from our ability to grasp our own place in the universe, regardless of any discernable functional interest this may relate to, and so to raise questions about our own nature, status, origins, significance and fate.
Putting all these points together, on the one hand as intelligent beings we are fully capable of constructing an infinite variety of environmental niches that are primarily and directly determined by our own desires, deliberations and actions; and on the other, the specifically intelligent aspects of our concrete ‘nature’ are so little predefined by our biological heritage that human beings are equally capable of living as savannah hunters, pyramid builders, factory workers, peasant farmers or evolutionary biologists – a collection of ‘niches’ so diverse that they would surely mark their occupants out as members of different species did we not know otherwise. At the same time, we are equally able to accept, reject and remake the conditions in which we exist by virtue of the values, goals, concepts, relationships and systems intelligence itself produces, and so to increasingly directly and intentionally drive our own further development.
As ever, the contrast with organic environments could not be more stark. As Robert Brandon has suggested, biological environments can be analysed into three interacting tiers (Brandon 1990). On the highest level, all organisms have a common ‘external’ environment consisting of all entities, structures and processes that are objectively real, even though they lack any biological impact, up to and including the entire universe. Below that level, an organism’s ‘ecological’ environment comprises the fraction of the external environment that actually bears on the lives of its members, including their ability to reproduce themselves. Finally, the fraction of the ecological environment that affects reproduction differentially, and so provides a platform for natural selection, Brandon calls the ‘selective’ environment.
These tiers are by no means static or impermeable. In fact, any notable shift in development or evolution alters these tiers and their boundaries. However, intelligence seems to represent a much more radical break. If Brandon can draw such distinctions and if it is true that we can collectively modify or replace any of the systems and processes involved at any of these three levels, then for intelligence these boundaries do not really exist. Unless our researches in quanta and quasars have reproductive implications of which we are not yet aware, the outermost, external environment is especially irrelevant to any study of intelligence, but the others are equally questionable. So the real scientific problem posed by intelligence’s ‘environment’ is neither to question Brandon’s categorisation or to locate external, ecological and selective boundaries for Homo sapiens, but to understand what it is about intelligence that renders all such boundaries and conceptions obsolete.
So intelligent worlds are quite different from organic environments. At the same time, they divorce us from the truly natural environment to a quite astonishing degree. So successful have we been in interposing cultural and technological systems between ourselves and the natural environment that, especially in industrial societies, it is hard to find any aspect of human existence that is not primarily defined on intelligence’s own terms. Every factory, every office, every piece of paper is composed of purely non-intelligent materials, but practically never do they impinge on our lives with a form, fit or function that is not fundamentally the result of their reconstruction into a tool, a resource, a system, and so on. As a result, both our collective relationship to nature and our personal lives are so absorbed in the struggle to render intelligible the representations, demands, opportunities, obligations and contradictions of employment, commoditisation, the media, industrialisation, bureaucracy, money and so on, that it takes an exceptionally well informed effort to detect, let alone encounter nature in natural form (Robinson 2004: Chapters 4-5). Even the recurring cult of the wilderness is expressed more through romanticism, documentaries and tourism than any direct acquaintance, and increasingly mediated by profiteering, regulation and romanticism (Teich, Porter and Gustafson 1997).
With few exceptions, the massively collective and artificial character of the contemporary human world is almost wholly ignored by would-be scientific studies of intelligence, which commonly limit themselves to adaptational, cognitive, phenomenal and individualistic frameworks. But this is to limit intelligence to extremely narrow aspects of what is in fact a world of immense social and symbolic, cultural and technological complexity and depth. It is fully equivalent to eliminating the study of evolution and heredity from biology, leaving only the isolated organism. As a result, the actual objects, processes, relationships and systems through which intelligence operates to the historical scale have been universally neglected, at least by students of intelligence as such. Indeed, history itself is seldom seen as the product of human intelligence, and although human activity and experience are widely accepted as being the products of history, only in quite narrow sections of the social sciences (e.g., the sociology of knowledge) do the implications of this fact have any impact.
Nevertheless, from an evolutionary point of view, it is the world of intelligence that delivers on both object permanence’s promise to supersede evolution and the intelligent subject’s potential to rise above biology altogether. This it does in a variety of ways. For example, intelligent beings create, share and develop a huge range of systems for action, organisation and evaluation through their ability to share a common world. You and I can share these words not because human biology provides mechanisms of direct communication in the manner of birdsong but because they are embedded in and mediated by systems (languages, intellectual disciplines, organisations, production and communications systems, and so on) that rely absolutely on our shared capacity for object permanence and the organisation of our objects into sequences and systems that allow us to debate their meaning.
However, by far the greatest implication of intelligence’s creation of the world is the corollary to which I have alluded several times, namely history. For any world created by collective intelligence is capable of radical transformation and development without any change in the underlying structures of either individual intelligence or its biological underpinnings. Had the cave painters of Chauvet or Altamira been born three hundred centuries later, they would have felt as much (or as little) at home in modern industrial society as we do ourselves, even though the vast difference between their actual lifestyle and ours would, in biological terms, qualify them as members of another species, if not a different genus. It is this historical character that has let so many generations take for granted a higher starting point than their parents, to disregard the burdens and threats that absorbed so much of their predecessors’ effort, and to find quite new opportunities open to them – and all without significant biological change. And this history itself has been made possible not only by object permanence but also because the social and material world we have collectively created both embodies and transcends the biological conditions any of us may inhabit.
By contrast, a biological niche is seldom dominated by the organism’s own products, and even when it is – in the case of a termite mound, for example – the product in question is typically rigidly determined by changes in the organism’s biology that are no more open to the organism’s understanding or control than the weather. But for human beings, not only all government and all organisations but art, philosophy, the social sciences and a great deal else assume the ability to recognise that we create our own world and may actively develop any aspect of it further. To repeat the same point on the largest possible scale, the way intelligence creates history ensures that the human ‘niche’ is indefinitely self-transforming, without material limit. Unlike the niche and environment of any other species, our ‘niche’ is ultimately identical with the human environment as a whole, and the human environment is ultimately identical with the universe.
So our ‘world’ is the arena of this unlimited development, and history is its medium, its realisation and, increasingly, its explicit inspiration and chosen proving ground. As social, political and economic policies grow more ambitious and the scope and power of our collective cultures and technologies expand both our impact on nature and our mutual relationships, history is increasingly the conscious product of our collective intelligent activity and increasingly organised into collective structures – political, economic, ideological, scientific and so on – that embody logico-mathematical rules, rational methods and quasi-intelligent systems.
As I have argued elsewhere (Robinson 2004), the analogy between individual intelligence and the structure of history as a whole began at the same moment as history itself, and it is not difficult to detect the resemblances between the familiar stages of individual cognitive development and major historical formations such as feudalism, capitalism, and so on. That is not to say that history is individual development writ large. The individualism of most research into intelligence makes it easy to misinterpret intelligence in this way (Allen 2006; Eller 2005; Robinson 2006), but the structure and development of intelligence on individual and historical scales are quite distinct (and often quite different) corollaries of the generic model of intelligence outline above (and expanded in Robinson 2004). Epigenesis, evolution and heredity are likewise all essential aspects of life, but only on the most abstract level is there any direct parallelism between these different and equally elementary aspects of life.
Not that the analogy is quite exact. History is only possible (beyond the simplest level) when the individuals who act it out internalise at least some of the structures that inform, determine and regulate it – an insight that is triggered with the first intimation of lineage or heritage. In other words, just as having any but the most primitive mind implies having a theory of mind, so participating in history proper presupposes that its inhabitants possess at least a modest theory of history, even if it is clothed in the rags of theology, property, philosophy, blood or superstition. By contrast, there is no need for any non-intelligent organism to have the least inkling of how evolution or epigenesis works or how the organism itself changes from one generation to the next, whereas the concepts of deliberate social change and of remaining true to certain purposes or values must both be present for history to ‘happen’. This places intelligence firmly back in a true world, for without at least such a personal apprehension and mastery of how social and symbolic systems work, the progressively greater conscious control historical intelligence exercises over its own activity, development and direction would be quite impossible. With some such apprehension, however, not only can artefacts, opportunities and lessons be deliberately passed on and more effective systems and processes built, but the very rules, methods, instruments and techniques whereby such structures are built, operated and evaluated can also be abstracted, improved and institutionalised, and historical change thereby accelerated.
It is primarily by virtue of the densely interlocking economic and political, cultural and technological, ideological and scientific systems that result from this process – this worldliness – that the question of human evolution has become a practical non-issue. If, as I argued in connection with object permanence, our general capacity for objectivity allows us to defend ourselves against the forces of variation and selection, it is the human world by which we surround ourselves that is the actual defence.
Of course, the world is far more than that. I have already mentioned some of our more immediate defences against specifically biological variation and selection: health education and public health systems, medical science, epidemiological research, mass sanitation, improved nutrition, a scientific attitude to disease, theories of evolution, and so on. One could easily add other defences against natural selection such as flood barriers, fire brigades, international aid (such as it is), swamp drainage and buildings capable of resisting earthquakes, not to mention disaster recovery for when all else fails. However, we are also the authors of far more powerful structures and systems whose evolutionary impact goes far wider and deeper than simply protecting us from infection, famine and natural disaster. Far from simply creating barriers, channels, vehicles and other mediators between the human world and its natural setting, human economies have progressively invaded, transformed and revolutionised nature itself. For better or for worse, human activity increasingly defines and directs not only the relationship between intelligence and the rest of nature but also what the rest of nature actually is. Through the spread of massive urban agglomerations, the conversion of huge tracts of wilderness to agriculture, pastoralism, mining and tourism, our impact on fish stocks, eutrophic algal blooms and the marine environment generally, pollution on a literally industrial and then global scale, and innumerable lesser actions, we have transformed the face of nature as a whole and changed or destroyed the niches of countless other species.
This brings me to my final argument in support of Wallace and the real importance of the fact that intelligent species live not in an environment but in a planetary-scale world: that, far from intelligence being subject to evolution, evolution is increasingly subject to intelligence. Sometimes this is intentional, as when breeding programmes are explicitly designed to create heritable differences more adapted to the selective pressures of markets. More often it is accidental, to the point where capitalist industrialisation has triggered off one of the worst extinction crises ever. However, as we come to appreciate the nature and consequences of our actions, the balance between the accidental and the intentional appears to be shifting. Hence not only the growing predominance of intelligence over non-intelligent nature but also its likely irreversibility. This has several strands.
Firstly, the sheer magnitude of the forces human beings now wield ensures that we are among the most important environmental factors for every other major vertebrate species. Through the food chains at whose heads the latter often stand, a huge number of other species is also impacted. Secondly, unlike the rest of nature, its intelligent inhabitants are capable of intending that variation and selection should occur, right down to specific genetic and developmental mechanics and precise inherited forms. And thirdly, it is in the direct interests of human beings that they should assume direction and control of the natural environment, not only because of the residual threats posed by disease, famine, earthquake, flood and so on, but also because the opportunity to extend our reach will not be resisted, so we should take care to do it ‘properly’.
This argument may seem to smack more of Victorian triumphalism than a realistic analysis of the contemporary human condition. A huge proportion of nature remains all but untouched by human activity; nature is awash with forces vastly more powerful than anything human beings can muster; and the ecological problems human beings are currently inflicting on themselves hardly suggest that we will be mastering nature at any time soon. But these arguments are not as convincing as they first seem.
The argument that nature remains largely unaffected by human beings assumes that intelligence’s pretensions will always be defeated by the sheer scale of the non-intelligent. However, if one compares the level of human influence on this planet today with our far more puny presence only three centuries ago, and then projects even a fraction of that rate of change and development onto the next few millennia, there seems to be little reason to believe that intelligence will never be able to operate on a truly planetary scale. It may well be that the current rate will not be sustained, but any plausible rate of development, even with setbacks of Dark Age proportions, is likely to be enough to transform the entire planet eventually.
The second argument, that there are forces in nature that are vastly more powerful than anything human beings can muster, is again empirically correct. For example, every day the Earth receives the equivalent of perhaps 160 trillion tons of TNT in solar energy alone. But the issue here is not the magnitude of these forces but how they are arrayed. Untamed nature is organised into the ecosystems on which human beings, like all living things, rely. However, if we could only learn how to operate such systems, they would not resist rearrangement into forms that suit human beings better. So even the most hostile of natural forces might conceivably be co-opted to the human ‘side’. Indeed, there is only one ‘side’ in this struggle: as intelligent beings, humans care about the outcomes of environmental changes and can organise to do something about them, whereas non-intelligent nature remains wholly indifferent both to human beings and to itself. If we wanted to restore the Sahara to the green and pleasant land it was some millennia back, the enormous forces we would need to reorganise to achieve this would not decline to be involved.
Finally, it is true that the ecological damage we are currently inflicting on the planet hardly suggests that we will be capable of any such control in the foreseeable future. If, as Bacon said, if nature is to be commanded, first it must be obeyed, we do not seem to have mastered nature’s most basic rules, and we certainly are not good at putting them into practice. We have not only scarred the face of the environment but also reduced its stability in the face of our actions, which is likely to exacerbate this problem before it is ever solved.
But what is the underlying problem? Simply a matter of technical competence to manage large-scale environments? Probably not. This is largely a self-inflicted problem, driven by strictly intelligent structures, primarily the nature of the economic system through which our relationships with nature are defined and controlled. Essentially, all political and economic systems are systems of interests, and as such will require considerable adjustment – if that is the word – before they support a constructive relationship with nature. Our dominant economic system, an increasingly global industrial capitalism, cannot be directed by our existing political mechanisms, and its accelerating invasion of nature is by no means motivated by a sense of disinterested ecological responsibility (Panitch and Leys 2006). So even if human beings can solve the enormous scientific and technological problems of creating a rational ecology, that is no reason to be optimistic about the outcome, which will have to be delivered by political and economic means. We may ultimately learn to manage nature as a sustainable totality, but that is not to say that the learning will be a pleasant experience. On the other hand, it is not as though we have any alternative: it is after all the sheer magnitude and power of these same systems and interests that mean that we cannot not assume responsibility for our relationship with the rest of nature, or that assuming that that responsibility can be performed adequately without a radical re-evaluation of our existing economic and political values, goals and systems.
This leaves human beings in a profound quandary. We are a long way from possessing the intellectual, technological or social tools needed to accomplish this ultimate feat except in trivial and risky ways, yet it is a task that cannot be avoided or shirked. And if it cannot be avoided then we must embrace it, for the only alternative is failure. We need not be optimistic about when, if ever, we will get it right: there is after all a natural limit to this challenge, namely the point at which the human environment is so completely impoverished that we are unable to continue taking action on a problematic scale. If we exceed that limit, so much the worse for us. But if we can instead create a truly rational ecology, then intelligence will be a little closer to the apotheosis we can scarcely yet imagine, yet for which we must certainly hope.
However, even if the worst comes to the worst, it still does not imply that it is not intelligence that leads evolution rather than vice versa. Even if the whole planet is reduced to an uninhabitable desert – not a scenario that any scientist seems to be contemplating at the moment – we still differ from previous victims of evolution’s vicissitudes in being able to recognise the dreadful thing we have done.
I have advanced three general reasons to believe that intelligence is immune to evolution. Firstly, our construction of reality in terms of objects (even when interpreted weakly) enables us to recognise and intervene in, nullify, deflect or undo the forces of natural selection. Secondly, both the abstractness of the process whereby it develops and the abstract nature of its end-product imply that, regardless of the species under consideration, its internal structure – the subject – must be the same for all forms of intelligence, so there is no basis for variation and selection. Finally, the building of a full-scale rational world progressively reverses the relationship between intelligence and evolution: it is not intelligence that is subject to evolution but evolution that is subject to intelligence.
So where does all this lead? In brief, if intelligence is immune to evolution, then one of the most fundamental assumptions made by a very large number of scientists in the study of human nature is simply false. Likewise, the language of adaptation is radically inappropriate for intelligent activity and development. I would certainly agree that intelligence originated by an evolutionary route (Robinson 2005), but it does not follow that intelligence shares the status of the lion’s claw or the horse’s hoof. If intelligence is not only a truly universal and invariant structure but also capable of comprehending the workings of evolution on every level and in every sense, then it is hard to see how it could be open to further biological variation and selection, any more than Planck’s constant, the periodic table or the value of pi.
As I understand it, this interpretation is quite consistent with Darwinism, and Eigen (1992), Prigogine and Stengers (1984), Kaufman (1993), von Bertalanffy (1950), Goodwin (1995) and many others could all be called upon to support the notion of abstracted structures that resist (and often drive) the pressures of variation and selection, and it is hard to see what complexity and chaos theory mean if it is not that abstract structures of this kind are very widespread. All I am adding is to suggest that in the case of intelligence such structures and relationships are internalised and put to active use in such a way that evolution is transcended.
Beyond that, it is clear that the human sciences need not try to trace human nature back to human biology, nor assume that human beings should consider them as ‘really’ or even as first and foremost organisms rather than persons. It is likely that the same could be said of a variety of other species, perhaps including dolphins, several non-human primates and a few birds. As a result, the human sciences are under no obligation to seek evolutionary or adaptive explanations for the things human beings do. Intelligence must be compatible with evolution, even if only in the minimal sense that it must sustain rates of biological reproduction that ensure the continued existence of intelligent beings. But that no more obliges one to look for biological explanations for intelligence itself than the fact that organisms must obey the laws of gravity or thermodynamics means that adaptation is ‘really’ a physical concept.
There is a great deal more to be said about the relationship between intelligence and evolution (Robinson 2005). However, the fundamental logic seems to be inescapable. One can only regret that Wallace’s view that our immunity to evolution reflected a qualitative break (Wallace 1870; Berry 2002 passim) was never followed up adequately, and that Wallace himself succumbed to the widespread mysticism of his day (Oppenheim 1985; Robinson 2004: Ch. 7). The nature of human and non-human intelligence and its unique relationship to biology might so easily have been grasped much earlier. But then, perhaps the reason why this did not happen was that Wallace, like most of science from his day to ours, had no clear definition of terms like ‘mind’, ‘intellect’ and so on. Hence the need for Piaget’s interpretation of intelligence, with all its portentous implications.
Aiello, L.C. and Wheeler, P. (1995). The expensive-tissue hypothesis: The brain and the digestive system in human and primate evolution. Current Anthropology. Vol. 36, No. 2, 199–221.
Allen, B. (2006). A history without the history. History and Theory 45: 1 (134ff).
Antinucci, F. (ed.) (1989). Cognitive Structure and Development in Non-Human Primates. London: Lawrence Erlbaum.
Baldwin, J.M. (1895). Consciousness and evolution. Science 2, 219-223
Baldwin, J.M. (1896). A new factor in evolution. American Naturalist 30, 441–451, 536–553.
Baldwin, J.M. (1902). Development and Evolution. New York: Macmillan.
Barkow, J.H., Cosmides, L. and Tooby, J. (1992). The Adapted Mind. Evolutionary Psychology and the Generation of Culture. Oxford: Oxford University Press.
Barnes, M.H. (2000). Stages of Thought. The Co-Evolution of Religious Thought and Science. Oxford University Press: New York.
Bauer, G. and Johnson, C. M. (1994). Trained motor imitation by bottlenose dolphins (Tursiops truncatus). Perceptual and Motor Skills 79: 1307-1315.
Bebko, J.M., Burke, L., Craven, J. and Sarlo, N. (1992). The importance of motor activity in sensorimotor development: a perspective from children with physical handicaps. Human Development. 35: 226-240.
Berry, A. (2002). Infinite Tropics. An Alfred Russel Wallace Anthology. London: Verso.
Bertalanffy, L. von (1950). The theory of open systems in physics and biology. Science, 3: 23-9.
Beth, E.W., and Piaget, J. (1966). Mathematical Epistemology and Psychology. Dordrecht: D. Reidel.
Blackmore, S. (1999). The Meme Machine. Oxford: Oxford University Press.
Bower, T.G.R. (1989). The Rational Infant. Learning in Infancy. New York: W.H. Freeman and Company.
Brandon, R. (1990) Adaptation and Environment. Princeton NJ: Princeton University Press.
Buss, L.W. (1987). The Evolution of Individuality. Princeton, NJ: Princeton University Press.
Campbell, D.T. (1974). Evolutionary epistemology. In The Philosophy of Karl R. Popper, edited by P. A. Schilpp, 412-463. LaSalle, IL: Open Court.
Cole, M., Engeström, Y. and Vasquez, O. (1997). Mind, Culture and Activity. Cambridge: Cambridge University Press.
Damerow, P. (1995). Prehistory and cognitive development. Invited Lecture to the Twenty-Fifth Annual Symposium of the Jean Piaget Society, Berkeley, June 1–3, 1995.
Damerow, P. (1996). Abstraction and Representation. Essays on the Cultural Evolution of Thinking. Dordrecht: Kluwer Academic Publishers.
Darwin, C. (1901). The Descent of Man and Selection in Relation to Sex. London: John Murray.
Dawkins, R. (1976). The Selfish Gene. New York: Oxford University Press.
Dawkins, R. (1982). The Extended Phenotype. Oxford: Oxford University Press.
Deacon, T. (1997). The Symbolic Species. London: Penguin.
Desmond, A. and Moore, J. (1992). Darwin. London: Penguin.
Dover, G. (2000). Dear Mr Darwin. Letters on the Evolution of Life and Human Nature. London: Weidenfeld and Nicolson.
Eagle, R.S. (1985). Deprivation of early sensorimotor experience and cognition in the severely involved cerebral-palsied child. Journal of Autism and Developmental Disorders. 15/3: 269-283.
Eble, G.E. (1999). On the dual nature of chance in evolutionary biology and paleobiology. Paleobiology, Vol. 25, No. 1 (Winter), pp. 75-87.
Eigen, M. (1992). Steps Towards Life. Oxford: Oxford University Press.
Eiseley, L. (1961). Darwin’s Century: Evolution and the Men Who Discovered It. New York: Doubleday.
Eller, J.D. (2005). Review of The History of Human Reason. Anthropology Review Database. February 18 2005.
Foss, B.M. (1969). Determinants of Infant Behaviour, 4. London: Methuen.
Fowler, J.W. (1981). Stages of Faith: The Psychology of Human Development and the Quest for Meaning. San Francisco: Harper & Row.
French, B.M. (1998). Traces of Catastrophe. A Handbook of Shock-Metamorphic Effects in Terrestrial Meteorite Impact Structures. LPI Contribution No. 954, Lunar and Planetary Institute: Houston.
Frisch, K. von (1974). Animal Architecture. New York: Harcourt Brace Jovanovich.
Gablik, S. (1977). Progress in Art. Rizzoli: New York.
Ghiselin, M.T. (1969). The Triumph of the Darwinian Method. London: University of California Press.
Goodwin, B.C. (1995). How the Leopard Changed its Spots. London: Orion.
Gottesman, M. (1973). Conservation development in blind children. Child Development, 44/4: 824-827.
Hahlweg, K., and Hooker, C.A. (1989). Issues in Evolutionary Epistemology. Albany: State University Press of New York.
Hallpike, C.R. (1979). The Foundations of Primitive Thought. Oxford: Oxford University Press.
Hallpike, C.R. (1984). The relevance of the theory of inclusive fitness to human society. J. Social Biol. Struct. 7: 131–144.
Hallpike, C.R. (1988). The Principles of Social Evolution. Oxford: Oxford University Press.
Hallpike, C.R. (2004). The Evolution of Moral Understanding. Prometheus Research Group: www.prometheus.org.uk.
Harrison, G.A. (1993). Human Adaptation. Oxford: Oxford University Press.
Huxley, J.S. (1947). Man in the Modern World.London: Chatto and Windus.
Ifrah, G. (1998). The Universal History of Numbers: From Prehistory to the Invention of the Computer. London: Harvill.
Inhelder, B. and Piaget, J. (1958). The Growth of Logical Thinking from Childhood to Adolescence. London: Routledge and Kegan Paul.
Inhelder, B. and Piaget, J. (1964). The Early Growth of Logic in the Child: Classification and Seriation. London: Routledge and Kegan Paul.
Kauffman, S.A. (1993). The Origins of Order: Self-Organization and Selection in Evolution. Oxford: Oxford University Press.
Kirkpatrick, K. (2001). Object recognition. In R. G. Cook (Ed.), Avian Visual Cognition [On-line]. Available: www.pigeon.psy.tufts.edu/avc/kirkpatrick/.
Kopp, C.B. and Shaperman, J. (1973). Cognitive development in the absence of object manipulation during infancy. Abbreviated version published in Developmental Psychology, 9, p.430ff.
Langer, J. (1998). Phylogenetic and ontogenetic origins of cognition: Classification. In Langer and Killen (1998), pp. 33-54.
Langer, J. (2006). The heterochronic evolution of primate cognitive development. Biological Theory 1(1) 2006, 41–43.
Langer, J. and Killen, M. (eds) (1998). Piaget, Evolution, and Development. London: Lawrence Erlbaum Associates.
LePan, D. (1989). The Cognitive Revolution in Western Culture. Vol. 1: The Birth of Expectation. Basingstoke: Macmillan.
Lourenço, O. and Machado, A. (1996). In defense of Piaget’s theory: A reply to ten common criticisms. Psychological Review, 103, 1: 143–164.
Marino, L. (2002). Convergence of complex cognitive abilities in cetaceans and primates. Brain Behav. Evolution 59: 21-32.
Maynard Smith, J. and Vida, G. (1990). Organisational Constraints on the Dynamics of Evolution. Manchester: Manchester University Press.
McGinn, C. (1989). Can we solve the mind-body problem? Mind, N.S., Vol. 98: No. 381: 349-366.
McKinney, M.L. (1998). Cognitive evolution by extending brain development: On recapitulation, progress and other heresies. In Langer and Killen (1998), pp. 9-31.
Mendes, N., and Huber, L. (2004). Object permanence in common marmosets (Callithrix jacchus). Journal of Comparative Psychology. Vol. 118(1) 103-112.
Moran, E.F. (2000). Human Adaptability. An Introduction to Ecological Anthropology. Second edition. Boulder, Colorado: Westview Press.
Morbidity and Mortality Weekly Report (1997). Fatal human plague – Arizona and Colorado, 1996. (Editorial note) 46 (27); 617-620.
Nissen, H.J., Damerow, P. and Englund, R.K. (1993). Archaic Bookkeeping. Writing and Techniques of Economic Administration in the Ancient Near East. Chicago: University of Chicago Press.
Norton, C.E. (ed.). (1877). Philosophical Discussions by Chauncey Wright. New York: Lennox Hill.
O’Hear, A. (ed.) (1995). Karl Popper: Philosophy and Problems. Supplement to ‘Philosophy’. Royal Institute of Philosophy Supplement: 39. Cambridge University Press.
Oppenheim, J. (1985). The Other World: Spiritualism and Psychical Research in England, 1850–1914. Cambridge: Cambridge University Press.
Oyama, S. (2000). The Ontogeny of Information. Developmental Systems and Evolution. Second edition. Durham, N.C.: Duke University Press.
Oyama, S., Griffiths, P.E., and Gray, R.D. (2001). Cycles of Contingency. Developmental Systems and Evolution. Cambridge, Mass.: MIT Press.
Panitch, L., and Leys, C. (2006). Coming to Terms with Nature. London: Merlin Press.
Parker, S.T. and McKinney, M.L. (1999). Origins of Intelligence: The Evolution of Cognitive Development in Monkeys, Apes and Humans. Baltimore: Johns Hopkins University Press.
Passingham, R.E. (1975). Changes in the size and organisation of the brain in man and his ancestors. Brain, Behavior and Evolution, 11: 73-90.
Pepperberg, I.M. (1999). The Alex Studies. Cognitive and Communicative Abilities of Grey Parrots. Cambridge, Mass.: Harvard University Press.
Pepperberg, I.M., Willner M.R., and Gravitz, L.B. (1997) Development of Piagetian object permanence in a grey parrot (Psittacus erithacus). J. Comp. Psychol. 1997 111(1):63-75.
Piaget, J. (1928). Judgment and Reasoning in the Child. London: Routledge and Kegan Paul.
Piaget, J. (1932a). The Thought and Language of the Child. Second edition. London: Routledge and Kegan Paul.
Piaget, J. (1932b). The Moral Judgment of the Child. London: Kegan Paul, Trench, Trubner and Co.
Piaget, J. (1950). Introduction á l’Épistémologie Génétique. 3 vols. Paris: Presses Universitaires de France.
Piaget, J. (1951). Play, Dreams and Imitation in Childhood. London: Routledge and Kegan Paul.
Piaget, J. (1952). The Child’s Conception of Number. London: Routledge and Kegan Paul.
Piaget, J. (1953a). The Origins of Intelligence in Children. London: Routledge and Kegan Paul.
Piaget, J. (1953b). Logic and Psychology. Manchester: Manchester University Press.
Piaget, J. (1968a). Épistémologie et Psychologie de l’Identité. Paris: Presses Universitaires de France.
Piaget, J. (1968b). Memory and Intelligence. London: Routledge and Kegan Paul.
Piaget, J. (1969). The Mechanisms of Perception. London: Routledge and Kegan Paul.
Piaget, J. (1971a). Biology and Knowledge. Edinburgh: Edinburgh University Press.
Piaget, J. (1971b). Structuralism. London: Routledge and Kegan Paul.
Piaget, J. (1971c). Mental Imagery in the Child. London: Routledge and Kegan Paul.
Piaget, J. (1972). Essai de Logique Opératoire. Paris: Dunod.
Piaget, J. (1974). Experiments in Contradiction. Chicago: University of Chicago Press.
Piaget, J. (1977). The Grasp of Consciousness. London: Routledge and Kegan Paul.
Piaget, J. (1978). Success and Understanding. London: Routledge and Kegan Paul.
Piaget, J. (1979). Behaviour and Evolution. London: Routledge and Kegan Paul.
Piaget, J. (1980a). Adaptation and Intelligence. Organic Selection and Phenocopy. London: University of Chicago Press.
Piaget, J. (1980b). Les Formes Élémentaires de la Dialectique. Paris, Editions Gallimard.
Piaget, J. (1981). Intelligence and Affectivity. Their Relationship during Child Development. Palo Alto: Annual Reviews.
Piaget, J. (1995). Sociological Studies. London: Routledge.
Piaget, J. (2001). Studies in Reflecting Abstraction. Hove: Psychology Press.
Piaget, J. and Garcia, R. (1989). Psychogenesis and the History of Science. New York: Columbia University Press.
Piaget, J. and Garcia, R. (1991). Towards a Logic of Meanings. Hillsdale, N.J.: Lawrence Erlbaum Associates.
Pollok, B., Prior, H., and Güntürkün, O. (2000). Development of object permanence in food-storing magpies (Pica pica). Journal of Comparative Psychology. Vol. 114(2) 148-157.
Prigogine, I., and Stengers, I. (1984). Order Out of Chaos. London: William Heinemann.
Radding, C.M. (1985). A World Made by Men. Cognition and Society, 400–1200. Chapel Hill: University of North Carolina Press.
Radding, C.M., and Clark, W.W. (1985). Medieval Architecture, Medieval Learning: Builders and Masters in the Age of Romanesque and Gothic. New Haven: Yale University Press.
Radnitzky, G. and Bartley, W.W. (1987). Evolutionary Epistemology: Theory of Rationality and the Sociology of Knowledge. La Salle, Ill.: Open Court.
Reiss, D. and Marino, L. (2001). Mirror self-recognition in the bottlenose dolphin: A case of cognitive convergence. Proc. Natl. Acad. Sci. 98(10): 5937–5942.
Richards, R.J. (1987). Darwin and the Emergence of Evolutionary Theories of Mind and Behaviour. Chicago: University of Chicago Press.
Robinson, R.J. (2004). The History of Human Reason. Prometheus Research Group: http://www.prometheus.org.uk.
Robinson, R.J. (2005). The Birth of Reason. Electronic document: http://www.prometheus.org.uk.
Robinson, R.J. (2006). The History of Human Reason: Reply to Critics. Electronic document: http://www.prometheus.org.uk.
Rose, H. and Rose, S. (2000). Alas, Poor Darwin. Arguments against Evolutionary Psychology. New York: Harmony Books.
Ruse, M. (1986). Taking Darwin Seriously: A Naturalistic Approach to Philosophy. Oxford: Blackwell.
Russon, A.E., Mitchell, R.W., Lefebvre, L., and Abravanel, E. (1998). The comparative evolution of imitation. In Langer and Killen (1998), pp. 103-143.
Smith, L. (1993). Necessary Knowledge: Piagetian Perspectives on Constructivism. Hove, Lawrence Erlbaum.
Smithurst, M. (1995). Popper and the scepticisms of evolutionary epistemology, or, What were human beings made for? In O’Hear (1995: 207-223)
Teich, M., Porter, R. and Gustafson, B. (1997). Nature and Society in Historical Context. Cambridge: Cambridge University Press.
Tishkoff, S.A., Reed, F.A., Ranciaro, A., Voight, B.F., Babbitt, C.C., Silverman, J.S., Powell, K., Mortensen, H.M., Hirbo, J.B., Osman, M., Ibrahim, M., Omar, S.A., Lema, G., Nyambo, T.B., Ghori, J., Bumpstead, S., Pritchard, J.K., Wray, J.A.. and Deloukas, P. (2006). Convergent adaptation of human lactase persistence in Africa and Europe. Nature Genetics 10/12/2006.
Tooby, J., and DeVore, I. (1987). The reconstruction of hominid behavioral evolution through strategic modeling. In Kinzey, W.G., The Evolution of Human Behavior: Primate Models. Albany: State University of New York Press. (183-237).
Vauclair, J. (1996). Animal Cognition: An Introduction to Modern Comparative Psychology. London: Harvard University Press.
Waddington, C.H. (1957). The Strategy of the Genes. London: Allen and Unwin.
Wallace, A.R. (1864). The origin of human races and the antiquity of man deduced from the ‘Theory of Natural Selection’. Journal of the Anthropological Society of London, 2: 158–187.
Wallace, A.R. (1870). The limits of natural selection as applied to man. In Contributions to the Theory of Natural Selection.
Wallace, A.R. (1905). My Life: A Record of Events and Opinions. Vol.2. London: Chapman and Hall Ltd.
Weiskrantz, L. (1997). Consciousness Lost and Found. A Neuropsychological Exploration. Oxford: Oxford University Press.
WHO (2005) Plague. Fact Sheet No.267. Revised February 2005.
Whyte, L.L. (1968). Aspects of Form. Second edition.London:Lund Humphries.
Wigner, E.P. (1960). The unreasonable effectiveness of mathematics in the natural science. Commun. Pure Appl. Math., 13, 1–14.
Wise, S.M. (2002). Drawing the Line. Science and the Case for Animal Rights. Cambridge, Mass.: Perseus.
Wynn, T. (1979). The intelligence of later Acheulean hominids. Man (ns), 14: 371–391.
Wynn, T. (1981). The intelligence of Oldowan hominids. Journal of Human Evolution, 10: 529–541.
Wynn, T. (2002). Archaeology and cognitive evolution. Behavioral And Brain Sciences. 25: 3: 389-438.
Young, M. E., and Wasserman, E.A. (2001). Stimulus control in complex arrays. In R. G. Cook (Ed.), Avian Visual Cognition [On-line]. Available: www.pigeon.psy.tufts.edu/avc/young/.
Zucca, P., Milos, N., and Vallortigara, G. (2007). Piagetian object permanence and its development in Eurasian jays (Garrulus glandarius). Animal Cognition. 10 (2): 243-258.
 Contrast Darwin’s reaction to Wallace’s later spiritualism (e.g., Darwin 1901: 72–73; also pp.195ff). On Darwin’s later attitudes to Wallace, see Desmond and Moore (1992, passim).
 The term ‘stage’ must be used cautiously in the Piagetian context – discrete stages and formal stage boundaries they lie more in the theorist’s logical analysis of the subject’s capabilities than in dramatic saltations in the subject’s internal structure and capabilities, which generally develop fairly continuously.