The idea that we are different from all other creatures is a central tenet of Western thought. The list of supposedly unique attributes is almost endless, including language, tool use, intelligence, morals, and aesthetics. Clearly, many of the ideas that pre-date Darwin by centuries must be revised in light of the fact that we are recently derived from primate ancestors and the enormous amount of information now known about non-human species. The general trend in evolutionary research has been to show that claims of human uniqueness were greatly exaggerated. Nevertheless, we clearly are unique in some respects so the challenge is to identify the real differences, not to deny any differences at all. In The Symbolic Species Deacon (1998) presents a sophisticated argument that our species is unique in its capacity for symbolic thought. According to Deacon, thinking symbolically doesn’t require an especially large brain or even a different brain from that possessed by our primate ancestors. In fact, it is actually possible to teach a chimpanzee or bonobo to think symbolically, more like us than their own kind. The problem is that it requires an arduous training process that has no counterpart in nature. Moreover, symbolic thought interferes with more basic forms of associative learning that are adaptive in most natural environments. If I pair the sound of the word ‘cat’ with an actual cat in a conditioning experiment, mice will learn to associate the two, but if I then say the word ‘cat’ many times without a cat actually being present, the object and its symbol will become dissociated again. In contrast, I can say the word ‘cat’ to you a million times and you will still associate it with a cat. More generally, symbolic thought requires symbols and what they stand for (p. 321 ) to remain associated in the mind even when they are not associated in the real world. Symbolic thought is like a lofty peak in an adaptive landscape that can be climbed only by first crossing a valley of low fitness. What made humans unique was a natural environmental context that made symbolic thought adaptive in its initial stages, allowing us, and us alone, to cross over to the new adaptive peak. Of course, social constructivism requires the capacity for symbolic thought. As Durkheim put it, ‘in all its aspects and at every moment of history, [human] social life is only possible thanks to a vast symbolism’ (Durkheim (1995) [1912]: 223). Deacon’s thesis therefore represents an important part of the new bridge connecting evolutionary biology with social constructivism.

Darwin machines: evolutionary processes built by evolution

All organisms—even bacteria and protozoa—are designed to be flexible, adaptively changing their properties within their own lifetimes in response to environmental change. In his panoramic review of brain evolution across the animal kingdom, Allman actually begins with a discussion of bacteria:

Some of the most basic features of brains can be found in bacteria because even the simplest motile organisms must solve the problems of locating resources and avoiding toxins in a variable environment. Strictly speaking, these unicellular organisms do not have nervous systems, but nevertheless they exhibit remarkably complex behavior: They sense their environment through a large number of receptors and store this elaborate sensory input in the form of brief memory traces. Moreover, they integrate the inputs from these multiple memory sensory channels to produce adaptive movements. The revolution in our understanding of genetic mechanisms has made it possible to determine how these brainlike processes work at molecular level in bacteria.

(Allman 1999: 3)

(p. 322 ) Another way to accomplish adaptive flexibility is illustrated by the mammalian immune system. Disease organisms are too numerous and evolve too fast with their short generation times to be combated with a set of rigid if-then rules. Instead, the immune system is a fast-changing evolutionary process in its own right, in which antibodies are produced at random and selected according to their ability to bind to antigens. The immune system is an evolutionary process, built by evolution.

This kind of adaptive flexibility extends beyond the immune system. Gerald Edelman, who received the Nobel prize for his work on the immune system, went on to develop a similar conception for the human brain (Edelman 1988; Edelman and Tonomi 2001). Deacon (1998) provides a lucid account of how brain development is fundamentally a Darwinian process, with growing axons competing for receptor sites. At a functional level, symbolic thought can be envisioned as a Darwinian process in a virtual world of mental representations.

Plotkin’s (1994a) term ‘Darwin machine’ aptly describes two essential features of an evolutionary process built by evolution. The word ‘machine’ indicates that the evolutionary process must be highly managed to lead to biologically adaptive outcomes. Anyone who has studied the immune system knows that it is mind-bogglingly complex in its genetic innateness. Antibodies that match antigens reproduce more, not by a lucky coincidence, but because the immune system is constructed that way. Nevertheless, the word ‘Darwin’ indicates that the process remains evolutionary despite being highly managed, with all the implications of evolution played out on a new stage. Open-ended and creative solutions to recent environmental challenges is perhaps the most important implication, but we should also expect the same kinds of historical contingencies and other constraining factors that cause adaptations to fall short of perfection in genetic evolution.

Social constructivists often associate evolution with an incapacity for change, in contrast to their own belief in the human potential for change. These associations only make sense if by ‘evolution’ we mean genetic evolution. When we include Darwin machines, the social constructivist position itself becomes evolutionary. Furthermore, genetic innateness can be seen positively as the complex machinery that makes rapid evolution possible. The idea that we can understand the human potential for change without paying attention to genetically innate mechanisms becomes as absurd as trying to study the immune system as if it were not genetically (p. 323 ) innate. With Darwin machines, the bridge from evolutionary biology to social constructivism has been considerably extended.

Groups as units of selection and adaptation

An important part of the sociological tradition is to view societies and cultures as organic wholes, whose properties cannot be reduced to their individual parts. This kind of holism fell upon hard times in both biology and the social sciences during the middle of the twentieth century. In the social sciences, the principle of methodological individualism claimed that all properties of societies can be reduced to the properties of their members and are not properly understood until this reduction has taken place (Watkins 1957). Rational choice theory offered a specific agenda in which individuals are assumed to behave as utility maximizers. In evolutionary biology, a consensus emerged that groups can evolve into adaptive units in principle but almost never do in reality. The reason is that group-level adaptations require a process of among-group selection, which is almost invariably weak compared to within-group selection (Williams 1966). As a separate argument, all adaptations were claimed to evolve at the level of genes, even when they benefited individuals or groups as vehicles of the genes (Williams 1966; Dawkins 2006 [1976]).

Despite the dominance of these ideas during recent intellectual history, much has happened in evolutionary biology to undermine their authority. One of the most important developments has been labeled ‘major transitions of life’. It used to be thought that evolution takes place entirely by small mutational change, but now a second pathway has been identified in which social groups become so integrated that they become a new higherlevel organism. One of the first to propose this radical new theory was Lynn Margulis (1970), who claimed that eukaryotic cells—the nucleated cells of all organisms other than bacteria—are actually symbiotic communities of bacteria whose members led a more autonomous existence in the past. Now it appears likely that similar transitions, from groups of organisms to groups as organisms, have occurred throughout the history of life, right down to the origin of life itself as social groups of cooperating molecular reactions (Maynard Smith and Szathmary 1995; Michod 1999).

Each transition requires a solution to the same kinds of social dilemmas that abound in human life. For example, consider a primordial cell in which the genes exist as independently replicating units. Some genes participate (p. 324 ) in the economy of the cell, producing resources that all can use. Other genes selfishly use these resources to replicate without contributing to the economy of the cell. The selfish genes are favored by within-group selection (they are more fit than solid-citizen genes within the same cell) while the solid citizen genes are favored by among-group selection (groups with an above-average frequency of solid-citizen genes contribute more to the total gene pool than groups with a below-average frequency). As long as within-group selection remains strong the cell will not function adaptively. The evolution of chromosomes neatly solved this problem by binding the genes together into a single structure that replicates as a unit. Chromosomes greatly reduce differential replication within the cell, making among-cell selection the primary evolutionary force. Notice that this represents a feedback process between traits that alter the parameters of social evolution (chromosomes) and traits that evolve on the basis of the alteration (solid citizen genes). This is what is meant by the term ‘niche construction’, which can be regarded as a close cousin of social constructivism.

The major transitions of life forever dispel the notion that higher-level selection is invariably weak compared to lower-level selection. Eukaryotic multicellular organisms such as you and me are shining contradictions of that claim. More generally, if the organisms of today are the social groups of past ages, then the concept of the social groups of today as like organisms, at least to a degree, no longer appears so outrageous. Indeed, when natural selection is carefully partitioned into within- and among-group components, group selection can be shown to be an important force in many species, even when it does not result in full-fledged ‘superorganisms’. In other words, many traits in the organic world evolve by increasing the fitness of collectives, relative to other collectives, rather than by increasing the fitness of individuals, relative to other individuals within collectives (Sober and Wilson 1998).

These developments in evolutionary biology provide a new background for the study of human evolution. People have always existed in many kinds of groups that merge and split for various activities. In any given context, traits can evolve by benefiting individuals relative to others in the same group, or by benefiting the whole group relative to other groups. Furthermore, traits comparable to chromosomes can evolve that reduce the possibilities for within-group advantage, concentrating natural selection at the between-group level. Cultural anthropologist and primatologist Chris Boehm (1999) has developed this thesis in his book Hierarchy in the Forest. (p. 325 ) According to Boehm, moral systems are the human analog of chromosomes, reducing the potential for individuals to profit at the expense of other members of the same group and converting the whole group into a potent unit of selection and adaptation. Consider the effects of moral systems on the three ingredients of natural selection: phenotypic variation, fitness consequences, and heritability. Behaviors that are strongly prescribed or proscribed by a moral system simply will not vary within the group as much as they would otherwise (unless variation per se is part of the norm). On the other hand, groups that adopt different norms can vary tremendously from each other. In technical terms, moral systems dramatically alter the partitioning of phenotypic variation within and among groups. These phenotypic differences also make a difference: murder, theft, adultery, rape, self-sacrifice on behalf of others, slacking, and most other behaviors that arouse moral passions are clearly relevant to basic survival and reproduction. Finally, moral systems are often elaborately constructed to replicate themselves from person to person and group to group through narratives and other practices deemed sacred. In short, moral systems have a transformative effect on the fundamental ingredients of natural selection, largely (although by no means entirely) converting social groups into functionally adaptive units, even when they are composed of genetically unrelated individuals.

These developments concerning levels of selection are intertwined with the other developments concerning symbolic thought and Darwin machines, giving them a collective dimension that would be missing otherwise. So many aspects of human life are collective, including the otherand group-oriented nature of morality, language, and any symbol that is not the personal invention of a single individual. Group-level functionality was taken for granted until the middle of the twentieth century, when it was denied altogether or viewed through the distorted lens of individualism. Now we are on the verge of achieving a complex but comprehensible middle ground, in which group-level functionality does exist, but only when special conditions are met. The collective dimension extends the bridge from evolutionary biology to social constructivism.

Adaptation and rational thought

Rational thought has a special status in the social sciences, in part because it provides the core of logical and scientific reasoning and in part because (p. 326 ) its barebone assumption of utility maximization provides the basis for endless theorizing. Rational thought is often treated as the gold standard against which other forms of thought are judged. Religious thought in particular is found so wanting by this standard that it becomes well-nigh incomprehensible.

In contrast, evolutionary theory judges all forms of thought in terms of what they cause organisms to do. Rational thought leads to adaptive outcomes in some contexts, which is why it has become part of our mental toolkit. It does not lead to adaptive outcomes in all contexts, which is why it is not the only tool in our mental toolkit. There is also a historical argument for why rational thought is only one tool in our toolkit—because it is a recent tool and most of the other tools are more ancient. Once these evolutionary factors become the basis for evaluating modes of thought, a host of alternatives to rational thought become explicable in terms of what they cause people to do. Emotions are evolved mechanisms for motivating adaptive behavior that are far more ancient than the cognitive processes associated with rational thought. We might therefore expect moral systems to be designed to trigger powerful emotional impulses, linking joy with right, fear with wrong, anger with transgressions. We might expect stories, music, and rituals to be at least as important as logical arguments in orchestrating the behavior of individuals and groups. Imaginary beings and events that never happened can provide blueprints for action that far surpass factual accounts of the natural world in clarity and motivating power. These other-worldly forms of thought, which include but are not restricted to religion, cannot completely eclipse rational thought, which is superior in some contexts, but the reverse statement is equally true.

Social constructivists are themselves uncomfortable with the gold standard of rational thought, but they don’t have a standard to replace it, other than the absence of standards associated with relativism. Adaptation, or rather adaptationism appropriately conceived (which acknowledges that not everything is adaptive) therefore provides a new way of thinking about alternative modes of thought, extending the bridge from evolutionary biology to social constructivism still further.

It should be obvious that these developments in evolutionary biology reach toward themes that have always been central to social constructivism. Let us now see how one sociologist reaches in the direction of evolutionary biology.

Earlier I stressed the current lack of consensus among evolutionary biologists on the subject of culture. Christian Smith reveals a similar lack of consensus among sociologists, who remain just as much in the rank speculation stage as far as the big picture is concerned, however well they have documented pieces of the puzzle. The conception of human nature and society that Smith is trying to establish within his own discipline includes the following elements. There is a universal human nature that transcends cultural differences. There is a core of truth in earlier sociological traditions (such as the work of Talcot Parsons) that needs to be preserved and built upon. Human culture is fundamentally a moral order and people are inescapably moral agents. There is more to morality than altruism. Norms do not exist as isolated packets but as part of a complicated normative system. People have a propensity for self-centeredness in addition to their inescapably moral natures. People are, fundamentally, believing animals who convey their beliefs largely through narratives. Stories shape people in addition to people shaping stories.

This conception of people, society, and culture is highly compatible with the developments in evolutionary biology that I have reviewed—so close that it seemed as I read Smith’s book that the bridge might be nearing completion. Alas, as soon as Smith started writing about biology and evolution (2003: 33–43) it became clear that the gap was still large in his own mind and his own effort to bridge the gap was proceeding in a different direction. According to Smith, the positions that he associates with sociobiology and evolutionary psychology are fatally flawed for a number of reasons. They attempt to explain moral systems as genetically adaptive. They tend to reduce morality to altruism. They cannot explain altruism toward non-kin. They have difficulty making the jump from genes to conscious and selfconscious people acting with moral intentions, and once this jump is made the basic assumptions of sociobiology and evolutionary psychology become unnecessary. They cannot provide a substantive account of morality and will lead to dire moral consequences if they become widely accepted.

Despite this negative account, which can be regarded as a bridgedemolishing effort, Smith acknowledges that we are biological creatures whose moral, believing properties must have come from somewhere. His (p. 328 ) positive account portrays morality as a by-product of human intelligence. Our abilities to anticipate the consequences of our actions, to make value judgments, and to choose among alternative courses of action combine to make us moral. Special emphasis is placed on self-consciousness as a capacity to step outside ourselves, preventing us from using our other mental capacities for more narrow utilitarian purposes.

Before I criticize these ideas, I want to express solidarity with Smith in two respects. First, there is much within evolutionary biology worth criticizing. Authors such as Buss (1999) and Cosmides and Tooby (2003) scarcely mention morality in their account of human nature and others such as Alexander (1987) attempt to reduce it to a form of self-interest, which is a non sequitur. The concept of selfish genes is entirely metaphorical and amounts to little more than newspeak for ‘anything that evolves by genetic evolution’. Morality is often simplistically equated with altruism. Inclusive fitness theory, with its emphasis on genealogical relatedness, does make altruism toward non-kin appear problematical. I agree with Smith that these are problems that need to be solved before we can understand morality from an evolutionary perspective. Far from denying these problems, I and many of my colleagues have been stressing them ourselves.

Second, I largely share Smith’s basic conception of human nature, society, and culture, as I have already recounted briefly here and in more detail elsewhere. My criticisms center not on the basic vision but on how it can be squared with evolutionary theory. Continuing the bridge building metaphor, I imagine myself as a person on one side of the gap calling over to Smith on the other side: ‘Hey! We’re over here!’

In this spirit, I will now attempt to show why Smith is heading in the wrong direction and how his bridge-building efforts can be redirected.

Adaptation vs by-product explanations of morality and religion

Any trait studied by evolutionary biologists can be an adaptation that enhances survival and reproduction or a non-adaptive by-product of the evolutionary process (or a combination of both). Smith is committed to a by-product interpretation of morality, which causes him to reject evolutionary accounts based on adaptation. His commitment extends beyond evolutionary considerations because he rejects all utilitarian and functional accounts of morality.

(p. 329 ) To see why Smith’s rejection of functionalism is premature, we need first to distinguish between individual-level and group-level functionalism. Consider three of Smith’s own examples. A woman returns a wallet to its owner even though she could have kept it without being detected. A citizen goes to the trouble of casting a vote whose impact on the election will be negligible. A soldier sacrifices his life in battle for comrades and country. These are intended as examples of people acting against their interests but in each case it is obvious that societies of people who act in these ways will function better than societies whose members behave otherwise.

As for these examples, so also for the more basic principles of morality. ‘Do unto others’ is the quintessential summary of morality that can be said while standing on one foot and is an excellent guide for a well-functioning society. All of the Ten Commandments that are not about serving one’s God are obviously functional at the societal level. The many hundreds of more detailed commandments that comprise Jewish law and have counterparts in other religious traditions are similarly functional at the society level. In most cases it makes little difference whether we interpret ‘function’ biologically or in the everyday sense of the word. At the most basic level, human welfare consists of such things as food, water, shelter, and freedom from persecution that are biologically adaptive. Moreover, the laws of the Hebrew Bible in particular are a sociobiological dream come true in their injunctions to be fruitful and multiply and more detailed commandments pertaining to reproduction. I am making these claims casually here but treat them more seriously in Darwin’s Cathedral and will return to the question of how they can be established scientifically to a skeptic’s satisfaction below. At the very least, Smith’s statement that religion ‘does not produce any obvious material benefit’ (2003: 95) cannot be taken as self-evident.

Another element of morality and religion that Smith regards as nonutilitarian is their categorical nature:

In these and all like cases, actions are performed at least in part because they affirm and express commitments to what are understood to be right, good, worthy, just and so on. And these are understood to entail imperatives independent of the actor’s own personal wishes and inclinations, and not because they might achieve some other valued outcome or benefit.

(Smith 2003: 10)

I do not mean to imply that moral systems are functional in each and every detail. The process of evolution involves many failures for each success and the results of this winnowing process are seldom completely functional. A sophisticated evolutionary perspective looks for adaptations without expecting them to be everywhere. A glimpse of what Smith is missing by prematurely rejecting adaptationism is provided by his own metaphor of human moral systems as like rafts upon the sea:

We see, then, that what any people, including ourselves, know about life and the world, about how life ought to be lived, is not founded on an indubitable, universal foundation of knowledge. These are not built on solid piles that have been driven down into a very bedrock of known reality that lies accessible beneath every human person. Rather, all of our knowledge and life practices—however obvious and well-founded they may seem to us—are built like large rafts on beams of particular trusted assumptions and beliefs that themselves float freely in the shifting seas of culture and history. And all of us, in our particular, historical communities of believers float together on those rafts, typically unable to see beyond our rafts to the open sea on which we float and thus accustomed to assuming our raft to be all that exists and is true.

(Smith 2003: 52–3)

The metaphor can be extended further. Suppose that the winnowing process works rather well. The rafts are not rafts at all but tall ships, elaborately constructed to ply the waters. Furthermore, there are many kinds of ships, since they must extract resources from their environment in addition to staying afloat and there are many ways of doing this. The ships have even crawled out on land and turned into terrestrial vehicles, inhabiting every region of the earth. The incredible diversity of vehicles is based in part on the many subsistence niches that can be occupied and in part on historical factors, since the winnowing process never unfolds in exactly the same way twice, even in identical environments.

All of this is missing from Smith’s conception of human nature, society, and culture because he has prematurely rejected evolutionary adaptationism in particular and functionalism in general in his account of morality. His rejection of functionalism is especially odd, given his desire to revive the sociological tradition of Talcott Parsons, one of the most flagrant functionalists of all time. Smith might think that he accepts some form of sociological functionalism while continuing to reject evolutionary adaptationism, but I argue in detail in Darwin’s Cathedral that any functionalist account (p. 332 ) in the social sciences must ultimately be based upon evolution, the only function-endowing process that exists apart from theistic explanations. The late social psychologist Donald Campbell made the same point when he stressed that all things functional were ultimately created by a process of blind variation and selective retention (1960).

So far I have defended the evolutionary position that Smith rejects. What about the evolutionary position that he accepts? Can morality be explained as a by-product of human consciousness and especially selfconsciousness? In the first place, this argument is woefully underspecified. No explanation is given for how these human capacities evolved. Are they biological adaptations that serve the selfish interests of individuals and their genes? If so, then why is morality any more authentic as a by-product of a biological adaptation than as a biological adaptation in its own right? If they are not adaptations, how can we explain them as by-products that produce morality as a second-order by-product? In the second place, it leads to a conception of morality that is excessively cerebral, as the following passage makes clear.

Self-consciousness gives rise to reflective distances between the self and its cognitions, emotions, and desires. And those distances provoke the quest for standards above and beyond the self’s cognitions, emotions, and desires by which they might be evaluated as worthy of thinking, feeling, and believing or not. [The philosopher Anthony] O’Hear writes: ‘the very fact of being self-conscious about our beliefs, of being in the full sense believers…initiates a process in which we search for what is true because it is true, rather than because it serves some interest of ours.’

(Smith 2003: 42)

One final comment will complete my discussion of adaptation vs byproduct explanations of morality and religion. Utilitarian explanations are (p. 333 ) often regarded as vulgar, as if morality and religion lose their beauty and majesty as soon as they are shown to be practical. I discuss this strange attitude at the end of Darwin’s Cathedral and will try to forestall it here with a final elaboration of Smith’s raft metaphor. Suppose that you are on vacation aboard a magnificent tall ship, gazing at a glorious sunset, marveling at God’s creation. Then the weather turns bad, so bad that within hours you are fearing for your life as the waves crash over the deck and the wind shreds the sails. Miraculously you survive, thanks to the resilience of the ship and the bravery of the crew. Suddenly you realize that the ship and crew have become beautiful, not the elements, and anything about the ship that interferes with its function has become hideously ugly. The stories of Joseph Conrad vividly convey the terror of a storm at sea and it is interesting that he described nautical life as appealing because of its moral simplicity. The problem was to survive.

The moral consequences of adaptationism

Smith regards sociobiology and evolutionary psychology as more than just factually inadequate; he also finds them morally corrosive.

Sociobiology and evolutionary psychology’s particular moral problems may be worth developing a bit, just to be clear about them. When human morality is redefined entirely in relation to reproductive fitness—so that morality is no longer driven by natural law or the will of God or self-evident inherent moral values—then we lose any real moral standard by which to judge actions. Genetic survival and extinction in a competitive environment is all that is. Beyond that we can have nothing evaluative to say about which genes successfully reproduce or how they do it. Indeed, we no longer even possess standards for value judgments about what constitutes progress in evolution. It is finally of no more value that humans survive than do bacteria. Why, on sociobiological grounds, should one be any ‘better’ than another? Furthermore, if some humans have genetic propensities that enable them to survive and thrive while, or even because, other humans die out, it is difficult, given the sociobiological account, to explain why this should not happen; or why, if it did, anyone should necessarily feel any moral concern or sense of tragedy about it. Some die. Some live. Natural processes work their way out. That is all. If any sociobiology or evolutionary psychology worth considering theoretically gives us any reliable normative direction, it is that the fittest genes should survive and the organisms that carry them should do what they need to do to ensure that outcome. Even then, any notion of ‘should’ makes little sense, really, since what will happen will simply happen by natural (p. 334 ) process. For sociobiology and evolutionary psychology to end up with any serious substantive morality requires smuggling in auxiliary assumptions and commitments that are alien to its own intellectual system…Sociobiology and evolutionary psychology’s case might be merely amusing, were it not for the monumentally misanthropic practical consequences that should follow any widespread embrace of its program. One can only hope that most moral, believing animals have more sense than to let that happen.

(Smith 2003: 37–8)

That man is the product of causes which had no prevision of the end they were achieving; that his origin, his growth, his hopes and fears, his loves and his beliefs are but the outcome of accidental collocations of atoms; that no fire, no heroism, no intensity of thought and feeling, can preserve an individual life beyond the grave; that all the labors of the ages, all the devotion, all the inspiration, all the noonday brightness of human genius, are destined to extinction in the vast death of the solar system, and that the whole temple of man’s achievements must inevitably be buried beneath the debris of a universe in ruins—all these things, if not quite beyond dispute, are yet so nearly certain that no philosophy which rejects them can hope to stand. Only within the scaffolding of these truths, only on the firm foundation of unyielding despair, can the soul’s habitation henceforth be safely built.

(Russell 1917: 93)

Science, theism, and the difficulty of testing evolutionary hypotheses

Evolutionary hypotheses are often criticized for being difficult to test, either because they involve events that took place in the distant past or because they are speculative adaptationist ‘just-so stories’. This is more than a mundane methodological issue: if evolutionary hypotheses can be tested and falsified with reasonable ease, resulting in the same accumulation of knowledge that we associate with the rest of science, then many arguments that marginalize the importance of evolution collapse. These arguments depend less on the importance of evolution than on the claim that we can never determine the importance of evolution.

With respect to Smith, let’s begin with his definition of religions as ‘sets of beliefs, symbols, and practices about the reality of superempirical orders that make claims to organize and guide human life’ (2003: 98). By superempirical, Smith means ‘an ordered reality that is not normally observable with the five human senses’. This term has the advantage of being broader than the usual term ‘supernatural’. It has the disadvantage that almost no one would have agreed with it 150 years ago. Prior to Darwin, the Christian conception of God was regarded as an eminently testable hypothesis that could be amply confirmed with the five human senses, either by themselves or when extended by the tools of science. Newton thought that he was studying God’s handiwork and one of the main motivations to study natural history was to understand the creator through his creations. Religious beliefs are not intrinsically superempirical—they become superempirical when they are driven from empirical inquiry by the scientific method. For example, it is perfectly possible empirically for the earth to be young, for each species to be separately created and unchanging through time, for organisms to be elegantly designed in every respect, and for life’s afflictions to be part of a benign grand plan when understood in greater detail. These claims could be true, but they are disconfirmed by the empirical evidence. In contrast, Darwin’s theory caused so many facts to fall into place (despite leaving many questions unanswered) that it compelled acceptance despite its disturbing implications. In short, the scientific method worked well enough for the subject of evolution (p. 336 ) to completely change the character of religion, forcing it to leave the empirical stage. Modern versions of ‘Creationism’ mimic the scientific method (the analogy with biological mimicry is worth exploring) but are a farce when examined carefully. The only alternative is to define God and other elements of religion in a way that cannot be falsified empirically—for example, by saying that he set the universe in motion but does not otherwise intervene.

Smith is careful to distance himself from ‘Creationism’, but his yearning to preserve theism in some form is clear. At first I was taken aback by passages such as the following.

Why are humans apparently unique among all animals in being profoundly moral animals? It may be impossible to answer this question definitively, but it is worth considering. Some people will say that humans are uniquely moral animals because they are made ‘in the image’ of a personal, moral God, who created them uniquely to reflect, know, and obey God. Other people will say that humans are moral because of the relatively large brains our species acquired through evolutionary development, which are neurologically capable of depths and complexities of evaluation and emotion unavailable to smaller brained animals. Maybe one or the other of these accounts is right, or maybe both are right.

(Smith 2003: 33)

Here, by contrast, is a theory that would be truly controversial, daring, and radical: human religions have existed and do exist everywhere because a God really does actually exist, and many humans—especially those not blinded by the reigning narratives of modern science and academia—feel a recurrent and deeply compelling ‘built-in’ desire to know and worship, in their various ways, the God who is there. Try publishing that, and we will find out who is controversial and daring. Of course, that theory, while not empirically verifiable, would certainly explain a lot. It is a most parsimonious theory. But prevailing assumptions of knowledge production rule it inadmissible. So (p. 337 ) we stick with other theories no more empirically verifiable or intellectually coherent but that at least fit our dominant narrative.

(Smith 2003: 117)

Passages such as these made me wonder if Smith was more intent on justifying his conception of theism than his conception of people as moral, believing animals. If so, then he is indulging in an intellectualized form of ‘Creationism’ and Oxford University Press should reclassify his book as theology. If not, then he should be happy to reject his conception of theism in favor of an evolutionary account that succeeds by normal scientific standards. To see how such an account is forthcoming, consider the part of Smith’s definition of religion that is not superempirical: ‘sets of beliefs, symbols, and practices…that make claims to organize and guide human life’ (2003: 98). This part is functional—surprising, given his earlier denial that morality has a functional basis. Religions (along with many other social organizations) are built and designed to tell people what to do. A system designed for this purpose is necessarily different from one designed factually to describe the natural world, just as a vehicle designed to move through water must be different from a vehicle designed to move over land. Furthermore, religions vary in what they tell their believers to do and this diversity can be explained with the same theoretical concepts and empirical tools that work very successfully for the study of non-human species. In this (p. 338 ) fashion, Smith’s theistic account of morality and religion will go the way of theistic accounts of the natural world.

I have used Smith’s book as a vehicle for addressing general issues in sociological and evolutionary theory. Moral, Believing Animals begins and ends with a conception of culture, society, and human nature that is not inherently theistic. There is a universal human nature that transcends cultural differences. This nature is moral and believing, giving rise to the cultural diversity that is the hallmark of our species. Establishing this conception within the social sciences (which in some respects involves reviving earlier traditions) will be a great achievement. Squaring it with evolutionary theory will be greater still. Empirically validating the conception so that it becomes the generally accepted ‘big picture’ for our species will be the greatest achievement of all. Evolutionary social constructivism works toward these goals. Smith’s theistic account of morality and religion does not and impedes scientific inquiry in general. I hope that the constructive intent of this chapter with respect to moral, believing animals is evident, even if some demolition with respect to theism was required along the way.