This is the slightly revised text of a lecture delivered at the University of Sydney in November:

The Enlightenment attempted to comprehend human institutions, from language to political forms, in strictly human terms. This challenged religion, which had until then provided the chief resource—for the Christian West, embodied in the Bible—for understanding these institutions. The proponents of the Enlightenment reduced God’s role in human history or eliminated it entirely. Thus was engaged, some 250 years ago, the first widespread debate between theistic and atheistic anthropologies.

In the past few years, in the context of the incomparably more advanced science of today, a number of works have soared high on the best-seller list—Dawkins, Dennett, Harris, Hitchens…— that do little more than recycle Enlightenment arguments for the marginalization or elimination of religion. Some continue to dismiss religion a la Voltaire as a more or less pernicious delusion concocted by priests; the more sophisticated explain it as a survival mechanism that has historically enhanced, as all selected behaviors must, reproductive fitness. In reply, some religious writers, deprived of the traditional appeal to ecclesiastical authority, have rather pathetically attempted to adapt scientific discourse to their own purposes. The more astute offer the evidence of human subjectivity; science may provide material explanations for biological adaptations, but our own conviction of their truth is of a different nature than the mechanical causality offered in explanation.

Generative anthropology antedates this current controversy by nearly thirty years. From the beginning, it has situated itself at the minimal point of contact between belief and non-belief, between God-creates-man and man-creates-God. This is not to say that we merely split the difference; our point is (1), to define what exactly it is that distinguishes the human from its predecessors, and (2), to hypothesize how this new element might have emerged from what preceded it. These desiderata are qualitatively sharpened by the reflection that the first question does not demand an “analog” answer—a series of traits (opposable thumb, upright stature, larger cortex…) that each bring their bearer a step closer to true humanity—but the “digital” one of… digitality, in other words, language, or more generally, representation. A protohuman can look like a human and walk like a human, but it isn’t human unless it can talk like a human. It is language that provides the basis of the subjectivity rightly alleged by the religious thinker as proof of a transcendental realm irreducible to material causality.

The place of GA in the current debate is not simply neutral. The combat is an unequal one, in which the secular side has the advantage of its adversary by the very fact of presenting their disagreement as a debate. Religious thinkers claim to reject the “God of the gaps” whose presence is attested only by lacunae in scientific knowledge, since this makes God a mere placeholder for science. Instead, they allege our intuitions of truth, beauty, and morality as proof that our minds cannot be reduced to objects in a causal chain. But however higher the ontological status of this “cannot” than the conundrums of genetic theory alleged by defenders of Intelligent Design, it defines a “gap” nonetheless: the argument remains “God must exist because science cannot explain X.” That the qualia of human subjectivity cannot be explained by evolutionary psychology cannot bear the weight of demonstrating the existence of a divine Being.

Religious believers argue from inexplicability, atheists from explicability. The first group need not reject the methods of natural science, but it considers, unlike the second, that human experience contains something that, not simply in fact but in principle, cannot be explained by natural science. This leads to a dialogue de sourds in which both sides rely on unchallengeable premises. How can we determine what science can or cannot explain? The only source of evidence for such a statement is the natural world, which includes the world of human experience; but no natural phenomenon can be excluded in principle from the purview of science.

What is required is a minimalistic approach that, instead of beginning in medias res and attempting to demonstrate the existence or nonexistence of God on the basis of whether the “evidence” can be otherwise explained, takes as its burden to provide a parsimonious model of how and why we conceived of God in the first place. This new way of thinking we call generative anthropology takes on the responsibility of furnishing a minimal model that plausibly demonstrates how beings like ourselves, capable of using language and of affirming faith in transcendental Beings, might have arisen within a world where no such faith—not necessarily to say, no such Beings—previously existed. Since the very notion of a transcendental being has no apparent necessity in the world as depicted by natural science, the burden of proof is on the scientist, not simply to determine whether such beings exist “in fact,” but to explain why it is we conceive of them.

Existence may be more than a simple attribute of the hypothetical divinity, as Kant claimed in rejecting Anselm’s “ontological proof,” but neither is it as crucial as it seems, whether to believers or nonbelievers. From an anthropological standpoint, whether God exists is a far less critical problem than the indubitable fact that the concept of God as the object of belief exists, for the latter is just as great a mystery as the former. Indeed, the claim that God does not “exist” makes his conceptual existence even harder to explain. To the extent that the argument between atheism and belief can be resolved, it is through the exhaustion of the difference between the existence of God and the concept of God. We may express this exhaustion by “defining” God as the being that is indistinguishably the signified (or meaning) and the referent of the signifier “God”—the being that makes the “ontological proof” undecidable. The originary hypothesis is offered as a plausible embodiment of this “definition.”

Attempts have indeed been made to explain the God-concept in naturalistic terms. The original Enlightenment thinkers saw no great difficulty in attributing the idea of God either to conniving priests or to human ignorance and self-delusion. A century later, Durkheim took a key step by recognizing—an achievement confirmed another century later by evolutionary psychology—that religious ideas and above all, religious organization possess survival value, not just for conniving or deluded individuals, but for the society as a whole. Nevertheless, where religious apologetics takes for granted an explanatory “gap” in natural reality that only God can fill, natural science just as uniformly takes for granted that the natural suffices to explain the transcendental. It is assumed without further ado that the myths and rites of religious representation are implicit in “natural” human attributes: language, supposedly of “practical” origin, and ritual, presumably universal among the higher animals. Language is conceived as an extension of animal communication systems, admittedly with a new neurological substrate, that can be explained by pragmatic advantages (“the food is over the hill”); the leap from designating phenomenal reality to postulating the existence of transcendental beings is presumed to require no explanation beyond its obvious expediency. Those things in nature, such as the weather, for which we cannot establish mechanical causality, we attribute to an “anthropomorphic” divinity. In Religion Explained (2001), Pascal Boyer provides plausible explanations for the attributes of transcendental beings, but without concerning himself with the nature of transcendence itself as the substantial basis of these attributes. Even assuming that our ability to conceive such qualities as omnipotence or immortality requires no transcendental agency, merely attaching these qualities to human-like subjects generates something closer to Superman than to God. Transcendence trumps all attributes; it is the basis of attribution itself.

To sum up, religion seeks in natural experience evidence that it is inexplicable without the transcendental, but can only produce “gaps” in the models by which we understand this experience, whereas science does away with the possibility of gaps by reducing the transcendental from an independent ontological realm to the level of a concept. In the debate between those who do or do not find “evidence” for the existence of a transcendental being, the first fail to determine the minimal conditions under which such a being is necessary, whereas the second fail to determine the minimal conditions under which such a being might be thought to be necessary. What both religion and science need is a clearer idea of where to locate the transcendental in human experience.


The originary hypothesis does not make it easier for either side to declare victory in the debate, but it reveals a symmetry between them that has the potential to increase their mutual respect and open further possibilities of dialogue. By substituting the reconstruction of an originary event for the search tous azimuths for “evidence,” GA escapes the fundamental weakness of metaphysics that all the “post-metaphysical” philosophies and quasi-anthropological doctrines of the past two centuries, from Feuerbach to Derrida, have been unable to shed. The empirical mindset that is appropriate, notwithstanding epistemological caveats, to the discovery of the natural world, is not suited to dealing with a situation where our very ability to formulate “empirical” truths about the natural world is the object of our explanation.

GA occupies a new territory at the meeting point of religion and human science. Its hypothetical originary event is a plausible model of these minimal conditions that offers a new way to think about the human and its origin. However important it may be to refine the details of the hypothesis, these details are qualitatively less important than the fact of the hypothesis itself: that there is an originary event and that any attempt to think the human independently of such an event is necessarily inconclusive, requiring either a leap of faith in the normal, religious sense, or a leap of faith to the claim that no such leap is required.

Rather than go over once more the hypothetical details of the originary scene, let us attempt to make more precise the limits of GA’s ambition to be a new way of thinking. Unlike “grand narratives” modeled on the Hegelian dialectic, GA has no pretension of being the anthropological last word. Yet neither does it adopt the spurious modesty of postmodernism by presenting itself as just one more fiction with no more claim to truth than any other.

A useful parallel is an example that I sometimes think I am the only person who continues to take seriously: Francis Fukuyama’s affirmation, inspired by the breaching of the Berlin Wall in 1989, that the world had realized Hegel’s “end of history.” Nothing is easier than to point out that (1), a priori, history continues; humanity is historical in essence, the “end of history” is an eschatological vision that cannot by definition be realized other than catastrophically, and more particularly, (2), what we normally call “history” has not ceased with the collapse of the USSR; 9/11 was a wake-up call in this regard, not to speak of the rise of China, which incidentally throws doubt on Fukuyama’s basic idea that liberal democracy is the ultimate form of social organization.

Yet, in an important sense, it is impossible to conceive that, whatever the details and qualifications, a more “advanced” form of human society can be found than that which maintains free markets in political as well as economic exchange. We may dream of a future society run by flawless computer programs, robots, or what have you, but it would no longer be a human society. Market exchange, where the whole cannot be predetermined by any of its parts, will always generate more degrees of freedom, and thus possess greater capacity for invention and creativity, than authoritarian systems; the market recycles into creative activity the resentments that in such systems are directed against the system itself. The Chinese model will pose a real challenge to Fukuyama only if China comes to economically outperform Western democracies such as the US while retaining its authoritarian political structure. (We may recall Khrushchev’s “we will bury you.”)

But even the definitive triumph of liberal democracy would not mean that we are truly at “the end of history,” any more than the triumph of heliocentrism marks the end of astronomy. Within the framework of liberal democracy, various national and international institutions would continue to interact, compete, and evolve, just as our own society continues to evolve. For example, the concept of liberal democracy does not tell us how to resolve the tension between global markets and national political systems. Historical change, not unlike biological evolution, focuses its efforts at different targets at different times; as old problems are solved, creative energy is diverted to new ones not previously anticipated.

Similarly, if GA claims to bring about the “end of history” of the metaphysical era in which, on the one hand, propositional language is taken for granted, and on the other, religious narratives attribute to a preexisting transcendent being the creation “in its image” of lesser beings—ourselves—possessing the capacity of transcendence, this does not spell the end of either religion or anti-religion. Even if GA realizes its ambition to become the meeting-place where the conversation can be pursued in more rigorous and parsimonious terms, both sides will react to this development in unpredictable ways. Nevertheless, just as liberal democracy has demonstrated its superiority as well as any social system can, I think that GA, even in its present state, gives proof of the qualitative superiority of the anthropology that begins with the originary scene.

I think that the parallel between GA and liberal democracy can be developed one step further. The familiar binary dispute between the political right and left is more paradoxical than we generally realize. Although most of us have concluded in favor of one of the two sides, we all recognize that in a democracy, there must be a debate between them. We want our side to win every election, yet we realize that a one-party system would be unhealthy, as would be a system based on ascriptive characteristics, such as the tribal-based parties often found in third-world countries. We each have our own ideas, but we recognize, however grudgingly, that the wisdom embodied in the system sometimes requires giving credence to opposing ideas, and modifying our own in consequence. The analogous debate between religion and atheism is not generally seen as having a similarly systemic function. Yet if the “ultimate” political system requires an ongoing debate between left and right, so perhaps does the “ultimate” anthropology.

In the United States, there is in fact a fairly close correlation between the two parties and the two positions on religious belief. This parallel has sharpened in recent years. The left and right have been at war at least since the French Revolution, but the corresponding struggle in the religious sphere—which in France went on unabated until after World War II—focused on the power of the Catholic Church rather than on the general question of religious belief. The immediate stimulus to the current spate of atheism books was the American left’s hostility to the Bush administration and its fear, sometimes articulated in near-hysterical terms, that the “religious right” intended to impose the teaching of creationism along with or in place of evolution, abolish legal abortion, forbid stem-cell research, and so on. Many in this group view Islamic jihadism as analogous to, or even as an understandable response to, Christian fundamentalism.

Beyond these polemics, the politicization of the religious question raises a much broader issue. Whatever else one can say about left-right political debate, the airing of political questions stimulates the search for new solutions in the hope of gaining the support of the ever-elusive majority. Neither side need be motivated by a “disinterested” search for the truth; in a liberal democracy, the only “truth” obtainable in these messy debates is a shifting series of compromises between moral values, economic efficiency, military preparedness, and so on.

What similar benefits, then, do we incur from the similarly messy debate between believers and atheists? The productivity of religious debate can be theorized only from the perspective of an anthropology capable of defining the common goal of both sides. That this common goal is indeed anthropological is intuitively obvious, but the idea of the “anthropological” has been so obscured by political issues that in itself, without the adjective “generative” preceding it, it is no longer very useful.

Social science, armed with evolutionary theory, explains religion as a mode of adaptive behavior. In turn, the more astute defenders of religion, realizing the inherent emptiness of the pseudo-sciences of Creationism and Intelligent Design, have reconceived the divinity as the guarantor not of an eschatological apocalypse but of an unending process of creation. These two models, that of evolutionary psychology, which sees religion within a biological context, but one that takes into account the existence of human culture and language, and that of “process theology,” in which, once more on the evidence of our subjective intuitions, the prior existence of the divine is subordinated to the construction of divinity in the real world, have far more in common than the polemical tone of their respective arguments would lead us to believe. Evaluating these models in the context of the originary hypothesis makes this similarity explicit, while clarifying where their difference lies. In the one case, language and culture are respected in fact, but not in their specific role in human ontology, so that their genesis is not explained. In the other, the human appetite for transcendence is respected, but only as a reflection of the divine, with the result that its provenance in human affairs is explained in maximal rather than minimal terms.

I do not fear that GA’s minimal explanation will put an end to this conversation. On the contrary, providing the debate with a clearer sense of its foundations can only make it more vigorous. Generative anthropology has no pretension of becoming the “end of history” of anthropology. It claims only, like liberal democracy, to stand at the end of one phase of history and at the beginning of another.