Religion in Human Evolution: From the Paleolithic to the Axial Age

by Robert N. Bellah

  External memory is a critical feature of modern human cognition, if we are trying to build an evolutionary bridge from Neolithic to modern cognitive capabilities or a structural bridge from mythic to theoretic culture. The brain may not have changed recently in it genetic makeup, but its link to an accumulating external memory network affords it cognitive powers that would not have been possible in isolation. This is more than a metaphor; each time the brain carries out an operation in concert with the external symbolic storage system, it becomes part of a network. Its memory structure is temporarily altered; and the locus of cognitive control changes.22

  But graphic invention and the external memory it makes possible are only the essential prerequisites for the development of theoretic culture, which is the ability to think analytically rather than narratively, to construct theories that can be criticized logically and empirically. Donald cites Bruner as describing the two modes of thinking evident in modern humans as narrative and analytic.23 And Bruner himself recognizes a distinguished precursor when he uses as the epigraph for his book the following passage from William James: “To say that all human thinking is essentially of two kinds-reasoning on the one hand, and narrative, descriptive, contemplative thinking on the other-is to say only what every reader’s experience will corroborate.“24 So analytic or theoretic thinking does not displace, but is added to, narrative thinking, a point essential to our understanding of the axial age.

  In one sense something like theoretic thought, the capacity to draw conclusions from instances outside a narrative context, goes all the way back: mimetic stone flaking surely required a degree of inferential thinking. At a practical level, “primitives” were as logical as we are, a major reason why Lucien Levy-Bruhl’s idea that they were “prelogical” has attracted such scorn.25 Even if we narrow our definition to something like conscious rational reflection, we can find instances earlier than the axial age. The practical need for calendrical accuracy in agriculture led even some preliterate societies to a kind of “primitive astronomy,” in which, Donald argues, many elements of modern science were incipient: “systematic and selective observation, and the collection, coding, and eventually the visual storage of data; the analysis of stored data for regularities and cohesive structures; and the formulation of predictions on the basis of these regularities … Theory had not yet become as reflective and detached as it later would; but the symbolic modeling of a larger universe had begun.“26 Begun, but, as perhaps in such fields as metallurgy as well, theory remained at the level of craft specialization, not challenging myth at the most general level of cultural self-understanding; there myth in the sense of ethically and religiously charged narrative remained largely unaffected by the new developments.

  What made Greece unique in Donald’s eyes was “reflection for its own sake,” going “beyond pragmatic or opportunistic science” and eventuating in “what might be called the theoretic attitude.“27 Donald does not relate his argument to the problem of the axial age, because he sin gles out Greece alone as the place where the theoretic attitude first arose, but Yehuda Elkana, while also focusing on Greece, relates his argument to the general axial problem in his contribution to the 1985 book edited by Eisenstadt, The Origins and Diversity of the Axial Age. His paper is entitled “The Emergence of Second-order Thinking in Classical Greece,” and what he means by “second-order thinking” is close to Donald’s “theoretic Following Donald, I am using the term “theory” in distinction to the term “narrative.” Elkana is concerned less with the distinction between narrative and theory than with, in Donald’s terms, the distinction between theory and “the theoretical attitude.” For Elkana, first-order theory can be quite complex, as can, for example, mathematics and the beginning of algebra in Babylonia, or the calendrical astronomy noted above, but it involves only straightforward rational exposition, not reflection about the basis of the exposition. Second-order thinking is “thinking about thinking”; that is, it attempts to understand how the rational exposition is possible and can be defended. One of the earliest examples is geometric proof, associated with Pythagoras in early Greece. Geometric proof asserts not only geometric truths, but the grounds for thinking them true, that is, proofs that in principle could be disproved, or replaced by better proofs. For Elkana the arguments of several of the pre-Socratic philosophers that the universe is formed from water or fire or mind, although clearly theories and not myths (we will have to ask later about the relation between such theories and myths), do not imply second-order thinking, as they do not seek to disprove the alternatives. One would think they did so implicitly, as each pre-Socratic offered in turn his alternative theory. The value of Elkana’s position, however, is not in the details, but in the help he gives us in seeing that “theory” precedes the axial age, at least in selected areas such as astronomy and mathematics, but that it is precisely the emergence of second-order thinking, the idea that there are alternatives that have to be argued for, that marks the axial age.

  Elkana quotes a passage from Momigliano that I cited earlier to make the decisive point: “New models of reality, either mystically or prophetically or rationally apprehended, are propounded as a criticism of, and alternative to, the prevailing Here we have, not theories about limited realms of reality, not even second-order thinking about a limited area of reality such as geometric proof, but second-order thinking about cosmology, which for societies just emerging from the archaic age meant thinking about the religiopolitical premises of society itself. It is second-order thinking in this central area of culture, previously filled by myth, that gave rise to the idea of transcendence, so often associated with the axial age: “Transcendental breakthrough occurred when in the wake of second-order weighing of clashing alternatives there followed an almost unbearable tension threatening to break up the fabric of society, and the resolution of the tension was found by creating a transcendental realm and then finding a soteriological bridge between the mundane world and the transcendental.” 30 But here Elkana, a historian of science, is, I think, skipping a beat. In the history of science the effort to make sense of what has come to be recognized as empirical anomalies that don’t fit existing ideas, leads to the creation of a new abstract theory, a new “order of reality” if you will, that succeeds in making sense of those anomalies. But “creating a transcendental realm” involves something more substantial than a scientific theory. Because transcendental realms are not subject to disproof the way scientific theories are, they inevitably require a new form of narrative-that is, a new form of myth. In Chapter 5 I noted that “mythospeculation”-myth with an element of reflective theory in it-already appeared in several archaic societies. The transcendental breakthrough involved a radicalization of mythospeculation, but not an abandonment of it.

  Akhenaten’s religious revolution in the middle of the fourteenth century BCE vividly illustrates the difference between myth and mythospeculation. It is not at all true that in a mythic culture there is no change-even the gods change. Some are forgotten, some demoted, some elevated to primacy. In Egypt the position of highest of the gods was indeed unstable: first Horus, then Re, then Amun or Amun-Re, then Ptah, then, in Ptolemaic times Isis, and so forth. None of these changes was traumatic; none of the gods who lost their primacy was denied existence. The way to change a mythic culture is to tell a different story, usually only a somewhat different story, which does not involve denying any previous story. The commonly remarked “tolerance” of polytheism, as noted by David Hume, for example,3’ is not the moral virtue of tolerance as we understand it today, but is part of the very structure of mythic culture. Some myths and the gods whose actions they recount may be more central than others, but the issue of truth and falsity doesn’t arise. The very idea of myth as “a story that is not true” is a product of the axial age: in tribal and archaic societies, believers in one myth have no need to find the myths of others false.

  But that is just what Akhenaten did: he declared that all the gods but Aten were false; he raised the cr
iterion of truth and falsehood in a way that drove a dagger into the heart of traditional Egyptian religion. As Jan Assmann puts it:

  The monotheistic revolution of Akhenaten was not only the first but also the most radical and violent eruption of a counter-religion in the history of humankind. The temples were closed, the images of the gods were destroyed, their names were erased, and their cults were discontinued. What a terrible shock such an experience must have dealt to a mentality that sees a very close interdependence between culture and nature, and social and individual prosperity! The nonobservance of ritual interrupts the maintenance of cosmic and social order.32

  But though Akhenaten cut to the root of traditional myth, he did not leave the mythic mode and, in some ways, was even quite conservative. The prime source of our knowledge of Akhenaten’s thought is “The Great Hymn to Aten,” which is still fundamentally narrative.33 Yet the “cognitive breakthrough” is clear enough. The Aten, the sun disk, is the source of light, and light is the source of life and of time itself. Ritual and myth are not abandoned, but they focus exclusively on Aten. James Allen has argued that, in finding light to be the fundamental reality of the cosmos, Akhenaten was more a “natural philosopher,” a precursor of the pre-Socratics, than a theologian.34 But Akhenaten was both. And what made him conservative was that he believed that Aten revealed himself only to him, the pharaoh, and only through the pharaoh to the people. In popular devotion, Aten was depicted together with Akhenaten and his wife, Nefertiti, all three as gods. In this respect Akhenaten’s religion reaffirmed the archaic unity of god and king, and however much a precursor, it failed to raise the critical question of the relation between god and king, the very hallmark of the axial age. Moreover, Akhenaten’s claim to be the exclusive channel for the relation of god and people took place in an age when personal piety, the direct relation of individuals to the gods, was on the rise.

  For many reasons, Akhenaten’s revolution failed: knowledge of his very existence was wiped out not long after his death, only to be rediscovered in modern times by archaeologists. The primary reason for the collapse, besides the fact that the revolution was far too radical for its time (other radical movements have survived on the margins of societies that rejected them) was that it was exclusively the intellectual product of its founder. When Akhenaten died, there were neither priests, nor prophets, nor a people to continue in the faith. Nonetheless, the fact that mythospeculation had made a cognitive breakthrough that would not be repeated for nearly a thousand years is indeed remarkable. It is an indication of the fact that, however slowly and painfully, the axial breakthroughs were the children of the archaic cultures from which they arose.

  But what I want to get at now, and what we will see more clearly when we examine the individual cases, is that “breakthrough,” that problematic word, does not mean the abandonment of what went before. Theoretic culture is added to mythic and mimetic culture-which are reorganized in the process-but they remain in their respective spheres indispensable. Theoretic culture is a remarkable achievement, but always a specialized one, usually involving written language in fields inaccessible to ordinary people. Everyday life continues to be lived in the face-to-face interaction of individuals and groups and in the patient activities of making a living in the physical world. It is first of all mimetic (enactive, to use Bruner’s term) and not in need of verbal explanation, but if linguistic explanation is necessary, it will most often be narrative, not theoretic.

  I have mentioned the fact that the face-to-face rituals of tribal society continue in disguised form among us. As an example, let me take the ritual handshake that is so much a part of our daily life. Arnaldo Momigliano tells us that the ancient Roman handshake, dexterarum iunctio, was an old symbol of faith, fides, that is, faith as trust or confidence, and that from very early times Fides was a Roman goddess. He says that there are good reasons for thinking that handshaking in Greece was an expression of pistis, the Greek equivalent of fides. Though normally the handshake simply confirmed the trustworthiness of an agreement, with perhaps an aura of divine protection, Attic grave reliefs suggest a further extension of the idea for they “show handshaking as a symbol of Faith at the parting between the dead and the living. Thus handshaking was not only a sign of agreement among the living, but the gesture of trust and faith in the supreme With us the handshake is hardly a conscious gesture, but nonetheless one does not expect to be attacked by someone with whom one has just shaken hands. A refusal of a proffered handshake, however, would make the ritual gesture conscious indeed: breaking the ritual raises ominous questions that would require an explanation.

  No one has argued more persistently than Randall Collins, following Durkheim and Erving Goffman, that daily life consists in endless “interaction ritual chains.” “Ritual,” he says, “is essentially a bodily process.” He argues that ritual requires bodily presence, and asks, rhetorically, whether a wedding or a funeral could be conducted by telephone or videoconferencing. His answer is, clearly, no. One could videotape a wedding or a funeral, but without the physical presence and interaction of the participants, no ritual could occur.36 But mimetic (enactive, embodied) culture does not just continue to exist alongside theoretic culture: it reclaims, so to speak, some of the achievements of theoretic culture. Hubert Dreyfus has shown in detail how skills learned with painstaking attention to explicit rules, through becoming embodied and largely unavailable to consciousness, are in the end far more efficient than they were at the beginner’s stage.37 His examples include driving a car and expert chess playing. In such cases the experienced practitioner knows “instinctively” what to do in challenging situations. “Critical thinking” (theoretic culture) at such moments would only disrupt the flow and produce serious mistakes. One can imagine such a process of embodiment going all the way back to Paleolithic stone chipping. What was initially learned by painful trial and error became, with practice, “second nature,” so to speak, even before there was any language to describe it. If we imagine that “moderns” live in a “scientific world” and have left behind such primitive things as ritual, it is only because we have not observed, as people such as Goffman, Collins, and Dreyfus have, how much of our lives is lived in embodied rituals and practices. This is not to say that ritual has gone uncontested: antiritual tendencies and even movements occurred in most of the axial breakthroughs, and periodically ever since. This is something we will have to consider closely as we go along. But in every case, ritual, when thrown out at the front door, returns at the back door: there are even antiritual rituals. Our embodiment and its rhythms are inescapable.

  If mimetic culture has interacted vigorously with theoretic culture once the latter has appeared, such is also the case with narrative culture. There are things that narrative does that theory cannot do. In I noted that narrative actually constitutes the self, “the self is a telling.“38 Not only do we get to know persons by sharing our stories, we understand our membership in groups to the extent that we understand the story that defines the group. Once theoretic culture has come into existence, stories can be subjected to criticism-that is at the heart of the axial breakthroughs-but in important spheres of life, stories cannot be replaced by theories. Because stories really have been replaced by theories in natural science, some have come to believe that such can occur in all spheres. Though efforts to create a science of ethics or politics or religion have rendered critical insights in those spheres, they have not succeeded in replacing the stories that provide their substance. When Aristotle, surely one of the greatest theorists of all times, begins his Ethics, he asks the question, what do people consider the highest good, and finds that the common answer is happiness. In short, he starts from opinion, from the stories people tell about what leads to happiness, and though he criticizes those stories, he doesn’t reject their substance. Aristotle agrees with the common opinion that happiness is the highest good-he brings his critical insight to bear in seeking to discern what will lead to true happiness. In short, he seeks to improv
e the common story with a better story, not with a theory. Some modern moral philosophers have sought to create an ethics based on “reason alone.” But when utilitarians say that ethics should be based on the consideration of the greatest good for the greatest number, they require a substantive account of the good to get started: they still need a story about the good. When deontologists try to get around this objection by distinguishing between the good, which is culturally variable, and the right, which is universal, they still require a story about the right that reason alone cannot produce. Efforts to create a “religion within the bounds of reason alone” run up against the same problem: they end up replacing old stories with new ones.

  Narrative, in short, is more than literature, it is the way we understand our lives. If literature merely supplied entertainment, then it wouldn’t be as important as it is. Great literature speaks to the deepest level of our humanity; it helps us better understand who we are. Narrative is not only the way we understand our personal and collective identities, it is the source of our ethics, our politics, and our religion. It is, as William James and Jerome Bruner assert, one of our two basic ways of thinking. Narrative isn’t irrational-it can be criticized by rational argument-but it can’t be derived from reason alone. Mythic (narrative) culture is not a subset of theoretic culture, nor will it ever be. It is older than theoretic culture and remains to this day an indispensable way of relating to the world.