Response to Louise Antony
By Victor Kumar and Richmond Campbell
We admire Louise Antony and hope to learn more from her work in the future. However, Antony’s critical review of A Better Ape in NDPR rests on misrepresentations of our ideas and mistakes about the science. We’ll explain why her five main empirical criticisms are groundless, then we’ll explain why her normative criticisms are also unfounded.
I. Antony’s empirical criticisms are groundless
First, a quick summary of Parts I-III of our book, where we lay out a theory of the evolved moral mind. Through gene-culture co-evolution, humans evolved moral emotions like trust and respect as well as the capacity to learn and internalize norms. We hypothesize that the co-evolution of moral emotions and norms generated values that exist in nearly all known human societies (harm avoidance, fairness, etc.). Moral reasoning evolved in part to help humans reduce moral uncertainty and resolve moral conflicts. More speculatively, we explore how the cultural evolution of major social institutions like family and religion shaped the moral mind in prehistory, expanding moral circles but also stabilizing social hierarchies.
1. Antony misrepresents our method
There are many problems with [Kumar and Campbell’s] story… What is sorely missing here is not just direct evidence for such extravagant claims, but even an idea of what such evidence might look like… For example…in support of [a claim about cooperative parenting] the authors cite Sarah Blaffer Hrdy’s 2009 book Mothers and Others…. [But] Hrdy has no direct evidence about family life pre-Pleistocene–how could she? … [K & C] insist that their own theory is not a mere “just-so” story. But their own account fails to provide the kind of specific evidence needed to evade this charge.
This criticism misrepresents the book by ignoring our explicit method. We begin by stating that “the most plausible theories of human evolution are informed by a wide range of evidence from many scientific fields,” such as primatology, developmental psychology, anthropology, archaeology, and at least a dozen other fields. “The aim is to triangulate” (6). Our method operates at a higher level of abstraction:
This book is an attempt to triangulate from the triangulators…. Well-confirmed scientific theories are our data. We sift through multidisciplinary evolutionary science, find ideas that are empirically robust, cast aside ideas that are unsupported or clouded by bias, build on what’s left, and construct a single coherent picture of human evolution…. By drawing on convergent evidence from many different sources, our evolutionary hypotheses about the moral mind in Parts I and II of the book aspire to “how-probably” (6–7).
In other words, while researchers like Michael Tomasello, Joseph Henrich, and Cecelia Heyes build theories of human evolution via diverse evidence and inference to the best explanation, our method is to synthesize and build on their theories. There might well be problems with this method. But to show that you would have to describe the method first, which Antony does not, much less critique it. Antony is evidently skeptical about any attempt to theorize about the minds of ancient ancestors, believing the evidence radically underdetermines theory. This is a response, however, not to us specifically but to research in the field at large.
2. Antony misunderstands language use in evolutionary theory
Another pitfall to adaptationist thinking is attributing directionality to evolution. K & C surely know that nature has no intentionality, that natural selection is a filter, not an inventor, and that the environment cannot evoke useful mutations. However, they persistently write in a way that encourages these ways of thinking. Consider these examples:
Culturally transmitted information…was so important for our survival that there was selection on our brains to evolve cognitive capacities designed specifically to take advantage of it. (72)
…natural selection remodeled human brains for the sake of enhanced social plasticity. (105)
Indeed, we are such an intensely cultural species that our biology has evolved in quite radical ways to facilitate culture. (65)
This is false. Phrases like “evolving to” and “for the sake of” are all bog-standard locutions in evolutionary theory, shorthand for biological function that arises out of random variation, inheritance, and natural selection. Antony’s criticisms here are reminiscent of Jerry Fodor’s failed criticisms of Darwinism.
3. Antony misunderstands cultural evolution
[K&C’s] official line is that the currency of cultural evolution is “information” (64), and that the measure of cultural fitness is not number of biological offspring, but number of “students”…. But K & C quickly and persistently conflate this normatively neutral and evolutionary respectable sense of “fitness” with the normatively loaded and misconceived ideas that the “fittest” ideas are also the best — the ones most likely to be useful…. In general, they continue, there is a “tendency to imitate individuals who are successful…. Standards of success that didn’t track reproductive fitness would not have lasted long” (68). (There is, by the way, no evidence cited in support of any of these claims.)
This, too, is false. None of the claims Antony attributes here are original to us. We are simply elucidating basic elements of gene-culture co-evolution, as developed primarily by Richard Boyd and Pete Richerson, along with their students and collaborators. According to this theory, human minds evolved biologically to acquire useful/adaptive information from group members. For example, humans have social learning mechanisms that involve attending to and imitating other humans that are successful and prestigious. Why? Because there was a biological advantage to drawing on accumulated knowledge in one’s group instead of figuring things out for oneself. Antony’s claim that we conflate descriptive and normative concepts is an artifact of her confusion about gene-culture co-evolution. In fact, the book draws that very distinction and explains how evolution can inform ethics without illicitly jumping the is-ought gap (10–12).
4. Antony makes three mistaken criticisms of our Darwinian arguments
Antony says, “K & C claim to recognize that “not all human traits are adaptations,” but they except “traits that exhibit ‘adaptive complexity,’” writing that in such cases, “Darwinian explanations are inescapable” (5). Traits with adaptive complexity are “composed of a set of interlocking mechanisms that could not exist except for the fact that they perform a crucial function” (5). She goes on to make three criticisms, each of which is mistaken.
First, Antony says, there is “circularity inherent in pre-classifying a set of traits as possessing adaptive complexity in order to argue that these traits are adaptations.” This complaint is grounded in a confusion about terminology. “Adaptive complexity” is a label (from Daniel Dennett) for a complex system that helps an organism to survive or reproduce, a system that has the appearance of being designed for that purpose. Once you see that this is what “adaptive complexity” means — not adaptation — the circularity vanishes.
Second, she says, “contrary to the authors’ confidence pronouncements, [there is] no consensus that an appeal to adaptation is either necessary or sufficient to explain complexity.” This is false. There is near universal consensus among evolutionary theorists that other than human intention, natural selection is the only thing that can explain complex design (like the eye).
Third, “justification of [K & C’s] strictly adaptationist story presupposes their central claims about human morality–that morality is based as much on sentiment as on reason, that moral emotions are separate from each other, that morality is pluralistic rather than unified, and that the involvement of language and analytical thought are incidental rather than intrinsic.” To begin with, we argue that norms and moral reasoning involve language and analytical thought. So, neither language or analytical thought is “incidental rather than intrinsic” to the moral mind. The other claims that Antony lists are just plausible conclusions of influential work in the field, by researchers such as Shaun Nichols, Jonathan Haidt, and Paul Bloom (all of whom we cite). We use evolution not to establish the truth of these conclusions but to explain how they make evolutionary sense. Once again, Antony is mistaken about knowledge in scientific moral psychology.
5. Antony misunderstands the significance of Chomskian moral nativism
[K & C] presume that the only argument for positive native moral structure is the need to explain “what different moralities have in common” (94). But this presumption about the methodology behind nativist views is false. K & C ignore poverty-of-stimulus arguments, the most powerful methodological tool there is for defending nativist hypothesis…. Susan Dwyer and John Mikhail have each argued that there are striking parallels between the structure and development of morality on the one hand and language on the other, suggesting that there is a “universal moral grammar” analogous to the “universal grammar” posited by Chomsky…. [K & C] culpably misrepresent the state of knowledge about these matters.
Anyone attending to research in cognitive science over the last two decades knows that poverty-of-stimulus arguments are (or seem to be) seriously challenged by the revolution in statistical inference learning. Moreover, poverty-of-stimulus arguments in moral psychology have been fatally undermined by research on moral learning, especially work from Shaun Nichols and collaborators. Simply put, the stimulus is far from impoverished given rich capacities for statistical inference. Antony accuses us of ignoring the state of play in the field, but it’s precisely because we are aware of current research in moral psychology that we believe the prospects for Chomskian moral nativism are dim.
To write a book covering the entire evolution of morality requires making judgments about which work is most worth engaging with. Antony repeatedly claims that the book displays ignorance about research germane to the evolution of morality. But anyone familiar with the field will know that a book on moral/social evolution that engages explicitly with Michael Tomasello, Sarah Hrdy, Frans de Waal, Paul Bloom, Richard Wrangham, Christopher Boehm, Richard Boyd and Pete Richerson, Herbert Gintis and Samuel Bowles, Joseph Henrich, Shaun Nichols, Cecilia Heyes, Kim Sterelny, Hugo Mercier and Dan Sperber, Jonathan Haidt, Joshua Greene, and others is well grounded in the research. Even if it ignores Dwyer and Mikhail.
II. Antony’s normative criticisms are unfounded
Antony saves her harshest criticisms for the ethical/political portion of the book. These criticisms are based on distortion of our ideas. We’ll explain our actual view in the course of unpacking each distortion.
[K & C’s] theory of moral progress is simple: injustice is the result of segregation, and justice is achieved when segregation is overcome. This will happen when privileged individuals are brought into the right kind of contact with members of disadvantaged groups, as they then come to see and understand the suffering of these groups and are motivated to make changes…. The authors would have you believe that slavery was abolished largely because of a sea change in public awareness of its brutality.
In fact, we do not claim that racism and other forms of oppression have abated in the past, or will dissolve in the future, simply once dominant groups see how victims suffer. The central aim in the final, normative part of the book is to use ideas about positive feedback loops in cultural evolution to shed light on one mechanism that underlies cumulative progressive change. Beginning with “inclusive moral progress” (expansion of the moral circle), we argue that social integration between ingroups and outgroups often leads to recognition of the harm outgroup members suffer and exposes the false ideology that denigrates them. This leads to attitude change, which fosters institutional reform, precipitating further social integration, and the loop continues. Progress accumulates.
Antony claims that this view makes false predictions about progress for women and Black Americans. This claim misrepresents the book. We devote an entire chapter to “egalitarian” moral progress (amelioration of unjust social hierarchies) and discuss at length the very cases Antony claims we cannot account for. We argue that integration of subordinated groups into roles occupied by dominant groups — roles with more power and social status, especially decision-making roles in major social institutions — contributes to cumulative egalitarian progress. These forms of integration lead to social knowledge and moral change, which fosters more integration: progress loops. Antony thinks this is “banal.” As she puts it, memorably, “Gee, why didn’t EVERY ACTIVIST ORGANIZATION THAT EVER EXISTED think of that?”
Synthesizing and building on insights of past work, what we are offering here is a new argument for an existing position, along with new inflections on it. For example, the need for democratic integration is not just about “political institutions,” as Antony says, but about the way in which diverse leadership in other major institutions, such as religious institutions, can shape moral minds and thereby positively impact people’s lives.
We are very explicit that our model highlights only one mechanism underlying moral progress. We don’t even claim that this mechanism is more powerful than others. So why don’t we focus on other mechanisms, such as the activists that Antony highlights? Because that would be a different book altogether. Our book is about one central and recurring idea: that morality evolves with knowledge and social structure. We hypothesize that the cumulative co-evolution of morality, knowledge, and social structure drove major transitions in human history.
Moreover, Antony leaves out the most important part: our purpose in focusing on this particular mechanism of moral progress is to contribute to non-ideal theory in ethics/political theory (189–98). In short, instead of imagining what a perfectly just world looks like (a hard task) moral and political theorists can identify psychological and social mechanisms that reliably lead societies to discoveries about moral progress (still hard, but easier). According to Antony, our view is that epistemic insight is the whole story about how moral progress happens. In fact, we are just trying to articulate how epistemic insight about moral progress often happens, conceding over and over that many other factors underpin moral progress. The book is intended for a general audience, so we don’t use technical philosophical language, but what we are doing is naturalistic moral epistemology.
III. Antony misunderstands the point of the book
Only one of Antony’s criticisms has any bite: that the book attends selectively to evidence and theories. Thus, Antony argues, we ignore evidence for certain forms of moral nativism, we do not rule out the possibility that elements of the moral mind are mere byproducts of selection, etc.
A Better Ape synthesizes theories from dozens of leading researchers on moral and social evolution, but it doesn’t rule out all rival hypotheses. Our approach is more rigorous in the 20+ moral psychology articles we’ve published in peer-reviewed journals (e.g., in Ethics, Nous, and The Journal of Philosophy). But it’s necessary to be somewhat selective in a book that covers the evolution of morality from ancestral apes to the present day, that presents a positive picture instead of getting bogged down in criticism of alternative theories, and that is accessible to scientists, students, and those outside academia. We think books like that are worth writing. We also think our attempt yields some original, interesting ideas — it would even be interesting to know that those ideas are false. Here are five of these ideas (notably, Antony engages with none of them):
1. Non-human apes lack deontic moral norms, but an even more fundamental difference is that humans have a richer set of moral emotions, particularly trust and respect, that evolved biologically in response to cooperative human culture. (Ch. 2)
2. There are deep moral similarities across societies not because any moral norms are biologically innate but because innate emotions have shaped the cultural evolution of norms. For example, groups share autonomy norms in part because these norms resonated with moral feelings of respect and were therefore favored in cultural transmission. (Ch. 4)
3. Complex reasoning evolved as a social activity scaffolded by morality. Moral norms about honesty, fairness, and autonomy improved our ability to reason together cooperatively, avoiding confirmation bias and other pitfalls of reasoning in isolation. (Ch. 5)
4. Human morality has a dark side: it undergirds unjust social hierarchies, and the reason is that hierarchical social institutions were selected for culturally after the agricultural revolution, and these institutions created new moral norms commanding submission to authority. (Ch. 7)
5. Humans do not gain moral knowledge through reasoning alone; they must learn about each other via integration in physical space and social roles. Progress accumulates in part because social integration and moral knowledge feed one another. Moral regress accumulates in part because social segregation and moral ignorance feed one another. (Ch. 9 and 10)
The principal running idea in the book is a very abstract empirical hypothesis: that major transitions in human evolution occurred through co-evolution between morality, knowledge, and social structure. This interplay seems to explain the evolution of our genus, our species, behaviorally modern humans in prehistory, and even the evolution of moral progress and moral regress in recent recorded history.
As we say, “Our aim is to start a conversation about evolutionary science and moral progress, not to offer the final word. We hope that critics can improve upon our theorizing” (228). In other words, we welcome strong criticisms — but conversation will be productive only if critics portray our ideas accurately.