American Scientist

Human Natures: Genes, Cultures, and the Human Prospect. Paul R. Ehrlich. xii + 531 pp. Island Press, 2000. $29.95.

Renowned biologist Paul Ehrlich has published on everything from butterflies to population-environment dynamics. His new book, Human Natures, considers the "gene-culture" interface, aiming to explore for nonscientists the range of what we know about human evolution, both biological and cultural. A central point is that humans possess not one nature, but multiple natures that arise in response to ecological and social conditions. Ehrlich draws on a diverse body of literature, attempting an in-depth synopsis of the relationship of human beings to the rest of the biosphere. We have evolved to notice and react to immediate changes in our environment. But today, we live in such evolutionarily novel environments that we confront problems nothing like those that faced our ancestors. And here is the link Ehrlich explores: What can we learn about genes, environment, culture and their interfaces that we can use to shape our future?

He begins (in chapter 2) by trying to tease apart the fact that genes dictate nothing and environments influence how (and when) genes can find expression. But then he seems to throw the baby out with the bath water, essentially arguing for the rest of the book that genes are irrelevant to anything we wish to ask about human behavior. Ehrlich goes to great lengths to argue (in chapters 12 and 13, for example) that genes really don't matter. He appears not to recognize how "variation in response to local conditions" has become a central paradigm in exploring human behavior in behavioral and evolutionary ecology, life history theory and evolutionary anthropology.

We all know it is a long way from testing trait-environment correlations to imputing genetic change, even in simple species. Yet, reproductive advantage to particular traits in specific environments is far from meaningless. This is a tangled and complex problem; so far none of us has gotten it exactly right. Even for very simple organisms, teasing out the relative importance for current conditions of natural selection, historical accident and phylogenetic constraint is difficult. In bacterial populations, Richard E. Lenski and Michael Travisano (1994) needed 50,000 generations and controlled conditions. How on earth should we expect to demonstrate cleanly for human beings the effects of natural selection as population-specific differences in equilibrium frequencies of particular alleles? If we cannot, should we assume that genes are irrelevant?

When we look at human behavior as problems of constrained optimization (as, for example, Lee Cronk, John R. Krebs and Nicholas Davies have done), several things become clear. Selection cannot see ahead but favors locally "better" versions of existing traits, not global optima. Whether this means that a particular form of allele will come to some equilibrium frequency is quite another matter. Remember that models of selection tend to focus on one or two alleles—but selection is operating on 28,000+ loci in humans, environments may change, and not all strategies may be equally available. So we can hardly expect clear demonstrations of genetic changes in humans due to natural selection to be common.

But shall we spurn the predictive ability of our half a loaf of information? For Homo sapiens as for other species, environmental conditions (both physical and biotic—including social—environments) constrain immediate costs and benefits for particular alternative actions; intergenerational transmission (both genetic and cultural) is important; and the past may constrain what options are available. The issue is not dichotomous (genes versus culture); rather, it is the impact of the relative contribution of all these elements.

Has natural selection been replaced entirely by cultural selection? Even Ehrlich wavers here, sometimes siding with "free will" and sometimes claiming that evolution has provided restraints. I found unsatisfying his discussion of ethics and free will. He observes that "Natural selection couldn't possibly operate at a sufficiently fine scale to promote neuronal connections that would program solutions to all [ethical] dilemmas. . . . There aren't enough genes to code the various required behaviors, it's difficult for selection to do just one thing, and the required speed of evolution is often much too fast."

But the complexities I've noted suggest that there will be no easy answer. Further, others have examined this problem—the ways in which ethical systems are likely to be constrained by our evolution—yet Ehrlich ignores them or cites them only for minor points. For example, Richard Alexander's Biology of Moral Systems (1987) is a major exploration of just what our evolutionary heritage means for patterns of ethical and moral systems, beginning with genetic self-interest, and the resulting conflicts of interests among individuals. Ehrlich cites Alexander's work, with others, only trivially (for example, in chapter 13, footnote 7, as support for the idea that the concept of blame is necessary for even simple human societies to function; and in footnote 20, as a source that discusses the naturalistic fallacy). At the core of whether we can convince large proportions of the human population to accept personal cost for the greater good is the question of how common truly genetically altruistic behavior is; Alexander treats this at length. Ehrlich's treatment is in footnote 44 of chapter 14, along with an assertion (on page 312) that "there can be behavior that is truly altruistic, and as far as we know, that behavior is unique to human beings." Yes—but what is at issue is not the existence of such behavior, but its likely frequency. Is it common enough that we can rely on human altruism to shape our global future? This is a problem central to Ehrlich's claim that we can do whatever we wish, if only we will wish it. Surely it is too important to be dealt with by assertion.

Unfortunately, I found this pattern consistently in Ehrlich's treatment of the literature I know best. Central references are sometimes cited casually, whereas secondary sources may be related in detail—and opposing sources are often cited without reconciling how they relate to each other. He cites several papers (in chapter 12, for example) as arguing that genes dictate behavior and then refutes them—even though they contain no such argument to be refuted. A typical Ehrlich response, given in the spirit of opposition, is as follows: "After all, even for behaviors directly related to sexual activities . . . predispositions clearly are very labile in response to environmental influences." That statement is actually the point of several of these papers, which examine variation in response to environmental conditions, gaining considerable insights by focusing on constrained optimization. Ehrlich imputes to the authors a "change in gene frequency over time" argument that none of them actually makes. Does Ehrlich confuse "old style" sociobiology, often focused on human "universals," with behavioral ecology and evolutionary anthropology, which explore environmental influences on behavior? All of this left me distrusting his mastery of the literature I know less well.

Ehrlich is tackling a very tough problem: He is striving to apply what we think we know about natural and cultural selection to "the human prospect." Can we use what we know to craft a more sustainable and equitable future? I agree with Ehrlich that the devil is in the details and that "There is no easy formula for understanding the human past or today's human natures or for projecting the human future." I also agree that "We'll never deal with the devils in the details unless we see the big picture." However, I was left feeling that he had muddied rather than clarified the situation.

We do have clues, potent ones at that, about how our evolutionary and ecological pasts may constrain us, what sorts of environments foster what variations in behavior, and clues about what strategies are likely to be more, or less, successful in different conditions. Regardless of the particulars of human social systems, we recognize basics that play out in similar ways. For example, it would be striking to find a society that persisted for a long time, and competed well against others, in which men did not strive for resources relevant to their mating success, parents did not strive for resources for their children's welfare, and women happily chose weak, poor men as mates in preference to strong, wealthy men. Beyond that, there is enormous—but nonrandom—variation in how we go about these things.

Ehrlich decries our propensity for consumption as an "evolutionary hangover." True—in our past, the conditions in which voluntary reduction of consumption proved reproductively profitable were almost certainly rare. So how shall we convince ourselves today? We could try exhortation and guilt (many environmentalists do), but there is a problem: We can invent cultures that absolutely contravene the direction of natural selection. For example, the Shakers choose celibacy. But how successful will such strategies be, competing with others? There are few Shakers alive today. Strategies that ask us to do things that have never been reproductively profitable, like extensive genetic altruism, are unlikely to become widespread.

Like Ehrlich, I care about crafting a sustainable future. How can we do this? Despite his cry for both the big picture and the devilish details, Ehrlich never gets past his detail-free claim that we could save the world, if we only would. He tells us that we could "just do" whatever we want, if only we had the will—leaving him precisely where philosophers of centuries ago stood. How has his biological knowledge informed his conclusion? For someone arguing that simple approaches to human behavior are useless, Ehrlich's arguments, in the end, are awfully simple.

I submit that one can test some obvious predictions about this set of problems, and that some problems and solutions will "match" better than others. If we can figure out how to work with our "predispositions," surely we have a better chance of getting more compliance. Why not apply both the detailed knowledge and the big-picture goals? If the devil really is in the details, let us get to work on them, and not end a long dissertation with the sentiment that we could do it, if only we would. That gets us no farther forward.