The Mirage of a Space Between Nature and Nurture
by Evelyn Fox Keller
Reviewed by Daniel W. McShea
Is there anything new to say about how we should understand the nature-nurture problem? The answer is yes, and it is not because there are conceptual matters still unresolved. It is because no one has offered a way to think about the problem that is simple and grabs the imagination. Absent a clarifying story, teachers continue to struggle to explain it to students. And some of us continue to write books and papers in which we say or imply things we do not literally mean about nature and nurture, genes and environment, heritability and plasticity -- things we later regret having phrased the way we did. So wouldn't it be nice if there were a small book that explained, clearly and simply, how to understand the problem, pitfalls and all; if there were a concise manual -- something like Strunk and White's famous style guide -- that we could just hand to our students; if there were a little manifesto that we could curl up with and reread every couple of years to restore to our thinking the clarity we know this difficult subject deserves? The Mirage of a Space Between Nature and Nurture, by Evelyn Fox Keller, may be just the book we've been waiting for.
Here is the issue she is addressing. We sometimes think of nature and nurture as distinct and separable causes, each with its own quantifiable contribution to a given trait. But at the same time, we know that every trait requires both, that genes alone produce nothing (as does environment alone, for that matter), and that therefore in a developing individual, both are 100 percent responsible for every single trait. In the metaphors Keller cites, drum and drummer are each 100 percent responsible for the beat we hear, bricks and mortar are each 100 percent responsible for the wall that is being built, and the person at the end of a hose aiming a stream of water into a bucket and the person turning the valve at the spigot both get 100 percent of the credit for filling the bucket. In such cases, it is absurd to think about quantitative partitioning of causes. So why do we slip? Why do we even occasionally hear ourselves suggesting, with a sober countenance, that Suzy's shyness might be partly genetic, when we know already for certain that it is both 100 percent genetic and 100 percent environmental?
Keller argues that much of the trouble has to do with linguistic practice, with slippages in usage and concepts. In her apt words, the nature-nurture debate is a "morass of linguistic and conceptual vegetation grown together in ways that seem to defy untangling." Undaunted, she takes on the task of untangling, focusing on two key points, two areas where words have historically set us up to muddy things that we know should be perfectly clear. First, it should be clear that difference makers (the factors responsible for observed differences in particular phenotypic traits) are not the same as trait makers (the developmental processes by which the trait came to be) -- that difference makers may be only trivially involved in the production of a trait. The address on an envelope makes a huge difference in where the letter goes but has little to do with generating the process that actually gets it there. And reciprocally, variation in a hugely important trait maker may produce only trivial differences in the result. The U.S. Postal Service gets my letter there about as reliably as the United Parcel Service, despite the many differences between the two organizations.
So the difference between difference making and trait making is clear. And in biology, we know that mutation studies in developmental genetics tell us only about difference making. So why, when we find a mutant gene that affects a trait (a difference maker), do we jump so easily to the claim that this gene is critical in the production of the trait (a trait maker)? Keller traces such slippages back to the first mutation studies in the early 20th century, and remarks:
It is hard to imagine that the early slippage was entirely accidental. To think of genes simply as difference makers would have been to detract from the power of the gene concept. Mapping difference makers and tracking their assortment through reproduction may have been all that the techniques of classical genetics could do, but the aims of these scientists were larger. What made genes interesting in the first place was their presumed power to mold and form -- in a word, their presumed power to act.... [And] the easy slide between genes as difference makers and genes as trait makers perpetuated the illusion (as widespread among geneticists as it was among their readers) that an increased understanding of the effects of gene differences would enhance our understanding of what it is that the entities called genes actually do.
Times have changed -- somewhat. As Keller points out, the modern focus is on expression patterns and on the role of noncoding DNA, not just on the genes themselves. And given what we know now about the cyclically interactive processes of development, today's biologists are much less likely to attribute causal agency to genes (or even to DNA). But the problem has not disappeared. A mutant allele associated with a speech and language disorder still gets labeled a speech and language gene. We continue to slip.
The second slippage has to do with alternative meanings of heritability
. Used colloquially, heritability
has to do with the transmission of genes from parent to child. "Where did you get those brown eyes?" I'm asked. "I inherited them from my mother," I reply, intending to convey that my mother passed along to me certain genetic brown-eye-color difference makers. But studies of nature-nurture in humans use heritability
in a very different and technical sense, what is called broad-sense heritability
-- a measure of the proportion of phenotypic variation in a population that is due to the total genetic variation (including variation arising from gene-environment interactions). What this measure tells us, essentially, is the degree of resemblance between parents and offspring in a population, regardless of how that resemblance is transmitted from parent to offspring.
Now suppose that it is discovered that some trait -- shyness, say -- has high technical-sense heritability. When we hear this, we are likely to think that it tells us something about heritability in the colloquial sense, something about how Suzy came to be shy. Indeed, we might leap to the conclusion that she is shy because she inherited certain difference-making alleles from her parents, alleles that directly affect her individual psychology, making her more likely to be shy. Of course, that does not follow at all. Suzy could be shy for some other reason -- because she has a speech impediment that makes her self-conscious about asserting herself socially, for instance. Indeed, this could be true of all shy individuals. And in that case, technical-sense heritability would still be high, even though inheritance involves a critical cultural factor. (And in a different culture -- one where speech impediments were more socially acceptable, for example -- the heritability of shyness might be much lower.)
So we know that this move from a measure of population heritability to a conclusion about the genetic contribution to a trait difference in an individual lineage is unwarranted, but we do it anyway. Here is Keller again:
Most behavioral geneticists would agree that a mistake has been made when explicit claims about the genetic basis of individual traits are inferred from measures of technical heritability, or at least I hope they would. More commonly, however, such claims are formulated simply in terms of heritable traits, with the implication of a genetic basis left just short of explicit.... But if one were to confront the authors of such claims (as I have) with this obvious source of confusion, a common response is: "Yes, of course. But that's not the way I use the term; I use it merely to refer to measures of heritability in the population."
Keller's story of how we came to this, of the roots of these slippages in the history of biology, especially in the writings of Charles Darwin
and of Francis Galton
, is a fascinating one. I will not retell it here. Instead let us turn to the question with which she titles her last chapter, "What's to be done?" We could try to change the words we use -- for example, by ditching the word heritability
for the population-level measure. But linguistic practice is highly resistant. Perhaps instead, Keller suggests, we can change the questions we ask. We are interested in nature-versus-nurture in part because we want to know which traits can be changed and which cannot, and for those that can be changed, we want to know how to change them. So instead of asking about genes versus environment, or heritable versus not, why not ask how malleable a trait is (at a given developmental age) and what factors actually affect it? Instead of asking, What are the genetic causes for trait X?, we could ask, What are the pathways that produce X? In answering this question, we would be interested in all of the resources involved in producing X, including DNA and its molecular products, of course, but also the many resources not manufactured by DNA. In making this suggestion, Keller is in good company. Essentially the same change of questions has been suggested by scholars in a school of thought known as developmental systems theory (William C. Wimsatt
and Susan Oyama
, for instance) and by some in developmental biology (Fred Nijhout, for example). These ideas have been in the air. Their time is coming.
In the meantime, while we await the coming change in the scientific culture, Keller's little essay is an excellent teaching resource -- and an excellent resource for reminding ourselves about the pitfalls of the current way of thinking. Anyone with an interest in the nature-nurture problem -- which is to say, almost everyone -- should read this book. Daniel W. McShea is an associate professor of biology at Duke University. He is coauthor with Robert N. Brandon of Biology's First Law: The Tendency for Diversity and Complexity to Increase in Evolutionary Systems (University of Chicago Press, 2010) and is coauthor with Alex Rosenberg of Philosophy of Biology: A Contemporary Introduction (Routledge, 2008).