Review - Behavioral Genetics in the Postgenomic Era
by Robert Plomin, John C. Defries, Ian W. Craig and Peter McGuffin (Editors)
American Psychological Association, 2002
Review by Robert Loftis, Ph.D.
Mar 26th 2004 (Volume 8, Issue 13)

      The tone of Behavioral Genetics in the Post-Genomic Era is authoritative and triumphalist. The book weighs in at 636 pages, with 26 articles grouped together in nine sections. The editors are sure to list their academic credentials after their names, and have longish biographies in the back extolling their accomplishments. The spine of the book bears the imprimatur of the 95,000-member American Psychological Association (APA), and major funding for the research in the book came from the NIH and Britain's Wellcome Trust. The book has three forwards, each by a different luminary in human biology, including Nobel Laureate James Watson. The content of these forwards lets the reader know what the editors want to do with all of this scientific authority. Behavioral geneticist Irving Gottesman calls the book "an advance base camp in the assault on Mt. Ignorance" and its endorsement by the APA "the end of a long cold war" (xii). The forward by Watson lets us know who the enemy in this cold war was, and what the spoils of victory are. By 1975 there was "incontrovertible evidence for genetic involvement in personality and intelligence differences" but "those on the radical left continued to shout 'not in your genes'" (xxii).  The rhetorical message of the book is clear: All right thinking people know that genes influence behavior, and we are here to tell you exactly how they do so. Any opposition to us is motivated by politics, not science.

      The message is powerful in part because all right thinking people do know that genes influence behavior. Consensus has it that human psychological traits arise out of the interaction of genes and the environment. The authority here is being used to push more than the interactionism that everyone agrees on and that the evidence indicates. The editors' agenda has two levels. First, they allow only a very narrow conception of how genes can interact with the environment, one that winds up giving genes the dominant role. This conception is far from scientific orthodoxy, and likely to be false. Their agenda is also manifest in the sorts of behavioral traits they discuss. Three whole chapters of the book are devoted to the albatross of human behavioral genetics: Spearman's g, representing general cognitive ability. It is clearly a very important part of the agenda of this book that humans can be ranked by intelligence using a single number, and that this number represents a strongly heritable trait. Although the editors never make any remarks about race, their emphasis on general cognitive ability allies them with the racist pseudo-science of Cyril Burt, Arthur Jenson, and Richard Herrnstein and Charles Murray. This is the more narrow political agenda that has led to the cold war mentioned in Gottesman's forward. There is actually a lot of useful information in Behavioral Genetics. Libraries would be wise to stock it, researchers and clinicians wise to consult it. Readers should be aware, though, that this is not a definitive statement of the state of a maturing science. It is another salvo in a war that has been going on in psychology since the 19th century, and despite their rhetoric, the editors of Behavioral Genetics are not on the winning side of this war.

      The narrow agenda of Behavioral Genetics is established by the framework of the book: introductory and concluding chapters by the editors, and a set of five chapters under the heading "Research Strategies" following the introduction. The focus of these chapters is on a single program for understanding the genesis of complex traits, the search for quantitative trait loci (QTL). Broadly, QTL are the locations on the chromosome for any gene that makes a small contribution to a trait that can be measured quantitatively. The assumption in Behavioral Genetics is largely that QTL will be individually Mendelian, but always interact in a simple additive fashion to create a nice bell curve distribution of the trait in the population. This assumption is often true in plant and animal breeding, but it is incredibly unlikely to be true of complex traits like those of human psychology. Indeed one of the few breaks from the triumphalism of the book is the repeated admission that "progress to date in identifying QTLs has been slower than expected" (534). Many explanations are given of this fact. In the introduction and conclusion, the editors stress the need to break the "1% barrier": the reason we have been unable to find QTL is that we have only been able to identify genes with large effects. If we can find genes that account for 1% or less of the existing variance, we can begin to detect QTL. Chapter 4, part of the "Research Strategies" section, suggests that QTL have not been found because standards of proof have been too high: an excessive concern with false positives has led us to miss true findings (56). The possibility that QTL have not been found for human behavior because psychological traits are generally not the product of the additive interaction of genes is not taken seriously. The volume does contain a review of nonadditive forms of gene H gene interaction (called epistasis), which even admits that almost all complex traits will be the product of nonadditive interaction, but it is for some reason not included in the section on research methods. Instead it is in the cognitive disabilities section, between chapters on Attention Deficit Hyperactivity Disorder (ADHD) and autism. The complications introduced by epistasis for the search for QTL are not addressed in any of the chapters promoting such a search. There are other problems, in addition to the neglect of epistasis. The focus of the book is on the 40,000 or so genes that code for proteins and their regulatory sequences. The roles of genes that code for RNA only and of epigenetic inheritance are neglected. On the whole, this book is misnamed. At a minimum it should be called Quantitative Behavioral Genetics in the Postgenomic Era. It would even more accurately be called Quantitative Behavioral Genetics, Using a Very Narrow Understanding of "Quantitative", in the Postgenomic Era.

      Why the focus on a single method? The three chapters on general cognitive ability, including one by the lead editor, Robert Plomin, seem to give answer. Since the 19th century beginnings of the modern era of research on human genetics, there has been an effort to identify a single factor, labeled g, that explains all significant variation in human cognitive ability, can be measured by IQ tests, and which is strongly heritable. For much of its history, this research has had openly racist aims, as documented thoroughly in books like Steven Jay Gould's classic The Mismeasure of Man. The usefulness of these ideas for racists is obvious. Once you have all of the premises about g I named above, one need only add the often-true premise that a minority group tends to perform badly on IQ tests, to get a cogent argument for the genetic inferiority of the group in question. Fortunately the first premise behind this argument is simply false: there is no reason to think that human cognitive ability reduces to a single factor. Arguments for the other two premises—that this factor is measured by IQ tests and is strongly heritable—are also often dubious, but the point is moot anyway, because the factor in question simply doesn't exist. The chapters on general cognitive ability, however, assume all of these premises. Plomin's chapter, in particular, is festooned with citations to Arthur Jenson and other major figures in the tradition of racist IQ testing. Plomin basically takes as his central question, "what are the QTL that underlie g?" In doing this he assumes three- fourths of a racist argument, asserting all the discredited premises, and leaving out only the premise about performance on IQ tests, the one premise that mentions race and is in fact sometimes true. This creepy style of argument casts a pall over the whole book.

      General cognitive ability is not the only psychological trait covered in this book, although it is the most extensively covered. There are two chapters each on addiction, ADHD, and depression. Single chapters are devoted to speech and language, reading disorders, autism, schizophrenia, dementia, anxiety, and personality. Most of these are well-defined traits, like autism or schizophrenia, with inheritance patterns that clearly indicate a genetic component. These chapters contain reviews of the current research, outlines of the most successful techniques, and suggestions for the direction of future investigation. These chapters also seem to focus less intently on the desire to simply identify QTL. They form the bulk of the book and provide a good reason for libraries to stock it. A detailed evaluation of any of these chapters, though, would have to be written by a specialist in the relevant disorder, which is beyond my capacities.

      Given the technical detail of this book, the target audience is clearly researchers and clinicians. Patients and their families, and academics who are not full-time researchers in genetics, can find useful information here but may have to work to dig it out. The glossary is missing many useful terms (e.g. nonparametric linkage analysis) and includes many unneeded ones (e.g. "nucleotide" and "half-sibling"). The real problem with this book is that someone looking for information on the genetics of autism might walk away misled about research on genes and intelligence, and on the direction of research into gene H environment interaction.

© 2004 Robert Loftis

J. Robert Loftis, Ph.D., Visiting Assistant Professor, Department of Philosophy, St. Lawrence University, New York.