WITHOUT A doubt, ELSI has worked hard to focus attention on ethical dilemmas arising from genetic research-the practical threats to our privacy, our sanity, our health insurance. And in doing so, it has often struggled with scientists (eager to do their work in peace) and the biotechnology industry (eager to get new products on the market as soon as possible). But when it comes to challenging the notion that human beings can be reduced to their genes, the ELSI panelists needn't take such an adversarial role: They can rely on the scientific research itself. In a sense, ELSI's greatest achievement may be the way it's helped familiarize nonspecialists with the intricacies of molecular genetics. For example, when the Working Group finally published its condemnation of The Bell Curve, it used arguments taken from the vanguard of genetics research: "The more scientists learn about human genes, the more complexity is revealed," the statement read. "This complexity has become apparent as more genes correlated with human genetic diseases are discovered.... Simplistic claims about the inheritance of such a complex trait as cognitive ability are unjustifiable; moreover, as the history of eugenics shows, they are dangerous."

Few geneticists would go as far as Harvard biologist Richard Lewontin, who contends that the complexity of the human organism renders specific genetic effects largely irrelevant to the etiology of disease. But recent genetic finds are raising researchers' awareness of the complexity of genetic interactions. At the dawn of the genome project, molecular geneticists often focused on simple genetic diseases. For example, in one of the first important accomplishments of positional cloning (the process that is now widely used to link genes to traits), researchers in 1983 found a location on chromosome 4 linked to Huntington's disease. This kind of "one gene, one disease" formula enticed scientists with the prospect of making relatively easy progress in the understanding and treatment of inherited diseases.

But the Huntington's model has turned out to be the exception rather than the rule, as evidenced by a recent letter that Stanford geneticists David Botstein and Neil Risch published in Nature Genetics. Calling for scientists to use more rigorous statistical efforts in interpreting linkage studies, it noted that over the last twelve years, sixteen different research groups have announced the discovery of genetic linkages for manic depression in fifteen different locations on eleven different chromosomes. Each research group was at first convinced it had found the key linkage for manic depression; none of the results could be reproduced. In the years since the genome project began, triumphal announcements about the genetic foundation of complex traits and diseases have time and again been confronted with studies that failed to reproduce the result-Dean Hamer's controversial 1993 discovery of a homosexuality gene is an obvious example. Given the growing evidence of complexity, genetics may have advanced to the edge of a new paradigm, Kenneth M. Weiss, a genetic anthropologist at Penn State University, suggested in a recent issue of Molecular and Evolutionary Phylogenetics. Some diseases are indeed easy to account for-sickle cell, the classic example, is caused by a single DNA misprint that leads to the creation of sickle-shaped hemoglobin that clogs the bloodstream. But even in the case of Huntington's disease, the variation in the mutation is large-and recently, researchers found several individuals with a Huntington's-type mutation on chromosome 4 who didn't suffer from the disease. In Weiss's view, many human traits may turn out to involve such complex and variable combinations of genes as to be almost unknowable. "These systems can't be predicted from reductionist principles," he said in an interview.

From the ELSI perspective, this is very useful news. But just because the science provides a potential critique of genetic reductionism, it doesn't mean that's the message most people will hear. "From the molecular genetics point of view, it's absolutely clear that we're demolishing the arguments of genetic determinism," says Cox. "But the facts have never got in the way of people who wanted to use genetics in a deterministic way in our society. And I think this is where my colleagues miss the boat. Just because something is scientifically right and proven, it isn't the end of the story. You have to consider how that fact is going to be parlayed into people's lives."

And that's where ELSI comes in, Cox says-if nothing else, by giving people like him a chance to mouth off to the press. Cox believes that all the grant money and public interest in genetics are leading to hasty publication of tantalizing but insubstantial results. As an example, he cites the article published by Science in late November announcing the discovery of a gene related to the brain's ability to use serotonin, which helps alleviate anxiety. Natalie Angier's punchy writeup of the announcement played on the front page of The New York Times under the memorable headline, grumpy, fearful neurotics appear to be short on a gene. While the researchers-including Dean Hamer-insisted in interviews that the impact of the "neuroticism gene" (which, incidentally, was carried by two thirds of the study subjects) was quite small, in Cox's view that was just window dressing for their rush to publish an essentially meaningless study. "The dynamic goes like this: The public is interested in genetics because it offers a way of understanding the world and giving them a modicum of control. So the geneticists, rather than realizing that what they're doing is perpetuating a social myth, start thinking there's inherent value in what they're doing. The scientists believe they're the dog and not the tail being wagged."

ELSI's crusade against genetic reductionism hasn't escaped the notice of Collins, who in 1995 wrote in the Proceedings of the National Academy of Science that publications about complex traits, however tentatively worded they may be, can easily be misinterpreted to deemphasize the role of social and environmental factors. In an assessment of the Human Genome Project to date, Collins wrote that the project "will need to foster a better understanding of the meaning of human genetic variation among members of the public and the professions, and expand its efforts to propose policy initiatives designed to prevent genetic stigmatization, discrimination, and other misinterpretation or misuse of genetic information." So even though it may sometimes speak through the voice of Francis Collins, ELSI is being heard.

Of course, there's a major downside to the new genetics paradigm ELSI promotes. For while the new complexity may run sociobiologists out of town, it also means the project offers diminishing hope for the sick-less predictability, after all, means less understanding and fewer immediate results in the form of medical treatment. "Expectations that new therapies will quickly grow out of genetic research have been a bit overblown," says Robert Weinberg, a molecular biologist at MIT's Whitehead Institute. It had been hoped that once a disease gene was isolated, scientists could figure out the proteins it coded for, and from there sleuth out biological misfires. But it turns out that, because of the complex interplay of effects, knowing the gene's chemistry won't provide a systematic shortcut to disease cures. In the short term, the main products stemming from the genome project are the diagnostic tests arising from the isolation of genes. And since these tests are of limited value, part of ELSI's job is to instill a little humility in the scientists who create them.