A few weeks ago, I attended a panel on gene editing at Harvard Medical School that covered some aspects of the science, ethics, and law of the practice. It was an interesting talk, in part because it largely covered the ethical issues of gene editing for human medicine and in other species as two sides of the same coin, rather than as fundamentally different conversations, as they are often treated.
Indeed, one member of the audience asked why there is so much focus on the safety and ethics of human gene editing, when the stakes, he argued, are much higher in the use of gene editing for environmental engineering. A botched human germline edit could harm a family; a botched gene drive could kill us all. It’s an interesting point. And because it suggests that we may want to be less than sanguine on the use of gene drives to eradicate malaria, on which I have previously been extremely sanguine, it is a point worth responding to.
While, of course, the analysis of risk for any particular technology will have to be assessed on its own terms, there are two good reasons to, in general, be more concerned with the safety of human germline interventions than with gene editing in agriculture and environmental engineering. My ethical assumption is that human life is what matters, and harm to the environment is only a problem to the extent that it harms human life. This is a contestable assumption. Some believe it is not true, and that nature has a more value in itself. While I do not have the space to defend it here, my assumption is probably more widespread and, at least, would seem to be shared by the audience-member at the panel, in his concern for the risks of genetic environmental engineering on human life.
The first reason to be skeptical of the claim that the use of genetic technologies in the environment is inherently more dangerous than it is in humans is that the background assumption—that interventions in the environment carry more systemic risk—is almost certainly false. That assumption, the foundation, too, of the “precautionary principle” of the Rio Declaration—that covenant of misanthropists and metaphysicians—simply does not withstand scrutiny.
The argument goes like this: (1) ecosystems are extremely fragile and incomprehensibly complex; and (2) they are globally interconnected; (3) the maintenance of human life requires certain specific ecosystems. Therefore, (4) because of (1), human intervention in ecosystems is extremely likely to be devastating, and because of (2) it will have worldwide effects; and, finally (5), because of (3) we will all die.
The problem is that to prove this, proponents rely on wholesale speculation. Indeed, the evidence belies the logic at every turn. One will search the history books in vain for a definitive instance where the precise kinds of interventions proposed by gene drives—the extinction of a species, the introduction of a new gene into an ecosystem—definitively harmed anyone, much less killed us all. Dozens of species go extinct every day and the human population and standard of living continues to rise. What are the chances that the handful of Anopheles species that carry malaria are going to be the species on which everything else depends? Mutations happen in nature every day and none of them are systemically threatening. Hell, life has been around on Earth for 3.5 billion years, and, obviously, not one of the literally quadrillions of genetic changes that has arisen in that time has been systemically threatening. In the 300,000-odd years of human dominion we have wrought much more damage than a precision elimination of Anopheles and this planet hasn’t killed us yet.
Sure, it is conceivable that the use of a gene drive will accidentally threaten life on Earth. But then that same standard of Murphy’s-Law-on-amphetamines must be applied to human gene editing as well, because we have apparently handed risk analysis over to oracular pundits of doom rather than people who understand probability. And, viewed in that light, the worst case of human germline editing is not that a family will be plagued with abnormalities down the line. It’s that we edit an embryo which immediately detonates with the force of a thousand suns. Unlikely? Obviously. Conceivable? Barely. But the same can be said for apocalyptic environmentalist speculation about gene drives. There is no difference in principle.
Second, if history is any guide, we do have actual evidence of the dangers of human genetic intervention and we do not with respect to interventions in other species. History, like law, proceeds by way of analogy. And traditionally transgenic genetic modification, while not gene editing, is the closest thing we have to compare. Jesse Gelsinger actually died of that. Genetically modified crops, in contrast, are now grown all over the world. Notwithstanding the perennial hysteria of precautionary principle types the introduction of these new genes into ecosystems has not had problematic systemic implications. Genetically modified food has harmed no one and helped millions. If anything, in retrospect, we could have afforded to move much more quickly on adopting transgenic agriculture. So we know, then, by analogy that there are real human risks to unprepared human genetic modification. But the risks in other species—even when the technology is used out of the lab and in the ecosystem—would appear to be wholly speculative.
Again, each technology is different. It is impossible to opine accurately on the risks associated with classes of technology in the abstract, and when seriously considering which to employ we will need to make an individualized assessment. It is possible that particular gene drives will kill us all. But for the foregoing reasons, academia’s emphasis on the risks of human, rather than environmental, genetic engineering mean their heads are in the right place.
James Toomey is a 2018-2019 Petrie-Flom Center Student Fellow.