Several of my children, in the summer before they began kindergarten, participated in a program called “safety city.” Among the ditties that seem destined to stick in my head forever is one they learned there:

The traffic light you see ahead
is sometimes green and sometimes red.
The red on top means “stop, stop, stop.”
The green below means “go, go, go.”

Noticeably missing from this little piece of poetic instruction is the yellow light.  Perhaps one has to be a little older to learn how to deal with it. We’re supposed to stop when the light turns yellow, but sometimes we’re already quite near the intersection. So then we tend to treat it like a warning sign and just “proceed with caution.”

I’m pretty sure I never thought much about the yellow light until I learned to drive. In the early stages of my driving experience I heeded it religiously. I never entered an intersection on a yellow light. I think I even slowed just slightly (a bad idea actually) as I approached an intersection, just in case the yellow light should suddenly appear. My approach was to proceed with caution.

Those days are gone. Now I mostly just proceed. Not only do I routinely enter intersections when the traffic light is yellow, I am likely to accelerate a bit as I approach, the better to proceed before the light turns red. To be sure, I do not run red lights. But over time, as life’s routines have accustomed me to intersections, the yellow has become less and less distinguishable from the green, even though in theory I am supposed to stop for it.  And I suspect that my experience is not all that uncommon. Knowing that I don’t run red lights, I have a pretty good conscience about sailing through yellow ones.

I found myself thinking about yellow lights as I looked through the exhaustively detailed report, “Human Genome Editing: Science, Ethics, and Governance,” recently produced by an advisory group formed by the National Academy of Sciences and the National Academy of Medicine. In most ways the salient moral issues do not seem to me to have changed much from the last time I thought at length about these issues roughly a decade ago.

One thing, however, has changed considerably—namely, the development of what is called CRISPR/Cas9, a new method for “editing” the human genome.  Attempts at gene therapy, although not terribly successful, have been around for some time. What CRISPR/Cas9 appears to offer, however, is an efficient and precise method for altering (both by addition and deletion) an organism’s genetic material. We stand on the brink of an age in which our capacity to modify the human genome may increase enormously. And not surprisingly scientists are eager to proceed with gene-editing research.

What sorts of research are under consideration or, at least, possible? Most obviously, basic laboratory research that does not, at least for now, attempt clinical applications. In addition to increasing our understanding of human biology generally, such research could increase our understanding of important areas of medicine, positioning us for future attempts at clinical applications. Researchers could increase their understanding of fertility and reproduction, perhaps opening doors for greater success with in-vitro fertilization (IVF) and related techniques. They might create disease models for study, with the hope of one day developing new approaches to treatment. 

Still more dramatically—and the issue that received the most notice in news stories about the Academies’ report—researchers might develop further the capacity to do what the report calls “heritable germline editing.” Most cells in the body are known as somatic cells (from the Greek word soma, meaning “body”). These cells are differentiated to form the body’s organs, tissues, skin, and bones. Somatic cells may undergo change (through mutation, for instance), but those changes die when the individual dies. They are not passed on to future generations. By contrast, germ cells, which give rise to sperm and eggs, are not differentiated into the various parts of the body, and changes in the germline could be inherited by future generations. Hence, to “edit” the germline is to exercise power not only over an individual patient but also over descendants of that individual into an indefinite future.

In principle, somatic-cell research aimed at developing possible treatments for disease is not controversial. It is simply an extension of medicine’s attempt to treat patients—to cure or ameliorate disease and debilitating injury. Genetic therapies might one day enable us to treat inherited genetic diseases such as cystic fibrosis or Tay-Sachs. Research in what is called regenerative medicine may increase our capacity to help those suffering from Parkinson’s or Alzheimer’s, or to produce sheets of skin for burn victims, or to replace heart muscle. To be sure, many things can go awry—and sometimes have—when we proceed from somatic-cell research to clinical attempts at genetic therapy. But at least for any work supported by federal government funding, we have in place a number of regulatory mechanisms aimed at ensuring, as much as possible, that attempts at genetic therapy will take place only after their safety has been rigorously studied and we have reason to believe the hoped-for benefits of the therapy are greater than its risks. In general, the report of the Academies regards this current regulatory structure as adequate for somatic-cell research, and there seems little reason to disagree.

The possibility of increasingly powerful and precise editing of germ cells raises more complicated moral questions of several different sorts. Why, apart from a general desire to expand the frontiers of our knowledge, might anyone want to do it? One reason is that heritable genome editing could, as the report puts it, “be the only or most acceptable option for prospective parents who wish to have a genetically related child while minimizing the risk of transmitting a serious disease or disability.” There are, of course, other ways to try to reach the same end, though I hardly wish to recommend most of them. We could use prenatal genetic diagnosis (PGD) of a fetus in utero and selectively abort any fetuses with genetic disease. Or we could use PGD of embryos in the laboratory, selecting only those embryos free of genetic disease for implantation through IVF. Still other alternatives exist. Prospective parents might use donated gametes; they might adopt children; or they might decide, given the risk that they may transmit genetic disease, not to have children. “These options, however,” the report notes, “do not allow both parents to have a genetic connection to their children, which is of great importance to many people.”

Before turning to other important moral issues raised by the possibility of germ-cell editing, it is—at least for Christians—worth thinking briefly about this seeming importance of genetic connection. The British theologian and ethicist Michael Banner has noted that, despite the seeming gulf between their views, advocates of IVF and Roman Catholic critics of those techniques almost try to “outbid” each other in expressing “profound respect for biological parenthood.” To be sure, the former do it in order to help infertile couples produce “a child of their own,” the latter in order to reject techniques of assisted reproduction. But Banner invites us to consider whether this may not be “a relatively minor disagreement between parties which share very similar understandings of the relationship between parent and child.” If he perhaps slightly overstates the case, it is nonetheless true that both views leave the rather fervent desire for a genetic connection between parents and children “solemnly in place on its contemporary pedestal.” Unless we break the power of that desire, I suspect that we will be unable to resist what will almost surely seem to be (as the report notes) a promising way to have genetically related children without passing on to those children a serious disease.

More generally, what should we think about germline editing, which “in the not-so distant future” may, the report concludes, “become a realistic possibility”?  Were it not so important, I would be tempted to characterize the central moral issue here as “intriguing” or “fascinating.” On the one hand, editing germ cells to eliminate serious disease has the potential to prevent suffering not only for a single individual but also for that person’s descendants. It eliminates the need for continued interventions at the somatic level for each new generation. On the other hand, that potential good is exactly the problem: germline editing may eliminate disease not only for a single individual but also for that person’s descendants. That is to say, whereas somatic-cell genetic therapy is simply an extension of what medicine has always sought to do, we might wonder whether germ-cell editing should even be characterized as medicine.  The “patient” is no longer a particular suffering human being; instead, the object of such proposed interventions is what Paul Ramsey once called “that celebrated nonpatient, the human species.” The possibility of unforeseen and unintended consequences is considerable, and it is not silly to wonder whether human beings—ourselves included—should ever seek to exercise this kind of control over those who will come after us. Recognizing such concerns, the report says: “Heritable germline editing trials must be approached with caution, but caution does not mean that they must be prohibited.” Which is to say, it sets before us a yellow light.

Inevitably, germline research raises another issue that has dogged our public discourse since the topic of stem-cell research became important several decades ago: namely, the use and destruction of human embryos. In some ways the report simply does not engage the moral issue, saying: “This report does not address those ethical arguments, and accepts as given the current legal and regulatory policies that apply in each country.”

Nevertheless, there are peculiarities in the report worth noting. Describing the various approaches taken in different countries, it says: “Views on the legal and moral status of the human embryo range from treating it the same as any other human tissue, to considering it a tissue deserving of some extra degree of respect, to viewing it as tissue that should be accorded the same respect or even the same legal rights as a live-born child.” What a peculiar formulation that is! I would think that those who believe a human embryo should have respect equal to that of a live-born child are likely to describe that embryo not as “tissue” but as one of us—a living human being in the earliest stages of its development, having the characteristics we all did at that stage of life.

There are also strange juxtapositions scattered throughout the report. At a number of places it notes that any research that destroys embryos or creates them solely for use in research may not be funded by the federal government’s Department of Health and Human Services. This is due to a provision known as the Dickey-Wicker Amendment that has been attached to the Health and Human Services appropriations bill every year since 1996. (The research can, of course, be funded by other sources, except in some states where it is illegal.) Yet, without ever recommending a change in this funding policy, the report just as often expresses a relentless pressure to proceed, to move forward with research that would involve the destruction of embryos. So, for example, shortly after noting the federal refusal to fund that has been in place for two decades, it states: “Important scientific and clinical issues relevant to human fertility and reproduction require continued laboratory research on human gametes and their progenitors, human embryos and pluripotent stem cells. This research is necessary for medical and scientific purposes that are not directed at heritable genome editing, though it will also provide valuable information and techniques that could be applied if heritable genome editing were to be attempted in the future.” Elsewhere it describes recommendations of a 1994 NIH Human Embryo Research Panel as “recognized within the scientific community as a more general evaluation of the ethics and acceptability of such research.” Those recommendations supported federal funding for research on embryos and—when deemed “necessary” to attain important scientific knowledge—the creation of embryos for use in research. It is worth recalling, however, that Congress rejected the funding recommendation of that panel, and President Clinton specifically withdrew from consideration for funding any research that created embryos simply for use in research.

At any rate, the effect of these juxtapositions is to exert a constant pressure to proceed with research that U.S. government authorities are unable even to consider supporting at the present time. Acknowledging the restrictions in place while asserting the necessity of the research has the effect of making the restrictions seem irrational. Insofar as that is the overall effect, it is not quite accurate to think of the report as prescinding from the moral and legal arguments. Via the claim of medical necessity, it presses for expanded, though always regulated, research that uses embryos. And it is worth reminding ourselves that we are not talking about research that might use just a few embryos, some very limited exception. What is deemed necessary is the use of countless embryos—always, of course, with caution and with relief of suffering as the ultimate end in view. There remains one further possibility on the gene-editing horizon. Beyond basic research or somatic-cell editing, beyond even heritable genome editing, there is the possibility of “enhancing” our traits and capacities. This is, as the report rightly notes, a notoriously difficult concept to define. We all understand in general that the idea has something to do with improving a trait beyond its typical level, moving beyond what is simply therapeutic, but it can be hard to be more precise than that. Moreover, it is quite possible that a product may have dual uses, some clearly therapeutic, others more like enhancement. So a treatment designed to maintain muscle mass in the elderly may also help to improve the capacities of wrestlers. Given these complexities, and given that a technique such as CRISPR/Cas9 is still in early stages of development, the report draws back somewhat, encouraging a focus “at this time” on therapy rather than enhancement: “Regulatory agencies should not at this time authorize clinical trials of somatic or germline genome editing for purposes other than treatment or prevention of disease or disability.” Reasonable as this no doubt seems and probably would be if one believed such research was “necessary,” one might wish the report had probed some of the deeper philosophical questions a bit more. Are there research aims that, however well intentioned, would seek to bestow traits of character and skills that have no value apart from the process by which they are developed and achieved?  Might the drive to enhance human capacities cause us to lose a sense of the “giftedness” of human life, making us, in Michael Sandel’s nice phrase, “inhospitable to the unbidden”? Perhaps these are not the sorts of questions we should ask a committee of the National Academies to address. Fair enough. But then we also need to acknowledge that there are limits to how instructive such a report can be.

Finally, I return to where I began: the meaning of a yellow light. On the whole, this report on human genome editing seems to me to set before us a yellow light, which should invite us to think about how we actually treat yellow lights. Surely they are useful. We recognize that it would be foolhardy for us just to forge ahead in research without taking due care to avoid possible unforeseen harms; hence, no green light.  But at the same time, many—and certainly the authors of this report—think it necessary to find a way to proceed with research that promises to relieve some ills of the human condition. We focus on avoiding harms and maximizing benefits—that is to say, not on the rightness or wrongness of what we are doing, but on the results of what we are doing. Therefore, no red light.

That leaves the yellow, and we proceed with caution. We embed gene-editing research in regulatory controls. We suggest that research to which some or many object should not be undertaken “at this time.” We focus on maximizing benefits and minimizing harms. All perfectly reasonable. But we do not ask ourselves whether there may be some research—even possibly beneficial research—that should not be done no matter what its benefits may be.

That is the nature of a yellow light. In theory it tells us to stop now; in practice it tells us to keep going—carefully. And whenever I see one, I always find myself thinking about what it means that we are proceeding with caution. On the one hand, we might congratulate ourselves. We are, after all, proceeding with caution, concerned lest we be misled by hubris. But on the other hand, it is still true that we are proceeding.  We keep on going. We don’t imagine that we should ever just stop. And so we keep on proceeding—doing so, moreover, with a good conscience, because we are cautious, thoughtful, and morally concerned. It’s just that, faced with the possibility of desirable consequences, we can never find a reason to stop.

No doubt it is generally wise to let a yellow light make us cautious. But there may also be moments when we should remember that there always remains another possibility and that moral seriousness might sometimes be measured by our willingness to be as wise as kindergarteners and to know when to “stop, stop, stop.”