TEMPTATION CAME, so the story goes, in the form of an invitation to eat forbidden fruit. The promised reward of that first disobedience was godlike status. Milton’s eloquent serpent offered a straightforward advertisement:
… your eyes, that seem so clear
Yet are but dim, shall perfectly be then
Opened and cleared, and ye shall be as gods,
Knowing both good and evil as they know.
Knowledge of good and evil? Later writers have worried about scientific research dangling divine powers before mortals morally ill-equipped to use them. Since at least Mary Shelley, a parade of novels, plays, and films have showcased the ambitious, often deranged, investigators who create monsters and sometimes threaten the survival of our species, and for whom ethical insight seems conspicuously absent. Their technical virtuosity dwarfs their moral sense. Fortunately, the disastrous consequences of their irresponsible tinkering are usually averted by the wisdom and courage of the genre’s heroes.
The great achievements of the biological sciences during the past half-century are manifested in techniques — godlike powers? — for modifying, even designing, organisms. Thirty years ago, then-current methods of DNA sequencing inspired prominent molecular biologists to campaign for the Human Genome Project. Advertised as having the potential to “tell us who we are” (perhaps an unintended echo of Milton’s serpent’s promise to Eve) and of leading directly to cures or treatments for hereditary diseases, it has provided — as commentators consistently predicted — a significant number of diagnostic tools. The project has also brought technical advances in DNA analysis. Sequencing today is cheap enough to allow many people in the affluent world to satisfy their curiosity about parts of their genomes. Yet the principal benefits lie in enhanced biological understanding — of the complex processes involved in “reading” the genetic sequences, of early development in some important types of organisms, and of fine-grained evolutionary relationships.
The Human Genome Project was accompanied by an officially funded effort to explore the ethical, legal, and social consequences of its anticipated results. Whether, as some cynics believed, that effort was simply window dressing intended to deter public protests that might impede research, it has surely helped allay the fears of outsiders. Worries about modern-day counterparts of Baron Frankenstein, creating uncanny and dangerous monsters inside their laboratories, were addressed in articles and books accessible to lay readers. A central part of the message consisted in explaining the limits of the available techniques. Those who imagined precisely designed genetic changes, with molecular scissors snipping out an unwanted sequence here and molecular tweezers inserting a favored sequence elsewhere, learned a sober truth. Working with a large population of cells, researchers injected them with some extra DNA, hoping that a few of them would incorporate it on some appropriate chromosome and that they could then detect when this had happened. Useful enough for a number of biological and medical purposes, but hardly a way to create a super race.
During the past few years, however, all that has changed. Molecular biologists have developed a far more sophisticated technique for gene editing. The new tool — CRISPR-Cas9 — combines two DNA sequences of types found in bacterial genomes. One part consists of Clustered Regularly Interspaced Short Palindromic Repeats: bits of DNA in which brief sequences that read the same backward and forward (palindromes) are regularly interrupted by other sequences. CRISPR sequences target the genetic material of potential viral invaders, thus serving as part of the bacterium’s immune system. They are aided in their work by proteins generated by nearby pieces of DNA, the cas genes. Effectively, the CRISPR sequences detect the intruders, and the cas proteins then destroy them.
The University of California at Berkeley biochemist Jennifer Doudna has been centrally involved in the international effort to harness the bacterial self-defense system for scientific ends. A Crack in Creation, co-authored with her former graduate student, Samuel Sternberg, provides a generous and inclusive account of the series of breakthroughs that culminated in a dramatically improved method for modifying the genetic material of cells in a wide range of species — including our own. Part I of the book explains the biological details and chronicles the discoveries, a testimony to Doudna’s pedagogical talents, honed in years of teaching undergraduates.  Yet, despite the clarity of her prose, readers may find the accumulation of detail overwhelming. No matter. The importance of the book lies in its lucid exploration of ethical questions, beginning in the prologue and taken up extensively in Part II. Even if talk of inactivation experiments and single-guide RNAs makes your head spin, skimming Part I will allow you to follow the ethical discussions.
In her prologue, Doudna explains how she was moved to tackle issues beyond her sphere of professional expertise, how concerns about what might be done with the new technique led her to discuss the implications with government officials and to help organize a “summit” on human gene editing. As she sees it, the old reassurances have ceased to apply. No longer is it possible to downplay worries about a renaissance of Frankensteins by pointing to the bluntness of the available instruments. Even though CRISPR-Cas9 needs further refinement before it is completely reliable — comparably reliable to other procedures used in contemporary medicine — the era in which it might be used on human bodies is now in prospect.
Some species — chimpanzees and dogs, for example — have characteristics, capacities for thinking and feeling, that ought to give researchers pause when they consider many forms of genetic interventions. And, as A Crack in Creation appreciates, proposals to edit the genomes of organisms — perhaps by producing mutant mosquitoes to eradicate malaria or Zika — need ecologically informed ethical scrutiny. Yet the nightmare scenarios center on deliberate manipulations of human beings. Doudna and Sternberg discuss a wide range of possibilities, among which it is useful to distinguish three major types. There are somatic interventions intended to cure or treat a person whose genome contains DNA sequences that cause, in normal human environments, a condition of disease or disability; here, the human body is modified in ways that do not affect sperm or eggs. Germline interventions, by contrast, attempt to prevent a hereditary disease from being passed on to future generations; in this case, the eggs or sperm must be altered. Finally, enhancements (whether germline or somatic) attempt to provide new DNA sequences intended to give rise to characteristics beyond those normally found in members of our species; these are the applications imagined in stories about super-races, endowed with prodigious strength or extraordinary intelligence.
In a thoughtful and sensitive discussion, Doudna and Sternberg argue in favor of somatic interventions, although they rightly caution that some heritable traits may resist this approach. If the DNA sequence responsible for the troublesome condition acts early enough in development, the attempt to use gene editing may come too late — damage will already have been done. Further, some diseases — schizophrenia is a well-known example — are dependent on sequences at many places in the genome. Nevertheless, there are well-understood diseases — cystic fibrosis, sickle-cell anemia, and Huntington’s — for which modification of cells of the appropriate type would enable people to live free of the disease. A child born with the long repeat characteristic of Huntington’s would no longer have to fear an inevitable onset of neural degeneration and acute physical decay, while still in the prime of life.
As Doudna and Sternberg recognize, some somatic interventions would be acts of great mercy. They would be rejected only by those who saw tinkering with human genetic material as proscribed on some principled grounds, most likely by appeal to a favored religious text. Yet what could be the basis for drawing a line between transformative, life-saving modifications of the DNA in some of the body’s cells, and other medical procedures through which terrible diseases are cured? What would distinguish the editing of the DNA in the lung cells of the cystic fibrosis sufferer from the bone marrow transplants that save the lives of some cancer patients?
With respect to the two other categories, Doudna and Sternberg are more cautious. After reviewing the case for somatic modification, they seem to draw a line: they propose to refrain “from using CRISPR technology to permanently alter the genomes of future generations of human beings, at least until we’ve given much more thought to the issues that editing germ cells will raise.” A later passage softens their stance. They consider some of the most popular reasons advanced to prohibit germline modifications, and shrewdly diagnose their flaws — concluding that because of “the pain that genetic disease causes families” an outright ban on germline editing would be premature.
Even that conclusion is too timid. Opposition to germline editing rests largely on fictions about the splendid design of our bodies (ignoring the many flaws, resulting from compromises made in our evolutionary past), the evils of eugenics, and the possibility that modifications may produce some dreadful unforeseen consequences. Although terrible things have been done in the name of eugenics, it is important to recognize that some form of eugenics is now inevitable.  Having acquired knowledge of the molecular structures underlying some of our traits, we are faced with decisions about the kinds of people there will be. To suppose that it is better not to use our knowledge is to express a eugenic choice, to declare that the world is better if we do not interfere. In many instances, that declaration is at odds with any deep appreciation of human suffering, and with the sympathy that pervades A Crack in Creation.
The more serious argument perceives risks involved in germline interventions. Human knowledge is partial, and so perhaps we will fail to recognize some dire consequence of eliminating a particular sequence from the genomes of all members of our species. Of course, it is very hard to envisage what might go wrong — in the course of human evolution, many DNA sequences have arisen and disappeared. Moreover, in this instance, assuming a version of CRISPR-Cas9 sufficiently reliable to use on human beings, we could presumably undo whatever damage we had done. But, a skeptic may inquire, why take any risk at all? Surely somatic interventions will suffice. No need to tamper with the germline, since we can always modify the bodies of the unfortunate people afflicted with troublesome sequences.
Doudna and Sternberg point out, in a different context, one reason why this argument fails: some genes associated with disease act too early in development (in utero, for example). There is a second reason for failure. In a world in which people are regularly rescued through somatic interventions, the percentage of later generations carrying problematic sequences is likely to increase, with the consequence that ever more resources would have to be devoted to editing the genomes of individuals.  Human well-being might be more effectively promoted through a program of germline intervention, freeing those resources to help those who suffer in other ways. Once again, allowing editing of eggs and sperm seems to be the path of compassion. (The problems could be mitigated if genetic testing and in vitro fertilization were widely available and widely used, leaving somatic interventions as a last resort for those who slipped through the cracks. But extensive medical resources would still be required, and encouraging — or demanding — pre-natal testing and use of IVF would introduce a problematic and invasive form of eugenics.)
What of enhancement? Here, the case against using tools of gene editing appears even stronger. Nevertheless, as Doudna and Steinberg partly appreciate, revulsion stems from fixating on a specific type of example. When ambitious parents hope their children will exhibit particular characteristics — being tall or intelligent, for example — the desire is often comparative: they want the kids to be taller or smarter than their peers. Genetic enhancement should not cater to that sort of wish. A society in which privileged people buy further biological advantages for themselves and their dependents is an ethically hideous prospect, as exemplified by the alphas, betas, gammas, and deltas of Aldous Huxley’s Brave New World.
When competition plays no role, however, genetic enhancement can be harmless, even benign. The losses affecting us as we age are familiar facts of human life. Hearing becomes less sensitive, and memory declines. Although the causal details underlying these processes are not yet fully known, it is easy to imagine that they might be discovered — and that the discovery could allow somatic interventions to preserve our youthful capacities as we age. People benefiting from those interventions would be genetically enhanced, equipped with abilities no normal human being has ever had. If the interventions were available to all, parts of the standard health protections delivered by all (enlightened) societies, it is hard to see what objections could be leveled against them.
In entering into dialogue with A Crack in Creation, I pay tribute to what Doudna and Sternberg have accomplished. They belie an old joke. A prominent neurosurgeon, seated at dinner next to a historian of science, declares his interest in his companion’s profession: “I’ve often thought that, when I retire, I’ll write on the history of neuroscience.” To which his neighbor replies, “How interesting! I’ve often thought that, when I retire, I’ll take up neurosurgery.” The care and patience with which Doudna and Sternberg discuss ethical questions reveals their understanding of the difficulties of the terrain they explore. Such thoughtful immigrants into bioethics are rare — and they deserve a warm welcome.
Yuval Noah Harari is also interested in the threats attending the human future, and impressed with the possibilities of applying biological knowledge to modify human genomes. But in Homo Deus, he paints on a far larger canvas. Scientific advances have provided our species with godlike opportunities. Computer technology and molecular biology together will transform human lives and what it means to be human. Most members of our species will become redundant. All of us will have to face the fact that we are not, and have never been, autonomous agents. The flaws in humanism will be exposed. A new religion in which the flow of data becomes central — becomes the reigning deity — will triumph.
Or will it? Harari is a historian, and his book is advertised as “a brief history of tomorrow.” Writing the history of tomorrow does not seem easy. It is surely not for the faint-hearted. Where are the sources? Perhaps they lie in trends discernible in the present. Yet, as Harari tells us, meditating on history does not typically yield reliable predictions. What, then, is the basis for Homo Deus’s provocative forecasts? Perhaps, despite the assertoric mode, they are not predictions after all? Are they invitations to consider possibilities? The book ends with three questions — whose topics are in fact theses, previously firmly announced and used to support many specific conclusions. Should the book have been composed of questions all along?
Harari thinks big, and his earlier book, Sapiens, a wide-ranging account of the history of our species, was rightly acclaimed. The intellectual courage on display in that book, the willingness to cross boundaries and draw on the findings of many different fields, pervades Homo Deus. But whereas the simplifications of Sapiens, its bold striding over complexities and nuances, could readily be forgiven, the corner-cutting in Harari’s futurology matters.
Many contemporary scholars are concerned about aspects of the world today that figure briefly in the book’s breathless discussions. The human future is threatened by our on-going habit of emitting greenhouse gases, by intra- and international inequality, by the proliferation of nuclear weapons, and by political movements that turn away from trans-national cooperation. Although these sources of potential disruption occasionally interrupt the flow of Homo Deus, they are mentioned only in passing, and breezily at that. Apparently untroubled by how they might disrupt his projections, Harari’s discussion moves on to the real motors of future change: increased biological knowledge and ever-advancing informational technology. Rightly impressed with the molecular biology of the past half-century, he projects a biological understanding of human behavior beyond the dreams of anyone who appreciates the difficulties of fathoming complex human traits. His confidence that artificial intelligence will replicate all human capacities seems more like an avowal of faith than a well-grounded prediction.
Readers of Homo Deus wait in vain for the Doudna touch — in other words, for a clear recognition of what has been achieved and a sensitive reflection on how it might valuably be employed. Harari’s stampede to the post-humanist future is unchecked by ethical ruminations. We’re destined to be dominated, even replaced, by our machines. That’s it, like it or lump it. But perhaps the refusal of ethical discussion is part of the point. Writing con brio, exultantly announcing a world many will find repugnant, playing the enfant terrible — is this a way of provoking thought about what we value and would hope to preserve? Not a history of tomorrow, but a dystopian vision that might scare us into changing our ways?
The gods glorified in Harari’s post-humanism are capricious, superficial, and cruel. Not for them the knowledge of good and evil promised by the serpent. The human beings who aspire to be this kind of deity are feckless children, addicted to the toys that dominate their lives. Homo Deus might best be read as a preface to a series of books — volumes emulating the mature wisdom shown by Doudna and Sternberg, and wrestling, as they do, with the details of a particular set of problems. Humanity surely needs more grown-ups.
 Reviewers of A Crack in Creation are faced with an unusual decision. It is a co-authored book, yet one written in the first-person singular. Because Doudna is well known for her development of CRISPR-Cas9, and because the authorial voice seems to be hers, I shall treat her as the sole author of passages in which she is recording the course of the research and expressing her growing concerns about its potential applications. The general discussion of ethical questions in Part II is, however, almost certainly the joint product of the two authors. It is also possible that many of the formulations throughout the book owe much to Sternberg.
 Argued at greater length in Philip Kitcher’s The Lives to Come: The Genetic Revolution and Human Possibilities, New York: Simon & Schuster, 1996.
 The point is well made by Russell Powell in “In Genes We Trust: Germline Engineering, Eugenics, and the Future of the Human Genome,” Journal of Medicine and Philosophy, 40, 2015, 669-95.