Introduction

On May 27, 2021, the International Society for Stem Cell Research (ISSCR) released its long-anticipated new guidelines for stem cell research,^[1] updating its earlier Guidelines from 2016.^[2] The same day, ISSCR working groups published five articles related to the updated Guidelines^[3] and the chair of the overall working group, Robin Lovell-Badge, published an article describing the need for the updates.^[4]

The updated Guidelines and their accompanying commentary cover a wide range of topics and provide well thought out and useful guidance on many and varied topics, from human/non-human chimeras to in vitro gametogenesis to human germline genome editing.^[5] But, inevitably, the headlines focused on the most controversial recommendation: the perceived elimination of a strict ban on keeping human embryos alive in a laboratory for more than 14 days (the so-called 14-day rule).^[6]

The headlines weren’t entirely right about the nature of the recommendation, but they didn’t get it entirely wrong either—and, at least in one respect, the ISSCR got its recommendations badly wrong. It proposed, at the very least, serious consideration of the elimination of the 14-day rule without suggesting any other time limit—or even recognizing the need for such a limit.^[7] The absence of any time limit is an important mistake, albeit one made for understandable reasons.

Here, I propose a different limit, based not on the characteristics of the embryo at any particular developmental stage but rather on the availability of alternative research tools. We should largely stop looking at the normal stages of development of human embryos and their characteristics at those stages in order to judge whether embryos outside a human uterus (“ex vivo”) can be used ethically. Instead, we should look primarily at the alternatives to using ex vivo embryos in research, mainly the availability of aborted or miscarried embryos and their value for the proposed research. This approach leads me to support, tentatively, an extension of research on ex vivo human embryos from 14 days to 35 days, or five weeks, after fertilization. And I also have a suggestion for how the new rule should apply to “embryo models,” human “embryo-like” things not derived from a fertilized egg or human gametes at all.

This article addresses those issues in five parts: the nature and history of the 14-day rule, the new ISSCR position, the problems with simply extending the 14-day rule, and why at least an interim ban on ex vivo (“outside the living” and thus in the laboratory) human embryo research past five weeks post-fertilization makes sense. And then there is a coda: what should we do about “embryo models”?

I. A Short Retelling of the History of an Embryo—and of the Fourteen Day Rule

Like Louise Brown, the first “test tube baby,” the 14-day rule was born as a result of the development of human in vitro fertilization (IVF) in 1978.^[8] IVF provided researchers with access to living human embryos that were not inside a woman. “Left over” embryos from IVF gave researchers the means to study early human embryonic development and to see how it worked, when it worked well, and when it went awry. But those “tools” created new ethical questions: was it ethical to use living human embryos as the objects of research, with the intention of destroying them when the research was done, and, if so, for how long?

Normally, about six or seven days after their creation, at the time of fertilization, human embryos have grown to the point where they are called blastocysts.^[9] Blastocysts are rough spheres, just at the edge of visibility to the naked eye (if the naked eye is sharp and the lighting is right).^[10] The blastocyst has more than a hundred cells, most of which make up the outside of the sphere, called the trophectoderm.^[11] These cells eventually develop into the placenta, amniotic sac, and most of the other “support” structures for the pregnancy.^[12] Inside the sphere is a much smaller number of cells, called “the inner cell mass,” most of which will eventually form the body of the embryo and, if things go well, the fetus, the baby, and the adult.^[13] At this stage, the blastocyst exits from the one of the two fallopian tubes in which the embryo was formed and moves into the uterus.^[14] It will then try to burrow into the uterus’s endometrium. If it succeeds, it will have “implanted” and established a pregnancy that might, or might not, result, about 260 days later, in a baby. ^[15]

Note that the timing of pregnancy and embryonic development can be very confusing. For one thing, embryonic development, at least in the early stages, is timed in terms of days of development since fertilization of the egg by a sperm.^[16] The “days of development” approach is necessary to account for the possibility that the embryo might have been frozen after fertilization and before use.^[17] An embryo might be six developmental “days old” even though it was created five years earlier if it had been frozen in the first few days after its creation. More confusingly, pregnancy is generally counted as weeks since the first day of the pregnant woman’s last menstruation, referred to as “last menstrual period” or “LMP.”^[18] That typically occurred about two weeks before a new embryo will have been created by fertilization and three weeks before it will implant in the uterus to “start” a pregnancy.^[19] Thus, when a woman is four weeks pregnant, by the LMP calculation, the embryo she carries is only two weeks old. And a woman at full term, about 40 weeks, is carrying a fetus that is 38 weeks old. (An embryo magically becomes a fetus, at least in name, at eight or nine weeks after fertilization and, therefore, at ten or eleven weeks “LMP.”)^[20]

In 1979, just a year after Louise Brown’s birth, an American advisory committee suggested that embryos in the laboratory could ethically be kept alive as an object of research but only for 14 days.^[21] A British commission headed by Mary Warnock in 1984 made the same suggestion, with more effect.^[22] The British commission’s position, unlike the American recommendation, was enacted into law in the United Kingdom in 1990. Another American advisory committee made the same recommendation in 1994, although again without prompting federal legislation.^[23]

All three groups picked 14 developmental days after fertilization as the appropriate limit for such research, though their reasons varied somewhat. The first group was influenced by the idea that implantation was generally completed by two weeks after fertilization.^[24] The second and third groups cited the formation of “the primitive streak.”^[25] The primitive streak is a layer of cells of different shading that for the first time distinguishes between the top and bottom of an embryo.^[26]Its formation marks the beginning of a several day process, known as gastrulation, when the embryo’s cells clearly differentiate into three different broad cell types.^[27] This is also the first visible sign of the cells that will become, among other things, neural tissue. And the formation of the primitive streak is thought to mark the end of the time when an embryo could split and create two identical twins.^[28]

None of the groups argued that the limit they proposed was a strong moral line or one that would necessarily last forever. Even at that time, some people wanted no experimentation with human embryos while others wanted no limits, or very distant ones, on such experimentation. Still, the 14-day rule was widely accepted^[29]—no doubt in part because at that point no one knew how to keep an embryo alive outside a uterus for more than seven days. It is easy to promise not to do something that you cannot do.

II. The New ISSCR Guidelines

The headlines were right: the new ISSCR Guidelines do recommend eliminating the 14-day rule—sort of. The earlier ISSCR guidelines had divided research into three categories: (1) permitted, (2) permitted with substantial oversight, (3) forbidden. ^[30] The new Guidelines subdivide groups 1 and 3 into two parts each.^[31] Research in Category 1A is permitted with no oversight; Category 1B is permitted with limited oversight.^[32] Category 2 remains permitted with substantial oversight.^[33] Category 3A is forbidden unless scientific or ethical views change; Category 3B is forbidden, period.^[34] The new Guidelines move research with human embryos after more than 14 days of development from the old Group 3 (forbidden) to Group 2 (allowed with substantial oversight).^[35] But the Guidelines, and the ISSCR commentary on them, say that researchers should not attempt such experiments unless their national jurisdictions allow them, which many currently do not, and unless the change has substantial public support—all of which makes research with human embryos after more than 14 days of development seem more appropriate for Category 3A.^[36]

This error is confusing—the ISSCR says that such experiments are no longer forbidden; instead, they are allowed with substantial oversight, but, on the other hand, only where jurisdictions and popular sentiment agree to it. But a second aspect of the new Guidelines is a serious mistake: they propose absolutely no new limit on the length of development of, and research with, ex vivo human embryos.

Nothing in the new Guidelines would prohibit, in jurisdictions that abandoned the 14-day rule, researchers from allowing ex vivo human embryos to develop for eight or nine weeks post-fertilization (when they change names, at least in utero, and become fetuses). Or for 12, 16, or 20 weeks. Or to viability, currently about 20 or 21 weeks post-fertilization. Or, for that matter, to 38 weeks post-fertilization and full term. ^[37] Of course, to this point, as far as we know, no one has maintained a human embryo for more than 14 days. We do know embryos in vivo can go much further—you and I and all of our fellow 7.8 billion living humans all did—but we do not know whether or how such extended gestation could possibly be sustained outside a human uterus. On the other hand, we do not know that, eventually, they might not be.

So the ISSCR wanted to remove the 14-day limit, but, I suspect, could not agree on another logically, scientifically, and politically defensible limit. To be fair, it undoubtedly expected jurisdictions that approve a change in the 14-day limit would set a developmental limit, or, more likely, different limits in different jurisdictions. It would have been better if the Guidelines had only said “morally, there has to be some limit, short of fetal viability, but we expect the approving jurisdictions to set such limits.”

III. The Problem of a “Not Expressly Limited Extension” of the Fourteen Day Rule

Whatever their backstory, the Guidelines clearly encourage going beyond 14 days. It is unusual for me to oppose expansions to biological research. But in this case, I do for two reasons: (1) the absence of an appropriate, and defensible, limit on the extent of development of these ex vivo research human embryos and (2) uncertainty about the need for this proposed research. Before I go into those concerns, though, I must discuss my view of the moral status of the human embryo.

I believe that human embryos, fetuses, and born people normally acquire additional moral status as they develop.^[38] An implanted and growing embryo has greater moral status than a zygote in a Petri dish or in a fallopian tube. A viable 21 week post-fertilization fetus (23 weeks by LMP) has higher moral status than a 10-week embryo. A live born baby has higher moral status than the same entity just before it was born. A competent born human has higher moral status, at least in terms of issues like autonomous decision-making or voting, than a newborn infant. Some of these differences may be based on the capacities of these different stages of what are all former zygotes: to what extent it can perceive things, or feel pain, or survive outside the womb. Other differences, notably between a fetus an hour before birth and a baby an hour after, may be based more on familiarity or tradition than logic—although, even there, the newborn has passed one more test by being born alive, one that historically many late term fetuses failed.

I would not condone intentionally harmful, let alone deadly, research on a newborn infant.^[39] Neither would I agree to it being performed on a fetus developed enough to be viable if delivered from its gestational mother. I am comfortable with research on blastocysts in the laboratory, before transfer and possible implantation into a woman’s uterus. But, leaving aside whether my level of comfort is meaningful to anyone other than me, that still leaves me with a gap of about 19 weeks of gestation in which I cannot find a compelling line. I see no logically compelling and non-arbitrary lines between fertilization and viability. Such lines have largely been discussed in the context of time limits for abortion rather than for embryo research, but the idea is the same: after some developmental stage a human embryo (or fetus) should not be destroyed—or kept alive ex vivo as a non-consenting experimental subject. It is to this problem that I now turn.

A. The Problems with Using Developmental Markers to Set Limits

1. In General

Why call for eliminating the 14-day limit without proposing a new one, or at least acknowledging that a new one will be needed? Because finding a limit is hard. Fertilization is a logical stopping place. Viability is also a logical, if somewhat vague and technologically moving, stopping place. But what to do with the roughly 19 weeks between those two points? Using the first one would eliminate all human embryo research but using the second would let it go on to a point many people, including me, would find intolerable. In the contexts of both embryo research and abortion, people have debated these lines endlessly and without consensus.

For some people, the answer is easy. Life and the right to life (and respect) begins at fertilization—even though the majority of fertilized eggs never become fetuses, let alone babies.^[40] The U.S. “Dickey-Wicker Amendment,” which since 1995 has almost entirely prohibited the National Institutes of Health from spending money for research in which embryos are harmed or put at risk of harm, seems to incorporate this view, at least with respect to government funding.^[41] (That last caveat is important because the Amendment might be viewed not, in itself, as an endorsement by the government of that view as much as an unwillingness to force taxpayers who hold that view to pay for such research.)

For others, it may be at the time limit for successful implantation in a woman’s uterus, although that line seems irrelevant to embryos that are never transferred to a uterus. I think the closest thing I know to a political expression of this view was former Senator Orrin Hatch’s position during the embryonic stem cell debates of the early 2000s that embryos that were ex vivo, and hence by definition not implanted in a uterus, could be used in destructive research.^[42] (At that time, human embryos had only been kept alive outside a uterus for about a week; I can only guess at Hatch’s view of research use of embryos up to the 14-day rule, let alone past it.)

But, if those dates seem too early, as they do to proponents of even today’s 14-day rule, what are possible lines after fourteen days (and the formation of the primitive streak)? I can see at least 10 but each seems, in different ways, unsatisfactory, at least to me. (All times given are post fertilization, not post LMP.) Here are those possible lines in roughly—but not exactly—developmental order based on sometimes uncertain estimates about when they occur:

1. Appearance of specialized tissue that will become the central nervous system (about weeks 3 and 4).^[43]

2. Detection of electrical impulses that will later signal a heartbeat (about week 3).^[44]

3. First apparent blood flow through what will become the heart (end of week 4).^[45]

4. First differentiated neurons with synaptic connections (late in week 5).^[46]

5. First neurons in the developing brain (the start of week 7).^[47]

6. Change of name from embryo to fetus (end of week 8).^[48]

7. Beginning of ability of the fetus to respond to sound (variously stated as at week 16 and weeks 21 to 23 ).^[49]

8. Viability (about week 21).^[50]

9. Beginning of the ability of the fetus to experience painful stimuli (about week 27).^[51]

10. Full term (about 38 weeks).^[52]

I will discuss these limits and give my own reactions to them, not in an effort to convince you that my views are correct, but rather to illustrate why finding a consensus (at least before viability) is so difficult.

The potential markers fall into four groups: limits based on the initial signs of a developing organ or tissue of special importance, limits based on the beginning of particular functions or abilities (including the ability to live outside the laboratory or uterus), one limit based on the transition in name, and one limit based on “completion” of the normal prenatal development process.

Please note that, as of today, we have no proof that any of these points can be reached by ex vivo human embryos or fetuses. The longest any human embryo has remained alive in the laboratory is 13 days,^[53] although embryos from Rhesus macaques have been reported to be alive ex vivo up to 20 days after fertilization.^[54]

2. Viability (and Full Term)

Of my ten options, only two of these possible lines are, to me, both logically plausible and (relatively) easy to ascertain: viability and full term. Even the line of viability is more statistical and changeable than it is clear-cut and permanent. There is no definite line that marks a firm border between babies that can survive and those that cannot. Instead, there is a survival curve that gets increasingly close to zero. Whether the end of the tail has been reached at any point is impossible to know.

In 1973, when Roe v. Wade was decided, viability was placed at about 26 weeks after the last menstrual period; nearly 50 years of medical progress has lowered that by about 3 to 4 weeks, or about 15 percent. But that is just when the chances of survival become significant; there are outliers. The current record holder for earliest surviving birth turned one in summer 2021 after being born 21 weeks and two days after his mother’s last menstrual period, probably a bit more than 19 weeks after fertilization.^[55] As technology improves, so will the odds for a baby’s survival at that level of development.

Still, viability, for all its uncertainty, is a line the U.S. Supreme Court has, at least as of this writing (if not of your reading),^[56] clung to as the moment when the state’s interest in preserving life (and the fetus’s interests in having its life preserved) become (usually) more important than the pregnant woman’s interests. And American law seems always to have taken the position that anyone (or anything) born alive from a woman is a human being, and, in the eyes of the law (and of the U.S. Constitution), a “person”—although, as far as I can tell this has never been expressly legislated or even clearly stated in case law.

For me, viability would serve as one clear end point for fetal experimentation. When a fetus could survive outside the uterus, or whatever uterus substitute is employed in the laboratory, its use in any kind of risky, let alone destructive, research, without (obviously) its consent, becomes almost entirely unacceptable. This, of course, is even more true when a fetus passes viability and reaches full term. My only exceptions would be clinical research that would benefit that particular fetus or research of truly minimal risk.

This is similar, but not identical, to the position of the Common Rule, as adopted by the U.S. Department of Health and Human Services (“HHS”), for research on fetuses (a term it defines as “the product of conception from implantation until delivery” [emphasis added] and hence includes what science and medicine call implanted embryos).^[57] For fetuses, the Rule allows research only if the relevant parts of ten conditions are met, including importantly,

The risk to the fetus is caused solely by interventions or procedures that hold out the prospect of direct benefit for the woman or the fetus; or, if there is no such prospect of benefit, the risk to the fetus is not greater than minimal and the purpose of the research is the development of important biomedical knowledge which cannot be obtained by any other means…^[58]

This regulation does not, of course, apply to the ex vivo “products of conception” in question because, by definition, they have not implanted. However, applying the regulation to “products of conception” that have reached the stage of viability seems sensible to me.

This discussion of viability raises an uncomfortable question, one that I would banish to a footnote except that it seems potentially too important. What would be the status of a fetus after the time and general development normally needed for viability if, as a result of intentional engineering or otherwise, it could not be viable? Assume, for example, a fetus with no lungs or with lungs that could not function. It could survive but not after its birth and the consequent discontinuance of oxygen delivery through umbilical cord blood. Should continuing research be allowed with such a doomed fetus, whether the defect was accidental or intentionally created? I do not currently have a good answer.

Again, the Common Rule may provide at least a straw in the wind. It has provisions for nonviable neonates (newborns), defined as “a neonate after delivery that, although living, is not viable.”^[59] It defines viable, with respect to a neonate, as “being able, after delivery, to survive (given the benefit of available medical therapy) to the point of independently maintaining heartbeat and respiration,”^[60] although the Secretary of HHS is authorized to publish guidelines to take into account medical advances that might bear on viability. The Rule states, “After delivery [a] nonviable neonate may not be involved in research covered by this subpart unless all of the following additional conditions are met,” the first two of the five conditions providing that

1. Vital functions of the neonate will not be artificially maintained;

2. Research will not terminate the heartbeat or respiration of the neonate….^[61]

It is hard to see how post-“viability” but non-viable fetuses used for research purposes could ever meet the first condition (that their vital functions not be artificially maintained) as whatever ex vivo apparatus has taken the place of a uterus would still, presumably, be in use. On the other hand, I can only speculate on details of a procedure that apparently is not even being developed yet. I confess that it is not clear to me why, logically, either of those conditions, set forth in the Common Rule for babies born alive and hence “persons,” should necessarily be applied to something that has not yet been “born,” and hence not yet a “legal” person. But, at the same time, they seem to me good positions.

Ultimately, my current (though not fully settled) position on ex vivo fetuses sustained past the normal point of viability becomes fairly simple. For purposes of research regulation, treat them the same as viable fetuses (or, if they are not viable for reasons other than their level of gestational development, like non-viable newborns). Note that it is very unclear when, if ever, it will become technologically possible to sustain an ex vivo fetus until the stage of normal fetal viability (except for “no time soon”). I would extend the same approach to an intentional destruction of a viable ex vivo fetus. Treat it under whatever laws govern traditional pregnancies, which, in many states and circumstances, would make it illegal under laws restricting late abortions. And, when the fetuses are untethered from whatever version of an artificial womb that has nurtured and sustained them, treat them as fully human babies and persons for purposes of the Constitution. That should, at the very least, cause some complicated problems for research institutions and judges. But, should technology enable it, we should no more discriminate against babies sustained outside a womb than babies created through artificial insemination, IVF, or, for that matter, non-marital sex.

3. Other Markers

Markers tied to perception or the ability to respond to perceptions are significant for me, as are any other indications of some form of consciousness, whatever we might mean by that term. So the ability to feel pain causes me, at the very least, concern about research that might inflict such pain. After all, we recognize the interests of even (vertebrate) non-human research subjects in avoiding pain (although in some cases we still authorize its infliction).^[62] The existence of a brain, or evidence of some kinds of brain functioning might also be relevant, perhaps largely (but not entirely) as a surrogate for the ability to feel pain. But, although every fact about embryos and fetuses remains controversial, or at least controverted, it appears at this point that these abilities require the development of the brain past the (fuzzy) point of viability at roughly 20 weeks after fertilization.^[63] As I would impose substantial restrictions on research from the moment of viability, any later markers would be redundant for my purposes.

Markers based on the first appearance of cells or tissues that will later become important for the functions of the tissues or organs they develop into hold no appeal to me. An argument from function may work, but an argument from the appearance of a cell type that will, if all goes well, lead to that function, does not. After all, one could say that the zygote marks the first appearance of a cell that could become a heart, a brain, or a conveyor of pain.

Heart-based markers, although popular with some legislators,^[64] seem particularly unconvincing. The heart is a pump, essential to (independent) life but not more so than a liver, a brain, or many other organs.^[65] The permanent cessation of its function has long been used as a marker for death,^[66] probably because it was relatively easy to detect (though much less so before the 19^th century invention of the stethoscope), but that does not make it a special marker of the beginning of meaningful life.

And the change in nomenclature from embryo to fetus (a change made only in humans; non-human mammals remain “embryos” until birth) seems completely arbitrary. The best explanation for this change I have seen is that it marks the time when all of the (major?) organ systems have come into existence,^[67] but coming into existence says nothing about coming to function. (The lungs, for example, never function prenatally, only after birth.^[68])

4. And Yet

So far this seems to be a problem with a fairly simple answer—if science makes it possible to keep ex vivo embryos and fetuses alive for not weeks but months, and if researchers can show that research on such ex vivo embryos and fetuses is likely necessary for important work, then permit experimentation on ex vivo human embryos and fetuses until the time of viability, albeit roughly defined. But that solution in turn poses two problems.

One problem is political. At, say, just short of 20 weeks post-fertilization, a fetus will be roughly the size of a coconut and may weigh almost one pound and be about 10 to 11 inches long.^[69] It will move in ways that have been apparent to its host for several weeks.^[70] It will look like a human baby—a very small one, but definitely a baby and not, say, a shrimp. It seems unbelievable that research on such a living entity would not prompt a strong popular revulsion and a backlash against all ex vivo fetal research.

Jon Shields has made this same point about the American public’s views of abortion:

Since the pro-life movement coalesced, its most important mobilization tool has been images from second- and third-trimester abortions. They’ve emboldened countless activists, giving them the confidence that they are waging a war for basic human rights. . .Without such clinics and the images that leak out of them, it may be harder for pro-life leaders to sustain the moral passions of their movement — as well as the fiction that most aborted fetuses resemble newborns…Like their pro-choice counterparts, pro-life activists simply can’t muster much feeling for embryos that are not recognizably human.^[71]

My other problem is personal. At some point after 14 days but before viability, for me the ex vivo human conceptus becomes something that neither should be kept alive without an intent to have it become a baby nor be used as a subject for experimentation.^[72] But I honestly do not know when I think that happens. I would get queasy well before the roughly 20 week start of viability. Some of that concern stems from the growing and developing brain and the likelihood that awareness, responsiveness, pain perception, and consciousness are getting closer and closer as development continues.

But I think some of my reaction is visual. I seem to be part of Shields’s generalization about the American public: the more a fetus looks like a baby, the more uncomfortable I am with using it for research (outside the exceptions noted above) or keeping it alive ex vivo with no intent to allow it to become a baby.^[73] Similarly, I would not support using a fully anencephalic infant as a research subject or extending its life solely for such a research use.^[74] I feel the same about a person who has no higher brain functions, one who may not be dead by neurological criteria under the U.S. Uniform Definition of Death Act or various other state and national definitions of death^[75] but who has convincingly lost the capacity for any sensation or consciousness.

I even have qualms about doing invasive research on someone who has been declared “brain dead” (dead through neurological criteria) but whose heart is kept beating and body maintained purely for invasive research. This last action was recently performed when a specially treated pig kidney was transplanted into a “living corpse,” the living body of someone who, from lack of brain function, was dead.^[76] I honestly do not know (yet) what I think of this or similar experiments. It might not be human “vivisection” on a “living human body” that used to be, but no longer is, a “living person,” but it comes awfully close for my comfort.

I worry that this is purely an emotional reaction on my part. I know I can try to justify it on spill-over grounds—the more we allow mistreatment of “things that look like persons” even if they are not, the more we encourage mistreatment of actual persons. But that’s an empirical assertion, albeit about a currently non-existent state of affairs, and one I’m not confident I believe.

Some, notably Leon Kass in a famous essay called “The Wisdom of Repugnance,” have tried to argue that such negative visceral reactions are, in themselves, entitled to be given normative weight in setting the “should.”^[77] I loathe this argument, in part because many of the things we hold dear today were viewed by most people with repugnance not very long ago, from religious freedom to racial and gender equality to same sex relationships.

Others have stressed the importance of a general concept of “human dignity,” not in terms of the dignity of individual humans but of the human species.^[78] I have been unmoved by these arguments as well, finding them generally the last resort of conservative bioethicists who cannot find a secular argument against something they strongly oppose that does no harm and that some will think does good. (Samuel Johnson’s statement that “patriotism is the last refuge of a scoundrel” comes to my mind here.^[79])

Finally, some ethicists, secular and religious, have argued for a special need to treat with dignity the remains and bodies of humans.^[80] These arguments, or ones based on them, may ultimately provide me with some rational cover for my aversion, but that remains a work in progress. They have not convinced me yet.

Nonetheless, in addition to an extremely strong political argument against such research, my personal view, whether or not totally emotional, remains powerful. I would not allow such research at some point well before viability—but well after 35 days.

Note that this concern about research on ex vivo fetuses before viability does not affect my views on abortion. My personal view is that abortion should be legal to the point of fetal viability and, in extraordinary cases directly involving maternal health, beyond. I support allowing abortion later in pregnancy past the point where I would support a continued ex vivo “pregnancy” solely for research purposes because in the latter case no person’s body is tied up (almost literally) with the “pregnancy,” nor, less literally, is any woman’s future. A laboratory simply has no interests comparable to the rights of a pregnant person to bodily autonomy and self-determination.

For the 14-day rule, I’m left with a sense that sometime after 14 days, but well before viability, a fetus (or possibly an embryo) should not be kept alive ex vivo without any intention to have it become a living baby, and it should not be used for any research not for its direct benefit that involves more than minimal risk. But I cannot say “when.” And my discomfort here gives me great sympathy for the ISSCR Committee that tried to grapple with extending the 14-day rule while finding a developmental marker to provide some limit to its expansion.

B. Using Potential Benefit as a Limit

But developmental markers may not be the only hope for such a limit. This section explores another: the need for —and hence potential benefit from—such research.

A key question for all research is whether the potential benefits outweigh the potential harms.^[81] Research with living ex vivo human embryos might be needed to improve scientific knowledge of how embryos develop and to discover better medical ways to avoid miscarriages or disabilities stemming from problems in those early weeks.^[82] Or it might not.^[83] Exploring that question leads down two paths. One asks whether there are questions about embryonic and fetal development where answers are likely to be important for human health and wellbeing, or for “science.”^[84] As a general matter, that seems highly likely. Nonetheless, for any particular proposal to use ex vivo embryos past 14 days (or, for that matter, up to 14 days), one should require a recitation of the specific questions the research seeks to answer and convincing explanations of why they are important and how this research could help lead toward answers.

The second path forks into two questions, asking both whether living ex vivo embryos are likely to be both useful for answering those questions and necessary for doing so. Utility should not be assumed. After 7 or 8 days, ex vivo human embryos are not in anything close to a natural situation.^[85] They have not implanted by burrowing into the endometrium of a human uterus.^[86] They are not connected to a host human by a placenta that brings them oxygen, nutrition, and a wide array of hormones and takes away carbon dioxide and waste.^[87] They are attached to a plastic structure and bathed in a mixture of fluids called a culture medium.^[88] Whether and to what extent their development mimics what happens to a real embryo, in a real human uterus, is deeply unclear.

This might become somewhat clearer through research with non-human primates. It may be possible to compare, say, 20 day ex vivo macaque embryos with in vivo 20 day embryos that were recently removed from a macaque’s uterus.^[89] It is possible that the development patterns will be quite similar, giving us some reason to hope—and perhaps even to expect—that the same will be true of 20 day human ex vivo embryos. It is also possible that they will not. Without substantial evidence that breaking the 14-day rule offers a strong hope of generating useful knowledge, why break it?

But the other question is also important. If the relevant information can be obtained without using post 14-day embryos, why do it? It does seem hard to imagine getting sufficiently detailed information from living, in vivo human embryos without deeply unethical action, although perhaps better remote imaging or monitoring could be valuable.

Studies of non-human primates could provide an approach less ethically fraught. Yes, every species will have its own variations, and these sometimes seem particularly strong in reproduction, even between closely related groups, but it is possible that non-human embryo research might provide answers to some of the specific questions (even if those answers turn out to be that the issue in question cannot be satisfactorily answered).

And, finally, how much of this research could be done with human embryos that had implanted in a woman’s uterus but were expelled through miscarriage or abortion? Although some people have concerns about the use of fetal tissue, particularly from intentionally aborted fetuses, the ability to use it in U.S. federally funded research is currently codified in federal law.^[90] If we can learn “enough” from this source, we may not need to use living ex vivo human embryos. After the first seven days of embryonic development, during which human embryos can develop normally outside the body, for research on human embryonic and fetal development, scientists have to rely on the results of miscarriages or induced abortions, but, as discussed below, at some stages of development that proves difficult.

Perhaps focusing on when this kind of research seems particularly necessary to obtain important results will help define a limit to ex vivo embryo or fetal research. At least, so the next section argues.

IV. A Modest Proposal

Our current understanding of human embryonic development has a “dark zone” during the period between seven (or, perhaps, 14) days and about four or five weeks post-fertilization.^[91] I propose that jurisdictions and advisory committees change the 14-day rule to ban, on at least an interim basis, on research on ex vivo embryos more than 35 days after fertilization, with research after seven and before 35 days allowed on a case-by-case basis. This entails a two-step approval process. The first step would require a general decision to allow it, based on a showing of the likely value of, and the need for, ex vivo human embryonic research during that developmental period. This would presumably be achieved by legislation, regulation, or a broadly accepted guidance document. The second step would require a specific approval, probably by a local committee, of the particular research protocol, based on a finding that the researchers had shown a need for their proposed research with such embryos. The committee could either be an existing human subjects research committee (an Institutional Review Board, or IRB in the U.S.) with a mandate expanded to include ex vivo embryos and fetuses or a new but similar committee.^[92]

Why this proposal? Because a gap exists in our knowledge of human embryonic development.^[93] We can see, in the laboratory, the development of normal embryos through roughly day 7, the time when, in the usual course of development, embryos need to begin to implant.^[94] We can also see, in at least a few laboratories, embryos up to day 13 that were normal, at least until the substitution of an ex vivo framework for normal implantation.^[95] (How accurately these embryos reflect implanted embryos between days seven and 13 is unknown.) And we can find miscarried or aborted embryos from about four or five weeks after fertilization.^[96]

But between those times falls a shadow. Certainly, some embryos have been found and examined during that stage of development.^[97] The Carnegie stages of embryonic development, based on actual embryonic or fetal remains, cover 23 stages in the first 60 days of human development; its developers must have found some embryonic remains from this period.^[98] But there aren’t many. And those few that were found and studied were probably the result of unusual circumstances or, more likely, chance.^[99]

This is also a crucial period in embryonic development.^[100] More than half of blastocysts that reach the uterus do not implant successfully.^[101] What happens from about the seventh day, when implantation should begin and proceed, through the next several weeks, by which time the possibility of pregnancy is gone? We do not know.

Some research efforts have been made.^[102] Dr. Heidi Cook-Andersen at the University of California at San Diego, for example, is studying “hatching blastocysts,” at five or six days of development, both those with normal and abnormal morphology (shape).^[103] This morphology is a moderately good predictor of success in implantation and pregnancy.^[104] She has found that, in the poor morphology blastocysts, one of their three cell types, the primitive endoderm, is much more likely to be expressing genes abnormally than the other cell types or than any of the three cell types in blastocysts with good morphology.^[105] That’s a hint that development of this primitive endoderm may be a cause of failure to become a successful pregnancy in the following weeks, but only a hint. Ex vivo embryos from 14 to 35 days might help confirm, or refute, that hint.^[106]

This gap suggests a time limit, based not primarily on the level of embryonic development but on the need for ex vivo embryos:^[107] allow human embryos to be kept alive for research for more than 14 days (should that prove scientifically both possible and important) but not for more than 35 days. Those numbers are driven ultimately by the average human menstrual cycle, of about 28 days with a “normal” range of 21 to 40 days.^[108] By 35 days after fertilization, and roughly 49 days after the last menstrual period, many women will be three weeks overdue for their menstrual period and may suspect, or (if they have tested themselves) know, that they are pregnant.^[109] The possibility of obtaining tissue from a miscarried or aborted embryo at that stage will be substantially higher than in the weeks before, when most women will not know they are pregnant.^[110]

One might choose 28 days instead of 35 days. A few things are likely to happen roughly between 28 and 35 days post-fertilization. Electrical impulses from cells that will eventually become the heart may be detected at about that time.^[111] The neural tube will have closed, encapsulating what will later become the spinal cord and brain.^[112] Those do not seem to be of great moral significance.

But the issue of the developmental markers is not irrelevant to this proposal. Whether the limit is set at 28 or at 35 days, few people would view the development state of the embryo as importantly advanced beyond 14 days.^[113] It is not very functional, it does not have any plausible possibility of perception, it is still more than 100 days from viability, and (for what it’s worth) it does not look much like a baby. The length of the embryo, from the crown of the head to the rump, is about a quarter of an inch at 28 days and a third of an inch at 35 days: between the size of a grain of rice and that of blueberry.^[114] The brain is far from functional, although it is about this time that its precursor cells begin to separate into forebrain, midbrain, and hindbrain.^[115] Neurons and synapses first appear in the spinal cord at about this time as well^[116], but behaviors like sucking and swallowing do not appear for another 10 weeks or so.^[117] The earliest evidence of “brain activity,” through EEG, seems to be about 45 days after fertilization.^[118] Although this proposed limit is driven by the potential benefits,^[119] it seems likely to be relatively “safe,” at least for most people, in terms of the embryo’s moral status.

I suggest this as an interim limit. In time, it might change, in either direction. Perhaps 28 or 35 days after fertilization may prove too many days, if useful tissue proves to be available from miscarriages or abortions before them. Or it may prove to be too few, if we see truly important questions that (1) cannot be answered with tissue from such “ended” pregnancies, or from non-human embryos, or, potentially, at least, from various “embryo models” and (2) can reasonably be expected to be answered with ex vivo embryos. And note once again the importance of that second point: unless we have some good reason to believe that a 40 day ex vivo embryo is a good model for a 40 day in vivo embryo, the justification for the ex vivo research disappears.

So, my proposal is that some “larger authority”—a legislature, an administrative agency, perhaps a widely respected non-governmental organization like a national academy or an international scholarly society—should consider the question of the value, and the need, in general, for research with embryos between 14 and 35 (or 28) days of development. If it finds a sufficient justification for such research, it would authorize it as a general matter. Before any particular research could be done, other authorities—probably local IRBs, embryonic stem cell research oversight committees, or other bodies—would have to review a protocol detailing the specific research and decide that the research may provide beneficial knowledge that likely could not be obtained through other means, including through work with non-human primates, earlier stage human embryos, or human embryonic remains.

V. Coda: What To Do About “Embryo Models”?

Something resembling a very tiny shrimp, about one-third of an inch long, floats in a laboratory dish surrounded by culture medium. On closer examination, it is clearly not a shrimp. It looks like a human embryo at about 30 days after fertilization. But on still closer examination, by an expert and probably with a microscope, it doesn’t quite look like a “normal” 30-day old embryo. And the expert asks, “Just what is that?”

Welcome to the world of human embryo models, or “embryoids,” “blastoids,” “gastruloids,” simbryos," “SHEEFs” (“synthetic human entities with embryo-like features”), or “SHELEs” (“synthetic human embryo-like entities”).^[120] These are human embryo-like “thingies” that were not produced by the merger of an egg and a sperm.^[121] Neither did they come about through cloning, also known as somatic cell nuclear transfer, where electrical stimulation of the product of a fused body cell with a human egg that has had its nucleus removed.^[122] Instead, they are created from human embryonic stem cells, derived from the inner cell masses of destroyed human embryos, or human induced pluripotent stem cells, derived from non-embryonic and non-germ cells (eggs and sperms) treated with special factors that revert them to a stage where they can become many, and perhaps all, human cell types.^[123] I will join the ISSCR Guidelines in referring to them as human “embryo models,” though I will note that neither the scientific literature nor the (largely clueless) popular literature has yet settled on a name for them. This research continues to expand rapidly, with the publication in August 2022, as I was revising the proofs for this article, two different laboratories produced surprisingly long-lived stem-cell derived embryos—although these were only in mice.^[124]

Human embryo models are a special case of a larger set of human organoids, clumps of human cells that develop into intestinal tissue, liver tissue, lung tissue, or brain tissue.^[125] These organoids, though, are limited to one (or possibly a few) tissue types.^[126] Although “human brain organoids” have generated some controversy, for the most part no one is concerned about tiny spheres of gut or liver or other specific tissues in laboratory dishes.^[127] But embryo models hold the promise, or threat, of creating not just a realistic model of the development of some parts of important human organs, but of leading to realistic models for all human organs and tissues…and potentially, of creating new babies.^[128]

We have, in some jurisdictions in the United States and elsewhere, laws about research with human embryos, created from the merger of an egg and a sperm.^[129] In a few jurisdictions, we have laws about human embryos created by cloning.^[130] And in many places, guidelines that are not legally binding are often followed, from groups like the U.S. National Academies of Sciences, Engineering, and Medicine^[131] or the ISSCR. But, with the exception of some very recent guidelines from the ISSCR, nowhere do we have clear laws, rules, or guidance on what to do with human embryo models, or even any recognition of the problems they pose. And that includes the 14-day rule—does it govern embryo models or not?

The ISSCR Guidelines take the position that the 14-day rule does not apply to embryo models.^[132] It was not created with them in mind (and, in fact, it was created decades before they became a possibility), the Guidelines say, and so, therefore, they are not covered.^[133] As a matter of interpretative approach, this is questionable. Many rules apply to things that were not contemplated at the time the rules were created. The Fourth Amendment applies to wiretapping even though telephones (or telegraphs) did not exist when it was adopted.^[134] If the new thing is similar enough to the old one, the rule may well apply.

I suggest therefore that if an embryo model is “similar enough” to a “normal” human embryo, it should be treated as a human embryo for statutory and regulatory purposes, including, but not limited to, the 14-day rule or any revision of it. And I suggest that “similar enough” would be met if the embryo model has a significant probability of being able to produce a living human baby. For this purpose, being able to develop a fetus to the point of viability, the level of development reached in about the twentieth week post-fertilization of a regular pregnancy, would count as a “living human baby.”

This is not a legal conclusion about whether courts or agencies would, or should, treat them as such under any given statute or regulation. That may well depend on the definitions used in such laws, as well as the history of their adoption and the other factors that go into legislative or regulatory interpretation. So, to be clear, I am arguing that new or amended policies, guidelines, or laws should expressly adopt this approach and that existing policies, guidelines, or laws should be so interpreted if possible.

The ISSCR Guidelines do, to some extent, take something parallel to this position. They apply higher scrutiny to “integrated” embryo models than to “non-integrated” models.^[135] Research proposing to use integrated models must, like research involving ex vivo embryos or derivation of new embryonic stem cells lines, be reviewed and approved “through a specialized oversight process.”^[136] Research with non-integrated embryo models need only be reported to such a process, not reviewed by it.^[137]

The ISSCR defines integrated models by saying,

[t]hese stem cell-based embryo models contain the relevant embryonic and extra-embryonic structures and could potentially achieve the complexity where they might realistically manifest the ability to undergo further integrated development if cultured for additional time in vitro… Blastoids are an example of a non-integrated stem cell-based embryo model.^[138]

The extra-embryonic structures, also called extra-embryonic membranes, include the chorion, amnion, yolk sac, and allantois.^[139]

The Guidelines define non-integrated embryo models by saying,

These stem cell-based embryo models will experimentally recapitulate some, but not all aspects of the peri-implantation embryo, for example differentiation of the embryonic sac or embryonic disc in the absence of extraembryonic cells. These stem cell-based embryo models do not have any reasonable expectations of specifying additional cell types that would result in formation of an integrated embryo model. Gastruloids are an example of a non-integrated stem cell-based embryo model.^[140]

By saying that integrated models, and only integrated models, “might realistically manifest the ability to undergo further integrated development if cultured for additional time in vitro,” the ISSCR comes close to admitting that such models might, at least potentially, be able to become viable fetuses.^[141] And the Guidelines imply, or assume, that because non-integrated embryo models, by definition, have no reasonable expectation of becoming integrated models, they could not reach viability.

Those conclusions seem reasonable and quite likely right. But before accepting them as marking a strong line between embryo models that should be subject to the 14-day rule or its successors and those that should not, some empirical evidence would be nice.

The simplest experiment is, alas, one that is ethically impossible—take various kinds of embryo models and transfer them into human uteruses to see if they implant and then develop to viability. But that effort could be undertaken with one or more species of non-human primates, using integrated and non-integrated embryo models from those species. And while, as mentioned before, species differ, when the perfect test is unavailable, less than perfect tests will often have to suffice. I do not know whether anyone will want to undertake such tests, which will involve time, money, and some possible public controversy over the use of non-human primates, all in pursuit of evidence for what will seem to many self-evident. But I think they should both be encouraged and funded to do so by the broader community interested in research with human embryo models.

Conclusion

Ever since IVF provided researchers with good access to living human embryos, those embryos have inspired enormous amounts of scientific research—and ethical and political controversy.^[142] Much of the controversy, I believe, stems from the felt need to find clear developmental markers that can divide appropriate from inappropriate research. Over 40 years of discussion have not yet uncovered such broadly acceptable magic markers. This article proposes that we (largely) give up that effort, at least for now, and try something different. The issues are so intellectually, morally, and emotionally difficult that I make my proposal with sincere (some might say unusual, for me) modesty, in the hope that, whether or not adopted, it may help move us toward a broadly acceptable way to carry out important research, research that may help relieve at least a few of the sources of human suffering.

---

1. Int’l Soc’y for Stem Cell Research, ISSCR Guidelines for Stem Cell Research and Clinical Translation (2021), https://www.isscr.org/policy/guidelines-for-stem-cell-research-and-clinical-translation [hereinafter ISSCR Guidelines].

2. Int’l Soc’y for Stem Cell Research, ISSCR Guidelines for Stem Cell Research and Clinical Translation (2016), https://www.isscr.org/docs/default-source/all-isscr-guidelines/guidelines-2016/isscr-guidelines-for-stem-cell-research-and-clinical-translationd67119731dff6ddbb37cff0000940c19.pdf.

3. Robin Lovell-Badge et al., ISSCR Guidelines for Stem Cell Research and Clinical Translation: The 2021 Update, 16 Stem Cell Reps. 1398 (2021); Amander T. Clark et al., Human Embryo Research, Stem Cell-Derived Embryo Models and In Vitro Gametogenesis: Considerations Leading to the Revised ISSCR Guidelines, 16 Stem Cell Reports 1416 (2021); Insoo Hyun et al., ISSCR Guidelines for the Transfer of Human Pluripotent Stem Cells and their Direct Derivatives into Animal Hosts, 16 Stem Cell Reps. 1409 (2021); Leigh Turner, ISSCR’s Guidelines for Stem Cell Research and Clinical Translation: Supporting Development of Safe and Efficacious Stem Cell-Based Interventions, 16 Stem Cell Reps. 1394 (2021); Eric Anthony et al., New Guidelines for Stem Cell and Embryo Research from the ISSCR, 28 Cell Stem Cell 991 (2021).

4. Robin Lovell-Badge, Stem-Cell Guidelines Why It Was Time for an Update, 593 Nature 479 (2021).

5. See ISSCR Guidelines, supra note 1.

6. Nidhi Subbaraman, Limit on Lab-Grown Human Embryos Dropped by Stem-Cell Body, 594 Nature 18 (2021); Nicola Davis, Time Limit on Lab-Grown Human Embryos is Relaxed by Experts, Guardian (May 26 2021), https://www.theguardian.com/science/2021/may/26/time-limit-on-lab-grown-human-embryos-is-relaxed-by-experts; Rob Stein, Controversial New Guidelines Allow Experiments on More Mature Human Embryos, Nat’l Pub. Radio (May 26, 2021), https://www.npr.org/sections/health-shots/2021/05/26/1000126212/new-guidelines-would-allow-experiments-on-more-mature-human-embryos; The Associated Press, New Guidelines Suggest Lifting ‘14-Day Rule’ on Growing Human Embryos in the Lab, NBC News (May 26, 2021), https://www.nbcnews.com/health/health-news/new-guidelines-suggest-lifting-14-day-rule-growing-human-embryos-n1268628.

7. See ISSCR Guidelines, supra note 1, at 13.

8. “In vitro” means “in glass” and, although the receptacles are now plastic, it is another way of saying ex vivo as opposed to in vivo, “in living,” in this case inside a living person’s uterus.

9. The following discussion is, of course, somewhat simplified but detailed enough, I believe, for the purposes of this paper without leaving the reader baffled and frustrated by unimportant details—although I always find that balance between accuracy and easy comprehension difficult to reach.

10. See, e.g., Mark B. Landon et al., Gabbe’s Obstetrics: Normal and Problem Pregnancies (8th ed. 2020).

11. Id.

12. Id.

13. Technically, the inner cell mass has two parts, the epiblast, which becomes the body of the embryo, and primitive endoderm (also known as the hypoblast), which forms the yolk sac and some other useful, though short-lived, structures.

14. Landon et al., supra note 10.

15. Id.

16. Id.

17. Id.

18. Id.

19. Id.

20. Id.

21. Ethics Advisory Bd., Dep’t of Health, Educ., & Welfare, HEW Support of Research Involving Human In Vitro Fertilization and Embryo Transfer (1979) [hereinafter Ethics Advisory Bd.].

22. Dep’t of Health & Soc. Sec., Report of the Committee of Inquiry into Human Fertilisation and Embryology (1984).

23. Ad Hoc Group of Consultants to the Advisory Comm. to the Dir., Nat’l Insts. of Health, Report of the Human Embryo Research Panel (1994) [hereinafter Ad Hoc Group of Consultants].

24. Ethics Advisory Bd., supra note 21, at 66.

25. Ad Hoc Group of Consultants, supra note 23, at 59.

26. Landon et al., supra note 10.

27. Id.

28. Ad Hoc Group of Consultants, supra note 23, at 59.

29. In some formulations the limit is expressed as 14 days or the formation of the primitive streak, whichever comes first.

30. Int’l Soc’y for Stem Cell Research, ISSCR Guidelines for Stem Cell Research and Clinical Translation 7 (2016), https://www.isscr.org/docs/default-source/all-isscr-guidelines/guidelines-2016/isscr-guidelines-for-stem-cell-research-and-clinical-translationd67119731dff6ddbb37cff0000940c19.pdf.

31. See ISSCR Guidelines, supra note 1, at 9.

32. Id.

33. Id.

34. Id.

35. Id.

36. Id. at 13.

37. See ISSCR Guidelines, supra note 1.

38. See Bonnie Steinbock, Moral Status, Moral Value, and Human Embryos: Implications for Stem Cell Research, in The Oxford Handbook of Bioethics 416 (Bonnie Steinbock ed., 2007). I largely agree with the idea that most rights come with “personhood.” As used here, personhood is not the same as merely having human DNA but rather requires some level of intelligence and self-awareness (in humans or in non-humans). In addition, some organisms that are not persons have interests or rights, such as the avoidance of pain, that we need to take into account. Peter Singer has famously (or infamously) said that a person must be a “rational and self-aware being so that not only are human embryos and fetuses not persons, but neither are newborn babies.” See Peter Singer, Practical Ethics 75 (3d ed. 2011), especially Chapter 4. He discusses his views on the moral status of human embryos and fetuses in Chapter 6. Some of the other classic philosophical statements of the “capabilities”-based approach to personhood and rights can be found in Joseph Fletcher, Humanness, in Joseph Fletcher, Humanhood: Essays in Biomedical Ethics 7–19 (1979), and Daniel Dennett, Conditions of Personhood, in The Identities of Persons 175–98 (A.O. Rorty ed., 1976). One can trace the concept at least to the 17th century and John Locke. For a review of various philosophical approaches to personhood, see John Janez Miklavcic & Paul Flaman, Personhood Status of the Human Zygote, Embryo, Fetus, 84 Linacre Q. 130 (2017).

39. I would make an exception for risky clinical research that was justified by the potential benefit to that newborn.

40. Maureen L. Condic has (often) argued rigorously, based on Aristotle and Aquinas, that a new human being is created at fertilization and that full moral status exists at, and can only come into being at, that point, with slight exceptions for identical twins and “two embryo” chimeras. See Maureen L. Condic, Untangling Twinning: What Science Tells Us about the Nature of Human Embryos (2020). I found the approach fascinating for its look into a very different style of argument than my own—but not, to me, close to convincing.

41. The Balanced Budget Downpayment Act I, Pub. L. No. 104-99, tit. I, § 128, 110 Stat. 34 (1995).

42. Transcript of Sen. Hatch Backs Expansion of Stem-Cell Funding, Nat’l Pub. Radio, (Feb. 16, 2007), https://www.npr.org/templates/story/story.php?storyId=7447914 (“I do not consider them babies until - they at least have to be planted in the womb to even have a chance to becoming a human being.”).

43. See Christopher J. Yuskaitis & Scott L. Pomeroy, Development of the Nervous System, 2 Fetal & Neonatal Physiology 1294, 1295 (Richard A. Polin et al. eds., 5th ed. 2017) (“Neurulation in humans occurs in two distinct phases: primary neurulation during weeks 3 and 4 of gestation leading to development of the brain and spinal cord (Figure 131-3), and secondary neurulation during weeks 5 and 6, with formation of the lower sacral and coccygeal cord.”).

44. See Ivana Bulatovic et al., Human Fetal Cardiac Progenitors: The Role of Stem Cells and Progenitors in the Fetal and Adult Heart, 31 Best Prac. & Res. Clinical Obstetrics & Gynaecology 58, 58 (2016) (“The human fetal heart is formed early during embryogenesis as a result of cell migrations, differentiation, and formative blood flow. It begins to beat around gestation day 22.”).

45. See Cheryl M. Tan & Adam J. Lewandowski, The Transitional Heart: From Early Embryonic and Fetal Development to Neonatal Life, 47 Fetal Diagnostics & Therapy 373 (2019).

46. See Kate Williams & Martin H. Johnson, Adapting the 14-day Rule for Embryo Research to Encompass Evolving Technologies, 10 Reproductive BioMedicine & Society Online 1 (2020).

47. Joan Stiles & Terry L. Jernigan, The Basics of Brain Development, 20 Neuropsychology Rev. 327 (2010).

48. See Fetal Development: Stages of Growth, Cleveland Clinic, https://my.clevelandclinic.org/health/articles/7247-fetal-development-stages-of-growth (last visited Jan. 3, 2022), (“Generally, it’s called an embryo from conception until the eighth week of development. After the eighth week, it’s called a fetus until it’s born.”).

49. Publications often state that the fetus (“your baby”) can begin to hear at 18 weeks of pregnancy, which is 16 weeks after fertilization. See, e.g., Fetal Development: The 2^nd Trimester, The Mayo Clinic (last visited Jan. 27, 2022), https://www.mayoclinic.org/healthy-lifestyle/pregnancy-week-by-week/in-depth/fetal-development/art-20046151. It is not clear that the scientific literature supports that. One 2015 study found possible responses to voice in fetuses in the second trimester (although none of the fetuses in the research were earlier than the 21^st week of pregnancy, and so the 19^th after fertilization). Studies found much clearer responses in the third trimester. See Viola Marx & Emese Nagy, Fetal Behavioural Responses to Maternal Voice and Touch, PLOS ONE, June 8, 2015, https://doi.org/10.1371/journal.pone.0129118. An earlier review article put hearing at 23 to 25 weeks of pregnancy, or 21 to 23 weeks after fertilization. Peter G. Hepper & B.S. Shahidullah, The Development of Fetal Hearing, 71 Archives of Disease in Childhood 81, 85 (1994), https://www.cambridge.org/core/services/aop-cambridge-core/content/view/49DFEBEDC7C279D5EC0F3FB035D51166/S0965539500001108a.pdf/the-development-of-fetal-hearing.pdf (23 to 25 weeks of pregnancy, which means 21 to 23 weeks after fertilization). Interestingly, the same Mayo Clinic source says it is the 25^th week of pregnancy (23^rd after fertilization) when "your baby might be able to respond to familiar sounds, such as your voice, with movement.

50. Brain M. Mercer, Periviable Birth and the Shifting Limit of Viability, 44 Clinics Perinatology 283 (2017); Bonnie H. Arzuaga & Ben H. Lee, Limits of Human Viability in the United States: A Medicolegal Review, 128 Pediatrics Persps. 1047 (2011).

51. The politics of state efforts to limit abortion made fetal ability to perceive pain especially important and contested. An excellent review of the question looking at a wide range of indicators, concluded that “fetal perception of pain is unlikely before the third trimester.” Susan J. Lee et al., Fetal Pain: A Systematic Multidisciplinary Review of the Evidence, 294 JAMA 947 (2005). For a less technical and more recent discussion. See also Sara G. Miller, Do Fetuses Feel Pain: What the Science Says, Live Sci. (May 17, 2016), https://www.livescience.com/54774-fetal-pain-anesthesia.html. The American College of Obstetricians and Gynecologists, which took an official position on the question, notes that the structures necessary to perceive pain do not exist before week 22 post-fertilization and that “the evidence shows that the neural circuitry necessary to distinguish touch from painful touch does not, in fact, develop until late in the third trimester.” Facts Are Important: Gestational Development and Capacity for Pain, https://www.acog.org/advocacy/facts-are-important/gestational-development-capacity-for-pain (last visited Aug. 30, 2022). But see John C. Bockmann, The ACOG Should Reconsider Fetal Pain, Charlotte Lozier Inst. (Aug. 3, 2021), https://lozierinstitute.org/the-acog-should-reconsider-fetal-pain/ (a recent publication by the pro-life Charlotte Lozier Institute, arguing that pain perception may be possible as early as 12 weeks).

52. I originally had an eleventh possible line, “substantial brain development,” but abandoned it when I couldn’t locate consistent sources providing such a timetable, or even agreeing on what constitutes “substantial” development.

53. Fiona MacDonald, Scientists Keep Human Embryos Alive in the Lab for the Longest Time Yet, Sci. Alert (May 5, 2016), https://www.sciencealert.com/scientists-have-smashed-the-record-for-growing-human-embryos-in-the-lab; Alessia Deglincerti et al., Self-Organization of the In Vitro Attached Human Embryo, 533 Nature 252 (2016); Marta N. Shahbazi et al., Self-organization of the Human Embryo in the Absence of Maternal Tissues, 18 Nature Cell Bio. 700 (2016).

54. See Yuyu Niu et al., Dissecting Primate Early Post-Implantation Development Using Long-Term In Vitro Embryo Culture, 366 Sci. 837 (2019); Huaixiao Ma et al., In Vitro Culture of Cynomolgus Monkey Embryos Beyond Early Gastrulation, 366 Sci. eaax7890 (2019), https://pubmed.ncbi.nlm.nih.gov/31672918/; Tao Tan et al., Chimeric Contribution of Human Extended Pluripotent Stem Cells to Monkey Embryos Ex Vivo, 184 Cell 2020 (2021). The normal macaque gestation period is about five and a half months, so if the relationship were to be linear, 20 days for a macaque would be equivalent to 32 days for a human embryo—if the relationship were linear, see id.

55. Sydney Page, A Newborn Weighed Less than a Pound and was Given a Zero Percent Chance of Survival. He Just Had His First Birthday, Wash. Post (June 23, 2021, 6:00 AM), https://www.washingtonpost.com/lifestyle/2021/06/23/premature-baby-survive-birthday-record/.

56. Whether that will be true at the time this is published remains deeply uncertain; personally, I believe it is likely the Supreme Court will abandon that tenet of Roe v. Wade sometime in late June 2022 when it (probably) will decide Mississippi v. Jackson Women’s Health Organization. See Henry T. Greely, Ex Vivo Embryos After Dobbs, J. L. & Biosci. (forthcoming).

57. 45 C.F.R. §46.202(c).

58. 45 C.F.R. §46.204(b).

59. 45 C.F.R. §46.202(e).

60. 45 C.F.R. §46.202(h).

61. 45 C.F.R. §46.205(c).

62. See, e.g., Larry Carbone, Pain in Laboratory Animals: The Ethical and Regulatory Imperatives, PLoS One 6(9) (Sept. 7, 2011).

63. See Lana Vasung, et al., Exploring Early Human Brain Development with Structural and Physiological Neuroimaging, Neuroimage 226 (2019), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6537870/.

64. See, e.g. Selena Simmons-Duffin & Carrie Fetal, The Texas Abortion Ban Hinges On ‘Fetal Heartbeat.’ Doctors Call That Misleading, NPR (May 3, 2022, 4:55 pm), https://www.npr.org/sections/health-shots/2021/09/02/1033727679/fetal-heartbeat-isnt-a-medical-term-but-its-still-used-in-laws-on-abortion

65. See id.

66. See id.

67. Fetal Development: Stages of Growth, The Cleveland Clinic, https://my.clevelandclinic.org/health/articles/7247-fetal-development-stages-of-growth (“By the end of the third month, the fetus is fully formed. All the organs and limbs (extremities) are present and will continue to develop in order to become functional.”).

68. See id.

69. Fetal Development: Stages of Growth, supra note 67.

70. See id.

71. Jon A. Shields, A Hard but Real Compromise Is Possible on Abortion, N.Y. Times (Oct. 19, 2021), https://www.nytimes.com/2021/10/19/opinion/abortion-pro-life-movement.html.

72. As noted above about post-viability ex vivo fetuses, I would make an exception to the second part of my sentence, and it is one found in the Dickey-Wicker Amendment itself, which incorporates, as to embryo research, 45 C.F.R. § 46.208, providing: “(a) No fetus in utero may be involved as a subject in any activity covered by this subpart unless: (1) The purpose of the activity is to meet the health needs of the particular fetus and the fetus will be placed at risk only to the minimum extent necessary to meet such needs, or (2) the risk to the fetus imposed by the research is minimal and the purpose of the activity is the development of important biomedical knowledge which cannot be obtained by other means.”

73. See Shields, supra note 71.

74. In this, I agree with the Common Rule’s treatment of research on non-viable neonates, among whom I would class fully anencephalic babies (children born with brains that entirely lack cerebrums).

75. See President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1981. Defining Death: Medical, Legal and Ethical Issues in the Determination of Death [On-line]. Available: https://repository.library.georgetown.edu/bitstream/handle/10822/559345/defining_death.pdf?sequence=1&isAllowed=y (last visited August 13, 2022).

76. Roni Caryn Rabin, In a First, Surgeons Attached a Pig Kidney to a Human, and It Worked, NY Times (Oct. 19, 2021), https://www.nytimes.com/2021/10/19/health/kidney-transplant-pig-human.html.

77. Leon R. Kass, The Wisdom of Repugnance, 216 New Republic 17, 19-20 (1997).

78. See, e.g., The President’s Council on Bioethics, Hum. Dignity & Bioethics: Essays Commissioned by the President’s Council on Bioethics (2008).

79. James Boswell, The Life of Samuel Johnson, LL.D.: Comprehending an Acct. of His Stud. & Numerous Works, Vol. 2, 478 (1791).

80. See, e.g., Matt Dias, Dignity After Death and Protecting the Sanctity of Human Remains, Voices in Bioethics (2015); Charles Foster, Dignity and the Use of Body Parts, 40 J. Med. Ethics 44 (2014); Joseph Meaney, Bioethics and the Dignity of the Human. Body, The Nat’l Cath. Bioethics Ctr. (Jan. 15, 2021), https://www.ncbcenter.org/messages-from-presidents/dignityofthehumanbody-pj5jh.

81. This requirement is found in the Nuremburg Code, the Declaration of Helsinki, and the Common Rule. “Permissible Medical Experiments,” Trials of War Criminals before the Nuremberg Military Tribunals under Control Counsel Law No. 10, Nuremberg, Oct.1946-Apr.1949. Washington, D.C.: U.S. G.P.O., 1049-1953 (“The degree of risk to be taken should never exceed that determined by the humanitarian importance of the problem to be solved by the experiment.”); World Medical Association, Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects, 79 Bulletin of the World Health Organization 373 (2001) (“Medical research involving human subjects should only be conducted if the importance of the objective outweighs the inherent risks and burdens to the subject. This is especially important when the human subjects are healthy volunteers.”); see also Criteria for IRB Approval of Research, 6 C.F.R. § 46.111(a)(2) (2017) (“Risks to subjects are reasonable in relation to anticipated benefits, if any, to subjects, and the importance of the knowledge that may reasonably be expected to result. In evaluating risks and benefits, the IRB should consider only those risks and benefits that may result from the research as distinguished from risks and benefits of therapies subjects would receive even if not participating in the research.”). The IRB should not consider possible long-range effects of applying knowledge gained in the research (e.g., the possible effects of the research on public policy as among those research risks that fall within the purview of its responsibility."), id.

82. Henry T. Greely, The Death of Roe and the Future of ex vivo Embryos, 9 J. Law and the Biosciences 1 (2022).

83. See id.

84. See id.

85. See Jinglei Zhai et al., Human Embryonic Development: From Peri-implantation to Gastrulation, 32 Cell 18 (2022).

86. See id. at 19.

87. See id. at 21.

88. Id.

89. See Alejandro De Los Angeles, Monkey Embryos Cultured to 20 Days, 29 Stem Cells Dev. 807 (2020).

90. 42 U.S.C. §289(g) (added in the National Institutes of Health Revitalization Act in 1993). Administrations can impose more onerous restrictions than those in the statutes, as the Trump Administration did in 2019 (only for those restrictions to be withdrawn by the Biden Administration in 2021), id. For the current status, see HHS Removes an Administrative Requirement for Human Fetal Tissue Research Proposals, NIAID Funding News (May 5, 2021), https://www.niaid.nih.gov/grants-contracts/hft-research-policy-change.

91. Zhai, supra note 85, at 19.

92. As far as I have discovered, no one has made quite my argument before. Two come closest. In 2021 Sophia McCully took a similar position, arguing for a 28 day limit. She notes that after 28 days aborted tissue is available, and she points to some disadvantages of using non-human primates. But she does not demand prior evidence that such research might be useful or urge a general regulatory scheme beyond referring to the United Kingdom’s Human Fertilisation and Embryology Authority. See Sophia McCully, The Time Has Come To Extend the 14-Day Limit, 47 J. Med. Ethics e66 (Epub ahead of print Feb. 2, 2021). In March 2021, Insoo Hyun and others argued in Science for a position similar to the ultimate ISSCR Guidelines but in terms similar to my proposal. The important difference from my proposal is that, like the ISSCR Guidelines, they do not set or suggest any specific time of in the embryo’s development when such research would have to stop, instead saying “Extended embryo culture should commence in small steps, with frequent interim evaluations and reassessments by regulators…Any research teams proposing to move beyond the 14-day mark would need to seek regulatory permission for each subsequent experiment.” Insoo Hyun et al., Human Embryo Research Beyond the Primitive Streak, 371 Science 998 (Mar. 5, 2021). John B. Appleby and Annelien Bredenoord call for an extension to 28 days based on the potential scientific value of research on that period but offer no reasons for stopping at 28 days. See John B. Appleby & Annelien Bredenoord, Should the 14-day Rule for Embryo Research Become the 28-day Rule?, 10 EMBO Mol Med e9437 (2018). They argue for a 28 day rule largely because of the scientific value of additional period. Kate Williams and Martin Johnson in 2020 argued for extending the research period but suggested that the “limit should be set as early as any relevant marker is found. . . . [W]e propose that a readily visible marker of development should be used instead, such as is exemplified by Carnegie stage 12 at 29–31 dpf, namely the acquisition of an upper limb of length 0.26–0.34 mm, 25–28 pairs of somites or maximum embryo length of 5.1 m….” Kate Williams & Martin H. Johnson, Adapting the 14-day Rule for Embryo Research to Encompass Evolving Technologies, 10 Reproductive BioMedicine & Society Online 1 (2020). With the possible exception of the Hyun et al., piece, I had not read any of them before drafting this article and none of them influenced it.

93. I learned this in a conversation with stem cell scientist Dr. Amander T. Clark of U.C.L.A., a conversation that inspired this idea and, ultimately, this article.

94. See Zhai, supra note 85, at 19.

95. See id.

96. See id.

97. See Ronan O’Rahilly & Rabiola Müller, Developmental Stages in Human Embryos: Revised and New Measurements, 192 Cells Tissues Organs 73 (2010).

98. See id.

99. See id.

100. Gavin E. Jarvis, Early Embryo Mortality in Natural Human Reproduction: What the Data Say, PubMed Central (Jun. 7, 2017), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5443340/ (“We see then a web of circumstantial evidence implying that there is a substantial amount of fetal wastage in humans. This insight arises from imperfect types of knowledge (as documented by Dr. Jarvis) but nonetheless, there is a signal consistent with the claim that approximately half or more of conceptions fail. More needs to be done to improve our understanding.”)

101. See id.

102. See, e.g., The Cook-Andersen Lab, https://cookandersenlaboratory.com/publications-and-presentations/ (last visited Aug. 13, 2022).

103. See id.

104. See id.

105. Cook-Andersen presented this research at the 2021 annual meeting of the American Society for Reproductive Medicine, at which I also spoke. Heidi Cook-Andersen, Uncovering the Molecular Determinants of Successful Implantation in the Human Blastocyst at American Society for Reproductive Medicine Scientific Congress and Expo, ASRM (Oct. 19, 2021), https://asrmcongress.org/asrm-2021-speaker-spotlight-heidi-cook-andersen-m-d-ph-d/. A paper describing that work is, as I am writing this, under submission. Email communication from Heidi Cook-Andersen to Henry T. Greely, Jan. 7, 2022. This 2019 doctoral dissertation from one of her students gives some indication of this research path: Jennifer Nicole Dumdie, Molecular Determinants of Oocyte and Embryo Developmental Competence (2019) (on file with U.C. San Diego), available at https://escholarship.org/uc/item/0kq611t3. The website for Cook-Andersen’s laboratory provides additional information. Cook-Andersen Lab, https://cookandersenlaboratory.com (last visited Jan. 4, 2022).

106. See id.

107. See id.

108. See id.

109. See id.

110. See id.

111. See J. Boullin & J.M. Morgan, The Development of Cardiac Rhythm, 91 Heart 874-75 (2005), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1768983/.

112. See Zhai, supra note 85, at 19.

113. See Zhai, supra note 85, at 19.

114. Fetal Growth Calculator, World Health Organization, https://srhr.org/fetalgrowthcalculator/#/ (last visited Aug. 13, 2022).

115. See Ronan O’Rahilly & Rabiola Müller, Developmental Stages in Human Embryos: Revised and New Measurements, 192 Cells Tissues Organs 73 (2010).

116. See id.

117. See id.

118. See id.

119. See id.

120. For embryoids, see Jianping Fu et al., Stem-cell-based Embryo Models for Fundamental Research and Translation, 20 Nature Materials 132 (2021). For blastoids, see Heiner Niemann & Bob Seamark, Blastoids: A New Model for Human Blastocyst Development, 6 Signal Transduction and Targeted Therapy 239 (2021). For gastruloids, see Susanne C. van den Brink et al., Single-Cell and Spatial Transcriptomics Reveal Somitogenesis in Gastruloids, 582 Nature 405 (2020) (“Gastruloids are three-dimensional aggregates of embryonic stem cells that display key features of mammalian development after implantation, including germ-layer specification and axial organization.”). For simbryos, see Julian Hitchcock, The Entelechy Test: Embryos and Simbryos, BioNews (Mar. 29, 2021), https://www.bionews.org.uk/page_155584. For SHEEFs, see John Aach et al., Addressing the Ethical Issues Raised by Synthetic Human Entities with Embryo-Like Features, 6 eLife e20674 (2017). SHELEs was used by George Church at a meeting I attended. George Church & John Aach, Stem Cells, Engineered Tissues, and Synthetic Embryo-Like Entities at The Petrie-Flom Center for Health Law, Policy, Biotechnology, and Bioethics at Harvard Law School Event: The Ethics of Early Embryo Research & the Future of the 14 Day Rule (Nov. 7, 2016), slides available at https://www.slideshare.net/petrieflom/george-church-and-john-aach-stem-cells-engineered-tissues-and-synthetic-embryolike-entities. One of the people in the audience was the renowned science and technology studies scholar, Sheila Jasanoff, who objected (I’m not sure how seriously) to this use of her name; the Church lab later changed the term to SHEEFs.

121. See id.; see also ISSCR Guidelines, supra note 1.

122. Id.

123. Id.

124. Shadi Taraza, et al, Post-Gastrulation Synthetic Embryos Generated Ex Utero from Mouse Naïve ESCs, Cell, 185, 3290-3306 (2022); see also Amadei, G., Handford, C.E., Qiu, C. et al. Synthetic Embryos Complete Gastrulation to Neurulation and Organogenesis. Nature (2022).

125. See Zhai, supra note 85, at 18, 25.

126. See id.

127. See, e.g., National Academies of Sciences, Engineering, and Medicine Committee on Ethical, Legal and Regulatory Issues Associated with Neural Chimeras and Organoids, The Emerging Field of Human Neural Organoids, Transplants, and Chimeras (2021); Julien G. Roth et al., Advancing Models of Brain Development with Biomaterials, 22 Nature Rev. Neurosci 593 (2021); Henry T. Greely, Organoids, Chimeras, Ex Vivo Brains – Oh My!, Neuroethics Blog (Sept. 4, 2018), http://www.theneuroethicsblog.com/2018/09/organoids-chimeras-ex-vivo-brains-oh-my.html.

128. See id.

129. For an excellent survey and summary of national laws on research with embryos (and embryo models, which become important at the end of this paper), see Kirstin RW Matthews & Daniel Morali, National Human Embryo and Embryoid Research Policies: A Survey of 22 Top Research-intensive Countries, 15 Regenerative Med. 1905 (2020).

130. See id.

131. Comm. on Guidelines for Human Embryonic Stem Cell Rsch., Guidelines for Human Embryonic Stem Cell Research (2005).

132. ISSCR Guidelines, supra note 1, at 65.

133. Previous restrictions on preimplantation human embryo culture (the “14-day/primitive streak rule”) were not written to apply to integrated stem cell-based embryo models." ISSCR Guidelines, supra note 1, at 65.

134. See Katz v. United States, 389 U.S. 347 (1967).

135. ISSCR Guidelines, supra note 1, at 9.

136. ISSCR Guidelines, supra note 1, at 6.

137. ISSCR Guidelines, supra note 1, at 10.

138. ISSCR Guidelines, supra note 1, at 65.

139. Bruce M. Carlson, Placenta and Extraembryonic Membranes, Human Embryology and Developmental Biology 117 (Bruce M. Carlson ed., 5^th ed., 2014).

140. ISSCR Guidelines, supra note 1, at 65.

141. Id.

142. Comm. on Guidelines for Human Embryonic Stem Cell Rsch., Guidelines for Human Embryonic Stem Cell Research (2005).