A reply to Lee Silver.
[This article was co-authored by Patrick Lee and Robert P. George.]
We thank Professor Silver for his response . On the basis of this exchange, we are confident that readers will be able to judge whether Professor Silver’s claim that human embryos are something other than what we say they are, namely, human beings in the earliest stage of their natural development, can withstand scrutiny in light of the findings of modern embryological science.
Are Human Embryos Human Beings?
Silver claims that while many people believe that human embryos are human beings, what sets us apart is that we contend that this is a matter of scientific fact, rather than a proposition of faith. But we are scarcely set apart by this claim. On the contrary, our claim that the human being that is, say, you the reader, is the same determinate and enduring being — the same individual member of the species Homo sapiens — who at an earlier stage of his or her life was an adolescent, a child, an infant, a fetus, and an embryo, is a fact confirmed by contemporary embryology and attested to by the standard works in the field. (We mentioned several of these works in the essay to which Silver responds; he cites none in support of his position.) Terms such as human “adolescent,” “infant,” and “embryo” do not refer to beings of distinct kinds; rather, they refer to the same kind of being — a human being, an individual member of the species Homo sapiens — at different developmental stages. You were once an embryo, just as surely as you were once an adolescent, and before that a child. So that readers will be left in no doubt, let us quote a couple of the standard texts. Keith Moore and T.V.N. Persaud, in The Developing Human: Clinically Oriented Embryology, perhaps the most widely used of the embryology texts, make the following unambiguous statement about the beginning of a new and distinct human individual: “Human development begins at fertilization when a male gamete or sperm (spermatozoon) unites with a female gamete or oocyte (ovum) to form a single cell — a zygote. This highly specialized, totipotent cell marked the beginning of each of us as a unique individual” (p.16, emphasis supplied). Ronan O’Rahilly and Fabiola Mueller, in their book Human Embryology and Teratology, say this: “Although life is a continuous process, fertilization (which, incidentally, is not a ‘moment’) is a critical landmark because, under ordinary circumstances, a new, genetically distinct human organism is formed when the chromosomes of the male and female pronuclei blend in the oocyte” (p. 8, emphasis supplied).
An even weaker point advanced by Silver is his complaint that we “have not budged” in our rejection of Ronald Bailey’s claim that embryos are analogous to skin cells or other somatic cells that can be used (it is assumed) in human cloning. People who recall our exchanges with Bailey on NRO know that our rejection of his proposed analogy has nothing to do with stubbornness. Rather, we showed by careful argumentation and the presentation of the relevant biological facts that somatic cells are analogous not to embryos — which are complete self-integrating organisms that are, as the standard scientific texts attest, new individuals of their species — but rather to gametes (sperm and egg) whose union can bring into existence a new and distinct individual in the embryonic stage of its development.
Parts and Wholes
Let us turn now to two key propositions in our argument that human embryos are human beings. Silver disputes these propositions. We said that human embryos are (embryonic) human beings because they are complete human organisms that “have an internal active disposition toward the mature stage of a human being.” Silver disputes this proposition. He claims that “we know that a single cell can separate from a 4- cell embryo and develop into a separate human baby (on rare natural occasions).” According to him, “Lee and George would argue that this cell has the ‘internal active disposition’ that makes it a human being.” But this would be inconsistent with our position that ES cells are not human beings because they lack such an active disposition to develop to maturity.
Silver here commits the error of confusing a part with a whole. When the cell is part of the whole multi-cellular organism, it lacks any disposition to develop on its own: for, if it is not split off, then it will continue to act in coordination with other cells along a trajectory toward the mature stage of development of the organism of which it is a part. If the splitting of the cell from the whole embryo results in a new embryo, this is a form of asexual reproduction, and an example of what evidently occurs in normal monozygotic twinning. So the fact that a single cell could be separated from a four-celled embryo and produce a second embryo is completely consistent with our position that embryonic-stem (ES) cells are not human beings because they lack this internal active disposition.
Silver then claims that “an internal active disposition” “is not a term that has any meaning in the context of cellular or molecular biology.” But this is mere bluster. The concept of a disposition or tendency is scarcely foreign to scientific contexts, where one speaks, for example, of the solubility of salt (sodium chloride) or of sugar (sucrose) and their variance in relation to temperature; for, solubility is a disposition, a tendency to do something in a given context. We added “internal” and “active” to distinguish it clearly from the (merely passive) capacity of something to be changed by being acted upon by an extrinsic agent. Thus, when biology books refer to the capacity for metabolism as being one of the marks of a living system, “capacity” refers to the same sort of thing we were referring to by the term “internal active disposition.” We could also have used the term “developmental trajectory” or “developmental program” to make the same point. An organism with a single developmental trajectory toward the mature stage of a human being is already a human being.
Moreover, dispositions or developmental trajectories in organisms have structural bases; that is, present in the organism are constituents and arrangements (organization) responsible for the active disposition or developmental trajectory. So, when Silver next claims, “If George and Lee want to claim otherwise [namely, that “internal active disposition” does have a meaning in the context of cellular and molecular biology] I’d like to know their perception of the molecular attributes that distinguish human being-cells from non-human-being, yet still fully viable and human cells” — we are happy to inform him. At the one-cell stage, comparing a zygote to, say, a stem cell, the one-celled embryo differs epigenetically from the stem cell (that is, while each has the complete genome there are differences in the activation or silencing of various genes).
But that is not all, for there are also differences with respect to several cytoplasmic factors. The embryo is such that, unlike a stem cell or any other somatic cell, its whole developmental program, both the epigenetic state of its DNA and various proteins and RNA factors in its cytoplasm (differences that are fully discernable at the molecular level) enable it to develop along a trajectory towards developmental maturity. At later stages of development, each of the cells of the embryo will differ from a separated stem cell or from other somatic cells — among other ways — in being modified by interaction with other cells within the embryo in such a manner as to contribute to homeostasis and the development of the organism as a whole. Therefore, “in the context of cellular and molecular biology,” a single-cell human being (i.e. a zygote) is distinct from any other viable human cell in terms of 1) its epigenetic state, 2) its pattern of gene activation, 3) its molecular composition, and 4) its subsequent pattern of development — all of which can be (and indeed are) easily determined by scientists without recourse to crystal balls, sacred texts, or articles of faith.
Are Embryonic Stem Cells Human Beings?
In an astonishingly misleading paragraph, Silver quotes from a quick summary given by a reporter from Science magazine about a meeting of scientists working on gene knockouts (a procedure in which, by a complicated process, a gene is deactivated — “knocked out” — in a mouse in order to discover what that gene’s normal function would be). Suspecting something fishy, we took the trouble to look. The scientists were examining how they could cooperate more effectively on an international level. In the past they had shared research data and materials in part by shipping mice that had been engineered with a certain gene knockout. Silver quotes the reporter summarizing their discussion as follows: “the participants agreed that it would be most economical to avoid trafficking in live mice and instead decided to maintain the knockouts as embryonic stem (ES) cells: clumps of tissue that can be frozen down and later grown up into full-fledged mice.” Professor Silver takes this as evidence that “[t]hose…who have worked only with early-stage mouse embryos” are more honest than scientists who work on human stem cells when speaking “about the relationship of embryos to stem cells.” In other words, Silver presents this quote as proof from authority that stem cells really are the moral equivalent of embryos (because they can somehow “by themselves” grow into embryos).
However, first, it is clear that the reporter is merely trying to summarize quickly the conclusions of a long meeting of stem-cell scientists; he is not attempting to select his words with exactitude, and much less is he addressing the issue in dispute between Professor Silver and ourselves. Silver is making much out of what is manifestly nothing more than a reporter’s shorthand manner of speaking. The article from Science is available online. We urge readers who are evaluating Silver’s arguments against our own to examine this article for themselves.
Moreover — and much more significantly — Silver fails to mention that the knockout procedure involves first knocking out a specific gene in mouse stem cells, then injecting those stem cells into an early mouse embryo, where the manipulated stem cells will be randomly incorporated into the body of the developing embryo. In some of the chimeric mice formed in this manner, the stem cells will contribute to the gonads, and therefore the germ cells produced by the mature animal will also lack the gene that was eliminated from the stem cells. In this case, by “back crossing” such “germ line deficient” mice (i.e., inter-breeding the male and female offspring of the mouse carrying the gene deletion), ultimately a mouse is produced that lacks the knocked out gene in all the cells of its body, and the phenotype that results from the loss of that gene’s action can be determined. The point the scientists were making was simply that it would be more economical and efficient to ship to other sites the manipulated stem cells rather than the mice with the gene knockouts, largely because many such knock-out mice are quite unhealthy, and therefore they are difficult to breed and expensive to maintain. When the scientists described the ES cells (or, more precisely, when the reporter recounted them as describing the ES cells as) “clumps of tissue that can be frozen down and later grown up into full-fledged mice,” it was understood — as part of the knockout procedure — that this later “growing up” process would involve injecting those ES cells into mouse embryos. Silver’s selective editing makes it appear that these scientists are asserting something they did not assert, and are speaking to an issue they did not address.
Professor Silver had argued in his book that aggregating mouse stem cells with mouse tetraploid embryos allowed the stem cells to develop “by themselves” into mature mice. We pointed out that the aggregation of the stem cells with tetraploid embryos (or embryo-like entities) does more than release an inner capacity in the stem cells. In fact, it generates a new and distinct organism, just as combining the nucleus of a somatic cell with an enucleated ovum (as in cloning) generates a new organism — in both cases, an embryonic member of the mouse (or other) species in question. In his reply, Silver now says the following:
Furthermore, as I wrote in my book, human ES cells have already been differentiated into placenta. This means that, in theory, the requirement for a second source of cells to reconstitute an embryo may be nullified, and ES cells — all by their lonesome — could develop into a fetus and human baby. This discussion was conveniently left out of Lee and George’s review.
This is supposed to show that stem cells really are equivalent to embryos because they can “by themselves,” or “all by their lonesome,” develop into mature members of their species. However, this argument is dubious at best, given that there is no evidence that an embryo could be generated by recombining ES-cells that have been “differentiated into placenta” with an undifferentiated ES cell, and there is sound scientific evidence to suggest embryos cannot be generated in this manner. In fact, the second article Silver cites in his book on this subject is much more cautious in its conclusions than Silver himself is. The authors write;
The human equivalent to trophoblast stem cells has not yet been derived, and it is likely that different growth factors will be required for their propagation.[note omitted] Although, in our current studies, BMP4 efficiently induced differentiation of human ES cells to trophoblast, these trophoblast cells propagated poorly, even in the continued presence of bFGF and fibroblast feeder layers (data not shown), suggesting that additional growth factors are required for their long-term proliferation. (Ren-He Xu, et al., “BMP4 initiates human embryonic stem cell differentiation to trophoblast,” Nature Bioetechnology 20 (December 2002), 1261-1264, at 1263)
Moreover, as partly recounted in Silver’s book, and more fully in the scholarly articles to which Silver refers, the stem cells (in the experiment Silver cites) by no means develop into embryos and fetuses without extrinsic causes in the form of external manipulations. No one denies or has denied that one can manipulate factors found in early stem cells and combined those factors with stem cells in such a way as to generate a new embryo. But just as combining factors from an ovum and a somatic cell to generate a new individual member of a species in the embryonic stage does not show that an ovum or a somatic cell used in the process was somehow already an embryo or its equivalent, the combining and manipulating of factors found in early stem cells does not show that a stem cell is an embryo or its equivalent. The fact that one can combine A (a stem cell) with B (another cell that is produced by a significant manipulation of a stem cell) to produce C, does not in the least prove that A or B were identical with C all along.
A Thing Either Is or Is Not a Human Being
The second proposition of ours that Silver contests, is that a thing either is or is not a human being, that is, that each human being came to be at once, not gradually. Readers will recall that our view is entirely consistent with the biological fact that fertilization is a process. Our point is that a new human being does not “partially” exist as the process unfolds. Rather, a new human being is what the process, if it reaches completion, brings into existence. (If fertilization fails, it is not correct to say that a “partial” human was brought into being, or that a human was “partially” brought into being.) Silver notes that we provided several arguments defending our position, but mentions only one, namely, in his words, “that it is simply commonsensical to most people.” He then notes that common sense is often mistaken, as “many examples from science demonstrate.”
Here again, however, what Silver says is highly misleading. First, in Silver’s article in Science and Theology News, he did not just claim that we presuppose the proposition that a being either is a human being or is not; rather, he claimed that it was a “hidden theological premise.” He even went so far as to allege — preposterously — that we adopted this premise from a literal interpretation of Genesis, according to which humans are made in God’s image, and God is absolute, so just as there cannot be a partial God, so there cannot be a partial human. His charge was that we were covertly injecting our “fundamentalist” Christian beliefs into secular-sounding arguments.
Of course, we pointed out that the inference we supposedly made starting from a teaching in Genesis was patently fallacious (i.e., God cannot be partial, humans are made in God’s image, therefore there cannot be partial humans — by the same reasoning one could infer that we must be omniscient, eternal, perfect, and so on). We then pointed out that the idea that human beings come to be at once was part of common sense; the point was to show (along with other arguments) that it was not a premise we had to take from the Bible or Church teaching: Since it is a commonsense idea, it is absurd to suppose that all those who make use of it are covertly smuggling in a literal reading of Genesis. Further, we made a point of stating expressly that just because this idea is part of common sense “doesn’t by itself prove it, but does provide support for it.”
After having given four reasons showing that no one has to appeal, openly or covertly, to revelation in order to hold this position (that something either is or is not a human being), we then moved on to a different task, namely, providing philosophical support for this position. We presented two distinct philosophical arguments. Silver offers no reply at all to either of these arguments. His only attack, and that a weak one, is on the single argument that we expressly said “doesn’t by itself prove” the position we were defending, but merely “provides support for it.”
In his book and article, Silver claims that the assumption that a thing either is or is not some kind of thing must be a hidden theological premise. But he now admits that it might have been adopted from common sense. Moreover, he now even admits that “[i]t is not surprising that so many philosophers have also held this position,” though he adds that, “[t]wo millennia ago, in the absence of modern scientific knowledge and biomedical technology, Aristotle would certainly have convinced me of its veracity.” Thus, Silver has implicitly abandoned the claim that this view depends on covert “fundamentalist” theological premises. But Silver’s attempt to discredit this view by tying it to a philosophical approach that has allegedly been falsified by modern science is equally misleading. First, he ignores the fact that we cited (to show, earlier, that this position was not a hidden theological premise) contemporary as well as ancient philosophers. (We specifically mentioned David Wiggins, Roderick Chisholm, Peter Van Inwagen, and E.J. Lowe, all recent or contemporary analytic philosophers.) Moreover — even if one disagrees with them — no one is likely to accuse any of them of naively following an outmoded, two millennia-old philosophical approach. Further, neither of the two philosophical arguments we presented to defend the proposition at the heart of the debate — arguments ignored by Silver — were lifted from ancient sources.
Daunted by Darwin?
Silver does present an argument against our position that a human being either is or is not — i.e., that there is no entity that is somehow between a human being and a non-human creature. Here is his argument:
The most serious challenge, of course, came from Darwin, whose theory of natural selection suggests that in the evolution of pre-human apes into human beings, there was no first human being. Instead, there appears to have been a continuum of evolutionary forms in a process during which no child was significantly different from its parents. The scientific implication is that some “things” might be in-between non-human and human.
But this argument is fallacious. Darwin proposes at least a partial explanation of the emergence of new species (whether it is a sufficient or complete explanation is another question, which we need not settle here). The idea is that some species evolved from other species, and that the human species evolved from lower animal species, at least with respect to the organic aspect of the human being (whether that is all there is to human nature is another question, which, again, we need not settle in the context of the current debate). But this does not show that there must have been some creatures which were in between humans and non-humans — creatures that were neither human nor non-human. It only shows that certain non-human animals evolved gradually (or perhaps not so gradually — the point is disputed among evolutionary biologists) to become more and more like humans until, at least partly through genetic mutations, the human species (as a distinct and identifiable species) emerged. Nothing in the Darwinian account suggests that there is or was a species that was neither human nor non-human — a species that was “in between” human and non-human.
In other words, if species B evolved from species A though an accumulation of minute genetic changes, that does nothing to show that the last entity just before the last genetic change which (wholly or in part) produced the new species B, must have been something that was partially B. Rather, that last entity prior to B was a non-B that was in a great many respects like a B.
Moreover, the addition of a minute genetic change could be — whether wholly or in part — responsible for a radical change in the properties possessed by a new being. It is entirely possible that an accumulation of minute changes to a system could together give rise to entirely new, emergent properties. There are numerous examples in the field of biology where minute changes in a gene (even the alteration of a single nucleotide) radically alter what the gene does, either completely eliminating its function (null mutations) or producing an entirely new function for the same DNA sequence (neomorphic mutations). In such cases, well-understood molecular events (i.e., DNA mutations) can give rise to emergent properties at the level of whole organisms. Small changes in the timing of key developmental events or in the structural organization of the brain could also produce radical differences in morphology or neurological function between closely related species. In other words, Darwin’s theory asserts that humans are descended from non-human animal species; Darwin’s theory does not say that those non-humans were or could in some way have been “partial” humans.
Embryos and Disorganized Growths
Next, Silver presents what he calls “another scientific challenge for Lee and George”: “If a perfectly normal human embryo is grown in a laboratory incubator, it can develop without any discontinuity into a teratoma.” We (Lee and George) would have to hold, he says, that during its first week in a petri dish, it is a human being. “But no one believes that an embryo-generated teratoma is a human being, even though it can be kept alive as a unique, integrated organism.” And this presents us with the problem: “So when during the continuous life of this organism did the human being suddenly disappear?”
Here Silver’s claims are deeply confused, as standard works of teratology, like the one we cited by O’Rahilly and Mueller, make clear. Even granting — for the sake of argument — that Silver has described the phenomenon accurately, there simply are no valid grounds for regarding — as Silver does — a teratoma as a unitary functioning organism, as opposed to a disorganized growth. Teratomas (unlike embryos) are not individuals of the species in the earliest stage of their development towards organismal maturity. Rather, teratomas are masses of cells, often containing some degree of sub-organismal order or organization (for example, hair, teeth, limb primordia), but lacking organismal wholeness, or indeed any unitary order at the level of a whole organism.
Embryos do not, however, by normal development, turn into teratomas. A “perfectly normal human embryo” cannot “turn into” a teratoma without an intervening event — the death of the embryo — that causes a radical discontinuity between the two. An embryo that dies, leaving physical remains in the form of a teratoma, is a distinct case from an embryo that develops to the fetal stage. Embryos literally turn into fetuses, just as adolescents turn into adults. There is no discontinuity; merely normal development. That is why we cannot say of an embryo and a teratoma what we can say of an embryo, fetus, infant, child, adolescent, and adult, namely, that they are the same entity or being at different developmental stages.
Silver’s argument is an effort to make something of the fact that an event may occur that irreversibly destroys an embryo’s capacity for self-organization and internally-driven growth in the direction of developmental maturity, leaving only a disorganized mass — a teratoma. If this happens, a tumor-like structure lacking the organization and capacity for self-directed development towards maturity as a member of the species is the physical remains of what was, before its death, a complete organism — a living member of the species in the embryonic stage. The teratoma is not continuous with the embryo, since a radical discontinuity — embryonic death — has occurred.
To simply return Silver’s challenge in a somewhat more familiar context; when a patient in a hospital dies, would Silver assert that “a perfectly normal human” develops “without any discontinuity” into a corpse? If a meaningful distinction between a patient and a corpse cannot be drawn (as Silver’s analysis seems to suggest) because the transition between the two states is “continuous,” are we required to view living and dead human bodies as the same, simply because the former seamlessly gives rise to the latter? This is, of course, absurd. Medical science routinely addresses the question of when a patient passes from life to death and medical research into this question continues to define the molecular and cellular correlates of death with greater and greater accuracy. The question of “when during the continuous life of this organism did the human being cease to exist?” can (and is) readily addressed by medical research, and does not present a significant intellectual or scientific challenge to our position.
Spontaneous Embryo Loss — What is Lost?
At the end of Silver’s reply there is a serious mischaracterization or misunderstanding of our position that must be corrected. Silver, and others, had argued that human embryos are not worthy of the full moral respect due to a human being because a high percentage of embryos formed in natural pregnancies fail to implant or spontaneously abort. To this we have at times presented, as a preliminary note to a detailed argument, the following point: “It is worth noting first, as the standard embryology texts point out, that many of these unsuccessful pregnancies are really due to incomplete fertilizations. So in many cases, what is lost is not actually a human embryo” (from Robert George’s Personal Statement in the Report of the President’s Council on Bioethics, on Human Cloning and Human Dignity, p.305).
In both his book and his reply on NRO, Silver misunderstands this to mean that we hold that all defects in fertilization — that is, malfunctions of various sorts in the fertilization process — result in entities that lack the active disposition to, or the developmental trajectory toward, the mature stage of a human being, and so are not actually human embryos. But after making the above point in his personal statement in the Report of the President’s Council on Bioethics on cloning, George went on to say:
“To be a complete human organism (a human being), the entity must have the epigenetic primordia for a functioning brain and nervous system, though a chromosomal defect might only prevent development to maximum functioning (in which case it would be a human being, though handicapped) (emphasis added). If fertilization is not complete, then what is developing is not an organism with the active capacity to develop itself to the mature (even if handicapped) state of a human.”
So our point has always been only that in some instances of “defective fertilizations” what results will lack the program for or developmental trajectory toward the mature stage of a human being, and so is not a human embryo. An embryonic human may certainly have the basic program or disposition orienting it toward the mature stage, but also have defects that prevent the full actualization of that program, as is the case with trisomy 21 and anencephalic infants.
Moreover, in his book Silver presents this merely preliminary note as if it were our main reply to the objection about so-called “embryo-wastage.” Our main replies to that argument, however, have always been that, first, the objection is simply a non sequitur, in fact a form of the naturalistic fallacy — that because something does occur in “nature” with predictable frequency then it must be morally acceptable when deliberately caused by human agency. Second, we have observed that historically, and in some places even today, the infant mortality rate has been very high, and so if this objection were sound, it would show that infanticide in such cultures would be morally acceptable, which (we hope Professor Silver would agree) is absurd.
Put us to the Test
Finally, at both the beginning and the end of his article (and in his book) Professor Silver claims that we have presented a position — that a human embryo is a human being — that is unscientific because it is unfalsifiable (that we cannot conceive of an experiment that would falsify it). We had rebutted objections that monozygotic twinning or formation of chimeras proves that a human individual does not begin at fertilization by pointing out that the conclusion simply does not follow: If monozygotic twinning occurs, evidently a new individual comes to be with the splitting of the original embryo; or, if two embryos combine to form a chimera, evidently either one absorbs the cells of the other or the two cease to be in the formation of a new multi-cellular organism. In his book, Silver labels our reply a “rationalization” and triumphantly claims, “Indeed, they have never described a test that could falsify the theory, which is why this is a matter of faith, not science.”
But again Professor Silver has gone off the rails. The test of whether a group of cells constitutes a single organism is empirically verifiable or falsifiable: that test is whether the cells form a stable body and function as parts of a whole, self-developing, adaptive unit. Again, one need only consult the standard texts. There is no scientific debate about this. It is certainly not a theological dogma or a merely scientific-sounding concept we have concocted. On the contrary, it is the test that has been in general use for years for determining whether an individual (at any developmental stage) remains alive or has died. In the life cycle of embryos that split or combine, there is, indisputably, a whole set of complex, coordinated activities performed by each of these embryos as a whole from day 1 to the day on which twinning or combination occurs (which may be as late as day 9, or, very rarely, even slightly later). This is empirical verification that a developing, single organism exists during that time. So there is a test for verification of the presence or absence of a unitary organism, and we have in our writings applied that test in a straightforward and uncontroversial way to early embryos. Where monozygotic twinning or combination occurs, the embryos prior to those events have clearly passed that test (see Helen Pearson, “Your Destiny, From Day One,” Nature 418 (2002), pp. 14-15). On the other hand, the test also can be failed, and in fact sometime is: In some cases where one thinks there might be a human being at the embryonic stage of development, it turns out that what is present is a teratoma or a complete hydatidiform mole, as evidenced by the lack of self-organization and internally directed growth towards organismal maturity. Thus, the test of whether a unitary organism is present is verifiable or falsifiable. Professor Silver simply ignores the evidence indicating its verification in the coming to be of human embryos at fertilization or monozygotic twinning.