The Precautionary Principle
Finally, I want to address one more argument in favour of the conception threshold. Let us assume again for the moment that the punctualist thesis is true. One question I have asked is whether it follows necessarily from that thesis, or from metaphysical dualism, that the critical moment when personhood begins must occur during the process of conception and not in a later phase of human development. For a long time, the doctrine of the Catholic Church was that ensoulment took place at forty days of gestation in the case of a male fetus and ninety days in the case of a female, and that only abortion of an ensouled fetus was tantamount to murder.27 This doctrine is not obviously incompatible with a punctualist view of personhood’s beginning. Indeed, it seems, on the face of it, to assume that punctualism is true, and that there is a distinct moment when a complete person arrives on the scene, through ensoulment. But earlier Catholic doctrine does not equate personhood with the very beginning of a human organism.
Perhaps, however, one might defend a precautionary principle in favour of presuming that humans become persons at conception. If punctualism is true, vagueness about when the ‘existential pop’ occurs will only be of the epistemic kind. There is a sharp borderline when personhood begins; we just cannot be sure where it is. When this epistemic problem is viewed alongside the moral risk of getting the threshold wrong, it could be argued that the only reasonable answer is to treat human organisms as if they were persons from conception, just in case. As Kaczor suggests, choosing a threshold of personhood is not like just choosing a minimum driving age. Specifically, what is at stake in choosing a personhood threshold seems far greater in the way of consequences of getting it wrong. Considering the gravity of the line-drawing exercise, perhaps a precautionary principle of assuming personhood from the earliest conceivable point is the soundest policy.
It is worth underscoring, first of all, just how integral the punctualist thesis is to the reasoning behind this precautionary principle. Those making the case for a risk-averse principle of this kind have presented analogous scenarios in which protagonists are unsure whether there is a tramp roaming around in a building about to be demolished, or are considering whether to risk shooting children accidentally 
by placing a firing range next to a playground. Should we blow up the building, or construct the firing range, if we cannot be sure whether or not they will result in the death of a person? Would we not do better to err on the side of caution and abstain? The very nature of these analogies reflects a clear prior commitment to the belief that for any week, day, or hour of gestation one might pick, the developing human being either is or is not determinately a person. That is, they assume the all-or-nothing, punctuated picture of how personhood begins. Asking whether a twenty-four-week gestated fetus is yet a person is like asking whether there is a tramp in the building; it is determinately true or false, and if we are unsure, we had better not take the risk.
The gradualist picture of personhood’s emergence does not suggest the same thing about the threshold of personhood. On that picture, it is not the case that between twenty-four weeks and twenty-three weeks and six days there is a determinate, independently existing answer to the question whether the fetus is yet a person or not, such that if we cannot find a reason for believing a transformation took place within that critical day, we run the clear risk that abortion at the earlier point kills a morally considerable person. Gradualism allows that the beginning of personhood is not a sharp threshold. Hence, while there may be good reason to limit abortion at, say, twenty-four weeks, no supportive argument need be made that the fetus is a radically different being one day or one hour earlier.
Perhaps, though, the defender of a precautionary principle can disavow punc- tualism and argue something different. That is, she might argue that whether or not the beginning of personhood is a gradual emergence of a property or a sharp borderline, it remains the case that we cannot know for certain when a morally considerable person has arrived on the scene. Even if personhood has no sharp beginning, we cannot be sure that a complete person has not materialized by twenty-four weeks’ gestation, twelve weeks, or even by the end of conception. Moreover, the stakes involved in making a mistake in either direction seem to go very strongly in favour of refusing abortion. If we wrongly assume that human beings have already achieved complete moral status from conception and on this basis deny abortion, a huge injustice will be done to countless women that are unjustly made to endure the burdens of unwanted pregnancy. However, if we mistakenly fail to attribute personhood to fetuses and for this reason permit abortion, the consequences will be countless unjustified deaths, a far worse prospect. Thus, the argument goes, any uncertainty should tilt us in favour of prohibiting abortion.
It is clear, however, that those who invoke precautionary reasoning must understand the uncertainty surrounding the beginning of personhood in a very particular way. Specifically, they must think that we have, in truth, absolutely no idea whether a fetus is a person at any given stage of gestation; hence we must presume that it is from the earliest conceivable point. We cannot so much as assume that the early human organism must acquire some physical properties in addition to human genes, or some psychological ones, before it is possible that it might be a person, since, for all we know, personhood does not require such things at all. This is surely the thought that is needed to ground a precautionary assumption that zygotes are persons. However, this construction of the epistemic problem involves a view of personhood as something that does not just transcend physical or psychological facts about human beings—that is greater than the sum of its base properties in the way that a painting is more than the sum of the brush strokes and droplets of paint which constitute it. Rather, it treats the property of ‘being a person’ as utterly separate from all other properties of creatures that are persons, such as that we could know everything it is possible to know about the physical and psychological nature of a creature and still be unable to say with enough confidence whether or not it is a person.
What could underwrite this sort of view about the nature of personhood? One might wonder whether it follows from metaphysical dualism. Returning briefly to the argument in chapter 5, someone committed to the dualist view of persons might question the whole relevance of transgenic, sentience, or brain-based continuums for the truth of the punctualist thesis. She might argue that personhood does not supervene at all on facts about human genes, sentience, or psychological states, but instead consists in a completely independent and separate property, irreducible to such things. It matters not then, she might say, that the coming-to-be of genetically complete, conscious, or sentient human beings involves a spectrum of innumerable variations, and that any variation one might propose as the absolute beginning of personhood cannot be non-arbitrarily distinguished from the previous one, because personhood has nothing whatsoever to do with any physical or psychological developments in human beings.
However, on a picture like this, which entirely disassociates personhood status from all other facts about human beings, puzzles will begin to arise regarding the limitations of the precautionary principle. For one, if personhood’s emergence does not supervene on any other facts about human biological or psychological properties, then there is no special reason for suspecting that human beings in particular are persons (let alone persons already as embryos) any more than there is reason for suspecting that rocks, trees, or non-human animals of all kinds are persons. Indeed, once the conditions for personhood are set loose from absolutely all other properties of personhood-possessing creatures, even basis for ascribing personhood to typical, mature human beings starts to give way. Mature humans are rational, self-conscious, communicative, capable of intelligent learning, and so on. But the view we are considering here treats the fact of personhood as separate from all such things. For all we know, then, mature human beings could be exactly as they are—thinking, feeling, and intelligent—and still lack personhood. Consequently, it will not look as though the precautionary principle has any particular reason to pick out genetically complete human beings as objects of our caution, or that there is any more reason to suspect that they are persons than to suspect the same about rocks, trees, or rabbits.
The defender of the precautionary principle will want to reply that only human beings could ever be candidate persons, since an individual human organism is the very least required for personhood. Thus, conception really is where the precautionary principle bottoms out. We should take caution with all human beings, but not with anything else. But I do not think she could make such a move fairly. If the precautionary principle could be limited by appealing to obvious basic conditions of personhood, this would go equally for the whole of gestation and beyond. Discussants would then be at liberty to argue that a being clearly lacking mental states is surely not a person, so we should take the precautionary approach of assuming personhood only once mental states of some sort (or whatever other features one thinks obviously required for personhood) are apparent. But if uncertainty about the beginning of personhood really is as thoroughgoing as precautionary reasoning needs to assume it is—if personhood does not depend on anything else about human beings that can be perceived—then even human genetic makeup cannot provide limits for the precautionary presumption. On the picture that such reasoning provides, the property of being a person might just as plausibly be possessed by a pair of gametes as by an embryo or fetus. True, gametes do not possess a complete set of human chromosomes, but on the view we are considering, this is not what personhood reduces to in any case; it is a property wholly independent from and separate to human physical properties.
Following this, it is difficult to see how the precautionary principle can limit the duty to take precautions at conception. Surely we should exercise caution with gametes too and treat all contraception as homicide. Nor, it seems, can the presumption be limited to living human beings. Are we entirely sure that dead human bodies are not persons? If personhood does not supervene on any mental or physical states whatsoever, it is not clear why they could not be. Given what one risks in making a mistake, and following precautionary reasoning, we had better take care and treat them as if they were persons, even if decomposing. These are clearly untenable implications of a precautionary principle.
Ultimately, an argument which relies on the complete disassociation of personhood status from any other facts about the physical or mental constitution of human beings runs the risk of casting off all conceptual constraints when it comes to what can be a person. If one’s metaphysical commitments about personhood entail that, for all we know, insects might be persons, this is, I think, an indication that one has taken a wrong turn. This is not to suggest that the question of who or what can be plausibly described as a person is merely a question about how we use language. I do not think this is so. But as Engelhardt helpfully points out, ‘syntactically deviant’ uses of language can provide insight into the outer limits of concepts.29 ‘That is’, he says, ‘when one recognises that it is nonsensical to speak of stones as having pains, one realizes something about the kind of objects stones are — what falls within and what exceeds the sense of their type.’30 
When personhood is not thought to depend on any base properties, even the hypothetical discovery that fetuses could do quadratic equations and contemplate the meaning of their life in the womb would not resolve the question of fetal personhood. For all we know, they may still not be persons. If this is the nature of the epistemic problem that sponsors the precautionary principle, then it seems to collapse the meaning of personhood altogether en route to advocating caution at the margins of life.
-  Cf Engelhardt (n 15).
-  See Margaret Brazier, ‘Embryos’ “Rights”: Abortion and Research’ in M Freeman (ed), Medicine,Ethics and the Law (Stevens 1988) and Francis Beckwith, Defending Life: The Legal and Moral CaseAgainst Abortion Choice (Cambridge University Press 2007) 60—1.
-  Engelhardt (n 15) 219.