Analogical Arguments and Sex Equality
The Use of Analogies in Abortion Argument
The last two chapters gravitated around Judith Thomson’s well-known violinist analogy argument about abortion. In the course of considering the Good Samaritan Thesis and the Justified Homicide Thesis, we considered various points of disanal- ogy between pregnancy and the violinist scenario, and asked whether those disan- alogies were morally relevant to our appraisal of the two cases. But some people might feel deeply unsatisfied with the extensive use of the violinist analogy in moral reasoning about abortion. For that matter, they may feel that all such analogies— conjoined twins, hypnotized knife attackers, cannibals at sea—fail to serve as any kind of useful guide to the moral and legal constraints on abortion that apply if we assume the fetus is a person. On the contrary, perhaps reasoning using cases such as these, which are so different from pregnancy in so many ways, is only likely to impede clear moral thinking on abortion.
The philosopher Margaret Little has argued that being pregnant involves a very unique kind of physical intertwinement that is not easily reflected by analogies of any kind.1 This is the fact that, for the pregnant woman, being pregnant is to be literally occupied by another life form:
To be pregnant is to be inhabited. It is to be occupied. It is to be in a state of physical intimacy of a particularly thoroughgoing nature.2
In this respect, even the conjoined twins example clearly differs. Although their intertwinement is certainly of a ‘thoroughgoing’ nature, it does not (and did not in the Re A case) involve the complete occupation of one person by another. If the fetus is a person, this kind of intertwinement is unique to pregnancy. Little argues more broadly that the nature of the union between woman and fetus cannot be properly analogized at all. For her, the unique physical intimacy that pregnancy entails ‘means that the fetus, the gestating woman, and their relationship do not fit ready-made categories’.3 In attempting to fit them into such categories, she warns that we will often appeal to analogies which are ‘at best awkward, at worst dangerous, but always distorting’.4 Attempting to understand pregnancy by considering
- 1 Margaret Olivia Little, ‘Abortion, Intimacy and the Duty to Gestate’ (1999) 2 Ethical Theory and Moral Practice 295.
- 2 ibid 301. 3 ibid 296. 4 ibid.
Arguments about Abortion: Personhood, Morality, and Law. First Edition. Kate Greasley. © K. Greasley 2017. Published 2017 by Oxford University Press.
situations which lack the same degree of physical enmeshment will always miss something important. Little draws on the feminist theorist Catharine MacKinnon’s well-known remarks that ‘nowhere in law is a fetus a fetus’, with a concept of its own. Rather, it must always be defined by some existing classification, whether a person or a body part.5 But why, MacKinnon asks, should we do things this way round? Why not treat the fetus and the pregnancy relationship as paradigmatic categories and analogize other things to them?
Little thinks we should expect that a tradition which has always imagined persons as physically separate will not fare well in analysing situations where ‘persons aren’t as it imagines them’. Hence it is not surprising if we find ourselves scratching our heads about what follows if a great number of persons actually exist inside of others.  But her main problem with traditional analyses of abortion is that they do not ‘take as pivotal the fact that gestation occurs inside of someone’s body’. Once one focuses on pregnancy’s unique form of physical intimacy, Little thinks it will become apparent that the moral permissibility of abortion does not depend on prenatal personhood. This owes to the fact that the kind of occupation entailed by gestation requires a special form of consent. Consequently, for Little, mandating continued gestation when that consent is missing is to inflict unjustified harm on a pregnant woman, much like an assault.
In effect, Little doubts that anything at all can be learned about the nature of pregnancy and abortion through analogies about violinists, conjoined twins, or, indeed, anything else. Her argument might be understood as the ‘weirdness objection’, introduced in chapter 2, in a more developed form. The weirdness objection, we saw, claims that exotic thought-experiments cannot tell us much about abortion because they are weird, whereas pregnancy is not. Little’s argument is that we should resist the temptation to analogize pregnancy to anything else, weird or not, since doing so is likely to direct us away from the unique and morally pertinent features of pregnancy—specifically, the bodily occupation of the pregnant woman by the fetus.
This scepticism about the usefulness of analogies might, in turn, speak to an even more general scepticism about the value of thought-experiments in ethics. Such hypotheticals typically require us to filter out any number of complicating realities in an effort to, in Stephen Mulhall’s words, ‘clear the scene of any other potentially polluting concerns’. To demonstrate this sort of disagreement about thought-experiments in ethics, Mulhall describes a widely cited study conducted by the psychologist Carol Gilligan:
[The conductor of the study] presented groups of students with a problem case: would it be morally permissible for an impoverished person to steal medicine from a chemist in order to save the life of her sick child? The men argued with one another about the immorality of theft and the sanctity of human life; the women inquired: ‘Why doesn’t she ask the chemist to give her the medicine she needs?’ Some will take it that the women had simply misunderstood the point of the problem case, and of problem cases in general; for to include such a dialogue between chemist and parent is to change the case, and thus to avoid responding to the particular issue of principle it was intended to abstract from the complexity of reality. Others will take it that the men had been distracted by matters of principle from attending properly to the concrete reality of moral experience and the possibilities of human fellowship.
When confronted with Thomson’s violinist analogy, the women in Gilligan’s experiment might well have pointed out that people never do, as a matter of course, find themselves in the situation of having their bodies connected up with that of an unconscious violinist, or anyone else. Can it possibly not matter that Thomson’s scenario is not a regular part of human life, or that if this really were to happen to you, you would be the first and only known person to be subjected to that bizarre imposition? Perhaps the relative normalcy of pregnancy cannot be dismissed as a morally irrelevant feature of the pregnancy situation. Rather, perhaps the fact that pregnancy is so embedded in everyday human life colours the entire meaning of pregnancy, and abortion, for those who experience it. There are, for instance, no websites and stores exclusively dedicated to selling merchandise to people currently supporting violinists, no ‘violinist leave’ standardly worked into employment contracts to cater for the very realistic eventuality that the Music Society might kidnap you for nine months at any time, and no general understanding of and allowance for the implications of hosting the violinist and the setbacks it entails. The reality of pregnancy is, of course, drastically different. It is everywhere, affects everyone, is not generally regarded as an unfortunate occurrence, and is often celebrated even when unintended. (By way of contrast, were anyone at all to celebrate the fact of their having been kidnapped and plugged into the violinist, we could only conclude that she was of unsound mind). It is, moreover, the way that human beings reproduce. All of these features, along with the intimacy of fetal occupation, affect the real life implications of pregnancy in ways that cannot be replicated in the violinist scenario.
In the course of her discussion, Thomson offers up some further analogies which better capture the normalcy of pregnancy, including the scenario in which ‘people-seeds’ routinely drift around in the air and might drift through windows and take root in homeowners’ carpets. The people-seeds might then begin to grow into ‘person-plants’ for whom the proprietors become responsible for keeping alive. In this scenario, Thomson thinks it obvious that you are not responsible for maintaining the life of the person-plant if that means continuing to suffer its imposition within your house, merely because you knew that people-seeds are generally drifting around and failed to batten down the hatches or stop any air from getting in.
The person-seed analogy appears to replicate the normalcy of pregnancy in the respect that the burdensome imposition is not extraordinary or wacky as the violinist situation is—not in the world we are imagining, anyway. However, Thomson’s alternative analogy fails to import any further likely consequences of our living in a world in which new humans standardly come to exist in this way, and the effect that these additional features might have on our appraisal of the moral problem with which we are presented. The normalcy of pregnancy in the world in which we live embeds it in a whole number of social realities which inform its wider meaning and burdensomeness, as well as the burdensomeness of avoiding it. These realities include the fact that pregnancy is the consequence of sexual intercourse, not the drifting of seeds to which we are totally unconnected; the fact that it is often celebrated, that it is the cause of our own coming into being, and so on. Thomson does not try to account, in her analogy, for the different ways our living might be structured if people-seeds were drifting around everywhere, or if they were the main source of human reproduction. How, for instance, might the social significance of leaving one’s windows open change if this were the case? Would someone not have developed a high-tech security system for keeping one’s windows shut, except when people-seeds are desirable, or innovative strategies for filtering them out, or alternatives for ventilating the home? These are the kinds of background circumstances that the women in Gilligan’s experiment might well want to know about, not because they miss the point of the inquiry, but because they understand the relevance of this surrounding context to the questions Thomson and her challengers are asking, which include considerations about the burdensomeness of avoiding pregnancy.
In his rebuttal of the weirdness objection, Boonin maintains that the mere strangeness of the violinist scenario cannot matter unless it constitutes a morally relevant difference. He writes:
If the differences do not themselves prove to be morally relevant, then the mere strangeness of the example in and of itself will provide no further reason for rejecting the argument’s soundness. 
I agree with Boonin that strangeness per se is not a reason to reject a thought- experiment in the course of abortion argument. But as the people-seeds analogy may demonstrate, it can be difficult to isolate the strangeness of an analogy from a host of features that are morally salient. As Mulhall argues, the moral significance that an issue has for us often owes to ‘the complex web of interrelated matters’ within which we actually encounter it. Rather than helping us to gain clarity about the rights and wrongs of the problem, extracting it from its actual context may just, in effect, ‘change the subject’.n
In the violinist’s case then, the concern may stand that putting its weirdness entirely out of our minds might only cause us to miss truly important differences between Thomson’s analogy and pregnancy that emerge from their different surrounding contexts. Indeed, we may miss the fact that the violinist scenario is really nothing like pregnancy except in the single sense that it too involves bodily inter- twinement with another human being for nine months.
Does this mean Little is right to claim that philosophers are in error whenever they argue about abortion by reference to broader categories or analogies? This will present great difficulty if we are asking ourselves what follows about abortion if the fetus is a person, for, other than by appealing to the rules about refusing aid or permissible killing which we accept outside of the abortion context, how can we reason about that matter at all? Pregnancy is of course unique. But so are many problems the moral implications of which we may want to assess, including Mary and Jodie’s predicament and that of Dudley and Stephens’s. Pregnancy is not unique in being unique. And like any other kind of life and death problem, discussants will find themselves in an argumentative vacuum if they cannot search outside of the pregnancy situation for more general principles that can help shed light on it. Those principles must be extrapolated from other contexts if they are to provide moral insight based on more general commitments, rather than generating an ad hoc conclusion about abortion.
This does not mean that the specific features of the pregnancy situation are irrelevant to moral and legal reasoning about abortion. There is always scope to differentiate pregnancy from a comparator case, and thus to refine the principle one is considering or show that it does not apply to abortion. But pregnancy’s uniqueness per se cannot be offered up as a point of distinction unless one explains in what respect it is unique and why this matters for our moral evaluations.
Still, perhaps the most outlandish thought-experiments lack usefulness for a very particular reason: that they cannot produce reliable intuitions from which to make further deductions. Take the following thought-experiment presented by Nicole Hassoun and Uriah Kriegel as part of an argument about the relevance ofpotential personhood for moral rights:
Suppose that many years from now, a space elevator is installed between Earth and Mars, and that an oyster finds its way to the elevator. At this point, the normal course of events should lead to that oyster’s becoming conscious in the absence of intervention. The oyster on the elevator is thus potentially conscious in the sense in which fetuses and neonates are—it is, so to speak, en route to consciousness. Yet it still seems intuitively permissible to kill the oyster.
The point these authors are trying to make is that if we think it permissible to kill the oyster, it seems we do not believe that being on an uninterrupted path towards full consciousness endows a creature with the fundamental right to life. If this is correct, then embryos and fetuses will not possess the right to life merely in virtue of being on that same path. But it may be objected that the oyster scenario is so other-worldly that it is difficult to draw any firm conclusions about it. How would we feel about killing oysters if there were space elevators to Mars onto which oysters tended to climb, at the end of which they were transformed into conscious beings? Perhaps such a scenario is so far removed from the possibilities around which our moral sensibilities are constructed that there is really no second guessing how we might assess it as a reality. If this is indeed the case, we may not be able to deduce anything useful about the grounds for attributing moral status to fetuses from our unreliable intuitions about the oyster scenario. Moreover, to help us know whether a creature’s being on a path to consciousness is sufficient for a fundamental right to life in circumstances analogous to pregnancy, we would have to construct the example so that, somehow, the implications of not permitting ‘oystercide’ were roughly comparable to the burdens of unwanted pregnancy and childbirth, including their particular effect on one already disadvantaged class of persons. Whatever imaginative twist to the thought-experiment could account for this calls for more creativity than I can muster. It might also be objected that, for true comparability, we should imagine a different reality whereby all clear cases of persons of which we are aware, including ourselves, start out as oysters, even though countless oysters never attain consciousness. How might this affect our judgements about oystercide? It is admittedly difficult to say. Like Thomson’s people-seeds analogy, it seems that much more of the surrounding context of pregnancy would have to be imported into the thought-experiment to reflect all of the morally relevant aspects. Once the relevant modificaions are made, however, we may well find ourselves hesitating to draw any firm conclusions at all.
The violinist and conjoined twins analogies are, in these respects, somewhat different. In the violinist scenario, the supposition is, I take it, that what takes place is extremely out of the ordinary. There is thus no need to perform the imaginative feat of deciding how everything else might be if this sort of occurrence happened all the time, and how that might affect our reactions, once every other morally relevant feature of pregnancy is built in. This difference might itself be considered an important point of disanalogy, as has been noted. The ‘weirdness objection’ may be apt insofar as our conclusions about abortion do not flow directly from our conclusions about unplugging the violinist when the latter case is so alien whereas the former is not. Quite apart from suggesting that the analogy is too weird to yield reliable reactions, the problem here is that the conclusion we do form about the violinist might not clearly carry over to abortion when all the morally relevant implications of the extraordinariness of the violinist case, as compared with the familiarity of pregnancy, are factored in. But this does not mean the analogy is unhelpful. On the contrary: distilling the meaningful differences between the violinist situation and pregnancy is precisely what may help us to isolate some morally relevant features of pregnancy and abortion.
The conjoined twins analogy is not quite the same in terms of sheer outlandishness. It is certainly extraordinary in one sense, that is, in its rarity. This is not the same extraordinariness that we encounter in the oyster case, because the conjoined twins case is real and, hence, perfectly possible. It is therefore a moral problem that can arise without anything else about the world around us having to dramatically change. Consequently, we need not hold our moral reactions to the problem quite as loosely as in the truly extraordinary oyster scenario. And if this is so, our moral evaluation of the conjoined twins problem can tell us something about the broader principles we accept regarding the killing of one person who is bodily dependent upon another. Again, some people will point out disanalogies arising between the conjoined twins case and pregnancy, even on the assumption that a fetus is a person. But attending to these disanalogies and whether they make a moral difference will again be an integral part of our reasoning about abortion.
Returning to Little’s argument, it may be suggested that she is in fact really pointing to a morally relevant difference between pregnancy and everything we might be tempted to analogize it to when she underscores the outright occupation of the pregnant woman by the fetus. As Little says, it is inside her, not just attached to her. This is one aspect of pregnancy’s uniqueness that clearly differentiates it from the violinist case and from the conjoined twins. Is it a difference that matters for the analysis of abortion as a category of justified homicide? Little claims that the special level of intimacy entailed by unwanted bodily occupation makes mandated gestation intolerable, whatever moral status one assigns the fetus. But it is interesting that she also regards the personhood category as particularly unsuited for conceptualizing the fetus because of the difficulty we will have imagining persons that are not ‘physically separate’. Little does not deny that such persons could exist. She only suggests that our conventional ideas about what persons are like will make it extremely hard to imagine them thus.
This observation could have more than one implication. Little, it seems, intends it to speak to the treacherousness of all analogies, extraordinary or not, which try to discern what would follow about abortion if the fetus were a person. It would be difficult, on her view, to see clearly on this question when the ‘person’ we are thinking of is so different from persons as we ordinarily think of them. Little takes it to endorse her wider point that when arguing about abortion, people should steer clear of subsuming the fetus under any of the traditional categories, such as ‘person’ or ‘non-person’. But the fact that discussants struggle to keep to the brief when trying to imagine what moral consideration is owed to persons entirely enclosed within others can equally reveal something about how they think of persons. Why do these thought-experiments stretch the imagination to breaking point? Why is it that discussants can only begin to formulate thoughts about what is owed to persons like that by analogizing them to partially separate persons, physically attached to others but still visible in the world and in direct contact with it, like the violinist or the conjoined twin? Rather than suggesting that all analogical argument in this topic is unhelpful or distorting, the struggle to account for the fetus’s complete enclosure whilst maintaining the personhood proviso could reveal something significant about our concept of a person.
The question also remains as to how moral and legal reasoning about abortion is to proceed other than by looking outside of the abortion context, since there is no other ostensible way to draw conclusions that are grounded on, and consistent with, the more general commitments that we have. Someone may press the objection that the violinist or conjoined twins scenario reflects some of the features of pregnancy but not all of its features. But this just misses the point of the examples. The analogous cases are not meant to resemble pregnancy in all respects, but to isolate certain features whose ethical significance we are testing. With the violinist, this is the bodily dependence of one person upon another at a considerable cost to the host. With the conjoined twins, it is the effect of bodily dependence by one person upon another when the weaker party’s dependence threatens to kill the other party, and where they themselves are beyond saving. If a discussant thinks that other features of pregnancy not captured in either case are also ethically salient to abortion, even on the assumption of fetal personhood (including the fact that the fetus is entirely enclosed within the woman), then she will have to explain what they are and why they are morally meaningful. But engaging in this kind of reasoning is the process of discerning and refining the more general principles we are prepared to defend, and what bearing this has on abortion morality.
Interestingly, it seems that even Little’s own analysis of abortion cannot do without reliance on more general categories and principles. Her argument is that pregnancy entails a particularly thoroughgoing degree of intimacy, one which cannot be legitimately forced upon someone. As she says, ‘To mandate that the woman remain pregnant is to mandate that she remain in a state of physical intertwinement against her consent.’14 Being in a state of physical intertwinement with another being must be consensual. Little derives this conclusion from some of the concepts and categories that we employ outside of the abortion context. She writes:
Just as sexual intercourse can be a joy under consent and a violation without it, gestation can be a beautiful experience with it and a harmful one without it. To think that the above concern insults the meaningfulness of pregnancy is simply to misunderstand the point. We don’t impugn how meaningful it is to have willing sex when we protest against a rape; the fact that sexual intercourse can be wonderful doesn’t mean we would think it appropriate for the state to conscript people into serving as prostitutes.1*
Little takes these thoughts about rape to indicate that ‘consent determines valence on matters of physical integrity ... a trip to a beautiful country turns into a kidnapping, a surgery into an assault, when they are done against one’s consent’d6 In other words, we accept the general principle that non-consensual bodily invasions are not acceptable, and that principle is meant to tell us that gestation without consent is no different, and that mandated pregnancy through abortion prohibition is consequently unacceptable. This is in truth not methodologically all that different from Thomson’s use of the violinist scenario to tell us that gestating a fetus cannot be morally obligatory, or from my use of the conjoined twins case to tell us whether abortion might amount to justified homicide in morality or law. Someone deploying Little’s original complaint might direct it back at her here and claim that we cannot learn anything about pregnancy abortion by thinking about rape, the two being such distinct things. But she would be right, as I see it, to respond that there is no way to make her argument against mandated gestation without appealing to extra-pregnancy contexts, and that the point of the examples is simply to prove the independent significance of consent in scenarios of bodily interference. One potentially significant point of disanalogy between the bodily invasion involved in rape and mandated pregnancy, however, is that the perpetrator of a rape is inflicting a non-consensual bodily invasion upon a person, whereas those who refuse to perform abortion are only failing to help end her non-consensual occupation by the fetus. Because of this difference, there may well be difficulty likening mandated pregnancy (through the prohibition of abortion) to bodily interference akin to assault.
If Little’s argument is wrong, it will be because morally pertinent differences like this hold between pregnancy and abortion and non-consensual sex or surgery, either meaning that the consent principle does not apply in the same way (perhaps because pregnancy is not the ‘invasion’ that rape and assault are), or because other considerations override it. I do not want to pronounce on Little’s argument here. My point is only that Little can no more manage to draw out the morally important features of pregnancy by discussing pregnancy alone than those who compare it to somewhat wilder scenarios.
-  Catharine A MacKinnon, ‘Reflections on Sex Equality under Law’ (1991) 100 Yale Law Journal1281, 1314.
-  Little (n 1) 297.
-  Stephen Mulhall, ‘Fearful Thoughts’ (2002) London Review of Books http://jeffersonmcmahan.com/wp-content/uploads/2012/11/London-Review-of-Books.pdf (last accessed 28 March 2014).
-  ibid 2—3.
-  Judith Jarvis Thomson, ‘A Defense ofAbortion’ (1971) 1 Philosophy and Public Affairs 59.
-  David Boonin, A Defense of Abortion (Cambridge University Press 2003) 141.
-  Mulhall (n 7) 3.
-  Nicole Hassoun and Uriah Kriegel, ‘Consciousness and the Moral Permissibility of Infanticide’(2008) 25 Journal of Applied Philosophy 45, 51 (emphasis in original).
-  Little (n 1). Little goes on to define exactly what kind of attitude towards the fetus ought to, onher view, accompany pregnancy, this being a simple openness to developing a relationship with thefetus which, if entered into, would ground some responsibility to sustain it.