An Essay called "Ethics of Abortion Redux"

Ethics of Abortion Redux

I support lenient abortion laws. I think that we should minimise the obstacles placed in the way of safe abortions at clinics. Nevertheless, unlike the majority of liberals and leftists, I don't think the ethics of abortion are clear-cut. I am right and they are wrong: the vast majority of liberals and leftists haven't thought very hard about the ethics of abortion (not surprising, given that very few people think hard about anything). I think it's obvious to anyone familiar with the argument of Don Marquis – a cruder version of which I came to independently, and wrote about in an old essay (published in September 2015) called "Deep Moral Dilemmas and the Futility of Formalisation" – that the matter is not straightforward. I will explain my reasons for supporting lenient abortion laws despite this much later on. But for now I will explain why the question of the morality of abortion is tricky, and why much of what "pro-choice" advocates say about abortion is as unrigorous as the religious malarkey of the "pro-lifers".

A lot (probably most) of what pro-choice progressives say about abortion is pure BS. Most pro-choice progressives seem convinced that, once you abandon religion, there is no argument whatsoever. This leads them to treat slogans like “Men would never tolerate having their reproductive rights infringed" as trump cards – obvious proof of male hypocrisy and the persistence of patriarchal domination within our society. Similarly, it leads them to see memes like that doctored image of Hillary Clinton, surrounded by women, signing away men’s right to ejaculate outside of reproduction, as rational takedowns. The problem is that one can only see the exertion of legislative control over abortion as unambiguous male hypocrisy if you presuppose that the analogy is completely clear – that there is not a hint of a morally relevant difference – between the reproductive role of females and the reproductive role of males. But the analogy is clearly not clear: as we all know, females incubate proto-human beings whereas males produce only gametes. And, as I will later argue, that is indeed a morally relevant difference.

A similar line that one often hears is the following: “If men had uteruses, there’s no way they’d be signing such draconian laws”. The problem with this slogan is that if men had uteruses and didn’t make such laws, then it’s hard to see in what sense they’d still be men. In fact, the sentence verges on total nonsense. It is essentially impossible to analyse the sentence in a way that makes any sense.  This is the best I’ve got: “If you imagine a world identical to our own, but except for the male (biological) sex there is a different sex which has the psychology of biological males in our current world and dominates the highest levels of government and business in the way that males do in our current world, but which reproduces in the way females do (and this would, of course, also mean that this sex has a vagina and ovaries and big breasts for lactating and wide hips for giving birth and that this sex ovulates, which would mean an entire different evolutionary history (which would mean that this sex looks like females in our current world in every possible way (gracile facial features, less body hair, etc), and would, of course, mean that the “women” in this world would have to be sexually male to produce the sperm for the “men”), then in this world the new sex we are imagining, which looks identical to the female sex but which we are calling “men”, would not make restrictive abortion laws.”

The main problem with this analysis, however, is that the best theory for the origin of patriarchy puts uteruses – the nature of female reproduction – front and centre, meaning that it is almost certainly not reasonable to imagine a world in which uterus-holders are the sex oppressing the other. The dominant theory for the origin of patriarchy (more precisely, the dominant theory for the fact that every civilisation ever is a patriarchy) holds that patriarchy begins when people settle down in chiefdoms or states. In chiefdoms or states, the wide kin connections and strong extended-family relations of hunter-gatherer societies fall apart (replaced by more ‘nuclear’ relations), and women no longer spend their time in the company of other women, with a communally decided group of different possible sexual partners (based roughly on genetic distance). Instead, women become extremely important to their (struggling, farm-working) fathers for their reproductive capacity – for their role as dowry generators. In order that “buyers” of women can be assured that their family line is secure (and because males are evolutionarily programmed to want only their sperm to enter women in whom they have sexual interest), virginity suddenly becomes extremely important, and in order to guarantee virginity, the veil emerges and women are kept to the home. And thus, women become chattel.

Anyway, that was incredibly perfunctory, but the point is that the uterus was very likely key to the origin of patriarchy, which means, in turn, that “Men with uteruses” would not be men; they would be women in every possible sense, including in the sense that they would be the oppressed sex.

 Finally, as for the "my body, my right" line (or the emphasis on “bodily autonomy” in general), it should be obvious why this isn’t a trump card: it presupposes that there is nothing wrong with abortion. Rapists aren't allowed to defend themselves by saying "my body, my right" – and why? Because one's bodily autonomy ends when it affects another's.

Now, there is, of course, a perfectly respectable way of arguing that there is nothing wrong with abortion (any instance of abortion, no matter what trimester the woman is in, no matter her personal circumstances and the effects a baby would have on her life), and this is the Singerian way. However, as we will later see, it does have implications that most find very unsavoury and deeply unsettling. Meanwhile, Don Marquis’ argument that abortion is seriously wrong (that individual abortions may be more or less wrong (depending on woman’s circumstance, etc) but that all abortions are probably more wrong than, say, any case of theft) has its own very troubling implications – primarily for population ethics. The fact that both these positions are unsatisfactory in their own way will ultimately lead me to a very odd conclusion. But we’ll get to that much later.

Because Don Marquis and Peter Singer both employ the same style of moral reasoning (the only possible style of moral reasoning, which essentially involves figuring out which moral position on the given contentious moral issue is most compatible with moral principles that people would overwhelmingly agree on), it is not surprising that they have a lot of common ground. Both Marquis and Singer agree that lots of ‘factors’ make individual acts of killing wrong, and that the differing presence of these factors in different cases explains  why not all acts of killing are equally wrong (killing Hitler is less wrong than killing Gandhi, etc). Both also use the word “killing” in such a way that entails that aborting a foetus is an act of killing (so there is no semantic gulf). Finally, both think there is a feature of a sizeable subset of acts of killing whose presence in a given act of killing is sufficient to make that act of killing wrong, irrespective of other factors. For Marquis, the key feature is deprivation of conscious, intelligent experience: he claims that the effect that the killing of humans (or proto-humans) who have a possible, conscious future has in depriving these creatures of future prospects – of future experience, of future joy, love and fulfilment, of future goals and achievements – is that feature of a sizeable subset of acts of killing whose presence is sufficient to make any act of killing with that feature wrong (incidentally, this principle may ‘justify’ the common belief that killing animals (other species) with a future is not necessarily seriously wrong: even other mammals probably don’t have the same richness of experience as humans, or the same ability to remember the past and to formulate long-term goals for the future, which means the deprivation is not as great). For Singer, the feature of a sizeable subset of acts of killing whose presence in a given act of killing is sufficient to make that act of killing wrong is the effect that that the killing of creatures with an ‘ideal desire to live’ has in denying these creatures their life. So whereas Marquis emphasises the importance of objective deprivation, Singer thinks what matters more is the victim’s psychology.

At this point, you may be wondering what work the “ideal” adjective does in Singer’s story. The answer is not actually straightforward. The obvious work it’s doing is covering every case where the person killed has a ‘good life’ and is more or less ‘content’ but who feels no specific desire at the point of their being killed to “live” (including cases where the person killed is asleep or temporarily unconscious and thus probably has no ‘desires’ at all (depending on one’s understanding of philosophy of desire)). I think Singer has also suggested that it may cover cases where the person killed is depressed and feels terrible about life and their future, but would, if they had not been killed, soon overcome their depression. However, this does seem a little dodgy (does it not?).

Marquis, meanwhile, thinks that – regardless of other factors – it is wrong to kill a person who is severely depressed and will never emerge from their depression, unless their depression is so bad that they won’t even have any future worth living. Singer would have to say, I believe, that killing a severely depressed person who will never emerge from their depression is not necessarily wrong, even if they do actually have some future worth living (the killing may, of course, be wrong because of the distress it will cause other people (family members, friends, etc)).

The above implication of Singer’s position is, I suspect, already a little bit troubling to some people, but that’s not half of it. There are actually far more troubling implications – and these implications are the reason for the controversy Singer has often faced (in particular, the massive backlash to Singer in Germany). It’s obvious why placing the emphasis on the ‘ideal desire to live’ makes abortion permissible, because there is no reason to think that foetuses are capable of having a “desire to live”, or probably even anything we could reasonably call a “desire” at all (and having an unwanted baby can, of course, ruin women’s lives, as well as the lives of those around her). But placing the emphasis on the ‘ideal desire to live’ also means that killing the mentally disabled, killing senile and impaired old people, and killing very small infants (at least), is not wrong in itself. That is to say, if killing a mentally disabled child wouldn’t cause anyone distress and if the killing was carried out in such a way that didn’t cause the mentally disabled child any suffering, then Singer believes it would be permissible. Even if it did cause his mother a nontrivial degree of distress, it would only be wrong in proportion to that amount of distress. I think many people find this to be a repugnant conclusion.[1]  

I have myself invented a thought experiment where our intuitions seem to strongly militate against the Singerian position (the thought experiment was originally contained in the aforementioned essay called “Deep Moral Dilemmas and the Futility of Formalisation”):
"Imagine, if you will, a healthy woman late in the first trimester of her pregnancy, roughly 10-12 weeks in (the start of foetushood). Suppose, if you can, that one day this woman contracts a very strange and disturbing urge. For whatever reason – be it insanity, iniquity or both – there is nothing in the world she wants more to do than scramble her foetus’ brains. We can also suppose she has three methods at her disposal. She could hire an unscrupulous and highly skilled surgeon who could use keyhole surgery to get inside the uterus, break through the weak skull and fiddle around carefully with the developing brain; she could ingest some kind of futuristic nanotechnology capsule which heads directly towards the uterus and, when on a conduit towards it, begins to release some kind of neurotoxic chemical; or she could just drink a lot of alcohol and take a lot of harmful drugs. Since her desire is not wanton malice, it is important to her that she does not do so much damage that the foetus dies. Indeed, she does want the foetus to survive to become a braindamaged baby and human. Thus, whichever method this woman opts for, we can be sure she will be careful not to take it too far."

We thus imagine that this woman brings a severely retarded child into the world, probably blind and unco-ordinated and maybe with autonomic issues, and that it’s all her fault. The terrifying thing is that Singer’s position entails that there is absolutely nothing wrong with this, except insofar as it causes other people distress – and it may not cause anyone distress. In fact, if we imagine that the woman kills the retarded child soon after it is born, without anyone’s knowledge (and painlessly), then Singer would say that nothing bad has happened at all. In fact, if she is such a sadist that this whole sick performance gives her pleasure, then Singer would be forced to the conclusion that this series of events is overall a good thing.

My friend HR tried to get around this thought experiment by means of the following argument:
"The scrambling is wrong at the moment of its occurrence for the same reason that my setting up on the footpath a caltrop which contains a neurotoxin that will seriously retard the first person to step on the caltrop is wrong at the moment of its occurrence: it drastically increases the probability that a misfortune will befall an unconsenting agent.

That the infant is born retarded – was never a non-retarded person, as the caltrop victim was – is immaterial: what matters is that the infant is indeed a person at the moment of its birth and thenceforth. It is when the infant becomes a person (whenever we judge that to be) that a victim of the mother comes into existence. (Whether the person retarded by their mother is eventually capable of comprehending their plight, or wishing it were otherwise, is also irrelevant: it is sufficient that all people considered sound of mind [like the caltrop victim] would consider mental soundness preferable to mental deficiency [and it is this preference that generates the initial intuition that the scrambler is acting immorally {and even if one took the ignorance-is-bliss line, in favour of retardation, the mother remains condemnable for acting in a way /she/ believes to be harming an eventual person}]).

This is a structural similarity between two cases that evinces the importance of victimisability in ethical problems.

But what if the caltrop contained not a retardant but a fatal neurotoxin? It is indisputable that the act is still deeply immoral – for it is a form of unjustified murder, an axiomatic sin in the world’s ethical systems. Is abortion, the antenatal analogue, equivalently immoral? No, and this is the kicker: because there never exists a person to be the victim of an act of killing. In the scrambling case, a person comes to suffer a misfortune initiated before their personhood. In the retardant caltrop case, a person comes to sustain an ill in the ‘normal’ way that people sustain ills – so too in the case of the fatal caltrop. But in the case of abortion, no person is /deprived/, to use Marquis’ buzzword, of a valuable future. There is no person. A future in which a foetus becomes a person is merely averted.

Your fatal error in interpreting your own thought experiment was to focus to the exclusion of all else on the fact of a particular future being averted in both the case of the scrambling (a future in which there exists a mentally sound person) and the abortion (a future in which a person exists). You treated this similarity as one that renders the two cases equivalent, but they’re not. You assumed that we impute wrongness to the scrambling /only/ because it involves the aversion of a preferable future, but this is incorrect. The scrambling is wrong because the preferable future of a /particular person/ is averted. You ignored the ontic criterion."

Now this is an interesting way of arguing (Singer plus an extra subtlety) but it's flawed, as HR conceded: the notion that what really matters is that everyone can 'see' a victim is just an ad hoc rationalisation of his intuitions, without basis in any familiar ethical principle. If the human being that resulted from the disturbed woman's pregnancy was severely retarded, and thus had no ability to conceive of himself as a victim and feel loss, and yet we still think that what she did was wrong, that can only be because the Marquisian objective notion of deprivation matters, not just the Singerian deprivation+ideal desire to live. 

Our intuitions about premature babies also seem to clash with the Singerian position. If a late-second-trimester baby is put in an infant incubator, I think 99% of people would say that it would be deeply wrong for a nurse (or someone else) to switch off the machine and allow it to die, even if the parents never found out that the baby had been killed and therefore never had to feel the distress of knowing that someone had intervened in their baby’s future.

Of course, the Marquisian position that abortion is seriously wrong (not as bad as murdering a human being out of the womb who has loved ones and specific goals, but still very wrong) and should be avoided if at all possible has some of its own problems. As we’ve seen, Marquis' big claim is that objective deprivation of future experience is what matters. Marquis claims to surmount the regress objection – 'But doesn't your view make wanking wrong?' – by putting the morally significant boundary at the point of clear individuation. The problem with putting the significant boundary at the point of clear individuation (which Marquis has suggested is after the 16-cell stage, when the cells are no longer 'totipotent' and no 'twinning' can occur) is that we can now split these totipotent cells and implant them into women to gestate. It seems that Marquis would say that all these humans in potentia should be born, or at least preserved for as long as possible – and this does seem to raise big problems for population ethics. And it might be worse than this: it’s not clear that he wouldn’t be forced to the conclusion that we should split the totipotent cells. Marquis would almost certainly reject this claim, and he would defend this rejection by saying that the fact that it’s bad to destroy any one of the post-split cells does not imply that it is good to create more potential life by splitting the totipotent cells to begin with. But I’m not sure of this response: I think it’s ultimately not clear why, if depriving life is such a terrible thing, it’s in no way a good thing to try to create more of it. How can life and conscious experience be simultaneously so valuable that losing the potential for any is a great tragedy, and so unimportant that it is not in any way actually a good thing to create more?

In general, if you look at the bigger picture – all the larger issues tangled up in the actual practical reality of abortion – Marquis’ position does seem, even if ‘right’, not ‘implementable’ in an ethical way. It would be natural to suppose that someone who feels that abortions are seriously morally wrong would want to stop women having abortions post-haste. However, the fact is that if just a small percentage of women around the world with unwanted pregnancies decided to forego an abortion because of government policy instituted as a result of this moral conclusion, that would lead to a massive jump in population (and would continue to have an effect on increasing the population if women were (somehow) continuously made to maintain unwanted pregnancies). This, in turn, would, of course, lead to the further degradation of the planet, further claims over limited resources, further poverty, further urban overcrowding, and exacerbate the terrifying, possibly existential (to civilisation) ecological disaster we are already facing.

Pro-choicers often make a similar point, though with a focus on how punitive laws affect female happiness and feminism rather than on ecology and society as a whole. They point out that banning abortion at clinics is certainly not going to eliminate all abortions – it may not even significantly reduce the overall rate of abortions – and probably leads to more misery in the world by forcing women into parlous situations or forcing them to have children which they do not want, at risk to their dreams and careers.

Marquis would probably claim that, because he can recognise the very bad things that would result from punitive law changes on abortion, he doesn’t really believe that it would be desirable to stop all women having abortions, unless this outcome was achieved by birth control. But whilst this would be a good response to this particular objection, this retreat might induce us to question whether he really thinks abortion is seriously wrong.

Of course, Singer, like any utilitarian, does himself have to deal with more theoretical issues in population ethics – those raised by Derek Parfit in Reasons and Persons. But clearly by having a permissive view of abortion, he avoids all of Marquis’ problems.

So, seeing as both positions face problems, what’s my conclusion? The answer is that I think the abortion debate reveals the limits of ethical reasoning. I think there's no clear fact of the matter; I think that abortions perhaps should be avoided (even first-trimester abortions), but I really don’t strong feelings about this (especially because it’s easier to empathise with the woman who already exists than the proto-person). That’s my strange view.

And so why do I support lenient abortion laws? Precisely because I don’t really have strong feelings about the ethics of abortion. The government shouldn’t legally prevent women from doing something that is not unambiguously seriously wrong (which is why even Marquis’ position doesn’t necessarily have strong legislative implications, as we just saw).

---

[1] To be fair, Marquis’ position itself suggests that killing a mentally disabled person may be less bad, all else being equal, than killing someone of high intelligence, because there is probably less to deprive.