A Utilitarian Framework for Evaluating the Morality of Abortion

by Christopher Carr

Jeremy Stangroom is a British author, philosopher, co-founder of The Philosopher’s Magazine Online – one of the premiere philosophy publications on the Internet – and the director of Philosophy Experiments – where users can participate in a variety of interactive thought experiments.  One of the more popular experiments is called Whose Body Is It Anyway; it is about the taboo taboo touchy touchy subject of abortion.  

I strongly recommend completing the experiment before continuing to read this blog post.

The Whose Body Is It Anyway experiment thus far has had two particularly interesting results: the first is that opposition to abortion tends to come disproportionately from the religious:

(B)y far the biggest predictor of whether a person is going to be opposed to abortion is religious belief. So, for instance, 83% of people with no religion support the right of a woman to have an abortion, compared to only 37% of Christians.

The second interesting result – a significant result indeed – is that people opposed to abortion tended to be generally unconcerned with the medical seriousness of miscarriage:

Near the beginning of Whose Body Is It Anyway?…you’re asked to rank the following medical issues in order of seriousness (focusing only on numbers of deaths): cancer, multiple sclerosis, miscarriage, stroke, heart disease and housemaid’s knee. Then, if it turns out that you’re against abortion, the activity complains if you haven’t ranked miscarriage as being a serious medical problem (since, for example, in the United States alone there are estimated to be more than a million miscarriages each year).

I have many problems with this interpretation, but let’s focus on only the most substantial ones – those related not to the particulars of the way in which Mr. Stangroom framed his experiment but to the issue itself: first, miscarriage is a natural phenomenon, similar to death by old age, whereas abortion is the direct and predictable result of human action, such as euthanasia.

For those participants who see this distinction as the most important factor – i.e. it’s okay for evil to exist as long as we don’t cause it, so we might as well not mess up the things we can control – a sentiment I sympathize with – this may be the most important factor in downplaying the seriousness of miscarriage.  Stangroom has addressed this problem as follows:

1. If you’re opposed to abortion, then likely you think a foetus has intrinsic moral value;

2. If a foetus has intrinsic moral value, then presumably you believe that its life if worth saving (all other things being equal);

3. It is estimated that up to 50% of pregnancies end spontaneously (i.e., as a result of miscarriage);

4. This is a serious problem because (a) we’re talking about the “deaths” of millions of foetuses, each one of which has intrinsic moral value; and (b) we’re talking about the “deaths” of millions of foetuses, each one of which has a life worth saving (all other things being equal);

5. Therefore, given the number of deaths that occur, miscarriage should be ranked as a serious medical issue;…

…The response that miscarriage isn’t a problem because it’s natural doesn’t work for two reasons. First, the fact it is natural doesn’t alter either of the points made in stage 4 of the argument above. Second, all sorts of medical conditions are “natural” – cancer, for example, or ALS (motor neuron disease) – but this doesn’t for a moment mean that we don’t consider them to be serious medical issues.

Indeed, even if miscarriage were “natural”, as I mentioned in my last post, our entire medical establishment and indeed all of civilization exists as a bulwark against the evils of the natural world, among which can be found cancer, heart disease, infectious epidemics, and all sorts of rare genetic disorders.  People who believe that life begins at conception should believe that miscarriage represents the most pressing medical issue of our time.  Nevertheless, I offered an extensive (and sometimes rather bitter) argument against this interpretation in the comments to Stangroom’s post, The Ethics of a Non-viable Pregnancy, which Stangroom has since addressed as follows:

I have been arguing that there is a strange asymmetricality in the attitudes of some people towards abortion and miscarriage (they are against abortion, but are not particularly troubled by the phenomenon of miscarriage).

Christopher Carr has put forward the following argument which aims to show that given a particular sort of view there might actually be nothing odd about this asymmetricality:

“If the value is a potential value, then miscarriage may not mean anything, or it may mean a literal fraction of what it would mean were the fetus further along in development or already born. For example, for a particular individual moral framework perhaps a human fetus is worth “one”, a human infant is worth “two”, a human baby is worth “three”, and thereafter all have reached full personhood of “four”. We would still consider miscarriage to be the destruction of valuable human entities, but we may be allowed to think that an eight-year-old child dying of leukemia represents a greater destruction of value. How we assign respective values depending on cause of death and degree of personhood could give us radically different results as to how we evaluate the medical seriousness of miscarriage in relation to other causes of death.”…

…Christopher’s argument here is relevant in that it does impact on how serious an issue some people should view miscarriage to be.

I don’t like the fact that I made a utilitarian argument, because I think in the case of abortion especially, it is likely that errors exceed our best estimations of moral value; that is to say that we have no idea what we’re talking about when it comes to assigning quantitative moral value to humans at various stages of development.  But for the sake of argument, let’s continue with this utilitarian framework (and it is notably the same framework adopted by the Supreme Court in Roe v. Wade allowing abortion, as well as that adopted by the Catholic Church in coming to the opposite conclusion).  I’d like to expand on it and see where it leads.

The following (admittedly arbitrary) numbers represent moral values of [right to self-determination (right to choose); fetus at conception, fetus at six weeks, fetus at six months, newborn infant, six-week old baby, full personhood at six months old] plus the logical conclusions about abortion which must necessarily follow from this array of potential attachments of arbitrary value to arbitrary categories of personhood.

For the following framework, the right-to-choose is valued as a scalar, unchanging quantity, and the degree to which personhood accrues as a function of time is measured in terms of percentage of staticness, with the corresponding moral values in brackets.  Ties go to anti-abortion crowd by convention (although I will stress again that it is only the framework and not the numbers that is significant):

A. null value of right to choose; 100% static personhood: [0; 1, 1, 1, 1, 1, 1] – abortion is immoral in all cases.

B. null value of right to choose; 50% static personhood: [0; 0, 0.5, 1, 1.5, 2, 2.5] – abortion is immoral in all cases.

C. null value of right to choose; 0% static personhood: [0; 0, 1, 2, 3, 4, 5] – abortion is immoral in all cases.

D. right to choose valued at one; 100% static personhood [1; 1, 1, 1, 1, 1, 1] – abortion is immoral in all cases.

E. right to choose valued at one; 50% static personhood [1; 0, 0.5, 1, 1.5, 2. 2.5] – abortion becomes immoral sometime between six weeks and six months.

F. right to choose valued at one; 0% static personhood [1; 0, 1, 2, 3, 4, 5] – abortion becomes immoral sometime between conception and six weeks.

G. right to choose valued at two; 100% static personhood [2; 1, 1, 1, 1, 1, 1] – murdering our own children is justified in all cases where we feel like it, since the right to choose has more moral value than human life (the de facto moral position of ancient Rome).

H. right to choose valued at two; 50% static personhood [2; 0, 0.5, 1, 1.5, 2, 2.5] – killing babies becomes immoral sometime between six weeks and six months old

I. right to choose valued at two; 0% static personhood [2; 0, 1, 2, 3, 4, 5] – abortion becomes immoral sometime between six weeks and six months.

Here is all that rendered in the kind of matrix we often see in game theory:

Given this framework, those identifying themselves as “pro-life” would be found in boxes A., B., C., and D.  Those identifying themselves as “pro-choice” would be found in boxes E., F., and I.  Those advocating rights to murder one’s own children (I don’t think there are any Romans reading.) would be found in boxes G., and H.

What I have put down here is of course just a set of several arbitrary examples in what is doubtlessly a continuum.  Nevertheless, I feel I have represented in good faith (in kind at least, since mathematically only the ratios matter) all possible utilitarian frameworks for evaluating the morality of abortion, and since I have done so, that eliminates the standard critique of utilitarianism – that it does not represent many possible scenarios.   Perhaps fittingly, depending on what numbers one chooses, any conclusion whatsoever could be reached using this framework.

What is most surprising about this ratiocinization (and the reason why I included two scenarios where moral value accrues as a function of development) are the meta-implications.  As one moves from the top of the matrix to the bottom – from attaching no value to the right of self-determination to attaching a maximum value to the right of self-determination (as far as it being legally permissible to murder one’s children) – the logical implications trend in the direction of liberal attitudes towards abortion.  No surprises there.

Nevertheless, as one moves from the left side of the matrix to the right side – from the direction of all life being of the same worth to an extreme accruement of moral value as a function of development – the logical implications trend in the direction of restrictions on abortion laws.

That is to say that there is a clear dominant strategy for the the perfect anti-abortion crusader, which is to minimize the value of the right to choose and to prefer strongly gradiential values attached to humans at different stages of devlopment, and this dominant strategy holds for all scenarios.  For pro-lifers, acknowledging that there is a continuum of moral value in terms of personhood is the clear rational choice from a game-theoretical perspective.  Keep that in mind as I address Stangroom’s criticisms of my argument one by one:

1. It’s worth starting off by saying that although it is a common view that the value of human life increases as the human entity develops from embryo through to full personhood, it is a very uncommon view amongst people who are opposed to all abortion (the overwhelming majority of whom think an embryo has equal intrinsic value to a human adult just by virtue of being biologically human); indeed, if you read the anti-abortion literature, you’ll find that many people deny that things such as sentience and personhood are the measure of value, precisely because this opens up the possibility that an embryo/foetus does not have the value of a human adult;

I think it’s safe to say as objectively as possible that a fetus does not carry the same moral value as an adult, and this is an “is” argument principally and not an “ought” argument: whenever fathers are faced with a choice during delivery complications to save the life of the mother or the life of the baby in such rare circumstances when this is called for, they almost always choose to save the mother, and indeed, society disapproves of fathers who fail to choose the mother’s life over the baby’s.

The interesting question is, why does the anti-abortion movement pretend that this is not the case?  Negotiation theory offers an explanation: it seems likely that anti-abortion people would object to a hierarchy of moral value correlative with development on the grounds that this would inherently justify abortion.  Admitting that a fetus is worth less than a “person” is a slippery slope to infanticide, some might think.  We must draw a line in the sand.

This seems to me to be a fairly reasonable argument for deontologist anti-abortionists to take.  For the sake of power in a binary negotiation scenario, the pro-life right has refused to give an inch, even comparing abortion to slavery and the holocaust.  This argument seems Macchiavellian, but it makes sense: force any compromise in your direction by defining the middle in relation to your extreme position.

Nevertheless, my matrix seems to suggest the exact opposite: that belief in a clear accruement of moral value as a function of development is conducive to restrictions on abortion laws.

2. Setting this point aside, the question remains as to whether it makes sense to see the value of human lives as being incremental, and yet to be resolutely opposed to abortion. This is actually quite a difficult question to answer, since it means weighing up the value attached to the embryo/foetus against the right to self-determination. However, to put Christopher’s argument in its best light, let’s say there isn’t a problem here: that it is possible to think that even the relativelylow value attached to the embryo/foetus in the early stages of pregnancy trumps our right to self-determination (this isn’t an incoherent view, by any means, it’s just it runs into difficulties when held up against the way that de facto we tend to treat the right to self-determination);

This would represent somewhere between scenario C. and scenario I., which is where I generally find myself.  Of course there is a moral value attached to the right to choose whether or not to terminate a pregnancy; the real question is, how does that moral right stack up to the right of the fetus to life at the particular stage of development?  Since the moral value of the right to self-determination is not a function of anything, but what we might consider a natural right, it does not change.

3. So how should this person view miscarriage? Well, it is true that they should not see any individual miscarriage as being as great a tragedy as the death of a human person (because embryos/foetuses don’t have the value of a fully developed human being). But equally, it is true that they can’t see any individual miscarriage as being morally neutral (even if we’re focusing only on the death of the embryo/foetus): they are required to think that miscarriage is a bad thing, in and of itself, because it means the ending of the life of something of value (albeit relatively low value);

Of course I agree, although quantifying this in my utilitarian framework above would require an incredible amount of knowledge, a lot of processing power, and insignificant rates of error.  But, let’s say that I multiply the number of deaths-by-cause from Stangroom’s experiment by the said personhood indices of scenarios C., F., or I. above.  I may find that the deaths from stroke represent a loss of moral value of 1,200,000 while the deaths from miscarriage represent a loss of moral value of 150,000.  Given these results, clearly stroke is a more pressing medical concern for society than miscarriage.  Yet my own opposition to abortion remains entirely coherent and consistent.

4. More than that, though, given that the right to self-determination is not trivial, and given that they think the value attached to the embryo/foetus trumps our right to self-determination, then it follows that they can’t think the value attached to the embryo/foetus is merely trivial. We can’t quantify this value, obviously, but we know it’s enough to rule out abortion, even though this means denying the right we have to self-determination. (The counter-argument here that the right to self-determination is trivial or unimportant isn’t persuasive because de facto we don’t think it is trivial or unimportant.)

I think I have already addressed this argument in good faith.

5. At this point, it becomes (partly) a numbers game. The sheer volume of miscarriages has to be factored into the argument. Okay, it’s true that one might think the numbers of people suffering cancer and heart disease means that these diseases/conditions are a more serious issue than miscarriage (given the fact that these conditions affect human persons to whom there is a higher value attached than to embryos/foetuses). But it is much less plausible if we’re talking about conditions such as kidney failure, Parkinson’s, ALS, etc., all of which we do consider to be serious medical issues. (There is a complication here to do with the possibility that one might attach arbitrary levels of value to the foetus, human person, etc., so that it turns out that just one human death is a greater tragedy than the deaths of say a million foetuses even though the value of a foetus is enough to rule out abortion…

I agree completely with Stangroom’s assessment here, and the fact that it is entirely a numbers game leads us into the lair of the relativist.  Relativism is an invincibility cloak or the equivalent of taking one’s ball and going home, depending on one’s philosophical perspective – and that is what the abortion debate seems to boil down to: personal, arbitrary, subjective assignments of value usually correlated with the particular moral upbringing which one received.

I don’t think it can ever be anything more objective than that for reasons already articulated.  This results in a public policy debate where the only certain conclusions are that I and mine do what we want to do, and you and yours do what you want to do, but is that such a bad thing?

Nevertheless, where the implications of this matter for public policy is that looking at scenarios E. and I. reveals that my framework counterintuitively suggests that placing fully gradiential value on human life at different stages of development results in a stronger argument against abortion rights.  As I have argued, it would behoove the pro-life crowd to embrace such a utilitarian framework.

It is often said that perfect is the enemy of the good.  Were pro-lifers to change tactics, embrace the framework I have provided above, and focus on reducing abortions performed at the margins, we could as a society engage in good-faith, dispassionate discourse on the subject of abortion, progress beyond political posturing and rhetoric concerning abortion, put an end to tasteless comparisons of abortion to slavery or the holocaust, offer good arguments for non-religious opposition to abortion, move in the direction of restrictions on abortion procedures, and continue to greatly reduce overall numbers.