Sunday, 6 September 2020

Thought Experiments and the Ethics of Abortion (2020)

From The Philosopher, Volume CVIII No. 2 Autumn 2020

      Thought Experiments and the Ethics of Abortion

By Keith Tidman



It is almost fifty years now since American moral philosopher, Judith Jarvis Thomson, published her signature argument in favour of abortion rights, pointedly titled ‘A Defense of Abortion’. In what was very much a novelty for the time, she depended heavily, though not exclusively, on thought experiments, building her argument around them to favour her unambiguous position on this complex subject. Yet, can something as consequential as the ethics of abortion hinge so largely on conclusions drawn from imaginary scenarios? In this essay, I focus on examining the merits of her approach in the context of one of practical philosophys most bitterly contested topics while also hoping to shed some light on her particular arguments.

Thought experiments are essentially short stories, works of fiction, in which experimenters are free to set the starting conditions and tweak the parameters before stepping back and observing events. It has sometimes been objected that in so doing the technique encourages people to assume the outcome at the outset. However, thought experiments depend more on logical coherence to advance their case, than the practical feasibility or plausibility of the experiments’ scenarios, which are seen as immaterial. This seems the case with Thomson’s thought experiments, a feature of thought experiments that some people judge as a weakness, whereas others view it a strength.

With Thomson’s central thought experiments, we should note one ostensible concession that she begins with — on a central issue. This is to grant opponents of abortion that human life starts at the moment of conception and seamlessly develops toward being first an infant and then a toddler, teen, and adult. This new life transitions from being a human, in foetal form, in the womb to being a baby born into the world. However, Thomson makes this concession with hesitancy, and with a curious twist she springs on readers at the end of her ‘defense of abortion’, which I’ll get to shortly.

Thomson first, briefly, suggests that a human being at conception is equivalent to an acorn. Her point is that an acorn is no more an oak tree than the foetus is a human. Thought experiments are all about assumed equivalencies, and although surely this one is offered largely unsupported and perhaps too soon in Thomson’s paper, I suggest that this challengeable analogy is the first of several misleading equivalencies she makes. 


Thomson does seem prepared to acknowledge, however, that it is impossible to draw an arbitrary line between when a collection of cells in the womb is not a person and when it suddenly, seemingly overnight, is. What one might regard as an arbitrary point of transition from not-human to human, made for the convenience of laws, which often require reducing matters to such lines in the sand. But, unsurprisingly, might not that line keep shifting according to what we scientifically understand to be sufficient to call something a human or person, thereby underscoring all the more the arbitrariness and randomness of such decisions?

In her desire to move beyond these problems, and get to the heart of her case, Thomson writes, ‘I’m inclined to think also that we shall probably have to agree that the foetus has become a human person well before birth’. The comment reveals that Thomson appears literal when her famous thought experiment explicitly compares the right to life of the foetus to that of a full-grown human being.



At the same time, given this, Thomson’s seeming afterthought that she was only ‘pretending’ that the foetus is a human being at conception may strike one as disingenuous, when it appears in off-putting fashion in the concluding two sentences of her article. Some object that the admission undermines the soundness and continuity of her original hypotheses, pinned on her earlier human-at-conception precept. Yet, the strategy could be argued to make sense in terms of clarifying one aspect of an issue. To use the language of experiments generally, Thomson eliminates one ‘variable’ so as to test another. And so, Thomson continues that ‘for the sake of argument, we allow the premise … that the foetus is a person at the moment of conception’. She offers this sizable concession to the right-to-life movement in order to move on and put to the test whether abortion is morally permissible even given the concession.

Thomson then addresses the core issue whether, given a general assumption granted to every human, of a right to life, an unborn baby in the womb must have that right, too. This is the point at which she introduces her first, and most famous, ‘violinist’ thought experiment. Thomson asks us to imagine a scenario where someone — she suggests ‘you’, the reader — wakes up to find herself in a hospital bed having, involuntarily and unknowingly, been medically hooked up in the night to a famous violinist. A situation, she says for playful effect, arranged by ‘The Society of Music Lovers’ who have ‘kidnapped’ you. The violinist’s life now depends on your kidneys cleaning the violinist’s blood of toxins. And that if the doctors were to unplug you, it would result in the violinist’s death. You’re asked, therefore, to stay attached for nine months, after which time the violinist will have become cured.

However, further important distinctions are buried in this curious scene. First is the violation of all semblance of responsible social, medical, and ethical protocols in the woman being attached, in this manner, to the violinist. Without any say-so. And the second distinction is between the notions of ‘killing’ and ‘letting die’. The woman, by opting to be unplugged, is letting the violinist die from what would amount to natural causes (diseased kidneys); she is not killing him in the active sense of, say, shooting him in the head or suffocating him with a pillow. This is an important, everyday distinction in matters of law. However, in my view, when Thomson asserts that ‘in unplugging yourself, you are killing’, she conflates killing and letting die. For opponents of abortion, terminations of pregnancies are a different kind of act from passively ‘letting die’ described in the scenario.

To build on her point, Thomson equates the woman’s involuntary tethering to the violinist as resembling a woman having become pregnant as a result of rape. Yet, even in these circumstances some people maintain that the foetus still has a special status, living as it is in the woman’s uterus. They contend that even in the case of rape, there’s a uniquely biological, mother–feotus relationship that must be factored in for ethical reasons. Thomson, however, goes on to declare unequivocally that ‘unborn persons whose existence is due to rape have no right to the use of their mothers’ bodies, and thus that aborting them is not depriving them of anything they have a right to’. But do we have to accept the reasoning, and how does the imaginary example of the violinist support Thomson’s argument?

One might argue, to the contrary, that the woman and human foetus share an unusually mutual bond, resulting in the foetus having a unique status in the woman’s life that no tethered violinist, no matter how brilliant a musician he or she might be, has any realistic hope of matching. In that vein, the woman has an extraordinary responsibility toward what Thomson refers to, in no uncertain terms, as ‘the unborn person’. As such, the human foetus perhaps has a special claim to the mother’s support, whereas the violinist, a stranger, doesn’t. What’s telling is Thomson’s use of the word ‘mother’ in her discussion here, with that word’s layered connotation and emotion. Usage that differs markedly from the neutral word ‘woman’, which one may use more typically to convey gender identification.

At the heart of this discussion are questions about how much sovereignty a person has over one’s body. The common claim by people who favour the right to choose is that women have the unbridled right to decide matters affecting their own bodies, and that such matters include decisions about aborting an unwanted foetus. That is, if a woman has autonomy in ‘deciding what happened’ to her body in the case of abortion rights, does that sovereignty over her body have any limits — limits that society might otherwise impose not just on women but men, too? And if there are limits as to what’s permissible, who or what spells out the constraints, and on what ethical, sociological, religious, or other grounds?

To return to the violinist example, if the woman did sacrifice months of her life to his support, might he, in the name of sovereignty over his body, take up the option to unplug himself? We would allow that, and yet we might consider the woman to have been unreasonably exploited, her having been denied a ‘say’ in his decision.

The situation changes when Thomson introduces the case of a woman’s life that (for some unspecified medical reason) is at high risk, and whether she ought carry her baby to term. At this point, the calculus changes, in that both the woman and the foetus have equal value as human. Or, as Thomson better puts it, ‘Both are innocent’, whereby consequentially neither the mother nor the human foetus has a ‘weightier’ right to life than the other.

The right to life that I mention above leads to one of Thomson’s other thought experiments. She presumes that the right to life ‘includes a right to be given the bare minimum one needs for continued life’. She proceeds by introducing the exception that ‘the bare minimum’ may be something the person actually has no right to be given. Thomson’s colourful, if implausible and peculiar, example is that in order to save her life, she needs ‘the touch of Henry Fonda’s cool hand on my fevered brow’.

Thomson asserts that she has no such right of expectation, even if it would be a nice gesture on Fonda’s behalf to offer such. And that, further, no one — no third party — has the right to force Henry Fonda to travel across country and do so. She ties this line of thinking to her violinist example, saying, ‘nobody has any right to use your kidneys unless you give him this right’. In short, arguing that a right to life does not mean that a person therefore has the unconditional right to use of another person’s body. Although, as presented earlier, some people regard the human foetus as an exceptional case, for reasons related to the unique relationship he or she arguably has with the mother.


In one thought experiment, so-called ‘people-seeds’ float in the air, on rare occasions coming through a window and implanting themselves in the carpets or on the surfaces of furniture to result in human foetuses and, in nine months, infants.


Thomson proceeds to describe two parallel, related thought experiments, whose purpose is the same: to question whether abortion is permissible in the case of accidental pregnancy. One thought experiment is where a woman opens a window in order to air out the room. A burglar takes advantage of the open window and climbs into the house, and attacks her leaving the woman pregnant. The matter turns on whether the woman is at least partially responsible for the break-in, based on her clear-cut exercise of free will and intentionality in choosing to open the window despite the risks, thereby giving the burglar ‘a right to the use of her house’ and, metaphorically speaking anyway, her body. The notion of so-called ‘contributory negligence’ is an issue in real-life criminal cases of assault and rape, but this example seems somehow too flimsy and tangential to illuminate the discussion.

The other thought experiment is where so-called ‘people-seeds’ float in the air, on rare occasions coming through a window and implanting themselves in the carpets or on the surfaces of furniture to result in human foetuses and, in nine months, infants. To protect herself against such an occurrence, as the woman doesn’t want a child, she installs mesh on the windows, yet the people-seeds manage to get in through a defect in the mesh and take root, with the result that one of them becomes a baby. The supposed parallel being, of course, between the failed mesh and failed contraception.

The seed-people thought experiment has a distracting ring of science fiction to it, however. It suggests again that a necessary underlying condition of thought experiments might be in their hewing close to what’s plausible in the real world, their allowable imaginative creativity notwithstanding. If some approximation to reality is desired, I found this particular thought experiment to falter, bordering on the bizarre.

However, putting the peculiarity of Thomson’s implausible thought experiment aside momentarily, if seed-people were really to exist, and the woman knew of their existence and the risks they pose, then there would be an interesting question to be asked over whether or not the woman would indeed have a responsibility toward the resulting human foetus and baby. There’s only so much responsibility with life’s conditions that one might legitimately remove from her shoulders because of either ignorance or disregard of safety measures. To borrow Thomson’s words, ‘there are some cases in which the unborn person has the right to the use of its mother’s body, and therefore some cases in which abortion is unjust killing’. This might be one of them.

However, in returning to the violinist thought experiment, Thomson now adds a wrinkle. She asks whether it makes a difference if the choice is to allow or disallow the violinist’s use of your kidneys by staying tethered to you for an hour instead of the original nine months. All the other conditions, particularly the fact that the hook-up was unknowingly performed while you slept, remain in place. Further, the presumption, justified or otherwise, is that there is no risk of harm to you. Thomson concludes, ‘It seems plain to me you ought to allow him to use your kidneys for that hour — it would be indecent to refuse’. But some people may not be persuaded, instead believing that the professed quaintness of ‘indecency’ — or the contrivance of a shortened timeframe — is not the real issue.

This question of a sliding scale, or ‘matter of degree’, of morality, comes up in Thomson’s abrupt switch to the case of the so-called ‘good Samaritan’ and the ‘minimally decent Samaritan’. The well-known ethical story of the good Samaritan concerns a person from that region who comes upon a man who has been attacked by thieves, leaving him stripped and half dead. Unlike earlier passersby who did nothing to help the man and went on their way (thus not acting even ‘minimally decently’), the Samaritan goes to unsparing lengths, at his own expense, to ease the man’s plight.

Thomson then revisits the mid-1960s murder of Kitty Genovese in New York, who was attacked and killed while some three dozen people heard the victim’s screams during the knife attack yet none rushed out to help Ms. Genovese nor called the police — which would be the ‘minimally decent Samaritan’ model. It may well be easy to agree with Thomson that ‘it was not morally required’ of any of the witnesses to run out and give direct assistance at grave risk to their own life. But surely, at the very least, there was a moral and humanitarian obligation to alert the police.

This discussion is at first couched in terms of whether, at the moment in history Thomson published her paper, fifty years ago, laws regarding abortion required women to be not just ‘good Samaritans’, but ‘very good Samaritans’, in sustaining the life of all unborn persons. But she doesn’t delve further into that subject. Instead, Thomson pivots to matters of morality. She asks ‘whether we must accede to a situation in which somebody is being compelled — by nature, perhaps — to be a good Samaritan’.

Thomson concludes by clarifying that although she ‘does not argue that abortion is not impermissible, I do not argue it is always permissible’. Accordingly, she notes that ‘minimally decent Samaritan’ principles likely apply in certain situations that propel the woman to bring the ‘child’, as she calls it, to term, and ‘in other cases resort to abortion is even positively indecent’.

The debate hinges too on whether, up to a certain point of maturity, the human foetus will not survive outside the ‘mother’s body’, even with the best of modern medical intervention. Significantly, over the decades such intervention has shifted survivability earlier and earlier. This is the ‘viability’ debate. Yet the issue is not easily reduced to mere practicality. After all, there are many other circumstances where human survivability is similarly unlikely without intervention, whether by familial, medical, or other involvement, such as newly born infants, some elderly people, and disabled or seriously ill members of society. These groups are not considered disposable, as worthless. It seems inconsistent to apply independent survivability as a morally acceptable, even sufficient, determinant of whom to protect.

Ultimately, I suspect that conclusive answers to the questions posed here, to confidently put one’s thumb on the scale in this tangle of rights and ethics, must reach deeper than many thought experiments, even as creative ‘works of fiction’ like Thomson’s, might be able to go. The fraught history of this subject, with both sides unrelenting in their claim to rectitude, doesn’t augur clear-cut resolution in the measurable future, inventive thought experiments notwithstanding — which may simply come down to whose partisan thumb happens temporarily to be on the political scale at any moment in time. What might seem beyond debate now may not prove to be forever.

Writing in 2020, with a presidential election looming, considerations in the United States around the right-to-life and right-to-choose factions have taken on decidedly ramped-up urgency and stridency, as each side jostles for state, congressional, and Supreme Court advantage. The result being that what had long been presumed as settled law — by a historical, landmark Supreme Court decision in 1973, just two years after Thomson’s paper was published— is again in the crosshairs, as fair for re-litigation. The U.S. presidential race has spotlighted the stakes couched by the arguments launched and deflected by both sides, with everyone fully anticipating the rancour to remain long after the election has been resolved. The moral stakes and heightened passions will make the issue even harder to discuss let alone resolve.

Except, perhaps, through materially groundbreaking shifts in the exacting and imaginary world of thought experiments to reframe the deliberation.


Editor's note.  Thomson’s A Defense of Abortion’ was first published in the journal Philosophy & Public Affairs in 1971. The full text can be read here.

About the author: Keith Tidman is a regular contributor to Philosophical Investigations, the sister site of The Philosopher.