On Judith Jarvis Thomson's "A Defense of Abortion:" Part 1

Judith Jarvis Thomson's essay “A Defense of Abortion” is an oddly-titled essay as there are very few defenses made. It is full of claims to be sure, but very little in the ways of actual philosophical arguments. What Thomson does instead is provide analogous hypothetical situations that she assumes require no further argument, that if the reader is simply presented with her examples the truth of the matter will be plainly visible. But if the truth of the matter were so plainly visible that all one needs to see it is to be presented, sans argument, with an analogous situation, then one wonders why there would be any controversy over the issue at all. As there is controversy, it is worthwhile to examine her essay and see whether her defense actually works.

Thomson starts her essay by addressing what she asserts to be the pro-life position. She claims that this section is ultimately irrelevant to her defense, but as it makes up a significant amount of her essay and its arguments are often referenced in the abortion debate it is worth addressing. According to her the pro-life argument goes like this: at some point between conception and birth the zygote, or embryo, or fetus obviously starts being a person. To pick any particular point after conception as the point personhood starts is to choose an arbitrary point, therefore we ought to suppose, or at least treat, conception as the point at which personhood starts. This, Thomson claims, is a slippery slope argument. She argues that a “clump of cells” is no more a person than an acorn is an oak tree. Now, one might wonder what exactly Thomson thinks an acorn is if it is not the same plant as the adult oak tree? It seems perfectly reasonable to think that an oak tree is merely the mature version of the same plant that starts as an acorn. Regardless, this is not a slippery slope argument by any means, an accusation Thomson doesn't try to prove. There are no logical inconsistencies or philosophically repugnant consequences of thinking that if we are unsure as to whether or not a living being is a person, we should treat it as such. If the military were testing a bomb and they saw on their radar something in the test area that could be a person, there is no one who would think that they shouldn't bother stopping and finding out before continuing.


The argument as Thomson describes it is that:

1. The start of personhood must be a non-arbitrary point.

2. There is no non-arbitrary point between birth and conception when the cells in question becomes a person.

3. The collection of cells is a person by the time of birth.

4. Therefore personhood must start at conception.


This is not a slippery slope argument. She can deny the premise that there are no non-arbitrary points between conception and birth, but formally the argument is perfectly valid. If a zygote is not a person, and a baby is, then there must be some point between the two at which personhood starts. If Thomson wants to claim that there is some non-arbitrary point at which that happens, then it is on her to demonstrate it. If we have no reason for thinking otherwise then we are morally obligated to assume that the living being which we know is a person at one point in its life is a person at all points in its life. This is not an argument that we have a moral obligation because we have epistemological certainty of when personhood begins, it is an argument that we have a moral obligation because we don't have epistomological certainty.

Denying that the unborn have personhood however is not the main thesis of her essay. Thomson claims that even if one grants the premise that personhood starts at conception, abortion is still defensible. Let us then examine the rest of the essay and see how well her argument stands.

One of the first thing that jumps out at the reader is the strangeness of Thomson's analogies for pregnancy. The analogy of the violinist hooked up to you against your will is in theory possible, but is exceptionally outlandish. The analogy of the people-seeds that float through the countryside has crossed beyond the outlandish and into the bizarre. Now an analogy being outlandish or bizarre doesn't invalidate it, but it does make it more than a little suspect. If we are talking about a common and ordinary occurrence then one would think that we would be able to come up with a common and ordinary analogy. Pregnancy is very common and very ordinary. Every female ancestor of every single person ever born underwent pregnancy, often more than once. So when the best analogies Thomson can come up with are rogue music appreciation clubs and floating people-seeds it is reasonable to wonder why she couldn't use analogies a little more mundane, and to want to take a closer look at them and see if they're really above board.

Thomson uses her analogies as a way to raise questions about her opponent’s views, and then treats those questions as counter-arguments. The problem is that simply raising a question does not alone disprove a position. If someone claims that life exist on Pluto, then one might ask, “If life exist on Pluto then they must not be breathing oxygen, so what are they breathing?” It's a good question, and it would also be a difficult one to answer. But it doesn't actually disprove anything. One could say, “aliens can't exist on Pluto because there's no oxygen, and all life requires oxygen.” This would at least be a valid argument. But Thomson does not do this. She simply assumes that such arguments follow from the questions she raises. If someone says, “I think life exists on Pluto because we have recorded structures that seem artificial on its surface,” then it is not a counter-argument to say “Well if there is life then we don't know what they breathe.” It is simply an unanswered question. Raising an unanswered question is not the same as proving something impossible. Thomson's violinist analogy is a perfect example of this. She claims that the pro-life advocates argue that the right to life outweighs the right to decide what happens to one's body, a kind of “right-conflict thesis.” But, she says, what about a case where someone is forcefully hooked up to a violinist in order to keep him alive? There it would seem that the violinist's right to life doesn't outweigh your right to freedom. This is an interesting scenario, and would raise interesting questions for the pro-life position (had that actually been the pro-life position; more on this later). But it does not actually prove it wrong, in fact it's more of a slippery slope argument than the pro-life argument she presented was, as it is claiming that it follows that if abandoning the violinist if justified then abortion must be justified, without providing any actual reason think so. The pro-life argument she gave doesn't make less sense due to the counter-example she offered. The moral principle that the pro-life argument appeals to, that the right to life is stronger than the right to bodily autonomy, is not overturned by her example. It might well prove that there's more to the issue than the rights-conflict thesis alone explains, but it doesn't alone prove the argument false. The violinist hypothetical is different from pregnancy, and Thomson needs to explain why those differences don’t matter if she wants to claim the analogy is valid. An analogy is not a replacement for an argument, it is a way of explaining a situation or concept. And all analogies are necessarily imperfect, they all have elements that make them different from what they are analogous to. They require explanations for why they apply to the issue at hand, and why their differences do not invalidate the comparison. It is these differences Thomson fails to address, and these differences that make her argument a failure.

People-seeds, felonious music lovers, babies that grow to gigantic proportions in inescapable houses, why does Thomson rely on such outlandish inventions? Why would anyone use these analogies at all? The justification might be that by setting up a reality of her own creation it allows her to keep the analogy simple and to the point. But there is nothing simple about her analogies. There is nothing simple about imagining a reality where reproduction works through “people-seeds” that float through the air. So what reason could she have for using it? If a magician is flourishing with his right hand it's because he doesn't want you to look at his left. Magicians want us to watch the flashiness of the magic, and ignore the ordinariness of the trick. People-seeds and terrorist violinists are very flashy. Pregnancy is very ordinary. And it is precisely this quality of ordinariness, which is present in pregnancy and absent in the analogies which render all Thomson's examples invalid. One can justify any extreme action if they invent an extreme enough situation. But pregnancy is not an extreme situation, it is a normal one. Not simply common, normal. People-seeds would not be normal, even if they were common. Pregnancy on the other hand is one of the most normal and natural acts possible for a human being.

It is in this normality that we find the key difference between Thomson's fictions and the reality of pregnancy. Like the magician's sleight of hand, her analogies only work if she includes the flourish. Killing can be justified in a situation as extreme as a kidnapping, and in a situation as bizarre and nonsensical as “people-seeds” it can at least easily be imagined as justified, because we have absolutely no context with which to analyze our situation or understanding of how any of it works, and it is easy to imagine possible justifications for almost any action if the context is obscure enough. If pregnancy was unnatural like people-seeds or evil like kidnappings, then it would be easy to deny any bonds of responsibility between a mother and a child. Thomson tries to pretend the abortion issue is about conflicting rights, but rights do not conflict in normal and natural relationships. Rights are only in competition when both parties are in competition. While this may be the case with felonious violinists, it is not the case with pregnancy. A parent and child are not conflicting parties. The strangest world Thomson creates in her essay is not one of people-seeds, but rather one where human rights play a game of survival of the fittest, with the rights of one person trying to overwhelm those of another. If a relationship is normal and natural, like the relationship between a mother and her child, then their rights should align. The relationship between a kidnapper and their victim is not a normal or natural one, it is an evil one. The relationship between a random person and a random “people seed” might not be evil, but it is at least arbitrary. But to argue that the relationship between a mother and her child is either evil or arbitrary is a complete denial of reality.

Part 2 can be found here.