In a recent piece over at richarddawkins.net, Richard Dawkins has tried to put Sam Harris’s “defence” of torture into perspective for us. Entitled “It’s What Moral Philosophers Do,” Dawkins argues that Harris has been unjustly vilified for doing what moral philosophers routinely do. He uses an example from P.Z. Myers, regarding the morality of abortion:
We can make all the philosophical and scientific arguments that anyone might want, but ultimately what it all reduces to is a simple question: do women have autonomous control of their bodies or not? Even if I thought embryos were conscious, aware beings writing poetry in the womb (I don’t, and they’re not), I’d have to bow out of any say in the decision the woman bearing responsibility has to make.
In response to this Dawkins justly says:
Now a reasonable person could disagree with him here. A humane rationalist could be pro-abortion under existing conditions, but anti-abortion under the counterfactual condition of the Myers thought experiment – the conscious, poetry-writing embryo. That is the whole reason why Myers found it worthwhile to invent his excellent thought-experiment.
And I think this is a good response to Myers’ thought experiment. If embryos in the womb had conscious life, as well as life plans and projects, then, it seems, abortion would have to be ruled out as morally appropriate. The point of the argument against the attribution of person-defining characteristics to embryos is precisely that they do not have them. The only one with life projects, hopes, fears, and other attributes that belong to persons, in the case of abortion, is the pregnant woman, and that is why it is inappropriate to arbitrarily define personhood in terms that would apply to embryos and foetuses, because doing so results in the abrogation of women’s rights to make decisions regarding their lives freely, and so is an unacceptable harm.
Imagining thought experiments may be what moral philosophers do, but the purpose of making thought experiments is to come to morally relevant conclusions on the basis of them. As a matter of fact, P.Z. Myers’ thought experiment regarding abortion shows clearly why, if the foetus was a conscious, poetry writing being, with the same sorts of hopes and fears and life projects as the woman in whose womb he or she is growing, then abortion would be an unalloyed moral wrong. What surprises me is that P.Z. Myers did not draw the consequences from his thought experiment that it seems intuitively obvious that he should, for the reason that personhood should not be attributed to foetuses is that they do not in fact have the kinds of consciousness and life prospects and projects that the thought experiment attributes to them. That is precisely why Peter Singer can ask about the rights of mature, fully conscious animals, contrasted with adult human beings with dementia. On what grounds do we give preferential moral treatment to mindless human beings, and none at all to adult cows? The anwsers are not clear, and they deserve close attention.
Harris’s argument regarding torture in The End of Faith (see pp. 192-199, “A Loophole for Torquemada?”) is similar to Peter Singer’s argument. What he does is to set up a moral equivalence between two situations, between the collateral damage caused by bombing, and the torture of a terrorist suspect in order to derive information from him. And Harris does not only dream up a thought experiment, he draws conclusions from it:
Because I believe the account offered above is basically sound, [he writes] I believe that I have successfully argued for the use of torture in any circumstance in which we would be willing to cause collateral damage. [198; my italics]
And presumably, in the light of the italicised words, he would be willing to see the institution of torture legalised for precisely those situations. In other words, what Harris does here is to argue that torture is a perfectly legitimate means of obtaining information in the situation, as it was, and perhaps still is in the so-called “war on terrorism”, where we are prepared to accept that dropping bombs or using drones to attack high value terrorist suspects will result in collateral injuries to innocent civilians.
Due to a neurological quirk we are quite prepared to cause harm where the consequences of that harm are not obvious to us, as in the case of dropping bombs from thousands of feet in the air, but that is merely an illusion, Harris says. As he puts it with accustomed exemplary clarity:
Given what many of us believe about the exigencies of our war on terrorism, the practice of torture, in certain circumstances, would seem to be not only permissible but necessary. Still, it does not seem any more acceptable, in ethical terms, than it did before. The reasons for this are, I trust, every bit as neurological as those that give rise to the moon illusion. In fact, there is already some scientific evidence that our ethical intuitions are driven by considerations of proximity and emotional salience of the sort I addressed above. 
Since the chapter in which this claim is made is entitled “The Science of Good and Evil,” it is in good company. But at this point Harris is doing something that moral philosophers do not do. He seems to think that it is possible to short-circuit the process, and to settle a moral conundrum — that is, our intuitive sense that torture is wrong — by adverting to scientific evidence. And, indeed, on the basis of this, he was prepared to say that torture is not only permissible but necessary in certain circumstances.
The question I want to ask — not by way of vilification of Harris — is whether Harris has really exhausted moral argument at the point where he reaches into his scientific bag of tricks. To start with, consider the “Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment.” According to Part I, Article 1, Subsection 1, this reads:
For the purposes of this Convention, torture means any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity. It does not include pain or suffering arising only from, inherent in or incidental to lawful sanctions. [my italics]
(Recall that the United States is a signatory of this Convention.) The terms of the convention are quite clear as to what constitutes torture. The fact that we consider these things to be wrong, and yet tolerate collateral damage in acts of war — though the latitude for such tolerance has increasingly narrowed over the years — is based, not on the proximity or distance between causing harm and our awareness of the harm caused, but on the lawfulness or otherwise of the acts involved in causing that harm.
It is the rule of law that Harris left out of consideration altogether in his thought experiment, and that he continues to leave out, even in his defence of his defence of torture. Now, notice, my point is not that, in some situations, people out of great concern might not justly resort, in extremis, to torture in order to elicit information from a suspect whom they have good reason to suppose is concealing information that may, for example, save the life of a child. This is the example to which Harris directs his readers in his article (to which Dawkins directs us), called “Wrestling the Troll.” Harris points us in the direction of the Stanford Encyclopedia of Philosophy entry under Torture. The author of the piece recounts the story of the theft of a car which had a baby in it, subsequently abandoned, and the captured thief, whose beating by the police extracts the location of the vehicle, and in which the baby is, as a consequence, saved. According to Harris his ”position on torture is more or less identical to the one prescribed in that handbook of evil, the Stanford Encyclopedia of Philosophy,” something you would never have known (he suggests) from reading the article in which he is taken to task, in which he is called one of the 5 most awful atheists around (who ruin it for everyone else).
However, the point about law is still being missed. The Standford Encyclopedia of Philosophy entry under torture makes it very clear that the question of the lawfulness of torture is of great importance. For example, the article says that
[t]he practice of torture is endemic in many, perhaps most, military, police, and correctional institutions in the world today, including democracies such as India. It is only in recent times and with great difficulty that torture in Australian prisons and police services, for example, has been largely eliminated, or at least very significantly reduced.
Even though it is against the law it is still very often resorted to, and not often in the kinds of circumstances considered by Harris. Also of importance is the fact that the ease with which torture is adopted by police departments means that, if legalised, it would have the tendency to spiral out of control. Indeed, according to the Stanford Encyclopedia article,
[i]n the light of the evidence it would be a massive understatement to say that historically the sub-institution of torture—whether in a lawful or unlawful form—has been no stranger to military, police, and correctional institutions. Further, there is now a great deal of empirical evidence that in institutional environments in which torture is routinely practised it has a massive impact on other practices and on moral attitudes. For example, in police organisations in which torture is routinely used the quality of investigations and, in particular, of interviewing of suspects, tends to be low. Careful, logically based, questioning on the basis of the available evidence is replaced by beating up suspects.
This is vital, and it is not mentioned by Harris, who defends himself on the basis of one case where beating up a suspect, provoked as it was by circumstances, actually worked. More important is the fact that Harris wrote his book The End of Faith just at a time when the question of the official use of torture by American authorities was being widely discussed, as well as practiced. Extraordinary rendition to places where suspects could be “legally” tortured was carried out routinely by the CIA, and the use of cultural and sexual humiliation was being widely used in Iraq to extort information from suspects, even though this fact was hidden by blaming it on low ranking service men and women at the Abu Ghraib prison. In that context it cannot be said that Harris’s defence of torture, by saying that it is only an illusion that we disapprove more of torture than we do of collateral damage, was altogether harmless, or, indeed, simply what moral philosophers do.
What Harris needed to do was to go beyond the short-cut scientific route to the establishment of moral equivalence, by considering many other aspects of the issue of torture, what the effect of institutionalising torture would be, and whether, in fact, there is no moral difference between collateral damage, and the deliberate causing of harm to individual persons in the quest of information. For the end of the story should not have come by means of claims about neurology, but by means of a thorough unpacking of the thought experiment, and this was not done. In the end, we are told that the difference between collateral damage and torture is simply neurological. Nothing is said about intention. Nothing is said about the value of international protocols. Nothing is said about the implications of institutionalising torture. Nothing is said about the impossibility of institutionalising collateral damage, or about the fact that increasingly accurate weaponry has reduced — or should have reduced — the amount of collateral damage in war. Nor is anything said about the justification of a “war on terror” in the first place, as if it were quite clear that this was a reasonable thing to undertake, or even to try to conceive of in military terms. My point here is not to vilify Harris, for I admire much that he has written, but I do not think that he is a moral philosopher, and I do think that his misguided belief that moral questions can be answered scientifically, and the correlative tendency to believe that our intuitions can simply be dismissed, on the basis of neurological evidence, as illusions, is warranted. In this case, it seems to me, Harris has been led into a moral cul-de-sac, and it would be helpful if he would try to get out of it.