Matthew H. Kramer, Torture and Moral Integrity: A Philosophical Enquiry, Oxford University Press, 2014

Uwe Steinhoff
Department of Politics and Public Administration
University of Hong Kong
Hong Kong

The blurb of Matthew Kramer’s book, Torture and Moral Integrity: A Philosophical Enquiry, states that the book “seeks to explain why interrogational and other types of torture are always and everywhere morally wrong.” This might give the prospective reader the impression that the book takes an absolutist stance against torture, but this impression would be misleading. The explanation of the discrepancy between the book’s self-presentation and what it is actually saying lies in the idiosyncratic terminology Kramer employs throughout the book.

Expressed in normal terminology, the book argues for the following: most forms of torture are wrong, are impermissible, but some forms of torture are permissible and infringe absolutely no deontological constraints. Still other forms of torture do infringe deontological constraints, yet they can still be morally justified and permissible as the lesser evil—in fact, they are morally obligatory. Moreover, while most forms of torture should be illegal, some forms of torture should be legal (296), some other forms of torture should always be excused (301), and still some further forms should at best be “punished” somewhat symbolically (296-298). Kramer might protest that this summary of his position distorts what he is saying. But it does not; on the contrary, it is Kramer who distorts the meaning of certain moral expressions by needlessly redefining them. To wit, he first distinguishes between “weak permissibility” and “strong permissibility” and then uses the term “permissibility” throughout the book to refer to the strong kind. He explains: “Some person X is [weakly] permitted to perform some action q if and only if X is not under any overtoppingly stringent obligation to not perform q” (5). And: “X is [strongly] permitted to perform q if and only if X is neither under an overtoppingly stringent obligation to not perform q nor under a non-overtoppingly stringent obligation to not perform q” (5). To understand this, one has to know that: “An overtoppingly stringent moral requirement exceeds in importance all the moral duties that run counter to it, or is unopposed by any competing moral duties. A non-overtoppingly stringent moral requirement R does not exceed in importance all the moral duties that run counter to it” (5). He distinguishes the meaning of his neologism “overtopping” from the meaning of the well-established term “overriding” by claiming (without textual evidence and against appearances) that philosophers who state that some deontological constraint, for example a right, has been overridden thereby mean to say that the overridden constraint ceases to be operative: for example, no compensation would be due to the persons whose rights have been overridden (10-11). He also distinguishes, in the same way, weak and strong forms of “justification” and “rightness,” using the unqualified term only for the strong forms (7-8). Likewise, he distinguishes between weak and strong forms of absolutism (8-9). Here, however, he interestingly and inconsistently uses the unqualified term “absolute” throughout the book for the weak form—which allows him to call himself an “absolutist” regarding the moral prohibition of interrogational torture (to be fair, on occasion he reminds the reader that this “absolutism” is only “weak”).

One of the strange consequences of the deontic logic implied by this terminology is pointed out by Kramer himself: in his framework, moral obligatoriness does not imply moral permissibility (16-18). One could argue that this alone already provides sufficient grounds to reject Kramer’s framework. But there are still further noteworthy consequences that Kramer does not mention. For example, while in common sense morality and in plausible systems of deontic logic “it is impermissible not to do x” implies “it is obligatory to do x,” this is not the case in Kramer’s framework. On the substantive level, moreover, Kramer’s account implies the impermissibility not only of interrogational torture (in the case that there is a deontological constraint against all kinds of interrogational torture: we will return to that), but of all kinds of actions against which there is a deontological constraint. In other words, on Kramer’s understanding of “justification” (which reserves this term for the strong form) there can be no such thing as a threshold-deontological lesser evil justification. For example, there is certainly a deontological constraint against (nonconsensually) pinching the noses of innocent people or against breaking one’s promises. But then, averting a genocide by pinching an innocent person’s nose or by breaking the promise to meet your friend for a coffee would be “absolutely impermissible” on Kramer’s account.

Another fact Kramer does not mention is that his position, which accepts that sometimes not engaging in interrogational torture also violates a duty, and a more stringent duty at that (13, 14, 212-215, 271, 280), thereby implies that not only engaging in interrogational torture is “absolutely” impermissible but refraining from certain kinds of interrogational torture is also “absolutely” impermissible. Thus, Kramer implicitly combines an “absolutist” stance against certain kinds of torture with an “absolutist” stance for those very same kinds of torture. While many readers will find Kramer’s penchant for paradox disconcerting, he himself is unfazed, defending his conflicted position by pointing out that “I ought to φ” does not logically contradict “I ought not to φ.” That is true, of course, but hardly a reason to embrace conflicts of this kind. What, for example, would we say of a law-giver or a morality giver (God?) who said: “I command you to do x, and I command you not to do x, my latter command is more stringent than the former, but the former stands nonetheless.” It would seem that such a law-giver or God, and by extension the command given, is either irrational or sadistic, gleefully enjoying confronting his subjects with impossible tasks. But why should one accept a moral system as valid that seems to be either irrational or sadistic (or at the very least grossly unfair)? Kramer does not offer any explanation for this, but such an explanation is direly needed. Note that the threshold deontological necessity justification does not run into such problems. While acknowledging that a right is “overtopped” and thus infringed, it does not claim that infringing the right is impermissible (as Kramer does). In contrast, the “damned if you do, damned if you don’t” stance of Kramer is akin to Michael Walzer’s (1973) paradoxical “Dirty Hands” position, as Kramer himself implicitly acknowledges at some point (270). Yet, Kramer does not engage the fiercely critical literature on “Dirty Hands” at all.

Kramer’s account also has certain probably rather unwelcome legal implications that he does not mention. Kramer states: “Because interrogational torture is morally wrong in all of its instances (including its morally optimal instances), and because it involves serious violence, the officials in a legal-governmental system cannot properly be allowed to perpetrate such torture with impunity” (275). It is unclear what the reference to violence is doing here. After all, firstly, according to Kramer, there is also psychological torture, and such torture, it would seem, cannot easily be subsumed under the label “violence.” More importantly, we do not punish people only for the use of violence. To wit, we would not only rightly punish a police officer who violently pinches an innocent person’s nose, but also a police officer who non-violently and idly stands by while a culprit he could stop is committing a rape or trying to kill someone. In fact, we would punish the latter police officer more severely. Thus, it is not the violence of an act that deserves punishment, but its wrongness. Given, however, that, as Kramer himself states, officers who opt against “impermissible” but nevertheless “optimal” torture would thereby commit “an even more serious moral wrong of fatal remissness” than they would by opting for it (280), they should accordingly be punished for this remissness, that is, for this “especially grave breach of their public-safety responsibilities (288). Thus, just as Kramer’s account implies, as already noted, a morally “absolutist” position both for and against certain kinds of torture, his account absurdly implies both the legal prohibition of certain kinds of torture and a further legal prohibition on abiding by the former legal prohibition. From his (not logically, but pragmatically) contradictory morality follows contradictory law.

Let us come back to the alleged deontological constraint against interrogational torture. According to Kramer, not all forms of torture are impermissible. The decisive normative difference between what he calls “ephemerally incapacitative torture” (“permissible”) and “calamity-averting interrogational torture” (“optimal” but “impermissible”) is, according to Kramer, that the former aims at making the tortured person commit an omission while the latter aims at making the tortured person commit an act. Kramer makes the act/omission distinction in the way Jonathan Bennett (1997) does. Bennett himself, however, does not think (as Kramer admits, 195) that this distinction by itself has any moral significance. Thus, Kramer has to provide his own argument as to why interrogational torture, aimed at making a person act (namely, say something), is “impermissible” (although sometimes obligatory). His argument boils down to the claim that interrogational torturers will “sull[y] their moral integrity by using another human being’s vulnerability to severe pain as a means of minutely directing his conduct” (196). But: so what? One of the usual police procedures against potentially dangerous suspects is to use those suspects’ vulnerability to fear of death and injury as a means of minutely directing their conduct: the police point their guns at the suspect and order him to lie down on the ground, with the implicit or explicit threat of opening fire if he doesn’t. Is this impermissible and deserving of punishment? It doesn’t seem so. But if it is, one will have difficulties finding people to join the police force, for who will be so irrational to do a job where one’s employer – the state – punishes one for one of the very things the employer requires one to do? Perhaps Kramer will reply that there is a normatively decisive difference between using a person’s vulnerability to pain (of course: being shot is also often painful) on the one hand and using her vulnerability to fear of death or injury on the other. However, first, that seems implausible, and second, Kramer has not provided any argument for such a view; in fact, he does not address the issue.

Moreover, why should “using another human being’s vulnerability to severe pain as a means of minutely directing his conduct” “sully” the torturer’s “moral integrity” in the first place? Because, says Kramer, the interrogational torturer thereby deals with that other person “from a position of godlike dominion” (196). Kramer also talks in this context of “overweening self-aggrandizement” (196), “hubris” (198), “hubristic overreaching” (201), and “superhuman ascendance” (203). Given that Kramer elsewhere warns against the use of hyperbole (169-173), this bombastic rhetoric is rather curious and invites suspicion. To wit, is it really plausible to claim that police officers who put a kidnapper into a painful armlock to make him tell where the kidnapped victim is thereby “arrogate to themselves the role of gods” (212)? Which self-respecting god descends upon interrogation rooms to put people into armlocks? Wouldn’t gods rather elegantly extract the information via telepathy, or already have it via omniscience? And could they not just snap their divine fingers to free the victim? The fact of the matter is that the police officers in the case Kramer is referring to did not arrogate the powers of gods to themselves by torturing; rather, they tortured because they are no gods and know it. Consider also the following (obviously fictitious) case of act-impelling (albeit not interrogational) torture. Frankenstein, a sadist, alters his body in such a way that he can swallow a woman alive and then slowly digest her—his ultimate sadistic fantasy. He acts on it. The woman in his stomach happens to have a pain-inflicting device and uses it on Frankenstein’s stomach walls with the intention that Frankenstein put his finger into his throat to make himself vomit. This is act-impelling torture; the woman tries to “minutely direct” (in Kramer’s sense) Frankenstein’s conduct. But is it really remotely plausible to say of this woman that she arrogates the role of a god to herself because she “act-impellingly” tries to escape the belly and the stomach acid of the beast? It does not seem so.

But let us also consider a case where someone literally acquires godlike powers. Evil God shows Eve an apple, telling her credibly that by eating it she will acquire godlike powers (not surpassing his, though). She will, among many things, be able to destroy galaxies with a mere thought. He credibly tells her that if she does not eat it he will destroy the world. She eats it and becomes like a god. Is her world-saving act of acquiring godlike powers violating a deontological constraint or sullying her moral integrity? Does she deserve criticism or punishment for saving the world in this way—the only way? Does she need to feel guilty for saving the world the only way she could? Intuitively (and outside of the theology seminar, but perhaps not even there), it does not seem so. Conversely, however, it seems that all four of Kramer’s claims here—that the police officers in his example “arrogate to themselves the role of gods,” that doing this “sullies their moral integrity,” that “sullying one’s moral integrity” violates a deontological constraint, and that there is not some other way of sullying one’s moral integrity involved in “ephemerally incapacitative torture”—are indeed mere claims. Kramer’s overblown rhetoric does not cover up the lack of argument for his paradoxical “absolute” prohibition of “optimal” interrogational torture so much as reveal it.

Uwe Steinhoff
Department of Politics and Public Administration
University of Hong Kong
Hong Kong

REFERENCES

Bennett, Jonathan. 1998. The Act Itself. Oxford: Clarendon Press.
Walzer, Michael. 1973. “Political Action: The Problem of Dirty Hands.” Philosophy & Public Affairs 2 (2): 160-180.