UPDATE: I stand by my original post. Let me take on Mr. Taylor's objections one by one.

1. I focused only on a fragment of his argument. That's right - I did. It was the fragment that made the argument I am criticizing in this post: the "fundamental trick" of treating the legal definitions of torture as if they change the colloquial meaning of words into technical terms. The rest of Mr. Taylor's article is not about the topic of my post. His article is, as he says, a "multi-point analysis," and I was focusing on just one of those points. Of course, quoting out of context can sometimes distort the meaning of a sentence or an argument when the context is restored. I don't think that happened here, and Mr. Taylor hasn't said that I got the meaning of his argument wrong on the point I was discussing. Readers may judge for themselves whether I misread Stuart Taylor. I don't think so.

2. He accuses me of ignoring the fact that the U.S. torture statute narrows the definition of "severe mental pain or suffering" to something that differs from the colloquial meaning. Wrong. Far from ignoring this point, I said so explicitly. I wrote: "That's not a narrow or technical definition (although Congress went on to give a narrow definition to the mental pain or suffering part)."

I didn't elaborate, because I don't think that mental torture is the issue in waterboarding. The issue is physical torture, and in particular, the infliction of severe physical suffering (one of the express terms of the torture statute). The main trick used in the Bybee-Yoo memos is to treat "pain or suffering" as a single term ("not distinct concepts of 'pain' as distinguished from 'suffering'"), and then focus on pain (see p. 11 here). But the statute uses two words, not one, and that is because there are forms of physical suffering that aren't pain. "Colloquial" common sense knows this: freezing cold, unbearable heat, itchy, nauseous, ache-all-over, unable to breathe -- all are suffering, none are pain. Unable to breathe: waterboarding is controlled suffocation. Even if suffocation isn't pain, it is a form of severe physical suffering. In a moment, I'll explain how OLC mangled the law to avoid this conclusion. But first, let's talk about the physical suffering of waterboarding.

Here's a famous description of waterboarding by Malcolm Nance, a former master SERE instructor, in a post titled "Waterboarding is Torture...Period." Nance's credentials: "SERE staff were required to undergo the waterboard at its fullest. I was no exception. I have personally led, witnessed and supervised waterboarding of hundreds of people." Here is his description:

Unless you have been strapped down to the board, have endured the agonizing feeling of the water overpowering your gag reflex, and then feel your throat open and allow pint after pint of water to involuntarily fill your lungs, you will not know the meaning of the word.

Waterboarding is a controlled drowning…. It does not simulate drowning, as the lungs are actually filling with water. There is no way to simulate that. The victim is drowning. … A team doctor watches the quantity of water that is ingested and for the physiological signs which show when the drowning effect goes from painful psychological experience, to horrific suffocating punishment to the final death spiral.

Waterboarding is slow motion suffocation with enough time to contemplate the inevitability of black out and expiration –usually the person goes into hysterics on the board.

Although Nance says that waterboarding begins as a "painful psychological experience," his description also includes "the agonizing feeling of the water overpowering your gag reflex." That's physical. Although the version of waterboarding OLC approved would require that it be stopped before "pint after pint of water" fills the lungs, the OLC itself writes, "the water may enter -- and may accumulate in -- the detainee's mouth and nasal cavity, preventing him from breathing. In addition, you have indicated that the detainee as a countermeasure may swallow water, possibly in significant quantities" so that it is necessary to use saline solution to avoid diluting the salt content of the detainee's blood. (Bradbury Memo of May 10, 2005, p. 13). OLC adds: "We understand that the effect of the waterboard is to induce a sensation of drowning. This sensation is based on a deeply rooted physiological response." That's physical too.

How did OLC avoid concluding that waterboarding inflicts severe physical suffering? As I just explained, Yoo and Bybee did it by treating "pain and suffering" as a single term (although Yoo facilitates his argument by misquoting the statute, which says "pain OR suffering" not "pain and suffering"). Yoo does go on to consider "physical suffering" as a separate term, and writes that "the waterboard could not be said to inflict severe suffering. The waterboard is simply a controlled acute episode, lacking the connotation of a protracted period of time generally given to suffering." The Levin Memo - OLC's substitute for the Bybee Memo - borrows this idea and reads into the statute a requirement that isn't there, namely that to be physical suffering an experience must have "some extended duration or persistence." In fact, the statute expressly states that mental suffering must be prolonged to be torture, but it imposes no similar requirement on physical suffering. Levin simply made it up. Nearly four years ago, Marty Lederman spotted this trick and speculated that Levin inserted it because waterboarding is done in short sessions. (The Yoo memo says that "the procedure would not last more than 20 minutes in any one application." When the Bradbury memos were released last year, they proved that Marty was right: Bradbury leaned on the Levin memo to deny that waterboarding imposes physical suffering.

To summarize: Yoo denied that suffocation is physical suffering (merely "a controlled acute episode" - but of what? If not suffering, then what?) because it doesn't last long. Levin builds the duration requirement into the statute even though it is not there. Both of them conjured up law from nowhere - law that not only isn't in the statute or any other legal source, but that the statute rules out. After all, when a statute expressly attaches a duration requirement to one form of torture (mental) but not to the other form (physical), you simply can't read the requirement into the other. Every lawyer learns a rule of statutory construction called "expressio unius," short for a Latin maxim that means "expressing it in one place excludes it from the other." This result mirrors common sense: why would Congress imply a duration requirement into the definition of suffering when everyone knows that severe suffering can happen for very short periods of time? To deny that forty seconds of suffocation is suffering is to deny the obvious.

3. That brings me to Mr. Taylor's criticism that I ignored the evidence of the 10,000 SERE veterans who "have almost unanimously reported that waterboarding caused no severe physical pain and no prolonged mental harm." The first thing to observe is that the source of this information is the classified Yoo memo, which does not say that the SERE veterans "almost unanimously reported" anything at all. It says that an official who trained 10,000 SERE students reported that waterboarding never caused prolonged mental harm, as evidenced by the fact that almost none of them dropped out of the program or required mental health services afterward. A small distinction, perhaps, but there is a difference between one official and the "almost unanimous" testimony of 10,000 military officers. And the Yoo memo, unlike Taylor's article, does not attribute the "no severe pain" conclusion to reports about SERE training. Mr. Taylor's attempt to enlist 10,000 officers claiming that waterboarding wasn't severely painful seems like careless reading and wishful thinking. Perhaps I've overlooked something in the Margolis memo, or perhaps Mr. Taylor has a different version than mine, in which much of the SERE discussion is blacked out.

But more significant is that Mr. Taylor discusses mental pain or suffering and physical pain, but ignores physical suffering -- the category that applies to waterboarding. Margolis makes the same mistake.

4. Mr. Taylor faults me for not talking about the specific intent requirement in the torture definition. Actually, I did mention intent when I described the core definition of torture as the intentional infliction of severe mental or physical pain or suffering. But I did not discuss specific intent. That matters, Taylor argues, because the specific intent requirement greatly narrows the definition of torture beyond the colloquial meaning.

Or does it? True, if I go through a painful medical procedure, I might say "That was torture," even though the doctors did not intend to torture me. (Probably the specific intent requirement was inserted into the statute to rule out cases like medical treatment that doctors know will cause severe pain.) Perhaps colloquial speech accepts the possibility of someone unintentionally torturing someone else - imagine a farmer who turns on the electric fence without realizing that a passerby is sitting on it. I agree that the definition in the torture statute rules out the possibility of unintentional torture.

But we're talking about waterboarding. Mr. Taylor lays great weight on a 2008 opinion by the Third Circuit Court of Appeals (Pierre v. Attorney General, 528 F.3d 180) that gave the same specific intent analysis as the Bybee Memo:

Specific intent requires not just the general intent to accomplish an act with no particular end in mind, but the additional deliberate and conscious purpose of accomplishing a specific and prohibited result. Mere knowledge that a result is substantially certain to follow from one's actions is not sufficient to form the specific intent to torture.

Even if we accept this interpretation, waterboarding meets the test. Strapping someone to a tilted board and pouring water over a cloth covering his mouth is not "an act with no particular end in mind." It has a "conscious purpose," and that purpose is to make the victim suffer. (And please, don't retort that the purpose is to get information or to safeguard national security. Those may be the motives, but the specific purpose of the waterboarding is to make the victim suffer.) A CIA interrogator is not The Accidental Waterboarder.

4. As for the fact that the use of the word 'torture' in the Lee case and the Ashcraft footnote were not part of a judicial holding interpreting a federal statute: true, but irrelevant. Both cases pre-date the torture statutes, so of course they were not holding that as a statutory matter waterboarding is torture. The job of the OLC lawyers was to interpret the language of the statute, and Lee, a federal case that labeled waterboarding "torture," is obviously relevant to an opinion on whether waterboarding is torture. At the time of the Bybee Memo no federal case had ever interpreted the torture statute. (There has been only one prosecution ever under the statute, the 2009 conviction of a Liberian torturer.) My whole point in my previous post was that even though the Lee court used the word "colloquially," there is not the slightest reason to suppose that the torture statute had transformed the meaning of the word.

Looking through Balkinization posts of yesteryear, it seems that Mr. Taylor was making similar arguments two years ago. See Marty Lederman's post here (although Marty unfortunately linked to the wrong Taylor article, and I've been unable to find the one he quotes from). Others have made similar arguments: Lederman skewered them here, here, and here. The arguments don't get better with age.