The constitution and our intervention in Libya

We're now attacking military targets in Libya. I think the cause is just, but of course the outcome is impossible to predict.

What is most striking is that we have entered the fighting without any direct authorization from Congress. There is no declaration of war, and there is no "specific statutory authorization," with which we began the war in Afghanistan and both our recent wars with Iraq. And there is, as far as I'm aware, no substantial objection to this from Congress -- even though apparently public support for this fighting is quite mixed.

I doubt that the Framers of our Constitution intended to grant the President the power to initiate armed conflict on his or her own. Glenn Greenwald in Salon reports that both Obama and now-Secretary of State Hillary Clinton, as Presidential candidates, appeared to say that the President could initiate defensive fighting -- and, as Greenwald says, Libya isn't that. But I think the fact that Obama has acted as he has, and with as little Congressional opposition as he's faced, confirms that today the President does indeed have some power to initiate armed conflict. If we understand the law not as its words' intrinsic meaning or its drafters' intentions, but as what those who apply the law have actually made of it -- that is, if we are "legal realists" -- then this is the law, the constitutional law, today.

Exactly how much power to initiate armed conflict the President has isn't clear, but it is clear that there are limits on this power. Our land wars of the past 20 years have all been authorized by Congress, and I don't see a basis in current practice for saying that the President can start a war of that dimension on his or her own. But the fighting in Libya, or at least our part in it, is much more restricted: it involves no U.S. soldiers on the ground (and President Obama assures us that won't change), and perhaps involves only a fairly brief series of attacks on well-specified military targets. How much more than this small-scale engagement the President currently has legal power to initiate we do not know; but this much, I think, is within his or her present powers.

That doesn't mean that we can go on attacking Libyan targets indefinitely. We have a statute on the books, the War Powers Resolution, which makes clear that this attack must end within 60 days (with a 30-day extension in some circumstances) unless it is affirmatively authorized by Congress. Those 60 days are sometimes described as a "blank check." I don't think they are a blank check -- that is, I don't think the President can do just anything for 60 days, though I do think he or she can do what Obama has done so far. But at the 60-day limit, this fighting must either be approved by Congress or ended. It is true that that rule has on one occasion been unambiguously violated -- by President Clinton, in another good cause, the 1999 bombing campaign against Yugoslavia to end its oppression of Kosovo -- and it's also true that Presidents and Congresses have pushed back and forth on what the War Powers Resolution required on a number of other occasions. But the War Powers Resolution is still on the books, and I don't think we have an accumulated practice that indicates it's no longer really binding.

The rule that Presidents can initiate some level of small-scale military engagement may be wise or unwise. It also may or may not be retained; the accumulated practice that is its basis could be altered. But I think it is now the law, and we will now have another occasion to view its consequences. I hope they will be good.

Do high LSAT scores and high college grades help -- or hurt -- lawyers?

In my previous post I tried to comment on some overall implications of the Shultz-Zedeck study. But several of the specific findings just cry out for discussion as well, on the complex relationships between measures of academic ability or achievement on the one hand and elements of lawyering effectiveness on the other.

Shultz and Zedeck generated an enormous quantity of data, including appraisals of study participants' competency on 26 different indices, as measured by the participants themselves, or by their peers, or by supervisors, or by peers and supervisors together, or by all evaluators together. In evaluating the extent to which various possible predictors in fact correlated with the subjects' lawyering competency, they focused (and I do too) on those instances where a predictor (such as the LSAT) correlated with a competency as appraised by at least two groups at a statistically significant level.

In the part of their study that focused on current Boalt law students, Shultz and Zedeck found that those with higher LSAT scores did better on a number of effectiveness measures, including not only "Analysis and Reasoning" -- presumably the principal focus of the LSAT itself -- but also "Writing," "Influence and Advocating," "Creativity" and "Problem Solving." (73) This finding is actually quite a comforting one for schools relying heavily on the LSAT in their admissions decisions, or in other words for almost all American law schools. It suggests that the abilities the LSAT measures are not just those of technical legal analysis but also include some more supple skills that surely are very much needed in actual practice. On one criterion, however, these students did worse, namely "Networking," and Shultz and Zedeck wonder whether high-LSAT students tend to be those who aren't as good at Networking or give it less attention. (73)

Results for the alumni/ae study subjects, who included graduates of both Boalt and Hastings, were broadly similar (53-54). Higher LSAT scores correlated positively with "Analysis and Reasoning," "Researching the Law," and "Writing"; and negatively with "Networking" and "Community Service."

Meanwhile, Boalt students with higher undergraduate grade point averages (UGPA) turn out to score lower on "Practical Judgment," "Questioning and Interviewing," "Developing Relationships," and "Community Service." All of those, the authors suggest, might reflect that those with higher grades are more focused on their books and less on engaging with others in the world. (73-74) Why wouldn't these same effects appear with higher LSAT scores? Perhaps the answer is that the LSAT tends to test just ability, while the UGPA reflects application -- and application to one's books takes time and energy that otherwise might be devoted to learning how to engage with others.

Interestingly, the same results didn't appear with the alumni/ae studied. For them, higher UGPAs didn't correlate at a statistically significant level on any of the competencies where Boalt students with higher UGPAs seemed to be weaker. (See Table 18.) Instead, among graduates, higher UGPAs correlated with higher competency in "Writing," "Managing One's Own Work," and "Diligence." (54) Perhaps the diligent people who earn high grades, and show some gaps in interpersonal competencies in their student years, apply themselves to their interaction with others once they are in practice and over time make up for their bookish focus as students. Perhaps those with high LSATs, who seem to be less skilled at Networking whether assessed as students or as graduates, just never do become interested enough in this task to master it.

The most dismaying feature of the impact of higher UGPA among the students is that it also results in lower scores on "Integrity." (73) All appraisals of Integrity -- by the students themselves, by their peers, by their supervisors, by their peers-combined-with-supervisors, and by everyone combined -- correlated negatively with UGPA, though the correlations for the appraisals by the students themselves, and by their supervisors alone, were not statistically significant. (Table 48.)

Shultz and Zedeck don't comment specifically on why this might be so. The bleakest explanation would be that those with the higher undergraduate grades sometimes didn't come by them honestly, and this lack of integrity shows up in their lawyering behavior as well. But I wouldn't read the finding this way, nor as an indication that people who study are in some way less fundamentally concerned with morality than their peers. What I suspect it indicates is that "Integrity" as a competency is a quality of action, and that those who have spent more time studying and less time in action have had less opportunity to develop themselves as moral beings. This might also explain why there's no statistically significant relationship between Integrity and students' LSAT scores; though students with higher LSATs do show signs of inattention to others in their lower "Networking" scores, they may be less inattentive to others than those who are putting in the work to earn higher UGPAs.

Interestingly, in the study of alumni/ae (at 54), higher LSAT scores did correlate negatively with self-appraisals of "Integrity," but positively with appraisals of Integrity by others (specifically, with appraisals by peers, and by peers-combined-with-supervisors, but not with appraisals by supervisors alone, see Table 18)). Shultz and Zedeck don't discuss it, but their Table 18 reflects the same kind of interaction between "Integrity" and the UGPA among the alumni/ae: a negative correlation between grades and Integrity as measured by self-appraisal, and a positive one for Integrity as measured by "Other" (that is, by peers and supervisors combined).

Shultz and Zedeck do address the LSAT - Integrity relationship, and suggest that the explanation may be that the "Self" ratings reflected participants' knowledge of their own secrets, while appraisals by others reflected the impact of successful manipulation of appearances by the guilty "Selves." (54) Presumably the same reasoning could explain the UGPA - Integrity relationship. And perhaps the reason that Integrity as appraised by all raters in the student study correlated negatively with UGPA, while Integrity correlates negatively with UGPA (and LSAT) for alumni/ae only when Integrity is appraised by the graduates themselves, is precisely that over time people learn to cover up their moral flaws from others.

But I prefer a different explanation, one that takes seriously the idea that integrity is an interpersonal quality. Over time, even those with relatively weaker interpersonal skills can learn what integrity requires -- so alumni/ae with higher UGPAs or LSATs are not in fact lacking in integrity in action. But if an inattention to others underlay their higher UGPAs and LSAT scores, perhaps that reflected a greater preoccupation with attention to themselves, and perhaps that inner-directed focus tends throughout their lives to make such people more self-critical than their peers are.

The main finding, however, is probably the one Shultz and Zedeck emphasize, namely that neither the LSAT nor UGPA nor an Index combining the two predicted much about the 26 elements of lawyering competence among alumni/ae (55) or among students (74). I don't think it follows from this overall point that intellectual ability and achievement are unnecessary to lawyering competence. Rather, I think what the finding underlines is how multifaceted good lawyering is, and how many different human attributes need to be nurtured to help people become effective lawyers.

Who will be a good lawyer?

Marjorie Shultz and Sheldon Zedeck in 2008 completed a formidable study, "Final Report: Identification, Development, and Validation of Predictors for Successful Lawyering." It builds on earlier work (which I haven't yet read but which they summarize here, at 24-27), through which they developed -- from extensive interviewing of lawyers, judges, law professors and students, and a number of clients -- 26 elements of effective practice. Perhaps even more excitingly, they adopted some existing psychological tests, and developed other tests of their own, which turned out to correlate -- much better than standard law school admission tools such as the LSAT did -- with which lawyers actually were performing effectively (that is, displayed those 26 elements of effective practice)(see 53-61).

Shultz and Zedeck are surely right (79-80) that these findings make further research along the lines they developed well worth undertaking. But they would also, I'm sure, acknowledge that we have a long ways still to go towards the goal of identifying new admission tools that will actually predict lawyering effectiveness. A marker of that distance is that the various new measures they had devised were apparently less powerfully correlated with the lawyering effectiveness of current law students than they were with the effectiveness of those already out in practice. In some respects, the new tests evidently did better than tests like the LSAT in predicting student effectiveness (76), but when the authors ran statistical multiple regressions to assess the value of all the potential predictors they had employed, it turned out that the LSAT seemed to provide the clearest predictive value! (77)

It isn't easy to know what to make of these results. One possibility is that law school -- or at any rate Boalt (the University of California at Berkeley), whose students were the ones studied in this part of the work) -- is not a setting in which it's easy to develop and demonstrate lawyering competencies. Perhaps it is only after graduation, when former students' attention turns definitively to practice, that the qualities Shultz and Zedeck's tests may predict can actually come into view. Another possibility is that today's law students aren't the same sorts of people as yesterday's, and that the psychological factors that might have predicted high competence in yesterday's students -- that is, in the current lawyers for whom the new tests correlated well with effectiveness -- don't work with today's students.

At any rate, there is clearly more to study here. Such studies won't be simple, as the 59 Tables accompanying Shultz and Zedeck's Final Report attest. But it is surely worth trying to figure out better than we now can who the most promising potential lawyers are among each year's applicants, and it is also surely worth building on the empirical effort Shultz and Zedeck have made to explain what the actual skills of successful lawyering are.

Terrorism, by Al Qaeda and the IRA

I personally have experienced two acts of terrorism: one by Al Qaeda and one by the IRA.

As for Al Qaeda: I was at my school, New York Law School, when the two planes hit the World Trade Center about half a mile south on September 11, 2001. As I sat at my desk that morning I'd heard two thunks, which I thought at first were boxes being dropped on the floor somewhere upstairs -- but after I heard the news of the attacks I realized those were probably the two planes hitting the towers. As the morning went on, I stood outside on our corner and saw the towers burning, and then the cloud of dust from the collapse of one or both of them billowed up to about one small block short of where the law school sits.

And the IRA: I was sitting in my parents' flat in Knightsbridge in London when we felt or heard an explosion. That was the IRA's, or Provisional IRA's, bomb at Harrod's. I'd forgotten the date, but Wikipedia reports that it was December 17, 1983; I'd have been in England visiting my parents for Christmas vacation. Wikipedia also reports that 6 people were killed in that car bombing attack, including a U.S. citizen.

The IRA and Al Qaeda aren't morally equivalent. In a New York Times article reflecting on Representative Peter King, currently the scourge of supposed Muslim extremism in the US but formerly a strong supporter of the IRA, Scott Shane notes that the IRA's efforts mostly did not target civilians, whereas Al Qaeda obviously took aim directly at US civilians in the World Trade Center. I'd say the IRA's cause was more just than Al Qaeda's too, though it must be said that both groups had real oppression (Protestant rule in Northern Ireland, autocratic Arab governments supported by the US) to complain of, and in my opinion both were fanatical in their response.

A lot of very bad things have been done in wartime, and I think it's hard not to say that sometimes the end does justify the means. At the same time, a century's effort has built a body of international humanitarian law -- the more elegant and vague name now often given to the "law of war" -- that aims to minimize the means used, no matter what the ends, and that effort is profoundly important. By its terms, it is illegal for everyone to deliberately target civilians, and both the bombings that touched my life were exactly that. If Peter King doesn't see that these acts similarly deserved to be deplored, he's missing a crucial point.

A personal note -- remembering my father

Amanda Sigler, a talented doctoral candidate at the University of Virginia, has just written an insightful biographical study of my father's work on his biography of James Joyce. (Amanda Sigler, "Joyce's Ellmann," Joyce Studies Annual - 2010, at 3-70.) It's startling to see one's father made the subject of study, but also enlightening. (As a good biography should, this study also makes reference to other people in its subject's life -- including my father's parents, my mother, my sisters, and, yes, me. All of us emerge unscathed.)

I'd always felt that my father's biographies were not only lives of the great writers he studied, but also expressions of himself. As much as he cared about each of his subjects (he wrote biographies of Yeats and Wilde as well as Joyce), and as much as he devoted himself to describing the details and the flavor of their lives, at the same time what he cared about in them he cared about in life as well, and so his accounts reflected his own response to life.

I learned from Sigler's article that my father knew this. It turns out that as he was working on the Joyce biography in the early 1950s, another scholar proposed that the two of them write about Joyce together. My father decided not to, and wrote to a friend and colleague, Ellsworth Mason, "I think I must proceed by myself; my notion of biography is that it should be a portrait of the writer as well as the subject, and I can't see how [the other scholar] and I would work together to constitute some sort of portrait a trois." (Quoted by Sigler at 20.)

What hadn't occurred to me, but Sigler sees, is that the process would work both ways. And this has to be right: as we infuse ourselves into what we do, so we ourselves learn from what we encounter. I always felt that I'd been raised partly on life experiences of James Joyce, and now I see better why that was. Sigler comments, for instance, that "for Ellmann, the little people were the important people, the small details essential to the composite whole. He had learned this lesson from Joyce, who insisted upon making ordinary men and women the subjects of his famous works." (Sigler at 58.) Joyce, as I learned myself when I read my father's work, once said, "I never met a bore" -- meaning that everyone is interesting if you engage with them. My father must have agreed, and the fact that these words of Joyce's stuck with me suggests that the lesson was passed on to me as well.

What's cause and effect here is hard to tease out. You study what you value, and you value what you study. But you're not a fixed entity either: everything you learn shapes you, every commitment of effort becomes part of your lifetime of choices. My parents denied that I was named for Joyce's Stephen Dedalus, and I'm sure that denial was sincere. Sigler (at 46) tells a similar story about my sister Lucy's name, and I believe my parents also denied that my sister Maud's name came from literature. But all these denials may also have been mistaken, for how could my parents (English professors both) help but value personally the writers and the words that they engaged so deeply with professionally?

We are what we study. We shape what we study so that it incorporates something of us. And we incorporate something of it into ourselves.

The Vaccine Injury Act case -- a problem for textualists

On February 22, 2011, by a vote of 6 - 2 in the case of Bruesewitz v. Wyeth LLC, the Supreme Court decided that the National Childhood Vaccine Injury Act of 1986 absolutely bars lawsuits based on allegedly defective vaccine design, as long as the vaccine was manufactured to its own specifications and came with proper directions and warnings about any risks. This result may be a good one. It is quite possible that potential litigation -- in particular, many pending cases about the possible relation between the DTP (diphtheria, tetanus & pertussis) vaccine and childhood autism -- might be so burdensome that private manufacturers would simply abandon the making of badly needed vaccines. But the various opinions in the Bruesewitz case leave me with the impression that whether or not this result is a good one, it wasn't the one Congress intended. The need for legislation does not enact it, as Justice Frankfurter once remarked.

Here, in any event, I want to put to one side the question of Congress' intentions (as manifested in the legislative history, whose various elements the justices scrutinize), in order to focus just on the text of the statutory provision at issue. A textualist must come to grips with the text. To be sure, a textualist can rightly consider other parts of a statute (or even of other laws) in order to determine the meaning of the particular bit of text at issue in a case, and I'm going to leave to one side as well the question of whether other parts of the Vaccine Injury Act weighed in the Court's favor. It seems to me as a general proposition, however, that a textualist must find something very strong in other parts of a statute or other laws to justify avoiding the apparent meaning of the words directly at issue, and I doubt that there was anything strong enough to justify the extent of the avoidance in this case.

Let's look at the statute. As Justice Sotomayor says in dissent, the act asserts as a general rule that state law -- this would be the body of law governing liability for defective products -- governs vaccine cases. (42 U.S.C. 300aa--22). It goes on to state exceptions, including this one, 42 U.S.C. 300aa--22(b)(1):

"No vaccine manufacturer shall be liable in a civil action for damages arising from a vaccine-related injury or death associated with the administration of a vaccine after October 1, 1988, if the injury or death resulted from side effects that were unavoidable even though the vaccine was properly prepared and was accompanied by proper directions and warnings."

What's wrong with the majority's argument that this statute bars all lawsuits for vaccine design defects, as long as the vaccine was "properly prepared" and came with "proper directions and warnings"? Here are several answers (most or all covered by Justice Sotomayor in her very effective dissent, which to my mind persuasively refutes the majority opinion by Justice Scalia, himself a master of statutory interpretation argumentation):

(1) The statute never says anything explicitly about design defects at all. If Congress had wanted to preclude all litigation based on design defects, it could have said "There is no liability for design defects." In fact, the statute doesn't even say in so many words that it has any relation to design defects at all. But it does, I believe: evidently vaccine litigation dealt with three issues -- warnings, preparation and design -- and the first two are explicitly referred to, suggesting that the discussion of "unavoidable side effects" is meant to refer to side effects from the third source, design. And while it is true that the statute fails to say "There is no liability for design defects," it also fails to say, "Liability may be found for design defects." So the argument from lack of explicitness is not conclusive, though I think the statute does come closer to explicitly affirming liability for design defects that aren't "unavoidable" than it does to explicitly denying such liability.

(2) The "if" clause (as Justice Sotomayor labels it): The statute says that manufacturers aren't liable "if the injury or death resulted from side effects that were unavoidable...." The use of the word "if" suggests that "if not" is also conceivable -- in other words, that some side effects are unavoidable but some are avoidable. But for the majority, as long as proper manufacture and warning are taken care of, there is no possibility of liability for design defect. In other words, there is no such thing as an "avoidable" side effect from a vaccine that is properly prepared and accompanied by proper warnings. This is an odd idea at best -- although evidently some courts in the years leading up to this statute had taken essentially this view -- and it's an idea that is very poorly conveyed by a clause beginning with "if."

(3) The meaning of the word "unavoidable": The majority maintains that if the design of a vaccine results in the risk of side effects, those side effects are unavoidable. But what if they could have been avoided by a different design? In that case, they just aren't "unavoidable" and so, for the statute to mean that all such side effects count as unavoidable, the word "unavoidable" has to have taken on some very odd definition, such as "avoidable, but not by this design" -- and textualists seek ordinary usage as a general rule, not idiosyncratic definitions, especially ones that are never spelled out.

To be sure, there's an exception to this focus on ordinary usage, for words that have become terms of art. There is in fact considerable discussion in the case of whether "unavoidable" was a term of art, but no one asserts that if it was a term of art, the consensus meaning it had acquired was "avoidable, but not by this design."

Somewhat remarkably, the majority claims that if "unavoidable" is read to mean "not avoidable by a different design" then "the word 'unavoidable' would do no work" (majority opinion at 7), on the ground that "[a] side effect of a vaccine could always have been avoidable by use of a differently designed vaccine not containing the harmful element." But Justice Sotomayor responds that "the harmful element" might be essential to the vaccine's efficacy, and that it's precisely in such cases that the side effects deserve to be called "unavoidable" -- whereas in other cases the side effects might have been avoided by better design, and would then be "avoidable." (Dissent at 14-15.)

(4) As Justice Sotomayor emphasizes, and the majority concedes, the net effect of the majority's reading is that 13 words of the statute -- italicized below -- turn out to have no meaning and to be completely superfluous:

"No vaccine manufacturer shall be liable in a civil action for damages arising from a vaccine-related injury or death associated with the administration of a vaccine after October 1, 1988, if the injury or death resulted from side effects that were unavoidable even though the vaccine was properly prepared and was accompanied by proper directions and warnings."

This is just a huge problem for a textualist. Though I don't claim to have read every case where such issues have arisen, I've never encountered a case where a textualist such as Justice Scalia accepted so extensive a violation of the well-known interpretive principle that statutes should be read so that every word has meaning. What the majority says is that this violation isn't determinative, because (says the majority) on the dissent's reading another set of 15 words become superfluous -- the ones italicized here (the "even though" clause of the statute):

"No vaccine manufacturer shall be liable in a civil action for damages arising from a vaccine-related injury or death associated with the administration of a vaccine after October 1, 1988, if the injury or death resulted from side effects that were unavoidable even though the vaccine was properly prepared and was accompanied by proper directions and warnings."

Scalia's argument is that for the dissent there's only one question -- were the side effects "unavoidable." (Majority opinion at 12.) But Sotomayor responds that the "even though" clause does have a function: it establishes that "unavoidable" side effects resulting from design defects are only exempt from liability if the vaccine was properly prepared and came with proper directions and warnings. If a manufacturer fails to prepare the vaccine properly or provide the directions and warnings that should accompany it, then even if the side effects of its design really are unavoidable, the manufacturer remains liable for them. It seems to me that this reading gives content to the "even though" clause, and therefore that the interpretive rule against finding portions of a statute superfluous quite clearly favors Sotomayor's reading.

(5) But the "even though" clause may also provide the strongest textual argument in favor of the majority's position. To say that "side effects ... were unavoidableeven though the vaccine was properly prepared and was accompanied by proper directions and warnings" seems to say that what might have made them avoidable was proper preparation and/or proper directions and warnings. The side effects are unavoidable even though -- despite, as Scalia says -- proper preparation, directions and warnings. Sotomayor's reading, on the other hand, seems to make "even though" mean "provided that" -- manufacturers aren't liable for unavoidable side effects provided that the vaccines were properly prepared and came with proper directions and warnings. Scalia gives this point a grammatical tag, telling us that the "even though" clause is "called a concessive subordinate clause by grammarians." (Opinion at 11.)

I think Scalia is right that Sotomayor's reading isn't "concessive." But the "even though" clause is awkward for all sides in this debate. After all, what makes the side effects unavoidable, on the majority's reading of the statute? The answer might be that we know these side effects are unavoidable "because" neither better preparation nor better directions and warnings could have avoided them. Otherwise their unavoidability is altogether undefined. "Because" is usually quite a ways from "even though," as Sotomayor argues (dissent at 17 n.14), but here, oddly, the meaning of these words seems to coincide. I think that on this score Scalia's reading is the more natural. Yet it's worth noting, as Sotomayor does, that Scalia's reading -- as discussed above -- actually means that the words "even though," along with the 11 words preceding them, lose all meaning, so that his emphasis on fidelity to the import of a concessive subordinate clause seems somewhat unsatisfactory. I think the main lesson to be drawn is that the statute is, truly, badly drafted. On balance, I also think Sotomayor's reading does a better job of giving meaning to as many words of the text as possible.

In short, I don't think the majority's reading of these words is easy to sustain on textualist grounds -- yet Justice Scalia is committed to textualist interpretation. Justice Sotomayor says that the majority's decision is "policy-driven," though she does so only in a footnote almost at the end of her opinion (dissent at 27 n.25). If I am right about the relative weakness of the textualist arguments in favor of the majority's position, that does suggest that something else -- policy -- drove the Court's thinking. It's possible to defend policy-driven statutory interpretation -- but not on textualist grounds. Indeed, textualists have been outspoken in objecting to other methods of interpretation, in which judges read statutory language in light of other evidence of legislators' specific intentions or broad purposes, as giving judges too much room to enact their own preferences into law. All of which makes Bruesewitz a problem for textualists.

Al Qaeda and the rise of freedom in the Middle East

Among Muammar Qaddafi's bizarre comments as his fall from power in Libya approaches, one particularly interesting one was his accusation that the rebels were under the thrall of Al Qaeda. His notion was that Al Qaeda had imposed its will by enticing young people to take hallucinogenic drugs, but -- that bizarre idea aside -- Qaddafi's basic point was not crazy: Al Qaeda has indeed opposed the autocratic governments of the Middle East. Not because Al Qaeda opposes autocracy, at least in the form of theocracy, but because it sees these governments as tools of the West and obstacles to Islamist change.

So is the fall of several authoritarian Mideast regimes good for Al Qaeda? That is one possibility. A commentator on the news today (Saturday, February 26) mentioned that one of the Libyan tribes that has now turned against Qaddafi practices a fundamentalist form of Islam and has declared Islamic rule in its part of Libya. More generally, it's clearly possible that fundamentalists will prove the best organized and most determined citizens in the countries that have thrown off their previous rulers, and will install themselves in place of the old, more pro-Western autocrats.

But something else is possible too. It's often been said, and I think correctly, that Franklin D. Roosevelt helped save American capitalism -- even though many capitalists hated him, and even though he sharply attacked them in turn. Precisely because FDR's government paid attention to the needs and the voices of the American people, he was able to generate reforms that contributing to reestablishing a workable social contract in this country.

Something of the same thing may turn out to be true in at least some countries of the Middle East. (It's a very diverse place, so I don't want to generalize too much -- and even so I'm speculating!) It may be that the best way to stem the tide of Al Qaeda is for governments to come to power that listen to their people. If so, the fall of the governments that we relied on may in the end turn out to be not Al Qaeda's victory but the source of its eventual downfall.