Monthly Archives: March 2009

Superior Coffee at 17th and Pennsylvania

Until 9:30 or so… Holding court at Caribou Coffee. Very Feist-heavy music…


The Importance of the G-20 Meeting

It looks as though Obama is going to get about half of what he asks for–which is a failure not on his part but on the part of the Europeans. Let’s hope that that is enough.Paul Krugman:>America the Tarnished: Like many other economists, I’ve been revisiting the Great Depression, looking for lessons that might help us avoid a repeat performance. And one thing that stands out from the history of the early 1930s is the extent to which the world’s response to crisis was crippled by the inability of the world’s major economies to cooperate.>The details of our current crisis are very different, but the need for cooperation is no less. President Obama got it exactly right last week when he declared: “All of us are going to have to take steps in order to lift the economy. We don’t want a situation in which some countries are making extraordinary efforts and other countries aren’t.”>Yet that is exactly the situation we’re in. I don’t believe that even America’s economic efforts are adequate, but they’re far more than most other wealthy countries have been willing to undertake. And by rights this week’s G-20 summit ought to be an occasion for Mr. Obama to chide and chivy European leaders, in particular, into pulling their weight.>But these days foreign leaders are in no mood to be lectured by American officials, even when — as in this case — the Americans are right…

Dan Froomkin on the Torture of Abu Zubaida

Dan reports on what the *Washington Post* now prints as front-page news–something it ould have printed three years ago:>White House Watch – Bush’s Torture Rationale Debunked: Abu Zubaida was the alpha and omega of the Bush administration’s argument for torture. That’s why Sunday’s front-page Washington Post story by Peter Finn and Joby Warrick is such a blow to the last remaining torture apologists. Finn and Warrick reported that “not a single significant plot was foiled” as a result of Zubaida’s brutal treatment — and that, quite to the contrary, his false confessions “triggered a series of alerts and sent hundreds of CIA and FBI investigators scurrying in pursuit of phantoms.”>Zubaida was the first detainee to be tortured at the direct instruction of the White House. Then he was President George W. Bush’s Exhibit A in defense of the “enhanced interrogation” procedures that constituted torture. And he continues to be held up as a justification for torture by its most ardent defenders. But as author Ron Suskind reported almost three years ago — and as The Post now confirms — almost all the key assertions the Bush administration made about Zubaida were wrong. Zubaida wasn’t a major al Qaeda figure. He wasn’t holding back critical information. His torture didn’t produce valuable intelligence — and it certainly didn’t save lives. All the calculations the Bush White House claims to have made in its decision to abandon long-held moral and legal strictures against abusive interrogation turn out to have been profoundly flawed, not just on a moral basis but on a coldly practical one as well. Indeed, the Post article raises the even further disquieting possibility that intentional cruelty was part of the White House’s motive.>The most charitable interpretation at this point of the decision to torture is that it was a well-intentioned overreaction of people under enormous stress whose only interest was in protecting the people of the United States. But there’s always been one big problem with that theory: While torture works on TV, knowledgeable intelligence professionals and trained interrogators know that in the real world, it’s actually ineffective and even counterproductive. The only thing it’s really good as it getting false confessions. So why do it? Some social psychologists (see, for instance, Kevin M. Carlsmith on have speculated that the real motivation for torture is retribution. And now someone with first-hand knowledge is suggesting that was a factor in Zubaida’s case. Quoting a “former Justice Department official closely involved in the early investigation of Abu Zubaida,” Finn and Warwick write that the pressure on CIA interrogators “from upper levels of the government was ‘tremendous,’ driven in part by the routine of daily meetings in which policymakers would press for updates…>”‘They couldn’t stand the idea that there wasn’t anything new,’ the official said. ‘They’d say, “You aren’t working hard enough.” There was both a disbelief in what he was saying and also a desire for retribution — a feeling that ‘He’s going to talk, and if he doesn’t talk, we’ll do whatever.'”‘ The Post story also makes it clear that some people with great reality-denying skills remain at the upper levels of the government: “Some U.S. officials remain steadfast in their conclusion that Abu Zubaida possessed, and gave up, plenty of useful information about al-Qaeda,” Finn and Warwick write. “‘It’s simply wrong to suggest that Abu Zubaida wasn’t intimately involved with al-Qaeda,’ said a U.S. counterterrorism official, speaking on the condition of anonymity because much about Abu Zubaida remains classified. ‘He was one of the terrorist organization’s key facilitators, offered new insights into how the organization operated, provided critical information on senior al-Qaeda figures…and identified hundreds of al-Qaeda members. How anyone can minimize that information — some of the best we had at the time on al-Qaeda — is beyond me.'” But who are these people? How can they still possibly believe this given all the evidence to the contrary? What are they doing still in government?>Author and investigative reporter Suskind first exposed the rampant fallacies of the administration’s Zubaida narrative in his explosive June 2006 book, The One Percent Doctrine. See my June 20, 2006 column for a summary. But mainstream news organizations, unable to match Suskind’s sources, largely refused to acknowledge his reporting. Indeed, in September 2006, when the White House for the first time publicly acknowledged the existence of a secret CIA detention and interrogation program, Bush had no qualms about putting Zubaida front and center. In a major speech, he proudly described how Zubaida — “a senior terrorist leader and a trusted associate of Osama bin Laden” — was questioned using the CIA’s new “alternative set of procedures” and then “‘began to provide information on key al Qaeda operatives.”>All lies and euphemisms. But all reported pretty much straight at the time by a mainstream media that, if it noted Suskind’s reporting at all, did so as an afterthought…

Republicans: The Stupid Party

Lee Fang:>Think Progress» Rep. Paul Ryan Concedes GOP Alternative Budget Would Increase The Deficit ‘A Lot’: Last week, the House GOP presented its alternative budget proposal. Members of the media, including conservative commentators, widely panned the document for being scant on details and appearing more as “campaign-style talking points.” Rep. Paul Ryan (R-WI), ranking member of the House Budget Committee, has said he will release yet another budget proposal, but this time with more specifics. Though Ryan has been most critical of the deficit impact of Obama’s budget, he has been unable to assess the deficit impact of his own budget. After being repeatedly asked this weekend by Bloomberg’s Al Hunt about “how large” the deficit would be under the Republican plan, Ryan finally respond, “A lot”:>>HUNT: But the Obama budget deficit is $1.4 trillion. How, roughly, how large will yours be?>>RYAN: Their budget deficit is $1.8 trillion. […]>>HUNT: Gimme an idea of how large yours will be?>>RYAN: A lot. Let’s put it that way.>>HUNT: Pardon me?>>RYAN: Now I can’t give you the specific numbers because we’re still waiting for some numbers back from CBO. But clearly we don’t want to have this kind of run up of deficits and debt….>As ThinkProgress previously noted, Rep. Mike Pence (R-IN) also stumbled and refused to offer a number when questioned by MSNBC about the deficit under the GOP plan.

The Bush Administration: An Oral History of the Bush White House. The threat of 9/11 ignored. The threat of Iraq hyped and manipulated. Guantánamo and Abu Ghraib. Hurricane Katrina. The shredding of civil liberties. The rise of Iran. Global warming

An excerpt:>Cullen Murphy and Todd Purdum: there were detailed discussions and briefings on cyber-security and often terrorism, and on a classified program. . . . he seemed—I was disturbed because he seemed to be trying to impress us, the people who were briefing him. . . . The contrast with having briefed his father and Clinton and Gore was so marked. And to be told, frankly, early in the administration, by Condi Rice and [her deputy] Steve Hadley, you know, Don’t give the president a lot of long memos, he’s not a big reader—well, shit. I mean, the president of the United States is not a big reader?

Tim Geithner Is Not a Tool of Wall Street

Edward Luce writes:>America’s liberals lay into Obama: The liberal backlash against President Barack Obama has begun with many prominent left-leaning economists in the US attacking the administration’s plans to bail out the banks.>Paul Krugman describes the toxic asset purchase plan as “cash for trash”. Jeffrey Sachs calls it “a thinly veiled attempt to transfer hundreds of billions of US taxpayer funds to the commercial banks”. Robert Reich depicts Tim Geithner, Treasury secretary, as a prisoner of Wall Street while Joe Stiglitz says the plan “amounts to robbery of the American people”. On the blogosphere and beyond, Democratic economists accuse Mr Obama – along with Mr Geithner, and Lawrence Summers, the president’s senior economic adviser – of taking dictation from the same financiers who have brought the economy to the brink of depression.>Mr Reich, who was Bill Clinton’s Labour secretary in the 1990s before resigning over the former president’s reluctance to pursue a strong public investment agenda, says that he and his colleagues fear a replay of the Clinton years under Mr Obama. Mr Reich now talks of the “Paulson-Geithner approach” to demonstrate what he sees as the continuity between Hank Paulson, George W. Bush’s last Treasury secretary, and the current administration. Mr Reich says bank nationalisation is the only answer to today’s crisis. “Bill Clinton chose to pursue a set of policies that Wall Street agreed with but at the expense of his long-term agenda of boosting public investment,” says Mr Reich. “Bill Clinton’s Wall Street agenda in the end brought America and the world crashing down with it. I hope we are not seeing history repeat itself with Mr Obama.”>Not every Democrat agrees. Brad DeLong, a former Clinton official, says that every banking crisis – barring the Great Depression – has been resolved by government recapitalisation of the banking sector, as Mr Obama is likely to attempt in the near future. Nor, says Mr DeLong, is it fair to paint Mr Geithner as a creature of Wall Street. “Hank Paulson is a man who grew up in American finance and cannot imagine a world in which America does well and its financial sector does badly,” he says. “Tim Geithner, by contrast, is a bureaucrat and a policymaker. He has never pulled down a multibillion-dollar bonus. They are not the same type of people.”>But in reality the division is as much political as economic. Most of Mr Obama’s liberal critics argue he should have gone to Congress already and asked for a lot of money for bank recapitalisation. His defenders say that would be political suicide until the populist mood on Capitol Hill has died down. “We have to ask ourselves: Do we want to revive our economy, or do we want to punish the bankers?” says Mr DeLong. “I don’t agree that we can do both.”I should have said “multi-million dollar bonus.” Tim Geithner has never pulled down a multi-million dollar bonus. He has never pulled down a multi-billion dollar bonus either, but rather few people have…

Cosma Shalizi Takes Me to Probability School. Or Is It Philosophy School?

After I accuse Cosma Shalizi of waterboarding the Rev. Dr. Thomas Bayes, he responds:>Cosma Shalizi: Cosma Shalizi Waterboards the Rev. Dr. Thomas Bayes: Hoisted from Comments: I am relieved to learn that the true model of the world is always already known to every competent statistical inquirer, since otherwise it could not be given positive prior weight. I would ask, however, when our model set became complete? And further, when did people stop using models which they knew were at best convenient but tractable approximations?>Less snarkily, these two examples are out-takes from what I like to think is a fairly serious paper on Bayesian non-parametrics with mis-specified models and dependent data:>>>described less technically here:>>>The examples were simple sanity-checks on my theorems, and I posted them because they amused me.Thus Cosma Shalizi takes me to probability school. Or perhaps he takes me to philosophy school. It is not clear.Let me give an example simpler than one of the ones Cosma Shalizi gave. Rosencrantz is flipping a coin. Guildenstern is watching and is calling out “heads” or “tails.” It is a fair coin–half the time it comes up heads, and half the time it comes up tails. Before Rosencrantz starts flipping, Guildenstern’s beliefs about what the next flip of the coin will bring are accurate: he thinks that there is a 50% chance that the next flip of the coin will be heads and a 50% chance that the next flip of the coin will be tails.Because Guildenstern starts with correct beliefs about what the odds are for the next flip of the coin, you might think that there is nothing for Guildenstern to “learn”–that as Rosencrantz flips, Guildenstern will retain his initial belief that the odds are 50-50 that the next flip of the coin will be heads or tails. But there is a problem: Guildenstern is not a human being but rather is a Bayesian AI, and Guildenstern is *certain* that the coin is biased: it thinks that there is a 50% chance it is dealing with a coin that lands heads 3/4 of the time, and a 50% chance it is dealing with a coin that lands tails 3/4 of the time, and its initial prediction that the next flip is equally likely to be heads or tails depends on that initial 50-50 split.What happens as Rosencrantz starts flipping? The likelihood ratio for an H-biased as opposed to a T-biased coin is 3z, where z=h-t and h is the number of heads and t is the number of tails flipped, which means that the posterior probabilities assigned by Guildenstern after h heads and t tails are:>P(H | z) = 3z/(3z + 1)>P(T | z) = 1/(3z + 1)And the estimate that the next flip will be heads is:>(3/4)P(H | z) + (1/4)P(T | z) = (3z+1 + 1)/(4(3z + 1))If the number of heads and tails are even, then Guildenstern (correctly) forecasts that the odds on the next flip are 50-50. If the n flips Rosencrantz has performed have seen two more heads than tails–no matter how big n is–then Guildenstern is 90% certain that it is dealing with an H-biased coin and thinks that the chance the next flip will be heads is 70%. If the n flips Rosencrantz has performed have seen ten more heads than tails–again, *no matter how many flips n there have been*–Guildenstern is 99.9983% sure that it is dealing with an H-biased coin and will forecast the odds of a head on the next flip at 74.9999%.How will Guildenstern’s beliefs behave over time? Well, this passage from Shalizi’s more complex example applies:>Three-Toed Sloth: The sufficient statistic z [for P(H)]… follows an unbiased random walk, meaning that as n grows it tends to get further and further away from [zero], with a typical size growing roughly like n1/2. It does keep returning to the origin, at intervals dictated by the arc sine law, but it spends more and more of its time very far away from it. The posterior estimate of the [probability of an H-biased coin thus wanders from being close to +1 to being close to [0] and back erratically, hardly ever spending time near zero, even though (from the law of large numbers) the sample mean [fraction of heads] converges to zero…So Guildenstern spends all of its time being nearly dead certain that it is dealing with an H-biased coin or nearly dead certain that is dealing with a T-biased coin–but it switches its belief occasionally–even though there is almost surely never any statistically significant evidence for H-bias against the null hypothesis that the coin is fair and 50-50. There is an allowable set of beliefs for Guildenstern that will lead it to make the right 50-50 forecast of the odds on the next flip: if Guildenstern simply continues to believe that there is no evidence either way for H-bias or T-bias. But Guildenstern’s beliefs are not those beliefs and do not converge to those beliefs: look far enough out into the future and you see that Guildenstern is almost sure either that the coin is H-biased or that the coin is T-biased, and has virtually no chance of being unsure about in which direction the bias lies. Thus Guildenstern’s processing of the data is not *sensible,* is not *smart,* is not *rational,* is not *human*–but it is Bayesian. For positive values of z, Guildenstern thinks “there are fewer heads than I would expect for an H-biased coin, but this could never come about with a T-biased coin; the coin must be H-biased: I am sure of it.” A sensible agent, a smart agent, a rational agent, a human agent would think: “Hmmm. Right now I am sure that the coin is T-biased, but 100 flips ago I was sure that the coin was H-biased. I know that as I get more evidence my beliefs should be converging to the truth, but they don’t seem to be converging at all. Something is wrong.” But there is nothing in the Bayesian agent’s little brain to allow it to reason from the failure of its beliefs to converge to the conclusion that there is something badly wrong here.But it seems, intuitively, that Guildenstern should be able to make good forecasts. The the prior that Guildenstern started with does admit of beliefs that would lead to accurate forecasts of the next coin flip: all Guildenstern has to do is to doubt that it has enough information to decide about the bias of the coin. Indeed, Guildenstern’s initial beliefs generate the right forecast of probabilities for the next flip. So, given that Guildenstern starts out with a set of beliefs that supports and generates the “right” forecast probabilities, given that there really isn’t enough information to decide about the bias of the coin–there can’t be, for the coin is not biased–and given that Bayesian learning is a kind of learning, why doesn’t Guildenstern simply keep its original beliefs and keep making good forecasts? Shalizi has identified a case in which it seems that a Bayesian agent should be able to learn enough to make good predictions, but cannot in fact do so.Shalizi has gotten his Bayesian AI Guildenstern to confess. But has he done so by legitimate means? Or by waterboarding? The probability theory question, or perhaps the philosophy question, is: Is this a problem for the Bayesian way of looking at the world? Or only a demonstration that torture elicits confessions?