Please visit the new home of Majikthise at bigthink.com/blogs/focal-point.

« Interview with co-author of Lancet study of Iraqi deaths | Main | Brad DeLong on the Lancet study »

October 13, 2006

Tim Blair and the web of belief

Quinepower

"The totality of our so-called knowledge or beliefs, from the most casual matters of geography and history to the profoundest laws of atomic physics or even of pure mathematics and logic, is a man-made fabric which impinges on experience only along the edges. Or, to change the figure, total science is like a field of force whose boundary conditions are experience. A conflict with experience at the periphery occasions readjustments in the interior of the field. Truth values have to be redistributed over some of our statements. Re-evaluation of some statements entails re-evaluation of others, because of their logical interconnections -- the logical laws being in turn simply certain further statements of the system, certain further elements of the field. Having re-evaluated one statement we must re-evaluate some others, whether they be statements logically connected with the first or whether they be the statements of logical connections themselves. But the total field is so undetermined by its boundary conditions, experience, that there is much latitude of choice as to what statements to re-evaluate in the light of any single contrary experience. No particular experiences are linked with any particular statements in the interior of the field, except indirectly through considerations of equilibrium affecting the field as a whole.

     If this view is right, it is misleading to speak of the empirical content of an individual statement -- especially if it be a statement at all remote from the experiential periphery of the field. Furthermore it becomes folly to seek a boundary between synthetic statements, which hold contingently on experience, and analytic statements which hold come what may. Any statement can be held true come what may, if we make drastic enough adjustments elsewhere in the system. Even a statement very close to the periphery can be held true in the face of recalcitrant experience by pleading hallucination or by amending certain statements of the kind called logical laws. Conversely, by the same token, no statement is immune to revision. Revision even of the logical law of the excluded middle has been proposed as a means of simplifying quantum mechanics; and what difference is there in principle between such a shift and the shift whereby Kepler superseded Ptolemy, or Einstein Newton, or Darwin Aristotle?"--W.V.O Quine [Emphasis added.]

Having dismissed statistical reasoning, Tim Blair goes on to reject peer review.

It's amazing the lengths some folks will go to avoid believing that the invasion of Iraq killed hundreds of thousands of people. I would have thought the big tough war bloggers would be telling the rest of us to suck up the hard truth and press onwards to the Glorious Outcome (whatever it is now). It is worth it, isn't it guys?

Instead, they're pulling the covers over their heads and saying, "What dead people?"

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c61e653ef00d83466a0bd69e2

Listed below are links to weblogs that reference Tim Blair and the web of belief :

» When proofreaders attack from Deltoid
Lindsay Beyerstein spanks Tim Blair: Having dismissed statistical reasoning, Tim Blair goes on to reject peer review. It's amazing the lengths some folks will go to avoid believing that the invasion of Iraq killed hundreds of thousands of people.... [Read More]

Comments

El Presidente said that those numbers are "just not credible", so that's all I need to know. I don't need to know how he knows those numbers aren't credible; I'm just going to assume that Our Leader's highly analytical mind was able to sift through the data and wade through all the minutiae in that report and find significant errors in the researchers' methodology and statistical analyses. Ya see, he's a fighter pilot, and you can't shoot no bull past a fighter pilot, ya see?

This Tim Blair fella knows the score. He might not be a fighter pilot but he's nobody's fool. Math? Statistics? I quit believin' what them math pointy-heads taught me when they tried to brainwash us with that algebra bullshit. Numbers is numbers; they ain't letters! Science? Peer review? If them scienticians with their fancy booklearnin' are so smart, how come they can't explain why my kids look like the mailman instead of me?

I'm glad we have this Blair guy on our side to give them ivy tower boys what fer. (I wonder if he's related to Tony "Yo!" Blair?) Now we need to give this guy a new assignment: Proving that no one died from Hurricane Katrina, 'cept maybe a few welfare queens who were too lazy to leave town. Hell, if we're lucky we could give Heckuvajob Brownie his job back.

Didn't Bush run the whole 2000 election against math, and the kind of people who will employ it in support of unpleasant conclusions?

The 2004 election was more in opposition to history and learning foreign languages.

I love when people try to sound smart. I used to do that in middle school, make stuff as wordy and thesaurosy as possible. I am glad I grew out of that. It's also much funnier if you subvocalize all that gobbley gook in the voice of the late-great Lionel Hutz.

Have conservatives latched onto the controversy over the Lancet-published study about improper anesthesia in U.S. executions? If I were them, I would.

That is a great quote. I frequently want to use metaphors from electrodynamics or statistical mechanics but hold back for fear of seeming too dorky.

There is a group of people, to which Tim Blair belongs, who feel they are not being heard. They have a sense of being persecuted. They have a sense that powerful forces, and large numbers of people, disagree with them. They believe they have truth on their side, if only people would listen. The problem, from the perspective of the TB faction, is that people aren't listening to them.

I was invited to Thanksgiving dinner by a friend of mine 3 years ago. She herself is a moderate, but her parents are both Christians and devout Republicans. This was in an affluent, all white suburb of Richmond, Virginia, USA. Everyone was very polite. Everyone there was well educated. I did notice that there was a deference to my friend's father. When he spoke his wife and kids and friends did not speak.

If Tim Blair is anything like my friend's father, then in his domestic world he is used to being listened to.

Speaking of smart conservatives who don't trouble themselves with facts when attempting to prove a point, Cliff Kincaid (wonder if he's related Thomas "Painter of Arson" and "Territorial Pisser" Kinkade) exposes a dangerous cabal in Washington:

HOMOSEXUAL REPUBLICANS ARE CLOSETED DEMOCRATS.

We are through the looking glass, people.

I hear that Lawrence. There are many flavors of opposition, but it's been interesting to note how often, in some of the most contentious struggles, both sides actually feel themselves to be the ones under threat.

``There is a group of people, to which Tim Blair belongs, who feel they are not being heard. They have a sense of being persecuted. They have a sense that powerful forces, and large numbers of people, disagree with them.''

Hm... Strikes me one could say the same about folks on the left. We know we are not being heard---at least not in the mainstream. To be sure, leftists are heard in some venues, e.g., in some parts of academia-land. And Chomsky did get to give a talk at one of the service acadamies (West Point, yes?) So, once in a great while...

``Just because you are paranoid, it doesn't meant that they aren't out to get you...''

The history of both social and official state action against people on the left over the course of US history is also well known. The obvious examples are the WWI espionage/sedition act prosecutions, and post war red raids, and the post WWII red scare, but there is more where that come from. If those on the left feel persecuted, we have some cause.

Oh, if you think that Tim Blair's stuff leaves something to be desired, you should have a look at the comments on his site in response to his posts about the John Hopkins study. There isn't the hint of an rational argument to be found, as far as I can see, just personal abuse and a sustained exercise in the ad hominem fallacy.

Oh, as this blog is advertised as ``Analytic philosophy and liberal politics'', it's nice to see some of the former. It would also be appreciated, by me at least, if the cartoon of Quine mentioned the co-author of The Web of Belief :-)

The one downside of great blogs like yours is they link to nutjobs like Blair. I read his posts and seethe.

Quine is giving an unjust account of the cohesion theory of truth, especially by pulling out the synthetic vs. analytic distinction. It's not about that. Instead, there's the idea that to have internal consistency in more than a trivial way there has to be a connection with the outside world. What difference is there between cohesion set ups of truth that take account of empirical evidence in their webs and series of interlocking analytic statements? I think he's really stretching the plausability of non-interconnectedness of analytic statements.

If I can say six analytic statements about something, which leads me to posit a seventh, more general statement, that upon empirical observation proves to be true doesn't that suggest some sort of cohesion behind the statements, some sort of structure, that might not be fully known, and that surely will be conditioned by the observer's knowledge and understanding of the situation, but that coheres and is empirical nonetheless.

Extending that out from a particular observor to a scientific community that has similar standards and similar backgrounds, where work is evaluated by teams and by a set, unchanging standard, couldn't it be said that the interrelationships and statements built up over time collectively by this community would be a sort of communal version of the cohesion theory of truth?

Additionally, this would take care of the problem raised by the synthetic/analytic split he talks about by making the subjective experience socialized to the point where it isn't particular subjective experience anymore but something that's consensually agreed on according to strict standards. I wouldn't call that synthetic in the sense of Hegel, for example, coming up with a theory of the evolution of consciousness that's ultimately hinged on his own subjective experience of what consciousness is. Instead, this sort of reasoning is backed up by something more.

Quine doesn't take this possibility into account.

I doubt that internal logic + empiricism would lead to a situation where any statement would be proveable based on how new info changes inter-relationships. There'd be a set number of changes which would be less than infinite that would be possible, leaving a paper trail, and that means that total reversal of everything would not be possible.

Plus, talking about hallucinations is immature if we're talking about science. Who, in a peer reviewed scientific community, would accuse another accredited person of hallucination?

This gets, finally, to the Bush-Blair take on the study in the Lancet. They aren't qualified to evaluate it. We have to take it on some sort of faith that this thing is right but we also have the capability to evaluate it on some level ourselves. These two faculties co-exist and they feed into one's own personal awareness of the cohesion of truth, but we have access to other, socially conditioned, standards. For instance, the Lancet study could be considered within the purely personal realm of what someone feels is the truth or it could be considered in reference to the social reputation of the Lancet along with that of it's methods and it's researchers. Also, the same could be said of one's capacity to evaluate the information. There's a person's first take but they can also be lead to another understanding of it by someone else's trail, i.e. if someone explains it to them. If that happens and the person is using socially accepted standards, taken from the flux of empirical-cohesive logical standards relating to the field, then he can be said to be leading the person to truth, out of his own small cohesion and into the larger cohesion talked about.

Then there's the criticism of power distorting the role of these things. Like 'those fucking liberals, they're all liberals they don't tell the truth'. Distortion is the key here. Even if they're liberals, those wily creatures, they still have other qualifications that they use so that any distortion, if taken within the whole context and not within the limited context suggested by one's own personal idea of the cohesion of truth, can be measured against the socially accepted standards and seen or corrected without necessarily damning the whole thing.

Or maybe it's all just language games that lead to overgeneralized particularities that are irreducable based on some socialized standard of the subjective component.

Ha.

Quine is giving an unjust account of the cohesion theory of truth, especially by pulling out the synthetic vs. analytic distinction. It's not about that. Instead, there's the idea that to have internal consistency in more than a trivial way there has to be a connection with the outside world. What difference is there between cohesion set ups of truth that take account of empirical evidence in their webs and series of interlocking analytic statements? I think he's really stretching the plausability of non-interconnectedness of analytic statements.

If I can say six analytic statements about something, which leads me to posit a seventh, more general statement, that upon empirical observation proves to be true doesn't that suggest some sort of cohesion behind the statements, some sort of structure, that might not be fully known, and that surely will be conditioned by the observer's knowledge and understanding of the situation, but that coheres and is empirical nonetheless.

Extending that out from a particular observor to a scientific community that has similar standards and similar backgrounds, where work is evaluated by teams and by a set, unchanging standard, couldn't it be said that the interrelationships and statements built up over time collectively by this community would be a sort of communal version of the cohesion theory of truth?

Additionally, this would take care of the problem raised by the synthetic/analytic split he talks about by making the subjective experience socialized to the point where it isn't particular subjective experience anymore but something that's consensually agreed on according to strict standards. I wouldn't call that synthetic in the sense of Hegel, for example, coming up with a theory of the evolution of consciousness that's ultimately hinged on his own subjective experience of what consciousness is. Instead, this sort of reasoning is backed up by something more.

Quine doesn't take this possibility into account.

I doubt that internal logic + empiricism would lead to a situation where any statement would be proveable based on how new info changes inter-relationships. There'd be a set number of changes which would be less than infinite that would be possible, leaving a paper trail, and that means that total reversal of everything would not be possible.

Plus, talking about hallucinations is immature if we're talking about science. Who, in a peer reviewed scientific community, would accuse another accredited person of hallucination?

This gets, finally, to the Bush-Blair take on the study in the Lancet. They aren't qualified to evaluate it. We have to take it on some sort of faith that this thing is right but we also have the capability to evaluate it on some level ourselves. These two faculties co-exist and they feed into one's own personal awareness of the cohesion of truth, but we have access to other, socially conditioned, standards. For instance, the Lancet study could be considered within the purely personal realm of what someone feels is the truth or it could be considered in reference to the social reputation of the Lancet along with that of it's methods and it's researchers. Also, the same could be said of one's capacity to evaluate the information. There's a person's first take but they can also be lead to another understanding of it by someone else's trail, i.e. if someone explains it to them. If that happens and the person is using socially accepted standards, taken from the flux of empirical-cohesive logical standards relating to the field, then he can be said to be leading the person to truth, out of his own small cohesion and into the larger cohesion talked about.

Then there's the criticism of power distorting the role of these things. Like 'those fucking liberals, they're all liberals they don't tell the truth'. Distortion is the key here. Even if they're liberals, those wily creatures, they still have other qualifications that they use so that any distortion, if taken within the whole context and not within the limited context suggested by one's own personal idea of the cohesion of truth, can be measured against the socially accepted standards and seen or corrected without necessarily damning the whole thing.

Or maybe it's all just language games that lead to overgeneralized particularities that are irreducable based on some socialized standard of the subjective component.

Ha.

Are you folks aware that Iraq Body Count had a rather nasty battle with defenders of the previous Lancet paper on Iraq? Much of it occurred in the temporary comment section of the medialens site. I started out on the Lancet side myself and ended up in the middle.

The gist of the IBC case against the 2004 paper was that there was a UNDP survey (also called the ICLS survey) in Iraq that covered the first 13 months and they found that there were 24,000 war-related violent deaths (excluding criminal murders) in the first 13 months of the Iraq War. The 95 percent confidence interval was 19,000 to 28,000. The Lancet paper in 2004 was in rough agreement, but somewhat larger if you looked at violent deaths and excluded the Fallujah outlier. Their figure was 57,000 violent deaths including criminal murders for the first 18 months. The IBC people argued that if you compare the same time frame, the 2004 Lancet estimate would be about 30 -40 percent larger than the UN number. Rough agreement, if you exclude Fallujah.

Okay, now the 2006 paper. It obfuscates things by saying that it agrees with the previous report. Well, not if you focus on violence. In table 4 there are 45 violent deaths for the first 13 months, which would correspond to 90,000 deaths. That's not directly comparable to the 24,000 figure in the UN survey, because the UN survey excluded criminal murders, but the first Lancet paper found that criminal murders were about 1/3 of the total. So maybe that's 90,000 vs. 36,000.

I believe myself that the Iraq war has killed hundreds of thousands, but probably in the low hundreds of thousands. The one time you can compare the 2006 Lancet paper with comparable numbers generated by a statistical sample we find it gives numbers that seem maybe 2-3 times too high. I don't think the reality-based community should be swearing fealty to this 400,000-900,000 confidence interval. The paper itself lists numerous reasons why its sample could be biased, and that could make the CI meaningless.

I think we can also probably expect Iraq Body Count to rip into this latest paper. There's bad blood between them and the Lancet team (particularly Les Roberts). Emotions aside, they'll probably make the sort of argument I just made. If they do it in good faith, they'll agree that another study needs to be done. But I suspect they'll try to dig the knife in.

"Tim Blair goes on to reject peer review."

Not just Tim: here's someone calling themselves daleyrocks over in comments at Greenwald's:

"I don't know what everyone is arguing about over this study. Of course it must be accepted. It has been PEER REVIEWED. That is the Good Housekeeping Seal of Approval for the left. It means unquestioning buy-in for absurd conclusions. Did you know that mankind is doomed in less than 15 years unless we act now? That's what 922 PEER REVIEWED papers from global warming alarmists would have you believe. "

Note that this same . . . connection . . . .pops up in Tim's comments, where someone argues that
"The objectivity of science is based not on peer review, but on the curiosity of mavericks who think a result is odd or interesting and check it out.

Climate change peer review works by killing curiosity."

Besides the 'we know it's wrong, how can we prove it' aspect, it wouldn't be surprising if the right is getting a bit peer-review-phobic: the theocons hear it in reference to evolution, the corporocrats in reference to global warming - it must be almost pavlovian.

Plus there's the whole 'if we can't have 100% certainty of Truth, then it's completely worthless, the same as guessing', which is partly stats illiteracy, but also part of a certain mindset.

So now it's the rightwing who are the reletivists? How amusing. Who would have thought that?

Emphasis on how little "what you believe" is affected by "what you know" works more against than for people in denial about how deadly our occupation of Iraq has been. Rejection of peer review and by implication the whole culture of the "tribe of science" [Stemwedel's term, I like it] is an admission that "how you come to know things" is, to conservatives, an arbitrary choice and faith alone can be relied upon as much as any set of facts. For some that equivocation of sources is necessary. Anyone's model and understanding of this world has been aquired at some cost, particulary the expense of finding you have been mistaken, and some people just won't spend another nickle on their beliefs or expose their current conclusions to the slightest risk: these we call conservatives.

Dan S:
Amused agreement here. I doubt any creature since the beginning of time has had the luxury of operating from a basis of complete information and certainty. There is not now one brain of any kind on this planet that is wired by birth to hold off action until it has all the data...lotta selection pressure against that. These constructed certainties are tissue thin to those who do not put them on as if they were armor.

"So now it's the rightwing who are the reletivists? "

Kinda-sorta, at least rhetorically. See for example Mat Bryan on Are Conservatives the New Postmodernists? or the Carpetbagger Report post Purveyors of doubt: Intelligent design, relativism, and the postmodern right, or just the infamous reality-based remark . . . and coturnix on why this might be the case.

greensmile - "tribe of science" is quite good, indeed - but can I borrow "some people just won't spend another nickel on their beliefs . . ."? 's awesome. :

"There is not now one brain of any kind on this planet that is wired by birth to hold off action until it has all the data"

Granted, I have been accused of being as close as we've yet come, but I dunno . . .

Did anybody else get the impression that there were an unusual amount of comments re: Lindsey's appearance over at the earlier Lancet thread?

Conservative is supposed to mean "playing it safe." Therefore, today, the true conservatives are liberals.

Please tell me there's an action figure that goes with that Quine card.

Analytic philosophy and liberal politics? Not even close Ms Philosopher Queen.

1 million PEER reviews papers are published every year, are you saying that they are all correct? because if you are you are the biggest fool to walk this planet.

http://www.senseaboutscience.org.uk/index.php/site/project/29/

True Liberals are not relitivists or post modernists, One of the greatest "liberal" thinker of all time, the great David Hume is actually the father of modern conservatism. Another GREAT liberal thinker is the greatest philosopher of Science in the history of mankind, namley Karl Popper. He solved Humes problem of induction with FALSIFICATION, or in simple terms for simple philosophers, every thesis has to admit its anti-thesis to be a genuine scientific statement.

Now you and your post-moderist relitivist "philosphers" point out in the lancets report where it has been falsified?

-s-

Lord, I'm invisible. Look folks, there really was a UNDP/ICLS survey conducted in Iraq in 2004 and for the first 12 months it really did find that war-related violence (excluding criminal murders) was between 19.000 and 28,000 for that period (midrange estimate was 24,000) and while that was only a bit lower than the comparable figure for the first Lancet paper, it's much lower than the approximately 90,000 violent dead figure you'd get from the recent Lancet paper. Admittedly that 90,000 would include criminal murders (as opposed to coalition or insurgent violence--Shiite death squads weren't the huge problem just yet.) You can google all this. You can find discussions (endless ones) at Tim Lambert's blog in 2004 through 2006, before this latest paper came along.

It's nice to set up a morality play with good liberals with SCIENCE on their side and bad conservatives who uphold the power of ignorance, and there are plenty of conservatives willing to fit that stereotype (the vast majority, IMO), but in the really reality-based world it's not at all clear that this latest Lancet paper makes sense. It might, you know, be wrong. The true excess death toll could be much lower--my own personal belief is that it's still in the hundreds of thousands and I'm normally the one yelling about American indifference to US-inflicted casualties in Iraq and why one can't trust Iraq Body Count statistics, but there really is reason to be somewhat wary of this 400-900,000 confidence interval. Honest.

>>but there really is reason to be somewhat wary of this 400-900,000 confidence interval. Honest.<<

I suppose its a bit like sending the tax man an estimate for you yearly earnings, you can guess with some certiainty you are going to get a bill for the top end 900,000 and a tax investigation to see what has gone wrong with your accounting methods.

1 million PEER reviews papers are published every year, are you saying that they are all correct? because if you are you are the biggest fool to walk this planet.

[Sense About Science's peer review page]

FWIW, SourceWatch includes Sense About Science on their list of front groups. Anyway, their little guide to peer review seems quite good (although I should add that I've never been involved in the peer review process, and have only general familiarity with it):

"When research findings have been reviewed and published in a scientific journal, this indicates they are sufficiently valid, significant, and original enough to merit the attention of other scientists . . . peer review is a kind of quality mark for science. It tells you that the research has been conducted and presented to a standard that other scientists accept.

I don't think anybody - at least anybody who knows what they're talking about - is claiming that peer review = unquestionable truth, that all peer-reviewed papers are correct. I think the main point is that this isn't, on the face of it, some incredibly shoddy piece of hackwork using a discredited methodology and lacking all credibility, but - unless something has gone quite wrong, and that's not the impression I get - a solid-enough scientific study, which needs to be addressed on its merits, not simply dismissed due to signals from one's gut. Frankly, all my gut ever tells me is that it's hungry. Do the current crop of will-and-gut conservatives have an extra brain in their gut, sort of like dinosaurs? (yes, yes, I know, it wasn't actually a brain . . . ) Or are the last few years really just the result of powerful people mistaking indigestion for insight? Who knows? Anyway, I'm off to eat some mac&cheese (or perhaps invade Iran).

The comments to this entry are closed.