Logomacy
Between logomachy and logomancy . . .

EconLog, Alcohol and Non-Linear Dosage Effects, Bryan Caplan: Library of Economics and Liberty Annotated

tags: statistics

Our years overlapped, but when I was an undergrad at Berkeley, I never met Aaron Wildavsky. My loss. Here’s a great passage he wrote (along with Adam Wildavsky) for Henderson’s encyclopedia:

Another questionable assumption is that cancer causation is a linear process, meaning that there is no safe dose and that damage occurs at a constant rate as exposure increases. This is known as the “linear no-threshold hypothesis.” Scientific evidence increasingly shows that there are indeed threshold effects.

His example:

Consuming two gallons of 100-proof liquor is an hour would be enough to kill most of us. If the linear no-threshold hypothesis applied to alcohol, one would expect that if 256 people consumed an ounce of liquor each, then on average one of them would keel over and die. It would only be a slight exaggeration to say that were the EPA to regulate ethyl alcohol… the same way that it regulates other chemical compounds, we would each be limited to sixteen-millionths of an ounce per lifetime.

No particular comment here, I just thought it was an interesting observation. And yes, I do plan on drinking a beer tonight

Gagarin was not the first cosmonaut - Pravda.Ru

As 40 years have passed since Gagarin’s flight, new sensational details of this event were disclosed: Gagarin was not the first man to fly to space. Three Soviet pilots died in attempts to conquer space before Gagarin’s famous space flight, Mikhail Rudenko, senior engineer-experimenter with Experimental Design Office 456 (located in Khimki, in the Moscow region) said on Thursday.

update: My friend Mac, who follows stuff about the Russian space program closely, doubts the authenticity of this story.  She says that these claims have been floating around for years, but there’s no new evidence presented here.

Ever since I decided, some time ago, that Wikipedia had serious problems in accountability and transparency in their meta-editing process, I’ve derived a certain sour satisfaction from articles like this:

Secret mailing list rocks Wikipedia | The Register
Controversy has erupted among the encyclopedia’s core contributors, after a rogue editor revealed that the site’s top administrators are using a secret insider mailing list to crackdown on perceived threats to their power.

Many suspected that such a list was in use, as the Wikipedia “ruling clique” grew increasingly concerned with banning editors for the most petty of reasons. But now that the list’s existence is confirmed, the rank and file are on the verge of revolt.

But this is a clear case of confirmation bias. I’m looking for information that confirms my low opinion of Wikipedia admins, while mostly ignoring the plentiful articles on how Wikipedia is the greatest thing since sliced bread and will change the world (in fact, at the moment every wikipedia page has a banner add that reads “You can help Wikipedia change the world!”).

Transterrestrial Musings

Thoughts On Objectivity

In both science, and journalism.

The notion that journalists are, or should be, or can be “objective” is perhaps the profession’s most fatal conceit. As Virginia Postrel says, what’s important is to be fair, something that they often don’t even attempt, as demonstrated by CNN and its performance in the debates.

I’m dubious. Even granting that “objective” is something of a term of art among journalists that doesn’t quite correspond to what a philosopher or scientist might mean in terms of attempting to avoid prejudice, bias, or wishful thinking, I don’t see how you can aim at fairness without first being able to assess what parts of your reporting might be unfair. And to do that, it seems to me you have to try to be objective…unless you’re just going to reduce everything to a procedure, as in “he said; she said” journalism.

From my point of view, the problem isn’t that journalists try to be objective where they should be trying to be fair–it’s that they’re so damn bad at objectivity. And it doesn’t reassure me that fairness over objectivity would be an improvement when the biggest critics of objectivity as a journalistic goal (e.g. Chomsky) want to downplay it precisely so they can hide their biases and better achieve their agendas. “Fake but accurate” is exactly what that approach is trying to legitimize. It is unfair that the journalist can’t present what he knows to be true based on his expert judgment, just because there’s no actual “objective” evidence. But because journalistic standards still require objectivity, he supplies fake evidence (and maybe even believes it to be true because of his biases), and with luck gets caught out. I say that if you believe that the journalist is obligated to provide the actual documents for other people to examine, and not just assert that they exist, you believe in objectivity not fairness; you believe that there is a truth of the matter that can be gotten at through examination of the evidence*, and not just a requirement to announce your biases.

Rand Simberg’s post above is in reference to a Virginia Postrel post on Objectivity. I haven’t read the book, but to infer anything about the appropriateness of objectivity as an epistemic virtue from a discussion of its history is to commit the genetic fallacy. I’m not at all sure whether the Daston and Gallison, the authors of Objectivity, would agree with Postrel’s take-away that “Real objectivity would turn the journalist into a C-Span camera, simply recording data without any sort of selection or pattern-making,” but I am sure that it is a core epistemic virtue for journalists to start by simply recording the data without any sort of selection or pattern-making. As the folks at Language Log have demonstrated over and over and over again, if you want the truth you have first accurately record what really was said. That doesn’t mean that you end there, and the journalist’s job is just to faithfully transcribe and then print it–but it has to start there.

* if it can be gotten at at all…

Marginal Revolution: Why Most Published Research Findings are False

There is increasing concern that in modern research, false findings may be the majority or even the vast majority of published research claims. However, this should not be surprising. It can be proven that most claimed research findings are false. - John Ioannidis

The argument is from a paper by John Ionnidis, but Alex Tabarrok gives a much easier to read analysis of the fairly simply Bayesian reasoning behind it. Essentially, this is the classic problem of false positives vs. true positives when the condition being tested for is rare in the population (e.g. presence of AIDs in non-high-risk groups, or in this case the truth of a hypothesis).

It might be tempting to argue that the case of a hypothesis under test being true isn’t typically as bad as the general assumptions being made to drive the argument, since the researchers presumably have some thought or intuition that drives them to pick a particular hypothesis to test (they’re not just throwing darts at a board), but consider that works both ways. Despite the common complaint that this or that study is “just another case of science proving what everybody already knows (and so a waste of money)”, I suspect very few researchers deliberately pick hypotheses that are widely believed to be true, particularly if there’s a lot of evidence and research backing up that belief. That’s not, generally speaking, believed to be the way to advance the frontiers of scientific knowledge. But in that case, the sample is biased in the other direction–a random hypothesis to test would include already-known-to-be-true hypotheses in the same proportion that they occur in the population of all hypotheses, so the hypotheses actually attracting attention are less likely to be true than random chance would dictate. Whether the scientist’s intuition towards selecting true hypothesis is a bigger bias than the elimination of all the ones believed to be true is something you can’t really be sure of, so I’d be really cautious about asserting that P(hypothesis is true) must be a lot better than Ionaddis’ calculations allow for.

In Trask’s Historical Linguistics there’s a very illuminating discussion (pp 78-82) of changes in phonological systems, in particular of Latin rhotacization–the change from an intervocalic /s/ to intervocalic /r/. There were several stages to this change, and the change was absolutely regular: every instance of pre-Latin intervocalic /s/ became intervocalic /r/ in Classical Latin. On the other hand, Classical Latin did have instances of intervocalic /s/, e.g. ecclesia, quasi, and visum. All of these turn out to have explanations, such as being borrowed after the shift had occurred (ecclesia, probably rosa), or not having intervocalic /s/ at the time of the shift (quasi, visum). Trask writes (ibid 82-83):

There are two important lessons here. First, a sound change normally happens at some particular time in the history of the language, and then stops. Consequently, the phonological history of a language consists of a series of changes, each acting on what’s left over from the last change. As a result of these accumulating layers of changes, the effects of earlier changes may be increasingly obscured by the effects of later ones. In our Latin example, various later changes reintroduced intervocalic /s/ into the language after the rhoticism had eliminated it; as a result we can’t immediately tell by looking at Classical Latin that the language had, centuries earlier, lost every single intervocalic /s/. We know this only because of patient and careful investigation by historical linguists.

Second, our policy of insisting that sound change must be regular is fruitful. If scholars had thrown their hands into the air and declared the troublesome words to be mere exceptions to rhotacism, there would have been no reason to worry about them. By insisting on their regularity, however, they were forced to find explanations for the odd cases, and, as you can see, they have been very successful in finding those explanations — and, as a result, they have wound up knowing rather more about the history of Latin than might otherwise have been the case. Even the few really nasty cases like rosa remain as puzzles to be investigated, and perhaps a future scholar will manage to find definitive solutions to these, too. But, without the regularity hypothesis as a guiding principle, there would be no reason for anybody even to look for such solutions.

The point I’m making about Intelligent Design¹ should be clear: ID is nothing but an exhortation for scholars (specifically biologists) to throw their hands in the air every time they come across something that seems puzzling. It’s not a scientific hypothesis, despite what its proponents claim, because it is anti-fruitful. It doesn’t require any kind of Popperian view of the philosophy of science or verifiabilty and falsifiability to see that saying “I don’t currently understand how this happened, so it must be beyond explanation” isn’t a research strategy at all. ID is sterile². To claim that biologists should take ID seriously is to claim that biologists should take something like Sidney Harris’s cartoon³ as a publishable result.
Life on Earth, too, is a series of changes, each acting on what’s left over from the last change. The hypothesis is, must be in order to be fruitful, that this is absolutely regular: following the laws of physics, chemistry, and biology, with no exceptions. As soon as you’re willing to throw your hands in the air and say, well this bit is inexplicable, you might as well pack it in; there’s nothing further that you’re going to learn, because you’re going to stop even looking. To do science, or really any kind of scholarly research, you have to start with the assumption “There’s an explanation for that”; ID’s “hypothesis” is “There’s no explaining that.”