« Disagreement Case Study 1 | Main | This Is My Dataset. There Are Many Datasets Like It, but This One Is Mine. . . »

February 27, 2007

Comments

I am an outsider to science, but a relatively close observer to one field, cryptography, which is in some ways more like mathematics or software development. From my perspective, I don't recognize the phenomena Robin describes. It seems to me that researchers seek citations of their papers as a mark of distinction, and that the field pushes forward in areas which are producing interesting results.

I don't know of any crypto papers that are admired for their eloquence. Mastery of detail, if manifested as a very complex paper with very complex proofs, has led to some bad results where incorrect proofs stood for several years, leading to something of a backlash against excessively complex proofs.

The most cited papers, particularly from a decade or two back, are the seminal ones that have established highly fruitful lines of investigation. I would think that it is the goal of every academic to write such a paper and secure his career. It doesn't seem to me that the kind of trumpery that Robin describes would go far to advance this goal. Ultimately, substance counts.

Hal, there are many kinds of eloquence. Almost by definition the most cited papers are "fruitful" and "interesting" in the sense of being mentioned by many future papers. I would say that is part of the process of validating their authors as impressive.

I think this overstates the scope of the problem. Researchers are human, and as such they are subject to human frailties. But there are people in this world who really do buy into the Enlightenment idea of seeking the the truth, and really are willing to pursue the truth even at some cost to themselves, and these are the people who are most likely to be attracted to careers in research. Moreover, there is a virtuous cycle here: the fact that people who are willing to sacrifice for truth concentrate in research means that such people will have an influence in determining what does and does not confer status, which increases the likelihood that status-seeking and truth-seeking will be roughly compatible.

David, no doubt there are some people more "willing to sacrifice for truth," and arenas of human behavior with more such people may well make more progress toward truth. But the question is why we should think that academia contains more such people, or that such people have greater effect in academia.

Bitter much?

Aaron, this topic goes to the heart of the question of when you can reasonably disagree with academic experts. The less you think ads are objective evaluations of product quality, the more you will feel free to disagree with those evaluations, at least if you have reason to consider yourself more objective. Similarly for academia.

You know, I had a serious response typed up here, but your post doesn't deserve it. It so reeks of bitterness and agenda and is so divorced from what one has to suffer through to obtain the marginal reward of an academic position that I'll leave it to others to mire their way through your mis- and preconceptions.

Aaron, I hope when you calm down you'll find time to explain yourself.

Prof. Hanson --

Two statements that seem inconsistent, at least on the surface:

1. "David, no doubt there are some people more 'willing to sacrifice for truth,' and arenas of human behavior with more such people may well make more progress toward truth. But the question is why we should think that academia contains more such people, or that such people have greater effect in academia."

2. This post: http://www.overcomingbias.com/2007/01/godless_profess.html
"If all we know about a view was that professors held it more, and elite professors even more so, we would be inclined to favor that view."

Is there something that I'm missing? Why aren't these two statements contradictory?

Stuart, academia can be an important store of knowledge in our society, even if it is not an especially efficient institution for increasing that knowledge store.

This reminds me of Margaret Mead's Coming of Age in Samoa, which was thoroughly debunked by Derek Freeman in 1984, but Mead partisans never recanted. How many Marxist professors ever said 'all that stuff about the superiority of socialist economies I did in the 70's, it was wrong'? People stake out a position, and after a certain time only death of the professors brings change, not new data, or new theories that contradict old approaches. So you are selling to students who have not made up their minds set indirectly, by making arguments that qualify as 'solid research' according to a set of conventions that the old professors deem 'scientific'. The key bug in the process is the finite lives of scientists, and the value of reputations, because this makes switching costs for old-timers much higher than for youngsters.

So, in the context of the earlier discussion about God, you'd be saying that professors might not be any better at advancing in their knowledge of God, but that their superior knowledge arises from accumulations of past advances? But how did those past advances come about?

Aaaactually, consumers did revolt against tasteless apples. That's why there are now choices in apples
in the supermarket at all. Until 20 years ago you had about two varieties of apples in most supermarkets,
um, red and green. Then all the sudden people woke up and figgered out that no crap those apples from
south of the equator taste good. Suddenly the apple contingent in WA state went bananas (pardon the phrase)
trying to make something other than red delicious. And within the span of no lie about two years the whole
country had many varieties of apples in the supermarket. Weeelll, the same thing is happening to funding
agencies when they have enough money. If the money is just tight tight tight, we can see that they will
always make the safe bet, look to the overachievers and assign dollars to zero risk projects with assured
payoffs. Even if the payoffs are puny. When agencies rapidly accelerate, they begin to talk about
risk versus reward, and actually try to parse it in the proposals' scores. There is at least some bit
of attention toward high risk high payoff. I don't think the premise of the post about progress being
a sideshow to the careerism is necessarily just plain inevitable. Under some conditions it won't be
100% inevitable. When fields are highly mature and when funding is not growing much, it probably is
inevitable, though.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31