Monday, November 14, 2005

The bounds of discourse   posted by Razib @ 11/14/2005 11:58:00 AM

This an expansion on the post below. What is the problem when individuals like Alexandra pose their questions and critiques? You see the problem before you, I'm taking time out to talk about meta-science rather than real science. Time and resources are scarce. Alexandra threw out some contentions related to Fisher's Fundamental Theorem of Natural Selection. From what I can gather, she seems to be curious as how to evolution dependent upon variation (adaptation) can occur when natural selection exhausts that variation. Godless addressed many of the confusions due to the Theorem previously, and it is illustrative that the answers were directed at a very different perspective (i.e., the salience of evolution in regards to human beings). The key is that evolutionary biology is not fixed in 1930, and the relevance of Fisher's idea is disputed within the field. I am personally of the opinion that it is a good starting point, but that evolutionary biology will never be amenable to reduction into a deterministic science with an Ideal Gas Law analogy as Fisher would have liked. Mutation, migration and random genetic drift, as well as the reality that "fitness is a bugger," play important roles in modulating response to selection in a Fisherian context. I suspect Sewall Wright was right, in any reasonable time frame the landscape is often rugged. I'm being short, terse and unspecific because time is finite, and I'm peeved at having to offer this apologia when it would behoove anyone genuinely interested in the topic to read Jim Crow's Perspective: Here's to Fisher, additive genetic variance, and the fundamental theorem of natural selection (Crow is more positive than most in field of evolutionary genetics toward the Fundamental Theorem from what I can gather, including one of Crow's former post-docs who I am acquainted with).

Or look at this assertion:

Some scientists make the common mistake of assuming that all advantageous mutations are successfully incorporated into evolution.. In reality, almost all advantageous mutations are eliminated by genetic drift and play no role in evolution.

"Some scientists?" How to evaluate the impact of a qualitative assertion. Were there surveys done? Do physicists count in terms of their opinions as regards natural selection? As for the last assertion, it seems a garbling of the Neutral Theory debate. The probability of fixation of a mutation is 2s, where s is the selection coefficient, all things being equal. The probablity of a fixation of a neutral allele is 1/(2Ne), Ne being the effective breeding population. The relative power of selection vs. random genetic drift is a function of the relationship to each to effective population size, the larger the population, the more relative power selection has as generation-to-generation sampling error due to random genetic drift is reduced, the smaller the population and the noise from random genetic drift is liable to swamp out selection. Given a small enough population, even deleterious alleles can be fixed (for example, in inbreeding that occurs in a population due to its extremely small size, even if breeding is random). Motoo Kimura's insight via Neutral Theory is that most substitutions, that is, replacement of one allele for another in a fixed state, might not be due to positive selection, but rather neutral fixation. That does not imply that selection is irrelevant, rather, its power might be constrained to particular loci and regions of the genome. Assuming that the vast majority of the time the selection coefficient is less than 10%, well, by definition the majority of favorable mutations would not be fixed. This is naked algebra (and probability distributions), not a great mystery.

I am eliding over the issues relating to the character of the genome (psuedogenes vs, functional sequences vs. introns, etc. etc.), epistasis, frequency dependent selection, heterozygote advantage, variation of fitness over time and space, and the general shiftiness of fitness in a host of contexts. There are many books have been written about this topic, and whole careers have been devoted to it. And yet here we have Alexandra ambling on in and presenting her talking points as if they are important and we should address them. It is rather frustrating. I am skeptical that she was in good faith examining evolutionary genetics and asking us for pointers to the literature, she was telling us how things really were, as if the secondary and tertiary information she recieved was the last word. You ask Alexandra why I insult you instead of engaging you, the broader problem is I wish you would engage the extent literature before you waste, yes, waste, my time. Someone with your intelligence could assimilate the information in Evolutionary Genetics, and so obviate the need for me to respond to your queries. More importantly, the basic concepts and their analytic elucidation take time and effort, and you can not assume that the questions you ask in 1-4 sentence formats really have 1-4 sentence responses, they do not. There are many parameters in these models, and each parameter is scaffolded by qualifications and conditionalities. For every minute of your time spent asking, there goes 15-30 minutes of my time responding. This doesn't seem fair to me. From your prespective I am sure it seems rude of me to insult you, but I have spent many hours studying population genetics and evolution, and it is an insult to the work I have put in to knowing this area to see that you expect that I should respond in a timely fashion to what seems to me like an uninformed query. To ask good questions in science one must already know many answers.

One problem I have with this issue in regards to Creationism is that I am skeptical that most non-evolutionary biologists, let alone a lay person, would know about Fisher's Fundamental Theorem. No matter the reality that evolutionary biologists on the whole don't see the theorem as explanatory in a general sense, use of "technical terms" like Fisher's Fundamental Theorem of Natural Selection seems to me a question of style and bluster rather than substance. I've seen this before, the use of the Second Law of Thermodynamics to disprove evolution. Today, we see it in irreducible complexity and no free lunch theorems. I think this sort of tactic is shit, it undermines our trust and value of genuine technical knowledge by its obfuscation through use in ideologically driven models. All science is to some extent politicized, but instead of being part & parcel of the method and dynamic of the scientific ecology, these tactics are more likely to yank the keystone species (trust) out of the network and send the system crashing. Creationists and Intelligent Design folk complain they don't get funding, well, I just heard that a recent NSF grant panel approved funding for 10% of the applications in the area of evolution and ecology, so don't come crying. If Creationists and Intelligent Design people want money from the government, they should find a way to make it relevant to human health so that they can tap the NIH money train.

Moving from this specific case to the general point, there is the persistent problem of how to judge the veracity of one's comments if one does not have specialized knowledge in a given field. Again, this is a topic of wide berth. Those of you who read this blog know that I try to be a "shit filter." If I don't know about a topic, but feel I should, I try to read some monographs. Being a non-specialist, and having finite time, I don't get far, but I try to make the best of it. Sampling error is going to be high, I'm only going to read a max of a dozen books on a topic, and a dozen papers. Nevertheless, I also try to take the temperature of the field and try to impart to readers of this blog what I perceive to be the dominant paradigm. I don't expect readers to be the same, time is finite and we allocate it according to our personal preferences and utility functions. Most readers of this blog are pretty intelligent, and I suspect they are used to being able to bluster and bullshit their way through everday discourse, but remember, that shit don't fly here homey, do your fucking homework. Don't repeat pointless cliches in language that connotes erudition, I am good at picking out historical errors and evolutionary mistakes, and there are dozens of readers here with specializations deep enough that they can pick you apart if you don't have your shit together (I know, it happens too frequently to me when I post). It's a good thing when you can admit you are wrong, a bad thing is being stubborn and refusing to accept the judgement that you overextended yourself and aren't in the same comfortable intellectual milieu where everyone agrees with you, or, where everyone is less intelligent than you.

Who judges what? That is the question. Here, I'm the final judge. I sure make a lot of mistakes, but the end, you're going to have to have faith. If you don't like the rules, no one is forcing you to play, and you certainly aren't invested in anything aside from a few minutes of your day (your investment is far less than the time and energy I've put into this weblog over the years).

Time is finite, and there goes 30 minutes.