As an evolutionary biologist working on theory, I feel trapped from both above and below.
From one direction we have everyone but biologists thinking that they understand evolutionary theory. Natural selection is such an easy thing to grasp, right? Offspring vary in size, speed, eyesight, and those that are better help the organism to have more offspring. Heritability ensures that evolution can take place. Creationists think they understand it, and while they are clearly off their rockers (i.e. they have a foregone conclusion, and that's not a way to do science, among other problems), at least scientists from other fields understand the premise of evidence. Except for philosophers (motto: "I think therefore I have something to contribute").
Jerry Fodor is a case in point. Listen to this conversation only if you have absolutely nothing better to do for an hour.
Rubbish. Very briefly, Fodor think pleiotropy (not that he knows that term) invalidates the theory of natural selection. Yeah, I know! Even ignoring that NS is presented as a theory, rather than just the mechanism of evolution that it is, saying that because two phenotypic traits can be linked (the example given is hearts pumping blood and making thump-thump noises) makes it impossible to talk about selection
for.
We should stress that every such case (and we argue in our book that free-riding is ubiquitous) is a counter-example to natural selection. Free-riding shows that the general claim that phenotypic traits are selected for their effects on fitness isn’t true.
You can read more reviews of Fodors book
here and
here. Suffice to say that I find it hugely annoying that philosophers who clearly have not understood the theory suddenly barge in with non peer-reviewed tripe to claim what evolutionary biologists are doing wrong. Call it arrogance, and see if I care. Call it arrogance when the plumber tells the mechanic to buzz off when the mechanic meddles with plumbing, or when the linguist calls the economist a fool for interfering in linguistics.
The point is just that if you want to switch field of work, please learn what there is to learn before you attempt to dismantle it. Otherwise, you wasting my fucking time! And I really think Sober thought so, too.
So that was the curse. Here is the damnation.
The field of biology is damned because it's hard, and because biologists aren't too clever.
Whoa, let me rephrase that.
The scope of what is to be explained is vast, and the complexity of it is vast. Physics (a field mostly considered harder than biology, because it uses mathematics) is mostly about natural processes that are very simple, and that makes it much easier to use complicated math to describe them, which is the triumph of physics, but also what makes it so hard to be good at.
Biology, on the completely other hand, is largely about counting frogs. Dissect the toad and describe what you see. Data collection, in other words. If we did physics that way, we'd go around observing apples falling from various heights of apple trees growing in different parts of the globe with slightly varying little g. And when we'd be done, we'd move on to pears.
The world of living things is so much more complex than the world of elementary particles, and that is what makes theory so hard, and why so little has been done so far. one way to advance theory in biology is by simulations (much used in physics, too), and it is here that I have met a fair amount of hostility in biology. The usual objection is that when you do simulations (and this could equally well apply to mathematical analysis) you construct a simpler model of the real world, by simulating, for example, only point mutations and natural selection in a static fitness landscape, and that's not realistic. And that is true. It's not realistic because it doesn't include every process we know of. Genetic drift, insertions and deletions, recombination and cross-over, gene duplication, transposable elements, other species, a changing environment, natural disasters, etc. etc. etc. But suppose we made a simulation like that, to the best of our ability, what would be gain? We could see that the output would be consistent with the real world, as far as we can observe it, and that would be good. But, like in physics, you have to keep some things constant in order to see what the effect of each component is. If you want to know the effect of point mutations, do make sure not to include indels and TEs, or you won't know which mechanism is responsible for you results.
Don't forget that understanding of nature
ends with theory, and that if we aren't allowed to theorize, then we're missing out. If everything we say about biology has to be backed up with data at every turn (and here I am mainly talking about getting papers published in peer-reviewed journals), then this enterprise of understanding nature might never get there.
P.S. I am not talking about trying to advance theory that is inconsistent with data, nor I am trying to dis backing simulations with data. </rant>