Contra Darwinism: A Debate

In Mind & Language, a more detailed article by Fodor on his objections to Darwinism (with replies from Daniel Dennett, Peter Godfrey-Smith, Elliott Sober, and Kirk Ludwig and Susan Schneider, and a rejoinder from Fodor). From Fodor's article:

An analogy (in fact, I think, quite a close one): For each person who is rich, there must be something or other that explains his being so: heredity, inheritance, cupidity, acuity, mendacity, grinding the faces of the poor, being in the right place at the right time, having friends in high places, sheer brute luck, highway robbery, or whatever. Which things conduce to getting rich is, of course, highly context dependent: It’s because of differences in context that none of us now has a chance of getting rich in (for example) the way that Genghis Kahn did; or in the (not dissimilar) way that Andrew Carnegie did; or in the (quite different) way that Andrew Carnegie’s heirs did; or in the (again quite different) way that Liberace did; and so forth. Likewise, the extreme context sensitivity of the ways of getting rich makes it most unlikely that there could be a theory of getting rich per se; all those how-to-get-rich books to the contrary notwithstanding. In particular, it’s most unlikely that there are generalizations that are lawful (hence counterfactual supporting, not ad hoc, and not vacuous)³⁸ that specify the various situations in which it is possible to get rich and the properties in virtue of which, if one had them, one would get rich in those situations.³⁹ This is, please notice, fully compatible with there being convincing stories that explain, case by case, what it was about a guy in virtue of which he got as rich as he did in the circumstances that prevailed when and where he did so.

I think adaptationist explanations of the evolution of heritable traits are sort of like that. When they work it’s because they provide plausible historical narratives, not because they cite covering laws. In particular, pace Darwinists, adaptationism doesn’t articulate the mechanisms of the selection of heritable phenotypic traits; it couldn’t because there aren’t any mechanisms of the selection of heritable phenotypic traits (as such). All there are is the many, many different ways in which various creatures manage to flourish in the many, many environmental situations in which the do so. Diamond (in Mayr, 2001, p. x) remarks that Darwin didn’t just present ‘a well-thought-out theory of evolution. Most importantly, he also proposed a theory of causation, the theory of natural selection.’ Well, if I’m right, that’s exactly what Darwin didn’t do; a ‘theory of causation’ is exactly what the theory of natural selection isn’t.

From the viewpoint of the philosopher of science, perhaps the bottom line of all this is the importance of keeping clear the difference between historical explanations and covering law explanations.

Comments

[Carlos]

Again? I'm still exhausted from watching people beat this to death last time.

[Phillip]

People have been beating this to death since Darwin published the book.

[McFawn]

This isn't your cliched Darwinism 'round and 'round. While this short blurb is a little hard to follow without reading all that it is responding to, the question about what kind of model of causation natural selection represents is central.

Does it simply give a comphrensible backstory to the organisms now on earth? Or does it actually explain the mode or mechanism of how those organisms came to be?

Darwin was really doing the former. Natural selection does not explain the mechanism of selection, rather it says that what we see now are surviving traits, which survived for all different reasons.

Seen in this way, Natural Selection becomes harder to refute, because it sounds more like a common sense observation rather than a radical new law of creation.

[Nick Smyth]

If anything, I hope that the posting of this article will give second thoughts to anyone who still thinks they can talk intelligently about evolutionary psychology without engaging with philosophy of mind.

[Carlos]

While this short blurb is a little hard to follow without reading all that it is responding to, the question about what kind of model of causation natural selection represents is central.

No, I mean this whole catfight between Fodor, Dennett, et al has already been through hundreds of responses during two previous visits to the original article and all the rebuttals here on 3QD. Now there are more rebuttals? Fine. I'm still exhausted. Chris? You still game?

[fred lapides]

I certainly have no background to argue about this sort of thing, but in para one, I like the
Donald Trump way of wealth. His daddy gave him ten million and said "now make something of yourself." It worked!

[Chris Schoen]

Yeah, I'm pretty pooped too. I think I've said my piece on this.

The original back and forth between Fodor and Dennett has been kicking around on the tubes for a year or two, (and IIRC, 3QD linked to those early drafts) but as best as I can tell Fodor's latest response (linked here) is new.

I say more power to him, but he's not going to get much traction until someone can come up with a better model (maybe the neo-Goetheans like Brian Goodwin). What Fodor's doing now is a little like saying we can logically show that the sun doesn't go around the earth, but we don't know why it looks that way.

Personally, I'm happy with the explanation of evolution being simply "things change," but people do like their explanations for things, and until they get one, I think Natural Selection is here to stay.

[aguy109]

Fodor’s argument amounts to an absurd, almost childish, denial of elementary statistics. He’s like the confirmed smoker who declares that smoking doesn’t cause death, because he once knew another smoker who was run over by a truck.
“..there aren’t any mechanisms of the selection of heritable phenotypic traits “
Rubbish! Hundreds of studies have been made on the inheritance of drug resistance amongst bacteria, predator selection of bugs on different color backgrounds, and so on
See the following link for an example of rapid evolution in sticklebacks in a Seattle lake, which evolved thick body ‘armor’ over just a few decades:
http://news.nationalgeographic.com/news/2008/05/080520-fish-evolution_2.html

[Picador]

Fodor's beef here seems to be that evolutionary biology has turned out to be a function of history, not Aristotelian logic. Organisms are how they are because of the history of their forebears and how they interacted with each other and the world. Fodor complains that evolution does not have "laws" the way Newtonian physics does, but only "laws" the way that history or economics do, which is to say highly contingently. To which one can only respond: duh. This doesn't mean that history is not a valid discipline, or that economists cannot distill principles with predictive power from their research.

Fodor says:

I think adaptationist explanations of the evolution of heritable traits are sort of like that. When they work it’s because they provide plausible historical narratives, not because they cite covering laws.

Sure, but you can make the same criticism of economics, or even physics. Each individual economic choice is made for context-specific reasons, but patterns emerge when you look at a large enough sample. For that matter, if you examine the behavior of particles crossing a membrane or objects settling to the bottom of a pile, you'll witness plenty of context-specific behavior that seems to defy a general "law": denser objects will sometimes find themselves buffetted upward for various reasons. Yet a law emerges from a large enough sample in the right overall conditions (e.g., gravity pulling in a specific direction more often than not).

It all does seem to hinge on a weak understanding of statistics.

[NIck Smyth]

Picador, do you understand what Fodor means when he says that a valid scientific theory has to posit causal laws, and that causal laws have to support relevant counterfactuals?

Do you know what a "counterfactual" is?

Just curious. In fact, I am extremely curious to know how many online discussions of evolution proceed without any knowledge of this sort. I'm willing to bet that it's close to the 99.9%-range.

[aguy109]

Nick, the statistical and probability issues that Picador and I have mentioned show that Fodor is wrong in demanding that “a valid scientific theory has to posit causal LAWS” because the word “law” as Fodor uses it means a cast iron rule that means that one thing ALWAYS causes the other.
When, in the late 19th Century Bolzmann first introduced his gas equation pV=nRT that links Pressure Volume Number of molecules of a gas and Temperature, people objected, just as Fodor is doing today with Natural Selection+, that this equation does not relate to the activity of INDIVIDUAL MOLECULES of gas when the temperature changes (something that is hard or impossible to measure), but only to the average activity of many millions of molecules.
Actually, however Bolzmann’s equation works pretty well in the internal combustion engine and a host of other examples, so no one objects to it anymore, and the use of probabilities, degrees of freedom, confidence levels and other basic statistical tools is basic to all quantitative scientific research , including Biology.
For example, as tree trunks in England became darker in color due to air pollution, pale colored moths resting on the bark became more visible to birds, who began picking them off in greater numbers. Darker colored individuals of that same species of moth were harder for birds to see, so the darker varieties became more common and a small evolutionary step was therefore demonstrated. These are not just “stories” – prey selection by birds has been extensively tested under controlled conditions.
Again, as with the gas molecules, NS does not predict that ALL light colored moths will be eaten and ALL dark colored moths will survive to reproduce, NS only deals with probabilities. Even a slight differential survival rate is enough to ensure selection of a favorable trait, because it is greatly amplified by the moths’ rapid breeding rate.

Fodor’s claim that evolutionary biologists don’t look at counterfactuals (opposing or contradictory evidence) is unfair, because
A – They do
and B, if they don’t look hard enough for alternative causes of a particular evolutionary or population change, their colleagues and competitors certainly will
Another fault in Fodor’s talk of ‘counterfactuals’ (ugly redundant word, Nick, you can keep it) is that the probabilities in the NS process are roughly quantify able (if they can be tested). It is not a case of all or nothing, but of a number of factors that effect the survival of a species. This presents major challenges to methodology, but it doesn’t mean that Evolution is simply "things change," as Chris thinks - meaningful conclusions can be drawn.
The most that can be claimed for Fodor’s arguments is that sometimes some researchers or popularizers of Science may talk too loosely about defining some genetic or other factor as the cause of a particular trait, especially in the field of human behavior.
I apologize for going on and on, about what must be well know to a lot of people reading this, its just that Fodor’s revival of an abandoned 19th Century argument seems so incongruous.

[Nick Smyth]

Carlos,

Please do go "on and on", I do so enjoy discussing these things.

While I understand what you're saying, I really don't think the analogy to Bolzmann is disingenuous. That law is testable, which means it supports counterfactuals (a word in common use in the philosophy of science and causation in general, so yes, I'll keep it). The law works because when you change the initial conditions (temperature, for example) you know in advance how different the result will be. Nothing even REMOTELY this predictive exists in evolutionary theory.

Of course all kinds of sciences make probabilistic predictions, but that doesn't preclude them adhering to a very high "testibility" standard.

There's a more fundamental difference here: the variables in Bolzmann's law are isolated from one another, which is what makes it testable in the first place. The temperature of the gas is not coextensive with the number of molecules or the volume of the container. They're genuinely distinguishable variables.

Fodor's basic point is this: the traits of organisms are not distinguishable in this way: they are sometimes co-extensive. So, suppose we posit some (probabilistic) evolutionary law that says "in environment E, animals that suckle their young are more likely to successfully reproduce". The difficulty here is that the trait in question (suckling) comes along with a host of other traits that occur in exactly the same organisms (and no others): having hair, having sweat glands, having a neocortex region, having three inner ear bones... etc.

All Fodor is saying is that even IF we establish the probabilistic law, we cannot know which of these traits has been selected for. It's not just difficult, it's impossible! In what sense, then do we have a genuine "hypothesis" or "probabilistic law" at all?

[Chris Schoen]

Even in the case of the peppered moths, the Fodorian view would be that it is plausible that less pigmented moths being more visible to predators resulted in their decreased fitness, but not demonstrable that this was so. We would have to know an unknown number of contributing factors to say so with certainty.

For example, since Kettlewell's original study, several assumptions about its findings have been challenged: e.g. "Do peppered moths habitually rest on tree trunks in the daytime? Do they do so in an "exposed" way (e.g. not in the crotches of branches where they're harder to see)? Do their predators rely on sight (in human-visible wavelengths) to locate prey? These challenges have largely been shown not to effect the reasonable conclusion that darker moths have a fitness advantage, at least with regard to the specific threat of bird predation, but how will we know when the last of these challenges has been fended off--when the last possible variable has been exhausted?

Now pretend for a moment that it has been discovered that "differential bird predation" was not the cause of the changes in allele frequencies in the population. Would this be considered by anyone evidence against natural selection? Or just a failure to find evidence in favor of it.

One thing that is well observed about life is that--unlike, say, molecules of argon--it is enormously creative. We cannot know in advance how organisms will respond to various environmental forces (or even what those forces might be). The light-colored typica moth is still abundant in the world, having survived the worst of melanic bark pollution. What exactly has been explained, and by what mechanism?

[aguy109]

Gentlemen, you are merely bringing up the old philosophical preocupation with knowledge: how can we know anything "with certainty"? This is like other mythical things, the "Perfect Sphere" the "Perfect Wife" and so on. In the real world, you measure those parameters that you can measure, using the limited time and research budget at your disposal, and you use accepted tools of statistical analysis to test your results against your research hypothesis. If you do it right and have a bit of luck (yes -LUCK!!!!) you can show that one of those factors is significant to a 90% degree of certainty (for example). Someone else can then try to support or contest your findings. This is not perfection, but it is science and its better than shrugging your shoulders and saying "who knows".
"Nothing even REMOTELY this predictive exists in evolutionary theory." Don't agree, what about say, the accuracy of predictions that bacteria resistant to antibiotics would spread in hospitals where those specific antibiotics were over used? Those predictions are confirmed daily. Another example: 50 years ago, paleontologists predicted that fossils would be found that supported the theory that birds are closely related to dinosaurs. Such fossils have since been found.
Nick, you contend that "the variables in Bolzmann's law are isolated from one another, which is what makes it testable in the first place. The temperature of the gas is not coextensive with the number of molecules or the volume of the container. They're genuinely distinguishable variables."
No, 'Pressure' is just a measurement that is easy to make using 19th Century apparatus, like bell-jars, tubing, mercury manometer etc. Pressure is caused by molecules bouncing against the chamber walls - A higher "temperature" means that the molecules go faster and bounce harder, so increasing the Pressure. And the "speed" of the molecules is not an actual speed of a particular molecule, but the root of the mean of the square of a measurement of their speed. In short, if you looked at an individual molecule, the properties leading to P and T would be no less coextensive than any 2 features of an animal.

[Chris Schoen]

Carlos, I guess I wasn't all that exhausted.

Aguy,

If you read the linked articles, Fodor is very explicit that he is not making an epistemological argument about what we can know about the selection pressure on a given trait (or organism, depending on where you think selection "happens.") That is, his concern (as I understand it) is not "How do we know which of two coextensive traits was selected for" but rather, On what, exactly is selection acting, by what mechanism, and how is this generalizable in a way that is not so diluted to be tautological or meaningless?

I'm neither as smart or erudite as Fodor; to me the epistemological component is compelling, but I agree it doesn't refute natural selection, just locks it in a black box.

Back to the argument. In the case of husbandry, there is a factual answer to the question of what was selected for, whether we breed ears of corn, puppies, or livestock. The breeder purposefully isolates the environmental pressures on (or, domesticates) the organisms being selected, directing their phenotypes toward a desired goal.

In nature, no such isolation exists, meaning that evolutionary change must be contingent, or context-sensitive. Probability analysis does not rescue us from this problem; it merely says that on average, rather than in every specific case, a given trait is fitness-enhancing. The question of why it is so, rather than fitness-decreasing, or neutral, is no closer to being answered.

What happens in adaptationist explanations, Fodor argues, is that both plausibility and history suggest the causal story, and natural selection is appealed to to burnish it with scientific respectability. That doesn't make the explanation false, in itself, but if you remove history and plausibility, natural selection alone has very little to say on how the leopard got its spots.

There is no law, or general principle, for example, governing camouflage. Some cryptic organisms employ it to evade predation, some to stay concealed when stalking prey. Some organisms employ it variably, to blend in with diverse environments, some only statically. Some organisms make do just fine without it. How does natural selection explain the variable fitness of crypsis? How would we use it to predict which organisms not currently carrying a cryptic trait might acquire one in the future, or which ones might lose one?

The reason that predictions about bacteria resistance are not good evidence for natural selection is that the resistance was observed before it was predicted. The same with peppered moths. If, noticing the darkening of trees in Manchester in the late 19th century, someone postulated that some moths would mutate to match the bark pigment, that would be one thing. But the pigmented moths were observed first (before Darwin, in fact) and the explanation came afterwards.

This is what Fodor (and others) mean when they say adaptationist stories are always post hoc. It's easy to say how an organism (or genome) responds to environmental pressure once the pressure has been identified and the response effectuated. It's much more difficult, if not impossible in principle, to anticipate what environmental pressures will transpire, and what the phenotypic response will be.

At the very least Fodor's argument (which I confess I'm not sure I completely understand, particularly the part about distinguishing counterfactuals of intentional and intensional states--yikes) should caution against overconfidence in adaptationist explanations. I'm fine with a "weak" Darwinism that says that permutations arise in populations, some to persist over time, some not. But the strong version that relies on metaphors of engineering and competition is becoming increasingly untenable.

[John M]

I like the standpoints of Chris S and Nick S. Intelligent (contested word, apologies) critique of the nostrums of 'Darwinism' is good stuff.

I think the falsifiability argument Nick makes is strong: if Darwinists are so strong in lambasting the anti-science of others, the least they can do is supply arguments that satisfy one of the key tenets of modern science. If you can't, in principle, prove wrong a proposition, then it may be common sense but it won't be science.

And how we even got to discourse that allows the phrase 'phenotypical behaviour' in relation to humans is disturbing. It's great that someone is saying "tell him how the frickin' mind works /qua/ mind, before telling me where our 'behaviour' comes from!".

What I hear is that when crooks are in the cell the night before trial the motto goes "work out the story and stick to it no matter what". It's really troubling that modern Darwinism seems to be trying the same.

Darwinists: let's have falsifiability and robust causality (not just explanatory narratives), otherwise consider yourselves pretenders in the ante-chamber, not nobles in the high court, of science.

[RXDIGG]

Category Antibiotics
http://neededpills.com/order_antibiotics_en-us.html
Buy Generic Biaxin (Clarithromycin)
http://neededpills.com/buy_biaxin_en-us.html
What is Biaxin (Clarithromycin)?
Biaxin is in a group of drugs called macrolide antibiotics. It fights bacteria in your body.
Biaxin is used to treat many different types of bacterial infections affecting the skin and respiratory system. It is also used together with other medicines to treat stomach ulcers caused by Helicobacter pylori.
Our official website http://www.NeededPills.com/