Intelligent design redux

[axocanth]

Hint: Pick up a science textbook from the year 1900, say, from your local library (if you can find one).

What proportion of knowledge claims contained therein do you suppose are still countenanced today?

 

[James R]

Just quickly, for now...

Yes, the circularity pertains to the very core claim of natural selection theory, variously stated as "survival of the fittest" or something similar. If the fitter just are those who, or those who are more likely to, survive and reproduce successfully, then "Certain organisms survive and reproduce relatively successfully because they are fitter (cf. better adapted)" is no more of an explanation than "Some men are bachelors because they are unmarried".

The problem with this is the assumption that 'fitness' can only be judged retrospectively. If that assumption were true, then you'd have a valid point. Whichever organisms managed to survive and reproduce successfully would be the ones retrospectively judged to be the 'fittest'. In that case, the theory of evolution would reduce to the tautological "Survival of the survivors".

Fortunately, this isn't the crux of the theory of evolution. Rather, it is a misunderstanding.

The theory of evolution posits that, in any well-specified environment, certain phenotypical traits will tend to lead to greater survivability and (more importantly) greater reproductive success for individual organisms, compared to alternative phenotypical traits. This hypothesis has the advantage of being testable prospectively, not just retrospectively. That is, we can attempt to predict which set of phenotypical traits we think might be beneficial to a group of organisms in a particular environment, and then watch to see how things turn out over a large number of generations. How are we to make the predictions? Well, we can refer to basic facts regarding such things as competition for survival resources (e.g. ability of an animal to get sufficient food) and competition to reproduce successful (e.g. ability of an animal to live long enough to find a viable mate and to produce viable offspring with said mate).

In other words, the theory of evolution is not just an empty statement of the bleeding obvious ("Survivors survive"). It says useful things about adaptation to the environment and such, which have predictive power, not just retrodictive power.

 

[axocanth]

@ James above

To escape the circularity, fitness (or adaptedness) has to be defined without any reference to survival and reproductive success. I'm afraid you'll find, on close scrutiny, that even when you think you have escaped the circularity, survival and reproductive success have been smuggled back in without noticing it.

I think what you're suggesting above is that natural selection can be characterized as the more well adapted surviving and reproducing more successfully, and you regard this an empirical, testable hypothesis? Is this right?

But now we have to examine the phrase "well adapted". Is it possible, for example, for things to be well adapted, or well engineered if you like. and not survive and reproduce successfully? Is survival and reproductive success (call it thriving, if you like) not already implicitly contained in the phrase "well adapted"? If organisms are a miserable flop at surviving and reproducing, in what sense can they be said to be "well engineered"?

If you disagree, give us a hypothetical example of a population of whatevers (you choose) which includes well adapted organisms which do not survive and reproduce successfully. Is it possible -- as a general principle -- to be really well adapted and not survive and reproduce like rabbits? Sounds a lot like trying to conceive of triangles that don't have three sides to me.


P.S. The reason I keep emphasizing "as a general principle" is because, as noted earlier, this whole can of worms gets extremely nuanced indeed. A so-called "propensity interpretation of fitness" has been introduced in an attempt to escape the original vicious circularity (i.e. the fittest just are those who survive and reproduce best). This alternative interpretation, then, allows for less fit individuals to prevail over more fit individuals. The odds being in your favor at the casino doesn't guarantee success, kind of thing.

But taken as a general principle, I submit we collapse back into tautology.

 

[axocanth]

P.S. Be wary of testing hypotheses that require no testing. Testing does not guarantee the empirical nature of a hypothesis.

For example, you can go knocking on doors if you like to test the hypothesis "All bachelors are unmarried men" -- Look! I found one! Here's another!

I trust you agree this would be an exercise in redundancy. Go knocking on doors if you like, but the hypothesis requires no testing. It is true by definition.

 

[axocanth]

The theory of evolution posits that, in any well-specified environment, certain phenotypical traits will tend to lead to greater survivability and (more importantly) greater reproductive success for individual organisms, compared to alternative phenotypical traits. This hypothesis has the advantage of being testable prospectively, not just retrospectively.

"Polar bears with white fur [in a given ecology] survive and reproduce more successfully than polar bears with green fur (mutants perhaps! -- or we could paint them!)" is a perfectly respectable empirical hypothesis. No one denies this. It may be true or it may be false. Only empirical inquiry will tell us which. But the above hypothesis is not the theory of natural selection, of course.

The theory of natural selection is a general theory purportedly applicable to any organisms in any environments. Right?

Now state the theory in general form, please.


Note: This critique applies only to natural selection theory, not evolutionary theory as a whole (see quote at top).

 

[axocanth]

The theory of evolution posits that, in any well-specified environment, certain phenotypical traits will tend to lead to greater survivability and (more importantly) greater reproductive success for individual organisms, compared to alternative phenotypical traits.

And the critical question is: Which ones?

 

[James R]

If you disagree, give us an hypothetical example of a population of whatevers (you choose) which includes well adapted organisms which do not survive and reproduce successfully. Is it possible -- as a general principle -- to be really well adapted and not survive and reproduce like rabbits? Sounds a lot like trying to conceive of triangles that don't have three sides to me.

You seem to be asserting that the theory of evolution is an empty theory. That would imply that if we were to take 100 different organisms and place them into a particular environment, there would be no way to start trying to accurately predict which of the 100 would be more likely to survive or reproduce in that environment. To put it another way, either 'natural selection' is a meaningful term, or it isn't. Your argument seems to be that it isn't, because it merely refers to an ex post facto process of distinguishing those organisms that survived from those that didn't. It sounds a bit like you're suggesting that there's no evolution at all, really - just the fact of some organisms being lucky enough to survive while others are not, or something along those lines. That is, evolution isn't really a theory at all - it's just a way of pretending that what is actually entirely random is something other than that.

Is that what you think?

Turning to your question, the matter of "well adapted" is always something that is specific to a particular set of environmental conditions. One cannot generalise about it. Thus, your demand that I give you an example of "well adapted organisms" that aren't successful is either a mistake - because it assumes that there is only one set of conditions that make all things "well adapted" - or it setting a deliberate trap by asking for a demonstration that you already know is impossible.

In case it is a mistake, you might like to consider that there are a great many examples of organisms (entire species, even!) whose environments have changed around them, leading to their slow or rapid extinction. In fact, more than 99% of all the species that have ever lived are now extinct. But every single one of those extinct species was "well adapted" at one time in the history of our planet. What went wrong for them was that evolution wasn't fast enough to keep them "well adapted" as the environment changed around them. "Well adapted" for one set of conditions did not equate with well adapted for a different set of conditions.

Think, for instance, of a certain meteor strike that wiped out a bunch of dinosaurs. That's a rather extreme example; there are many others that didn't require such a radical change in environmental conditions.

 

[axocanth]

You seem to be asserting that the theory of evolution is an empty theory.

I'm saying the theory of natural selection is an empty theory.

That would imply that if we were to take 100 different organisms and place them into a particular environment, there would be no way to start trying to accurately predict which of the 100 would be more likely to survive or reproduce in that environment.

Not at all. In any particular case, we can -- at least in principle -- determine which particular traits are advantageous. It's an empirical matter.

What I'm saying is the theory -- being a vacuous tautology -- isn't going to help you one bit with this. Which traits are advantageous cannot be derived from the theory.

A vacuous tautology isn't going to help you do anything.

 

[axocanth]

You seem to be asserting that the theory of evolution is an empty theory. That would imply that if we were to take 100 different organisms and place them into a particular environment, there would be no way to start trying to accurately predict which of the 100 would be more likely to survive or reproduce in that environment. To put it another way, either 'natural selection' is a meaningful term, or it isn't.

It's perfectly meaningful . . . in the same sense that "dogs are dogs" is. No one has any difficulty understanding what "dogs are dogs" means.

I'm saying both are meaningful but devoid of any empirical content. Put another way, they are truths of language, not truths of the world.

 

[James R]

axocanth:

I think you're wrong about natural selection being an empty theory - an idea with no content, so to speak.

Have you read On the origin of species?

In it, Darwin spends a lot of time talking about artificial selection before moving on to the idea of natural selection.

We humans are familiar with the idea that we can artificially select for desired traits in animals and plants by selectively breeding. Darwin noticed that nature does the same kind of thing, selecting for adaptive traits via the process Darwin called "natural selection".

Now, maybe you're right. Maybe Darwin was an idiot who wrote a book about nothing, really. Another alternative is that maybe the idea of evolution by natural selection isn't as empty a notion as you think it is.

I guess we'll see if the name "axocanth" is as widely recognised and lauded as the name Darwin 200 years from now.

 

[axocanth]

Turning to your question, the matter of "well adapted" is always something that is specific to a particular set of environmental conditions. One cannot generalise about it. Thus, your demand that I give you an example of "well adapted organisms" that aren't successful is either a mistake - because it assumes that there is only one set of conditions that make all things "well adapted" - or it setting a deliberate trap by asking for a demonstration that you already know is impossible.

Exactly! You cannot generalize. Or perhaps I should say you can, but you inevitably degenerate into vacuous triviality -- "Those with traits advantageous to survival and reproduction tend to do so better than those without".

James, do you think the theory "Armies with traits advantageous to military victory tend to do better than those without" is worthy of any interest to scholars? Does it explain any particular military victory? Or it a merely a vacuous truism?

Why, then, do you consider natural selection to be an explanatorily powerful theory? (if indeed you do)

 

[axocanth]

Now, maybe you're right. Maybe Darwin was an idiot who wrote a book about nothing, really. Another alternative is that maybe the idea of evolution by natural selection isn't as empty a notion as you think it is.

It ain't just me, sir! If you scroll back you'll see both Ernst Mayr and Steven Stanley assert that natural selection is a tautology. But they think it's nonetheless useful. I respectfully suggest they are philosophically confused about this. I admire both men. Even the greatest of men can get confused at times. No?

I'm not calling Darwin an idiot. You're an intelligent man, James. No straw men . . puh-lease!

 

[axocanth]

P.S. Even if I'm right and natural selection is a vacuous notion, it does not follow that Darwin "wrote a book about nothing". Far from it! There's a lot more contained within than just his theory of natural selection. It's another straw man, James, with all due respect.

 

[James R]

James, do you think the theory "Armies with traits advantageous to military victory tend to do better than those without" is worthy of any interest to scholars? Does it explain any particular military victory? Or it a merely a vacuous truism?

Well, imagine a situation in which military historians wrote about military conflicts for many centuries, without ever articulating any ideas about why certain armies won certain battles while others lost. (Or, maybe they articulated the idea that armies won only because God was on their side - but that's another story.)

Then, one fine day, a certain military historian realised that - Eureka! - all those armies that won their battles did so because, in certain ways - peculiar to the particular circumstances of the battles, and varying from battle to battle - they achieved "military superiority" over the enemy.

In some cases, that military superiority was primarily due to a technological advantage. In other cases, it was primarily due to better tactics. In yet other cases it was primarily due to strength of numbers. But, in all cases, suddenly reasons could be identified to explain heretofore unexplained military victories (or defeats).

This new idea of "military superiority" would not be a vacuous idea. Rather, it would be an umbrella idea (and a new label), useful for drawing together a wide range of empirical observations that nobody had previously recognised as having anything in common.

Darwin's theory of evolution by natural selection is a bit like that.

Sure, after the fact it's easy for you to scoff and say "Well, Darwin was really just stating the bleeding obvious. What's all the fuss about, then?" But lots of things seem obvious after the fact. It seems obvious that military superiority wins battles, but somebody had to notice the patterns first. In the case of evolution by natural selection, that somebody happened to be Charles Darwin. Prior to him, very few people had noticed what you now regard as the bleeding obvious.

Perhaps you ought to recognise the historical advantage you find yourself with.

Why, then, do you consider natural selection to be an explanatorily powerful theory? (if indeed you do)

Because, like so many of the greatest general theories, it draws together many strands of knowledge that would otherwise seem entirely unconnected. As a wise man said, now we know that nothing in biology makes sense, except in the light of evolution. And we didn't know that before Darwin and Wallace.

 

[axocanth]

James, very impressive reply above. Bravo! I think you put that all very nicely.

But I still need to ask again: Do you feel the allied victory in WW2, say, is explained by "Armies with traits advantageous to military victory tend to do better in warfare than those without." ?

 

[James R]

James, very impressive reply above. Bravo! I think you put that all very nicely.

But I still need to ask again: Do you feel the allied victory in WW2, say, is explained by "Armies with traits advantageous to military victory tend to do better in warfare than those without." ?

Do you feel that the revolution of the Earth around the sun is adequately explained by Newton's Law of Gravity or - if you prefer - Einstein's theory of General Relativity?

General theories do not attempt to explain specific cases. If they did, they wouldn't be general theories.

 

[axocanth]

re above

James, do you feel there a common "umbrella" explanation which adequately explains all military triumphs which is not utterly trivial? E.g. my theory of battle victories above.

Or do you feel -- as I do -- that each particular battle or war has to explained individually with no substantive theory uniting them all? Anything invoked to unite them all will be inevitably vacuous?

Note, this is not the typical case in science. Newton (and everyone else!) does offer a substantive (i.e. non-trivial) account of all phenomena relevant to the domain in question.

General theories do not attempt to explain specific cases. If they did, they wouldn't be general theories.

You can't be serious! You mean what Neptune is doing right now can't be explained by either GR or Newtonian mechanics?

 

[axocanth]

Do you feel that the revolution of the Earth around the sun is adequately explained by Newton's Law of Gravity or - if you prefer - Einstein's theory of General Relativity?

Yes! The respective explanations may be correct/incorrect or true/false. But they are certainly not vacuous or trivial!

 

[axocanth]

James, real world crap beckons. Sigh! Back later. Thanks for giving me plenty of food for thought.

 

[axocanth]

Do you feel that the revolution of the Earth around the sun is adequately explained by Newton's Law of Gravity or - if you prefer - Einstein's theory of General Relativity?

General theories do not attempt to explain specific cases. If they did, they wouldn't be general theories.

Just to add a bit more to what has been already said, James. We see, then, that Darwin lies on the border between, on the one hand, what are sometimes called nomothetic explanations (i.e. law-like explanations), typical of the natural sciences (esp. physics), and on the other hand, idiographic explanations, typical of history, say, which appeal to the individual causes and factors relevant to any particular case (e.g. WW2).

I don't think anyone supposes that the history of human warfare (or anything else!) can be captured in a substantive, non-trivial, all-encompassing law. Do you? Human affairs are far too messy, varied, heterogeneous, diverse, and so on. These things have to be explained on a case by case basis.

Natural history isn't all that different, as I see matters anyway. There are six million stories in the big city, and there are an awful lot more out there in the natural world.

Darwin evidently thought he could capture biological change -- in all its dizzying diversity and complexity -- in a simple one-liner. It was representative of 19th century thinking; "in the air", as it were. Hegel was doing it, Marx was doing it. I have serious doubts that it can be done at all, and it certainly wasn't done (IMO) by Darwin's one-size-fits-all empty truism.

We may have to settle for natural history instead. Would that be so awful?


Edit: See also my post #1446