Falsifiability? Let's not be Naive.
Achilles942: I'm posting the following video to try to illustrate the naivete of our members' understanding of falsification in science:
Now, on one hopelessly naive understanding of falsification in science, when observation/evidence is at odds with theory, the theory is falsified and must be rejected.
Lest it be objected that I'm attacking a strawman, consider the following quote, which I regard as fairly typical, from kittybobo34 in the "Dark Matter and the Age of the Universe" thread, top of page 2:
"Faith in a theory? The theory works until some evidence refutes it, don't see where faith comes in"
Whether Kitty, and others like her, realize it, what she is espousing is a long-discredited, vaguely Popperian model of theory change in science.
Thomas Kuhn offers what I believe to be a far more accurate picture. As a sketch (my own paraphrase):
"Under conditions of 'normal science' (Kuhn's technical term), that is, the vast majority of scientific work, the truth of the overarching theoretical paradigm is largely -- one might say "dogmatically" -- taken for granted. No attempt, by and large, is made to attack or falsify the paradigm. To do so runs the risk of excluding oneself from the science game altogether.
All major theories are born into an ocean of anomalies, i.e., cases where the evidence does not sit comfortably with the theory. So-called 'normal science' consists of a process of reconciliation; MAKING recalcitrant evidence fit theory. Far from trying to falsify the regnant paradigm, the task of the 'normal scientist' is to find some way to make recalcitrant data (another man's falsifying evidence) compatible with it.
If there can be any talk of 'testing' at all here, what is under test is the SCIENTIST, not the theory. Any failure to assimilate awkward evidence with theory is regarded as a failure of the scientist, not the theory/paradigm. The theory is just fine, thank you very much"
(end of Kuhn paraphrase)
Achilles942: Now, back to Dawkins.... at 0:25 we are told, "There has got to be a series of advantages all the way, in the feather. If you can't think of one, then that's your problem, not natural selection's problem. Natural selection, er, um, well, I suppose that is a matter of faith on my part..."
Of course, there is another logical conclusion that could be drawn from "I can't think of any way to reconcile theory with recalcitrant evidence", namely, the theory is false. Needless to say, this possibility does not even enter Dawkins' mind.
So, does this sound to you like:
(i) A la Popper, a man hell-bent on having a theory falsified whenever evidence conflicts with theory?
(ii) A la Kuhn, a man who takes the truth of his theory for granted, and in any case where evidence is at odds with theory, and you can't rectify the situation, that's your own failure, not a failure of the theory? Where does the failure lie: with the theory or the lack of the scientist's ingenuity?
Now, compare with Kitty's gambit: when the evidence is at odds with my theory (sharks and horseshoe crabs, etc.), there must be an "unrecognized force" at work. My theory is just fine, thank you very much.
Does she sound like a scientist hell-bent on falsifying a pet theory? Or a person trying very hard to do precisely the opposite: PROTECT her theory against falsification?
Again, I emphasize, I make no normative judgements on the rightness and wrongness of all this. That's the scientists' business.
What I do want to stress again, is the tension between bragging about the 'falsifiability' of a pet theory, while simultaneously taking every measure possible to prevent it from being falsified.
Achilles942: Whatever else we can say about falsification, one fact from the history of science glares out at us: Scientists, as a whole, NEVER renounce and reject a major theory, regardless of how good or how dire its "evidential fit", until an alternative presents itself.
To the best of my knowledge, there exists not a single precedent in the entire history of science where a major theory was declared to be false and abandoned, even when known to be in conflict with observation (i.e., evidence), in the absence of a new bandwagon for the relevant scientists to jump on.
What do you expect, after all? For scientists to do themselves out of a job?
And when such a "paradigm shift" occurs at all, it invariably occurs in a piecemeal manner; a gradual conversion process of individual scientists from old to new until the successor paradigm holds sway, the entire process often taking several decades. Also commonly seen is for a small cabal of die-hard defenders of the old theory going to their graves quite unconvinced by the virtues of the new.
Such a phenomenon puts the lie to simpleminded comments such as
"Faith in a theory? The theory works until some evidence refutes it, don't see where faith comes in"
"If you can disprove evolutionary theory, come forward an collect your Nobel Prize"
(Ignoring the further absurdity that there does not even exist a Nobel Prize for biology)
It may be the case that in a hundred, or two hundred, years from now we will be told that the theory of evolution, say, was long ago "disproven" (i.e., falsified), just as we're often told today that the theory of phlogiston was disproven.
This, I suggest, will be a rewriting of actual historical events.
As for now, NO MATTER WHAT evidence comes to light, -- a whole basketful of pre-Cambrian cuckoo fossils, say -- anyone expecting the entire community of evolutionary biologists to stand up en masse and declare their theory falsified can only be described as living in Cloud Cuckoo Land.
And, far more commonly seen, the same applies to anyone provoking a ToE skeptic to produce evidence that would cause the entire community of evolutionary biologists to stand up en masse and declare their theory falsified.
Fiendishly perspicacious as always, Australian philosopher David Stove, with the theory of inclusive fitness in mind, has this to say...
"Scientists sometimes (as is well known) continue to work with a theory which
they themselves know is false. Laymen, when they hear of such a case, are apt to
be audibly critical of the scientists' conduct; but of course they have no BETTER
theory to suggest, and the only result is, that the scientists grow angry and
impatient with their lay critics. But these features of scientists' behaviour are not
ones which deserve esteem, and still less, imitation. They are DEPARTURES FROM
rational behaviour, not forms of it. They arise only because professional scientists,
without the guidance of SOME theory however unsatisfactory, do not know what to
do with themselves. But laymen have other occupations, and the indignation they
feel, when scientists stick like limpets to a theory they know is false, is not only
natural but rational. A rational interest in science, as distinct from a professional
one, is an interest in what is true, or probably true, or probably close to the truth: in
that, and in nothing else. If a scientific theory is certainly not even NEAR the truth,
then, whatever attractions it may have for scientists, it is of no interest to a person
who is simply trying to have rational beliefs and no others. That is how things
actually stand, of course, with the theory (for example) that the blood is stationary,
or that the earth is shaped like a bullet, or that it rotates from east to west. It is also
how things actually stand with the theory of inclusive fitness.
When a proposition is obviously false, and is nevertheless widely and fervently
believed, it is a reasonable inference that it possesses some powerful attraction for
the minds of those who believe it: powerful enough, anyway, to outweigh its
obvious falsity. Take, for example, the theory that human beings are immortal. The
falsity of this proposition is obvious now, but it always was as obvious as it now is:
it is not as though we have lately discovered the first disproofs of this theory - we
have not. Yet it was generally believed in western Europe for most of two thousand
years, and (on the whole) was believed most fervently by precisely the people
whose intelligence and education best entitled them to rank as intellectual
authorities. What the attraction of the theory was in this case, is too obvious to need
The theory of inclusive fitness is in an analogous position nowadays, if what I
have said about it earlier in this essay is true. That is, it is obviously false, and is
nevertheless widely believed, and believed most fervently precisely by the people
best entitled to rank as authorities on evolutionary biology. It therefore must
possess some powerful attraction for the minds of those who believe it. But what is
Achilles942: Since David Stove brought up the theory of inclusive fitness (or kin selection), it might be worthwhile reflecting for a moment on why this theory was introduced in the first place, as well as its ramifications for falsification in science.
Let's remind ourselves first of claims typically made by defenders of Darwinian, or neo-Darwinian, theory:
C1: "The theory fits all the facts; no exceptions"
C2: "A theory is clung to only for so long as it is consonant with the facts/evidence. As soon as theory conflicts with facts/evidence, it has been falsified and must be abandoned."
Now back to ToE, the theory of evolution (or at least one version thereof)...
In its original, unadulterated form, with a species-wide, constant, ruthless struggle of all against all, survival of the fittest and all that, with selection acting on individual organisms, the existence of altruism presented an immediate and obvious objection to the theory, namely, there should be none. It's an objection that, needless to say, Darwin himself was well aware of.
Or, at the very least, whenever altruism dared to make an appearance, it should be immediately eradicated through the merciless winnowing of natural selection.
But the existence of altruism in many species, not least of all our own, is an obvious and incontrovertible fact.
Here, then, we have yet another case where facts -- or "observation" or "evidence" if you prefer -- are, prima facie, at odds with theory.
(I emphasize "prima facie" since, as I tried to make clear in the OP, rarely if ever in science does a situation arise where evidence flat out contradicts a theory. Appeal can always be made to "hidden variables", "unrecognized processes", etc.)
C1 above already begins to look less than convincing.
What to do, then? Well, here are some options for the scientist:
O1: In accordance with C2 and Sir Karl Popper's strictures, declare the theory falsified and reject it.
O2: Do nothing. Ignore the awkward evidence, perhaps hoping that someone smarter than oneself can come up with a patch. This is not at all unlike the state of many religious people who feel uncomfortable with the existence of both a benevolent God and so much evil in the world, yet continue to cling to their faith nonetheless.
O3: Find some way to reconcile the embarrassing (cf. "falsifying" ) evidence with theory -- exactly as Thomas Kuhn tells us is the main occupation of "normal science". In other words, put a patch on the puncture.
I know of no Darwinian who has ever chosen the path of O1, in this particular case, or any other similar case.
O2 is common enough in science as a whole.
As for O3... enter the theory of inclusive fitness.
Moral of the story once again is: theory being at odds with evidence is a run-of-the-mill phenomenon in science. Given that there is no such thing as a logical refutation in such matters, a judgement call has to be made between -- apart from doing nothing -- declaring the theory to be false, or doing everything in one's power to PREVENT a cherished theory from being falsified.
No prizes for guessing which route is almost invariably taken.
Achilles942: In these forums I continue to see naive comments on falsifiability such as the following:
"Relativity is a falsifiable theory: It makes predictions that can be tested by experiment. [...]"
( "Einstein's theories are false", p1)
"What's naive about that?", one might well ask. Well, I'd invite the reader to consider the following implications -- both of which are false -- which appear attendant upon remarks like the one above...
(For simplicity I'll use "T" to stand for "theory", for we might as well be examining ANY scientific theory here; not just relativity)
(i) In the event that a particular observation came to light which is ruled out by T, the scientific community would be logically compelled to reject T as false.
(ii) By being falsifiable, T enjoys a certain scientific virtue that the theory of God's existence, say, does not.
I repeat, both these implications are indefensible.
Let's suppose, first, that that a certain observational prediction (OP) can be derived from our theory, T, and that OP fails to be observed. Then by simple logic...
Premise 1: If T is true, OP must be observed
Premise 2: OP is not observed
Conclusion: T is false
Hey presto! Looks like we've just falsified our theory, T, with impeccable logical rigor.
Alas, the above schema bears no resemblance whatsoever to real world science.
As Pierre Duhem first pointed out over a century ago, and whose work was later expanded upon by W V O Quine, scientific theories cannot be tested in isolation this way. If only they could! Scientific testing is a "holistic" business, as they say. What is tested is never an individual hypothesis/theory/law, but a holistic package thereof.
Supposing we wish to test Newton's law of gravitation (NLG), what observational consequences can be derived from it for us to test? Ans: absolutely none... by itself.
Enter auxiliary hypotheses and background knowledge ... *drumroll*
Observable consequences/predictions can only be derived from a scientific theory in conjunction with a host of other statements often referred to as "auxiliary hypotheses" or "initial conditions" or "background knowledge". I'll just refer to them collectively as AHs. In the case of testing NLG, the auxiliary hypotheses might include -- among a great deal more -- information on the position and velocity of heavenly bodies.
The actual testing schema, then, unlike the naive caricature above, looks like this (for deterministic theories only):
Premise 1: T, in conjunction with AHs, entails OP
Premise 2: OP is not observed
Well, what can be logically concluded? Ans: only that something is wrong somewhere in the holistic package of T in conjunction with the innumerable plethora of AHs. It may be that T is false; it may be that T is true and the problem lies somewhere in the AH assemblage.
Logic alone cannot tell us where the problem lies. Logic alone cannot determine for us how praise and blame ought to be distributed through the "T & AH" holistic bundle. This, in a nutshell, is the so-called Duhem-Quine thesis. A judgement call has to be made.
So, in conclusion, and returning to where we began:
(i) Under such circumstances as those sketched above, any particular scientist, if he so wishes, is free to announce that he believes T to be false.
Is he ever compelled by logic to do so? Ans: No.
(And, of course, they almost never do this. What almost invariably happens, instead, is that T is jealously guarded, and the blame put on the AHs -- remember those "unknown forces? )
(ii) In this respect at least, the scientist is in precisely the same position as the proponent of the God theory. The latter may also, at any time he pleases, renounce his faith. Just like the scientist, however, he is never compelled by logic to do so. And very rarely does.
Achilles942: One conception of how theoretical change in science occurs, apparently widely believed by both laypeople and scientists alike, goes roughly like this (my own paraphrase):
"A new scientific truth triumphs when a previously held theory is falsified by evidence. The scientific community, as a whole, are convinced by the weight of the evidence -- as any rational person would be -- that the old theory is false and reject it en masse. A new theory is subsequently adopted to take its place."
Unfortunately, charming story though it is, the resemblance it bears to the actual history of theory change in science might be described as scant.
Physicist Max Planck is far less naive...
"A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it."
Achilles942: Continuing the theme of the previous two posts, and having seen the importance of the Duhem-Quine thesis, I'd caution the reader to be very skeptical indeed whenever you hear a scientist speak of a "definitive falsification" or "knock-down refutation" or some similar locution for, as I've been at pains to show, if this is understood to mean "conclusively demonstrated to be false" there exists no such thing in matters empirical.
When scientists do -- not infrequently -- speak this way it needn't imply any intent to deceive. Being busy with their own areas of expertise, scientists, by and large, tend to be ignorant of the philosophy and actual (as opposed to "Whig" ) history of science.
It's not uncommon, just to take one example, to read of the aether theory being "definitively disproven", perhaps by the 1887 Michelson-Morley experiment, or by Einstein's 1905 paper on special relativity.
The actual historical facts of the matter are that the aether continued to be defended well into the 20th century by first-rate scientists until -- cf., the Max Planck quote in the previous post -- its defenders were all stiff as a board and six feet under. Consider:
"As J. J. Thomson said as late as 1909, "The ether is not a fantastic creation of the speculative philosopher; it is as essential to us as the air we breathe. . . . The study of this all-pervading substance is perhaps the most fascinating and important duty of the physicists"
-- "The Advancement of Science and its Burdens", Gerald Holton, p79
So, what *can* we say about the supposed disproof (i.e., falsification) of the aether?
"No scientist today believes the aether theory to be true" -- entirely unobjectionable
"The aether theory was definitively shown to be untrue" -- nonsense!
Finally, it's not my position here to endorse the various claims made by proponents of Creationism or the Intelligent Design (ID) community. These are not positions I subscribe to myself, though insofar as they challenge orthodox hegemony, I personally see them as serving a valuable function. You, dear reader, are perfectly entitled to feel otherwise.
What I would suggest, however, is that next time you hear a high profile scientist declare that such-and-such an ID claim has been "definitively debunked" (or suchlike), you take the dismissal with a healthy pinch of salt.
Let's continue not to be naive in such matters.
Achilles942: Well, boys and girls, despite my heroic efforts in this thread to expose overly naive ideas about falsification in science, I continue to see fairy tales about that pesky pre-Cambrian rabbit being promulgated in the science forum.
On page 107 of the "Most Atheists Don't Know About Science" thread, in response to my enquiry about what she thinks would happen in the unlikely event that a rabbit fossil were to be unearthed in pre-Cambrian strata, kittybobo34 explains:
"Yep , that would end the ToE"
(ToE = theory of evolution)
The naiveté is not confined to our own science forum, of course. It's the kind of simpleminded remark routinely rehashed ad nauseum by the likes of high profile scientists such as Richard Dawkins, apparently equally clueless about the philosophy and history of science.
Well, what's so naive about it? Here are a few reasons why:
First of all, the remark presupposes the existence of a universally agreed upon "theory of evolution". There is, of course, no such beast; scientists make all manner of claims on the topic of evolution, and those claims are not infrequently mutually contradictory.
What is being suggested, then, by kittybobo34 and others like her, is that the discovery of that pre-Cambrian lagomorph would put the quietus to a beast that does not even exist in the first place.
Not to be a spoilsport, though, let's just indulge the fairy tale raconteurs for now, and pretend that the chimera is real: ToE exists.
The history of science provides us with not a single precedent, that I'm aware of anyway, whereby a major theoretical paradigm has been abandoned in the light of recalcitrant data until an alternative becomes available.
Even then, the revolution, as I described in an earlier post... "invariably occurs in a piecemeal manner; a gradual conversion process of individual scientists from old to new until the successor paradigm holds sway, the entire process often taking several decades. Also commonly seen is for a small cabal of die-hard defenders of the old theory going to their graves quite unconvinced by the virtues of the new."
The suggestion, then, that one measly piece of puzzling evidence -- one scrawny, antediluvian bunny -- would cause the entire community of evolutionary biologists to immediately and en bloc renounce a cherished theory appears, quite frankly, about as likely as Scotland winning the next World Cup.
But, hey, we can all dream, I suppose.
And, to continue the fairy tale, supposing they DID all give ToE its ignominious marching orders, what would be the alternative?
The only serious rival to evolution seems to be special creation and that means .... G...G....G.... God!!!
Now, boys and girls, are we seriously expected to entertain the possibility that, inasmuch as evolution has been definitively refuted, scientists everywhere would start attending Mass, dipping themselves in the River Jordan, and perhaps handling rattlesnakes to boot?
Let's not be silly. We're all mature adults here.
Well, what WOULD happen if -- wonder of wonders -- a pre-Cambrian rabbit were to rear its ugly head?
We've already seen in this thread that -- based on the holistic nature of theory testing (Duhem-Quine thesis) -- there is no such thing in empirical science as a "definitive refutation" or a "knock-down falsification", despite what scientists themselves are wont to say about these things. A prized theory can ALWAYS be defended by playing around with what I referred to earlier as "auxiliary hypotheses" and "background knowledge".
When facts and theory are at loggerheads with one another -- a routine occurrence throughout the sciences -- the existence of a new particle might be posited, for example, or an unknown planet, or a new level of selection (to add to our already prodigious collection), or unrecognized forces at play, or... you name it.
As Thomas Kuhn taught us to see more clearly (see OP), the principal business of normal science is that of assimilation; reconciling intractable data with theory. "Puzzle solving" is what Kuhn calls it. As the history of science bears witness, that rabbit would most likely be regarded not as a refutation, but rather as a puzzle to be solved; a challenge to the scientists, not the theory.
What is now under test is not ToE, but the ingenuity of the individual scientists. Some way must be found to reconcile awkward evidence with theory.
Now, in the case of our pre-Cambrian rabbit pest, how exactly this would pan out is anyone's guess. These scientists are no dummies. They'd come up something I have little doubt -- exactly what I cannot say -- or die trying.
What I can say with virtual certainty -- pre-Cambrian rabbit, tarsier, ringtailed lemur or not -- is that you should not expect to see evolutionary biologists handling rattlesnakes any time soon.
Achilles942: One man's fish is another man's poisson.
And one man's "falsifying evidence" is another man's "puzzling evidence" or "anomaly" or "problem for the theory".
Despite what you might have heard science fans, or even scientists themselves, say about their favorite theory "fitting all the facts", countless philosophers and historians of science have pointed out that all major theories are born into an ocean of anomalies, that is, instances of theory NOT fitting the facts.
Now, pay close attention to the choice of words. In every such case, a critic of the theory in question might claim:
"Your theory does not fit the facts. The evidence falsifies your theory"
Those under the spell of a darling theory are unlikely to see matters this way. Even assuming the objection is acknowledged at all, the response is probably going to sound something like:
"Pfft and meh! Don't be ridiculous. That's not falsifying evidence. It's just an anomaly (or a "puzzle" or whatever). We're working on it. Clearly, we don't see the whole picture yet. My theory is as fit as a fiddle, thank you very much"
I've been told myself, for example, in the science forum that what might APPEAR to be a fact-theory mismatch -- horseshoe crabs, say, that didn't receive the memo telling them that they were supposed to be busy evolving -- is in fact no such thing. There are clearly "unknown forces" at work.
Few theories in the history of science have been more successful than Newtonian mechanics. Though now regarded as false, it held sway for two centuries or more, widely regarded with certainty or near-certainty as being true.
Imre Lakatos, in his essay "Newton's Effect on Scientific Standards", describes the situation around the year 1700, and the reaction of Newtonian defenders to the myriad anomalies plaguing their theory ...
"But Newtonians had little doubt that their programme would finally digest all the 'exceptions'; and this required a great deal of self confidence, for 'exceptions ', or 'anomalies', ' recalcitrant instances', abounded. It is characteristic, for instance, that nobody thought that the well-known fact that the comets' tails seem repulsed rather than attracted by the Sun, was a refutation of Newton's theory, although it was acknowledged as a problem -- or 'puzzle', as Kuhn would call it -- within Newton's research programme. Halley hoped its solution would be inserted in the first edition of the Principia. While it was still in the press, he wrote to Newton: 'I doubt not but this may follow from your principles with the like ease as all the other phenomena; but a proposition or two concerning these will add much to the beauty and perfection of your Theory of Comets.' Although Newton did not reply, no Newtonian was unduly worried.
The same tranquility was displayed at the many divergences between Newton's theory of the Moon and the observations. These divergences were regarded as problems but few thought there was anything wrong with the research programme: it was rather the researchers who were at fault. Newton's 'theory of the Moon' was in fact first published many years after the first edition of the Principia, in I702, in David Gregory's 'Astronomiae Physicae et Geometricae Elementa'. It calmly states that Newton's theory 'agrees VERY NEARLY with the phenomena as he had proved by very many places of the Moon observed by the celebrated Mr. Flamsteed'. But we have to remember that Newtonians never let the authority of observations prevail against their research programme; with the help of their positive heuristic they produced one theory after the other to accommodate counterexamples; but frequently they ignored observational counterevidence altogether: they knew not only that theories had to be constantly tested by observations but also observations by their theories."
Read it all here
Had Wireclub been around in 1700, then, we might well have heard an overzealous Newtonian apologist boast ...
"My theory fits all the facts. So there!"
Should you have believed him? Ans: of course not. To do so, you'd have to be
And next time you DO hear an overzealous neo-Darwinian apologist in the Wireclub science forum bloviating ...
"My theory fits all the facts. So there!"
Should you believe her?
Achilles942: I'm not sure if anyone understands (judging from past experience -- no one) but anyway...
When the scientistic loonie says to me "Show me a pre-Cambrian rabbit, and I will renounce my theory immediately. I'm a reasonable man"
It sounds to me eerily like the religious nutter who boasts "My God theory is falsifiable too. If it rains pre-Cambrian rabbits tomorrow I will renounce my God theory. I'm a reasonable man.
And the Islamic extremist who claims "Show me your proof that God does not exist and I will immediately renounce my faith. I'm a reasonable man"
Of course, we all know, nothing shown to any of them will have any effect whatsoever.
I despair for humanity.
All of your theories are, in principle, subject to negative evidence.
No matter what anyone produces, you'll still be defending your shite.
Got any cassowaries, Vine?
theHating: "Of course, we all know, nothing shown to any of them will have any effect whatsoever.
I despair for humanity."
Humanity: doing fine without colin's forced perspective on science.
Achilles942: A slight digression into the explanatory vacuity of natural selection...
At 1:50 our evolutionary expert (Barry Lynn) describes natural selection thus:
"Random selections which make a species more likely to survive are beneficial. That's a very simple idea and it explains why in fact some species survive and others do not."
First minor point to note is that Mr Lynn seems to be a little tongue-tied. By "random selections" he presumably means "random mutations"; selection, according to orthodoxy, is a non-random process. Moreover, also on the orthodox account, selection acts on individual organisms, not species.
(Feel free to make the relevant substitutions if you so wish. It makes no difference to the elephant in the room which follows.)
This can all be forgiven as being (presumably) slips of the tongue, easy to make in an impromptu situation.
Far more serious is the supposed explanatory power which Mr Lynn imputes to natural selection -- and this IS orthodoxy. Listen very carefully again:
Now, what can possibly be meant by "beneficial" in the first sentence? Beneficial to what? Ans: beneficial to survival (and reproduction), of course.
Thus by substitution, we get: "Random selections which make a species more likely to survive increase the likelihood of that species' survival".
You're quite right, Mr Lynn. It is a very simple idea. It's as simple as "Dogs are dogs" or "Diseases which cause people to die cause people to die" or "Survival of the fittest".
Or as David Berlinski says at 1:52, "Whatever happens happens", and again at 2:30, "What survives survives".
As for the putative "explanatory power" of your principle of natural selection, Mr Lynn, it enjoys precisely the same degree of explanatory power as any other tautology, viz., none whatsoever.
Mr Lynn's entire characterization of natural selection, quoted above, might thus be paraphrased as...
"Some species have traits which make them more likely to survive. This increases their chances of survival. This is a very simple idea and it explains why some species do in fact survive. Their survival is explained by their superior ability to survive."
1:24 - "It's always easy to persuade yourself that you've understood something when you haven't understood a thing" -- David Berlinski
Achilles942: For crying out loud, folks, what would YOU say to the fellah who told you excitedly about his cutting-edge new theory about sport ...
"My stunning new discovery is that traits which make a sportsman more likely to win are beneficial. That's a very simple simple idea and it explains why, in fact, some sportsmen do win and others do not"
First point to note is that our cutting edge theory of sport is TRUE, but trivially so (cf. "dogs are dogs" ). That is to say, its truth can be known without leaving the comfort of our armchairs.
But is this really what you would consider to be a satisfying scientific explanation?
For contrast, consider an alternative proposed explanation:
"Long legs help sportsmen to win competitions. That's a very simple simple idea and it explains why, in fact, some sportsmen do win and others do not"
This, on the other hand, is a perfectly respectable empirical theory/explanation. It is non-trivial; its truth or falsity CANNOT be known just by closing your eyes and thinking about it.
Oh, and it's almost certainly false. (Think of weightlifting, say)