Beware of Stephen J. Gould


30


Eliezer_Yudkowsky

Followup to:  Natural Selection's Speed Limit and Complexity Bound

If you've read anything Stephen J. Gould has ever said about evolutionary biology, I have some bad news for you.  In the field of evolutionary biology at large, Gould's reputation is mud.  Not because he was wrong.  Many honest scientists have made honest mistakes.  What Gould did was much worse, involving deliberate misrepresentation of science.

In his 1996 book Full House: The Spread of Excellence from Plato to Darwin, Stephen J. Gould explains how modern evolutionary biology is very naive about evolutionary progress.  Foolish evolutionary biologists, says Gould, believe that evolution has a preferred tendency toward progress and the accumulation of complexity.  But of course - Gould kindly explains - this is simply a statistical illusion, bolstered by the tendency to cite hand-picked sequences like bacteria, fern, dinosaurs, dog, man.  You could equally well explain this apparent progress by supposing that evolution is undergoing a random walk, sometimes losing complexity and sometimes gaining it.  If so, Gould says, there will be a left bound, a minimum at zero complexity, but no right bound, and the most complex organisms will seem to grow more complex over time.  Even though it's really just a random walk with no preference in either direction, the distribution widens and the tail gets longer.

What romantics, ha ha, those silly evolutionary biologists, believing in progress!  It's a good thing we had a statistically sophisticated thinker like Stephen J. Gould to keep their misconceptions from infecting the general public.  Indeed, Stephen J. Gould was a hero - a martyr - because evolutionary biologists don't like it when you challenge their romantic preconceptions, and they persecuted him.  Or so Gould represented himself to the public.

There's just one problem:  It's extremely unlikely that any modern evolutionary theorist, however much a romantic, would believe that evolution was accumulating complexity.

There was once a time when many evolutionary biologists had a romantic conception of progress, evolution climbing ever-higher mountains of complexity, dinosaur to dog to man.  And there was a hero who challenged that widespread misconception.  The hero was George Williams, his challenge was successful, and his reputation rests securely in evolutionary biology today.

In a population at equilibrium, harmful mutations will be eliminated by death (or failure to reproduce) at the same rate they are introduced by copying errors.  A very severe mutation may be eliminated by an embryo that fails to develop, but a mutation that's lethal only one time out of 10,000 may spread to 10,000 people before it starts to be eliminated.  It takes the same amount of selection pressure to support minor or major adaptations; whether the adaptation was a big one or a small one, at equilibrium, mutations must be eliminated at the same rate they are introduced by copying errors.

A population cannot sustain too high a selection pressure - too many deaths or failures to reproduce - without dying out.  And it requires the same amount of selection to support any given amount of DNA against the degenerative pressure of copying errors.  This, in turn, implies an upper bound on the amount of DNA that can be sustained by selection against the degenerative pressure of copying errors.

The upshot, as George Williams wrote:

A certain amount of information is added by selection every generation.  At the same time, a certain amount is subtracted by randomizing processes.  The more information is already stored, the more would mutation and other random forces reduce it in a given time interval.  It is reasonable to suppose that there would be a maximum level of information content that could be maintained by selection in opposition to randomizing forces...

The view suggested here is that all organisms of above a certain low level organization - perhaps that of the simpler invertebrates - and beyond a certain geological period - perhaps the Cambrian - may have much the same amounts of [meaningful] information in their nuclei.

Saying this did not make Williams a heroic, persecuted martyr.  He simply won.  His arguments were accepted and biology moved on.  The book quoted above is Adaptation and Natural Selection, now showing its age but still considered a great classic.  The shift to a gene's-eye-view in evolutionary theory is sometimes called the "Williams Revolution", the other founders being Hamilton, John Maynard Smith, Trivers, and Dawkins as popularizer.  In short, Williams was not exactly Mr. Obscure.

And Williams wrote in 1966, thirty years before Gould wrote Full House.

If Gould had simply stolen Williams's ideas and presented them as his own, then he would have been guilty of plagiarism.  And yet at least the general public would have been accurately informed; in that sense, less damage would have been done to the public understanding of science.

But Gould's actual conduct was much stranger.  He wrote as if the entire Williams revolution had never occurred!  Gould attacked, as if they were still current views, romantic notions that no serious biologist had put forth since the 1960s.  Then Gould presented his own counterarguments to these no-longer-advocated views, and they were bizarre.  Evolution is a random walk in complexity, with a minimum at zero complexity and no upper bound?  But there is an upper bound!  Sheer chance explains why dogs are more complex than dinosaurs?  But they probably aren't!

Why did Gould behave thus?  Two observations:  One, to bring order to a scientific field, it must first be in chaos.  Two, plagiarism is a crime that everyone understands.

Gould undid the last thirty years of progress in his depiction of the field he was criticizing, pretending that evolutionary theory was in chaos, so he could depict himself as heroically bringing order to it.  If Gould had also redid the accepted solutions as his own, he would have been caught, tried, and cast out of the public eye.  Newspaper editors may not be interested in mathematical arguments about evolutionary biology, but they understand stories about plagiarism and theft.  Once Gould's copying had been laid out next to the original, and eminent scientists attested to the identity, it would have been over.

So instead Gould committed a crime so bizarre that it couldn't be explained to editors.  He stole Williams's chaos.

(Incidentally, Gould's notion of a random walk in complexity has the same quality as the rest of his argument.  A genome acquires a beneficial allele at a readily calculable speed and probability, and until the complexity reaches equilibrium, new adaptations will tend to be acquired faster than old adaptations are lost to copying errors or environmental shifts.  The fewer adaptations have been acquired by a genome, the fewer are likely to be lost to a given event.  If complexity starts far below the equilibrium level, it will tend to increase.)

All this that I have said, was a common pattern throughout Gould's "work".  And all this that I have said, is no news to professional biologists.  Here's John Maynard Smith:

"Gould occupies a rather curious position, particularly on his side of the Atlantic. Because of the excellence of his essays, he has come to be seen by non-biologists as the preeminent evolutionary theorist. In contrast, the evolutionary biologists with whom I have discussed his work tend to see him as a man whose ideas are so confused as to be hardly worth bothering with, but as one who should not be publicly criticized because he is at least on our side against the creationists.  All this would not matter, were it not that he is giving non-biologists a largely false picture of the state of evolutionary theory."

John Maynard Smith was a genuinely great evolutionary biologist, the sort of man that Gould pretended to be.  But some readers may have to take my word for this, since the names of eminent scientists are often less well-known to the general public than the names of fast-talking scoundrels such as Uri Geller or Stephen J. Gould. 

I am not calling Gould a scoundrel because he was wrong; honest scientists can make honest mistakes.  But Gould systematically misrepresented what other scientists thought; he deluded the public as to what evolutionary biologists were thinking.

It is as if someone presented geocentric epicycles as the current belief in 21st-century astronomy, sharply criticized the complexity of all those circles orbiting circles, and argued for their own simpler model of planets that move in straight lines.

Did Gould deliberately lie?  If not, he executed one of the most epic feats of self-deception in the history of marketing.  The eminent John Tooby speaks:

"Although Gould characterizes his critics as "anonymous" and "a tiny coterie," nearly every major evolutionary biologist of our era has weighed in in a vain attempt to correct the tangle of confusions that the higher profile Gould has inundated the intellectual world with.  The point is not that Gould is the object of some criticism -- so properly are we all -- it is that his reputation as a credible and balanced authority about evolutionary biology is non-existent among those who are in a professional position to know...
These [major evolutionary biologists] include Ernst Mayr, John Maynard Smith, George Williams, Bill Hamilton, Richard Dawkins, E.O. Wilson, Tim Clutton-Brock, Paul Harvey, Brian Charlesworth, Jerry Coyne, Robert Trivers, John Alcock, Randy Thornhill, and many others."

If Gould, after receiving that many corrections, managed to still not know the actually current beliefs in evolutionary biology, he must have had neutronium earplugs.  I'm not saying it's impossible, though, because it's amazing what people can not-know when their reputation depends on it.  But there comes a point in self-deception where it becomes morally indistinguishable from lying.  Consistently self-serving scientific "error", in the face of repeated correction and without informing others of the criticism, blends over into scientific fraud.

And after all this, Gould is widely believed, by the general public and even by many scientists outside evolutionary biology, to be an evolutionary theorist of honorable reputation!  It is as if Immanuel Velikovsky had managed to make himself into the public face of astronomy.

If you have read one of Gould's books, you are not to blame; but you must now do your best to un-believe it all - especially all the implied beliefs in evolutionary biology that Gould seemed to be attacking.

And so as not to be accused of plagiarism myself, many others have said much of what I said here - only in politer academic language, with longer sentences, and without that specific example.  I thought it deserved a sharper presentation, for the benefit of the popular audience that Gould deluded; and a clear-cut example of Gould's "work", to show what the fuss is about.  Many academic writers on Gould could not speak as sharply as Gould deserved.  As I have no fear for my own reputation, I will say it plainly:  One way or another, knowingly or unknowingly, Gould deceived the trusting public and committed the moral equivalent of deliberate scientific fraud.