Thursday, 8 May 2014

The evolution of genius. The necessity is for the production of enough geniuses, and a society which values products of genius


I had a valuable conversation with Michael A Woodley yesterday, in which he gave me a new and very significant insight into the nature of genius in its social context.
Michael had noticed and recognized the crucial importance of a comment by HJ Eysenck, in his 1995 book Genius - that noted that many geniuses were short lived.

(There are a particularly large number of mathematicians in this short-lived category; but there have been many short-lived examples among geniuses of all types.)

Furthermore, Michael already knew that many geniuses have either zero children - or few/ zero children who survive.

(Shakespeare had zero direct descendants, so far as known. Same for Beethoven and Schubert. Same for Pascal and Turing. Same for Newton and Maxwell.)

Therefore, there is a high probability of reproductive extinction for geniuses - and this implies that geniuses do not need to live beyond early adulthood and do not need to reproduce.

I believe that this insight of Woodley's is the missing key; when added to the insights of Eysenck, he has essentially solved the problem of genius, as I understand it.


One thing that must be emphasized is that each individual genius potentially has a disproportionately massive impact on their society as a whole.

For instance, it is plausible that the unknown Byzantine inventor of Greek Fire quite possibly ensured the survival of Constantinople for several hundred years beyond what would otherwise have happened.

This is totally different from the way that genetics works!

So a single genius may affect - may massively-enhance - the likelihood of survival and growth of a whole human society - even when their genetic impact on that society is near-zero or actually zero.


Which means that the problem in explaining the emergence of a society where there is a high concentration and number of geniuses (e.g. Western Europe from the late Middle Ages and until recently) reduces to the simple problem of first producing geniuses and secondly of society being such as to appreciate and use the products of genius.

In other words, geniuses do not need to live long or have children, they simply need to survive long enough to make their contribution. And having made their contribution they do not personally need to be rewarded for it - because it is likely that they will not live very long anyway.


Why should geniuses so often be short lived (and sterile)?

Plausibly, because that the genius has a brain of exceptionally high complexity (in relevant ways), which means (from the predictions of general complex systems theory) that in the first place this kind of super-brain is harder to grow and therefore more likely to have developmental faults; and in the second place that an exceptionally complex brain is harder to maintain and therefore more likely to malfunction fatally (in many possible ways).
(The minority of long lived and fertile geniuses would include those few who have had the good fortune to develop an extra highly complex brain without any significant or potentially fatal faults; and therefore whose brain is more robust and easier to maintain over the long-term.)
So, what is needed to produce a society with a high concentration of geniuses is:

1. Enough potential geniuses are born
2. Enough of these potential geniuses survive to early adulthood (at least)

These two factors are enough to ensure that geniuses will produce - because the innate tendency to practice genius is a property of geniuses: it takes very little in the way of encouragement, and indeed geniuses will 'do their thing' even in the face of considerable dis-couragement.
The final step is that:

3. Society must value and use the products of genius.

Not all societies do this - or, more often, they only use certain restricted products of genius - such as decorative arts, improvements in statecraft or military technologies.

But, to reiterate, the society need not value or reward the genius himself, indeed the genius can safely be neglected, defrauded of his credit, or even killed (after making his discovery) since he will probably be dead soon anyway.

So long as the genius has the minimal psychological appreciation/ positive valuation - perhaps from just one or two or a few persons - then he will do his thing anyway.


So what kind of society produces geniuses?

Probably all societies produce sporadic geniuses, but only a few produce a high enough concentration of geniuses to make a difference to the society as a whole.

Although even one single genius who produces a particular invention may make a vast difference.

But the genetic pre-requisites of genius are already known: a society with a sufficiently high average general intelligence ('g') and also an average personality of a moderately high level of Psychoticism

('Psychoticism' embodies both creativity and the personality traits necessary for a genius to produce a distinctive contribution - see )

This will ensure a sufficient production of geniuses - then these must survive to adulthood in sufficient number.

I think this implies that the society must have a sufficiently large proportion of 'middle class' people - in other words those whose economic role involves a high level of technical or cognitive skills - and where possession of technical or cognitive skills gives a high probability of a higher than replacement-level of survival for your children.

In other words, what is needed is a sufficiently large and successful middle class.


But what kind of society tends to value the products of genius?

My guess is that it needs to be a society where the middle class of skill-workers has a substantial role in serving itself - because where the skill workers serve the upper classes, then only the types of genius which serve upper class interests will be used - mostly innovations in statecraft and war, and to a lesser extent in upper class luxury goods such as the arts and crafts and architecture, cooking and clothes and the like.

But when a middle class person can sell their skills to other middle class persons, then conditions are ripe for the recognition and use of a wide range of products of genius - including those which increase the functionality of society in many ways - innovations in economics, industrial organization, agriculture, transport, power machinery, communication and so on.

But a society of mandarins, a society in which the skilled middle class survive by serving the ruling class, a society in which the middle class of predominantly officials and bureaucrats, is NOT, on the whole, a society which will be likely to recognize or use the innovations of genius.


All that is somewhat speculative.

But Woodley's insight has simplified, and thereby solved, the basic problem of how genius is produced - by clarifying that genius as such is not, cannot coherently be, regarded as a direct product of natural selection: genius does not happen because it brings reproductive success - quite the reverse.

Genius is a product of heredity - but is itself (on average) a genetic dead-end.

And, having had enough geniuses born and survive to adulthood - the crucial factor is that society as a whole recognizes and uses the innovations which individual (and often short-lived) geniuses produce.



Bruce Charlton said...

Comment from Charles Murray

One slight problem with this discussion is that there's no evidence that geniuses die younger than others. On the contrary, their average lifespans are far higher the average life expectancy in earlier centuries and (though I haven't done the homework to verify this) look to be higher than average life expectancy for people who survived infancy.

From my inventories of "significant figures" in Human Accomplishment (2003), here are the mean lifespans for people in the top 5 percent of the people who qualified for the inventories (which in turn required that they be mentioned in at least half of comprehensive histories/chronologies/biographical dictionaries of their fields). The sample is limited to people born after 1000 CE, when birth & death dates for famous people are usually reliable:
> Art.China 72.68182
> Art.Japan 71.375
> Art.West 66.55556
> Lit.Arabic 67.54545
> Lit.China 67.5
> Lit.India 61.14286
> Lit.Japan 64.94118
> Lit.West 63.33333
> Music.West 60.08333
> Phil.China 70
> Phil.West 65
> Science 70.14563
> Total 68.3139
> If we look at the specialties within the sciences, here are the results:
> Astronomy 74.72727
> Biology 74.26667
> Chemistry 58.33333
> Earth Sci 70.57143
> Mathematics 64.63636
> Medicine 69.36364
> Physics 70.4375
> Technology 72.6
> Total 70.14563
> Suppose we restrict the sample to the greatest geniuses, the top 1 percent of the significant figures:
> Art.China 74.125
> Art.Japan 72
> Art.West 89
> Lit.Arabic 67.54545
> Lit.China 66.6
> Lit.India 61.14286
> Lit.Japan 60.5
> Lit.West 67.5
> Music.West 56.75
> Science 73.16667
> Total 69.8764
> Astr 78.75
> Biol 73.66667
> Chem 51
> Eart 78
> Math 79.66667
> Medi 70
> Phys 76.2
> Science 70.52381
> Tech 83.5
> Total 73.16667
> I didn't record data on number of offspring, but given a population of people with (at the least) the normal lifespan, there's no reason to assume their number of offspring was below average, and considerable reason, given the positive relationship between socioeconomic status and number of surviving offspring in earlier centuries, that they contributed more to the gene pool than the average person.

So it's okay to argue that geniuses contribute greatly to civilization independently of their contribution to the gene pool--that could be true no matter how long geniuses live or how many offspring they have--but don't invoke speculations about early death and few offspring. They appear to be without foundation.

Bruce Charlton said...

@Charles Murray

Thanks for taking the time to summarize this data.

The point I was trying to get across was exactly that point you summarize as "geniuses contribute greatly to civilization independently of their contribution to the gene pool".

Also, that an examination of the detailed personality and biography of geniuses makes it clear that many are simply 'not built' for high reproductive success. They not only are NOT that kind of person, they are often almost the opposite. And yet such people as Pascal are very obviously of the genius type - i.e. made to be a genius, not a patriarch.

Since the chances are that any specific individual whose life strategy is to aim at a revolutionary breakthrough in their field is engaged in an extremely high risk enterprise - with failure much more probable than success; as I argued in relation to revolutionary science:

This (plus the numerous examples of individuals who make major breakthroughs but do not benefit from them, die paupers etc) means that IF genius was to be increased in proportion in a population by the usual way of naturally-selection (by amplification of the representation of relevant genes in following generations) then this means that to be naturally-selected in the normal way, the relatively small proportion of *successful* geniuses would have to have an *enormous* reproductive payoff.

Which it seems clear they do not - or, not usually. JS Bach is one exception - with twenty children, half of whom survived to adulthood. I think that would have to be the kind of scale of genius reproduction for it to be naturally selected

What of the longevity of the Men of Achievement that you mention? Well there are various explanations. Firstly, longevity in men, beyond young adulthood has no necessary correlation with reproductive success. As a frivolous counter example, eunuchs have increased life expectancy and zero offspring. But I would certainly acknowledge that under some social conditions, male longevity is a great help in producing the 'resources' needed to successfully raise children - so you are quite correct about this association - all else being equal.

Secondly there is a reverse causality in the sense that to accomplish a lot, and to become acknowledged as accomplishing a lot, longevity helps a lot.

But my main point is that averages and correlations may conceal more than they reveal if they do not apply at the individual level - and from this angle I would say that it is crystal clear that an increase in geniuses is not accounted for by enhanced reproductive success.

So, I hold by the main point that the evolutionary explanation of high concentrations of genius in certain times and places does need to focus on their production, and not on their re-production.

wmjas said...

I looked up the top 20 figures on Murray's "combined sciences" list and checked how many children they had. No information is available about the children (if any) of Ptolemy or Euclid. For the remaining 18, the average number of children to survive to adulthood is 1.61 -- well below replacement. Fully half of them (Newton, Lavoisier, Descartes, Huygens, Faraday, Hooke, Leibniz, Berzelius, Maxwell) had no offspring at all (not counting Descartes's one illegitimate daughter, who died in childhood). The most conspicuous outliers was Darwin (7 children who survived to adulthood), followed by Kepler and Euler (5 each).

I haven't yet looked at geniuses in non-science fields, but I expect the results to be similar.

Bruce Charlton said...

@wmjas - Very helpful contribution - thanks.

For completeness (and I am *not* trying to get you to do this!) the Men of Genius would need to be compared with some kind of 'matched' group - the difficult thing is 'matched for what?', and how on earth can we find out this information...

But perhaps ideally, somebody such as their brothers (similar background, opportunities, similar intelligence and personality on average etc) - this would compare 'those who took the path leading to becoming a successful genius' - with similar people who either did not take that path, or failed for one reason or another.

However, I would strongly predict that a matched group would have a fertility level *way above* two, and would indeed on average raise to adulthood considerably *more* than two children - since geniuses mostly come from the middle classes (or upper) and the work of Gregory Clark suggests that this group had the highest reproductive success in most of the times and places relevant to most of these people.

wmjas said...

I've now checked the top 20 in philosophy as well (excluding Plotinus, for whom I could find no information), and the average is even lower: 1.21 children who survived to adulthood. (This figure would be even lower if I didn't include the 5 illegitimate children Rousseau claimed to have fathered, for whom there is no evidence outside of the Confessions.)

Bruce Charlton said...

Thanks again William!

James Purcell said...

The average number of children of the Doctors of the Church and Church Fathers is approximately 0.

Bruce Charlton said...

Comment from Kristor - "One question that immediately springs to mind is the effect of monasticism – especially of the Christian sort – on a population’s production and nourishment of genius. I should think that the effect would be massive. Monasteries would perfect for social maladepts. It’s almost as if they had been intentionally designed for Asperger syndrome patients. A comforting daily routine, solid food, a healthy daily slug of “real world” work to keep things down to earth, lots of time for contemplation, a huge library, lots of other smart fellows about, loads of quiet, no need to worry about women."