A perpetually scary question is: What if something is going increasingly wrong medically with large numbers of people and we just don't have the conceptual tools to notice the trend? About a decade ago it was common to worry about an increase in autism, but since then it seems like it has become accepted that we shouldn't worry about that because that's maybe just a matter of changing definitions on diagnosis and insurance forms.
But then again maybe the increase was real, and we just couldn't count it? It would be pretty horrifying if 50 years from now we realize that there was an autism epidemic and it had some simple cause that we could have fixed, but we didn't do anything about it because we couldn't deal with the methodological issues of agreeing upon a definition of autism.
So, it's worth talking about the history of thinking about autism.
A British doctor writes:
'Autism' has changed meaning since I was at medical school. Then it was about a severe type of mental retardation - the kids did not talk, but rocked back and forth head banging etc; and seemed not to regard other people as people but as-if inanimate - did not react to loud noises etc. The autism bit was simply the lack of human reactions but mostly these kids were simply severely mentally handicapped, although they tended to look 'normal' and like their parents (not syndromal like Downs) .
Nowadays, Asperger's syndrome has radically re-shaped the perception of autism - Asperger's was never mentioned 35 years ago but was revived by Uta Frith of London University and her disciples. And of course here we are talking about people with high intelligence, advanced language - but who are relatively uninterested in socializing, socially clumsy etc (probably a majority of the people, men, in maths, physics, etc).
Both of these get called 'autistic' on the basis of a 'autistic spectrum' which supposedly connects them - but I don't see any connection whatsoever between a silent mentally handicapped kid head-banging in a cot year on year, and Sheldon Cooper from the Big Bang Theory...
Europeans were ahead of Americans in autism research.
My very vague impression is that the American conception of autism has followed the opposite course: that if you go back far enough, the American stereotype was of autistics as moderately functioning with perhaps some savant capabilities, but extremely annoying: e.g., Dustin Hoffman's Oscar-winning role 25 years ago in Barry Levinson's "Rain Man."
My recollection is that Hoffman's now-often criticized depiction was widely accepted in 1988 as realistic. (By the way, Hoffman's performance remains theatrically mesmerizing. Recently, I was walking past a TV showing "Rain Man" and found myself standing there several minutes later, still agog at Hoffman's bag of tricks.)
I may be totally wrong about this, but my impression is that in America since then the term "autism" has spread to all forms of non-cooperative mental retardation. A lot of this is driven by checkboxes on forms. If I had a severely difficult child and I heard I could now get more help if I checked the "autism" box, I definitely would, whether or not my child behaved like the 1980s American stereotype of an autistic.
So, Europeans started out with a picture in their heads of autism as severe retardation and have spread toward including Asperger's, while Americans started out with a picture in their heads of autism as severe Asperger's and have since spread toward including severe retardation.
I guess that's scientific progress.