http://books.google.com/books?id=DyMjW21HwHwC&pg=PA89&lpg=PA89&dq=reading+literacy+england+1600&source=bl&ots=SL1ct7yRfW&sig=0Hz6txLaE3_51PwzxFQTxSdEka4&hl=en&ei=4NHmSdjPF8LgtgeVvvCVBg&sa=X&oi=book_result&ct=result&resnum=1 30 percent of english men could read in 1600. I would say that is a considerable amount. Only 10 percent of women could read at that time. By 1700 50% of men could read.
So I would bet that the 30 percent that could read in 1600 out bred those who couldn't. It's possible that the increase in reading literacy was partly genetic in origin.
Rational/logical/analytical abilities could help with learning.
There are obviously a lot of complex interactions at work in our evolutionary past that I think we are just begining to understand.
Being able to learn and absorb written material allows a person to create more wealth. This is presumably why people go to school at all. Being able to accumulate more wealth as a result of learning allows a person to have increased reproductive fitness. So when you say "Access to better food, more sanitary living, less stress all indicate higher survival rates", these things could be the result of learning information. Like if you learned how to build something from a book that could improve your surrounding living conditions. Of course this could be highly variable depending on an individual person's evolutionary past.
Genes have been correlated with reading ability. http://www.physorg.com/news142091390.html I would imagine their could have been selection pressure on these genes in the recent past.
See here for more info. http://jmg.bmj.com/cgi/content/full/44/5/289
Well here's a paper about the changes in surnames in Britain between 1600 to 1851. This is recent evolutionary selection.
http://www.econ.ucdavis.edu/faculty/gclark/Farewell%20to%20Alms/Clark%20-Surnames.pdf
Wealthier people quickly out bred poorer people. I think its highly probable that genes were selected for to be able to better seek out and absorb written material. Being able to learn information has a huge fitness benefit to propagating your genes. There was plenty of scientific reading material available at this time (1600-1851) that could have conferred a fitness benefit.
Any factual information you learn about the world (through reading) can improve your odds of avoiding death. Death was very common in the recent past.
http://www.udel.edu/educ/gottfredson/reprints/2007evolutionofintelligence.pdf
I'll expand on the controversial stance as to why this is. It's obviously extremely complicated and I can't really do it justice in this specific comment. However I will try to give a cursory explanation. I personally tend to think that rational thinking may in part be genetically encoded in the brain and is a trait that may be somewhat distinct from aspects of general intelligence. I think its very likely that their has been differential selection pressures on male and female brain's over the course of evolution. There has been a lot of evolutionary selection pressure over the last 10,000 years in humans. You can think every time an important invention arose in a specific location of the world, then their were resulting shifts in human evolution as a result of that new technology. Genes that allowed their host vehicle to best exploit that new technologies propagated.
So in a society where the written language was more common (for instance), there was selection to build brains that could seek out and absorb more written material. People who had more analytical/rational/logical brains that could seek out the most salient written material were probably selected for. Those who spent a lot of time reading nonsense or were not good/logical readers were not as reproductively fit. They would probably have less information in painting a correct world picture and would be less likely to avoid dangers to their life. To me, it seems obvious that evolutionary selection pressures between men and women were different in this regard. This is especially true in societies where women were relegated to a much different societal roles than men. Gene expression patterns have been shown to be different in female vs. male brains, for instance.
Their is probably a huge variation of human brain abilities depending on where a person is located in the world and their unique evolutionary history on the branch of life. In places where a written language was uncommon until recently, then you might conceivably expect a different neural architecture to be in place. Evolution does not necessarily lead to improved abilities in all traits over time, though. For instance chimps have a superior numerical working memory than humans. So their is probably a wide swath of individual variation with people having specific aspects of intelligence that are quite different. I definitely believe in the construct of general intelligence, but I do think there are subsets to intelligence (working memory being one example, creativity being another that is a harder trait to define, logical/rational thinking).
Gregory Cochran has posited the unique history niche of ashkenazi jews as leading to their higher verbal IQ over non-jewish europeans. This is another instance of a historical branch on the evolutionary tree of life resulting in a shifted bell curve when comparing two population groups. I would imagine in certain historical evolutionary niches, men were probably in positions where they need to seek out and digest written material in a rational/logical way and women less likely to be in those positions.
However until you have a complete understanding of history, gene expression profile changes over the course of history, how they relate to brain/behavior changes, sex differences in phenotypical gene expressions and meme propagation, your always going to have an incomplete picture as to why any specific "macroscpic phenotypical trait" is different when comparing macroscopic "groups". All group categorizations are ultimately going to be fuzzy and phenotypical traits are as well. The evolutionary tree of life is complex and a group classification is ultimately going to have to have some arbitrary cut off. I think a lot of "macroscopic phenotypical traits" can be found on a bell curve. So the bell curve for any phenotypical trait (like a majority of Less Wrong readers being male) has a long universal evolutionary history behind it that defies easy synopsis.
About the literary statistics, it specifically says "These figures may be pessimistic because reading was taught before writing at school". So the figures may actually underestimate the literacy rate. I don't have the time to found other sources.
I'm not saying the increase in literacy rates was all genetic in origin. The fact that literacy increased on a societal scale probably had to do with new inventions that made it easier to publish and distribute written material (I'm not sure of the whole history to be honest so that is somewhat of a guess). However these societal changes may have also led to genetic changes as well. Read the "10,000 year explosion" for more examples of how societal changes lead to gene frequency changes that led to more societal changes. Some animals have rudimentary reading capabilities. Being able to read may just be an offshoot of some other brain process (reading facial expression for example). So evolution is often good at co-opting brain processes for alternative uses. With a wider distribution of reading material it is possible that evolution selected gene frequencies that altered reading ability (speed of reading, logical analysis, etc.)
I don't think the it matters if the dyslexia gene is the same in females versus males. The phenotypical effect of the gene on the host (male vs. female) would likely be different. The fact that men and women have different hormones could alter the expression of any single gene (for example). The phenotypical effect of any gene in an organism is going to depend on a multitude of factors (like environment, other genes).
That's all I can say for now.