Here's a comparison for you, Patrick: They haven't succeeded in figuring out an autopilot for a car yet. What makes people think it's possible for a plane, which has a whole additional dimension of movement?

By the way, I know the Mythbusters' episode peeves you, but they were very clear that the question was whether an untrained pilot could land a plane with a lot of help from a live pilot. And I think they were well aware that it was only a simulation; they rated it only "Plausible," not "Confirmed." (Both Jamie and Adam have said over and over again, on multiple shows, that you cannot verify without a full-size test. They never "Confirm" on a simulated test.) And it was the simulation expert who said the plane could durn near land itself, so complain about him.

With all due respect, I am fully aware of how simplistic the analysis was. That's my point: even a very simple check of the odds suggests how unlikely it is that all ten were from the same gender. Check my later post for additional comment.

I get the journal with the articles you mention -- and the counterpoints. The logic is that since genes related to mental retardation occur disproportionately on the X chromosome, they are more likely to get expressed in XYs than someone with an XX which could have a healthy gene inhibiting it.

When relating this finding to this article, there are three problems:

1. Mental retardation is not intelligence. This may spread the range on the lower side, but not the upper side of the bell curve. There are no genes currently identified as being associated positively with intelligence. To take a silly example, having flat tires slows down your car from the average, but having overinflated or larger tires does not make you faster!

2. Intelligence is not the same as writing good stories. Writing novels is a highly complex task, involving a wide range of abilities, not all of which are intelligence-related (e.g., empathy or Emotional Intelligence, if you like). The reasonable assumption is that there is a constellation of traits and trained skills necessary to be a successful writer. (Plug: My book on motivating writers discusses some of this.)

3. It is by no means certain that intelligence is a unitary concept in any case, though a number of my colleagues will disagree with me; but in that case you could have a host of genes predicting different elements of intelligence.

It ain't gonna be simple...

Meant "if men are given the edge inappropriately". Sorry, typo.

As hobgoblin and others have pointed out, you shouldn't question the statistics offhand. I'm a member of (and former president of the Internet Chapter of) Sisters in Crime, which promotes a level playing field in mystery fiction by supporting women, who have consistently been underrepresented in contracts, high-profile publications, reviews, and awards, though it is much better today. I've been tracking similar awards for some years now. Perhaps more importantly, I'm a research scientist by training, so have some experience not just with statistics, but their limitations.

The inputs are indeed as important as the outputs -- it is theoretically possible that all the top books by women weren't as good at the top ten of the men, or vice versa. All I said is that you shouldn't bet on it.

Women write and publish nearly half the books out there today; if you assume a normal distribution of talent in both genders, then you can assume equal chances. The farther away the list gets from 50-50, the less likely it is. Assuming all else is equal -- which admittedly never happens -- in a large enough population, the odds of having all ten from one gender is the same as having ten flipped pennies in a row come up heads.

Now, it is possible that there is incremental bias here as well. Even a relatively small bias -- say, men are given the edge appropriately about 10% more often than women -- can lead to major outcomes after enough iterations. For example: if 10% more men get an agent, then 10% more sell their book, then 10% more get the hardcover deal instead of a paperback original, then 10% more get a little extra marketing money...you get the idea. No one is very biased, but the final outcome is extremely biased. Or some kinds of books are harder for women to publish than men, and this means some very good books are not published. There's a reason several SF writers, especially early ones, wrote under male names (e.g., Andre Norton, Julian May, etc.).

I certainly don't think PW intentionally came up with a list that happened to be all men; but I think it would serve them well to re-examine their assumptions around what makes for a "best book," and it would be very interesting to trace backwards how that relates to genders.

And, as I said: it's not a bet I would take, in a fair world.

While it is entirely possible for a list to randomly come up as all male or all female in a given year -- since you can't predict who will and who won't have a great book anyway -- what are the odds that ten out of ten are by the same gender?

The top five list: 1 in 32

The top ten list: 1 in 1024.

That's .098% chance. Roughly one-tenth of one percent.

Don't bet on it.

Do monsters await in the bowels of the earth, or in ourselves? Novelists have been attempting this sort of thing since John Fowles' "The Magus"

Just since 1966? Excuse me?