Writing about intelligence is like running a ferry service between two different planets.

On one, everyone assumes that g, general intelligence, is a real and important trait, in which heredity plays a big role, and which is accurately measured by tests. On the other planet, g is a social and historical phenomenon--a belief one set of people have about themselves, or about other people. Most of the time the natives on each world speak happily only to each other. When you land with your news of the other place, the usual reactions are incredulity (those people don't exist!), anger (no one can believe that!) and/or a droning recitation of the sacred texts (please see these 114 papers listed below). See the comments on my Pinker-Gladwell post for examples.

Is it possible to foster communication between these two worlds? It seems like a job for serious journalism. Yet those who venture into the subject end up being flamed as shills for one side or the other.

Maybe a better approach is to describe controversies within each school, as that's probably where the ratio of thought to rhetoric is highest. Here's a case in point (and thanks to Frank Forman for cluing me in): there's a controversy over whether it is possible to say that a nation has a stable and characteristic level of intelligence.

On Planet History, this idea wouldn't be taken seriously for a minute, because it claims today's Egyptians, Jews and Asians are identical with their ancestors (at least for the important trait of intelligence). As Patricia Cohen explains here, historians and anthropologists know that any people's ties to its ancestors are a combination of fact, myth and selective emphasis, re-created in each lifetime, and shaped by the needs of the moment. This cultural activity occurs on top of the genetic facts, and is not congruent with them. While there are genetic differences between populations in the Middle East, those contrasts don't follow the same borders as the cultural ones.

For example, Patricia Cohen's last name is traditionally associated with descendants of the kohanim, the priestly caste of Biblical Judaism. A decade ago a team of geneticists found a genetic marker common in men who identified themselves as kohanim. But the "Cohen Modal Haplotype" is also found in some Yemenites, Uzbeks and Italians. Addressing that fact, the discoverers recently reported that they've created a more precise marker, which is found only in Jews. That is significant news at the genetic level of analysis--but it doesn't change the fact that there are many Jews who are kohanim but who don't have the genetic marker. They remain kohanim because kohan is a cultural concept, not a genetic one.

When it comes to the shared history that defines a people, as Marc Bloch wisely said, it is the present that shapes the past. Cohen's piece yesterday revisited the issue in the context of a pot-stirring book, The Invention of the Jewish People, which reminds its readers that this is just as true about the foundational beliefs of Israel as it is for any other nation.

On Planet IQ, though, matters are different. There, theories are circulating which claim to explain group differences in everything from sperm quality to longevity to income via variances in "national IQ." So, in that world, the idea that today's Egyptians are in some sense "the same" as their ancestors is hotly debated.

In the January 2010 issue of the journal Personality and Individual Differences, Jelte M. Wicherts, Denny Borsbooma and Conor V. Dolan say today's average IQ test scores cannot be used as a measure of ancestral intelligence, for several reasons. Migration and gene flow from "outsiders" are on the list, but most convincing, I thought, is the simple fact that population-average IQ scores haven't even held constant over the few decades since IQ tests were invented. Instead, scores have been rising steadily since the tests were developed (that's the "Flynn Effect," named after the psychologist James R. Flynn). Wicherts and the other authors point out that their own country has seen IQ averages rise by some 3 points per decade over the past 60 years. "Knowing that national IQs can fluctuate by more than a standard deviation over 50 years,'' they write, "we may question the relevance of contemporary national IQs to peoples that lived thousands of years ago.''

In a reply, Richard Lynn argues that the Flynn effect marks a rise in all scores throughout the world--meaning that it simply exists on top of an unchanging 10,000-year-old gap in average intelligence among races. In their reply to this reply, Wicherts and company say: pyramids. The obvious intelligence and success of ancient Egyptians doesn't jibe with the unremarkable IQ scores of present-day people in that country.

And that's a pretty striking argument: Over the course of history, riches, intellectual advancement and political power have passed from one people to another. If some groups have always been more intelligent than others, then history's ups and downs are evidence that intelligence isn't so important after all. On the other hand, if group differences in smarts are nonexistent or always-shifting, then intelligence-level isn't a stable trait for any ethnic group. "National IQ," if there could be such a thing, can be either constant or important. But how could it be both?