October 07, 2013
Haafu and half
Harry B. Harris Jr. is receiving his fourth star as a Navy admiral and assuming command the U.S. Pacific Fleet. What makes this appointment unique is that Harris is the son of a Navy enlisted man and a Japanese woman, Fumiko Ohno Harris.
Shisaku muses about how the brain trusts in Japan and China might read these particular tea leaves. But I was also struck by how Harris is a kind of mirror image of the Japanese actor Masao Kusakari. I'm not just talking about physiognomy.
Kusakari's father was an American G.I. He was killed in Korea, so Kusakari was raised by his mother in Japan. You can easily imagine one of those "trading places" scenarios.
When a haafu is Japanese/Caucasian, the Japanese genes usually dominate, except for height. Masao Kusakari is six feet tall, way above average for a Japanese born in 1950. Rangers pitcher Yu Darvish (Iranian father) is a towering six feet five.
In reality, the actual genetic mixing and matching is all over the map. Though with enough data it should plot out as a nice Guassian distribution.
Risa Stegmayer, co-host of NHK's Cool Japan (American father, Japanese mother), doesn't look especially Japanese, especially seated next to the very Japanese Shoji Kokami. (She speaks both English and Japanese fluently and without an accent.)
These variations can be found in "Yamato" Japanese too, though long periods of geographical and political isolation trimmed the tails of the distribution curve pretty short.
In the time-travel comedy Thermae Romae, for example, Hiroshi Abe (below) plays a Roman architect and Kazuki Kitamura is cast as Ceionius Commodus. In terms of their physical appearances, I was perfectly able to suspend disbelief.
Just to make sure, however, whenever Abe's character ended up back in Japan, the casting director surrounded him with "Japanese-looking" Japanese so he could observe how different they are (he describes them as having "flat faces").
This is pretty much axiomatic in population studies. Given a large enough cohort, the Guassian distribution of a common trait will reveal bigger differences within the cohort than than the mean differences across similar cohorts.