The Guardian is putting out a call for people to write about archaeology and anthropology.

Finished up Rothman's Slave Country, which was good. Dug into Johnson's River of Dark Dreams. It's great, but frequently grim. Reading this stuff takes a psychological toll and we all cope in our own ways. Johnson likes to use irony and the occasional double entendre. He's of the school that you should use strong words for strong things, so his Mississippi valley is made of blood and sweat, but also "semen" and "shit". Now and then it's downright literary. In a discussion of manure, he finally declares shit (his word) to be liquid capital. That's true in an agricultural setting, in the minds of the reforming planters he's writing about there, and also a bit funny to my inner twelve year-old. Given the unrelenting bleakness, you take what you can get.

As promised yesterday, a map of the economic output of Japan by Koku per Km^2. Here's the middle Tōkaidō (Edo-Kyoto highway) section. Unfortunately I haven't figured out how to import the Tōkaidō itself. I'll get around to it eventually.

Map for the Meiji is here

The methodology is quite simple. The 国立歴史民俗博物館(National Museum of Japanese History) have their digital archives here. In it is the data base of late Tokugawa and early Meiji (1872) land survey data, which is paired with a municipality code for each entry. The code is a bit outdated, but is easy to match to the current code.

From that, it's easy to compile how much koku is produced per year in every single modern municipality. And then just plot that into Google Fusion Tables, with polygon data for each municipality pulled off the internet (specifically a Google Drive user who's user name is japonyol who kindly published the data). And load that into Google Map/Earth.

There are some inaccuracies:

1. On rare instances the data base is missing data, so I had to guess on some entries that don't have municipality code (i.e. over a hundred year old villages that don't exist anymore) which municipality falls into based on its name and my google-fu/Japanese wikipedia. Other times the data base itself misses some koku numbers, resulting on the sums of each province been a few hundred to just over a thousand hundred under the total displayed.
2. The boundaries are also stolen off wikipedia and are very approximate. In fact a lot of the times we don't even know where the actual boundaries of the provinces are. Though most follow the borders of modern municipalities, there are definitely overlaps which I encountered a couple times a day when sorting through the archive.
3. Due to me using modern municipalities because the data is readily available (and really I don't have enough knowledge to go any deeper), a problem readily visible is that some modern municipalities covers a city and then a ton of mountainous hinterland (Toyota city, Aoi Ward-Shizuoka City, Shinshiro city, and many more). So what should have been considerable concentration in the flat land gets diluted by all the mountains the municipality includes. Unfortunately not knowing how the numbers (both koku and area) divides, I don't know how to fix the problem. So if you see areas that has some flat land and then a tonne of mountains, just move those areas up a few shades in your head. Sorry.

The red are missing data. I'll put in a 0 for them next time. The red on the right is Mt. Fuji. The red on the left is a modern artificial island.

Here's a map of the Sengoku period.

This is made using the assumption that while the overall productivity of land increased greatly, the relative concentration of productivity within a province remained the same. From there I applied the percentage of overall production of each modern municipality to its pre-modern provinces. The totals for each province in 1598 (and other years) could be found here, from the 大日本租税志 a record of Japanese tax and economy published in the Meiji period.

There are two things wrong with that assumption:

1. Increased concentration around the domain castles (Edo, or in this case Nagoya, Okazaki, Hamamatsu, and Sunpu)
2. The widespread opening of new lands throughout the Edo period (so wide spread a ton of old villages actually have "new farms" in their names).

I am hoping these two effects more or less cancel each other out. It's not perfect, so just use it as an even rougher estimate.

Outlined in the map are the respective realms of the Imagawa and the Oda around the time of Okehazama. Of course Okehazama took place in 1560 and the numbers are for 1598, so it's even less accurate. But I don't think it'd be too far off. The Oda are in Orange (except Nobunaga's personal castle of Kiyosu, FML) and the Imagawa are in Yellow. The battlefield location is marked, and the dots around it are castles involved (black is unknown which side it was on at the time of battle).

Looking at that I actually think it's just as amazing as the victory of Okehazama is that Imagawa Ujichika and Yoshimoto are able use careful diplomacy to expand from Sunpu to conquer Tōtomi, most of Mikawa, and drive into Owari and come close to ending the Oda. To put it into perspective, Imagawa's three provinces come to 695,875 koku. Owari, Mino, and Ise, the first three provinces Oda Nobunaga would conquer, totals to 1,678,842.

EDIT: And I used the wrong colour for Kiyosu Castle FML.

Anybody got any recommendations for books/resources on the history of UAW, or labor in Detroit in general? (Layman here.)