Research Experiences Compared: A Conversation

with Prof. Mataji Umemura


Mataji Umemura: You seem to be at a loss sometimes.

Konosuke Odaka: How did you know that, Professor?

Umemura: When I look over the materials I've been sent, I can tell. And you've been wandering around the dohyo1 [work area], and hanging around.

Odaka: One must start from where there is no dohyo. Now, the dohyo has finally been finished.

Umemura: You'd be well advised to take as much time as you can.

Odaka: Right. That's because, as you know, this is a group of amateurs.

Umemura : That's true at the start for anyone. At the beginning, everyone is an amateur.When I was doing the LTES project with Prof. Ohkawa years ago, we didn't know much at first. We didn't even know that the Ministry of Agriculture and Commerce2 had a statistical yearbook.

Odaka: What? Is that true, Professor? You didn't know about that...

Umemura: It's true, all right. We only knew about the Statistical Yearbook of the Japanese Empire.3 Also, we learned about the Census of Manufactures4 after we started working.

Odaka: Hearing about this makes me feel better.

Umemura: At the beginning, I spent a lot of time carrying a rucksack around used bookstores in Kanda5 with Mr. (Tsutomu) Noda.

Osamu Saito: You gathered statistics like that?

Umemura: There wasn't much then in the way of materials at the school. We learned different things from the owners of the used bookstores. They, in turn, learned a lot, step by step, as they gathered these sorts of materials. In due course of time, Hitotsubashi University became the price leader for old statistical materials. The bookstores also saw that if they collected these sorts of materials, they would be able to sell them.

Odaka: So in the beginning, you had little idea of what materials were where?

Umemura: That's right. At first, you had the old base. Once you've left it behind, you have to start from scratch. At any rate, starting out is always like that. When you look back later, even the little things, you just didn't know at the start.
You know, I started my investigation of the Edo era because I couldn't understand what was going on in the period preceding the Matsukata deflation [of the 1880s]. Now understand why; we couldn't find anything because Matsukata had destroyed the statistics compiled by Okuma.6

Odaka: It was part of a process of solving a puzzle about a transitional period, wasn't it?

Umemura: But Prof. Ohkawa must have thought that I was wasting time.

Odaka: Often we don't understand about past statistical figures and the circumstances in which they were collected.

Umemura: When I worked on Labor Force,7 I started from the premise that the figures provided by the population censuses and enterprise statistics8 were compatible. So, we used the census as a benchmark and extrapolated from the enterprise statistics because we thought that the premise was correct.
More recently, however, I have come to discover instances where the figures aren't compatible. Maybe the premise itself is incorrect. This is an example of the sort of problem that needs to be studied carefully. Should you have an opportunity to revise the time series, I'd like for you to watch out for relevant comments I left in Labor Force showing where I thought work remained to be done.

Saito: We've checked the first population census for Taiwan, and found out that the quality of the statistics is pretty good.

Umemura: That's excellent!
Speaking of study method, one needs to be careful about weights and measures. Weights and measures varied a lot in early modern Japan, too, according to the region until the early years after the Meiji Restoration.

Saito: It was in 1886 when Japan signed on to the Metric System Treaty.

Umemura: James Nakamura's main thesis centered on the measurement of rice yields in the mid-19th century. So the issue was essentially about weights and measures. This was obvious from the start, although we kept quiet about it [the weights and measurements issue].9

Odaka: If there are changes in a system of measurement, it does indeed make a big difference in statistical figures.

Umemura: The agricultural calendar may provide a precious clue to such a question. The planting season and the sequence of the farm work, for instance, was long decided. There were valid reasons for old customs like these; the farmers observed these practices year in and year out. That's why if you clarify the agricultural calendar and look at and investigate the statistics, you may be able to find valuable information not only about the nature of the materials but also hints about how to solve the problems.
A friend of mine is doing an intensive study of customs in agricultural villages in Java. You might ask him to tell you about what he has uncovered.



Prof. Umemura was, along with Tsutomu Noda and Kazushi Ohkawa, one of the three head researchers of the Long-term Economic Statistics (LTES) of Japan project conducted during the 1960s and based at the Institute of Economic Research at Hitotsubashi University. Konosuke Odaka and Osamu Saito are professors at the Institute of Economic Research.



Notes.

1 A dohyo is a sumo wrestling arena.

2 This ministry, which existed from 1881 to 1925, was a forerunner of the contemporary Ministry of Agriculture and Fisheries and Ministry of International Trade and Industry (MITI). The Ministry of Agriculture and Commerce statistics are the Noshomusho tokei hyo.

3 Nippon Teikoku tokei nenkan.

4 Kojo tokei hyo.

5 Kanda is a central Tokyo district known for its used bookstores.

6 Matsukata Masayoshi and Okuma Shigenobu were among Japan's most important political leaders of the 19th century. Matsukata succeeded Okuma as minister of finance in 1881 and reversed his basic policies.

7 Volume 2 of Choki keizai tokei [Long-term economic statistics]. 14 volumes were published from 1967 to 1987 by Toyo Keizai Shimposha. A one-volume English version, Patterns of Japanese Economic Development, co-edited by Kazushi Ohkawa and Miyohei Shinohara, was published by Yale University Press in 1979.

8 Enterprise statistics = jigyosho tokei.

9 James I. Nakamura, Agricultural Production and the Economic Development of Japan, 1873-1922 (Princeton: Princeton University Press, 1966).