We are gradually learning much more about the social foundations of classical Japan, thanks to a cluster of new and important studies on its population, agrarian technology, economy, and climate. In the book under review here, William Wayne Farris, who holds the Sen Soshitsu XV Distinguished Chair in Japanese History and Culture at the University of Hawai‘i at Manoa, has provided an extremely useful compendium of debates on demographic history, to which he adds new research, insights, and conclusions. The story provides the back story for his recently published Japan's Medieval Population: Famine, Fertility, and Warfare in a Transformative Age (University of Hawai‘i Press, 2006). The two are really companion volumes.
Farris's main conclusions, which he considers a hypothesis in need of further study, are these: the period from the eighth to the twelfth century saw either stasis, or more likely decline, in population due especially to foreign-borne epidemics. Drought, soil exhaustion, and meager irrigation facilities also led to frequent crop failures, while food shortages led to chronic or recurrent malnutrition. Even in the twelfth century, when some historians have argued that there was a significant expansion of the arable, Farris thinks that cold and damp summers led to frequent crop failures. If the population recovered in the twelfth century, he thinks it was only to eighth-century levels. Should his conclusions prove correct, they will significantly impact many of the ways we think about the Heian Period (794–1185). The work in this book is therefore foundational for future research.
In his Introduction, Farris provides readers with a useful overview of earlier debates concerning Japan's early demographics. We learn that, among others, in 1927 Sawada Goichi argued that the age of Shōmu Tennō (r. 724–749) represented a high point when the population reached 6 to 7 million. More recently, Kitō Hiroshi has hypothesized two distinct demographic cycles: one in Jōmon times, followed by one encompassing the long span from Yayoi-Kofun times up to the twelfth century. Kitō has also argued that demographic stasis began sometime in Yayoi or Kofun times, for unknown reasons, and that Japan's population did not resume significant growth until after the twelfth century.
These ideas make up the baseline from which Farris takes up the argument. He begins by recalculating population figures for three historical moments (750, 950, 1150), using elements from earlier theories while incorporating new data and insights. This process gives readers a good sense of how demographic calculations are made, and the choices and assumptions they rest on. For the eighth century, Farris chooses to use the total number of administrative villages (gōri) cited in an eighth-century legal commentary (the Rissho zanpen), which he then multiplies by the number of residents in an administrative village as Sawada calculated. He favors Sawada's figure because it allows for regional variations. After adding in non-cultivators (esp. slaves and urbanites), Farris arrives at a population of from 5.8 to 6.4 million persons. Then, moving to the mid-tenth century, Farris uses similar figures from ninth- and tenth-century legal compilations to estimate a population of from 4.4 to 5.6 million persons. These results represent a significant decline from the eighth-century totals. And for the mid-twelfth century, following Kitō's lead, he applies figures for arable fields from a fourteenth-century encyclopedia, the Shugaishō, the figures of which are thought to date back to the twelfth century. He divides that arable figure by an estimate of the amount of land needed to support an individual, thereby arriving at a population of from 5.8 to 6.3 million people. That is quite close to the eighth-century population, and on that basis, Farris hypothesizes that the Heian population essentially remained static between the eighth and twelfth centuries.
In Chapter 2, Farris turns to explaining this stasis. As readers of his first book, Population, Disease and Land in Early Japan (Harvard East Asia Institute, 1985), may remember, Farris himself suggested an answer twenty-five years ago: he argued there (after the world historian William McNeill) that from 700 to 1200 the Japanese populace suffered a high degree of pestilence, and that given its sparse settlement, the populace did not develop appreciable immunity, making gains in population impossible. Now, in Daily Life, Farris elaborates that argument more fully, setting out both written and archaeological evidence and considering its broad ramifications. Since official histories were no longer written after the early tenth century, Farris has had to comb through the individual and fragmentary documentary sources in Takeuchi Rizō's Heian ibun; he has also scanned courtier diaries and other compendia such as the Hyakurenshō and Nihon kiryaku annals. Particularly interesting in this regard is his discussion of archaeological finds that evidence rites to banish sickness or propitiate angry spirits. This reader would have liked more translations from the sources, providing both flavor and content. A helpful chart like that in Fujikawa Yū's classic on the subject (Nihon shippei shi, Heibonsha, 1969, pp. 11–71) could also have been added. But Farris nonetheless provides the gist, which suggests continuing and significant die-offs during epidemics into the twelfth century. He sums up his findings this way: “Ravages of epidemic disease in turn had widespread ramifications, including substantial depopulation and its concomitant phenomena: field abandonment, village desertion, and a shrinking tax base” (p. 51).
Beyond pestilence, Farris also argues that frequent famines decreased fertility. The written record indeed shows that until 900 poor harvests and starvation were reported somewhere in the realm every three years. While such records decrease when the official histories stop around 900, other sources suggest continued waves of heat and drought up to 1100. Dependence on mountain run-off for watering fields also made agriculture vulnerable to the vagaries of weather. And, in the twelfth century, cold and damp weather, possibly due to volcanism, resulted in famine, sickness, and social disintegration. Farris's findings here agree with Bruce Batten's recent discussion of climatic developments during classical times.Footnote 1
Diminished population, Farris thinks, had important ramifications that challenge current explanations of developments in Heian history. For instance, Farris thinks that it could well have been disease (not greed or weak structures) that rendered the Chinese-style ritsuryō system of government unworkable. He also suggests that greater mortality from disease in the Western Seto region could have given the advantage to eastern warrior families who rose to prominence during Heian times. Meanwhile, unrelenting vulnerability to pestilence surely influenced religious developments – the strong influence of Latter Days (mappō) fears are one instance. And while Farris agrees that nobles did cannibalize the provincial tax system, he does not think that greed was the reason, nor does he believe that private estates bankrupted official coffers. Rather, “demographic loss led to shrinkage of the tax base” (p. 58). By his reading, only as demographic stability improved in the twelfth century did a new fiscal system, based on public and private holdings (the shōen kokugaryō system), nurture a new degree of fiscal cooperation between elites at court and in the provinces. Farris's ideas here are sure to stimulate future research.
In Chapter 3, Farris continues his exploration of the social and technological markers of demographic stasis, including conditions in the agrarian regime, in the labor market, and in industry, trade, and urban centers. He agrees with other historians that a cycle of land clearance and abandonment prevented expansion of the arable – some 20 to 30 percent of paddy are thought to have been habitually out of cultivation. And he sees diffusion of the plow and draught animals as labor-saving efforts that had limited productive effect, given the evidence of ongoing frequency of famine. Unfortunately Farris's sources do not yield many numbers, although we hear of one estate (Ōba no mikuriya) in Sagami Province where production totals from twelfth-century paddies seem to have been no higher than those in good paddies during the eighth century. Meanwhile, in terms of social organization, Farris thinks that settlement in small hamlets with frequent migration separated marital partners in the duolocal marriage system of the time, limiting fertility and spreading infection. He points, too, to signs of labor shortages, such as the discontinuation of corvée, difficulty in finishing large-scale building projects, and decreased production of silk thread, woven goods, and ceramic wares. In the case of iron production, he thinks the reversion to vertical shaft furnaces during the ninth to eleventh centuries was a labor-saving effort. From this period he thinks that trade and the use of money shrank. Recovery and urban growth began again only because of the vibrancy of Sung Chinese trade penetrating the archipelago by 1050. Ports like Hakata then began to grow again, as did new urban centers inland (p. 83). What is not clear is whether the urban population grew more numerous than it had been in the eighth century.
I am neither a demographer nor an economic historian, but I find the substance of Farris's story here plausible. There will be debates as to his chronology and reading of the record – when was there a new negotiated settlement between central and local authorities, and did it not predate the twelfth-century shōen kokugaryō system? And when did Sung trade begin strongly affecting the Japanese economy – is 1050 a bit late? Was the diffusion of the plow and draught animals as inefficacious as Farris thinks? Should bilateral kinship be seen necessarily as a sign of minimal pressure on resources? Were duolocal residence and bridewealth frequent practices in Heian commoner marriage – how can we know when most of our evidence concerns courtiers, and when the fragmentary evidence that we do have for the countryside pertains to elites rather than cultivators, and that same evidence shows a broad spectrum of marital residence practices? As Farris himself indicates in his footnotes, there are lively ongoing debates concerning these issues, the outcome of which will affect the conclusions of this book. I also suspect that Charlotte von Verschuer, author of Le Riz dans la Culture de Heian, Mythe et Realité (Collège de France, Institut des Hautes Études Japonaises, 2003) – not listed in Farris's bibliography – will have counter-arguments to make. Readers of French will want to take a good look at her work, which insists that we give far too much emphasis to rice agriculture and not enough to the production of the other foodstuffs that formed 75 percent of the average person's diet in Heian times.Footnote 2 For the most part, Farris's story of the agrarian regime and its constraints on population growth focuses only on rice agriculture.
Farris has done a great service by advancing the debate on classical demographics and daily life. The reviewer surely hopes that researchers inside and outside Japan and in various fields will respond enthusiastically by giving more attention to issues Farris raises herein. This very readable, engaging, and straightforward book (total pages of text number only slightly over one hundred) will surely be widely read and pondered both by specialists and by those with a general interest in the social and economic history of Japan.
