Guest Post: Counting the Carnivores: Who Ate Meat in Early 20th Century China?

The Historical Cooking Project is pleased to publish Thomas DuBois's work regarding meat consumption in China. The post's emphasis on methodology serves to remind us about the important role source material has in shaping the ways we understand the past. 

Counting the Carnivores: Who ate meat in early twentieth century China?

China’s food history is a minefield of myths and misunderstandings.

Tasty as it is, General Tso’s chicken had nothing to do with the real general Zuo Zongtang, and the dish is virtually unknown in China. And despite it’s newfound popularity, the spicy flavors of Sichuan Sauce would certainly have been unavailable to the historical Hua Mulan who, had she actually existed, would have lived a thousand years before the first chilies reached China.

beef shop in Yunnan, China (photo by author)
There have nevertheless been great advances in uncovering the content of historical diets in China. Archeologists have unearthed physical evidence from tombs and lake beds, literary scholars have mined historical novels, travel accounts and scriptures for clues, and a new generation of amateur enthusiasts in China have recreated recipes from historical novels like the Dream of the Red Chamber 紅樓夢. This year marks the publication of the first English-language translation of Yuan Mei’s eighteenth century Recipes from the Garden of Contentment (Suiyuan shidan 隨園食單). This multi-year project was the work of Sean Jy-Shyang Chen, who also posts individual entries in his award-winning blog, “The Way of Eating.”

The need to understand historical diets becomes strikingly clear in light of China’s current influence on the global markets for food. Since the 1980s, China has undergone its own “livestock revolution,” emerging as by far the world’s largest producer and consumer of pork, and among the top consumers of beef, chicken, and even dairy. From Brazil to Australia, the impact of China’s demand for meat has transformed markets for meat, live animals and animal feed.

How new is China’s appetite for meat? The answer depends on what one means by “new.”

From 1980 to 2016, China’s annual meat consumption rose from 13.6 to 50kg per capita, a fourfold rise in a single generation. But that starting point followed many decades of created austerity, including at least one massive famine. Looking back further, the evidence starts to look wildly contradictory. Two eminent historians have separately stated that “the Chinese diet is largely/was overwhelmingly vegetarian.”[1] A famous and endlessly quoted study vividly depicts the Chinese countryside as crowded with destitute peasants barely eking out a precarious existence, like a “man standing permanently up to the neck in water.” [2]

But other sources show Chinese people eating meat regularly. Gastronomic tracts like Recipes from the Garden of Contentment feature all manner of meat dishes, and pioneering work on living standards and food consumption in China has surmised that meat consumption in eighteenth-century China could have matched or even exceeded that in Europe. [3] One late nineteenth-century travel writer estimated that southern Chinese diets consisted of about five parts vegetable and one part animal, while northern diets were about half and half. [4]

So which is it? Did people actually eat meat in China? If so, who, what kind, and how often? In an agrarian country that equaled Europe in size and diversity, answers will no doubt be extremely local.

No one source provides the key, but thinking back to those wonderful Annales studies that pieced together an understanding of medieval diets from as unlikely a source as labor contracts[5], I gathered three very different sets of data—price surveys from near the end of the Great War, profiles of China’s livestock industries from the 1930s, and dietary surveys conducted over a span of three decades—to draw a composite picture of the early twentieth century.

Placed side by side, each of these sources not only revealed something quite unique about diverse habits of meat consumption within China’s vast territory, they also showed different ways to approach the problem.

Although fought primarily in Europe, the Great War deeply affected food supplies in Asia. Provisions were sent from Asia to feed troops and refugees, and a dearth of merchant shipping disrupted vital cargoes of rice, sugar and milk to hungry port cities like Saigon, Singapore and Manila.[6] In Japan, war-driven inflation drove up prices of basic foodstuffs to the point of provoking nationwide riots.

Late in the war, the Japanese Ministry of Agriculture and Commerce began tracking prices on the continent, especially in China, which was becoming an important supplier of certain foods (such as soybeans and beef) to the home islands. Methods were hardly scientific or even standard: consulates simply sent someone out to the market to record the prices of the day, without accounting for time of year, or using a standard list of measures, or even currencies (prices were quoted in Chinese yuan, Russian rubles, Shanghai dollars, and even currencies such as guantie 官帖 that were leftover from the already-defunct Qing dynasty).

Prices for meat, fish and vegetables in Zhengjiatun, Taishō 9 (1920). From Market prices of daily necessities in Zhengjiatun [鄭家屯ニ於ケル日用品相場表ノ件]
Japan Center for Asian Historical Records

So what can prices show? 

Although collected without a common method, these prices can still show relative relationships within a single market, such as what items like meat cost locally in terms of other materials such as grain or wages. By these standards, meat seems to have been more expensive in the south of the country than in the north. With a day’s wage, a skilled laborer in the southern city of Hangzhou could have bought about 1.5 jin (at the time, one jin equaled roughly .6 kg) of pork, but twice as much in the Manchurian city of Tieling and nearly eight jin in the town of Siping.

Intriguing, until we remember that wages were also highly seasonal.

Market prices are more revealing of which kind of meat people preferred. Here again we can piece together a broad sense of geography, with the south of the country generally favoring pork, while the north preferred beef. This would make sense since Chinese medicine regards beef is regarded as a “warming” meat, and thus suitable for cold weather. Surprisingly, chicken was the most expensive meat in almost all of the 21 markets surveyed.

These conclusions are suggestive, but tentative. Even beyond inconsistencies in the data, the real problem is that prices show only possibilities, not actual choices. They can take us on a trip to the market, but they don’t tell us if anyone is buying.

While prices hint at the level of demand for meat, another source from a decade later clearly describes supply.

During the 1930s, a new Chinese government took steps to rationalize, modernize, and of course tax the country’s diverse economy. To do so, they followed Japan’s lead in counting everything, producing a “tsunami of statistics” (to coin historian David Faure’s phrase[7]) that included two detailed outlines of animal industries.

The first is a profile of urban abattoirs in fifteen cities. These include month-by-month figures on the number of animals killed, weight and price at slaughter.

Monthly slaughter statistics of cattle, sheep, pigs, chickens and ducks
in the city of Xining, Qinghai. From Urban animal slaughter statistics
 [各都市牲畜屠宰量統計], 1934. Source held in Academia Sinica, Taiwan. 
Abattoir surveys show that animal slaughter was both regional and seasonal. Production of cold weather meats like mutton and beef peaked in winter, just as prices dropped, showing the dual effects of seasonal demand and the practice of culling animals before the harsh northern winter set in. Prices varied considerably between cities. Meat was cheap in the western city of Xining, which was adjacent to pasture but far from transportation, and as much as five times higher in coastal Qingdao, which was a major center of meatpacking and export.

A second set of surveys presents a nationwide view of animal products. While abattoir data captured only a thin slice of urban centers, figures for products like hides and pig bristles (which were used for brushes) show the true scale of animal slaughter in a country that was still overwhelmingly rural.

Estimated sheep population, by province. In Investigation of the distribution of
important livestock resources in China
[支那ニ於ケル重要畜產資源ノ分布ニ關スル調查], 1940.
Held at Cornell University Library.

The numbers are astonishing: over 14 million head of cattle, 61 million sheep, and 21 million pigs for the year 1936 alone, and based on herd sizes, we can estimate that animals were being killed almost as soon as they reached maturity. Even if meat was simply a by-product of export demand for hides, clearly a massive amount of it was being produced. Without refrigeration, and only a small percentage of the meat being preserved by salting (the origin of products such as Jinhua金華 cured ham, or ganba干巴 dried beef), we must assume that most of the meat was distributed through wet markets for local consumption.

Dried beef for sale in Yunnan market (photo by author)
But like prices, even these detailed production data are incomplete. Knowing how much meat was out there does confirm a certain level of aggregate consumption, but is silent about the all-important details, specifically who and how. For these, we must look elsewhere.

The final part of the picture comes from two decades of food intake surveys, part of what Jia-Chen Fu calls the progressive “scientization” of nutrition research and policy in China.[8]

The first of these comes from 1922, just as northern China was emerging from two years of famine. Seeking to quantify rural diets under more normal circumstances, China-based researcher William Henry Adolph tasked students returning home for the summer with measuring and recording all of the food eaten by the family. This rough effort would be the first of many such surveys, conducted across China, and growing increasingly sophisticated over nearly three decades.

Diets of Han and Muslim villagers in Songpan. In Zheng Ji, Gu Xueqi.
“Diets of middle class Han and Hui in Songpan” [松潘中等漢回人膳食之調查] 
Kexue 24, 2 (1940): 110-116.

One survey records the average daily diet of an office worker in 1943 Chongqing as follows:

Grain (rice): 469g
Bean products: 112g
Fresh vegetables: 268g
Vegetable oil: 15g
Meat: 52g
Seasonings: 14g
Other (dried fruit, egg, milk): 21g
TOTAL: 951g

Measured out, these ingredients might have looked something like this:

(Incidentally, the most expensive item on this menu would likely have been the cooking oil, which explains the prevalence of water-based dishes like hot pot.)

While these surveys sample only very small numbers, usually just a few dozen respondents, they are the first capture such detail as gender, ethnicity and income, thus revealing actual patterns of meat consumption. While previous sources could suggest broad regional divides, dietary surveys show that the largest divisions were in fact those between households. Even within the same village or neighborhood, one family might consume meat as 13 or 14 percent of their diet, nearly the same as the American average, while their neighbors (especially poorer ones) were essentially vegetarian.

So what does all this mean?
The simplest takeaway message is that Chinese people were indeed eating meat, prodigious amounts of it, well before the economic growth and emergence of global supply chains that drive consumption today.

But for me, the greater revelation was how each source told a different story about meat in China and thus just how layers of meaning are contained even within this seemingly straightforward question. The simplest and most commonly cited measure—grams of intake per capita—actually tells us very little. The real story of meat or any other food, in China or anywhere else, has to include this much wider field of considerations.

[1] Bray, Francesca, Science and Civilisation in China. Volume VI: Biology and Biological Technology. Part 2: Agriculture. Cambridge University Press, 1984, 5; Simoons, Frederick J. Food in China: a cultural and historical inquiry. CRC Press, 1991, 293.
[2] Tawney, Richard H. 1979. Land and labour in China. Octagon Books, 38.
[3] Allen, Robert, Jean-Pascal Bassino, Debin Ma, Christine Moll-Murata, and Jan Zanden, “Wages, Prices, and Living Standards in China, 1738–1925: In Comparison with Europe, Japan, and India,” The Economic History Review 64 (2011): 8-38.
[4] James Hyde Clark, Story of China and Japan, Oriental Publication Co., 1894, 123.
[5] Le Roy Ladurie, Emmanuel. The Peasants of Languedoc. University of Illinois Press, 1974.
[6] Doeppers, Daniel F., Feeding Manila in peace and war: 1850-1945, University of Illinois Press, 2016. I discuss this period in a forthcoming article about the condensed milk trade.
[7] David Faure, “They went to the people but did they hear them? Comments on Field Research in China the 1920s and 1930s”in Out of the Archive: A Reader on fieldwork research on modern Chinese history, eds. Jan Kiely and Thomas DuBois, Routledge, forthcoming 2019.
[8] Jia-Chen Fu, “Scientising Relief: Nutritional Activism from Shanghai to the Southwest, 1937–1945,” European Journal of East Asian Studies, 11, no. 2 (2012): 259–282.

Thomas DuBois is a historian of modern China, best known for his three books on Chinese religion. Previously Associate Professor of History at the National University of Singapore and Senior Research Fellow at the Australian National University, he is currently a visiting researcher at Fudan University in Shanghai. His present project on China’s animal industries has produced articles on beef production chains in Manchuria, competition between Nestlé and Borden to sell condensed milk in Asia, and the political life of dairy in socialist China. His publications are available at He also maintains a food blog at