How Much Were People’s Lives Worth in the Ming Dynasty?

How much was a human life worth in the Ming dynasty? I mean literally, how much money did a life cost? Although one might wonder whether such a question is even a valid one (how can we quantify the value of a human life?), my recent reading of some official correspondence has led me to believe that part of this question can be answered.

The Ming state regularly paid money to reward soldiers who killed or captured rebels in battles, encourage former rebels to surrender, and (less frequently, it seems) redeem captured Ming subjects whose lives were in danger. As it turns out, some extant sources record not only the amount of silver paid for such endeavors, but also the differing prices assigned to people of different gender and age. They do not reflect the economic value of an individual on the labor market, but they do, in a way, show how much the state was willing to pay for the lives (or deaths) of law-abiding subjects and/or dangerous rebels.

Below is a list of numbers that I have found through a very random reading (skimming) of the writings of Ming officials. Some of my following descriptions might turn out to be wrong after more thorough reading and further research, but I thought I would share my notes here, as I’d be interested to hear what others think. Obviously there will be regional and chronological variations, but so far I find the numbers to be surprisingly consistent. Some speculative and bold conclusions based on the sources that I have seen so far:

  • An adult male was worth about 1-5 taels?
  • Women were worth about half of men, and children were worth even less?
  • A horse might have been worth more than a man?

… Or am I getting something wrong? We’ll see when I find more materials in the future…

The Numbers

The head of a bandit/rebel: 1 tael, 1 3 taels, 2 5 taels 3

A bandit/rebel captured alive: 0.2-2 taels 4

  • Main bandit: 2 taels
  • Minor bandit: 1 tael
  • Woman: 0.5 tael
  • Male or female child: 0.2 tael

Money given to bandits/rebels who surrendered: 1-3 taels 5

  • Strong person (man?): 3 taels
  • Old or young person: 1 tael
  • Person who came with a strong horse: 10 taels
  • Person who came with a weak horse: 5 taels

Ransom for buying back captured Ming subjects: about 3 taels per person 6

  • The exact amount was to be determined after further discussion, but the memorializer suggested that men of ages 16-60 were to be worth twice as much as women, the old, and the weak.

The Sources

  1. 李化龍 (1554-1611) 平播全書 堵截事 (四庫存目叢書 15.2a): “我兵微有斬獲威勢。亦振全賞半賞。如議支給。一級賞銀一兩。”

  2. 盧象昇 (1600-1638) 盧象昇疏牘 鼓練鄉勇 (明末清初史料選刊 p. 64): “鄉兵殺賊。與官兵一體給賞。殺真賊首級一顆。賞銀三兩。斬獲小頭目一名。賞銀十五兩。斬獲大頭目一名。賞銀三十兩。”

  3. 李化龍 平播全書 縣購規則疏 (四庫存目叢書 2.10a-15b): “查例苗級一顆。賞銀五兩。今加倍。應賞銀十兩。”

  4. 李化龍 平播全書 俘獲賞格 (四庫存目叢書 10.21a-22b): “但生擒一名。解驗的實。首賊賞銀二兩。從賊一兩。婦女每口五錢。幼男女每口二錢。”

  5. 楊嗣昌 (1588-1641) 楊文弱先生集 上宰相書 (續修四庫全書 45.13a-16a): “收降亦自有法。朝廷但給萬金。付遼撫鎭。聽其自來。強壯賞銀三兩。老小賞銀一兩。拐馬來者。臕壯賞銀十兩。疲弱賞銀五兩。”

  6. 張岳 (1492-1552) 小山類稿 乞立存活被虜人口賞格疏 (四庫全書 5.1a-3a): “男女長幼相折補。每口用銀。大約三兩。則銀一千兩。可活三百餘口… 能送出被虜人口數多。男子十六嵗以上至六十嵗。一口准一功。老弱婦女。二口准一功。”

Indexing Images of Everyday Life from Medieval Japan

This post is about a very interesting reference book, An Index to Painted Scrolls: Everyday Life in Medieval Japan 絵巻物による日本常民生活絵引 (Kadokawa shoten 1964-68; Heibonsha, 1984). The project was originally conceived and carried out under the editorial supervision of Shibusawa Keizō 澁澤敬三 (1896-1963), a powerful businessman who was once president to the Bank of Japan (1944) and Minister of Finance (1945-46), but was also known as a researcher in folklore studies.

Essentially, this is an index to scenes in medieval Japanese paintings that concern daily lives of the common people, with an amazingly attentive eye to details. The book was not published until after Shibusawa’s death, but his short essay dated 1954 explains that he had initially conceived of the project more than ten years ago, when he noticed that scenes from medieval paintings could offer excellent sources for studying the lives of the common people, if only they could be “indexed” properly.1 Starting in 1955, he gathered a group of collaborators who made sketches of scenes from the scrolls, numbered individual components within these scenes, and labeled them using carefully-defined categories. In all, the group indexed 26 works (most of which are painted scrolls) from around the late Heian to the Muromachi period [date to be confirmed].

A scene with eighteen components labled.

日本常民生活絵引 総索引
A section of the index.

What makes this book especially interesting and useful is that it indexes not just objects but also actions of people — from sitting, sleeping, fighting, to doing laundries, just to name a few. In this way, the book not only makes it easier to find images of particular topics, but also makes it possible to trace chronological changes in the depiction of the mundane objects and acts of everyday life. (A few months ago, I used this index to look for images related to medieval Japanese transportaiton, and was able to locate many images of roads, bridges and boats.) It is also noteworthy that all of this was undertaken at a time when material culture as an area of investigation did not receive as much attention as it does today.

More recently, there has been a project on Systemizing Non-Textual Sources for Humanities Researh (人類文化研究のための非文字資料の体系化) based in Kanagawa University, where a group of researchers published the first two volumes of the Index in English translation under the title Multilingual Version of Pictopedia of Everyday Life in Medieval Japan. This project also created several additional indices to East Asian paintings, including three on early modern Japan, one on Qing China (姑蘇繁華図), and one on Chosŏn Korea. Most of these publications are available for download on the project website.

  1. Shibusawa Keizō 澁澤敬三, “Can Image-Indices Be Made?” 絵引は作れぬものか, in Emakimono ni yoru Nihon jōmin seikatsu ebiki 絵巻物による日本常民生活絵引 (Heibonsha, 1984), vol. 1, viii.

趣文共賞 A Eulogy for Burned Books

Note: This post is part of a series where I reproduce random sources that I find interesting, with minimal annotation and commentary. Punctuation and translation (if given) are done by me unless otherwise stated. Corrections and comments are greatly appreciated.

I was recently looking through the literary collection of Wu Guolun 吳國倫 (1524-93; jinshi 1550), an official and prolific writer who came to be known as one of the “later seven masters” of the Ming. Among the many poems included in his collection is a curious one titled “Mourning for Burned Books.” As Wu explains in his preface to the poem, his entire library of books had been burned by a “violent guest” who apparently started a fire because he did not like the books in Wu’s collection. It seems that Wu was not even at home when this happened, since he writes that he had “heard” of the news recently. Wu was from Xingguo Subprefecture (興國州) located in Wuchang Prefecture, Huguang Province. Although the poem does not make explicit the location of this library, it was likely located in his native place in Xingguo.

I find this piece interesting mostly because of what it tells about access to private libraries: apparently a guest whom Wu may or may not have known was able to enter his (or his extended family’s?) library in his absence. Several lines of the poem (妻孥存鼠竄 城郭任鴟張; 却遂吾兒嬾 虚增過客傷) also seem to suggest that Wu was upset that his family members had let too many untrustworthy guests enter the library. (For more on the topic of library access, see Joseph McDermott, A Social History of the Chinese Book, pp. 134-47.)

The fact that Wu reacted to what must have been a devastating news about the loss of his treasured books by producing yet another literary text is somewhat incomprehensible to my modern senses, although it might have been perfectly reasonable for a literatus who used poems to express his feelings on a daily basis. I have not read the poems itself closely, and I have only translated Wu’s short preface below. The full text of the poem is included here for the curiosity of readers who can appreciate these texts much better than me.

Source: Wu Guolun 吳國倫, Danzhui dong gao 甔甀洞稿 (Siku cunmu congshu edition), 18.17b-19a. (The text below was taken from Airusheng’s 愛如生 Database of Chinese Classic Ancient Books 中國基本古籍庫 and checked against the Siku cunmu congshu edition.)

悼焚書 Mourning for Burned Books


In my twenty years of travel as an official, the books that I purchased have numbered nearly ten thousand volumes. Building a library to store them, I have treasured them greatly. Every time I returned home, I would quickly decline the guests and sit down on top of the library, feeling at ease and forgetting that I was poor. And so, very few of the youths in our village even knew that I was poor. Recently, I heard that there was a violent guest who came up to the library. When he searched through my boxes of books and saw nothing that satisfied his intentions, he started a fire, burned the books, and left. Quietly mourning my books, I have composed forty rhymed couplets to console myself.

萬卷從吾好 千金不易裝
少年窺槖籥 垂老厭圭璋
散帙高連屋 懸籖疊滿牀
牛毛原汗漫 雞跖亦煇煌
天啓河圖秘 霞生石墨香
寄身編竹穩 遊思結繩長
几上赤虹現 楹端彩鳳翔
秦碑留故蹟 汲塚弔餘芳
露綴三經席 星標六藝場
百家馮羽翼 諸體叶笙簧
野叟飡芝訣 園公種樹方
靈蛇蟠入次 繡虎伏成行
緝理茅茨閣 週遭薜茘墻
幔雲飄旖旎 簾影動瀟湘
户牖都鉛槧 山川此棟梁
敢同莘野樂 差異竹林狂
歳月資弦誦 精英屬表章
正須防散逸 誰更卜災祥
宦拙家逾遠 時艱力未匡
夢魂驚燕雀 消息閧豺狼
舊業千羣象 斯文厄九陽
鬼神憎篋笥 兵燹縱縹緗
豈復秦時暴 無容魯壁藏
何曾分玉石 渾欲戰玄黄
井吐商辰色 煙昏太乙光
總爲蝌蚪崇 遂作螙魚殃
噀雨遥難制 廻風近莫禳
熛氛騰赤館 烈熖倒靑囊
未乏燋頭客 其如白額郞
妻孥存鼠竄 城郭任鴟張
誤改于公户 慙非魏士鄕
絲桐歸火帝 符籙獻空王
重比周彝徙 珍逾孔履亡
刼灰期未辨 裨竈語能詳
錦目殘烽擲 珠函敝帚當
輪今悟糟粕 角且叩宫商
却遂吾兒嬾 虚增過客傷
爲儒甘刺豕 託隱慕屠羊
市肆歸堪借 言筌靜欲忘
有形俱幻化 何用熱中腸

The Print Run in Movable-Type Printing

This is a rather specialized post that might be of interest to book historians only. While collecting sources for my dissertation research, I recently saw a curious colophon in a book by the Ming official Xu Xuemo 徐學謨 (1522-93), Collection of Extraneous Things Learned in Shizong’s Reign (Shimiao shiyu lu 世廟識餘錄).1 A biji-style compilation on events of the Jiajing reign (1522-66), the book was printed by Xuemo’s son Xu Zhaoji 徐兆稷. As Zhaoji’s colophon explains, the book was printed using a borrowed set of movable types:

It has been more than ten years since this book was completed, yet we have been too poor to print it. We could only borrow movable types to print a hundred copies for now, keeping them in our family collection and not daring to let them circulate in the world. Yet even with movable type it takes quite a lot of trouble, so [the printing] could not be continued on. It is hoped that the reader will understand this. By Xu Zhaoji.

Xu Zhaoji’s colophon

I was excited to see this colophon because a book that states its own print run is quite rare as far as I know. Also interesting was the number a hundred — I was aware that books printed with movable type tended to have lower print runs compared to those printed using woodblock, but somehow I had expected that even these books would have had a print run of several hundred. Assuming that typesetting constituted the most time-consuming part in the process of book-making, would it not be the case that the great cost associated with typesetting would have been much larger than the combined costs of paper, ink, printing, and binding, so that once someone had prepared the typeset pages, he would have wanted to print something more than just a hundred copies?

First page after the colophon — preface by Xu Xuemo

I imagine there must exist some detailed studies on these issues, but since I have not been able to locate one yet, I decided to compile a list of other known print runs. To do this, I went through Zhang Xiumin’s 张秀民 History of Printing in China 中国印刷史 (Shanghai renmin chubanshe, 1989), which contains a substantial section on the history of movable type printing from Song to Qing. The table below is a summary of the known print runs mentioned in this book, drawn from either the extant books themselves or premodern catalogues and second-hand descriptions.2 In order to retain the ambiguity in some of the numbers, in describing the print runs I have retained the original Chinese expressions provided in Zhang’s book, some (but not all) of which have been quoted from the original sources.

Year Title /
(Place of Printing)
Print Run Zhang
Page #
1298 旌德縣志 王禎 (旌德) * 百部 674
1574 太平御覽
周堂 (浙江) 百餘部 689
1597 城真寓存稿
李登 * 數百本 680
徐兆稷 (嘉定?) * 百部 680
郭相奎 (杭州) 百許部 680
(杭州) 百餘部 680
1720-25? 古今圖書集成
雍正帝? (北京) 六十至六十六部? 718
1774-? 永樂大典各種佚書 乾隆帝 (北京) 五至二十部或
1805 甫里逸詩 周氏易安書屋 百部 707
1830, 1831 南疆繹史勘本
李瑤 (杭州) 八十部 (1830)
百部 (1831)
修凝斋集 若干部 697
淮城信今錄 百部 697
數部 703
寒支集 若干部 703

Contrary to my original expectations, the list suggests that print runs of a hundred or lower were more often the norm. It is also notable that there are several cases (marked by asterisk signs * ) in which the sponsor of the printing was the author himself or someone in his family, including the case of Xu Xuemo. I am not sure whether these numbers are supposed to be surprising, but to me, they indicate the relative ease with which personal works of this type could be printed, albeit in small numbers. It would be interesting if we could learn more about how, in the minds of the Ming-Qing writers and readers, these small-scale printed copies differed from works that were reproduced and circulated in manuscript form.

Added on 2013.02.04:

I noticed after writing this post that there was a short article by Inoue Susumu 井上進 on Ming movable-type printing,4 where I learned several additional points:

  • Inoue makes the same speculation that the print run of movable-type printing was about one hundred, citing four of the examples that appear in the table above. Additionally, he cites two Japanese examples that support this number (p. 104).

  • The article also discusses the print run of block-printed books. These were about 200-300 copies for low-demand books such as local gazetteers, and ran up to 1,000 copies or more for books that were in higher demand (p. 104-5).

  • Despite Xu Zhaoji’s colophon that I quoted above, it seems that Xu Xuemo’s Shimiao shiyu lu 世廟識餘錄 was later reprinted using woodblocks, first in 1608 (by Xu Xuemo’s grandson) and then in 1614. Extant copies of these two editions are now held in Japan and Taiwan respectively, and I have not been able to see either of them. (See Inoue p. 109 and p. 448 footnote 43.)


  1. Edition in Siku quanshu cunmu congshu 四庫全書存目叢書, shibu vol. 49. For more on Xu Xuemo, see Chao-ying Fang and L. Carrington Goodrich eds. Dictionary of Ming Biography (Columbia University Press, 1976), 585-87.

  2. Zhang Xiumin’s book is very encyclopedic and informative, but it is not sufficiently rigorous in its citations. I have not checked the original sources to confirm the accuracy of the information reproduced in here.

  3. The numbers of juan given in square brakets [ ] were added by me using book catalogues. They may be different from the particular editions described by Zhang.

  4. Susumu Inoue 井上進, “Mindai katujikō” 明代活字考, in Min Shin gakujutsu hensenshi : shuppan to dentō gakujutsu no rinkaiten 明清学術変遷史ー出版と伝統学術の臨界点, (Tokyo: Heibonsha, 2011), 90-109.

A Quicker Way to Filter Google Search Results by Language

Many people must have the experience of searching a word on Google and receiving results in languages different from the one originally intended. This is especially so for students of certain fields of East Asian Studies such as premodern Japan and Korea, where one might look up a proper name in Chinese characters, only to find a long list of Chinese-language pages in the results page.

One would expect that there is a simple operator that specifies the desired language of the results, but this does not seem to be the case. Of course, you could go into Advanced Search and choose the language from a drop-down menu, but this requires at least three clicks and a lot of scrolling. When working with multiple search terms and constantly shifting from one language to another, you would want a more efficient method.

As it turns out, there is a small trick that allows you to filter the results by language without using the Advanced Search page. A more detailed explanation is available on this page, but essentially this involves adding a text string at the end of the address generated by Google:

Filtering Google by Language Screenshot

The codes for languages that I use most frequently are as follows:

  • Japanese: &lr=lang_ja (Sample)
  • Korean: &lr=lang_ko (Sample)
  • English: &lr=lang_en (Sample)
  • Chinese: &lr=lang_zh (Sample)
  • Simplified Chinese: &lr=lang_zh-CN (Sample)
  • Traditional Chinese: &lr=lang_zh-TW (Sample)
  • Other languages: For a list of codes for all available languages, see this chart.

It would be very frustrating to type these strings every time I need to specify a language, so I use TextExpander to help ease this process. TextExapander is a Mac application that automatically produces a longer string of texts when you type a short trigger. (You can also find less expensive alternatives for both Mac and PC.) For example, I have configured TextExpander in a way so that I can type “;gjp” to have it automatically convert into “&lr=lang_ja,” and the trigger “;gkr” converts into “&lr=lang_ko” and so on. In this way, all I need to do is to type the short trigger at the end of the address bar, and I can easily narrow down the search results by language.

This is the quickest method that I have been able to figure out so far. If anyone knows of a better way, I would appreciate the information very much.

Happy searching!