Navigation – Plan du site
Homicide and Organised Murder in a Global Perspective

Figures of Deterrence in Late Imperial China. Frequency, Spatial Repartition, and Types of Crimes Targeted by Dismemberment under the Qing Dynasty

Jérôme Bourgon et Julie Erismann
p. 49-84


Quoique l’empire chinois ait employé un système de codification juridique sophistiqué pendant des siècles, on a très rarement tenté d’en mesurer l’impact sur la société. La numérisation offre la possibilité de collecter des données à une échelle suffisante pour évaluer les pratiques pénales. Cette recherche constitue l’esquisse de recherches futures, et se limite aux exécutions par démembrement (lingchi), forme la plus dissuasive de la peine de mort, car elle est la mieux documentée et la plus aisée à retrouver dans les archives. Une base de données d’environ 1 140 sentences de lingchi exécutées sur environ 260 ans sous la dynastie des Qing permet l’examen statistique de la fréquence, de l’évolution dans le temps, de la répartition géographique et de la nature des crimes concernés. Initialement, cette peine visait les crimes contre l’État tels que la rébellion ou le brigandage, mais nos données montrent que le lingchi fut employé de manière croissante pour conforter la hiérarchie interne à la famille, en visant les enfants indignes et les épouses meurtrières. Plutôt que l’arme rudimentaire servant à écraser les soulèvements souvent décrite par les historiens, le lingchi constituait un dispositif sophistiqué de soutien d’un projet confucéen complexe.

Haut de page

Texte intégral


1Numbers are rare in Chinese history, which forces China historians to rely on empiricism and rule of thumb. It may seem that legal history is less hampered by the shortage of quantitative data than economic or social history, since there are plenty of codes and casebooks to explore. Yet evaluating the efficiency of the legal system in general, or the implementation of death penalty and its deterring effect on the Chinese populace, the lack of reliable figures becomes a serious obstacle.

  • 2 Lee (1991).
  • 3 Lee (1991, p. 124).
  • 4 Buoye (2000).
  • 5 Lee (1991, pp. 116-117), Buoye (2000, pp. 238-239).

2As far as we know, there have been only two attempts from historians of China to treat crime and punishment statistically. Both rely on the xingke tiben, the routine memorials written by the provincial governors presenting cases of capital sentencing from their jurisdictions to the Board of Punishment. These documents were ultimately registered by a bureau of the Censorate specialized in penal matters (xingke), and conserved in the Grand Secretariat archives. In 1990, James Lee used copies of xingke tiben available at the Utah genealogical society to establish a first survey of 22,000 capital cases transmitted by four provinces (out of eighteen)2. With a yearly average of 2,000 to 3,000 death penalties submitted to the Assizes, only a minority, fluctuating between 12 and 40%, were executed. The system deliberately made executions uncertain, in order to enhance the role of the imperial mercy. Such uncertainty, added to the unevenness of recording throughout the provinces of the great empire, makes any general assessment challenging.3 In his innovative book, Thomas Buoye focused on 385 homicide cases caused by dispute over land and debt in the Guangdong province during the second half of the 18th century.4 Relying on Lee’s figures for the rest of China as a point of reference, Buoye subtly analyzed the social and economic causes of dispute leading to increases in homicide rates. These two pioneering studies regard judicial archives as documents for social history, and are interested in crime and punishment as indicators of social conflicts. They are specifically concerned about the gap between statistics and the elusive reality, the notorious ‘dark figure of crime’.5

  • 6 Chaunu (1968).

3This article is less concerned with ‘social realities’ than with the nature and significance of punishments. What we search in figures of executions are not so much reflections of social conflicts than messages from the State about proper social behavior. Ultimately, we aim at assessing the human cost of crime deterrence, in order to assess whether deterrence was efficient in preventing violence, or whether the remedy was worse than the evil. Whereas Lee and Buoye dealt with the legal definition of crimes as a distortion of the true criminality, we focus on the penal agenda of the State in its conceptualisation of crime and staging of punishment, as well as the dogmatic canonical background. This opens for a new set of questions : Were punishments correctly conceived, and did they reach the targeted objective ? Or did they go in a different direction, and turned counterproductive in relation to the values they claimed to defend ? Such dogmatic issues cannot be studied in fragments through provincial or periodical samples ; they require general evaluation on the scale of the whole empire, and for the whole Qing era. Admittedly, such objective is far beyond our current documentation and capacity of treatment. As a way to remove the obstacle, this article proposes a short-cut, by bringing together the growing part of the Qing archives that are now easily accessible through digitization. It has become possible to get what Pierre Chaunu called a “pesée globale”, a “comprehensive weighing”, of crimes and punishments as social practice.6 As approximate and incidentally flawed it may be, this comprehensive assessment raises issues that would not be apparent in a more limited and fragmented study. As such, we hope to open new field of reflection on law and crime in China.

Why focus on lingchi executions ?

4This paper focuses on executions by dismemberment, lingchi chusi in Chinese, which are more commonly known by the expression “death by a thousand cuts”. In reality the practice involved no more than eight to ten cuts. Why focus on lingchi, that constituted the most extreme form of deterrence in the Qing penal system ? As the harshest form of death penalty, it was assured the highest degree of visibility as well as certainty, due to its exclusion from the judicial revision process. Like other crimes, lingchi sentences were systematically conveyed through memorials drawn up by the provincial judge (ancha shi) on behalf of the provincial governor under whose authority cases were investigated, processed and submitted to the Board of Punishments in Beijing. Crimes committed in the Beijing metropolitan area were adjudicated directly by the Board of Punishment in a procedure called “immediate adjudication” (xianshen). Once examined by the specialized services of the Board, the cases were assigned to the “Three Legal Courts” (Sanfasi) composed by the directorates of the Board of Punishments, the Censorate (Duchayuan), and the Court of Revision (Dalisi). Lingchi sentences were then transmitted to the emperor for confirmation by the emperor, like all capital sentences.

5This judicial process was somewhat perfunctory, however, since lingchi was not an ordinary death penalty, it did not appear among the “Five Punishments”, the list of legal penalties in Article One of the Chinese penal codes, which had been in force since the Tang. The Fifth and most serious penalty was subdivided into “death by strangulation”, supposedly a milder form of execution because it left the body in its integrity, and “decapitation”, more severe since “head and body were separated” (shenshou yichu). While the Chinese legal system was based on the principle that the nature of crimes and punishments must be clearly and publicly defined and announced, so that people would be deterred from committing crime, the absence of the harshest punishment of all in the list of the legal penalties is intriguing. It makes dismemberment a terse and duplicitous device of deterrence.

6Lingchi was peculiar in another respect which was hidden in the depth of the code. In various sections, death penalty was allowed in order to fit itemized crimes, each time specifying whether the sentence required “immediate execution” (lijue), or whether, as in most cases, extenuating circumstances justified preventive “detainment in jail” (jianhou), meaning that the death sentence was suspended until its reexamination by the Autumn Assises. This procedure effectively reduced executions to a small number of “confirmed sentences”, while most death sentence were finally commuted to banishment or pardon. These broadly dispensed mitigations were welcome in a harsh legal system, but lingchi completely derogated from them. The few articles of the code that mention this penalty – the detail of which will be examined below – always implied “immediate execution”. Accordingly, lingchi sentences were excluded from the revision procedure of the Autumn Assizes. Introduced in the code by the conquering Mongols who established the Yuan dynasty, lingchi was originally intended as a sheer instrument of terror, and it remained so both under the Chinese Ming dynasty and the Manchu Qing dynasty. Its application therefore had to remain outside the boundaries of mercy and pardon, although the latter was quite generously dispensed in the penal system as a whole.

  • 7 On this swift procedure of execution, see Suzuki (2003).

7In addition, legal provisions could be circumvented to make the execution more “immediate”. Prior to its legalization and codification by the Mongols, lingchi had started as an expeditious device against rebellions, and these shady origins resurfaced in the procedural shortcuts which were at the disposal of provincial governors. Indeed, a significant number of lingchi were meted out through a speedy process, reverently dubbed : “Respectfully requiring a sovereign order” (gongqing wangming). In practice this meant that the provincial governor had already proceeded to the execution, and requested a post-facto imperial rescript to give this arbitrary execution an appearance of legality.7

8This plethora of derogations deprived the accused of the due process which was normally guaranteed by the revision system. This is undoubtedly the evil face of the imperial legal system, reserving in many respects a sphere for extra-legal measures and draconian justice where legal procedures were systematically distorted or ignored. Although a curse for the population, the lingchi penalty and its procedures are a blessing for the statistician, who can assume that all identified lingchi sentences were executed with very few exceptions. For this particular punishment sentence meant execution, with only a negligible margin of discrepancy.

9Another advantage of lingchi for historical enquiry comes from the formal nature of the memorials sent by the provincial authorities to the central boards. These reports mostly consisted of the lists of convicted persons, categorized according to the sentences they received. Only those who were ‘short listed’ were named, while the others were left unnamed under the character “deng”, meaning etc., et alii. For instance : “Condemned to immediate decapitation are Wang X, Zhang Y, and 18 others”. Lingchi condemnations were judged serious enough to be presented separately in a single memorial, where each individual was identified by name with the crime he or she had committed. Whenever lingchi sentences were mentioned in a routine memorial including several different types of penalties, lingchi sentences were placed at the top of the list. In group executions of rebels or bandits, the ringleader was usually condemned to lingchi and his full name was used to identify the case, while his subordinates who were condemned to decapitation were mentioned only cursorily by the et alii stock designation.

10This conspicuousness has allowed the identification of a significant sample of over 1100 lingchi sentences executed between the Qing conquest in 1644 and the abolition of lingchi and other ‘cruel penalties’ in 1905. This set of cases is briefly described below and more details are provided in the appendix.

Common assumptions about lingchi and the issues they raise

  • 8 Bodde, Morris, (1973).
  • 9 Bodde, Morris (1973, p. 112, n.51).

11My first encounter with lingchi execution was in Bodde and Morris’ Law in Imperial China, the textbook wherein many legal historians of China learned the ropes.8 In this book, the description of the Qing legal system ends with a section headed “Western glimpses of Chinese punishments”, describing the cases selected, the authors explain, “because of their seeming accuracy and objectivity, as well as detail”.9 The most memorable, for most readers, is the long and minutely detailed description of a series of decapitations of bandits, ending with the dismemberment of their ringleader, witnessed by Thomas T. Meadows, a Senior Officer at the British Consulate in Canton in the 1850s. Cherished among Sinologists for the perceptiveness and reliability of his views on the Qing administration, he is also known for his pleas for British support to the Taiping rebellion, although his efforts were to no avail. He wrote two books on Qing China. In one he included what remains the only accurate eyewitness account of execution by dismemberment. Here is a description of the dismemberment of the ring leader :

  • 10 Meadows (1856, pp. 655-656). Quoted in Bodde, Morris (1973).

As soon as the thirty-three were decapitated, the same executioner proceeded, with a single-edged dagger or knife, to cut up the man on the cross, whose sole clothing consisted of his wide trousers, rolled down to his hips and up to his buttocks. He was a strongly-made man, above the middle-size, and apparently about forty years of age. (…) As the man was at the distance of twenty-five yards, with his side towards us, though we observed the two cuts across the forehead, the cutting off of the left breast, and slicing of the flesh from the front of the thighs, we could not see all the horrible operation. From the first stroke of the knife till the moment the body was cut down from the cross and decapitated, about four or five minutes elapsed. We should not have been prohibited from going close up, but as may be easily imagined, even a powerful curiosity was an insufficient inducement, to jump over a number of dead bodies and literally wade through pools of blood, to place ourselves in the hearing of the groans indicated by the heaving chest and quivering limbs of the poor man. Where we stood, we heard not a single cry ; and I may add that of the thirty-three men decapitated, no one struggled or uttered any exclamation as the executioner approached him.10

12The impact of this breathtaking rendition is put into context by Bodde and Morris who stressed the exceptionality of the event :

  • 11 Bodde, Morris (1973, p. 111) – (my italics).

The fact that these accounts from north, central and south China have to do with events occurring respectively in 1860, 1857, and 1851 is important because all three dates fall within the period of the Taiping rebellion (1850-1864). This conflict, probably the most destructive civil war of all time, devastated much of the richest land of China, killed twenty or more millions people, and very nearly overthrew the Manchu government. These then were years of crisis, very different indeed from the period only thirty or forty years earlier to which the bulk of the cases translated in this book belong. Hence it may be not wholly fair to project what is said about penal institutions during the one age back to the earlier age of political stability.11

  • 12 For reflections on this silence, see Bourgon (2007) ; Brook, Bourgon, Blue (2008, chapter 1). The s (...)

13These few lines raise a lot of questions. First, why rely on “Western glimpses”, why not compare them with Chinese accounts ? The answer is that there are no such accounts. As curious as it might seem, no substantial account of a lingchi execution from a Chinese writer can be found for the whole Qing period ; Chinese literati did not write about such things.12 More importantly, to what extent are these accounts by Meadows and other Westerners representative of Chinese Law under the Qing ? The italicized passages above hint that the shocking aspects of repression and deterrence were due to the extraordinary conditions that faced the Qing State at that time, such that “it would not be wholly fair to project to the earlier age of political stability”. There is no reason to challenge this judgment in general, nor to deny the depth of the mid-19th century crisis caused by the Taiping rebellion and other contemporary uprisings as well as their tremendous impact on the Qing state and Chinese population. A naïve reader of Bodde and Morris may infer from their comment that dismemberment and other deterring devices were extreme measures employed mainly during an age of turmoil, in sharp departure from the normal course of justice in the heyday of the Qing.

14Hence the question that underlies this article : to what extent is Meadows’s account of a bandit’s execution by dismemberment and beheading representative of Chinese law under the Qing ? Was lingchi a tool of deterrence used in periods of exceptional turmoil, which significantly receded in peaceful times ? An implicit assumption by Bodde and Morris is that lingchi was mainly employed against rebellions and gangs of violent bandits. In another part of their book the authors provide a complete list of the crimes that were punished by lingchi. Nevertheless it is clear that in their view this punishment was primarily a technique of mass terror intended for reestablishing State authority and public law and order disrupted by revolts and armed banditry. Other crimes punishable by dismemberment, such as murders of elder relatives, bloody feuds between families or lineages, are deemed marginal and anecdotal compared with the major disruptions of the public order. Must we regard this kind of executions primarily as a tool of political terror used by the State against society ?

15Our data gives insight on the following five issues :

  1. Frequency : how many lingchi executions took place during the 261 years between the Qing conquest in 1644 and the abolition of the ‘cruel penalties’ in 1905 ?

  2. Trend : do we discern an upwards or downwards trend in the use of lingchi during the Qing period ?

  3. Periodization : were there periods which saw an abrupt increase or decrease in lingchi executions ? Were some Qing Emperors more prone than others to authorizing ‘slicing’ executions ?

  4. Localization : were lingchi executions spread evenly throughout the empire, or did some provinces or regions witness the executions more frequently than others ?

  5. Dissuasion target : Among the few major crimes for which the code provided for a lingchi sentence, which crimes were most frequently subject to linchi execution ? Were there significant changes across provinces and time ? Were certain types of crimes more targeted or repressed in particular provinces, or during particular periods ?

Problems and potential of the collected data

16We have gathered a sample of 1,140 persons killed by the lingchi process across 1,017 executions. The discrepancy between the two figures results from the fact that several dismemberments could take place in the same execution, as the dramatic case of the Monk Snowfield and his followers will show. Let us start by presenting our source (for further details see appendix 1).

17Our two major sources are the “Archives of the Grand Secrétariat” (neige daku dang’an) and the judicial chronicles in the Shenbao which reproduced the “Beijing gazette” (jingbao). The archives preserve the record of 357 persons dismembered in 338 executions between 1644 and 1854 ; the Shenbao supplies us with the record of 453 persons dismembered in 453 executions. This makes up the first sample of 810 lingchi, a good basis for our enquiry, but it presents two vexing flaws. First, the periods covered by each source are grossly unequal in length : 210 years for the archival source versus 33 years for the newspaper source. Moreover, the eighteen years from the last year of the archival source, 1854, to the first year of the journal source, 1872, are entirely devoid of data. As a matter of fact, the gap is wider, since the Grand secretariat archives show a sharp drop of executions from the early 1840s down to almost none in 1854. Given that these are the years of the high tide of lingchi and many other forms of violence caused by the Taiping, Nian, and Muslims rebellions, it is clear that these small numbers, or none at all, most likely reflect the disruption of the Qing administration rather than a real decrease in executions. During these thirty years, the rebels as well as the officials acted outside the law and off the record. Even though the lingchi sentences imposed on the Taiping leaders as well as on the chiefs of other rebellions were recorded, most punishments occurred through summary “executions on the spot” (jiudi zhengfa) which were rarely reported. The magnitude and lawlessness of violence for these two decades renders all judicial statistics unreliable.

18Similar problems might negatively affect the data of previous periods. For instance, the Grand Secretariat records only 34 executions for the 60 years of reign of the Kangxi emperor, i.e. around 1.4 lingchi per year. The average is much lower if we consider that a single collective execution held in 1667, that is, during Oboi Regency, counts for 27 lingchi, or 80% of the total number of the individuals executed. If we consider the number of the executions, rather than the individuals, the number falls to eight, which is barely more than one every ten years. Either Kangxi’s was the Grand lenient reign that admirers of imperial ‘légendes dorées’ like to venerate, or the data are flawed. Our inclination for the second view prompted us to compensate the gap in time and flaws of data by referring to three complementary sources. These are collections of administrative and judicial documents issued by State agents, either officially or unofficially (see appendix 1). Such sources allow raising the rate of executions under Kangxi to a more credible 3.9 per year. However, there are still spans of over ten years entirely “blank” of all executions under Kangxi or Yongzheng reigns, which lead us to suspect deficiencies in judicial registration.

19Resorting to complementary sources could not compensate for the sharp discrepancy between the last thirty years, corresponding exactly to the reign of Guangxu, and the rest of the Qing period. It simply allowed reducing the Guangxu share of the lingchi executions toll from 55% to 40%, but still a remarkably high proportion ; this question will be discussed later. The imbalance might result from the difference of coverage between the sources kept or issued by the central Qing administration, and the Shenbao, a modern newspaper that daily extracted information from the Beijing gazette as well as from other local reports thanks to an efficient network of correspondents. Hence a possible underestimation of the numbers prior to Guangxu, which only a systematic survey of the judicial archives in Beijing would be able to rectify.

  • 13 The legal term in such case was not “putting to death by dismemberment” (lingchi chusi), but “slaug (...)

20Despite the problems indicated above, our two major sources provide for complete and accurate information for over 1,100 lingchi sentences. We are provided with the name of the individuals condemned to dismemberment, along with the names of their followers sentenced to decapitation ; the place where the crime was committed ; the name or title of the official or the government services that supervised the execution at the provincial and central level (or both) ; and, less frequently, the local magistrate who started the investigation. Details or at least outlines of the cases are given, informing us of the nature of the crime, the reasons for the sentence in Chinese described as “the name of the crime” (zuiming). Finally, the gender of the criminal is normally specified except for a few cases. In brief, these are highly reliable execution registry and the executions most certainly took place except when the condemned died in prison. Even in the latter cases, which are duly specified in the data, laws and practices usually commanded that the convict’s body be dismembered, so that they can also be counted as lingchi executions, even though their deterrent effect on the public is more questionable.13

21The complementary sources are less reliable. They contain the name of the condemned, the authorities that confirmed the sentence, the “name of the crime”, and the date of sentencing. In around one hundred cases, however, the locality is not indicated, or even the province, where the case originated. We laid down as basic requirement for our samples that the case include the name of the condemned, a clear statement of the sentence to lingchi, and the nature of the crime. On the whole, this set of samples is sound enough to attempt a “comprehensive weighing” of lingchi executions under the Qing and to find answers to the five questions raised above.

Frequency of linchi sentences and trends throughout the Qing period

  • 14 On the abolition decrees of 1905, see Bourgon (2003, pp. 851-862).

22The 1140 lingchi sentences pronounced between 1644 and 1905 produce an average of 4.5 executions per year. It amounts to 5.5 per year if we do not take into account 65 ‘blank years’ for which we suspect that the records have been lost. From the linear diagram (see b. in Appendix 2), we start with a low average under the Shunzhi, Kangxi and Yongzheng reigns, especially in the last two decades of the 17th century and the first three decades of the 18th when the number of lingchi is at its lowest. From the 1730s we observe an increase that continues throughout the 60 years reign of Qianlong. The upward trend picks up after 1770 and reaches its peak under Jiaqing during the first decade of the 19th century. Then, under Daoguang comes a dramatic decrease ending with the ‘blank years’ in the 1860-1870s, which can be ascribed to the disruption of the judiciary caused by the huge rebellions of the mid-century. The mid-1870s saw the beginning of the noticeable rise which occurred in the Guangxu era : with more than thirteen per year, the average is ten times higher than in Kangxi era (four times if the ‘blank years’ are deducted), three times higher than in the Qianlong era. 439 executions over thirty years represent a little less than 40% of the total for the whole Qing era. Strikingly enough, by the time of the abolition, the number of executions was at its highest with 22 executions in 1904, which marks the second highest record year after 1883. Certainly, 1905 might have reached a new top if the abolishing decree had not brutally interrupted the rising trend in late April of that year.14 These changes are at odds with the expectations inferred from the European cases : in Europe the abolition of the “tormented executions” and judicial torture came after a long period of gradual decline, during which these kinds of practices were increasingly regarded as barbaric and backward. How are we to explain this discrepancy ?

23The diagram shows a regular trend upward from the last years of Kangxi (after 1700) to the reign of Jiaqing (1810). If, as we suspected, the trend downward that started during this era can be explained by the judicial breakdown, we can compensate for the lack of data by conjecturing a continuation of the trend upward from the peak of the Jiaqing reign in 1806 to the peak of the early Guangxu reign. Since the gap in the diagram corresponds to the bloodiest years of the civil conflict, when lingchi executions of the kind that Meadows witnessed took place in great numbers. It is reasonable to imagine even higher peaks under Xianfeng and Tongzhi – between 1850 and 1875 – than under Daoguang : thus the line would reach a peak around 1860, over twenty executions per year, which then would slightly decrease and remain at around fifteen per year under Guangxu.

24Even if we refrain from such bold statistical speculation, the general trend upward seems undeniable with the available data. At the present stage, it might be interpreted as a confirmation of Bodde and Morris’ assumption of a growing recourse to lingchi as a reaction to the unprecedented civil conflicts of the mid-19th century. The most suitable interpretation would be as follows : once the critical years of settlement under Shunzhi and the regencies of the early Kangxi reign passed, lingchi was seldom employed during most of the Kangxi and Yongzheng reigns, marking the acme of the dynasty. It was more frequently employed under Qianlong, whose reign was plagued by millenarian revolts and banditry during its last years. Then followed a troubled period during the early 19th century, mostly concealed by the absence of archival data. Finally, once the State administration was restored and a better recording of judicial activities by the new media became available, the change appears dramatic : lingchi had become the unmistakable device of mass deterrence in a Chinese empire in general turmoil.

25This interpretation seems coherent, but it relies on the idea that lingchi was mostly a tool for repressing rebellions and restoring ‘public order’. Conceptually, it places the dynamics of the evolution not so much in the law but rather in society which was marred by ethnic, social, and political conflict. In such a perspective, lingchi is just a tool of repression in the hand of the State. The momentum is external to law, which reflects events passively : more rebellions to repress implies more lingchi sentences and executions. This paper will contend that, contrary to these interpretations, law is dynamic and creative : by determining what is criminal, it shapes the social reality in which the crimes take place. Far from a mere tool, lingchi was a conception of society and power cast in a dogmatic mold and grounded in the most sacred Classics. The trend of sentences therefore had its own dynamics, relatively independent from the socio-political events, no matter how tremendous the impact of such events.

Targets of deterrence : crimes liable of lingchi and their share in effective lingchi sentences

26Crimes subject to lingchi in the Qing code fall in the following categories :

    • 15 Law against Robbery or banditry (“Qiangdao”, art.261 in the Qing code) did not provide for lingchi (...)

    Rebellions, high treason, conspiracies (real or attempted) against the dynasty, and armed banditry. In analyzing the statistical and diagrammatical data, all the forms of collective actions that subverted or disrupted public order have been grouped under the same category of “Revolt”, marked by R in the diagrams of the appendix15 (see diagram c. and d. in appendix 2)

    • 16 Art.287 : “Killing three persons in a same family” (Sha yijia sanren). The term “familicide” is cur (...)

    Killing three persons or more in another family. This was clearly a measure against family extermination, or “familicide”, that marked particularly vicious forms of blood feuds between lineages. These are the F crimes in the diagrams.16

    • 17 Art.284 : “Plot to kill grand-parents or parents” (Mousha zu fumu fumu) ; art.286 : “Plot to kill t (...)

    Murder, attempted murder, or similar action, by a younger person on the elder person, or an inferior on the superior in the same family or clan. This is the general category of “parricides”, marked by P. This category can be split into 5 sub-categories, according to whether the victim of the crime was the father (Pf), the mother (Pm), other close relatives (aunt, uncle, elder brother or sister all regrouped under the collective noun of zunzhang, that can be translated as “senior elder”, hence Pse in the diagrams) ; the killing of the household head (Phh) by a slave or hired domestic ; and the killing of the husband by his wife, which I propose to call viricide, as a Latin symmetric of uxoricide – marked Pv in the diagrams.17

    • 18 Art.287, last paragraph ; and art.288 : “Picking up parts of the body of a living person” (Zhaishen (...)

    Inhuman crimes such as maiming and desecrating of the body of an enemy killed in a feud, or taking a body part of a living person for cannibalism or witchcraft (or both), have been ranged in the category of “atrocities” (A).18

27These categories of crimes affect different realms, which can be regarded as concentric circles. The first circle concerns the public order of the empire, and the crimes in question have the potential for disrupting large areas or even the whole territory, as was the case of the massive rebellions of the mid-19th century. For crimes within this category, lingchi fulfills its ‘classical’ function of protecting the State and the dynasty. Crimes of the second category delineate a more restricted circle : a village or a district, the order of which could be disrupted by blood feuds between rivaling families engaged in ‘private wars’ or vendetta that affected local communities in many parts of the empire. In this field, lingchi does not defend the State itself but local peace and order.

28The third circle is the family, in its twofold legal definition : Firstly, as a household including the head of the family, all relatives living with him, slaves and hired servants. Secondly, as including all relatives linked by duties of ritual mourning, even those living separately. These two kinds of kinship – the existential one and the ritual one – were taken into account in the legal framing of a “parricide” sentence, thus condemning to lingchi a domestic servant suspected of having plotted the death of the family head as well as a younger brother accused of having killed his elder brother even if living in separate dwellings. Moreover, a junior member of the family who killed a distant senior relative, for instance an aunt visiting his parents, was also liable of lingchi. Here, dismemberment was intended to maintain a strict subordination of the ‘younger junior’ (beiyou : those who were of a younger generation, or junior in age) vis-à-vis their senior or elder relatives (zunzhang : those relatives of an older generation, or elders of the same generation). This subordinate status was for life : one remained the lifelong ‘younger junior’ of one’s elder brother or sister, which changed only by the arrival of new members in the family who became one’s ‘younger junior’.

29The fourth circle is the most intriguing, since it delineates the individual body : death by dismemberment for desecrating the human body, either in the course of a vengeful feud or with the intent to use parts of the bodies for witchcraft or wicked sectarian cults. Here, lingchi serves to defend a certain idea of human dignity, at least its bodily integrity. Despite the original significance and intentions behind making this a crime of lingchi, such cases became rare under the Qing, with most of them occurring at the height of blood feuds (see diagram c in appendix 2).

  • 19 This old classic may seem outdated in the Qing, but its meaning had been revived under the Ming, an (...)

30The underlying connection between these four realms can be easily understood as a reflection of the relationship between the Empire or Universe, the family, and the Self which is at the foundation of the Chinese conception of an orderly State. The ‘Great Study’ (Daxue), a Confucian Classic compiled in the 4th century BC., established the nexus between self-cultivation, or the ability of the true Gentleman (Junzi) to govern himself, the ability to rule his family, and consequently to govern and pacify the empire19. The same dialectic relationship between macrocosm and microcosm, with the familial hierarchy at its center as an epitome of the Filial Piety, reflects the Confucian cardinal virtue to unify the three realms, which permeates the whole range of lingchiable crimes. No matter how horrible a form of execution, and how legally questionable its status as a clearly codified penalty, and irrespective of the procedural shortcuts in its implementation, lingchi was perfectly in line with Confucian orthodoxy in its conceptual background, suitable for the crimes that violated filial piety.

31The four categories of crimes had their specific realms, so that their respective proportions are indicative of the main trend in the repression of the most heinous crimes. The code merely provides us with a theoretical definition of the most heinous crimes. To identify which of these crimes were most effectively repressed in practice, we need to turn toward the lingchi sentences pronounced by the judicial courts. The data allow for calculation of the share of each of the four categories of lingchi sentences, and for representing the figures in pie charts covering the reign of eight Qing emperors (see c. “Four categories of lingchi sentences for eight Qing emperors reigns” in the appendix).

32Some structural characteristics become immediately apparent. Thus, the relative proportion of each category of crime during the reign of Shunzhi is strikingly similar to that of the reign of Guangxu, and to the intermediary reign of Daoguang. Condemnation for parricides represents around two thirds of the crimes, while the last third is shared between revolts and familicides : there are more revolts (30%) than familicides (7%) under Shunzhi (30% vs 7%) and also under Daoguang (22% vs 8%), but under Guangxu there are slightly more familicides than revolts (14 vs 16%). During the long 18th century, under Yongzheng, Qianlong and Jiaqing rules, the occurrence of parricides remain under or around 50%, and revolts somewhere between 33 and 45%. The share of revolts tends to increase throughout the time, until the early 19th century. Finally, there are the anomalies of the Kangxi and Tongzhi periods, during which revolts accounted for 94% of all crimes under Kangxi and 100% of parricides under Tongzhi. This aberration, however, simply indicates that the data are flawed. Apart from parricides and revolts, familicides would require a special study : they remain stable between 7 and 9% for most of the period, and jump to 16% under Guangxu. Judicial archives as well as novels by popular authors, confirm that familicides and their punishment with lingchi formed a basso continuo of local life throughout the empire. They occasionally came with “atrocities” (maiming of bodies, dead or alive), which slowly declined from 4% to 2% of the sentences between the Qianlong and the Daoguang reigns, and disappear thereafter.

  • 20 Spence (1979, pp. 89-98).
  • 21 See Carlitz (2007, pp. 234-257).
  • 22 A sketch of the Wang Weiqin case appears in the introduction of Brook, Bourgon, Blue (2007).

33Under the late-Qing, familicides became the most chronicled form of crime. Some famous cases offer fascinating examples of interconnection, if not confusion, between reality and fiction, jurisprudence and gory tales. This is well illustrated by the historian of China, Jonathan Spence, who included a chapter in his best-selling Death of Woman Wang which mingled his description of a real case of “killing four people in the same family” with a novel of the popular early-Qing novelist Pu Songling.20A very similar case occurring in the late 18th century was turned into a popular novel by Wu Woyao a century later.21 This novel was published at the moment when a real case made the headlines, and recently established modern newspapers reported the investigations and the trial in a novelistic form, which adopted all the stereotyped characters known from fiction. This was a true case, however, which ended with the execution by lingchi of Wang Weiqin and the beheading of his elder son and four of his collaterals in October 1904.22 Fiction and real cases share stereotyped elements : a feud between a peaceful family of peasants and a powerful bully enjoying strong connections with the local administration leading to the almost complete annihilation of the former by the latter’s Mafia-like clan. Among the details of spectacular violence that inundated the public media, the most revolting was the high numbers of victims : nine in the 18th century case that was turned into a novel, twelve in the Wang Weiqing case, the latter including six children aged between two and thirteen, as well as two women. Both of these blood feuds only turned into legal cases because one single victim managed to survive, and sought justice in the courts. In the real case well as in the fictional account, this became the starting point of a long odyssey, strewn with ambushes organised by the powerful and influential bully, until the nearly-dead plaintiff met his or her savior in the guise of a ‘Clear Sky’, an upright and benevolent mandarin who received the complaint and confronted the villain. As a happy denouement of the fictional plot, the feud is eventually reversed when the villain is executed by lingchi, his kin and followers condemned to death or exile, his wealth confiscated and given to the victim in compensation.

34In such cases, the legal penalty strikingly mirrored the circumstances of the crime, the State committing a familicide in a bout of retributive justice. This was the only category of crimes for which lingchi sentences met broad popular consensus. The legitimacy of the State was enhanced when it redressed the abuses of the local gentry, and repressed feuds by turning their own methods against them. For this type of crime, public opinion and popular culture undoubtedly supported the State’s effort to use terror as deterrence.

35There is enough evidence to establish that the use of lingchi could not be ascribed solely to the repression of revolts, even though this component was always present and could become overwhelming depending on time and location. Even taking into account the potential flaws of our data, it seems reasonable to argue that the majority of lingchi sentences were not pronounced against rebels or armed bandits, but against parricides. Throughout the period under examination, revolts account for hardly one third, while parricides account for more than half of the lingchi sentences, and familicides for ten percent combined with 1.5% of “atrocities”. However, revolts and bandits are dominant for a longer period, when we combine the reigns of Kangxi, Qianlong, Jiaqing and Tongzhi, i.e. 155 years out of 261, while parricides are dominant only at the beginning (under Shunzhi) and, more preeminently, at the end (under Guangxu), whose thirty years’ reign saw more than 40% of all lingchi sentences. Therefore, the predominance of parricides over revolts is established in the very last decades of the empire. In other words, it seems safe to infer a shift from a revolt-focused lingchi toward a parricide-focused lingchi during the 19th century. This shift occurred in the 1820s, if we take seriously the striking similarity between Daoguang’s and Guangxu’s pie-charts. To explain this shift, let us first consider of the case of revolts, to turn to parricides next.

Lingchi as restoration of public order

36The idea that lingchi was devised mainly for repressing rebellions and bandits is firmly grounded in famous historical events, such as the dreadful fate of famous Taiping leaders. Our data include eloquent examples of that kind, most notably the repression of a revolt led by Buddhist Bonzes, whose chief was called Snow Field (Xueye). It took place on the sixth year of Kangxi (1667), when the reign was still under the regency of Oboi, two years before he was tried and imprisoned by the young Emperor. This was a period of violent conflict among the ruling elite. Oboi had the entire clan of one of his co-regents executed, a fate that came to haunt him and his kin some years later. This background might account for the merciless treatment of Snow Field partisans, of whom nothing is known otherwise. The memorial to the Board of Punishments reads :

The Governor General of Shanxi and Shaanxi provinces and other places, concurrently in charge of military affairs and logistics, vice-president at the Ministry of War, and Censor Lü Chongling respectfully reports in re : the rebelled monk has surrendered.

  • 23 For the wives and children of the condemned ‘being confiscated’ meant to be deported and reduced to (...)
  • 24 Neige daku, Archives of the Grand Secretariat, ‘Basic material’ (jiben ziliao) 065912-001.

Your Servant has under his control the chief rebel, the monk Snow Field, who connived with Li Faming et al. for printing fake seals and spread libels full of absurd ideas and foolhardy bravadoes. Your servant has proceeded to interrogations and established that Snow Field was the vicious head of a “Complot of Sedition”, punished by dismemberment according to the Law. Those who plotted with him for faking seals and spreading libels, that are [the monk] Illuminating Light and ten others who followed the rebellion and received the libels, as well as Zhou Shaofeng and thirteen others, each being found guilty of participating to a plot of rebellion, all of them (i.e. 27 persons), without distinction between leaders and followers, are according to the Law, condemned to death by dismemberment. There has been no contestation of the sentence. As for those who were aware of the situation, who concealed the prohibited items, or deliberately let escape [prisoners ?], namely Wang Sanye and four others, as well as the members of the family of the chief rebel such like Qi Fengxi and twelve others, they are all condemned to decapitation. (Here follows a list of accessories, sentenced to various non-capital punishments) The wives and the children of the criminals are all, according to the law, confiscated (ruguan)23, as well as their wealth and lands, with specifications detailed in the appended folios.24

37This harsh sentence reveals some salient characteristics of lingchi execution against rebels, which can be sketched out as follows :

  • Collective incrimination conforming to legal key-terms in the statute and sub-statute : “all of them, without distinction between leaders and followers” (jie, bu fen shoucong). Here, the abnormally high number of leaders condemned to dismemberment suggests that the rebellion was considered a serious threat for the dynasty, although the Bonze Snow Field and his unfortunate companions left no other trace in the records.

  • combination of lingchi for the main culprits and decapitation for accessories, with exposure of the head for both categories ;

  • ‘chain indictment’ (lianzuo) of relatives and neighbors who were supposedly aware of the plot but did not denounce the plotters to the authorities ;

  • Derived penalty for wives and children – slavery, castration for males, or even death, deportation and ‘confiscation’ (ruguan) for females, meaning that they became State slaves.

  • 25 On the Boxers’ execution, see Hevia (2003) ; and Bourgon (2012a).

38This case is singular because of the high number of lingchi sentences in proportion to other derived penalties. In cases that posed much more serious threats for the dynasty, like the capture of the Taiping general Shi Dakai and his troops in 1863, a greater number of people were executed, but two thirds of the rebels were pardoned, one third beheaded, and only Shi Dakai was dismembered. This great rebellion of the mid-19th century might well mark the end of the use of lingchi against grave political crimes. Under the Guangxu reign, the share of “revolts” in the pie chart is the smallest during the Qing period, with 22%. Moreover, most of these “revolts” were in fact armed banditry, which became endemic throughout the empire at the turn of the 20th century. It is worth noticing that none of the modern revolutionaries who plotted an armed coup to topple the late Qing were condemned to lingchi ; all died in fighting, or were beheaded when captured. The same can be said of the Boxers, whose presumed leaders and supporters were all beheaded, mostly under pressure from the Western Powers.25

39Therefore, it seems reasonable to infer that the “lingchi against great rebellion” with its apparatus of mass terror, collective incriminations and serial executions, made its last round in the suppression of the Taiping, the Nian and Muslims rebels in the 1860s, and then almost disappeared over the last thirty years of the Qing. The dramatic increase of the lingchi sentences during this period is entirely due to the increase of the parricides, accounting for more than 70% of the total, as the pie chart clearly shows.

40As shown in table d., the category of “revolts”, banditry included, represents little less than one third of the lingchi sentences for the entire period, a proportion that had fallen to 15% in the last decades of the Qing. By contrast, parricides represent more than half of lingchi cases for the entire period, with more than two thirds occurring during the last decades. For every rebel dismembered for challenging public order, two or three rebellious children or wives were dismembered for breaching the family hierarchy in the household.

Lingchi used for restoring hierarchical order within the family

  • 26 See Bourgon (2012b).

41Parricide does not exist as a legal concept, and there is no Chinese word for it.26 Our category is a compound of various crimes labeled and listed in distinct articles of the Qing code. Practically, lingchi sentences for “parricide” split into a dozen of sub-categories in our chart (see appendix 2, table d. and related pie chart), which can be outlined as follows, in order of decreasing frequency :

Murder or attempt against :

percent of total parricide

1. Husband by his wife (viricide)


2. Seniors-elders (zunzhang) by younger relatives


3. Father


4. Mother


5. Household Head by slaves or hired servants


6. Mixed cases




42Greek tragedy, Roman Law, and Freudian psychology taught us to construe ‘parricide’ primarily as the murder of a father by his son, or secondarily of a mother by one of her children. In Qing China however, parricide was a much broader category, involving distant relatives, which includes murder or attempted murder by slaves and servants against the household head. The killing of a parent comes only in third and fourth places, after offences against ‘seniors-elders’ (zunzhang). The category covering conflicts between domestic slaves and their master has the lowest number throughout the period, nearly disappearing under the Guangxu reign (just 4 cases recorded, or less than 1%). Yet this was very different by the beginning of the Qing, as we will see below. Most striking is the prevalence of spousal killing, where “offences against husband” comes in the first place with 25% of lingchi. The “parricide”, in Qing China, was primarily a wife killing her husband, or “viricide”.

43The statistics reveal significant changes throughout the Qing period. Under Shunzhi, in the mid-17th century, the killing of a master by one or more domestic slaves was still the most important category of lingchied parricides, accounting for 47% of all such executions, ahead of the viricides, which already surpassed 40%. Master killings recede gradually along the Qianlong and Jiaqing eras (with 7% and 4%, respectively), and drop to less than 1% over the last thirty years of the Qing under Guangxu. During the same years, viricides reach their peak, with almost one third of the “parricides” (31%), and almost one quarter (22.5%) of all the lingchi sentences. Inasmuch as lingchi sentences reflect the inner conflicts in the household, there is a clear shift from slave-against-master killings toward a wife-against-husband killings, while other parricides do not show any striking change.

  • 27 Women represent less than 10% of the executed in 18th century London, according to Gatrell (1994, p (...)

44The sex ratio of the condemned people is highly unusual when compared with studies of other parts of the world, where most executions concern men, in an approximate proportion of 10 to 20 men for one woman.27 In Qing China, of 1140 persons condemned to lingchi, 210 were registered as women, 929 as men, and 59 were of unspecified gender ; this means that women represented 18.5% of the lingchi sentences. This is in itself a remarkable proportion for the harshest form of punishments. Moreover, the proportion of women increases sharply during the late 19th century : starting with 31.5% of women under Shunzhi, the ratio falls to 2.4% under Kangxi, and remains between 10 and 15% for the 18th and most of 19th century, with a striking, although erroneous, record showing no woman executed under Xianfeng and Tongzhi. Then comes the horrendous last thirty years, with 114 women out of 446 lingchi sentences, or 25.6%. Of course, this overrepresentation of women is a direct consequence of the prevalence of parricides over the other categories of lingchi sentences, and of the prevalence of the ‘viricides’ in the category of parricides. This double prevalence is overwhelming under Shunzhi and Guangxu, the first and the last reign of our period.

  • 28 This squares with James Lee’s observations that “le seul ‘supérieur’ contre lequel se dressaient ma (...)

45In short, lingchi was increasingly used to repress ‘private conflicts’ within the family rather than revolts, banditry, and other ‘public crimes’. Among family crimes the one most frequently punished with lingchi was the viricide, the killing of a husband by his wife. During the last thirty years of the Empire, one in four ‘great criminals’ sentenced to the harshest punishment was a woman who had killed her husband or another ‘superior relative’.28

Spreading deterrence : the spatial repartition of lingchi in the Qing empire

46Thus far, lingchi as a means of mass deterrence has been evoked in its theoretical, qualitative, and quantitative aspects. The crucial issue remains how deterrence was conveyed in practice from the centers of State power toward the provinces and local communities. Capital executions, notably those applying the harshest forms of violence, involve a particular relationship between the center and the provinces or the non-Han marches of the empire (Mongolia, Tibet, Xinjiang). They suggest regional peculiarities which our data leads us to capture.

47This is the most challenging issue, however. First, the flaws in the data that obscured changes over time pose even more serious obstacles when we turn to the distribution between regions. Major differences set apart the eighteen provinces of China proper from the immense crescent of Manchurian, Mongolian, Turkestan and Tibetan steppes and mountains that were incorporated through the Manchu conquest of China. Apart from the well-known and persistent differences in climatic conditions, population density, ethnic composition, etc., they constituted two different worlds in terms of institutions and rules. Chinese provinces were ruled for centuries by the imperial administration, and submitted to codified law and centralized judiciary. The outer margins were left to their own aristocratic elites, under the initially loose, but gradually tighter and more intrusive supervision by a special administration, the Court for Ruling the Frontier-Marches (Lifanyuan). Although the majority of its personnel consisted of non-Han, mainly Mongolian, origin, this Court tended to promote the Chinese way of regulation, codified law, and harsh punishments. But this general statement does not explain why Tibet had almost no lingchi sentences, while Mongolia and Manchuria have the highest rate of lingchi executions per capita, although all three were under the jurisdiction of the Lifanyuan. One of the aims of this research project is to gain a clearer understanding on how the legal and judicial components of the imperial state functioned as a coherent system.

  • 29 Zhao, Xie (1984) ; Cao (2001).
  • 30 Here we took example on recent studies of executions in modern and pre-modern Europe : Gatrell (199 (...)

48The second problem deals with the population. For an evaluation of the effectiveness of deterrence in any given area, examining the numbers of executions is not enough ; an estimation of the size of the population is needed, in order to calculate a ratio between the two figures. There is much uncertainty about population figures for each province during the Qing. For this project we had to rely on data found in Chinese books with relevant information despite their drawbacks and possible inaccuracies.29 We also had to deal with two different sets of figures which were hardly compatible : on the one hand, the population in vast areas numbering millions or tens of millions ; on the other hand, 1,140 individuals executed over 261 years in the whole empire, with two or three dozen people at most for a given province. The ratio can measure the toll of the penalty on the total population – very small indeed. If we focus, not on the executed, but on possible onlookers, the ratio can hint at the deterrence effect, once we take into account those elements impossible to quantify, such as the ‘media coverage’ of the event : its advertising prior to execution, the publicizing and staging of the execution itself, and, more importantly still, the memory of it as kept or emphasized by oral or written accounts and narratives. Admittedly, local memories can vary significantly according to cultural and social factors, but it seems safe to suppose that the ratio of lingchi executions per habitant for a given period has an immense impact on this memory, a basic ingredient of deterrence. But what is the time span for this kind of memory when such deterrence effect loses its impact ? We used imperial reigns like Guangxu, which is convenient for designating the last thirty years of the empire. But for the sake of comparison, we cut the periods into twenty-year blocks, presuming that this was the ‘generation that witnessed or heard about’ any particular execution30. We took the average number of the population reported for each of the periods of twenty years, and calculated the ratio that resulted in six categories, ranging from 1 lingchi sentence for less than 500,000 inhabitants, to 1 lingchi sentence for more than 1,250,000 inhabitants, every twenty years (see maps c, d, e).

49We start with a rather fragmentary framework for the first two decades 1644-1663, due to the lack of data for many provinces. At the dawn of the dynasty, when its grip on the Chinese territory was still uncertain, lingchi sentences were concentrated in the centers of power : Zhili province, the metropolitan area centered on Beijing ; and Manchuria, that is in fact Shengjing – Moukden under its Manchu name – which was the original capital of the Qing and kept after the conquest of China as part of its institutional and jurisdictional functions to the exclusive usage of the Manchus. The high rate of Gansu deals either with Muslim resilience that was to influence the province in the following centuries, or more likely with the gradual involvement of the dynasty in Mongolian rule in the northern border. The latter cannot be confirmed, since we have no figures for Mongolia proper. At this first stage, the high rates of lingchi sentences are seen in the Northern region that was directly under the rule of the conquerors, while central China reveals lower rates ; for the South and West there are no data. As will be shown further on the nature of the crimes punished with lingchi, there is no indication of rebellious attitudes of these Northern regions, since “parricide”, not “revolt”, was the most targeted crime.

50The map for the two decades, 1728-1747 shows a very contrasted picture, with high rates of lingchi sentences in the South-Western regions (Yunnan, Sichuan, Guizhou) singularized by their many ‘ethnic minorities’ and a strong Muslim component in the population. As the “nature of the crime” will show below, these are rebellious regions already at this time – the years of the end of Yongzheng and the start of Qianlong rules – and this is confirmed in our second time-unit : the map for 1770-1789 shows high rates of lingchi sentences in Yunnan-Guangxi-Sichuan, and also the adjacent Hubei and Hunan, all regions where the White Lotus rebellion originated, which drained armies and finances of the late Qianlong and early Jiaqing reigns. On the South-Eastern coast, Fujian province and its dependence of Taiwan, reveal high rates, mostly explained by feuds and banditry. But the most striking are the high incidents in the ‘non-Han’ Northern and Western vast crescent comprising Manchuria, Mongolia, and Xinjiang. The grim honor of being the areas that saw the most incidents of lingchi goes to these three areas under Daoguang (1821-1851), and Guangxu, up to lingchi’s abolition in 1905. The blind spot in the middle of the century prevented us from verifying whether the Taiping, Nian and other rebellions augmented the rates of lingchi sentences in the regions which were most affected. One notes, however, that regions where the Taiping originated, like Guangxi, and the regions where they spread, like Jiangsu, Anhui Zhejiang, and the regions where the rebels fled and tried to entrench, like Sichuan and Guizhou, saw higher incidents of lingchi. The high rates in Guizhou are partly explained by the recurrent revolts of the Miao. The clearest case of all is Xinjiang’s, the Chinese Turkestan with its original populace of Uyghurs, prone to rebellion and separatism such as the independent khanate with Russian support. Here, as we will see, lingchi sentences have been a tool for repressing and deterring separatism.

51The impact of rebellions on the occurrences of lingchi sentences would need further research. However, the category of “revolts” makes for less than one third of the total number for the whole period, and much less still during the last decades that followed the great rebellions. If insurgent spirit and deeds were targeted in places like Xinjiang, Guizhou or Yunnan, this is but one aspect of a more complex picture, as shown in our last category of documents : maps indicating the nature of crimes that were predominant in the lingchi sentence pronounced in each province (see maps f, g, h).

Localizing crimes targeted by lingchi sentences

52The last three maps all show a sharp contrast between two categories : 1. Provinces where “parricides” are overwhelmingly prevalent, with at least 40% of the sentences ; 2. Province where “revolts” are the most targeted crimes, with at least 40% of the sentences. For the whole period (map f), category 1 is dominant in all the Northeastern region from Manchuria, Mongolia, and Zhili, the metropolitan province, to the lower Yangzi provinces of Jiangsu and Zhejiang, throughout the central provinces, like Anhui, Hubei, down to the Hunan and West to Sichuan. The second category, where “revolts” is the main course of lingchi sentences forms a peripheral arc from the Southeastern coastal provinces of Fujian and Guangdong, up to the North-western Xinjiang, including the ethnically instable Yunnan, Guizhou, and Xizang, that is Tibet. Close to Xinjiang, Gansu has an even greater proportion of “revolts” than expected for a province where an important Muslim minority periodically revolted well before initiating the great Hui rebellion of the 1860s. The adjacent Shaanxi presents a similar picture.

53Thus a contrast appears between a Southern and Western ‘outer’ China on the one hand, where lingchi was mainly a tool for repressing or deterring rebellions and banditry, and a Northern, Eastern and Central China on the other hand, where “parricides” were the main target. These two Chinas are highly unequal in population and density given that the Chinese population is concentrated in the Eastern Coastal provinces and in some central provinces like Hubei or Sichuan, while the peripheral areas are sparsely populated. The prevalence of “parricides” over “revolts” in our data reflects this difference in demographic weight, which should not lead us to overlook the importance of the spatial dimension. If “revolts” amounted to less than one third of the lingchi sentences for the whole empire, they mostly occurred in the vast border areas with mixed and unruly populations, which form buffer marches of great geo-strategic importance for the Empire. In such regions, lingchi was undoubtedly the emblematic manifestation of imperial supremacy, or the tool for its restoration, in other words the State terror device that the term conveys in the Western literature.

54For most of the provinces of ‘inner China’, however, lingchi was imposed not to affirm imperial supremacy or restore public order, but to restore local and familial order. As for familicides, their general progression, as mentioned above, increased from around 7% to nearly doubling in the last decades of the Qing, as well as their permanence in local life. Geographically, lingchi sentences for familicides are in greater proportion in Northern and Central China : around 45% in Manchuria, 50% in Anhui, 30% in Shanxi, over 20% in Zhili. In the unruly Yunnan province, lingchi sentences are equally shared between parricides, revolts, and familicides. There are stunning bursts of “atrocities”, in Zhejiang or Fujian, meaning that local conflicts lead to inhumane conducts, such as the maiming of corpses or dismembering alive, with statistically significant frequency. Given that parricides were already a major cause of lingchi from the beginning of the dynasty, the progression of familicides appears as a major change, which seems to complement the decline in revolts and banditry. Under the Qing, and particularly in the last decades, lingchi was increasingly employed to deter dominant lineages or local bullies from vanquishing their opponents by exterminating them with their kin, and thus maintain at least the simulacra of peace and order at the local level.

55The data do not allow us to go further in the geographical analysis. As limited as our survey is at the present stage, it hints at a penal typology of provinces, which would help restitute its spatial dimension to the legal system.

Some tentative comparisons with Europe

56How should we assess the dismemberment of 1140 people over 250 years, with a frequency of around five incidents per year ? Do these numbers confirm that China lavishly resorted to ‘cruel punishments’, as Westerners would assert increasingly vociferously from the mid-19th century ? How does the Chinese practice of execution compare with the equivalents of the Western countries in the same period ?

57The comparison is not easy. The first difficulty is that reliable and regular data are hard to find prior to the mid-19th century except on a very small scale : covering one city or town, or at best a whole province. Since the judicial systems of Europe were not centralized to the same degree as the Chinese, historians have to piece together scattered information to reach general estimation on the execution rate in England or France. The second difficulty deals with the quality, rather than the quantity, of the data : the type of execution and the categories of targeted crimes in the West do not coincide with the Chinese ones. For instance, it has been calculated that around 35,000 death sentences were pronounced in England and Wales between 1751 and 1800, of which about 7,000 were actually executed. These figures cannot be compared directly with our 1,140 lingchi cases, since the latter number refers only to the particular form of penalties, which occurred much less frequently than ‘ordinary’ capital punishments. A sound comparison could be made only with a general account of all the capital sentences and executions in China, a figure that would be considerably higher than our present data. Producing such statistics is perfectly feasible through a systematic survey of the judicial archives kept in Beijing and Taipei, but it would require much more time and manpower than we have been able to summon until now.

  • 31 See the famous description of Damiens’ écartèlement in Foucault (1977) and its critique in Friedlan (...)

58If comparing lingchi sentences with the general statistics of death penalties in Europe is not feasible, can we attempt a comparison between lingchi and the forms of particularly cruel executions in Europe ? Indeed, it is tempting to compare lingchi to the French écartèlement, the emblematic tortured agony used against those convicted of lèse-majesté that Foucault famously – and wrongly – presented as the epitome of death penalty in pre-modern Europe, or to the English quartering, which was reserved to public enemies guilty of great subversion or high treason.31 The problem here is that these types of penalties were extremely rare, with no more than one or two dozen incidents found during the 17th and 18th centuries in Western Europe. As a penalty for lèse-majesté, parricide, blood feuds, individual acts of atrocity as well as revolt or armed banditry, lingchi is yet not comparable with the harshest and most spectacular punishments of Europe.

  • 32 Garnot (2009, pp. 479-480).
  • 33 According to Pieter Spierenburg (1984, p. 213), there were 40 executions by breaking on the wheel i (...)

59In fact, there is one specific type of punishment in Europe that allows comparisons with lingchi : the wheel. Under the two different forms it was practiced in Germany – to be broken alive with a heavy cartwheel – and in France – to be broken alive with a bar while tied on the wheel – this punishment exhibited many similarities with lingchi. The punishment targeted the most heinous criminals, such as bandit ringleaders, serial murderers, parricides, and other culprits of atrocious acts. Moreover, the ‘breaking’ and ‘slicing’ process were very comparable in their operations and significances : the methodical destruction of the criminal’s body was aimed at symbolizing the eradication of the evil and at inspiring awe and fear to the spectators. We have no statistical data on wheel executions, equivalent to our lingchi data. Yet a recent work of synthesis provides certain elements of comparison. Benoît Garnot, an outstanding historian of justice in France, has gathered together figures from different jurisdictions to reach a conclusion of “an average of two hundreds death penalty per year for the whole kingdom of France (0.74 per 100,000 inhabitants)” during the second half of the 18th century. It is a figure that must be decreased by “at least one third”, to account for pardons, contumacies, etc. So we are left with a general estimation of around 140 criminals executed per year in mid-18th century France.32 So much for death penalties of all kinds. Garnot has also estimated the proportion of tormented executions, in particular “the breaking alive on the wheel”. In 18th century Burgundy, the “wheel is decided for one quarter of all the condemned”, and this rate increased as time went by ; “at the end of the reign of Louis XV, the inflation of the tormented death penalties (l’inflation des supplices) seems to be the rule here and there, to be decreased later”. In Provence, for instance, the “breaking on the wheel” rose to 53% between 1761 and 1770, with a peak of 72% between 1781 and 1790. On the basis of Garnot’s calculation, one can infer that “breaking on the wheels” represented at least one quarter, more likely one third, of the total number of executions in the second half of the 18th century. Therefore, it is easy to deduct that, among the 140 criminals who were executed each year over the second half of 18th century France, between 35 and 45 were “broken on the wheel”.33 If these figures are compared with the 60 years of the Qianlong reign (1736-1796), it would result in more than 2,000 deaths by the wheel, almost twice the total amount of lingchi we have found for the whole Qing period. Of course, these approximate figures should not be extrapolated. But they suggest that the use of lingchi was not more frequent than the use of the wheel, in an empire much bigger in size and more populous than France. Interpreted strictly as legal repression and deterrence of the most heinous crimes, the reign of Qianlong may not have been any more ‘cruel’ than the reigns of Louis XV or Louis XVI ; but of course, all this depends on how far the present sample is representative of the lingchi sentences executed under the Qing. This can only be confirmed or disproved by further research.


  • 34 On the etymological and legal meanings of lingchi, see Brook et al. (2008, chapter 3) ; Bourgon (20 (...)
  • 35 Striking cases of ‘familial crimes’ have been collected and aptly commented by Lauwaert (1999).

60Lingchi was from its origin an anomaly in Chinese law : belatedly and imperfectly codified, loosely and expeditiously handled, excluded from revision procedure. It was applied to criminals for their presumed crimes and attempted crimes without consideration of their intent, in complete contradiction with the sound principles of the regular legal system. An enclave of extraordinary law from the outset, lingchi had been shaped for the defense of the Confucian “Three bonds” (sangang), in which the two last bond – the father-son and the husband-wife relationships – are modeled on the first, the sovereign-subject bond. For centuries, the assimilation of regicide and parricide that authoritative commentators read into the Annals, one of the Five Classic of the Confucian tradition, was mainly a doctrinal justification of political power. Those who challenged the dynasty or the public order by rebellion or treason committed an attempt of regicide-parricide, and thus were doomed to lingchi and the extermination of their clan.34 An anomaly from a legal point of view, lingchi conformed to a logic that recommended using mass terror against public enemies. The primary interest of our data is to show the dramatic shift of a mass terror device from the public to the private sphere. Under the Qing, less than one third of the lingchi sentences targeted crimes threatening State power while more than half targeted crimes committed within the family, with 15 to 40%, depending on regions, aimed at limiting the extermination of rival families.35 In the last decades of the empire, lingchi was still used as deterrence against rebels in remote border areas, but in most of the Chinese provinces it targeted chiefly rebellious children or wives, as well as local bullies prone to kill any opponent together with his family.

61Even at this initial stage, our investigation in penal geography promises an important outcome in charting out for the spatial dimension of law. The Qing had a well-developed legal system, highly systematic in labeling crimes and apportioning punishments ; but the system could live and grow only by its interrelation with local realities through a complex institutional network that is most aptly seized when restored in its spatial dimension. The Qing code provides us with only a list of crimes liable to lingchi, but in order to know what kind of crimes were effectively targeted by lingchi, we have to delve into the archival cases, and into their spatial repartition and evolution through time. Beyond the sheer figures about harsh punishments, and the comparisons of the level of cruelty between China and Europe that these data seem to reveal, a key interest of such an enquiry is to map the penal agenda of the State throughout its territory. The present data and the maps we drew from them as well as the associated analyses are intended as an invitation to further and more ambitious research into the central archives in Beijing and Taipei, as well as in local archives that become more accessible each year.

  • 36 Anonymous (n.d., ca. 1900).

62Figures are cold abstraction, however, which need to be fleshed out by this illustration of the subject most targeted by the harshest penalty, the arch-enemy in the late Qing : the wife who plotted the killing of her husband with the help, or under instigation, of her lover. Here she is, as shown in an illustrated handbook on the “Five Punishments” that circulated in the 1900s (see appendix 3), logically subjected to lingchi. This beautiful, desirable young woman languidly contemplates her severed limbs, while her lover is waiting to be beheaded at the background. In all its suggestive crudity, this illustration comes, not from a popular titillating novel, but from a stern legal handbook.36 For a reader trained by reading Bodde & Morris, and so many authors, to face violent rebels and bandits, it comes as a disconcerting dismay. One feels like a moviegoer who expected to watch Spartacus and is shown The Postman Always Rings Twice instead. The dismay is instructive, however. The fact that instead of the Great Traitor – Chiyou, the Archetypal Rebel who was defeated and dismembered by the Founding Yellow Emperor in the great myth on the origins of the Chinese civilization – we find a fragile, though unfaithful and murderous wife is like a parody of the Confucian great project of total governance. It is as though the majestic progression schemed in the Great Study : Governing the Self, Ruling the Family, Pacifying the World, had shrunk back, as though the chaos that had shaken the State to its foundation had gained the very core of the family, the spousal couple, that was indeed to get through some turbulences in the century to come. This is the threat that lingchi sentences were supposed to conjure, a threat against the dogmatic foundation of the Confucian order, much more serious than social upheavals or political protests. This is, however, where Chinese Law originated its raison d’être and dynamics and where its meaning and progress must be sought.

Haut de page


Anonymous, Jinshan xian baojia zhangcheng : Fu quanjie tiaokuan, Zuiming tushuo [Commented charts of Crimes and Punishments, appended as articles of exhortations and warnings to a Regulation on the local police of Jinshan district], ca.1900.

Bastien, P., Une histoire de la peine de mort. Bourreaux et supplices 1500-1800, Paris, Seuil, 2011.

Bodde, D., Morris, C., Law in Imperial China : Exemplified by 190 Ch’ing Dynasty Cases, Philadelphia, University of Pennsylvania Press, 1973 [orig. publ.1968].

Bourgon, J., Abolishing “Cruel Punishments” : A reappraisal of the Chinese Roots and Long-Term Efficiency of the Xinzheng Legal Reforms, Modern Asian Studies, 2003, 37, 4, pp. 851-862.

Bourgon, J., Supplice chinois, Bruxelles, La Maison d’à côté, 2007.

Bourgon, J., Obscene Vignettes of truth. Construing photographs of Chinese executions as historical documents, in Henriot, C., Yeh, W. (Eds), Visualizing China. Moving and Still Images in Historical narratives, Leiden, 2012a, pp. 39-91.

Bourgon, J., Lapsus de Laïus. Entre régicide et parricide, l’introuvable meurtre du père, Extrême-Orient, Extrême Occident, 2012b, 1, pp. 313-339.

Brook, T., Bourgon, J., Blue, G., Death by a Thousand Cuts, Cambridge MA., Harvard University Press, 2008.

Buoye, T., Manslaughter, Markets, and Moral Economy : Violent Disputes over Property Rights in Eighteenth-Century China, Cambridge, CUP, 2000.

Cao S., Zhongguo renkou shi – di wu juan : Qing shiqi [A History of the Chinese population : 5. The Qing Period], Shanghai, Fudan daxue chubanshe, 2001.

Carlitz, K., Genre and Justice in Late Qing China : Wu Woyao’s Strange Case of Nine Murders and its Antecedents, in Hegel, R.E., Carlitz, K. (Eds), Writing and Law in Late Imperial China ? Crime Conflict, and Judgment, Seattle, University of Washington Press, 2007, pp. 234-257.

Chaunu, P., La pesée globale en histoire, Cahiers Vilfredo Pareto, 1968, 6, 15, pp. 135-164.

Evans, R., Rituals of Retribution. Capital punishment in Germany 1600-1987, Oxford, OUP, 1996.

Foucault, M., Discipline and Punish. The Birth of Prison, New York, Vintage, 1977.

Friedland, P., Seeing Justice Done. The Age of Spectacular Capital Punishment in France, New York, Oxford University Press, 2012.

Garnot, B., Histoire de la justice. France, XVIe-XXIe siècles, Paris, Gallimard, 2009.

Gatrell, V.A.C., The Hanging Tree : Execution and the English People, 1770-1868, Oxford, OUP, 1994.

Hevia, J., English Lessons : The Pedagogy of Imperialism in Nineteenth-Century China, Durham, NC., Duke University Press & Hong Kong University Press, 2003.

Lauwaert, F., Le meurtre en famille. Parricide et infanticide en Chine (XVIIIe-XIXe siècles), Paris, Odile Jacob, 1999.

Lee, J., Homicide et peine capitale en Chine à la fin de l’empire. Analyse statistique préliminaire des données, Études chinoises, 1991, X, 1-2, pp. 114-134.

Meadows, T.T., The Chinese and Their Rebellions, Viewed in Connection with Their National Philosophy, Ethics, Legislation, and Administration, London, Smith, Elder & co., 1856.

Rowe, W.T., Saving the World. Chen Hongmou and Elite Consciousness in Eighteenth-Century China, Stanford, Stanford University Press, 2001.

Spence, J., Death of Woman Wang, London, Penguin, 1979.

Spierenburg, P., The Spectacle of Suffering : Executions and the Evolution of Repression : From Pre-industrial Metropolis to European Experience, Cambridge, CUP, 1984.

Suzuki, H., Gong Qing Wang Ming— Kyôsei ômei kô : Shindai shikei saiban ni okeru ‘kengi’ to ‘teirei’ [Gong Qing Wang Ming : A study of capital sentence under the Qing, from “expedients” to “established rules”], Hôseishi kenkyû, 2003, 53, pp. 47-80.

Zhao W., Xie S., Zhongguo renkou shi [A history of the Chinese population], Beijing, Renmin chuban she, 1984.

Haut de page


1. Sources

a. The “archives of the Grand Secretariat” (neige daku dang’an) include a collection the memorials sent by provincial authorities to the Central boards for revision and transmission to the emperor for all kinds of administrative affairs, among which were capital cases. These archives are kept in Taipei, in the “Old Palace” (Gugong) Museum, where parts of the documents that were primarily in Beijing have been moved. Their digitalization provides us with 357 persons dismembered in 338 executions between 1644 and 1854.

b. Judicial reviews of the Shenbao (or Shanghai News). The Shenbao was the first modern news journal in Chinese daily published in Shanghai from 1872 up to 1949, with interruptions due to civil war and Japanese aggression. Right from the first year onward, the journal systematically reproduced abstracts of the ‘Beijing gazette’ (jingbao), an unofficial copy of official documents – mostly legal and judicial– issued by the central boards in Beijing. Until very recently, researchers had to use a facsimile of the original format, whose general layout made its reading quite laborious and a statistical surveys almost impossible. Its recent digitalization and online access has permitted to find accounts on 453 people executed in 453 executions between September 1872 and April 1905.

c. The “illicit sexual intercourse” were a part of the “Routine memorials of the Penal service of the Censorate” (xingke tiben). The sentence transmitted to the Board of Punishment were routinely reviewed and archived by a service of the Censorate specialized in penal affairs. This bureaucratic process has produced millions of documents now stored in the Archive n°1 at Beijing’s Forbidden city. In recent years, the Institute of History and Philology of the Academia Sinica in Taiwan has copied important samples and made them available online.

d. Leading cases from the ‘Board of Punishments’ (Xingbu) part of the ‘Official administrative collected compendium with cases and rules of the Great Qing’, (Qinding Da Qing huidian shili) covering the Guangxu reign (publ. 1899). This was a huge collection of all texts that constituted the central services at the capital. The part devoted to the ‘Board of Punishments’ included the Qing code, and a selection of precedents or shili that were likely to be inserted into the code as sub-statutes or tiaoli, and thus become new laws enforced on a particular matter.

e. The ‘Comprehensive compendium of penal cases’ (Xing’an huilan) is a huge collection of judicial materials, including cases adjudicated at the local level and transmitted to the ‘Board of Punishment’ by the provincial judiciaries as well as memoranda exchanged between services of the Board. The original collection in 60 volumes covers around ten years from 1823 to 1834, and has been followed by many sequels and reprints up to the 1890s. Our cases are abstracted from a PDF version of the recent edition by the Chinese Academy of Social Sciences Institute of Legal Studies.

f. The ‘Veracious registers’ (Shilu) is an official day-by-day record of the decisions taken by the Qing emperors in connection with their ministers and subordinate services. There is one ‘Veracious register’ per reign, giving a total of ten for the Qing, all available in PDF and searchable with OCR devices. Death sentences reviewed by the emperor were cursorily registered therein.

2. Statistics and diagrams

2.a. Yearly frequency of lingchi per reign, with account of “blank years


Lingchi sentences

Yearly average

Year count with “blank years”

Shunzhi (1644-1661)




Kangxi (1662-1722)




Kangxi blank year deducted


40 years deducted from 61

Yongzheng (1723-1736)




Qianlong (1737-1796)




Jiaqing (1797-1820)




Daoguang (1821-1851)




Xianfeng (1852-1861)



5 years deducted from 9

Tongzhi (1862-1874)



8 years deducted from 12

Guangxu (1875-1908)




Year unknown



Whole period (1644-1905)




Whole period minus 65 “blank years”




2.b. General trend of lingchi sentencing 1644-1905

2.b. General trend of lingchi sentencing 1644-1905

2.c. Four categories of lingchi sentences for eight Qing emperors reigns

2.c. Four categories of lingchi sentences for eight Qing emperors reigns

2.d. Share of lingchi sentences for each category of crimes (whole period)




share (%)











Parricides (total)




Nature of parricides

Pv (viricide)




Pse (senior-elder)




Pm (mother)




Pf (father)




P (unknown)




Phh (household head)




Pc (combined) :



P (father and mother)




P (father and household head)




P and revolt




P (viricide and senior-elder)




P (viricide and senior-elder and familicide)




P (viricide and father)







2.e. Nature of the parricides

Nature of parricides for the whole period (1644-1905)

Nature of parricides for the whole period (1644-1905)

Nature of parricides : Shunzhi (1644-1662)

Nature of parricides : Shunzhi (1644-1662)

Nature of parricides : Qianlong(1736-96)

Nature of parricides : Qianlong(1736-96)

Nature of parricides : Guangxu (1875-1905)

Nature of parricides : Guangxu (1875-1905)

3. Maps

3.a. Rate of lingchi sentences by province for 1644-1662

3.a. Rate of lingchi sentences by province for 1644-1662

3.b. Rate of lingchi sentences by province in 1728-1747

3.b. Rate of lingchi sentences by province in 1728-1747

3.c. Rate of lingchi sentences by province, 1770-1789

3.c. Rate of lingchi sentences by province, 1770-1789

3.d. Rate of lingchi sentences by province 1821-1850.

3.d. Rate of lingchi sentences by province 1821-1850.

3.e. Rate of lingchi sentences by province 1875-1905

3.e. Rate of lingchi sentences by province 1875-1905

3.f. Nature of the crimes leading to lingchi sentences in each province for the whole period (1644-1905).

3.f. Nature of the crimes leading to lingchi sentences in each province for the whole period (1644-1905).

3.g. Nature of the crimes leading to lingchi sentences in each province under Qianlong (1736-1796).

3.g. Nature of the crimes leading to lingchi sentences in each province under Qianlong (1736-1796).

3.h. Nature of the crimes leading to lingchi sentences in each province under Guangxu (1875-1905).

3.h. Nature of the crimes leading to lingchi sentences in each province under Guangxu (1875-1905).
Haut de page


2 Lee (1991).

3 Lee (1991, p. 124).

4 Buoye (2000).

5 Lee (1991, pp. 116-117), Buoye (2000, pp. 238-239).

6 Chaunu (1968).

7 On this swift procedure of execution, see Suzuki (2003).

8 Bodde, Morris, (1973).

9 Bodde, Morris (1973, p. 112, n.51).

10 Meadows (1856, pp. 655-656). Quoted in Bodde, Morris (1973).

11 Bodde, Morris (1973, p. 111) – (my italics).

12 For reflections on this silence, see Bourgon (2007) ; Brook, Bourgon, Blue (2008, chapter 1). The short eyewitness report of Wang Weiqin’s execution by the hydraulic engineer Li Yizhi evoked in this chapter does not make a real exception, since Li depicts the arrival of the condemned, the place and the crowd, but has not a word about the actual execution.

13 The legal term in such case was not “putting to death by dismemberment” (lingchi chusi), but “slaughtering the corpse” (lushi), a nuance that we can ignore for the present purpose.

14 On the abolition decrees of 1905, see Bourgon (2003, pp. 851-862).

15 Law against Robbery or banditry (“Qiangdao”, art.261 in the Qing code) did not provide for lingchi sentences, but only decapitation with immediate execution (zhan lijue), with possible exposure of the head (xiaoshou) for the ringleader. However, a gang of a certain magnitude in number and criminal records could be assimilated to a rebellious scheme, just like dissident groups for political or religious purposes. Our focus on State conceptions and practices justifies the grouping under the single category of “Revolt” of what could be termed “crimes against the public order”. They generally fall under art.254 : “Plot of Rebellion and Great Sedition” (Moufan dani), and art.255 : “Plot of High Treason” (Moupan).

16 Art.287 : “Killing three persons in a same family” (Sha yijia sanren). The term “familicide” is currently used for crimes committed by a spouse, commonly the husband, killing the other spouse with the children. But the term also includes the killing of a family by strangers, which is common for our Chinese cases.

17 Art.284 : “Plot to kill grand-parents or parents” (Mousha zu fumu fumu) ; art.286 : “Plot to kill the parents of one’s deceased husband” ; art.314 : “Slave beating the household head” (Nubi ou jiazhang) ; art.315 : “Spouse or Concubine beating their husband” (Qie qi ou fu).

18 Art.287, last paragraph ; and art.288 : “Picking up parts of the body of a living person” (Zhaisheng zhege ren).

19 This old classic may seem outdated in the Qing, but its meaning had been revived under the Ming, and placed at the core of the statecraft current by Qiu Jun’s ‘Complement to the extended meaning of the Great Study’ (Daxue yanyi bu), which had a deep influence on Qing statecraft activists, such as Chen Hongmou. See Rowe (2001, p. 117).

20 Spence (1979, pp. 89-98).

21 See Carlitz (2007, pp. 234-257).

22 A sketch of the Wang Weiqin case appears in the introduction of Brook, Bourgon, Blue (2007).

23 For the wives and children of the condemned ‘being confiscated’ meant to be deported and reduced to slavery.

24 Neige daku, Archives of the Grand Secretariat, ‘Basic material’ (jiben ziliao) 065912-001.

25 On the Boxers’ execution, see Hevia (2003) ; and Bourgon (2012a).

26 See Bourgon (2012b).

27 Women represent less than 10% of the executed in 18th century London, according to Gatrell (1994, p. 8) ; Richard Evans (1996), give similar rates for Germany, apart from bouts of repressive fury against witches, and later against infanticide mothers in the mid-18th century.

28 This squares with James Lee’s observations that “le seul ‘supérieur’ contre lequel se dressaient massivement les femmes était leur mari”, see Lee (1991, p. 127).

29 Zhao, Xie (1984) ; Cao (2001).

30 Here we took example on recent studies of executions in modern and pre-modern Europe : Gatrell (1994) ; Evans (1996) ; Bastien (2011).

31 See the famous description of Damiens’ écartèlement in Foucault (1977) and its critique in Friedland (2012, p. 2013).

32 Garnot (2009, pp. 479-480).

33 According to Pieter Spierenburg (1984, p. 213), there were 40 executions by breaking on the wheel in Amsterdam between 1651 and 1750, i.e around 0,4 per year. Estimating the average population to around 200,000, this makes a rate twice higher than in France, and much higher than the rate of lingchi in China according to our data.

34 On the etymological and legal meanings of lingchi, see Brook et al. (2008, chapter 3) ; Bourgon (2012b).

35 Striking cases of ‘familial crimes’ have been collected and aptly commented by Lauwaert (1999).

36 Anonymous (n.d., ca. 1900).

Haut de page

Table des illustrations

Fichier image/jpeg, 1,1M
Titre 2.b. General trend of lingchi sentencing 1644-1905
Fichier image/jpeg, 164k
Titre 2.c. Four categories of lingchi sentences for eight Qing emperors reigns
Fichier image/jpeg, 456k
Titre Nature of parricides for the whole period (1644-1905)
Fichier image/jpeg, 52k
Titre Nature of parricides : Shunzhi (1644-1662)
Fichier image/jpeg, 56k
Titre Nature of parricides : Qianlong(1736-96)
Fichier image/jpeg, 52k
Titre Nature of parricides : Guangxu (1875-1905)
Fichier image/jpeg, 56k
Titre 3.a. Rate of lingchi sentences by province for 1644-1662
Fichier image/jpeg, 116k
Titre 3.b. Rate of lingchi sentences by province in 1728-1747
Fichier image/jpeg, 120k
Titre 3.c. Rate of lingchi sentences by province, 1770-1789
Fichier image/jpeg, 112k
Titre 3.d. Rate of lingchi sentences by province 1821-1850.
Fichier image/jpeg, 112k
Titre 3.e. Rate of lingchi sentences by province 1875-1905
Fichier image/jpeg, 120k
Titre 3.f. Nature of the crimes leading to lingchi sentences in each province for the whole period (1644-1905).
Fichier image/jpeg, 120k
Titre 3.g. Nature of the crimes leading to lingchi sentences in each province under Qianlong (1736-1796).
Fichier image/jpeg, 116k
Titre 3.h. Nature of the crimes leading to lingchi sentences in each province under Guangxu (1875-1905).
Fichier image/jpeg, 114k
Haut de page

Pour citer cet article

Référence papier

Jérôme Bourgon et Julie Erismann, « Figures of Deterrence in Late Imperial China. Frequency, Spatial Repartition, and Types of Crimes Targeted by Dismemberment under the Qing Dynasty », Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 18, n°2 | 2014, 49-84.

Référence électronique

Jérôme Bourgon et Julie Erismann, « Figures of Deterrence in Late Imperial China. Frequency, Spatial Repartition, and Types of Crimes Targeted by Dismemberment under the Qing Dynasty », Crime, Histoire & Sociétés / Crime, History & Societies [En ligne], Vol. 18, n°2 | 2014, mis en ligne le 01 octobre 2017, consulté le 20 novembre 2017. URL : ; DOI : 10.4000/chs.1487

Haut de page


Jérôme Bourgon

After studying and passing grades in history departments, Jérôme Bourgon specialized in Chinese history. In 1994, he defended his Ph.D thesis in ‘History and Civilization’ at the EHESS, with a study of Shen Jiaben, who introduced Western law in late imperial China and was the last jurist and historian of the Chinese legal tradition. Hereafter, Bourgon researched various aspects of Chinese law, such as the questionable relationship between ‘customs’ and civil law, the various genres of legal books in imperial China, and penal law as seen through the ominous ‘Supplice chinois’ []. He published two books on this topic. Since 2011, he leads an international project aimed at a complete and accurate translation of the Qing dynasty penal code, as well as a new understanding of law through its differentiated implementation on the various parts of the empire territory, hence its title “Legalizing Space in China” []. He is currently writing a general history of penal law in China, in which this study on lingchi takes place.

Julie Erismann

Julie Erismann has a doctorate in geography (Univ. Jean-Moulin, Lyon, France) – She is responsible for the statistics, diagrams and maps in this article.

Haut de page

Droits d’auteur

© Droz

Haut de page