Introduction
There are many methods human beings use to kill each other, some of which are more socially tolerated than others. Beheading a person, however, is a way of killing another human being which in many cultures bears particularly strong associations with the most violent exercises of authority, barbarism or brutality.Footnote 1 The importance of a cultural context which assigns powerful meanings (and traditionally Chinese people did display great prejudice against dismemberment) to beheading results in this often being regarded as an exceptionally horrible form of death; other considerations – such as the fact that beheading may be less painful than other ways to die, or that the dismemberment may have been inflicted after life was already extinct – fade into insignificance.Footnote 2 Headhunting is here understood as a specific form of killing that involves not merely severing the head from the body, but also significant ritual attention to the head after it has been removed (Hoskins Reference Hoskins and Hoskins1996a: 12–4). Taking a head is an action loaded with social and religious meaning, and it is often governed by very strict rules as to who can be killed, and how the whole process should be sanctioned by conducting appropriate ceremonies before and after the killing.Footnote 3 However, since in many cultures an association is made between headhunting and barbarism, societies which practise headhunting, and which naturally do not denigrate themselves in this way, often redefine the nature of the activity in which they are engaged.Footnote 4 Indeed, as will be demonstrated below, ancient Chinese accounts of headhunting were significantly modified from the Eastern Han dynasty onwards, with a view to obscuring this part of their history, and this is likely to have occurred because such practices had come to be seen as barbaric and uncivilized.
The importance of maintaining a careful distinction between killing people by cutting their heads off and headhunting can be seen from the fact that there were many reasons in ancient China why one person would have wished to cut the head off another, and some were entirely unrelated to any ritual purpose. One of the most important reasons that heads were taken was to document success in battle: military rewards were calculated on the basis of the number of severed heads an individual warrior presented to the authorities. This kind of accounting is recorded in the “Jingnei” 境内 (Within the Borders) chapter of the Shangjun shu 商君書 (Book of Lord Shang), a Warring States era text associated with the historical figure of Shang Yang 商鞅 (d. 338 bce), in which the author proposed that social status be made dependent on the number of enemy combatants killed in battle by each individual – a practice which was institutionalized in the kingdom of Qin, and subsequently introduced across the empire after the unification of China:
Any man who can take one [enemy] soldier's head, should be rewarded with a promotion of one grade in rank, and receive an additional one qing of fields, and further nine mu of land attached to their residence (Jiang Reference Jiang1986: 119).Footnote 5
能得甲首一者, 賞爵一級, 益田一頃, 益宅九畝.
According to both historical and legal texts, during the Qin dynasty, soldiers were indeed rewarded with land and rank in the event that they killed an enemy in battle and took his head (Yunmeng Shuihudi Qinmu bianxiezu 1981: plate 69 [“Junjue lü” 軍爵律]; Hulsewé Reference Hulsewé1985: 82). As described in the Hanshu 漢書 (History of the Han Dynasty), this system continued to apply, with a twenty-level system of ranks – the orders of honour – being used by the government to reward those who had performed meritorious service to the state, in particular killing enemy combatants (Ban Reference Ban1962: 19A.739–40; Zhangjiashan ersiqihao Hanmu zhujian zhengli xiaozu 2006: 51–2 [“Hulü” 戶律]; Gao Reference Gao1982: 33–57; and Yang Reference Yang2003). Although this must have led to a massive increase in the numbers of heads being severed on the battlefield, and potentially significantly increased the chances of decapitation among any unfortunate people who happened to be in the vicinity, this is not headhunting in the strict sense of the word.Footnote 6 There is no suggestion that these heads served any purpose once they had been entered into the official ledgers that would allow the soldiers concerned to receive their rewards, and it is likely that once documented, the decapitated heads were simply buried with other remains.Footnote 7 Similarly, beheading is known to have been used as a punishment in early China – criminals who were convicted of serious crimes were executed in this way.Footnote 8 (The simple fact of beheading should not be considered headhunting, though in the event of a severed head being ceremonially exposed, sacrificed on an altar, or experiencing some other kind of ritual appropriation, penal beheadings may be considered to fall within the spectrum of headhunting). However, in addition to military and penal beheadings, there is a large body of literature recording headhunting in ancient Chinese, dating back to at least the time of the Shang dynasty, not to mention suggestive archaeological discoveries, which serves to demonstrate the importance of this practice.
Headhunting in Huaxia culture
The early history of headhunting in China is extremely obscure, not least because there has often been considerable resistance among academics to the idea that the ancient inhabitants of the Huaxia cultural region could have engaged in practices typically considered as “barbaric”.Footnote 9 Hence, in spite of the fact that heads and headless bodies have been discovered in considerable numbers in excavations at the Shang dynasty capital city of Anyang, great disquiet has been expressed by some scholars at the idea this could represent human sacrifice or headhunting (Zhang Reference Zhang1988: 196). Nevertheless, the fact remains that the earliest textual evidence of the taking of human heads for use in sacrifice can be found in inscriptions excavated at Anyang.Footnote 10 One important group of such texts consists of inscribed human skull bones, of which to date fourteen examples have been discovered. The best preserved of these skulls incorporates an inscription recording the sacrifice of Yi Fangbo 夷方伯 – a regional lord – to the deceased Shang king Zuyi 祖乙 (seventh generation monarch), and it is assumed that the skull on which this text was inscribed was that of the unfortunate Yi Fangbo himself (Guo and Zhongguo shehui kexueyuan lishi yanjiusuo Reference Guo1982: 12.4808 [#38758]; and Huang Reference Huang2005). There are also a number of oracle bone inscriptions which record divinations as to whether a particular head was acceptable for sacrifice; for example, one fragmentary inscription reads: “…Can the skull [taken in] Wei territory be used in sacrifice to Female Ancestors Geng and Wang Bin?” (… yong Weifang xin yu Bi Geng Wang Bin? … 用危方囟於妣庚王賓?) (Guo and Zhongguo shehui kexueyuan lishi yanjiusuo Reference Guo1982: 9.3463 [#28092]). Another inscription on the same subject reads: “Were the skull [taken in] Qiang territory to be used in sacrifice, will the king receive divine protection thereby? / If we were to use [the skull] taken in Qiang territory in sacrifice at the ancestral temple, will the king receive divine protection thereby? (Qiangfang xin qi yong, wang shou you you? / Qi yong Qiangfang [xin] yu zong, wang shou you you? 羌方囟其用, 王受有佑? / 其用羌方【囟】於宗, 王受有佑?) (Guo and Zhongguo shehui kexueyuan lishi yanjiusuo Reference Guo1982: 9.3463 [#28093]). As with many other peoples who practised headhunting, there is clear evidence that the Shang elites prioritized the capture of live high-status individuals for sacrifice over the presentation of heads (Hoskins Reference Hoskins and Hoskins1996a: 24). This can be seen from the inscription: “… captured Wei Rou … one thousand five hundred and seventy heads were taken … Rou was offered in sacrifice to [the deceased Shang king] Zuding” (… qin Wei Rou … guo qian wubai qishi … yong Rou yu Zuding 擒危柔… 馘千五百七十… 用柔於祖丁) (Guo and Zhongguo shehui kexueyuan lishi yanjiusuo Reference Guo1982: 12.4541 [#36481 正]). The individual named Rou has been identified with the Weibo Rou 危伯柔 (Rou, the Chief of Wei) who is mentioned in another oracle bone inscription (Guo and Zhongguo shehui kexueyuan lishi yanjiusuo Reference Guo1982: 9.93463 [#28091]). In this instance taking such an enormous number of heads was no doubt highly prestigious, but the capture of an important enemy leader was worthy of specific record and mention by name.Footnote 11 That this practice continued through the Zhou dynasty can be seen from many bronze vessel inscriptions.Footnote 12 For the purposes of this paper, the examples given will be taken from some of the most famous and historically important inscriptions. For instance, one section of the “Xiao Yu ding” 小盂鼎 text reads:
[Yu] reported: “His majesty ordered me to … [two characters illegible] attack the Guifang … [five characters illegible], capturing alive two leaders, taking four thousand eight hundred [and] twelve heads, taking prisoner thirteen thousand and eighty-one people, capturing … [two characters illegible] horses, capturing thirty chariots, capturing three hundred and fifty-five heads of cattle, and thirty-eight sheep.” Yu again reported: “… [four characters illegible] to the end. I went on campaign, I captured one chief, I took two hundred and thirty-seven heads, I took prisoner… [two characters illegible] people, I captured one hundred and four horses, I captured one hundred … [one character illegible] chariots …”. … [The king] ordered Yu to enter the gate with his heads, and present the Xilü [captives], [two characters illegible] entered and sacrificed by burning in the Zhou royal ancestral temple… . (Zhongguo shehui kexueyuan kaogu yanjiusuo 1985: 5.247–8 [#2839]; Shirakawa Reference Shirakawa1965: 12.682–718; and Li Reference Li1987).Footnote 13
告曰: “王令盂以囗囗伐[鬼]方, 囗囗囗囗囗, 執嘼[酋]二人, 獲馘四千八百又十二馘, 孚[俘]人萬三千八十一人, 孚[俘]馬囗囗匹, 孚[俘]車卅輛, 孚[俘]牛三百五十五牛, 羊卅八羊. 盂或[又]告曰囗囗囗囗, 乎蔑. 我征, 執嘼[酋]一人, 獲馘二百卅七馘, 孚[俘]人囗囗人, 孚[俘]馬百四匹, 孚[俘]車百囗輛… 令盂, 以厥馘入門, 獻西旅, 囗囗入燎周廟…”
The “Xiao Yu ding” account, dated to the reign of King Kang of Zhou, stresses the religious context in which the presentation of heads takes place. It is repeatedly emphasized that the ceremonies in which the severed heads are part take place in the ancestral temple; furthermore, it is likely (though the text is illegible at this point so it is not entirely clear) that the heads end up being burned as part of a liao 燎 scent-sacrifice.Footnote 14 Meanwhile, the “Guo Ji Zibai pan” 虢季子白盤 inscription, dated to the reign of King Yi of Zhou, again provides clear textual evidence of the practice of headhunting and the ceremonial presentation of severed heads. This particular inscription begins:
On Dinghai day, First Auspiciousness, in the first lunar month of the twelfth year [of the reign of our king], Guo Ji Zibai made this precious basin. The greatly illustrious Zibai was brave and strong in military matters, bringing good government to the four corners of the world. He attacked and fought with the Xianyun people on the north bank of the Luo River. He cut off five hundred heads and took fifty prisoners captive: for this reason he was placed in the vanguard. The magnificent Zibai presented the heads he had taken to the king … (Zhongguo shehui kexueyuan kaogu yanjiusuo 1994: 16.177 [#10173]; and Shirakawa Reference Shirakawa1970: 32.800–13).Footnote 15
佳(唯)十又二年正月初吉丁亥子虢季子白乍(作)寶盤. 不(丕)顯子白, 壯武於戎工, 經維四方. 搏伐玁狁于洛之陽. 折首五百, 執訊五十, 是以先行. 桓桓子白, 獻馘於王…
These inscriptions, and others of the same ilk, do not necessarily describe in detail the ritual importance of headhunting.Footnote 16 However, even in situations where it is not explicitly recorded that the severed heads were presented in the ancestral shrines, or that they were subsequently burnt in a liao scent-sacrifice, a ritual element is implicit in the fact that the taking of heads was recorded on bronze vessels. These bronzes were made to be used in sacrifice to the ancestors – giving a religious dimension to situations such as the presentation of severed heads to the king. The evidence of headhunting in ancient China is not restricted to bronze vessel inscriptions, being also recorded in a wide variety of Zhou dynasty texts – historical, philosophical and ritual. These historical records include an account in the Yi Zhoushu 逸周書 (Surviving Zhou Documents) of the victory King Wu of Zhou 周武王 (r. 1049/45–1043 bce) obtained over the last king of the Shang dynasty. This account not only records an astonishing number of heads being taken, but also documents their subsequent sacrifice to the ancestors of the Zhou ruling house:
King Wu accordingly went on campaign in the four directions, whereupon he destroyed ninety-nine [enemy] states, taking the heads of one hundred and seventy-seven thousand seven hundred and seventy-nine [defeated soldiers], capturing three hundred and ten thousand, two hundred and thirty prisoners-of-war, and making six hundred and fifty-two states submit to his authority… .Footnote 17 King Wu then lined up [his prisoners] at the south gate [to the capital city], where they were sacrificed, all in full battle dress. This was done before the heads were sent in.Footnote 18 While King Wu presided over the sacrifice, the Grand Preceptor carried in on his back King Zhou of the Shang dynasty's head that was suspended from a white battle-standard and the heads of his two wives [hanging from] red battle-standards; again this was done first before the heads were sent in and sacrificed by burning in the Zhou dynastic temple (Huang, Zhang, and Tian Reference Huang, Zhang and Tian1995: 461–7 [“Shifu pian” 世俘篇]).Footnote 19
武王遂征四方, 凡憝國九十有九國, 馘魔[磿]億有十萬七千七百七十有九, 俘人三億萬有二百三十, 凡服國六百五十有二… 武王乃夾于南門，用俘皆施佩衣衣先馘入。武王在祀, 大師負商王紂縣首白旂, 妻二首赤旂; 乃以先馘入, 燎于周廟.
References to headhunting are also found in the Zuozhuan 左傳 (Zuo's Tradition). Many centuries after the events described in the previous quotation, in the aftermath of the stunning defeat inflicted by King Cheng of Chu 楚成王 (672–626 bce) on Lord Xiang of Song 宋襄公 (650–637 bce) in the battle of Hong 泓 in 638 bce, there was a somewhat unusual display of the severed heads taken by Chu soldiers:
Early in the morning on Bingwu day, the principal wife of Lord Wen of Zheng (673–628 bce), Lady Mi, and Lady Jiang, went to call upon the ruler of Chu at Keze.Footnote 20 The ruler of Chu ordered Master Jin to display the captives and heads. The gentleman said: “This is not ritually correct. A woman, when greeting or seeing off [a guest], should not go out of the door; even when meeting her brothers, she should not cross the lintel. Military matters should not be allowed to impact upon women” (Yang Reference Yang1993: 399 [Xi 22]).
丙子晨, 鄭文夫人羋氏, 姜氏, 勞楚子於柯澤. 楚子使師縉示之俘馘. 君子曰: “非禮也. 婦人送迎不出門, 見兄弟不踰閾. 戎事不邇女器.”
The objection made here is evidently not to the ceremonial display of human heads, but to the fact that it was made to women. The comments here attributed to the “gentleman” show that in the gender-segregated society of the Eastern Zhou dynasty, women were not supposed to be present in a military camp at all, and that given their presence, it was not considered appropriate for them to view the captives and heads: the presentation should be made in the presence of men, and the victory should be announced to the ancestors. The tenor of these remarks indicates that the ceremonial value of severed heads was a very important consideration in Huaxia society, and that heads were regularly collected and preserved after battle for this specific use. The same practice can be seen in the account of the battle of Chengpu 城濮, which took place in 632 bce; not only is the conduct of the battle itself described in unusual detail, but the ceremonies which took place after Lord Wen of Jin's 晉文公 (r. 636–628 bce) return to the capital are also carefully documented. The Zuozhuan places his actions after this important confrontation with the armies of Chu in the context of Lord Wen's difficulties in establishing his own authority following many years of political uncertainty in Jin – offering the heads was part of a process for confirming his position as the ruler:
In the autumn, in the seventh month, on Bingshen day, he put his army in good order so that they might enter [the capital city of] Jin in triumph. He presented captives and offered up heads, then he poured a libation and issued rewards, and summoned [the other lords] to a meeting to punish those who had treated him badly.Footnote 21 [After that] he killed Zhou Zhi Qiao in order to set an example to the country: the people then greatly submitted to his authority (Yang Reference Yang1993: 472 [Xi 28]).
秋, 七月, 丙申, 振旅愷以入于晉. 獻俘, 授馘, 飲至, 大賞, 徵會, 討貳. 殺舟之僑以徇于國: 民於是大服.
In this instance, the significance of the gathering of severed heads on the battlefield is clear: they were to be presented in the ceremony that concluded the triumphal entry of the marquis of Jin into his capital. Participating in this kind of ritual is here presented as a crucial aspect of establishing the authority of the ruler, and the more heads were present on this occasion, the greater the display of might. Although some aspects of the ceremonial presentation of severed human heads may have taken place behind closed doors, in the presence of a selected audience, it is clear from these descriptions that other parts of this ceremony did involve public participation, in which the power and success of the ruler was revealed to all his people. Furthermore, there are some instances of headhunting in a military context, in which a dead enemy commander had his head severed to be turned into a trophy; for example, this was the case with the head of the Earl of Zhi 智伯, captured and executed in 453 bce. His inveterate enemy, Viscount Xiang of Zhao 趙襄子 (r. 457–425 bce) had the skull cleaned and lacquered for use as a drinking vessel (Sima Reference Sima1959: 86.2519). In this kind of instance, victory over an enemy was not merely commemorated by taking a prestigious head, but by creating a lasting memento of the occasion. Although the skull of the last Earl of Zhi seems to have been used as a cup within a purely secular context, there is no doubt that his head was the subject of ongoing attention.
Headhunting in the Han dynasty
Headhunting continued to be openly accepted as a part of Huaxia culture until at least the time of Emperor Wu of the Han dynasty 漢武帝 (r. 141–87 bce), who is known to have been an enthusiastic proponent of the practice. His interest in taking heads can be seen in strong evidence at the time of the conquest of Nanyue 南越 in 111 bce.Footnote 22 There were two victims involved: Zhao Jiande 趙建德, the last king of Nanyue, and his Chancellor, Lü Jia 呂嘉. The importance of these head-takings for Emperor Wu are well-recorded in ancient historical texts: he renamed the place where he was staying when he heard of the fall of Nanyue and the death of its king as Wenxi xian 聞喜縣 (Hearing Good News County); while his location when he heard of Lü Jia's death was also renamed as Huojia xian 獲嘉縣 (Capturing Jia County) – both of these place-names have remained in use right up to the present day (Ban Reference Ban1962: 6.188; Yu et al. Reference Yu, Liang, Xia, Lan and Tan1988: 231–3). In the case of Lü Jia in particular, Han dynasty sources repeatedly stress that Emperor Wu was delighting in the taking of his head (de Lü Jia shou 得呂嘉首) (Ban Reference Ban1962: 6.188; Xun Reference Xun1973: 14.2a). What is more, Emperor Wu's ambassador to the Xiongnu is recorded as having discussed Han imperial head-taking with their ruler, the Chanyu Wuwei 烏維單于 (r. 115–105 bce):
At this time, the Son of Heaven went on a progress to the border region, and when he arrived at Shuofang, he stationed one hundred and eighty thousand cavalry there in a show of strength, and sent Guo Ji to report to the Chanyu as quickly as possible. The Chanyu granted an audience to Ji, and Ji said: “The head of the king of Nanyue is already hanging from the North Gate at the Han [Palace].Footnote 23 Now if the Chanyu is in a position to be able to do battle with the Han, the Son of Heaven has come in person in command of troops, and is waiting at the border. If on the other hand the Chanyu is not in a position to be able [to do this], he should face south and declare himself a vassal of the Han. Otherwise his only choice is to run far away, and hide himself north of the desert, where it is cold and [life is] harsh, and there is no water or grass, for there is nothing else he can do.” When he had finished speaking the Chanyu was absolutely furious … He kept Guo Ji and would not allow him to go home, sending him to live in the regions above the northern sea. However, in the end the Chanyu was not willing to raid the Han borders and he rested his officers and men, getting them to practise their shooting and hunting skills. He repeatedly sent ambassadors to the Han, asking for a marriage alliance with honeyed words (Sima Reference Sima1959: 110.2913; Ban Reference Ban1962: 6.189).Footnote 24
是時天子巡邊, 至朔方, 勒兵十八萬騎以見武節, 而使郭吉風告單于… 單于見吉, 吉曰: “南越王頭已懸於漢北闕. 今單于 (能) 即 [能] 前與漢戰, 天子自將兵待邊; 單于即不能, 即南面而臣於漢. 何徒遠走, 亡匿於幕北寒苦無水草之地, 毋為也.” 語卒而單于大怒… 而留郭吉不歸, 遷之北海上. 而單于終不肯為寇於漢邊, 休養息士馬, 習射獵, 數使使於漢, 好辭甘言求請和親.
In making his remarks, Emperor Wu's ambassador shows the significance attached at the Han court to the display of the severed heads of enemies in making manifest the emperor's own power and authority. There is no evidence of Han dynasty headhunting concerning other inhabitants of the conquered kingdom of Nanyue. Although many people are likely to have been killed, this is not recorded, nor were their severed heads the subject of any subsequent concern. This suggests that for Emperor Wu and his generals, only the most prestigious persons were worthy of headhunting (Hoskins Reference Hoskins1996b: 227). The ambassador's references to the public display of severed enemy heads – reinforced by the appearance of this passage in the two dynastic histories which cover the Western Han dynasty: the Shiji and the Hanshu – also suggests that the ruling elite of this period would have fully agreed with modern scholars of headhunting, that it is not the death of an enemy that is the most important factor in such practices, but the attentions focused on the severed head (McKinley Reference McKinley2015: 464–72). Through these practices, which here seem to have mainly involved the display of the king of Nanyue's head at the gate to the palace, the victor's world-view is reinforced: his is the correct, indeed the only possible, perspective on the situation.Footnote 25 Both the Shiji and the Hanshu accounts describe how the Xiongnu Chanyu was persuaded by being told of this show of headhunting into sending ambassadors to the Han to sue for peace: this emphasizes that Emperor Wu's views had prevailed.
On destroying knowledge of headhunting
A key element in the elimination of knowledge of the history of headhunting within Huaxia culture was not to bring this practice to an end, but to redefine the nomenclature. Tang Shunzhi 唐順之 (1507–60), in the Ming dynasty, was one of the first scholars to have noticed the existence of this problem. Working exclusively with transmitted texts, he reached the erroneous conclusion that prior to the unification of China, casualties in war were calculated by collecting ears and noses, and that during the Qin and Han, people collected heads for the same purpose (Tang Reference Tang1986: 1.21a–21b [“Shang” 賞]). The obscuring of the history of headhunting in China rests on the two characters guo 馘 and guo 聝. The first is an ancient word, found in Shang dynasty oracle bone inscriptions and Zhou dynasty bronze vessel texts, and means “to kill an enemy and take his head”. The second character is one thought to have been invented in the Eastern Han dynasty – appearing first in the Shuowen jiezi 説文解字 (Explaining Graphs and Analyzing Characters) dictionary, which glosses guo 聝 as meaning: “to fight a battle and sever the ear [of an enemy combatant]” (junzhan duan er 軍戰斷耳) (Duan Yucai Reference Duan1964: 598 [“Erbu” 耳部]).Footnote 26 It seems that the practice of using severed ears to count enemy dead was not originally used in Huaxia culture; however, after the introduction of this new character and the promulgation of its meaning, the presentation of decapitated heads after battle to claim rewards did decline, to be replaced by this new method. Furthermore, although the second of these characters seems to have been extremely rarely used, the meaning of the first character was gradually subsumed into the meaning of the second, thus destroying any knowledge of ancient Chinese people as takers of heads (Wei Reference Wei2014; Wang Reference Wang2005; and Liu Reference Liu1992). Over time, this later interpretation of the ancient character of guo 馘 resulted in more and more ancient texts being misread, as the second character entered usage through commentaries which promulgated a different interpretation of past violence. For example, in the Shijing 詩經 (Book of Songs) ode “Huangyi” 皇矣 (How August), there is what might be thought to be an unequivocal reference to the extensive ceremonies which concluded a successful headhunting expedition:
Captured prisoners advance in serried ranks; then severed heads are carefully placed.
They are used in sacrifice to Heaven; they are used in sacrifice to Earth.Footnote 27
They are used in sacrifice to the state altars; they are used in sacrifice to the ancestors (Kong Reference Kong1999: 1034–5).
執訊連連, 攸馘安安.
是類是禡, 是致是附.
In the Han dynasty Mao 毛 commentary on this ode, which is addressed to the glory of King Wen as the founder of the Zhou dynasty, the new meaning of the term guo has already been incorporated: “Guo means to capture. To kill those who do not submit to authority and present their left ears is called guo” (guo, huo ye. Bufu zhe sha er xian qi zuo er yue guo 馘, 獲也. 不服者殺而獻其左耳曰馘) (Kong 1999: 1035 [“Huangyi”]). This kind of interpretation was frequently emphasized by imperial-era commentators through their quotations of the Mao commentary, though beginning in the Song dynasty there seems to have been a particular stress on the term anan 安安 (“carefully”) as indicating a lack of violence (bubao 不暴, or buqingbao 不輕暴: “not easily moved to violence”) appropriate to King Wen's reputation for virtue, in spite of the obvious horror of the action of dismembering another person – a reading of the text which seems to have been derived from the highly influential commentary produced by the brothers Cheng Hao 程顥 (1032–1085) and Cheng Yi 程頤 (1033–1107) (Cheng and Cheng Reference Cheng and Cheng1986: 4.24b).Footnote 28 Interestingly, the Cheng brothers seem to have been well aware of the older tradition that the term guo referred specifically to severed heads, because they gloss it in this way: “Guo means to behead prisoners” (guo, zhanhuo ye 馘, 斬獲也). This gloss is virtually never quoted by other Song dynasty or later imperial era scholars, even though they were quite happy to cite the Cheng brothers’ appreciation of King Wen's virtue.
A further example of the new reading of the character guo can be found in the commentarial tradition concerning a passage in the “Wangzhi” 王制 (Royal Regulations) chapter of the Liji 禮記 (Records of Ritual), which describes the proper religious ceremonies to be gone through when undertaking a military campaign. The original text reads:
When the Son of Heaven was about to go out on campaign, he would offer a lei-sacrifice to God on High, an yi-sacrifice to the state altars, a zao-sacrifice to his deceased father, and a ma-sacrifice to the lands where he would go on campaign. He would report his actions to his ancestors [at the temple]; and he would fix his strategy at the National Academy. When he went out on campaign, he would capture those who had committed crimes; on his return, he would make sacrificial offerings and pour libations at the National Academy, and present his prisoners and the severed heads [of his enemies] (Kong Reference Kong1999b: 371).
天子將出征, 類乎上帝, 宜乎社, 造乎禰, 禡於所征之地. 受命於祖, 受成於學. 出征, 執有罪; 反, 釋奠于學, 以訊馘告.
In the Eastern Han dynasty, the commentary by Zheng Xuan 鄭玄 (127–200 ce) on this short passage includes a gloss on the difficult word guo: “Xun and guo are those who are captured alive and those who have their ear cut off” (xun guo, suo shenghuo duan'er zhe 訊馘, 所生獲斷耳者).Footnote 29 This interpretation, in line with contemporary thought on the meaning of this term, was again highly influential in the later imperial era; for example Kong Yingda's commentary on the same passage says: “Xun are those who are alive; guo are those who are dead and have had their ears cut off” (xun shi sheng zhe, guo shi si er jie er zhe 訊是生者, 馘是死而截耳者) (Kong Reference Kong1999b: 373 [“Wangzhi”]). This kind of unanimity on the meaning of this character was common throughout the imperial era and well into modern times. It is probably safe to say that by the time of the Tang dynasty, educated Chinese people were generally unaware of their ancestors’ traditions of headhunting. Thus, in the Suishu 隋書 (History of the Sui dynasty), compiled by Wei Zheng 魏徵 (580–643), the practices of the people of the Ryukyu Islands (Liuqiuguo 流求國) had to be explained in some detail:
It is their custom to serve the deities of the mountains and seas; they sacrifice to these gods with meat and wine, but when they fight battles and kill people, they take those who have been killed to be sacrificed to their gods. Sometimes they erect brushwood into a little hut, at other times they suspend skulls from trees in order to shoot arrows at them… . The residence of their monarch has piles of skulls at the foot of the walls and they consider this to be a fine thing (Wei Reference Wei1973: 81.1824).
俗事山海之神. 祭以酒肴, 鬭戰殺人, 便將所殺人祭其神. 或依茂樹起小屋, 或懸髑髏於樹上以箭射之… 王之所居多聚髑髏以爲佳.
This kind of statement is indicative of the way in which attitudes had changed. The concluding words, “they consider this to be a fine thing”, presupposes that contemporary readers will not agree; however, had this kind of practice been discussed with an earlier generation of members of the ruling elite, they would most likely have completely understood the desirability of making a show of strength through the ceremonial display of captured heads. Although Huaxia peoples might traditionally have employed different manifestations of the ritual incorporation of enemy dead into their own cultures through the presentation of severed heads, there is every chance that they would have recognized what the Ryukyu islands’ rulers were trying to do, and agreed that this was, in general, a very fine thing indeed.
Conclusion
This study of headhunting in ancient China raises a number of interesting points. The first concerns the way in which, in spite of being a well-documented part of traditional Huaxia culture and forming a defining feature of their practice in warfare, it is nevertheless known today only to a handful of experts on the period. The texts which form the basis of this study are still widely read by historians and literary scholars, and yet what might seem to be unequivocal references to headhunting are ignored, because of the successful redefining of the terminology which occurred in the Eastern Han dynasty. Nearly two thousand years of reinterpretation has its own weight, and even the most clear-cut of descriptions of ceremonies carried out to celebrate the taking of enemy heads have been elided into the apparently more acceptable practice of presenting the severed ears of enemy soldiers as a means of counting the dead. That it is possible to eliminate so many centuries of headhunting from the history of Chinese warfare and religious practice with success, in turn raises the question of what else has been lost from the record? What other practices, important in their own era, have been written out because of changing attitudes about what is acceptable?
The second point of interest concerns the subsequent history of headhunting in China. The creation of a new term, which in turn allowed for the reinterpretation of the past (and indeed the development of a new “traditional practice” of taking ears), did not mean that headhunting ceased entirely. Heads continued to be taken and displayed in public, and ceremonies continued to be held around them, for many centuries after the original specialized terminology concerning this practice was removed from general use. The later, imperial era history of headhunting – and general acceptance of this custom – is beyond the scope of this paper, but it is undoubtedly an extremely important subject for further research. It remains to be seen whether the elimination of old vocabulary resulted in a new set of words developing, or whether headhunting simply became a described but otherwise effectively undocumented practice.
The final point concerns Emperor Wu of the Han dynasty himself. The reign of this ruler is often seen as the apogee of power for the dynasty, and he is frequently credited with the introduction of Confucianism, with its condemnation of violence and its prejudices against dismemberment, as state orthodoxy. Again, the vexed question of the relationship between this monarch and the imposition of Confucianism is beyond the scope of this paper, but nevertheless it is worth considering the contradiction that Emperor Wu was on the one hand a supporter of Confucian thought and on the other an enthusiastic proponent of headhunting. This is an interesting example of the way in which old ideas – traditional assertions of power and authority – still hold sway even as newer forms of rule are introduced. It is not at all impossible that one monarch should wish to encompass both, and that headhunting still had a role to play in the government of the Western Han dynasty, when demonstrating that the emperor meant to be obeyed.
[師]於城濮. 述[遂]朝周襄王于衡澭[雍]. 獻楚俘馘) (Li 
𥄉’ 及其相關字. 
