Hostname: page-component-68c7f8b79f-lvtpz Total loading time: 0 Render date: 2025-12-16T06:38:00.736Z Has data issue: false hasContentIssue false
  • English
  • Français

Cooking meat in ancient China: luan 臠, zi 胾, and the origins of the kebab

Published online by Cambridge University Press:  04 December 2025

Olivia Milburn Open the ORCID record for Olivia Milburn [Opens in a new window]
Olivia Milburn*
Affiliation:
School of Chinese, University of Hong Kong, Hong Kong, China
*
Email: omilburn@hku.hk
Rights & Permissions [Opens in a new window]

Abstract

Considerable evidence survives of the cooking and eating of kebabs as a major form of meat consumption in early China. Not only are there numerous artistic depictions in both painting and low-relief stone sculpture of this practice, but there are also some very early excavated skewers, grills, and indeed preserved meat kebabs, not to mention references in contemporary literature, and this evidence significantly predates any documentation of kebabs in the Middle East. However, in spite of this wealth of documentation, this tradition has gone largely unexplored, partly due to scholars failing to understand the relevant terminology and partly due to an unjustified belief that all kebab cooking must derive from the Middle East. This article explores the indigenous ancient Chinese tradition of kebab cooking, focused on grilling and roasting of smaller (luan) or larger (zi) pieces of meat seasoned with soybean pastes and sauces, which developed independently of other similar culinary practices elsewhere. This analysis is focused on literary evidence of the Chinese kebab, with particular reference to the contents of a recently discovered very early cookbook, dating to the Han Dynasty, excavated from the tomb of Wu Yang, first marquis of Yuanling, who died in 162 BCE.

Keywords

cookbooksearly Chinakebabsmeat cooking

Information

Type
Article
Information
Journal of the Royal Asiatic Society , First View , pp. 1 - 16
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of The Royal Asiatic Society.

Introduction

This article explores early Chinese culinary traditions, as recorded for the period circa 500–100 BCE, relating to the cooking of meat that was first prepared by being cut into small pieces (kebabs) and then quickly cooked in dry heat, usually on skewers or a turned spit (shish kebab or shashlik).Footnote 1 In other words, the term ‘kebab’ denotes a foodstuff defined by both method of preparation and manner of cooking.Footnote 2 In the present day, most people in China discussing pieces of meat affixed on skewers would use the term chuan 串, with the Chinese graph a helpfully pictorial representation of the concept. When this graph appeared in early Chinese literature, however, it was in the meaning of ‘custom’ or ‘habit.’Footnote 3 The terms used in ancient China for meat cut into pieces for quick cooking in dry heat were either luan 臠 or zi 胾, with the distinction between the two apparently based on size. This division was promulgated by the Shuowen jiezi 説文解字 (Explaining Graphs and Analysing Characters) dictionary, which noted that there were two different ways of combining the rou 肉 (meat) radical with the phonetic luan 䜌 ( and 臠): both of these graphs are pronounced luán: ‘luan means emaciated … Alternatively, meat that has been chopped into pieces is called luan’ (, 臞也。⋯ 一曰切肉, 臠也).Footnote 4 Subsequently, this dictionary provides a further gloss: ‘zi are large luan’ (胾, 大臠也).Footnote 5 This would become an enormously influential reading, which was frequently quoted in commentaries on references to meat cookery in classical texts from the Han Dynasty onwards.

Although kebabs are today ubiquitous in many countries around the world, largely thanks to ongoing Islamic influence, this is a relatively recent development and there are considerable variations as to which types of this dish have been popularised in different places.Footnote 6 The modern dominance of Middle Eastern kebabs and their associated vocabulary has served to obscure the presence of a much older and very similar tradition of meat cookery in China. However, considerable textual evidence survives in Chinese literature from the pre-unification to early imperial period of the importance of kebabs in elite cooking, and as will be discussed below, the earliest unequivocal reference to shish kebabs in any language is found in the Han Feizi 韓非子. Multiple recipes for how to make kebabs also appeared in the earliest known Chinese-language cookery book, the Meishi fang 美食方 (Gastronomic Recipes) or Shifang 食方 (Food Recipes) bamboo manuscript, excavated in 1999 from the tomb of Wu Yang 吳陽, first Marquis of Yuanling 沅陵侯 (d. 163 BCE).Footnote 7 In addition, the desiccated remains of Han Dynasty mutton shish kebabs were discovered in a tomb in the cemetery at Changle Village 常樂鎮, Zhongwei City 中衛市, Ningxia Province, in 2012.Footnote 8 A number of Eastern Han Dynasty artistic depictions of kebabs have been excavated in modern times, including a painted pottery stove showing a female cook holding up three kebab skewers just prior to putting them onto the grill excavated from a tomb in Wei County 蔚縣, Hebei Province, in 1996;Footnote 9 as well as numerous low relief stone sculptures of kitchen scenes depicting the preparation and cooking of shish kebabs.Footnote 10 This means that both literary and artistic portrayals of kebab cooking and physical evidence that lumps of meat were indeed being placed on skewers to make shish kebabs in China predate the earliest references to be found in Arabic and Persian literature by more than 1,000 years.Footnote 11 Accordingly, this article will argue that cooking and eating kebabs was an important element in early Chinese culinary culture (particularly amongst the elite who could afford to eat meat regularly); that the assumption that all kebab cooking traditions represent cultural influence from the Middle East is unjustified; and that an insufficient understanding of the vocabulary associated with this style of meat cookery has ensured that the dominance of kebabs within Chinese cuisine in antiquity has been ignored.

Smaller pieces of meat: luan

In early Chinese discussions of meat cookery, luan appears much more rarely than zi, and it can sometimes be used in non-culinary contexts to simply mean ‘lumps of flesh’. For example, at the end of the Xin Dynasty when the first and only emperor–Wang Mang 王莽 (r. 9–23 CE)–was beheaded, his corpse was pulled to pieces (luanfen 臠分) by the dozens of people fighting to participate in the dismemberment. Luan may be used here to describe the smallness of the parts to which his body was reduced and thus by extension to emphasise the rage of the populace, or the choice of this word may perhaps have been intended to be dehumanising.Footnote 12 However, far more commonly, luan applied to meat kebabs, as can be seen from the famous story about a spoiled dish of shish kebab from the Han Feizi:

In the time of Lord Wen [of Jin (r. 636–628 BCE)], the official in charge of the kitchen served a dish of roasted meat with a hair wrapped around it.Footnote 13 Lord Wen summoned the chef and complained to him: ‘Do you want me to choke? Why was there a hair wrapped round the roasted meat?’ The chef knocked his head on the ground and bowed twice before he exculpated himself, saying: ‘I have committed three crimes deserving of the death penalty: I plied the whetstone against my knife until it was as sharp as [the sword] Gan Jiang to cut up the meat.Footnote 14 The meat was cleaved and yet the hair was not—this is my first crime. I plied the wooden [skewers] so that they speared the kebabs and yet I did not see the hair—that is my second crime. I held [the kebabs] over the stove as the charcoal was crimson and flames vermilion, so that the roasted meat was done and yet the hair was not burned—that is my third crime.Footnote 15 Is there perhaps someone in your household who is secretly envious of me?’Footnote 16

文公之時, 宰臣上炙而髮繞之, 文公召宰人而譙之曰: “女欲寡人之哽邪? 奚為以髮繞炙。” 宰人頓首再拜, 請曰: “臣有死罪三: 援礪砥刀, 利猶干將也, 切肉, 肉斷而髮不斷, 臣之罪一也。援木而貫臠而不見髮, 臣之罪二也。奉熾爐, 炭火盡赤紅, 而炙熟而髮不燒, 臣之罪三也。堂下得無微有疾臣者乎?”

The clever chef in this story provides a very detailed account of the cooking of this dish to indicate that he cannot possibly be held responsible for its spoiling: this must be the work of an envious server after the food had already left the kitchen. Since this description includes references to cutting up the meat neatly, skewering each piece one by one, and then roasting them over hot charcoal, it is clear this dish of small pieces of meat cooked in dry heat was a form of shish kebab. During the course of the Han Dynasty, artworks would be produced to illustrate just the scene described in this story, with a chef holding skewers over the stove.Footnote 17 It is also worth noting that as a member of the Han royal family, the author of this text, Han Fei 韓非 (d. 233 BCE), found it entirely plausible that aristocrats ate this kind of food—grilled and roasted skewered meat was a gastronomic delicacy relished by the kings and aristocrats of early China. Although not a recipe derived from a cookery book, this account appears to represent the earliest known literary reference to the shish kebab in any language. Furthermore, to highlight the distinction made between luan and larger kebabs, or zi, Han Fei also used this term in a different section of his text:

When little children play together, they have soil as their rice, mud as their stew, and [pieces of] wood as their kebabs, but at the end of the day they have to go home to eat, for soil rice and mud stew can be used for play but not for eating.Footnote 18

夫嬰兒相與戲也, 以塵為飯, 以塗為羹, 以木為胾, 然至日晚必歸饟者, 塵飯塗羹可以戲而不可食也。

Shish kebabs continued to form part of elite fine dining after the unification of China. Stoves and grills have been excavated from a number of early imperial era tombs, including the very fine small bronze grill, elegantly balanced on four feet, with holes in the sides for slotting in skewers and ring handles for transportation, that was excavated from the tomb of Zhao Mo 趙眜, Emperor Wen of Nanyue 南越文帝 (r. 137–124 BCE), the ethnically Chinese ruler of an independent southern kingdom with its capital at Guangzhou.Footnote 19 This object is of such high quality as to suggest that it was intended to allow a chef to cook shish kebabs in the royal presence. In the same period, kebabs could also feature at Han Dynasty imperial banquets in the capital city, according to at least one account of the test set for Zhou Yafu 周亞夫 (d. 143 BCE) by Emperor Jing 漢景帝 (r. 157–141 BCE). Invited to dine in the company of the emperor, the meat dishes were served in such a way as to test Zhou Yafu’s obedience and humility, qualities that he singularly failed to display in practice:

[His majesty] summoned Yafu and bestowed food upon him, but he only served big pieces of meat (zi) with no small shish kebabs (luan), and he also did not provide him with chopsticks. Yafu was angry about this and turned his head to ask the mat attendant to get him some chopsticks. Yafu then proceeded to eat. When he left, his majesty followed him with his eyes and said: ‘That moaner is no vassal for a young ruler.’Footnote 20

召亞夫賜食, 獨置大胾無臠, 又不置箸。亞夫心不平, 顧謂掌席者取箸。亞夫前食。既出, 上目送之, 曰: ‘此怏怏非少主之臣也。’

In other accounts of these events, preserved in the Shiji 史記 (Records of the Grand Historian) and Hanshu 漢書 (History of the Han Dynasty), Zhou Yafu is said to have been served big pieces of meat (again described by the term zi 胾) but not meat chopped into pieces (qierou 切肉).Footnote 21 As has been exhaustively discussed by commentators for the last 18 centuries, this does not make sense, for there is no reason why large lumps of meat would be more difficult to eat without implements versus meat chopped into smaller pieces.Footnote 22 However, if this story is read in conjunction with the Han Feizi, and luan (and by extension qierou) are understood to refer specifically to shish kebabs served on the skewer, Zhou Yafu would not have needed chopsticks since he could pick up the skewers by hand.Footnote 23 By serving him only large pieces of cooked meat with no means to pick them up, Zhou Yafu was forced to either ask for chopsticks or go hungry. His failure to take the second option irritated Emperor Jing and convinced him that Zhou Yafu would not make a suitably subservient minister for his young heir. As a result of his poorly considered behaviour, within months of this dinner, Zhou Yafu was arrested and thrown into prison, where he subsequently starved to death.

It is difficult to understand the story of Zhou Yafu and exactly what went wrong at this imperial banquet. However, it is entirely possible that the entire situation was a trap, and one that the victim did not have the wit to get out of. The account in the Yanzi chunqiu 晏子春秋 (Spring and Autumn Annals of Master Yan) of Yan Ying’s 晏嬰 (d. 500 BCE) very delicate and dangerous diplomatic mission to the king of Chu recounts an incident in which he was given an orange to eat and a knife to peel it with. Nevertheless, Yan Ying ended up eating the fruit whole: the king had presented him with the fruit and etiquette demanded that he eat it with gratitude, but his majesty had not given him permission to remove the peel.Footnote 24 As events proved, Yan Ying was fully capable of successfully negotiating this kind of damned-if-you-do-and-damned-if-you-don’t situation, while Zhou Yafu was not.

It is possible that not all references to luan pertained to shish kebab, since it may also have referred to pieces of meat which had been quickly cooked in other ways. The Huainanzi 淮南子 gives the Han Dynasty aphorism: ‘By tasting one kebab of meat, you can know the flavour of an entire pan’ (嘗一臠肉, 知一鑊之味).Footnote 25 The huo 鑊 was a kind of bronze frying or roasting pan; as noted by Zheng Xuan 鄭玄 (127–200 CE) in his commentary on this word in the Zhouli 周禮 (Rites of Zhou): ‘A pan is an implement in which you can roast meat or fish or salted meat, and when it is done, it is finished in a ding-cauldron’ (鑊, 所以煮肉及魚, 臘之器, 既熟, 乃盛於鼎).Footnote 26 If pieces of meat were roasted or pan-fried first, they could no doubt be garnished in any number of ways in the ding prior to serving. However, it is likely that the Huainanzi represents an unusual use of the term luan, for other references to this saying make use of a slightly different form of wording. Thus, the Lüshi chunqiu 呂氏春秋 (Spring and Autumn Annals of Mr Lü), compiled at the behest of Lü Buwei 呂不韋 (d. 235 BCE), says: ‘By tasting one piece of meat, you can know the flavour of an entire pan and the seasoning of a whole ding-cauldron’ (嘗一脟肉, 而知一鑊之味, 一鼎之調).Footnote 27 It is likely that in this instance the older text preserves the original wording and that luan in pre-unification and early imperial literature did generally specifically denote small shish kebabs.

Larger pieces of meat: zi

Zi is by far the most common term used in Chinese literature for pieces of meat that had been fried, grilled, or roasted. Furthermore, it appears in much older works of literature, as can be seen from the fact that this word figures in the hymn ‘Bigong’ 閟宮 (The closed temple) preserved in the Shijing 詩經 (Book of Songs), traditionally attributed to Lord Xi of Lu 魯僖公 (r. 659–627 BCE). In this song, the ancestral line of the lords of Lu is celebrated, recounting their position as a junior branch of the Zhou imperial house. Included in the ‘Bigong’ is an account of the meat dishes presented in sacrifice to the ancestors: ‘[Pork] roasted whole, kebabs, and stews,/ [Offered in] baskets and wooden bowls, and on large trays’ (毛炰胾羹, 籩豆大房).Footnote 28 The use of kebabs in sacrifice is also documented in the Yili 儀禮 (Ceremonies and Rituals), which describes the formal ceremonies for an ancestor in which an impersonator of the dead man (shi 尸) was served various fine foods in order to present them as sacrifices to the deceased. These offerings included (amongst many other foodstuffs) a pair of dishes of dried meat and kebabs (zi) which would be served simultaneously on earthenware dishes, with a pickled meat condiment. The impersonator was supposed to eat a kebab, before the ceremony proceeded with the presentation of a dish of fish.Footnote 29

Early texts dealing with the complex etiquette observed at social events such as banquets are very detailed in setting out exactly how tables should be laid, including placement of kebabs alongside other dishes served to elite diners. In these writings, the term used for grilled or roasted meat that had been cut into pieces for cooking is always zi. For example, the Liji 禮記 (Record of Ritual) gives a very detailed description of how comestibles should be arranged before each guest:

As for the correct etiquette for presenting food [at a banquet], meat served on the bone should be on the left and kebabs on the right, with the grain placed on the person’s left and the thick stew placed on the person’s right.Footnote 30 Finely diced and roasted meat is positioned outside,Footnote 31 with fermented meat paste and soybean sauce inside.Footnote 32 [Raw] spring onions and steamed spring onions can be added if required, while wine and posca are placed to the right.Footnote 33 If [sliced] cured meats are served, then the folds should be on the left side and the ends on the right.Footnote 34

凡進食之禮, 左殽右胾。食居人之左, 羹居人之右。膾炙處外, 醯醬處內。蔥渫處末, 酒漿處右。以脯修置者, 左朐右末。

It is worth noting in this description that the items set to the guest’s right hand (i.e. the dominant side for most people) are the kebabs, the stew, and the drinks. The staple grain was placed by the left hand, as it still would be today. This would seem a more sensible reading than the commentators who have assumed a yin-yang system underpinned this placement of comestibles.Footnote 35 Banquet eating is also mentioned in the Yili, which also describes a very complicated arrangement of different meat dishes, orientated by cardinal points rather than to the left and right of the diner. So detailed do these instructions appear that many people over the centuries have attempted to draw the layout of dishes at such sumptuous meals, only to discover that there is considerable information missing.Footnote 36 However, it is evident that when a lord feasted government officials, it was appropriate to serve a wide range of meat dishes, including kebabs made from various different animals:

South of the roasted [beef], there is fermented meat paste, and to the west of that there are beef kebabs, fermented meat paste, and sliced raw beef. South of the sliced raw [beef], there is roast mutton, and to the east of that there are lamb kebabs, fermented meat paste, and roast pork. South of the roast [pork], there is fermented meat paste, and to the west of that there are pork kebabs, mustard seed sauce, and sliced raw fish.Footnote 37

炙南醢, 以西牛胾, 醢, 牛鮨; 鮨南羊炙, 以東羊胾, 醢, 豕炙; 炙南醢, 以西豕胾, 芥醬, 魚膾。

Even on less formal occasions, surviving ritual texts hint at a wide range of food consumption practices and taboos that are very difficult to understand at this remove. Thus in the ‘Neize’ 内則 (Regulating the interior) chapter of the Liji, the text informs us: ‘When grandees ate their main meal of the day, if they had sliced raw meats, they did not have cured meats; while if they had cured meats, they did not have raw meats; while knights did not have both stew and kebabs’ (大夫燕食, 有膾無脯, 有脯無膾; 士不貳羹胾).Footnote 38 According to texts like the Mozi 墨子, this kind of self-restraint was also associated with the frugal dietary practices of the sage kings. The luminous virtues of Yao 堯 and the vastness of the domains that he ruled stand in marked contrast to the simplicity of his diet and the restraint he shows in cooking methods—slow cooked stews and quick cooked kebabs were not offered at the same meal in order to save fuel and economise on expensive ingredients:

In the past, when Yao governed All-Under-Heaven, to the south he pacified Jiaozhi while to the north he forced Youdu to surrender, while to east and west as far as where the sun rose and set, there was no one who did not respectfully submit to him. Yet even when he was served [the foods] he liked best, he did not [eat] both glutinous and non-glutinous millet, he did not double up stew and kebabs, as he ate from an earthenware bowl and drank from an earthenware cup, measuring out his drink with a spoon. As for the rituals of raising and lowering the head, bowing down and rising up again, and parading about in an ostentatious manner: these are not things which the sage king did.Footnote 39

古者堯治天下, 南撫交阯北降幽都, 東西至日所出入, 莫不賓服。逮至其厚愛, 黍稷不二, 羹胾不重, 飯於土塯, 啜於土形, 斗以酌。俛仰周旋威儀之禮, 聖王弗為。

In this short passage, the Mozi advocated a wide range of economical behaviours—the sage king limits his consumption of foodstuffs, drinks, utensils, and fuel, as well as practising a parallel economy of physical action, asking neither himself nor others to bow and scrape. Such abstinence and self-control did not appeal to everyone. Certainly, meat eating was something that many people in early China indulged in if they could, and kebabs were a very common form of cooked meat. For example, the teacher in the ‘Dizi zhi’ 弟子職 (Duties of the disciple) chapter of the Guanzi 管子 is served kebabs by his students when it is time for him to eat his main meal. This text, written in four-characters-per-line rhymed verse and describing the daily routine from morning until night in the life of students studying under a master, is thought to have been composed before the third century BCE (possibly in the context of the Jixia Academy 稷下學宮 which was extant circa 318–284 BCE), though the dating remains highly controversial.Footnote 40 In addition to circulating as an independent text, it was subsequently included in the amorphous compendium of material known as the Guanzi, though it does not appear to have anything to do with Guan Zhong 管仲 (d. 645 BCE), the focus of much of this eponymous compilation.Footnote 41 Rather than describing what these students learn, the ‘Dizi zhi’ describes what they do, focusing on the round of chores set to them by their teacher within a closed hierarchical community, which includes serving his dinner:

For some reason, thanks to the vagaries of textual transmission, there is not a single reference in the transmitted tradition to larger kebabs, or zi, that gives any indication of how they were cooked. There is ample evidence that zi could be made from the meat of many different animals, including sheep, pigs, and cattle; that they involved the meat being cut into pieces and they should be pure, bone-free pieces; and that this meat was then presumably cooked in some way prior to serving. It is this lack of precision about the cooking of zi that makes the Shifang collection of recipes so valuable.

The evidence of the Shifang

The earliest known Chinese cookbook, the Shifang, was excavated from the tomb of a quite well-recorded individual: Wu Yang, the first marquis of Yuanling. During the Qin Dynasty, his grandfather, Wu Rui 吳芮 (d. 202 BCE), had been a very popular administrator of the indigenous population of Poyang 鄱陽 in what is today Jiangxi Province, and as the regime spiralled into civil war, he and his son-in-law, Ying Bu 英布 (d. 195 BCE), raised the local Baiyue 百越 people to fight in support of Liu Bang 劉邦 (r. 202–195 BCE), who would eventually become the first emperor of the Han Dynasty. For his role as a long-standing, loyal supporter of the nascent dynasty, Wu Rui was rewarded with the title of king of Changsha 長沙王.Footnote 46 On his death, he was succeeded by his oldest son, Wu Chen 吳臣, who became King Cheng of Changsha 長沙成王 (r. 202–194 BCE).Footnote 47 As one of Wu Chen’s younger children, Wu Yang could not inherit his father’s main title; instead, in 187 BCE, he was honoured with the marquisate of Yuanling, a position which he held until his death in 162 BCE. As befits the scion of such an important family, Wu Yang is mentioned in both the Shiji and Hanshu: the two main historical accounts of this period.Footnote 48 The interest in gastronomy revealed by the excavation of his tomb came as a complete surprise to scholars; interestingly, this text on fine dining seems to be strongly associated with northern, imperial modes of eating, with an emphasis on exceptionally expensive ingredients like meat and wine.Footnote 49 It is not for nothing that early Chinese elites were known as ‘meat eaters’ (roushizhe 肉食者).Footnote 50

The excavators of the tomb at Yuanling found the Shifang text reduced to some 300 fragments, making it very difficult to interpret. This goes some way to explaining why so important a text has been generally overlooked in current scholarship on early Chinese foodways.Footnote 51 The surviving material makes it clear, however, that this book was entirely devoid of any reference to dietetics or the medical benefits of foodstuffs, being instead focused entirely on gastronomy. Furthermore, as with other early cookbooks from around the world, it presumed that the reader already knew how to cook and merely required suggestions for improving their repertoire. From surviving slip headings (unfortunately one of the most easily damaged areas), it is evident that the Shifang originally consisted of at least 148 numbered recipes. Of these, recipes 46 and 47 described how to make pork and chicken kebab dishes. In the transcription of these two strips below, the subscript numbers indicate the numbering allocated to each bamboo fragment at the time of excavation, □ is used to indicate an illegible character, […] indicates a preferred reading for the character, 【】indicates a graph suggested by modern scholars for an illegible character, and … indicates where the strip has been broken. This text states:

16 Recipe 46, To Make Pork Kebabs: First kill [your pig], then burn off all the hair, then with your hand moving against the grain pull out [the last remaining hairs].Footnote 52 Spit-roast [your meat pieces] using soybean sauce and dripping [for basting; three characters illegible] …Footnote 53

16 卌六, 為豚胾方: 先刺殺, 乃燒齊毛, 以手逆搯之掾。炙以【醬汁】□□□…

34 Recipe 47, To Make Chicken Kebabs: First kill [your chicken], then scald it in hot water. Once all its feathers have been removed, stir, to get rid of its [three characters illegible]…Footnote 54

34 卌七, 為鷄胾方: 先刺殺, 乃熱湯潦之, 毛盡撓, 去其□□□…

Since both these bamboo slips are badly broken after the initial instructions for preparation, it is impossible to know more about how these dishes were cooked. However, slip 16 specifically uses the term zhi: to roast on a spit.Footnote 55 In addition, slips 152 and 192 give the headings for recipes for making meat kebabs of an unknown kind; and slip 205 gives the heading for a dish of lamb kebabs.Footnote 56 This last recipe specifically advocates using shoulder of lamb, which remains today the most favoured cut of meat for making kebabs. If any of the accounts preserved in this text of marinades and basting ingredients—wine, salt, ginger, soybean sauce, fermented bean paste, and so on—pertained to these kebab recipes, the damage to the bamboo slips prevents us from recognising this.Footnote 57

These are not the only references to kebabs made in the Shifang. Another recipe in this text begins with a discussion of meat stewed slowly in a cauldron over the fire, before proceeding to describe what appears to be a side-dish of kebabs (or perhaps a garnish to the original stew) that would go towards making a very elaborate and expensive banquet experience. The transcription below uses the reconstruction of the basic stew recipe from the Shifang proposed by both Yao Lei and Gao Yizi, which involves reading three broken slips together:Footnote 58

256 … Once the meat has been soaked, use water and bitters to wash it. When the meat is white, then 195 strain the liquids in order to make a thick soup in the ding-cauldron, before bringing it to the fire. Souse 24 the meat and then stir the ding, before bringing it to the boil and skimming off the scum.Footnote 59 Once it is cooked, harmonise the flavours with fine wine, soybean sauce, and salt, and taste for seasoning. Once it is well-harmonised and the meat is cooked, the dish is complete. • Prepare a shoulder of beef and organ meats (?), cut into kebabs, [one illegible character] yellow…Footnote 60

256 …【肉】漬, 以水苦洒 [洗], 肉白【乃】195 清閭水以洎鼎, 富 [煏]。釀 24 肉撓之鼎, 㵒 [沸], 斟去其沫, 孰 [熟] 煮之, 和以美酒, 醬, 鹽, 嘗其和=, [和] 甘, 肉孰 [熟], 成。• 治牛肩[富+臽]Footnote 61 胾, □黄⋯

Zhang Chang’s discussion of the preparation and cooking methods described in the Shifang strongly supports reading zi as a form of kebab, since these recipes instruct the cook to cut up the meat into pieces and roast them on a spit.Footnote 62 When boiling, stewing, and other slow cooking methods are mentioned, it is not in contexts where it unequivocally pertains to zi. The form of cookery described in the Shifang is, however, deeply alien: the only flavourings available to the Han Dynasty gastronome were salt and an array of fermented pastes and sauces, together with ginger, which had originally been imported from Southeast Asia and quickly naturalised in Chinese cooking.Footnote 63 If Sichuan pepper was known, it is not mentioned in any surviving fragment of the text; likewise there appear to have been no sweeteners available, not even honey.Footnote 64 Although Wu Yang’s chefs no doubt made every effort with marinades and bastings, this would have been a salty, umami-rich, meaty cuisine, making the most of every cut from the animals.

Kebabs continued to be popular in China into the medieval period, with a number of recipes preserved in the Qimin yaoshu 齊民要術 (Essential Skills to Preserve the People) by Jia Sijie 賈思勰 (fl. 530s), an official of the Northern Wei Dynasty (386–535). The Qimin yaoshu is the earliest surviving cookbook in the transmitted tradition, containing 280 recipes, including a number of recipes for kebabs. Cooks are instructed to cut their meat into one-inch cubes (fang cun luan qie 方寸臠切) or perhaps a little larger, which were to be marinaded before being spit-roasted.Footnote 65 The marinades recommended in this text were broadly similar to those in use more than five centuries earlier and would have been recognised by the cooks of the Shifang. Such cookery was still a long way away from the modern globalised Xinjiang kebab, with its usual marinade consisting of Sichuan pepper, supplemented with cumin (a plant native to Iran) and chilli from the New World.

Conclusion

There was no word for a kebab in the language of ancient China, but that does not mean that the concept was alien to them or that they lacked a vocabulary to speak about this kind of meat cookery. Instead, they spoke about pieces of meat (luan and zi), and a method of cooking: roasting (zhi). Ample evidence has survived from early Chinese texts, both excavated and transmitted, about the cooking of kebabs, and this can be supplemented by shish kebabs found preserved in ancient tombs, illustrations of kitchen and banqueting scenes in early imperial era art, as well as excavated kitchen equipment and pottery tomb models which clearly depict meat that has been cut into pieces and roasted on skewers. These significantly predate any possible influence from the Middle East and are (to date) the oldest known examples of shish kebab cookery from anywhere in the world. Accordingly, this must represent an indigenous Chinese tradition of meat cooking, which subsequently became overlaid with the Middle Eastern tradition, probably as a result of Islamic influence from the seventh century CE onwards.

No surviving account of early Chinese kebab cooking explains why this dish was adopted so enthusiastically in this culinary tradition, but it is likely that as with Middle Eastern kebab cooking, it was the result of poor access to fuel. Essentially, fuel poverty in two different parts of the world led, in antiquity, to the development of very similar methods for preparing and cooking meat. Accordingly, difference was expressed in the form of marinades and basting ingredients, with the ancient Chinese tradition featuring soybean sauces and various other fermented pastes. The unfamiliarity of this form of cookery is likely to be a factor in why so many scholars have failed to recognise that these literary references pertain to kebabs. At the same time, thanks to the dominance of Middle Eastern kebab traditions in the modern world, this has been anachronistically projected back into remote antiquity. However, there is no need to seek an explanation involving cultural diffusion to explain the presence of shish kebabs in early China.

Conflicts of interest

None.

References

1 While it might be thought that cutting meat into small pieces for cooking is universal, this is not in fact the case, with Alan Davidson, The Oxford Companion to Food (Oxford, 1999), p. 442, suggesting that this practice was traditionally correlated with poor availability of fuel. As late as the time of the Macartney Embassy to China in 1793, there were vociferous complaints from the British contingent about this unfamiliar style of cookery; see Sir G. Staunton, An Authentic Account of an Embassy from the King of Great Britain to the Emperor of China (London, 1797), vol. ii, p. 5. Although visitors to the Middle East were almost universally enthusiastic about kebabs of every kind, information about this style of cookery was slow to spread; a topic extensively documented in Priscilla Mary Içın, Bountiful Empire: A History of Ottoman Cuisine (London, 2018). For an interesting example of bafflement on the part of a nineteenth-century American cook as to what a kebab could possibly be, analysing the description of ‘Moorish cubbub’ contained in James Riley’s (1777–1840) Sufferings in Africa; see Anonymous, The Cook not Mad, or, Rational Cookery (Watertown, NY, 1831), pp. 94–95.

2 For an early Chinese reference to the importance of shape in determining cooking terminology, the Jijiu pian 急就篇 [Quickly master characters] primer by Shi You 史游 (fl. 48–33 BCE), contains the line: ‘Stewed meat, finely sliced meat, spit-roasted meat, and kebabs each have their own shape’ (膹膾炙胾各有形); see Zhang Chuanguan 張傳官, Jijiu pian jiaoli 急就篇校理 [Quickly Master Characters annotated and arranged] (Beijing, 2017), p. 272.

3 For example, the Xunzi 荀子 says, ‘The laws of the land prohibit picking up abandoned items because of the concern that people will become accustomed to taking things without any consideration for social divisions’ (國法禁拾遺, 惡民之串以無分得也). Wang Xianqian 王先謙, Xunzi jijie 荀子集解 [Collected explanations of the Xunzi] (Beijing, 2008), p. 518 [‘Dalüe’ 大略]; see also the gloss on the character chuan in Guo Pu 郭璞 and Xing Bing 邢昺, Erya zhushu 爾雅注疏 [Approaching Elegance dictionary with commentary and subcommentary] (Beijing, 1999), vol. ii, p. 43.

4 Duan Yucai 段玉裁, Shuowen jiezi zhu 説文解字注 [Explaining Graphs and Analysing Characters with commentary] (Nanjing, 2007), p. 304 [‘Roubu’ 肉部].

5 Shuowen jiezi, p. 313 [‘Roubu’].

6 Arkadiusz Blaszczyk and Stefan Rohdewald (eds.), From Kebab to Ćevapčići: Foodways in (Post-)Ottoman Europe (Wiesbaden, 2018). Regional developments in the localisation of doner kebab in particular have been extensively documented in the research of Maren Möhring; see, e.g., Maren Möhring, ‘Döner Kebab and West German consumer (multi-)cultures’, in Hybrid Cultures–Nervous States: Britain and Germany in a (Post)Colonial World, (eds.) Ulrike Linder, Maren Möhring, Mark Stein, and Silke Stroh (Amsterdam, 2010), pp. 151–166; and Maren Möhring, ‘From Berlin to the globe: the transnational story of döner kebab’, in Routledge Handbook of Turkey’s Diasporas, (eds.) Ayca Arkilic and Banu Senay (London, 2024), pp. 446–458.

7 The first report of the excavation of this text, which included transcription of a couple of excerpts, was published in Hunansheng wenwu kaogu yanjiusuo 湖南省文物考古研究所, Huaihuashi wenwuchu 懷化市文物處, and Yuanlingxian bowuguan 沅陵縣博物館 (eds.), ‘Yuanling Huxishan yihao Hanmu fajue jianbao’ 沅陵虎溪山一號漢墓發掘簡報 [Preliminary report on the excavation of Han Dynasty tomb M1 at Huxishan, Yuanling], Wenwu 文物 [Cultural relics] 1 (2003), pp. 36–55. At this juncture, the text was known by the former title. It was renamed in the publication of all the manuscripts found in this tomb; see Hunansheng wenwu kaogu yanjiusuo 湖南省文物考古研究所 (ed.), Yuanling Huxishan yihao Hanmu 沅陵虎溪山一號漢墓 [Han Dynasty tomb M1 at Huxishan, Yuanling] (Beijing, 2020), pp. 145–153.

8 No archaeological report on the excavation of the Changle cemetery has yet been published, and work there has been ongoing since 2008; see Zhongguo kaoguxue hui 中國考古學會 (ed.), Zhongguo kaoguxue nianjian 中國考古學年鑒 [Yearbook of archaeology in China] (Beijing, 2009), pp. 450–451. The foodstuffs excavated from tomb M17 at this site have proved particularly interesting, with the scientific analysis (including discussion of the specific types of meat used to make the kebabs) detailed in Meng Ren, Renfang Wang, and Yimin Yang, ‘Diet communication on the early Silk Road in ancient China: multi-analytical analysis of food remains from the Changle Cemetery’, in Heritage Science 10.46 (2022), pp. 1–12, with the kebabs illustrated in fig. 2, pl. i. The authors of this study seem to believe strongly in the Middle Eastern origins of the kebab, perhaps reinforced by the presence of foreigners in this cemetery; see Zhou Yawei 周亞威 and Wang Renfang 王仁芳, ‘Ningxia Zhongwei Changle Hanmu chutu Ouluoba renzhong de shengwu kaoguxue kaocha’ 寧夏中衛常樂漢墓出土歐羅巴人種的生物考古學考察 [Bioarchaeological investigation into the Europeans excavated from Han tombs in Changle, Zhongwei, Ningxia], Kaogu yu wenwu 考古與文物 [Archaeology and cultural relics] 2 (2018), pp. 136–140.

9 Wang Xiaojuan 王曉娟, ‘Han caihui shinü kaorouchuan tu taozao lüekao’ 漢彩繪侍女烤肉串圖陶灶略考 [Research note on Han Dynasty painted stoves depicting serving maids grilling kebab skewers], Shoucang jie 收藏界 [Collector’s world] 5 (2019), p. 16.

10 Examples of such stone sculptures have been found at sites across Shandong Province, though a handful are also known from other parts of the empire; see Yu Qiuwei 于秋偉, ‘Handai huaxiangshi zhi ‘Baochutu’ fenxi’ 漢代畫像石之 ‘庖厨圖’ 分析 [Analysis of cooking scenes on Han Dynasty pictorial stones], Zhongguo bowuguan 中國博物館 [Journal of the National Museum of China] 2 (2010), pp. 105–110. Art historians have generally accepted that these illustrate kebab (rouchuan 肉串) cookery, though the assumption is usually also made that this represents Middle Eastern influence; see, e.g., Li Xin 李欣, ‘Kaogu ziliao suojian Handai “shaokao” fengsu’ 考古資料所見漢代 ‘燒烤’ 風俗 [Han Dynasty barbeque customs as depicted in archaeological materials], Sichuan wenwu 四川文物 [Sichuan cultural relics] 1 (2016), pp. 77–81; Yang Yanjun 楊艷軍, ‘Cong Jiyuan diqu chutu wenwu kan Handai de shaokao wenhua’ 從濟源地區出土文物看漢代的燒烤文化 [The barbeque culture of the Han Dynasty seen from excavated objects from the Jiyuan region], Jiyuan zhiye jishu xueyuan xuebao 濟源職業技術學院學報 [Journal of Jiyuan Vocational and Technical College] 9 (2018), pp. 24–28; and Wang Renxing 王仁興, ‘Han Jin Beichao kao rouchuan tuxiang yanjiu: jian ji dangdai kao rouchuan de shijie fenbu’ 漢晉北朝烤肉串圖像研究: 兼及當代烤肉串的世界分佈 [Research into illustrations of grilling kebab skewers from the Han, Jin, and Northern Dynasties: together with the global reach of kebabs in the modern world], Meishi yanjiu 美食研究 [Gastronomy research] 1 (2017), pp. 1–7.

11 See Renaud Soler, ‘Food and poetry: kebab imagery in Persian and Turkish poetry’, Diyâr 2 (2024), pp. 288–322; and Nawal Nasrallah, Annals of the Caliphs’ Kitchens: Ibn Sayyār al-Warrāq’s Tenth-Century Baghdadi Cookbook (Leiden, 2007), pp. 364–370, for the earliest recorded references to Middle Eastern kebabs. Much later in recorded history, Islamic foodways would indeed have a significant impact on Chinese eating from the Mongol period onwards, including a (re)introduction of kebabs, as evidenced by Yinshan zhengyao 飲膳正要 [Proper essentials in eating and drinking] by Hu Sihui 忽思慧 (fl. 1314–1330). This development is described in Paul D. Buell and Eugene N. Anderson, A Soup for the Qan: Chinese Dietary Medicine of the Mongol Era as Seen in Hu Szu-hui’s Yin-shan cheng-yao (London, 2000), pp. 62–80. For the wider influence of Near Eastern food on modern ideas of what it means to cook pieces of meat, with particular reference to the dominance of Arab or Persian terminology, see Blaszczyk and Rohdewald, From Kebab to Ćevapčići.

12 Ban Gu 班固, Hanshu 漢書 [History of the Han Dynasty] (Beijing, 1962), 99B:4191–4192.

13 As noted by Zhang Feng 張鳳, ‘Handai de zhi yu zhilu’ 漢代的炙與炙爐 [Han Dynasty roasting and cooking stoves], Sichuan wenwu 2 (2011), pp. 58–60, zhi 炙 referred to any meat that had been spit-roasted. This term was applied both to foodstuffs that were roasted whole and to kebabs placed upon skewers for cooking.

14 The legendary sword Gan Jiang was made by the eponymous swordsmith for King Helü of Wu 吳王闔閭 (r. 514–496 BCE); see, e.g., Yuan Kang 袁康 and Wu Ping 吳平, Yuejue shu 越絕書 [Lost history of Yue] (Shanghai, 1985), p. 80 [‘Ji baojian’ 記寶劍]; and Zhou Shengchun 周生春, Wu Yue chunqiu jijiao huikao 吳越春秋輯校匯考 [Spring and Autumn Annals of Wu and Yue, edited and collated, with associated research] (Shanghai, 1997), p. 40 [‘Helü neizhuan’ 闔閭内傳]. Accordingly, any reference to Gan Jiang in a story about Lord Wen of Jin is highly anachronistic. The other version of this story from the Han Feizi does not include this detail and sets these events in the context of a banquet for guests; see Chen Qiyou 陳奇猷, Han Feizi jishi 韓非子集釋 [Collected explanations of the Han Feizi] (Beijing, 1958), pp. 595–596 [‘Nei chushuo xia’ 内儲說下]. All subsequent quotations from this text refer to this edition, unless otherwise stated.

15 The translation of this line follows the commentary given by Zhang Jue 張覺, Han Feizi jiaoshu xilun 韓非子校疏析論 [Han Feizi with commentary, subcommentary, and detailed analysis] (Beijing, 2011), vol. ii, p. 622 n. 2.

16 Han Feizi, pp. 595–596 [‘Nei chushuo xia’]. The Han Feizi gives two slightly different versions of this story, and a further version is known only from early quotations of the Shuoyuan 說苑 [Garden of stories], though it does not form part of the standard transmitted text; see Xiang Zonglu 向宗魯, Shuoyuan jiaozheng 說苑校證 [Garden of Stories with commentary and verification] (Beijing, 2000), p. 545 [‘Yiwen’ 佚文].

17 See, e.g., Li Guilong 李貴龍 and Wang Jianqin 王建勤 (eds.), Suide Handai huaxiangshi 綏德漢代畫像石 [Han pictorial stones from Suide] (Xi’an, 2001), p. 96.

18 Han Feizi, p. 638 [‘Wai chushuo youshang’ 外儲說右上].

19 Guangzhoushi wenwu guanli weiyuanhui 廣州市文物管理委員會, Zhongguo shehui kexueyuan kaogu yanjiusuo 中國社會科學院考古研究所, and Guangdongsheng bowuguan 廣東省博物館 (eds.), Xi-Han Nanyue wangmu 西漢南越王墓 [Tomb of the Western Han Dynasty king of Nanyue] (Beijing, 1991), vol. i, pp. 78–79; and vol. ii, pl. xxxv.

20 Xun Yue 荀悅, Qian-Han ji 前漢紀 [Records of the Former Han] (Taipei, 1984), 9:288.

21 See Sima Qian 司馬遷, Shiji 史記 [Records of the Grand Historian] (Beijing, 1959), 57:2078; and Hanshu, 40:2061. Qierou is often translated as ‘sliced meat’ but this is not correct; slicing meat in ancient China was termed zhe 聶.

22 The problem of what exactly went wrong at this banquet in 143 BCE has exercised scholars since at least the early medieval period, with the Jijie 集解 [Collected explanations] commentary on the Shiji quoting remarks to this effect by Meng Kang 孟康 (fl. 220–254) and Ru Chun 如淳 (fl. 220–280); see Shiji, 57:2078 n. 3. Yan Shigu’s 顏師古 (581–645) analysis of these events is given in Hanshu, 40:2061–2062 n. 3. This topic is considered further in the light of the recipes contained within Wu Yang’s cookbook in Yao Lei 姚磊, ‘Han Jingdi de da kuai rou’ 漢景帝的大塊肉 [Emperor Jing of the Han Dynasty’s big pieces of meat], Dushu 讀書 [Reading] 7 (2021), p. 95.

23 A marvellous Han Dynasty carved stone from Shandong depicting the deity Xiwangmu 西王母, or Queen Mother of the West, being served shish kebabs makes it clear that meat cooked in this way was indeed served still on the skewer to be held by hand. See Wang Jishen 汪繼甚 (ed.), Han huaxiangshi xuan 漢畫像石選 [Selected pictorial stones of the Han Dynasty] (Shanghai, 2000), p. 80, pl. lxxxv. At the time of publication, this depiction was thought to depict fruit; subsequently, following further research into this work by Yang Zhishui 楊之水, ‘Shuo “Shaoyao zhi he”’ 說 “勺藥之和” [On ‘harmonising ingredients’], Zhongguo lishi wenwu 中國歷史文物 [Journal of the National Museum of History] 2 (2004), pp. 51–63, it has become widely accepted that this stone sculpture shows the deity about to eat kebabs.

24 Wu Zeyu 吳則虞, Yanzi chunqiu jishi 晏子春秋集釋 [Collected explanations of the Spring and Autumn Annals of Master Yan] (Beijing, 1962), p. 396 [‘Neipian zaxia’ 内篇雜下]. For a translation of this story, see Olivia Milburn, The Spring and Autumn Annals of Master Yan (Leiden, 2016), p. 350.

25 This particular aphorism appears twice in this text; see He Ning 何寧, Huainanzi jishi 淮南子集釋 [Collected explanations of the Huainanzi] (Beijing, 1998), pp. 1157 [‘Shuishan xun’ 說山訓], 1193 [‘Shuilin xun’ 說林訓]. These chapters are generally agreed to have been written in imitation of the collections of aphorisms found in the two ‘Shuilin’ 說林 [Forest of persuasions] chapters of the Han Feizi; see John S. Major, Sarah A. Queen, Andrew Seth Meyer, and Harold D. Roth, The Huainanzi: A Guide to the Theory and Practice of Government in Early Han China (New York, 2010), pp. 617–624. For the original text, see Han Feizi, pp. 417–479 [‘Shuilin’].

26 Zheng Xuan 鄭玄 and Jia Gongyan 賈公彥, Zhouli zhushu 周禮注疏 [Rites of Zhou with commentary and subcommentary] (Beijing, 1999), pp. 95–96 [‘Tianguan’ 天官; ‘Zhongzai’ 冢宰; ‘Xiangren’ 享人].

27 Chen Qiyou 陳奇猷, Lüshi chunqiu xin jiaoshi 呂氏春秋新校釋 [Spring and Autumn Annals of Master Lü, newly collated and explained] (Shanghai, 2002), p. 944 [‘Chajin’ 察今]. Most commentators on this text have accepted that lie 脟 is a synonym for luan 臠. However, no transmitted or excavated text suggests that luan or zi were ever served in a ding; furthermore, the same commentators agree that pieces of roasted or grilled meat were always served in a dou 豆. This contradiction goes unexplained.

28 Kong Yingda 孔穎達, Mao Shi zhengyi 毛詩正義 [Correct interpretations of the Mao tradition of the Book of Songs] (Beijing, 1999), p. 1413 [Mao 300: ‘Bigong’].

29 Zheng Xuan 鄭玄 and Jia Gongyan 賈公彥, Yili zhushu 儀禮注疏 [Rites and Ceremonies with commentary and subcommentary] (Beijing, 1999), p. 920 [‘Shaolao kuishi li’ 少牢饋食禮]. For a complete translation of this passage, see John Steele, The I-li, or, Book of Etiquette and Ceremonial (London, 1917), vol. ii, p. 170.

30 The importance of the stew in early Chinese cookery is discussed in Gilles Boileau, ‘Some ritual elaborations on cooking and sacrifice in Late Zhou and Western Han texts’, in Early China 23 (1998), pp. 89–113.

31 Finely diced meat or fish (kuai 膾) was traditionally served raw; see Xiao Fan 蕭璠, ‘Zhongguo gudai de shengshi roulei xizhuan: kuaisheng’ 中國古代的生食肉類餚饌: 膾生 [Kuaisheng: an ancient Chinese raw meat delicacy], Zhongyang yanjiuyuan lishi yuyan yanjiusuo jikan 中央研究院歷史語言研究所集刊 [Academia Sinica Bulletin of the National Research Institute of History and Philology] 6 (2000), pp. 247–365; and Dong Yuejin 董躍進, ‘Gudai yukuai de yuanliao he chifa’ 古代魚膾的原料和吃法 [The components and consumption of raw fish in antiquity], Yangzhou daxue pengren xuebao 揚州大學烹飪學報 [Yangzhou University journal of cookery] 26.1 (2009), pp. 31–33.

32 Zheng Xuan’s commentary on the Zhouli provides an account of how this fermented meat paste was made: ‘You must first dry your meat and then hash it, before mixing it with yeast and salt, sprinkling it with fine wine, and mashing it into a jar: after 100 days it will be complete’ (必先膊乾其肉, 乃後莝之, 雜以梁麴及鹽, 漬以美酒, 塗置甀中: 百日則成矣); see Zhouli, p. 138 [‘Tianguan’ 天官; ‘Zhongzai’ 冢宰; ‘Xiren’ 醯人]. This was considered an indispensable condiment if minced meat was served. In this article, use of the term ‘soybean sauce’ for jiang 醬 is preferred to avoid any confusion with modern soy sauce.

33 When this passage has been translated, various suggestions have been offered for jiang 漿, including syrup, broth, and vinegar; see James Legge, The Sacred Books of China: The Texts of Confucianism: The Lî Kî (Oxford, 1885), p. 79; Roel Sterckx, ‘Sages, cooks, and flavours in Warring States and Han China’, in Monumenta Serica 54 (2006), pp. 1–46; and David R. Knechtges, ‘A literary feast: food in early Chinese literature’, in Journal of the American Oriental Society 106.1 (1986), pp. 49–63, respectively. David Knechtges is correct in identifying jiang as a form of vinegar, but instead of being a condiment, this was a drink. Accordingly, it has been translated here as posca, which was the Roman term for a drinking vinegar.

34 Kong Yingda 孔穎達, Liji zhengyi 禮記正義 [Correct interpretations of the Record of Ritual] (Beijing, 1999), pp. 56–57 [‘Quli shang’ 曲禮上].

35 Commentators on this passage have expended extensive efforts in attributing the position of each setting in a yin-yang configuration, as well as arguing about exactly what shape of serving vessel these dishes were presented in; see, e.g., Sun Xidan 孫希旦, Liji jijie 禮記集解 [Collected explanations of the Record of Ritual] (Beijing, 2007), pp. 51–52 [‘Quli shang’]. This is in spite of neither topic being mentioned in the original text of the Liji.

36 Different possible table layouts are discussed in detail in Gao Jiyi 郜積意 and Zeng Xingui 曾新桂, ‘Lijing Zheng zhu “zhuan yao fang” tushi’ 禮經鄭注 “饌要方” 圖釋 [An illustrated explanation of Zheng Xuan’s commentary on the ritual classics that ‘food should be laid out in a square’], Xueshu yuekan 學術月刊 [Academic monthly] 55.1 (2023), pp. 173–190.

37 Yili, p. 490 [‘Gong shi dafu li’ 公食大夫禮]. For an alternative translation of this passage; see Steele, I-li, vol. i, p. 251.

38 Liji, p. 848 [‘Neize’].

39 Sun Yirang 孫詒讓, Mozi xiangu 墨子閒詁 [Exposing and correcting the Mozi] (Beijing, 2001), pp. 164–166 [‘Jieyong zhong’ 節用中]. For an analysis of how this development in the persona of Yao relates to other classical masters’ texts in the context of a creative engagement with this character, see An Haimin 安海民, Chen Juan 陳娟, Yue Ting 岳婷, and Tang Xiaomeng 唐肖萌, ‘Xian-Qin zhuzi shuo Yao’ 先秦諸子說堯 [Pre-Qin philosophers’ accounts of Yao], Qinghai shifan daxue xuebao (Zhexue shehui kexueban) 青海師範大學學報 (哲學社會科學版) [Journal of Qinghai Normal University (Philosophy and Social Sciences edition)] 6 (2016), pp. 66–74.

40 For studies of this important text concerning education in community living, including the vexed issue of dating, see Zhang Liangcai 張良才, ‘Cong Guanzi “Dizi zhi” kan Jixia xuegong de jiaoxue yu shenghuo guanli’ 從管子 “弟子職” 看稷下學宮的教學與生活管理 [Viewing the teaching and organisation of daily life in the Jixia Academy from the perspective of the ‘Duties of the disciple’ chapter of the Guanzi], Guanzi xuekan 管子學刊 [Guanzi journal] 3 (1994), pp. 39–42; Chen Qiang 陳強, ‘Lun Jixia xuegong xuesheng jiaoyu yu guanli de guifan huafanshi: yi Guanzi “Dizi zhi” wei shijiao’ 論稷下學宮學生教育與管理的規範化範式: 以管子 “弟子職” 為視角 [On the standardised paradigms of student education and management in the Jixia Academy: from the perspective of the ‘Duties of the disciple’ chapter of the Guanzi], Guanzi xuekan 2 (2017), pp. 40–42; and Oliver Weingarten, ‘What did disciples do? “Dizi” 弟子 in early Chinese texts’, Harvard Journal of Asiatic Studies 75.1 (2015), pp. 29–75.

41 For references to the ‘Dizi zhi’ as an independent text in antiquity, see Hanshu, 30:1718. For modern printings with commentary, see, e.g., Hong Liangji 洪亮吉, Dizhi zhi jianshi 弟子職箋釋 [Compiled explanations of the ‘Duties of the disciple’] (Beijing, 2000); and Gui Wencan 桂文燦, Dizi zhi jiegu 弟子職解詁 [Expositions on the ‘Duties of the disciple’] (Guangzhou, 2015). In addition to being included in the Guanzi, the ‘Dizi zhi’ with a commentary by Zhu Xi 朱熹 (1130–1200) was also sometimes appended to the Xiaojing 孝經 [Classic of filial piety]; see, e.g., Ren Zhaolin 任兆麟, Xiaojing: fu Dizi zhi 孝經: 附弟子職 [Classic of Filial Piety with the ‘Duties of the disciple’ appended] (Changsha, 1938).

42 Cuo 錯 is here being read as a loan word for cuo 措, ‘to put in position’ or ‘serve’. This meaning is attested to in numerous commentaries; see, e.g., Kong Yingda 孔穎達, Zhou Yi zhengyi 周易正義 [Correct interpretations of the Zhou Book of Changes] (Beijing, 1999), p. 277 [‘Xici shang’ 繫辭上].

43 In his translation of this passage, W. Alleyn Rickett assumes that caigeng 菜羹 refers to a single dish: ‘vegetable stew’; W. Alleyn Rickett, Guanzi: Political, Economic, and Philosophical Essays from Early China (Princeton, 1998), pp. 287–288. However, it is more likely that this denotes two separate dishes: vegetables that were consumed first and then a meat stew.

44 As numerous commentators have pointed out, jiang 醬 in this line of the poem should be read jiang 漿 (i.e. posca), since this refers to the diner being given a choice of drinks; see, e.g., Hong Yixuan 洪頤煊, Guanzi yizheng 管子義證 [Guanzi interpreted with verification] (Shanghai, 1995), p. 550 [‘Dizi zhi’]; and Xu Weiyu 許維遹, Guanzi jijiao 管子集校 [Collected commentaries on the Guanzi] (Beijing, 1956), p. 959 [‘Dizi zhi’].

45 Li Xiangfeng 黎翔鳳, Guanzi jiaozhu 管子校注 [Guanzi annotated with commentary] (Beijing, 2004), p. 1146 [‘Dizi zhi’].

46 Shiji, 7:316 and 8:380; Hanshu, 34:1894.

47 Wu Chen would successfully navigate the rebellion of Ying Bu in 195 BCE, with the able assistance of his prime minister, Li Cang 利蒼 (d. circa 186 BCE); see Shiji, 19:978–979; and Hanshu, 16:618. Li Cang, his wife, and son were buried in the famous tombs at Mawangdui 馬王堆, ensuring that the artistic and cultural legacy of the kingdom of Changsha bulks large in understandings of elite Han Dynasty life. For the full excavation reports on these magnificent tombs, see Hunansheng bowuguan 湖南省博物館 and Zhongguo kexueyuan kaogu yanjiusuo 中國科學院考古研究所 (eds.), Changsha Mawangdui yihao Hanmu 長沙馬王堆一號漢墓 [Han Dynasty tomb M1 at Mawangdui, Changsha] (Beijing, 1973); and Hunansheng bowuguan 湖南省博物館 and Hunansheng wenwu kaogu yanjiusuo 湖南省文物考古研究所 (eds.), Changsha Mawangdui er sanhao Hanmu 長沙馬王堆二三號漢墓 [Han Dynasty tombs M2 and M3 at Mawangdui, Changsha], two vols (Beijing, 2004).

48 Shiji, 19:985; Hanshu, 16:618.

49 For a pioneering study of Han Dynasty cuisine, with particular focus on the ingredients favoured by the titled urban elite who were in a position to be recorded in literature and buried in lavish tombs with paintings and texts, see Hayashi Minao 林巳奈夫, ‘Kandai no inshoku’ 漢代の飲食 [Han Dynasty food and drink], Tōhō gakuhō 東方學報 [Journal of oriental studies] 48 (1975), pp. 1–98. For a parallel ingredients-focused analysis of the Shifang, see Hu Qi 胡綺, ‘Cong Huxishan Hanjian Shifang guankui Handai yinshi wenhua’ 從虎溪山漢簡食方管窺漢代飲食文化 [Exploring Han Dynasty food culture from the Huxishan bamboo slip Food Recipes manuscript], Hanzi wenhua 漢字文化 [Chinese script and culture] 21 (2022), pp. 168–174.

50 See, e.g., Yang Bojun 楊伯峻, Chunqiu Zuozhuan zhu 春秋左傳注 [Zuo’s Tradition of the Spring and Autumn Annals with commentary] (Beijing, 1990), pp. 182–183 [Zhuang 10]; and Shuoyuan, p. 271 [‘Shanshui’ 善說].

51 The first study of this text in English can be found in Donald Harper, ‘The cookbook and gastronomy in ancient China: the evidence from Huxishan and Mawangdui’, Hunansheng bowuguan guankan 湖南省博物館館刊 [Journal of Hunan Provincial Museum] 1 (2004), pp. 164–177. For references to its significance in the history of Chinese food, see Roel Sterckx, ‘Food and philosophy in early China’, in Of Tripod and Palate: Food, Politics, and Religion in Traditional China, (ed.) Roel Sterckx (New York, 2004), pp. 34–61. There is one thesis so far on this text: see Zhang Chang 張暢, ‘Yuanling Huxishan Hanjian Shifang jishi ji xiangguan wenti yanjiu’ 沅陵虎溪山漢簡食方集釋及相關問題研究 [Collected explanations of the Han Dynasty bamboo slip manuscript Food Recipes from Huxishan in Yuanling and research into related questions] (unpublished MA dissertation, Wuhan University, 2022). In addition to a detailed online commentary on the text produced by Yao Lei, there is a study of the plants mentioned; see Han Yijie 韓亦傑, ‘Huxishan Hanjian Shifang zhong suojian zhiwu zashi’ 虎溪山漢簡食方中所見植物雜識 [Identifying plants seen in the Han Dynasty bamboo slip manuscript Food Recipes from Huxishan], Zhongguo wenzi 中國文字 [Chinese script] 7 (2022), pp. 249–261; and a detailed analysis produced by Gao Yizhi 高一致 and Huang Kaixin 黃開心, ‘Huxishan Hanjian Shifang kaoshi congzha’ 虎溪山漢簡食方考釋叢札 [Notes on research and explanations of the Han Dynasty bamboo slip manuscript Food Recipes from Huxishan], Chutu wenxian 出土文獻 [Unearthed documents] 3 (2024), pp. 92–100.

52 The final phrase in this sentence is very controversial. This translation follows Gao Yizhi 高一致, ‘Shishuo Huxishan Hanjian Shifang zhong “fu” ji xiangguan neirong’ 試說虎溪山漢簡食方中 “富” 及相關內容 [An attempt to explain the term fu in the Han Dynasty bamboo slip manuscript Food Recipes from Huxishan and related content] (www.bsm.org.cn/?hanjian/8376.html; accessed 29 July 2025), but an alternative translation is proposed by Shi Guangze 石光澤, ‘Huxishan Hanjian Shifang zhaji erze’ 虎溪山漢簡食方札記二則 [Two notes on the Han Dynasty bamboo slip manuscript Food Recipes from Huxishan] (www.bsm.org.cn/?hanjian/8388.html; accessed 29 July 2025), which would read this as ‘use your hands to thread [the meat pieces] onto skewers’.

53 Hunansheng wenwu kaogu yanjiusuo, Yuanling Huxishan yihao Hanmu, p. 145. The original transcription says the last five characters are illegible. However, Gao Yizhi, ‘Shishuo Huxishan Hanjian Shifang zhong “fu” ji xiangguan neirong’, argues that the first two graphs can in fact be read, and this amendment has been generally accepted.

54 Hunansheng wenwu kaogu yanjiusuo, Yuanling Huxishan yihao Hanmu, p. 146.

55 This follows the gloss by Zheng Xuan quoted in Liji, p. 669 [‘Liyun’ 禮運], which defined this word as meaning ‘to spear [meat] on a spit above the fire’ (貫之火上). Guan 貫 indicates that the meat was pierced right through, and since zi refers to a larger piece than luan, it would seem reasonable to imagine a spit rather than a skewer.

56 Yao Lei 姚磊, ‘Huxishan Hanjian Shifang de bianlian paixu’ 虎溪山漢簡食方的編連排序 [The compilation and ordering of the Han Dynasty bamboo slip manuscript Food Recipes from Huxishan] (www.bsm.cn/?hanjian/8354.html; accessed 29 July 2025), discusses all surviving recipe titles from this text, noting the division of the text into kebab dishes, stews, congees, grills, rice dishes, and perhaps brewing. The text of these slip heads can be found in Hunansheng wenwu kaogu yanjiusuo, Yuanling Huxishan yihao Hanmu, pp. 149–151.

57 Gao Yizhi and Huang Kaixin, ‘Huxishan Hanjian Shifang kaoshi congzha’, pp. 92–93, argues that the term rui 芮 in the Shifang pertains to marinades, which was a particularly important aspect of the cooking documented in this text.

58 See Yao Lei 姚磊, ‘Du Huxishan Hanjian Shifang zhaji (er)’ 讀虎溪山漢簡食方札記 (二) [Reading notes on the Han Dynasty bamboo slip manuscript Food Recipes from Huxishan (II)] (www.bsm.org.cn/?hanjian/8357.html; accessed 29 July 2025); and Gao Yizhi, ‘Shishuo Huxishan Hanjian Shifang zhong “fu” ji xiangguan neirong’, respectively.

59 Niang is apparently used in the Shifang to refer to sousing a dish; e.g. both slips 46 and 97 give instructions for stuffing a game bird (a swan and a goose, respectively) with glutinous rice, fermented meat and soybean paste, and ginger, then sousing it before boiling it, and afterwards harmonising the flavours with fine wine, soybean sauce, salt, and jujubes; see Hunansheng wenwu kaogu yanjiusuo, Yuanling Huxishan yihao Hanmu, p. 146 and p. 148. This dish is very reminiscent of Korean samgyetang 蔘鷄湯. Zhang Chang, ‘Yuanling Huxishan Hanjian Shifang jishi ji xiangguan wenti yanjiu’, p. 95, incorrectly suggests that niang in this context means ‘stuffing’.

60 Hunansheng wenwu kaogu yanjiusuo (ed.), Yuanling Huxishan yihao Hanmu, pp. 152, 150, and 146. Strip 94 gives a very similar recipe, also with a side dish or garnish that involved kebabs made from a shoulder of lamb; see ibid, p. 148.

61 The meaning of this character is unknown, but since the word before it denotes the stomach, it is likely to refer to some other kind of organ.

62 Zhang Chang, ‘Yuanling Huxishan Hanjian Shifang jishi ji xiangguan wenti yanjiu’, p. 104.

63 According to the numerical analysis of Hu Qi, ‘Cong Huxishan Hanjian Shifang guankui Handai yinshi wenhua’, pp. 170–171, ginger is mentioned 27 times in the Shifang, making it the most important flavouring after soybean sauce and salt. Ginger was undoubtedly a particularly important spice in pre-unification cookery that had been enthusiastically adopted by Chinese eaters; the first mention of ginger in any language is found in the Lunyu 論語 [Analects of Confucius], which informs us that the Master ate it regularly but was careful to consume it only in moderation. Yang Bojun 楊伯峻, Lunyu yizhu 論語譯注 [Analects of Confucius with commentary and translation into modern Chinese] (Beijing, 1980), p. 103 [10.8: ‘Xiangdang’ 鄉黨].

64 The Eastern Han Dynasty dictionary Shiming 釋名, thought to have been composed in around 200 CE, does include a reference for how to make meat stuffing flavoured with Sichuan pepper: ‘Han means “stuffing”. You stuff your roast with finely minced meat that has been flavoured with ginger, Sichuan pepper, salt, and black beans, and once the meat has been filled up inside, then you close and roast it’ (䐄, 銜也。銜炙細密肉和以薑椒鹽豉巳乃以肉銜裹其表而炙之也). See Liu Xi 劉熙, Shiming 釋名 [Explanation of names] (Beijing, 2016), p. 59 [‘Shi yinshi’ 釋飲食]. By the early medieval period, han denoted a kind of kofta or meatball, flavoured with ingredients including ginger, Sichuan pepper, orange peel, and celery, and cooked on skewers; see Miao Qiyu 繆啓愉, Qimin yaoshu jiaoshi 齊民要術校釋 [Essential Skills to Preserve the People with commentary and explanation] (Beijing, 1998), 9:622.

65 Qimin yaoshu, 9:616.