From conspicuous consumption to discourses of connoisseurship

A wider use of imported aromatics, mostly in the form of wood and resin, is perhaps the most noticeable change in the Chinese sensorium since the arrival of Buddhism. As a point of comparison, the aromatics buried in the Han-Dynasty Mawangdui tomb (circa 150 BCE), which belong to the pre-Buddhist olfactory world, were all indigenous in origin.Footnote ⁹ In the Tang era, aromatics were part of a broader vogue for exotic imports that also included foreign clothes, food, and music, as Edward Schafer details in his book, The Golden Peaches of Samarkand.Footnote ¹⁰ Foreign aromatics arrived in China on ships. In 743, when the Buddhist Monk Jianzhen 鑒真 (688–763) was preparing for his voyage to Japan in the port city of Guangzhou 廣州, he saw ‘innumerable’ foreign ships anchored in the Pearl River 珠江, laden with ‘mountainous piles’ of aromatic drugs (香藥), pearls, and other rare and precious goods, and the vessels were said be from India (婆羅門), Persia (波斯), Malaya (崑崙山), and Arabia (大食), each 60 to 70 feet deep.Footnote ¹¹ Jianzhen's eye-witness inventory speaks to the voluminous transport of precious commodities on maritime trade routes. He may have paid close attention to these ships because he himself purchased close to 1,000 jin of aromatics that year to be transported via ship to Japan. The aromatics he purchased included musk (麝香), agarwood (沉香), armour aromatic (甲香),Footnote ¹² dragon brain aromatic (龍腦香), benzoin (安息香), land-infusing aromatic (熏陆香), and others.Footnote ¹³ Aromatics are compact, dense, and often resinous and therefore less perishable, thereby lending themselves towards long-distance transport on vessels moving across the Indian Ocean. This is also evidenced by another encounter in Jianzhen's experience: a pirate based in Hainan Island, Feng Ruofang 馮若芳 (n.d.), who was said to have pillaged the cargoes and crew of ‘two or three Persian ships each year’, entertained his guests by extravagantly ‘burning frankincense as if they are candles’, 100 catties at a time.Footnote ¹⁴ This testifies to the pirate's wealth in general, but also to his unmitigated access to frankincense on board these passing ships in the South Sea.

Recent studies related to Jianzhen's eye-witness account in the coastal port and on Hainan Island have shown that a maritime trade route from the Persian Gulf to China via the Indian Ocean experienced a ‘permanent upswing’ in the late Tang to early Song Dynasties; of all the commodities that entered China via the sea, aromatic drugs (xiangyao香藥) were the most significant in value and quantity.Footnote ¹⁵ A study by Wen Cuifang concludes that, in the medieval period, it was aromatics rather than metal currency that constituted a major import commodity that helped balance the export of Chinese silk.Footnote ¹⁶

At its farthest reach, a maritime trade route connected the (inland) capitals of the Chinese empire, Chang'an, and the caliphate seat of Bagdad, as well as many port cities along the South Sea in Sumatra, India, and Ceylon. In official histories such as the ‘geography’ (dili 地理) category of the New Tang History, the grand councillor Jia Dan 賈耽 (729–805) recorded ‘The route to foreign countries across the sea from Guangzhou’ (廣州通海夷道), which begins with the port city on China's south coast.Footnote ¹⁷ This recorded route is consistent with those described in Arabic sources such as The Book of Routes and Realms by Ibn Khurdādhbih (circa 825–911) and the anonymous Accounts of China and India compiled in 851.Footnote ¹⁸ These sea routes have been corroborated by recent archaeological finds as well. Excavated Chinese pottery dating to the eighth to tenth centuries dot the coastline of the Indian Ocean, the Bay of Bengal, and the ports of Indonesia, attesting to locations that received exports of Chinese ceramics.Footnote ¹⁹ Two recently excavated shipwrecks near Indonesia also testify to this maritime traffic in the ninth and tenth centuries. The Belitung (circa 850), a ship that carried over 60,000 pieces of Tang ceramics on its outbound leg, sank near Sumatra.Footnote ²⁰ The Chinese ceramic cargoes of the Intan (circa 940) are, similarly, believed to have been intended for exchange with commodities from Southeast Asia, including aromatics in the form of resins and wood.Footnote ²¹ For example, agarwood or ‘sinking fragrance’ (chenxiang 沉香) was imported into China as both an ingredient for blended incense as well as a material that could be carved into religious objects of devotion.Footnote ²²

According to Yan Liu's count, the state-commissioned pharmacopoeia compiled in 659, Newly Revised Materia Medica (Xinxiu bencao 新修本草), describes 11 kinds of aromatics of foreign origin, such as agarwood, frankincense, patchouli, storax, dragon brain, benzoin, sandalwood, and armour aromatic.Footnote ²³ Foreign aromatics also figured prominently in personal perfumes, which were burned in censers and fumigated onto clothing. In Tang poems from the ninth century, aristocratic youths are described as wearing clothing fumigated with fragrance from far-flung places. The poet Zhang Xiaobiao 章孝標 (791–873) describes a day in the life of such a young man, and he begins with the couplet: ‘At dawn, he leaves the barracks for a quick hunt/ Foreign fragrances permeate his sleeves’ (平明小獵出中軍，異國名香滿袖熏).Footnote ²⁴ We find a similar depiction in a ‘Ballad of a young man of Chang'an’ (長安少年行) by Li Kuo 李廓 (d. 851), which begins with the couplet: ‘Wearing a Yangzhou-styled hat aslant/ heavily infused with fragrances from foreign lands’ (剗戴揚州帽，重熏異國香).Footnote ²⁵

Imported aromatics also played an increasing role in the production of incense during the Tang era, constituent to an important type of religious commodity. A recipe for blended incense associated with Huadu 化度 Temple in the north-west portion of Chang'an survives from the Song-Dynasty treatise on aromatics, Xiang pu 香譜; it calls for one part dragon brain and one part musk (麝香), along with more copious amounts of agarwood (沈香), white sandalwood (白檀香), storax (蘇合香), and armour aromatic (甲香).Footnote ²⁶ Of the ingredients listed in the recipe, most are of foreign origin. In an entry about aromatics in a tenth-century prose collection Qing yi lu 清異錄 (Records of the Pure and Marvellous), we see an explicit statement of a preference for imports. The entry is entitled ‘Beggar child's fragrance’ (乞兒香):

Having now arrived in China from places such as Sumatra, Borneo, or parts of present-day Vietnam, these aromatics were beginning to be catalogued, evaluated, gifted, displayed, compared, and competed against in the hands of the elite. Because imported aromatics were costly, anecdotes that describe their consumption in large quantities in effect showcase a wilful act of conspicuous consumption. For example, in Tang anecdotes, Emperor Yangdi of the Sui 隋煬帝 (569–618) is said to have burned ‘fiery mountains’ of agarwood as his nightly entertainment. Similar behaviour is often recounted as typical of doomed last emperors, and was used as pointed admonition to Emperor Taizong 太宗 (r. 626–49) by his officials.Footnote ²⁹ Two centuries later, when a Persian merchant named Li Susha 李蘇沙 presented Emperor Jingzong 敬宗 (809–27, r. 824–27) with materials for a pavilion made of agarwood, an official again remonstrated against such opulence, pointedly comparing it to building terraces made of jasper (不異瑤臺瓊室)Footnote ³⁰—a reference with a specific and time-worn association with dynastic decline.

Not all acts of wanton extravagance were from benighted rulers however; some were also from the uber-riche in the capital of Chang'an, such as a Tang merchant Wang Yuanbao 王元寶. The following record similarly illustrates the extreme of this consumptive craze for aromatics:

Similarly, Yang Guozhong 楊國忠 (d. 756), who ascended to power through his family ties to the royal consort Yang Guifei 楊貴妃 during the reign of Xuanzong 玄宗 (r. 712–56), was considered a parvenu of a different sort. In an entry entitled ‘The four-fragrance pavilion’ from the same collection, the Kaiyuan Tianbao yishi, his extravagance with expensive aromatics is described as overreaching propriety:

While the entries cited above show the historians’ contempt toward the merchants and power usurpers, what is less discussed is the way in which they also reveal an anxiety toward an unbridled consumption of expensive aromatics. As records of the ‘misuse’ of aromatics, they point to a tension or slippage in socially perceived value and their deployment: Emperor Yangdi of the Sui augured dynastic collapse with his use of agarwood, while Yang Guozhong and Wang Yuanbao, parvenu and merchant, consumed aromatics in amounts that were clearly deemed ‘wrong’. Yet their behaviour was not defined by sumptuary codes and fell into a no man's land of social regulation. As Arun Appadurai points out in his article on commodities: ‘Sumptuary laws constitute an intermediate consumption-regulating device, suited to societies devoted to stable status displays in exploding commodity contexts, such as India, China, and Europe in the premodern period’.Footnote ³³ Yet another crucial observation that we should make is that sumptuary laws are inherently designed for the optics of consumption, such as clothing, carriage, and the sizes of dwellings. What we see in these Tang-era anecdotes is that consumption of olfactory luxury goods was noticed as extravagance and transgression, yet it curiously escapes the social regulations per se. Even though agarwood-scented banisters displayed immense wealth and status, aromatics as such were in fact not regulated by Tang sumptuary laws that regulated other material goods that signalled social status. Thus, we see in these anecdotes the unstable cultural valuation of a relatively new luxury commodity for which there were no established protocols for proper measures of consumption, such that these ‘wrong’ ways of consuming aromatics fall into the interstices of social regulation. Yang Guozhong, the merchant Wang Yuanbao, and the pirate Feng Ruofen, who burned mounds of frankincense, are each represented as failing to practise the ‘right’ way to consume these expensive foreign aromatics.

As the art historian Craig Clunas observes of sumptuary regulations: ‘they begin with categories of person which are assumed to be immutable and then assign to those categories specific types of object’. This contrasts with the aspirational drives of connoisseurship-related texts, which, he notes, ‘begin invariably with types of thing, specifically described, and then relate to them essentially social categories like “vulgar” or “elegant”’.Footnote ³⁴ This is precisely what Tang elites had begun to do with aromatics, and imported aromatics in particular, as evidenced in anecdotal writing that reveals the ways in which they self-consciously cultivated an appreciation for aromatics to distinguish themselves from the negative examples of consumption cited above. One valuable source for this burgeoning discourse of connoisseurship can be found in the genre of biji 筆記 (notebook jottings), and in particular the aforementioned Qing yi lu 清異錄 attributed to Tao Gu 陶穀 (903–73), who was active during the Five Dynasties. The collection consists of 661 entries of short anecdotes about people and events throughout the Tang and Five DynastiesFootnote ³⁵ organised into 39 categories (門); they are ordered hierarchically and in a fashion that is consistent with leishu (categorical books or ‘encyclopaedia’) from this time. As is conventional for the leishu genre, the categories begin with astronomy (天文) and geography (地理) and move on to terrestrial plants, including grasses and trees (草木), varieties of fruit (百果), and vegetables (蔬菜), followed by interior decorations (陳設), wine and ale (酒漿), tea (茗荈), fine food (饌羞), and others, before ending with the liminal categories of ghosts (鬼), spirits (神), and demons (妖).Footnote ³⁶

Entries in Qing yi lu pertaining to aromatic practices are rich and numerous, and have not been adequately studied.Footnote ³⁷ They are primarily found under the category of ‘fumigating and burning’ (燻燎), with 24 entries detailing aromatics, incense, and aromatic woods. Just as importantly, entries relating to (mostly domestic) aromatics are also scattered among other categories such as ‘plants’, ‘flowers’, and ‘fruits’, which discuss the botanical ingredients and their olfactory features. For example, in the fruit category is an entry about the fragrance of melons, vividly entitled ‘Choosing with the nose’ (鼻選).Footnote ³⁸

There are several entries in Qing yi lu that describe olfactory practices through a connoisseurial lens, manifested as an implicit or explicit emphasis on rankings and lineages of scents. For example, under the category of ‘plants’ (草木), the lan 蘭 flower is crowned as a ‘master fragrance’ (香祖) by the intensity and durability of its aroma:

This awareness of a lineage, hierarchy, or ranking for the quality of scents is evident in other entries as well. An important part of connoisseurship is the process of ranking and evaluation, in the form of competitions. Qing yi lu describes an aromatics competition called ‘The Fragrance Tournament’ (鬬香):

This competition seems to have been a recurring event among the households that were all imperial affines and therefore at the top echelon of Tang society. We do not know the identity of the aromatic that emerged as the champion in that Qing yi lu entry, but the passage suggests that it came from a privileged source, most likely through an imperial tribute. Interestingly, the term for empress—jiaotu 椒塗—is itself related to an aromatic, the Sichuan pepper (椒), which is native to China. Starting in the Han Dynasty, ‘the pepper chamber’ (jiaofang 椒房) began to be used to refer to the living quarters of the empress and royal consorts, as the Sichuan pepper was incorporated into the plaster on the walls due to its warming properties and suggestion of fertility.Footnote ⁴²

Comparing rare aromatics and establishing a ranked list constitutes but one form of connoisseurship pertaining to aromatics; another is that of comprehensive display. An entry in Qing yi lu documents an ‘aromatics banquet’ (香燕) in which various specimens are showcased for appreciation, this time in the court of Emperor Yuanzong 元宗 (r. 943–60) of the Southern Tang (937–75):

This olfactory pageantry emphasises the variety and foreign origin of the aromatics as well as their versatility (as drugs, as perfume, etc.). During a time at which regional courts were vying for imperial ambitions, the display in turn creates a kind of spectacle to advertise its power beyond its own geographical limits.

During the late medieval era, aromatics were also being appreciated in the private homes of the gentry class. It is here that we come across activities that emphasise selection, discernment, and private valuation, all of which constitute what Craig Clunas calls the ‘invention of taste’.Footnote ⁴⁴ An entry entitled the ‘Fivefold suitability’ (五宜) describes another form of connoisseurial activity: the art of finding the right aromatic to blend with fragrances from flowers. It features Han Xizai 韓熙載 (902–70), an official who served in Emperor Yuanzong's court (mentioned above), whose pleasure-seeking antics were well known in his time.Footnote ⁴⁵ This entry, however, connects Han to an aesthetic venture in the lesser-known realm of olfactory discernment:

Although this particular anecdote is tantalisingly short, it presents the kind of connoisseurial engagement with blending scents and situating them in a socially mediated experience, in contrast to the aforementioned cases of undisciplined and extravagant consumption of aromatics in which a corrupt emperor or a parvenu merchant simply burns agarwood. The evaluative terms, such as he 和, miao 妙, and yi 宜, along with the mention of Han's discourse, suggest that there may have existed more writing on connoisseurial matters from this era that, like the early medieval Hexiang fang (Recipes for Combining Aromatics) by Fan Ye, have since been lost.

The first kind of connoisseurial engagement suggested by this entry is that of the aromatic pairing of scents. Although it does not explain the reasoning behind each aromatic pairing, its overall logic is to combine aromatics of resinous, woody, and animalic sources with the seasonal scents of in-situ botanical fragrances found in the cultivated garden. The mention of Han Xizai's discourse of the ‘Fivefold suitability’ suggests yet another common penchant in connoisseurship, namely that of the numerological component of rankings, lists, and categorisations. The numerological category of five evokes the role of the five elements (五行)—‘one of the most important generators of the possibility of cosmic knowledge’ in the context of patterns that might serve as models to make sense of objects through groupings.Footnote ⁴⁸ It is worth noting that, of the five aromatics mentioned in this account, four of them—camphor, sandalwood, agarwood, and musk—are imports or largely imports from foreign states or from the outer reaches of the empire. This inventory thus speaks to the nature of aromatics that were popular in the tenth century and have become increasingly appreciated to enhance a garden environment shared by the elite, perhaps in a way analogous to Han Xizai's famous night revels.

What a swindle story tells us: ‘leaked’ scents and the diffusion of dragon brain

In understanding the process by which foreign aromatics gained traction in Chinese scent culture and became assimilated into daily use, one discernible pattern is that some imported aromatics began their role in Chinese olfactory culture first as rarities gifted to rulers, who first disseminated them to those in an inner circle, thus beginning a process of transculturation. In this regard, rare aromatics share similarities with the rise in popularity of tea, which was also frequently given as a high-level gift in the Tang era.Footnote ⁴⁹

Even from the early medieval era, there were examples of this pattern in literary traces, in which we see a foreign and unfamiliar scent ‘trickle down’ as a gift bestowed from the highest echelon of power. When Cao Pi 曹丕 (187–226, r. 220–26) transplanted into his garden an aromatic plant of Roman origin called midiexiang 迷迭香, he invited a group of literati to compose rhapsodies on it. The rhapsodies composed on this occasion are the only extant early descriptions of this foreign aromatic plant and, in effect, capture an occasion on which a new and foreign oddity was incorporated into the collective imagination through a conventional literary form.Footnote ⁵⁰ Another literary record can be found in the collection of anecdotes compiled during the Wei-Jin period, Shishuo yinyu 世說新語 (A New Account of Tales of the World). This scent-oriented narrative unfolds around an unidentified ‘rare and unusual fragrance’ (奇香) originating from the ‘western regions’ and said to last for over a month when worn on a person. The courtier Jia Chong 賈充 (217–82) receives this fragrance as a gift from the emperor and, to his surprise, smells this scent one day on one of his officials named Han Shou 韓壽, leading to his realisation that his daughter has initiated a liaison with this young man.Footnote ⁵¹ Thus, due to both its rarity and durability, the fragrance serves as a tracer for the otherwise invisible connections of desire and intimacy.

Building on these early medieval precedents, this section of the article draws on anecdotes from the late-Tang and Five Dynasties era to focus on the descriptions of dragon brain aromatic that highlight its strong and lasting scent, and, just as importantly, its olfactory reception. I hope to show that, from the ninth and tenth centuries, dragon brain was an imported aromatic that seemed to strongly exemplify the abovementioned ‘trickle-down’ effect, straddling a dual identity as royal status symbol and commodity, thereby offering us glimpses into the pathways for the transculturation of imported scents.

First, as a brief background to the nature of dragon brain aromatic, the Chinese term longnao xiang in this context mostly refers to Borneo camphor, formed in the wood of the evergreen tree Dryobalanops aromatica, whose natural range during the Tang era was in the Malay region. The English term ‘camphor’ can refer to the aromatic products of two other tree varieties: Cinnamomum camphora from southern Japan, Taiwan, and Ryukyu islands, as well as Blumea balsamifera, which is a common variety found in Southeast Asia and southern China. All three varieties yield aromatic material that is distinct in its properties.Footnote ⁵² In contrast to aromatics such as agarwood, which was used in its resinous woody form, Dryobalanops aromatica yields a camphoraceous oleo-resin (oil) in addition to solid camphor (borneol).Footnote ⁵³ Before the term longnao xiang denoted Borneo camphor as seen in Tang sources, the term longnao appeared in a more literal sense in at least one early medieval text as an appended commentary to the entry “long gu” 龍骨 (dragon bone) in Bencao jing jizhu 本草經集注 by Tao Hongjin 陶弘景 (456–536), under the category of ‘the three grades of crawlers and animals’ (蟲獸三品). As indicated by both its categorical heading and its description, in this work, it refers literally to constituents of a dragon rather than metaphorically to the botanical by-product of a tree.Footnote ⁵⁴ How the term ‘dragon brain’ became detached from the creaturely dragon and became associated with Borneo camphor awaits further investigation.Footnote ⁵⁵

Textual records from the seventh to tenth centuries mention dragon brain in official accounts of tribute gifts, in medical treatises, and in anecdotal literature. In the first type of sources, emphasis is placed on its rarity and exotic origins. For example, in the official history of the Sui Dynasty (581–618), dragon brain is mentioned as a tribute item from the state of Chitu 赤土; another reference in the New Tang History mentions the tribute of dragon brain from the state of Wucha 烏茶 in the Indo-Malay region.Footnote ⁵⁶ Perhaps also due to its name, which has a powerful association with the throne, dragon brain is often featured in royal rituals: in the annals of Emperor Xuanzong 宣宗 (810–59), it was mentioned that there had been a long-standing custom in the palace of strewing dragon brain and saffron underfoot for the emperor.Footnote ⁵⁷ These references, though by no means precise in their denotation of dragon brain, largely point to a converging set of characteristics of dragon brain as far-flung and precious—something gifted and reserved for royal consumption.

Medical treatises from this period also contain descriptions of dragon brain. The early Tang state-commissioned pharmacopoeia Xinxiu bencao 新修本草 (Newly Revised Materia Medica) describes its origin as being from the state of Polü (婆律國), associated with north-west Sumatra, and its sapor (味) and major medical uses.Footnote ⁵⁸ The text also mentions the preferred method of storage to prevent it from wasting away.Footnote ⁵⁹ The latter suggests that the aromatic described may have been a combination of the borneol from the Dryobalanops aromatica tree (which does not evaporate or volatilise) and that from other camphor trees (which does volatilise when exposed to air). The pharmacological treatise Waitai miyao 外臺秘要 (Arcane Essentials from the Imperial Library) from 752 contains six medical formulas using dragon brain.Footnote ⁶⁰ In these recipes, however, dragon brain is not distinguished by its scent or its olfactory effects and is treated as any other medical ingredient.

The third category of texts containing references to dragon brain is from anecdotal literature, and is perhaps the most capacious. Some of the shorter anecdotes overlap with references to dragon brain in the abovementioned types of sources; however, it is in anecdotal literature that we find the most emphasis on dragon brain as a scent and olfactory experience. For the purpose of this article with a focus on olfactory perception, I will look closely at four substantive and resonant anecdotes drawn from two anecdotal collections written in the ninth century and two from the tenth, all of which emphasise the nature of dragon brain as a socially rooted scent, sometimes in surprising ways.

First, anecdotes about dragon brain show that it was used in its crystalline form, as an oil, or as the wood of the tree; it functioned as a pharmacological ingredient, as personal perfume, and as ingredients in food and drink. Because of its putative cooling effect, there was an imperial practice in the court of Jingzong 敬宗 for making ice-chilled ‘refreshing breeze rice’ (清風飯) during the dog days of summer; it used powdered dragon brain as an additive.Footnote ⁶¹ Its oleo-resin form seems to have been used in scenting wine, as well. The ‘wine and ale’ section (酒漿門) of the Qing yi lu notes that, during the winter, the grand councillor Pei Du 裴度 (764–839) served his guest a drink called ‘fish wine’, which was made by first congealing dragon brain and carving it into the shape of small fishes, before immersing them into a flagon of hot wine.Footnote ⁶² From the late-Tang era, the longest and most systematic description of the source tree of dragon brain is from the ninth-century collection Youyang zazu 酉陽雜俎 [Mixed Morsels from Youyang], which details its taxonomy, origin, botanical properties, and methods of harvest:

Whatever lingering ambiguities there may have been regarding the terminology for dragon brain, accounts of its olfactory effects focus on its extraordinary scent. In this entry, Emperor Xuanzong 玄宗 (r. 712–56) receives a tribute gift of dragon brain from Jiaozhi 交趾 in Indochina, then bestows it onto Precious Consort Yang Guifei 楊貴妃 (719–56) as a rare personal perfume:

The newly acquired epithet of ‘auspicious’ in particular is presented as a Sinitic quality that was not inherent in the aromatic itself, but was acquired after its arrival in China. Either because of its scarcity or due to the reference to a royal emblem in its name, dragon brain aromatic evoked an association with the imperial inner circle.

In the second half of this anecdote, the scent of dragon brain begins to take on a larger-than-life importance. It recounts that, one day, as Yang Guifei was watching the emperor play a game of chess, her kerchief was blown away and landed on the headgear of a courtier nearby. The courtier went home and kept his headgear wrapped in a brocade pouch. The subsequent scene takes place after the An Lushan Rebellion and the author did not need to mention to readers that Xuanzong was forced to put his favourite concubine to death during his exile. The anecdote only tells us what happens years later, after order was restored:

In this instance, a rare scent was presciently ‘archived’ and stored away; it later allowed the grieving Xuanzong to access the memory of Yang Guifei, whose death and absence are contrasted with the enduring scent of the dragon brain, which shows a preternatural ability to transect time and evoke memories and emotions, in the manner of the Proustian madeleine a millennium later.

In the case of her kerchief, Yang Guifei's signature scent stayed within the confines of the palace and with the person who originally bestowed the gift but, in other cases, the scent of dragon brain was not easily contained; it breeched physical and social barriers and hierarchies by wafting beyond these confines. The incident in question is recorded in the collection Duyang zabian 杜陽雜編 (Miscellaneous Compilations from Duyang) by Su E 蘇鶚 (fl. 885–88) and begins with a description of the extravagant habits of Princess Tongchang 同昌 (849–69) as she travelled around the city ensconced in luxurious objects and subsequently leads to a discussion about whether the whiff of aroma someone caught in a wine shop was the legendary dragon brain:

The Guanghua 廣化 Ward, where this anecdote is set, was a residential ward near the Xingqing 興慶 Palace in the north-eastern quadrant of Chang'an.Footnote ⁶⁸ This, then, was urban space that was outside the palace gates and yet proximate to it—a spatial detail that is relevant to the issue of how scent can be ‘leaked’ unintentionally. The narrative depicts the volatility and power of the dragon brain scent lodged in a brocade jacket, which allows those outside of the princess's inner circle to access her opulence as well as royal status, bringing traces of the inner court to urban, public spaces such as the wine shop. Compared with the ostentatious display of opulence mentioned at the beginning of the anecdote, the brocade jacket functions as something that intrigues rather than boasts and, precisely due to the tenuousness of the scent, occasions a greater hold on the imagination.

Another way to read this anecdote is to see it as narrative evidence that a scent, however exclusive and luxurious, cannot be fully restricted because its olfactory pathway cannot be controlled or even predicted. We see another incident of ‘scent leakage’ in an anecdote from the collection Qing yi lu entitled ‘Snow fragrance fan’ (雪香扇), which describes an incident from the court of the Later Shu 後蜀 (934–65) in the Five Dynasties. Here, a prized aromatic again transgresses palace walls and ‘leaks’ from the inner palace:

Here, the vector for scent propagation is a fan instead of a jacket or headgear; the tension here is between dragon brain being a closely guarded ‘royal’ signature scent on the one hand and its being emulated or popularised in the outside world by the privileged connoisseur. The anecdote suggests that a small mishap is all that it took to unleash the closely guarded aromatic into wider circulation. Once again, it is significant that the practice is scent-related rather than one that comes under the purview of sumptuary regulations, designed on the basis of sight-oriented consumption. Yet, here it is nonetheless still applicable that the ‘leaking’ of the royal scent shows an inherent contradiction, as pointed out by BuYun Chen, of sumptuary measures: ‘Through the prescription and proscription of specific objects, the laws plainly reify the objects as symbols of rank and status—thereby opening them up to usurpation and imitation’.Footnote ⁷⁰

We do not know exactly how much dragon brain was circulating outside of imperial use, but the fact that a small number of medical recipes refer to it (even without mentioning its scent) makes it likely that it was also available at least to some; Wen Cuifang estimates that the price for dragon brain during the Tang was 50,000 wen of copper coin for two liang of the aromatic.Footnote ⁷¹ The following narrative is particularly evocative for this reason, because it demonstrates that, at least in the popular imagination, dragon brain was known and identifiable outside the palace and that, whether as lore or as practicable knowledge, its connoisseurs were not limited to imperial consorts and eunuchs. As with many informal and anecdotal narratives from this time, it shows what was plausible to the audience of the narrative at the beginning of the tenth century, and therefore does not need to be factually accurate in order to illustrate pertinent perceptions from this time. The narrative describes a band of conmen who swindled a Buddhist monastery during the reign of Emperor Yizong 懿宗 (r. 859–73) by, remarkably, impersonating the plain-clothed emperor (textual emphasis mine):

What does this swindle narrative tell us about dragon brain? First of all, scent plays a central role in this swindle as a claim to authenticity. The swindlers’ aim is to counterfeit an emperor counterfeiting a non-emperor, in which the dragon brain scent is the olfactory clue that would give the emperor away. Furthermore, the swindlers show an entrenched understanding of the psychology of the monks in a temple close to the imperial palace: the monks know that the emperor may be visiting in plain clothes to give alms but are complicit in not giving the emperor away. The ingenious swindlers show that they understand this complicity better than the monks know it themselves and, in this ‘recursive theory of mind’, dragon brain is an unmistakable olfactory marker of the subsurfacial and invisible hierarchy of the world.Footnote ⁷⁴

With its olfactory premise, the account can only be plausible if dragon brain occupied a paradoxical place in the cultural imagination in this late-Tang moment: it must have been simultaneously restricted and yet part of the popular consumptive imagination—that is, something that signalled an imperial presence and yet was familiar enough for those outside of the palace to recognise it, and, just as importantly, was plausibly available to the counterfeiters.

In this context, it is interesting to note that, despite observations of its powerful scent compared with other aromatic ingredients, dragon brain was rarely mentioned in Tang recipes for fumigating clothing (薰衣香方). In the recipe collection Bei ji qianjin yaofang 備急千金要方 (652), five such recipes are included, each of them calling for various quantities of agarwood, musk, sandalwood, storax, and armour aromatic, but none of them included dragon brain. Similarly, among the five fumigant recipes preserved in the pharmacological treatise Waitai miyao 外臺秘要 (Arcane Essentials from the Imperial Library) from 752, including one bestowed by Emperor Taizong during the Zhenguan 貞觀 era, none of them contained dragon brain.Footnote ⁷⁵ This pattern is consistent with what this swindle story implies: that dragon brain was not a scent for popular use as a fumigant perfume.

However, this is also a story of a savvy connoisseurship of scent reaching beyond enforced social hierarchy, balancing uneasily between the aromatic as a commodity and as an authenticating signifier. In his article ‘The cultural biography of things’, the anthropologist Igor Kopytoff proposes two diametrically opposed poles when considering an object. At one pole, an object can be a ‘perfect commodity’ when it is ‘exchangeable with anything and everything else’; its opposite, the ‘perfectly decommoditised object’, is one that is singular and unexchangeable. Kopytoff notes that there are two forces that oppose the drive toward commodification: 1) what he calls ‘culture’ or the drive for discrimination and 2) one in which the power-holding state places prohibitions to preclude certain objects from being commoditised because, as he puts it, ‘power often asserts itself symbolically precisely by insisting on its right to singularize an object, or a set or class of objects’.Footnote ⁷⁶ Here, I propose that this tenth-century swindle narrative serves as an incidental cultural biography of dragon brain: it captures the precarious moment at which dragon brain transitioned from a singularised object reserved for the uppermost echelon of society to one that was de-singularised and had the potential to be commoditised as an object to be acquired by the savvy. As Sophie Volpp points out in the case of the circulation of python robes during the Ming era, the prestige of such robes ‘depended on the fiction that it was manufactured exclusively under imperial auspices’.Footnote ⁷⁷ The same can be said of the prestige of dragon brain as a fumigant fragrance at this moment in its olfactory history. This narrative therefore echoes the pattern found in aforementioned accounts (relating to Princess Tongchang and to Meng Chang) in which dragon brain, as a rare and closely guarded scent, is ‘leaked’—even fictively—into the urban space at large in the capital city. This leak speaks to not only its being an object with potential for exchange, but also its ontological status as a scent, which has a unique mobility, as Jonathan Reinarz puts it, to cross boundaries that are otherwise difficult to breach.Footnote ⁷⁸ While dragon brain may have begun its early life in China as an imperial tribute item or elite gift, by the time of the anecdote (in the late-Tang era and during the Five Dynasties), it could no longer be contained.