Genuine Words: Deception as a War Tactic and a Mode of Writing in Third-Century China

Abstract

The late Eastern Han (roughly the Jian’an era, 196–220) through the Three Kingdoms (220–265) period witnessed a historic turn at which war, deception, and writing coincided in more intricate ways than ever before. By closely examining three cases of fabricated letters, I untangle the complex phenomenon of “writing in deception” from the lens of war tactic and mode of writing. My basic question is how and why writing created or facilitated new possibilities for deceit in war during this period. More broadly, how was a voice created in deception different from one that was “genuine”? What does “writing in deception” reveal about the nature and perception of literature or wenzhang 文章? My inquiries shed light on a fiercely pragmatic approach to writing, while exposing the line between “deceptive” and “non-deceptive” writings to be extremely fine and tenuous. Ultimately, these cases of fabricated letters challenge us to think about authorship and other properties of a literary work, such as authenticity, legitimacy, genuineness, and sincerity, as the results—that is, the effect and affect—of writing, and not the other way around.

Keywords:

• deception
• literature
• war tactic
• literary voice
• literary effect and affect

According to one account, the widespread civil wars at the end of the Han dynasty (206 BCE–220 CE) were inaugurated by a piece of fabricated writing. At the time, Yuan Shao 袁紹 (154?–202), who would become the leader of the coalition forces that rose against Dong Zhuo 董卓 (138–192), who ransacked the Han court, was about to launch his campaign.¹ However, the Regional Governor of Ji Province (Jizhou mu 冀州牧) named Han Fu 韓馥 (d. ca. 192), himself presiding over large population and abundant military supplies, was very suspicious of Yuan Shao.² Thereby, he dispatched his retainers to contain Yuan Shao, making it impossible for the latter’s army to proceed. Just then, Qiao Mao 橋瑁 (d. ca. 190), Governor of Dong Commandery (Dongjun taishou 東郡太守), “wrote in deception” (zhazuo 詐作) a “Proclamation by the Three Dukes at the Capital” (“Jingshi Sangong yishu” 京師三公移書), in which he made charges of “crimes and evildoings” against Dong Zhuo. He then circulated the fake proclamation to the provinces and commanderies, intending to galvanize anti-Dong Zhuo forces. Apparently not suspicious of the fake proclamation, Han Fu ultimately followed its call and completely changed his attitude toward Yuan Shao. In a letter sent to Yuan Shao, he speaks of Dong Zhuo’s “evils,” presumably reiterating Qiao Mao’s language, before permitting Yuan Shao to move forward with his campaign.³ One of the most violent and chaotic periods in Chinese history commenced henceforth.

Using writing to deceive was not a new phenomenon in the late Han.⁴ In accounts about such use of writing in earlier times, however, there is no evidence that the tactic played a major role in wars. In this light, the account above is indicative of a historic turn in which war, deception, and writing coincided in more complex and intricate ways. This account, believed to be composed contemporaneously with the events it described, has the potential to help us understand this historic turn in two ways:⁵ first, by showing us, with its recounting of events, what might have taken place at the time; and second, by showing us, with how it recounted the events, the attitude at the time toward what had actually taken place. By closely examining such contemporaneous or near-contemporaneous accounts about the use of writing for deception in war, I situate the phenomenon historically to explore how it signaled important shifts, not only in military strategy or tactics, but also in writing or literature—that is, wen 文 or wenzhang 文章 (“patterned composition”).

The basic question to be addressed in this article is how and why writing created or facilitated new possibilities for deceit in war during this period. This question concerns the conception and practice of writing in this period. While earlier accounts about using writing for deception are only sporadic, they portray the practice in a negative light, as such relegating it to the marginal space of the morally dubious. The accounts examined here, as we will see, reflect a drastically changed attitude. As such, while they bring a previously hidden phenomenon to light, they further raise a whole host of intriguing questions about writing itself: how did deceptive writing exactly deceive? How was a voice created in deception different from one that was “genuine”? How was fabricating the voice of, say, an enemy different from writing on behalf of a ruler or an acquaintance? My inquiries will ultimately shed light on a fiercely pragmatic approach to writing that became evident in the late Han (roughly the Jian’an era, 196–220) through the Three Kingdoms (220–265) period. While this approach might be deemed a consequence of wars and conflicts, it nonetheless poses the challenge—for later critics and for us as well—of how to place and evaluate works of deception. This challenge compels us to consider “writing in deception” not just as a war tactic, but also as a mode of writing, that is, to consider it within the realm of writing and literature; from this lens, the line between “deceptive” and “non-deceptive” writings, as we will see, is exposed to be extremely fine and tenuous.

Deception and War

Early Chinese traditions were deeply concerned with the issues of fabrication/fakery (wei 僞) and deception (zha 詐). Thinkers and theorists since the Spring and Autumn period (770–476 BCE) and Warring States period (475–221 BCE), including but not exclusive to moral-ritualists in the tradition of Confucius (551–479 BCE), expounded repeatedly on the problems of deceptive behaviors, often using the words wei and zha in describing them.⁶ The fundamental purpose of ritual teachings, as encoded in works such as Li ji 禮記 (Records of Rites) and Zhou li 周禮 (Zhou Rites), was to “make evident sincerity while expelling fakery” (zhu cheng qu wei 著誠去偽).⁷ Often, the ritual codes that “forbid” (jin 禁) acts of deception in these works also imply that violators would face punishment.⁸ The threat of deceptive acts was conceived broadly, not only to moral, social, and political order, but also to wealth and material possession. But there was exception in one field of thought: military strategy (bingfa 兵法). Here, the employment of deception is openly discussed and even promoted. The Sunzi bingfa 孫子兵法 (Sunzi’s Strategy of War) from the fifth century BCE, whose “fundamental thesis is that war is based on deception,”⁹ for example, famously contains a chapter on the use of spies (jian 間), not only to gather intelligence, but also to disseminate disinformation, the intent of which is to cause disruption in the enemy camp.¹⁰ In the work of Han Fei 韓非 (ca. 280–233 BCE), the employment of deception as “a tactic of military maneuvering” (junlü zhi ji 軍旅之計) is also given approval.¹¹ Such military discourse that promotes the use of deception underscores the perception of war as a differentiated sphere of engagement; here, the conventional objection to deception does not necessarily apply.

In spite of the promotion of the use of deception in works such as Sunzi bingfa and Han Feizi, it is worth noting that there were other contrary opinions. In addition to Xunzi 荀子 from the fourth century BCE and Lüshi chunqiu 呂氏春秋 (Mr. Lü’s Spring and Autumn) from the third century BCE, in which the use of deceptive tactics to win wars is viewed as a limited, if not illegitimate, strategy,¹² the Wei Liao zi 尉繚子, a work from the late fourth century BCE, calling victory by deception “crooked victory” (qu sheng 曲勝), claims that these tactics are “imperfect models for the general.”¹³ As such, even in the realm of military thought, the use of deception was not unanimously embraced. Historical accounts of wars won through deceptive tactics attest to the fact that deception was commonly practiced in the battlefield, but the retelling of such events often reveals that an underlying unease toward deception persisted. In one such account in Sima Qian's 司馬遷 (ca. 145–ca. 85 BCE) Shi ji 史記 (The Grand Scribe’s Record), for example, the famous Han general Li Guang 李廣 (184–119 BCE) is said to have lured (you 誘) more than eight hundred Qiang 羌 rebels into surrendering before killing all of them, a tactic that Li Guang himself called zha.¹⁴ But in the account, we learn of Li Guang’s deception at the point when Li Guang was asked about his “regrets” (hen 恨), and his deceptive and brutal tactic was identified by a fortune-teller as the cause of the denial of his promotion to the rank of marquis (hou 侯).¹⁵

A fundamental shift in the attitude toward the use of deceptive tactics emerged amidst the collapse of the Han central court, as violence in the capital cities gave way to constant and widespread wars among aspiring warlords, leading eventually to the triangulation of the Three Kingdoms period. This shift is signaled not only by the promotion of the Sunzi bingfa by Cao Cao 曹操 (155–220), the most powerful warlord of the period, but also, more indicatively, by contemporaneous or near-contemporaneous accounts of the practice of deception in war.¹⁶ Chen Shou’s 陳壽 (233–297) Sanguo zhi 三國志 (Records of the Three Kingdoms), the official history of this period, is the most informative resource in this regard, but other works, many of which survive in fragmentary form through Pei Songzhi’s 裴松之 (372–451) commentary to Sanguo zhi, also help fill out the context against which the use of deceptive tactics during the period is to be understood.¹⁷

In the following pages, I discuss three cases involving the strategic use of altered or fabricated letters in a war context. At least three other cases of fabricated or deceptive writing, identified as zha or jia 假 (“feigned”), are found in Sanguo zhi and Hou Han shu, but the fabricated texts in questions are not cited in these cases.¹⁸ Another case in which the deception is described but the text itself is not cited occurs in a work known as Wu lu 吳錄 (Record of Wu).¹⁹ Additionally, there is a related case, discussed later in this article, in which a letter is written for the purpose of “deceiving and inciting” a general in the enemy camp.²⁰ By engaging these accounts, specifically with the three selected cases, my article seeks to shed new light on the nature of writing both within and beyond the context of war during this period.

Circulation and Materiality

As noted above, the late Han through the Three Kingdoms period was possibly the first period in Chinese history in which the employment of deceptive writing became evident as a major war tactic. Qiao Mao’s story, cited at the beginning of the article, already hints at the rising use of writing for war propaganda, specifically in aiding and even substituting human agents in creating and disseminating misinformation. The contrast with earlier stories, such as that of Tian Dan 田單 (fl. 284 BCE) dispatching “turned agents” (fanjian 反閒) into the State of Yan 燕 to spread the rumor about Yue Yi’s 樂毅 (fl. 284 BCE) intent to empower himself, highlights writing as an alternative medium for wartime deception.²¹ Behind this rise of writing was no doubt a material reality different from that in which communication was largely unaided by writing.

In the first case of deceptive writing to be discussed below, the rivalry among the three kingdoms, Wei 魏 (220–265), Shu 蜀 (221–263), and Wu 吳 (222–280), plays out in none other than the form of a letter. The year was 233, and the Wei was ruled by Cao Rui 曹叡 (204–239; Wei Emperor Ming, r. 226–239). At the time, the Kingdom of Wu under Sun Quan 孫權 (182–252) and the Kingdom of Shu, guided by the powerful minister Zhuge Liang 諸葛亮 (181–234), had formed an alliance against the Wei. As they schemed to fight the Wei, Sun Quan and Zhuge Liang exchanged letters frequently. In this instance, a border commandant (bianhou 邊候) of Wei intercepted a letter that Sun Quan wrote to Zhuge Liang. Liu Fang 劉放 (d. 250), a courtier of Wei, who was known for his talent at composing “letters and proclamations” (shu xi 書檄), proceeded to alter the letter.²²

The Sanguo zhi account of the incident reads as follows:

The border commandant obtained Sun Quan’s letter, and Liu Fang then altered the words, changing the original language here and there while making it still fit together. He gave the letter to Man Chong (d. 242),²³ General for Expedition to the East, who sealed it before he showed it to Zhuge Liang, making it appear as if [Sun Quan] would like to submit [to Wei]. Zhuge Liang passed the letter to Bu Zhi (d. 247) and other major generals of Wu, and they in turn presented it to Sun Quan. Fearful that Zhuge Liang would be suspicious of him, Sun Quan took pains to explain and excuse himself.²⁴

邊候得權書，放乃改易其辭，往往換其本文而傅合之，與征東將軍滿寵，若欲歸化，封以示亮。亮騰與吳大將步騭等，騭等以見權。權懼亮自疑，深自解說。

First of all, this account relates a sense of the materiality that was at work as Liu Fang altered and handled the letter, hinting, as such, at the technology of writing that made his act of deception possible. As is known, during this time the materials used for writing included bamboo, wood, silk, and the more recent invention, paper, in addition to others.²⁵ In this case, as well as in the other two cases that I will discuss below, there is no way to ascertain what exactly was the material used for the letter. The description of the process of forgery in the passage cited above is perhaps most easily explained by assuming that the materials involved were bamboo slips, as the actions depicted—altering, changing, and refitting—could very well suggest unstringing the slips, then altering, adding, or replacing them to make them fit well as a text, before finally restringing them.²⁶ While this assumption cannot be verified, the more important point here is that, both in process and as a “product,” writing was not the same as speech.

Incidentally, the account also offers a sense of the circulation of the altered letter. The letter’s journey began with Sun Quan and went back to him. Coming full circle, the letter’s travels were also a process of transformation: by the time it returned to Sun Quan, the original writer of the letter, it was no longer the same letter. Compared to relying solely on using human agents, that is, on speech, the manipulation of writing must have caused disinformation to circulate differently through the human network, perhaps more efficiently and effectively than before. Rather than deploying human agents to sow discord as Tian Dan did, Liu Fang sent an altered letter into the human network instead. But what was it that writing could achieve that human agents by themselves could not? I will return to this question in the concluding part of this essay.

Alteration by “Historicization”

There is no extant record of the changes that Liu Fang made to Sun Quan’s letter. But luckily, there are accounts of another case of intercepting and altering a letter that offers some glimpses into what changes might have been made. This case allows for speculation on the considerations behind the act of altering a letter, hence raising questions about the nature of writing itself.

The case concerns a brutal battle between two warlords, Yuan Shao and his rival Gongsun Zan 公孫瓚 (d. 199), that took place in 199.²⁷ At the time, Yuan Shao’s army had completely encircled Gongsun Zan’s entrenchment. To break out of the encirclement, Gongsun Zan planned to launch a sudden attack from two sides—from within using his own army and from outside by coordinating with a rescue army. As part of the scheme, he sent a secret agent to deliver a letter to his son Gongsun Xu 公孫續 (d. 199?). In the letter, he commands his son to dispatch a rescue army, further adding a note instructing him to light a fire as a signal when the rescue army arrives. However, the letter was intercepted by a scout working for Yuan Shao, who then instructed Chen Lin 陳琳 (d. 217), one of the most accomplished writers from the period who at the time was serving under Yuan Shao, to alter the letter. The story ends with the tragic death of Gongsun Zan. Yuan Shao lured Gongsun Zan’s soldiers out of their defense by lighting a fire as a false signal and then defeated them in an ambush. Yuan Shao had also built underground passages to launch direct attacks on Gongsun Zan’s main quarters. Desperate under siege, Gongsun Zan killed his wives and children before committing suicide.²⁸

Based on an account in Dian lüe 典略 (Canonical Summaries), a third-century work, Gongsun Zan’s original letter reads as such:

The attack of the Yuan is like that of gods and ghosts: as sounding drums and horns ensued from the ground, cloud ladders and siege engines danced above on my towers. As days end and months pass, I have no one to rely on. You should pledge your “smash head” loyalty to Zhang Yan (fl. 199), and quickly send a light cavalry this way.²⁹ Upon arrival, light a beacon fire in the north, and I shall emerge from within. If not, [then I die, and] after I die, even though the world under Heaven is large, how could you find a place to settle in, were you to desire it?³⁰

袁氏之攻，似若神鬼，鼓角鳴于地中，梯衝舞吾樓上。日窮月蹴，無所聊賴。汝當碎首於張燕，速致輕騎，到者當起烽火於北，吾當從內出。不然，吾亡之後，天下雖廣，汝欲求安足之地，其可得乎！　

According to Pei Songzhi, who based his account on the third-century (?) work Xiandi chunqiu 獻帝春秋 (The Springs and Autumns of Han Emperor Xian), the letter as altered by Chen Lin reads as follows (underlines added):

I have heard that in the past, at the time of the waning Zhou, corpses stiffened as blood flew, but I did not believe it to be so. How could I have imagined that I myself would be hit by the same situation today! The attack of the Yuan is like that of gods and ghosts: as sounding drums and horns ensued from the ground, cloud ladders and siege engines danced above on my towers. As the day ends and months pass, I have no one to rely on. You should pledge your “smash head” loyalty to Zhang Yan, and quickly send a light cavalry this way. Upon arrival, light a beacon fire in the north, and I shall emerge from within. If not, [then I die, and] after I die, even though the world under Heaven is large, how could you find a place to settle in, were you to desire it?³¹

蓋聞在昔衰周之世，僵尸流血，以為不然，豈意今日身當其衝！袁氏之攻，似若神鬼，鼓角鳴于地中，梯衝舞吾樓上。日窮月蹴，無所聊賴。汝當碎首於張燕，速致輕騎，到者當起烽火於北，吾當從內出。不然，吾亡之後，天下雖廣，汝欲求安足之地，其可得乎！　

If extant sources are to be believed, Chen Lin only added several phrases at the start of the letter, leaving the rest intact.³² As innocuous as they appear, Chen Lin’s phrases added a new element to the letter: historical reference. Specifically, his addition refers to the demise of the Zhou dynasty (ca. 11^th c.–4^th c. BCE). Historical references were certainly already commonplace in earlier writings, but they seemed to be used in particularly concentrated ways during this period.³³ In writings, a historical reference often functions to “prove” or make evident the way things should be, or rather, simply the way things are. The citation of historical references was, in this sense, meant to mask what was discursive as demonstrative. In other words, history was made into, and utilized as, the site of li 理, that is, the “inherent pattern” or “universal principle” of the world.³⁴ On the surface, historical references appear to serve the prognostic function: the citation of a past event or a past figure foretells or predicts the outcome of an ongoing event or the fate of a living figure. But in fact, the way writing works is that it can camouflage an “after-the-fact justification” through historical references’ perceived prognostic mode, making it appear as if the “justification” were written before the fact, turning it as thus into a “prognostication.”

To be sure, much remains unknown: we have no information on exactly when, amidst the brutal battle, Chen Lin altered Gongsun Zan’s letter, nor on what the Yuan Shao camp did with the altered letter. But early medieval recounting of this case hinted at the propagandistic intention behind the practice of intercepting and altering an enemy’s letter. Chen Lin’s addition of the gruesome image of the Zhou dynasty’s decline at the start of the letter created a defeated or defeatist image of Gongsun Zan, while signaling the end of the current Han dynasty. The targeted audience of the altered letter was likely not just Gongsun Zan’s son, the intended addressee of his original letter. The Yuan Shao camp could very well have made the altered letter after Gongsun Zan’s death, circulating it to propagate the notion that Gongsun Zan saw his own demise coming and that the Han was in irreversible decline. Extant sources have noted at least one Guan Jing 關靖 (fl. 199), a subordinate of Gongsun Zan, who launched a “suicide attack” at Yuan Shao’s army after Gongsun Zan’s death. In this light, the need to “close the case” of Gongsun Zan by rationalizing through history—that is, to suggest that his loss and death were inevitable—was likely aimed at, among others, discouraging and suppressing his followers from further resistance. Furthermore, it is worth considering the fact that eight years earlier, in 191, Yuan Shao had attempted to replace the Han emperor held hostage by Dong Zhuo with his own candidate, going as far as using a jade seal to legitimize his choice.³⁵ For ambitious warlords such as Yuan Shao, spreading the message that “the current Han dynasty is coming to an end” was a part of their propaganda campaign, as is also seen in the case of Cao Cao, who, in a famous song deemed propagandistic by Jean-Pierre Dièny, essentially declares that the Han’s fortune has expired.³⁶

Historical referencing—ultimately a legitimizing strategy—was by no means the only rhetorical device that could be manipulated to create deception. That it happened to be the rhetorical device used in this case underscores the delicate nature of tampering with an enemy’s letter for strategic gains. The crucial skill involved, as this and the first case both suggest, was in maneuvering words and phrases in subtle ways, perhaps just enough to shift the tone, perspective, attitude, and/or implication of the original text in the intended direction. The subtlety of the operation highlights the power of the written language in conveying complex meanings, but it also calls the written language as a form of meaning into question, exposing it fundamentally as a construct.

How, then, is a piece of writing that is fabricated or deceptive ultimately different from one that is “authentic” or “genuine”? These cases of fabricated or altered letters help focus the issue not on authenticity or genuineness per se, but on authenticity and sincerity as an effect and affect of writing; thus, they probe directly at what writing is capable of. Completely usurping the voice of another for wartime propaganda, as the next case will show, was already within the realm of what writing could achieve by the third century.

Genuine Words

The letter in this third case was entirely fabricated, from beginning to end. It was composed by Hu Zong 胡綜 (183–243), a celebrated writer in the State of Wu, in the voice of Wu Zhi 吳質 (177–230), a prominent general and adviser in the State of Wei, to fabricate the news that Wu Zhi had abandoned his own state and surrendered to the State of Wu.

In 229, the ruler of Wu, Sun Quan, officially claimed emperorship and established his capital at Jianye (in modern Nanjing). Upon this occasion, Hu Zong was promoted to be palace attendant (shizhong 侍中), basically a confidential adviser to the emperor. In late 229, as the two states continued to struggle against each other, some of the Wei “capitulators” (xiangren 降人) in Wu—that is, officials or military men of Wei who had surrendered to Wu—brought the information that Wu Zhi, who at the time was a commander-in-chief (dudu 都督) with the title Awe-Inspiring General of Hebei (Hebei zhenwei jiangjun 河北振威將軍), had fallen into distrust by the Wei court.³⁷ Seizing on this information about the rift between the Wei court and one of its most renowned advisers and generals (Wu Zhi was a most favored and trusted confidant of Cao Pi 曹丕 [187–226], the founding emperor of Wei who had died just three years prior), Hu Zong proceeded to fabricate a letter of capitulation attributed to Wu Zhi, undoubtedly with the assent or even command of Sun Quan.

At more than 1,500 characters, Hu Zong’s fabricated letter of capitulation is the longest of its kind from the period that still exists today. As a rare remnant, it gives us unique insight into the day’s practice of fabricating an individual’s voice in writing. The Sanguo zhi presents the letter in three parts, the first of which opens with a narrative about the great chaos at the time:³⁸

As the heavenly cord is loosened and cut off, the four seas are divided and collapsing. The masses are wan and worn, and the genteel men are spread far and wide. Soldiers and bandits bring even more hardship, leaving towns with no residents, but only wind, dust, smoke, and fire wherever one goes. Since the time of the three eras, there has never been a great chaos as extreme as that of the present time.³⁹

天綱弛絕，四海分崩, 羣生憔悴, 士人播越, 兵寇所加, 邑無居民, 風塵煙火, 往往而處, 自三代以來，大亂之極，未有若今時者也。

This narrative was a familiar one in the second and third centuries, echoing the refrain of “ours is a chaotic age” that prevailed during the period. It recalls, for example, Cao Cao’s description of a world turned topsy-turvy by Dong Zhuo in his famous poem “Shallot Dew” (“Xie lu” 薤露):⁴⁰

Lines 9–12:

賊臣執國柄	=	The treacherous subordinate held control of the state;41
殺主滅宇京	=	he killed the sovereign and demolished the capital of the realm.42
蕩覆帝基業	=	Recklessly he toppled the imperial base and enterprise;
宗廟以燔喪	=	the Ancestral Temple, thereby, burned to the ground.

The poem further refers to mass displacement and hardship:

Lines 13–14:

播越西遷移	=	Those displaced and scattered were sent westward;43
號泣而且行	=	they wailed and cried as they journeyed onwards.

In “Wormwood Village Song” (“Haoli xing” 蒿里行), another well-known poem by Cao Cao, there is also description of the death and empty space caused by large-scale wars:

Lines 11–16:

鎧甲生蟣蝨	=	Military armors bred nits and lice,
萬姓以死亡	=	ten thousand folks perished in death.
白骨露於野	=	White bones lie exposed in the wilderness;
千里無鷄鳴	=	across a thousand li there is no cock’s crow.
生民百遺一	=	Of the living, one out of a hundred remains;
念之斷人腸	=	brooding on it severs one’s bowels!44

If Cao Cao’s poems spoke to a deep, collective sense of chaos and victimhood during the period, how about Hu Zong’s fake “chaotic age” narrative? Does it not read exactly like the “genuine” accounts by other contemporaneous writers?

While it strikes a tone of collectivity at the start, Hu Zong’s letter shifts quickly into a personal voice. In the rest of the first part, “Wu Zhi,” referring to himself as chen 臣, “your servant,” directly addresses Sun Quan, referred to as bixia 陛下, “your Highness.” Here, “Wu Zhi” expresses his regrets for not having the chance to serve Sun Quan yet and his great admiration for the Wu ruler, using language such as this to portray him in the most majestic light:

Your Highness equals Caelum and Terra in virtue, and shares with the sun and the moon in brightness. Your demeanor of divine mightiness is naturally endowed upon you. Embodying and extending the august standard, your [moral] influence flows through ten thousand li.⁴⁵

陛下齊德乾坤, 同明日月, 神武之姿, 受之自然, 敷演皇極, 流化萬里.

The letter bemoans at the beginning that “[t]he heavenly cord is loosened and cut off,” but towards the end of the first part, it speaks of Sun Quan “ordering the heavenly cord” (zhengli tiangang 整理天綱).⁴⁶ The message is essentially that Sun Quan is the ruler mandated by Heaven to restore and preside over this world.

The second part of the letter is the longest and the most complex. The tone also turns even more personal. In this part, “Wu Zhi” speaks of his intimate relationship with former Wei rulers and his devotion to the Caos. The language was crafted to appear genuine and, thereby, to move its intended audience, as this passage shows:

In the past I was befriended and received by the Caos. We appeared as ruler and minister on the outside, but we were like kin [lit. bones and flesh] in private. As gratitude and devotion bonded us together, we became inseparable. Subsequently, I was given an appointment in a remote area, where I commanded the army in Hebei. At this time, I had great ambitions and high hopes; wishing to live and die with the Caos, I feared only that I would fail to establish merits and not succeed in my undertakings.⁴⁷

臣昔為曹氏所見交接，外託君臣，內如骨肉，恩義綢繆，有合無離，遂受偏方之任，總河北之軍。當此之時，志望高大，永與曹氏同死俱生，惟恐功之不建，事之不成耳。

“Wu Zhi” then alludes to the decline of the Wei rule under Cao Rui, the current Wei emperor, by describing a toxic court environment taken over by infighting, slanders, and betrayal. He portrays himself as being victimized by slander and defamation, his safety in jeopardy and his heart in great anxiety. As he concludes that the mandate of Heaven is not with the current Wei emperor, he keeps circling back to his plan to leave the Wei and join the Wu: never once admitting to having “surrendered” (xiang 降) or “rebelled” (pan 叛), he nonetheless justifies his attempt not only by describing at length Wei’s deteriorating treatment of him, but also by referencing well-received historical figures known for switching allegiance under difficult circumstances. He goes on to deliberate on the choice between “the integrity of a determined gentleman” (zhishi zhi jie 志士之節) and “the principle of establishing merits” (ligong zhi yi 立功之義), that is, whether to remain loyal to one’s sovereign or to build a great career out of one’s life.

In this part of the letter, “Wu Zhi” further refers to his past attempt to present a secretive plan to Sun Quan through a Wei official named Zhou Guang 周光 (fl. ca. 200).⁴⁸ Using the lack of response from the Wu ruler as the pretext, he wonders out loud about Sun Quan’s possible doubt about him:

Zhou Guang has been gone for a year and I have not heard any response—I wonder if my intent has indeed reached you or not? … I surmise that Your Highness, not having bestowed clear condolence upon me, must have thought your servant Zhi would not make such a move [i.e., rebel against Wei and collaborate with Wu], given his thorough adherence to the way of benevolence and righteousness. You might have thought what Zhou Guang said was mostly untrue, or there were other [unknown] circumstances associated with what he said.⁴⁹

光去經年，不聞咳唾，未審此意竟得達不？ … 臣私度陛下未垂明慰者，必以臣質貫穿仁義之道，不行若此之事，謂光所傳，多虛少實，或謂此中有他消息.

In a way, through Wu Zhi’s voice, Hu Zong is directly and preemptively addressing readers’ or listeners’ possible doubts about the authenticity of the letter—better put, he is appealing to their sense of authenticity. Thereafter, “Wu Zhi” explains himself again before finally, toward the end of this part of the letter, presenting a rather elaborate military scheme to Sun Quan. This military scheme is complete with actual place names and specific details, including “Wu Zhi’s” promise to supply “more than three thousand horses.”⁵⁰

In the last part of the letter, “Wu Zhi” returns to the topic of switching allegiance, but this time his perspective shifts to the ruler and, instead of historical models, he cites contemporaneous figures. He uses Cao Cao’s trust and employment of Xu You 許攸 (fl. 196), who left Yuan Shao to serve Cao Cao, to make the point that a successful ruler knows how to utilize talents. At the end, he brings up Yan Fu 閻浮 (fl. early 3^rd c.) and Zhao Ji 趙楫 (fl. early 3^rd c.), two Wei generals who supposedly intended to submit themselves to Wu but were killed by Wei because Wu did not respond to them soon enough.⁵¹ With that, he urges Sun Quan to take action quickly:

Now your servant, with deep sincerity, entrusts his life to you from afar. If you still harbor doubts [toward me] and do not take action in a timely manner, causing your servant to be cut off in isolation and suffer great calamity, then I am afraid all the bold and stalwart men in the world who wish to accomplish meritorious deeds will not dare entrust their lives to Your Highness again. I wish Your Highness will give thought to this matter. August Heaven and Sovereign Earth have indeed heard these words of mine.⁵²

今臣款款，遠授其命，若復懷疑，不時舉動，令臣孤絕，受此厚禍，即恐天下雄夫烈士欲立功者，不敢復託命陛下矣。願陛下思之。皇天后土，實聞其言。

By the end of the letter, “genuineness” or “sincerity” comes into full display—that is, as affect—in the fabricated voice of Wu Zhi, creating the effect of (possibly) moving the audience. But who was this voice intended to move?

Clearly, this letter was a piece of pro-Wu and anti-Wei propaganda. The Sanguo zhi further relates this information: “By the time this composition became widely circulated, Wu Zhi had already been summoned into the court to be Palace Attendant.”⁵³ All considered, it would seem that the letter was intended to be widely circulated. As mentioned, the letter keeps returning to one issue, that is, “Wu Zhi’s” plan to leave Wei and submit to Wu. This issue, often referred to as qujiu 去就 (“leaving [one lord] to approach [another]”) in contemporaneous discourse, was of utmost concern to the elite men of the time. It was entangled in a host of concerns about loyalty and filial piety (zhongxiao 忠孝), social status and reputation, livelihood and security, as well as battleground wins and losses.⁵⁴ A wrong decision on the issue could cost one’s life, and a right one could bring handsome rewards. As elite men debated the issue intensely, aspiring rulers saw the opportunity to persuade and recruit them.⁵⁵ Underlying Hu Zong’s fake letter of capitulation is in fact an argument, targeting elite men in Wei such as Wu Zhi, for them to prioritize their ambition over their loyalty. As the letter announces, in the voice of Wu Zhi:

As such, I set aside the integrity of a determined gentleman, and keep my thoughts on the principle of establishing merits.⁵⁶

是以忘志士之節，而思立功之義也。

Furthermore, the military scheme outlined in the letter might offer some indication of the regions that this message was intended to reach. The scheme proposed, on the one hand, to have Sun Quan dispatch his armies internally to the Huai River and the Si River to occupy Xiapi, and then for the two provinces, Jingzhou and Yangzhou, to respond in coordination.⁵⁷ “Wu Zhi’s” army, on the other hand, would come sweeping through from Hebei down to the south.⁵⁸ In the letter, “Wu Zhi” concludes by saying, “[once] our forces and formations are joined, the roots and sprouts would have taken hold forever.”⁵⁹ This military scheme, as such, could be read as a kind of call-to-arms to all the fighting men in the large stretch of area, connected by the place names mentioned in the letter that basically formed an encirclement of the Wei territory.

The message of the letter could not be clearer: leave the Cao Wei court and join the Sun Wu. But the letter is couched in personal details, including snippets of Wu Zhi’s biography and names of his several associates, and in a personal voice that shifts between pleading and mourning, self-deprecating and self-justifying, expressing both deep anxiety and agitated urgency. This letter was a great performance; it wanted its audience to believe by appealing to their reason, and most of all, by stoking their emotions.

Did the letter have any impact? If it was meant to persuade or even put pressure on Wu Zhi to surrender, it obviously did not succeed or did not have a chance to succeed. As the Sanguo zhi reports, by the time the letter was widely circulated, Wu Zhi had already been summoned back to the Wei central court. But one should not underestimate the ability of a written text to spread misinformation. The word used in the Sanguo zhi to describe the letter’s circulation is liuxing 流行, “spread by traveling.”⁶⁰ This choice of word hints at the fact that a text, through its circulation, can have a considerable influence on the shaping of opinion and perception. In light of this, it may not be far-fetched to speculate further on the impact of the fake Wu Zhi letter. As already noted, the letter was composed in late 229, and the final turn of Wu Zhi’s life followed closely: he was summoned back to court in 230 and died in the summer of the same year. There is no mention of the circumstances of his death in the dynastic history, but he had apparently so enraged the Wei emperor that after his death he was given the posthumous title Chouhou 醜侯, Shameful Marquis, and it would take his son well more than a decade to successfully petition the court to change his posthumous title to Weihou 威侯, Mighty Marquis, around 244.⁶¹ Could the fake letter of capitulation have had some impact on Wu Zhi’s life after all? Having a letter in his voice denouncing his emperor out in circulation could have only harmed his situation.

“Deception” as a Mode of Writing

What was unique or different about the deception that was created by writing? At one point in the fake Wu Zhi letter, the written text as a medium of communication is mentioned and juxtaposed against kouchuan 口傳, “passing a message through someone else’s mouth.”⁶² The unwitting suggestion there was that, unlike “passing a message through someone’s mouth,” which might be resorted to in a state of rush (cangcu 倉卒), the letter allowed one to explain oneself much more fully; it was, in that sense, a more substantial and reliable representation of the self in the first-person voice. If, however, a human agent were to convey a message from his dispatcher to a third party, he would have to speak for his dispatcher on the spot, which, in the language of the fake Wu Zhi letter, could result in “more falsehood and less actuality” (duo xu shao shi 多虛少實).⁶³ In early medieval China, the perceived first-person voice was that which lent the medium of writing its particular power to move, persuade, and make-believe; it was this kind of power that forgers drew on to fabricate and manipulate letters. This, however, does not mean that the forgers were wholly or necessarily relying on their own anonymity to succeed in their deceits. Lu Yun 陸雲 (262–303), a writer in the Western Jin dynasty (265–316), once described the published works of his brother Lu Ji 陸機 (261–303) using this phrase: zi xing tianxia 自行天下, that is, “they traveled the world on their own.”⁶⁴ As Lu Yun implied, a written text could detach from the author and take on its own life. This idea hinted at the notion that, having become “a thing of its own,” a well-written text gained a power and efficacy regardless of its author.

The three cases examined above furthermore showcase a group of elite men skilled in writing. In their day, Liu Fang, Chen Lin, and Hu Zong, the three forgers in the cases, were all highly recognized by various powerful warlords and rulers for their cai 才, “talent,” particularly their wencai 文才, “literary talent.”⁶⁵ Their stories draw out the concept of “literary talent” or, more broadly, the concept of “literature” (wen or wenzhang) at this time. Literary talent in all three cases basically means the skill to create and manipulate “voices” by means of writing, whether the voice of an enemy ruler, a desperate father, or a surrendering general. The ability to “write in the voice of another” by employing rhetorical devices to create an impression of justifiability, inevitability, or authenticity and/or the affect of sincerity or genuineness was important to being considered a “literary talent” at the time. Extrapolating from the experiences of these writing men, we can further ask: how was forging the voice of an enemy ruler fundamentally different from writing for one’s own ruler? And, how was fabricating the voice of an enemy general different from writing on behalf of an acquaintance?

As mentioned, Liu Fang, Chen Lin, and Hu Zong represented a group of elite men valued for their writing skill. Within a ruling regime, their main function was to produce prose writings that served diplomatic or other official purposes, as these Sanguo zhi descriptions of their work bring to light:

Yuan Shao put [Chen Lin] in charge of literary composition.⁶⁶

袁紹使典文章.

Emperor Taizu [i.e. Cao Cao] appointed both Chen Lin and Ruan Yu (d. 212) as Military Adviser to the Libationer and had them take charge over the record keepers. Of the letters and proclamations dealing with military and state affairs, most were composed by Chen Lin and Ruan Yu.⁶⁷

太祖並以琳、瑀為軍謀祭酒，管記室，軍國書檄，多琳、瑀所作也.

Ever since Sun Quan was directing [state] affairs, almost all of the announcements, records, and commands, as well as letters and contracts dealing with neighboring states, were made by Hu Zong.⁶⁸

凡自權統事，諸文誥策命，鄰國書符，略皆綜之所造也。

Liu Fang was good at composing letters and proclamations. Among the edicts and commands issued by the Three Progenitors [i.e. Cao Cao, Cao Pi, and Cao Rui], whether to summon or instruct [someone], many were composed by Liu Fang.⁶⁹

放善為書檄，三祖詔命有所招喻，多放所為。

The types of wenzhang listed in these descriptions, including letter (shu), proclamation (xi), announcement (wengao), record (ce), command (ming), and contract (fu), were all arguably “proxy texts,” for they were all written for or on behalf of a ruler, a court, or a regime.⁷⁰ The extent to which these types of writing facilitated and were integrated into the state bureaucracy of the period requires a separate study, but, as the contexts specified in the descriptions cited above (“[managing] military and state affairs”; “directing [state] affairs”; “dealing with neighboring states”; “[issuing] edicts and commands”) make it clear, these writings were crucial to the everyday function of a ruler and especially in regime-to-regime interaction. The works of early medieval writers such as Liu Fang, Chen Lin, and Hu Zong underscore their practice of “writing in the voice of another” as central to the nature and conception of literature in the second and third centuries.

Writing in the voice of another went far beyond aiding a ruler or serving a state: works written in the voice of an acquaintance, a family member, or simply a known person were quite prevalent at this time.⁷¹ In fact, even the autobiographical writing of this period was arguably a manner of writing “in the voice of another.” Matthew V. Wells has argued that the autobiographical writings of the period reflect “a multiplicity of ‘selves,’ each shaped by the formal demands of different genres,” which were in turn shaped by textual traditions (in other words, other works, especially those deemed canonical) and socially determined expectations and conventions.⁷² In this sense, so-called self-writings were not as much the expression of the “self” as it is as that of the “self” as textually and socially constructed. To write as the “self” was, therefore, to speak in the voice of a constructed self. It was as much of an intersubjective process as in any other mode of “writing in the voice of another.” Construed as such, to write was, put another way, to impersonate. At the same time, it required the author to draw on collective imagination: the success of impersonation depended on the audience’s belief that the voice they heard was real or authentic.

Simply put, “writing in deception” was one of the modes of “writing in the voice of another.” But how was it different from the “non-deceptive” modes of “writing in the voice of another”? At one level, the answer can be similar to how modes of imitation are differentiated in modern literary theories. Gérard Genette, for example, focuses on the writers or, rather, their goals or intention, in distinguishing between pastiche and “true forgery,” which he describes as “a perfect imitation.”⁷³ As he points out, “The true pasticheur wants to be recognized—and appreciated—as such. The author of an apocryphal text does not. His goal is to disappear.”⁷⁴ In making this statement, Genette concludes that Akakia and Bataille, who (in)famously wrote La Chasse spirituelle in Rimbaud’s style and claimed it to be the authentic work of Rimbaud before acknowledging their own act of forgery, had acted in failure, for they “wanted to have their cake and eat it too.”⁷⁵ To be a successful pasticheur or forger, one has to be clear about his or her goal: either to be recognized or to disappear. Importing this insight for the differentiation of “writing in deception” from other “non-deceptive” modes, we might conclude that it was the intention or goal to deceive, achieved in part by concealing one’s identity as the true author, that distinguished the former. This conclusion notwithstanding, we might wonder if the three forgers in our cases were somehow guilty of “wanting to have their cake and eat it too,” for, if not, how did it come to be known that they were the forger? A definitive answer to this question, which must rely on detailed knowledge of how the letters in question were circulated, transmitted, and documented, is out of our reach. The question itself, however, has two implications: first, the attitude toward forgery by writing was complex and diverse in early medieval time; and second, the efficacy of a written text was “potent” in its own way in the period.

In the second and third centuries, even deceptive writings could be recognized, if not celebrated, as a representation of a writer’s literary talent. Chen Shou’s treatment of Liu Fang and Hu Zong’s writing skill in Sanguo zhi is a case in point. It is right after he points out that “Liu Fang was good at writing letters and proclamations” that Chen Shou offers the anecdote about the latter’s alteration of Sun Quan’s letter.⁷⁶ There is no indication of any reservation about Liu Fang’s deceptive act on Chen Shou’s part: to him, Liu Fang’s successful alteration of the letter was a prominent example of his literary talent. Similarly, to demonstrate Hu Zong’s outstanding writing skill, Chen Shou includes his fake letter of capitulation in its entirety in the Sanguo zhi, putting it side by side with Hu Zong’s other representative works, including a fu written at Sun Quan’s command, and a meng (“covenant”) written to mark the alliance between Wu and Shu.⁷⁷ Chen Shou did make it clear in passing that the letter of capitulation was wei, “fabricated,” but without judgment.⁷⁸ This matter-of-fact attitude toward fabricated writings was consistent in other second- or third-century accounts about fabricated writings examined above.⁷⁹ The underlying implication of this attitude appeared to be a separation of literary skills from moral cultivation.

In earlier times, wei was directly opposed to de 德, virtue, as the Shu jing 書經 (Classic of Documents) teaches:

Let reverence and restraint be the virtues, while carrying no falsehood in yourself. Practice these virtues, then your heart shall relax, becoming daily more at ease; practice falsehood, then your heart shall struggle, becoming daily more burdened.⁸⁰

恭儉惟德，無載爾偽。作德，心逸日休。作偽，心勞日拙。

But in the early medieval period, while rulers such Yuan Shao, Cao Cao, Sun Quan, Cao Pi, and Cao Rui actively recruited men with exceptional literary skills, they often overlooked these men’s behavioral transgressions, that is, acts deemed inappropriate according to traditional codes of ritual etiquette, li 禮.⁸¹ This phenomenon reflects the tendency during this period to recognize literary talent as independent from ritual propriety that, as a form of bodily enactment, was inseparable from the person. This tendency can be further observed in Chen Shou’s assessment of Liu Fang: he praised Liu Fang for his literary skills (wenhan 文翰) but added that Liu Fang was lacking in “self-cultivation” (zixiu 自修).⁸² In this comment, “literary talent” stands separate from “self-cultivation,” the realm of moral and ritual training. In an age when deception could be delivered in writing materials through the craft of talented writers, a work of wei, like other works, was not a manifested form of a writer’s moral and ritual cultivation or the lack thereof, but rather of his literary skills and his utility to the enterprise of the state. Fiercely pragmatic, this instrumental view of literature departed significantly from the socio-political paradigm built on moral and ritual codes.

Incidentally, the pragmatic use of writing could have been encouraged by the efficacy that written texts possessed in the period. In another case of deception by writing, the Wu general Lu Xun 陸遜 (183–245) “feigned a reply letter to [the Wei border governor] Lu Shi 逯式” (jia zuo da Shi shu 假作答式書), in which he created the impression that Lu Shi had been in secret contact with him about surrendering to Wu.⁸³ What is interesting here is Lu Shi’s response: when his soldier brought the letter to him, he was “fretful and fearful” (huangju 惶懼) and subsequently sent his wives and children back to Luoyang, escorting them himself.⁸⁴ Even though he clearly knew that the letter was “feigned,” Lu Shi seemed to feel powerless in defending himself against the letter, opting for damage control instead. Recalling the case involving Liu Fang’s alteration of Sun Quan’s letter, we notice a similar response: while Sun Quan must have known immediately that his letter had been altered, he felt “fearful” anyway, worrying that Zhuge Liang would be suspicious of him because of the altered letter, and, instead of exposing the letter as fake, he “took pains to explain and excuse himself.”⁸⁵

Once crafted, a text gained efficacy by being read, heard, and felt to be “true” or conveying “truth,” regardless of who the author was or if the author was known. In this textual culture, it was possible that a letter speaking in the voice of a certain individual could still be effective even if it was known to have been fabricated by another individual. If so, then letter forgers could indeed “have their cake and eat it too.”

*  *  *

The pragmatic approach to literature during this period elicited diverse responses in the following centuries. One example comes from the historian Pei Songzhi. Even though he took no issues with other accounts about deceptive writing in the Sanguo zhi, in his note on Lu Xun’s use of a feigned letter against Lu Shi, he is critical of both Lu Xun, the deceiver, and Chen Shou, the narrator. He depreciates Lu Xun’s tactic by calling it “a minor act of deception” (xiao zha 小詐), adding that “treating this [act by Lu Xun] as [a] fine [example]—this, too, is not to be taken as a model” 以斯為美, 又所不取.⁸⁶ Whereas Chen Shou had focused on Lu Xun’s tactical skill (i.e., his cai or “talent), which, in this case, was manifested through writing, Pei Songzhi is evaluating Lu Xun’s—and Chen Shou’s—actions and decisions in terms of how Lu Xun and Chen Shou came across as a person. Later on, the literary critic Liu Xie 劉勰 (ca. 465–ca. 522), in reviewing xi (“proclamation”), a highly functional and propagandistic genre used to great effect in denouncing an enemy in the late Eastern Han and Three Kingdoms period, explicitly acknowledged its function as “a military tactic of deception” (bing zha 兵詐), spelling out “unscrupulous craftiness” (juegui 譎詭) and “dazzling persuasion” (weiye 煒曄) as its general principles.⁸⁷ Liu Xie’s formulation reveals an attempt at embedding the instrumental view of literature within the pursuit of literary and aesthetic values, which constituted a focus of attention during his time. Did his claim of the aesthetics of xi betray a rehashing of the unease with the use of deceptive tactics? Or was it simply a clever reconciliation of the pragmatic and aesthetic functions of literature? While Liu Xie’s statement deserves to be understood in its own historical context, it is the first in Chinese history to explicitly acknowledge that literature can act—and has acted—as a tool of war, deceiving, persuading, and manipulating for the purpose of bringing down the other side. Liu Xie mentions war deception again in his chapter on “Yang qi” 養氣 (“Cultivating Breath”), noting there that “during the Warring [States] era, the tactics turned deceptive, as surprised attacks and embellished persuasions were used” 戰代枝/技詐, 攻奇飾說.⁸⁸ He seems to suggest that the phenomenon marked a turn from the earlier tradition of using “pure language” (chun yan 淳言), leading to the eventual draining of “thoughts” (lü 慮) and “feelings” (qing 情)—and hence, “breath/energy” (qi 氣).⁸⁹ There is certainly no praise for deception here, highlighting Liu Xie’s attempt at both acknowledging the place for deception as a mode of writing and relegating it to a marginal position in literary history.

The diverse commentaries elicited by the practice of “writing in deception” prove the indelible significance of the works and experiences of the generation that lived through the period of war and chaos after the collapse of the Han empire. While Pei Songzhi and Liu Xie’s comments reflect its mixed reception in later history, “writing in deception” as a mode of writing brings the nature and conception of writing in third-century China into sharp relief. In this respect, it is worth linking the discussion to the writing of poetry from the same period. In thinking about the fluid relationship between the author and the speaker—what I have referred to as “the voice” here—in the conceptualization of poetry in the third and fourth centuries, Stephen Owen has identified four categories of poems that reflect “a range” of such a relationship: 1) poems that are anonymous, i.e. of unknown authorship; 2) “poems of unknown authorship that seemed appropriate for a certain person to have uttered”; 3) poems by known authors who “assumed the persona of some historically known character,” later called daizuo 代作; and 4) poems by known authors speaking in their own voices about their own cases.⁹⁰ As Owen has alluded to a wide variety of textual materials (“stories, ideas, political positions, popular analogies … ”⁹¹) in contextualizing his discussion, it would seem reasonable that, by replacing the word “poems” in the descriptions of the four categories of author-speaker relations with “writings” or “works,” the four categories can also be applied broadly to literary writings in the period in general. If this is the case, which category would the altered or fabricated letters discussed in this article fit into? The answer would have to be: all four categories. By this answer, I am not speaking of “what these letters truly were,” but of how they could have been received, imagined, or situated at the time. It is hard to ascertain how conscious an audience in a particular context at the time was in differentiating “the author” from “the speaker/voice,” but, as Owen’s discussion indicates, the literature of third-century China challenges us to think of authorship, just as how we should think about authenticity, legitimacy, genuineness, sincerity, and other properties of a literary work, as the results—that is, the effect and affect—of writing, and not the other way around. If, as Owen points out,

a known author … speaking in his own voice about his own case … is a privileged mode in poetry, and its status exerted a pressure on all the other modes of composition and utterance. … That is, poems are in search of authors and particularly in search of an author who is speaking about his own case,⁹²

then it was this “quest” for an author who was also the speaker that the writers of altered or fabricated letters greatly manipulated. Even more crucially, as “[t]he property of authorship is inseparable from acts of valuation; valuation, in turn, gives force to judgments of authenticity,”⁹³ it was ultimately certain ways or habits of valuation that the forger-writer was manipulating. As the three cases of altered or fabricated letters demonstrate, writing worked “with” acts of valuation not simply to comply with, reinforce, or subvert them, but also to instrumentalize them for its own purposes.

Acknowledgments

I want to express my greatest thanks to the anonymous reader of my article and to Xiaofei Tian for their insightful suggestions and corrections, which greatly benefited my revision. I also want to extend my deepest gratitude to Robert Joe Cutter, Kang-i Sun Chang, Xiaofei Tian, and Richard VanNess Simmons for giving me the chance to present earlier versions of this article to excellent audiences, whose comments and questions greatly helped my exploration in this study. All remaining mistakes and flaws are my sole responsibility.

Additional information

Notes on contributors

Meow Hui Goh

Meow Hui Goh is Associate Professor in Chinese Literature at The Ohio State University. She specializes in the literature, literary history, and cultural history of medieval China, and is the author of Sound and Sight: Poetry and Courtier Culture in the Yongming Era (483–493).

Notes

1 For a modern account of Dong Zhuo and Yuan Shao’s rise, see Rafe de Crespigny, Fire over Luoyang: A History of the Later Han Dynasty 23–220 AD (Leiden: Brill, 2017), 449–73.

2 Unless otherwise stated, my English translation of Chinese official titles follows Charles O. Hucker, A Dictionary of Official Titles in Imperial China (Stanford: Stanford University Press, 1985).

3 Cited in Pei Songzhi’s 裴松之 (372–451) commentary to Chen Shou 陳壽 (233–297), comp., Sanguo zhi 三國志 (1959; rpt. Beijing: Zhonghua shuju, 1982), 1.6; also see Fan Ye 范曄 (398–445), comp., Hou Han shu 後漢書 (1965; rpt. Beijing: Zhonghua shuju, 2001), 74a.2376–77.

4 Two examples of the use of writing for deception in earlier time are in Sima Qian 司馬遷 (ca. 145–ca. 85 BCE), Shi ji 史記 (1959; rpt. Beijing: Zhonghua shuju, 2002), 12.458; Hou Han shu, 38.1280.

5 The source of this account, which is cited in Pei Songzhi’s commentary, is Yingxiong ji 英雄記, attributed to Wang Can 王粲 (177–217). For this work, see the “Bibliographical Treatise” 經籍志, in Wei Zheng 魏徵 (580–643) et al., comps., Sui shu 隋書 (1973; rpt. Beijing: Zhonghua shuju, 2014), 33.960.

6 Some examples are found in Li ji zhengyi 禮記正義, 2:37.1084; Zhou li zhushu 周禮注疏, 1:14.368; and Chunqiu Zuozhuan zhushu 春秋左傳注疏, 2:29.838. In Shisanjing zhushu 十三經注疏, ed. Li Xueqin 李學勤, 13 vols. (Beijing: Beijing daxue chubanshe, 1999). In these examples, wei and zha are sometimes used together as a compound word (zhawei) and sometimes separately as two words. The Er ya 爾雅 (ca. 3^rd c. BCE) glosses zha as wei ye 僞也. Er ya zhushu 爾雅注疏, in Shisanjing zhushu, 13:2.36.

7 Li ji zhengyi, in Shisanjing zhushu, 2.38.1116. This formulation of the efficacy of rituals does not necessarily imply that sincerity exists a priori to ritual performances. For a study of the tension between rituals and sincerity that includes consideration of Confucianism, see Adam B. Seligman, Robert P. Weller, Michael J. Puett, and Bennett Simon, Ritual and Its Consequences: An Essay on the Limits of Sincerity (Oxford: Oxford University Press, 2008), especially chapters 1, 4, and 5.

8 Zhou li zhushu, in Shisanjing zhushu, 1:14.368.

9 Krzysztof Gawlikowski and Michael Loewe, “Sun tzu ping fa 孫子兵法,” in Early Chinese Texts: A Bibliographical Guide, ed. Michael Loewe (Berkeley, CA: The Society for the Study of Early China, 1993), 446.

10 Chen Qitian 陳啓天, annot., Sunzi bingfa jiaoshi 孫子兵法校釋 (Taipei: Zhonghua shuju, 1944), 215–30.

11 Chen Qiyou 陳奇猷, comp., Han Feizi jishi 韓非子集釋 (Beijing: Zhonghua shuju, 1958), 15.192.

12 See Li Disheng李滌生, comp., Xunzi jishi 荀子集釋 (Taipei: Taiwan Xuesheng shuju, 1979), 311–17; the “Yishang” 義賞 chapter, in Wang Liqi 王利器, annot., Lüshi chunqiu zhushu 呂氏春秋注疏, 4 vols. (Chengdu: Bashu shushe, 2002), 2:14.1484–97.

13 Christopher C. Rand, Military Thought in Early China (Albany: State University of New York, 2017), 93–94. Wei Liao zi is traditionally attributed to Wei Liao 尉繚 (fl. late 4^th c.). For the discussion in Wei Liao zi on which Rand’s statement is based, see Wei Liao zi, annot. Liu Zhongping 劉仲平 (Taibei: Shangwu yinshuguan, 1984), 68.

14 Shi ji, 109.2874.

15 Shi ji, 109.2874.

16 To Cao Cao is attributed a work titled Sunzi lüejie 孫子略解. See Yan Kejun 嚴可均 (1762–1843), comp., Quan sanguo wen 全三國文, 1.1055a, in Yan Kejun, comp., Quan shanggu sandai Qin Han sanguo liuchao wen 全上古三代秦漢三國六朝文 (Beijing: Zhonghua shuju, 1958). A “Sunzi bingfa xu” 孫子兵法序 by Cao Cao is still extant. See Yan Kejun, comp., Quan sanguo wen, 3.1070b–71a.

17 One such work is Yingxiong ji, mentioned in an earlier note, which provides the account about Qiao Mao’s fabricated proclamation.

18 Sanguo zhi, 58.1352, 51.1205; Hou Han shu, 10b.453–54.

19 See Pei Songzhi’s commentary to Sanguo zhi, 46.1097.

20 The letter is cited in Sanguo zhi, 60.1387–90.

21 The Shi ji account of the Tian Dan story reflects the use of human agents and verbal means in earlier time of war deception: “Tian Dan … thereupon gave free rein to turned agents to infiltrate Yan, who spread the words that … ” 田單 …  乃縱反閒於燕, 宣言曰. Shi ji, 82. 2454.

22 Liu Fang’s biography is in Sanguo zhi, 14.456–62. Liu Fang was a Filial and Incorrupt (xiaolian 孝廉) who rose to serve four Wei sovereigns and, at the time of his death, was one of the highest-ranking officials in the Wei regime.

23 Man Chong’s biography is in Sanguo zhi, 26.721–25.

24 Sanguo zhi, 14.457. With no access to the altered letter, we do not know exactly why Sun Quan was worried that Zhuge Liang would be suspicious of him and what suspicion he was worried about, but the context here suggests that Liu Fang must have somehow created the impression that Sun Quan was untrustworthy through the changes that he made to the letter. I thank Antje Richter for helping me with the translation of this passage.

25 For a description of the different materials used for letter writing during the Han period, see Eva Yuen-wah Chung, “A Study of the Shu (Letters) of the Han Dynasty (206 B.C.–A.D. 220)” (Ph.D. dissertation, University of Washington, 1982), 194–211. For different aspects of the materiality of a letter in the early medieval period, including the calligraphy and the use of transporters, see Antje Richter, Letters & Epistolary Culture in Early Medieval China (Seattle: University of Washington Press, 2013), 17–34.

26 I thank the anonymous reader at EMC for suggesting this plausible explanation for what took place here.

27 Sanguo zhi, 8.261. Gongsun Zan’s biography is in Sanguo zhi, 8.239–44.

28 Sanguo zhi, 8.244–45.

29 Sui shou 碎首, “smash one’s head,” is a term often used to express or describe a subordinate’s willingness to die for speaking truth to a ruler; here, it means pledging complete loyalty or allegiance. Zhang Yan, called the “Black Mountain bandits” (Heishan zei 黑山賊), rose as a regional power during the Yellow Turban rebellion. Known for his military skill, he attracted tens of thousands of followers. In the post-Dong Zhuo era, he often collaborated with a warlord to fight against another warlord. In the battle between Yuan Shao and Gongsun Zan, he took sides with the latter. Zhang Yan’s biography is in Sanguo zhi, 8.261.

30 Pei Songzhi’s commentary to Sanguo zhi, 8.246–47. Pei Songzhi cited Dian lüe, which has been attributed to the Wei figure Yu Huan 魚豢 (see Sui shu, 33.961), as the source of Gongsun Zan’s alleged letter.

31 Pei Songzhi’s commentary to Sanguo zhi, 8.247. The “Bibliographic Treatise” in Sui shu lists a Yuan Ye 袁曄, who is not attested in extant sources from Jin or earlier periods, as the author of Xiandi chunqiu. Sui shu, 33.957. Some scholars have identified him as a Jin dynasty figure. See Yang Jicheng 楊繼承, “Wei Jin shiji zhong de Cao Cao chengwei—ji yu Pei zhu suo yin ‘yishi’ de kaocha” 魏晉史籍中的曹操稱謂—基於裴注所引“佚史”的考察, in Wei Jin nanbeichao Sui Tang shi ziliao 魏晉南北朝隋唐史資料 40 (2019): 52.

32 It was also possible that Dian lüe did have a longer citation, which was excerpted in Pei Songzhi’s commentary. In this sense, Chen Lin did change the letter (geng qi shu 更其書), rather than simply “added to the letter” (yi qi shu 益其書). I thank Xiaofei Tian for suggesting this possibility to me.

33 One reflection of the concentrated use of historical references is in the prose works of this period, such as memorials (shu 疏, biao 表, zhang 章 or zou 奏) and disquisitions (lun 論 or yi 議), where the conveyance of meanings and ideas are sometimes completely reliant upon the clustering of historical references and canonical citations. For example, see Yang Fu’s 楊阜 (fl. 213–235) memorials in Sanguo zhi, 25.704–5, 707–8.

34 By this statement I am not implying that there were fixed or unchanging patterns or principles to which history was or could be attached. Rather, what I mean is the idea that “[h]istorical anecdotes were polemical, political, and philosophical weaponry” and that “[n]ew judgements often came to be attached to anecdotes as they were transmitted.” David Schaberg, A Patterned Past: Form and Thought in Early Chinese Historiography (Cambridge, MA: Harvard University Asia Center, 2001), 4.

35 Sanguo zhi, 1.8; Pei Songzhi’s commentary in Sanguo zhi, 8.241–42.

36 The poem is none other than “Xie lu” 薤露. For Dièny’s discussion of the poem, see his Les Poèmes de Cao Cao (155–220) (Paris: Collège de France, Institut des Hautes Ètudes Chinoises, 2000), 27–31.

37 Sanguo zhi, 62.1414.

38 The Sanguo zhi describes the three parts of the letter as san tiao 三條, “three slips” (Sanguo zhi, 62.1415), which, though not entirely clear in meaning to us, hints at the material form of the letter as it was known. Incidentally, this description is not the only instance in which a piece of writing is referred to by tiao in the Sanguo zhi (for other examples, see Sanguo zhi, 15.482, 59.1368). In one account already cited as an example earlier, while the narrative about the letter in question describes it as consisting of qi tiao 七條 (“seven slips”), the text of the letter contains the letter writer’s reference to the rest of his letter as appearing in “other pieces of paper” (Sanguo zhi, 60.1387). This example suggests that tiao was used as a measure word for “paper letters” (zhishu 紙書), if not exclusively, perhaps as a transference from its use for “bamboo-slip letters” (jianshu 簡書).

39 Sanguo zhi, 62.1415.

40 Shen Yue 沈約 (441–513), comp., Song shu 宋書 (1974; rpt. Beijing: Zhonghua shuju, 2000), 21.605–6; Lu Qinli 逯欽立, ed., Xian Qin Han Wei Jin nanbeichao shi 先秦漢魏晉南北朝詩 (Beijing: Zhonghua shuju, 1995): 347.

41 The “treacherous subordinate” refers to Dong Zhuo.

42 The reference here is to Dong Zhuo’s murder of Emperor Shao 少 (Liu Bian 劉辯; r. 189). This and the next two lines also evoke Dong Zhuo and his soldiers’ destruction of the capital city Luo Yang by looting, killing, and burning.

43 When Dong Zhuo moved Emperor Xian 獻 (Liu Xie 劉協; r. 189–220) from Luoyang to Chang’an, he also forced several hundreds of thousands of households to make the westward migration.

44 Song shu, 21.606; Lu Qinli, Xian Qin Han Wei Jin nanbeichao shi, 347.

45 Sanguo zhi, 62.1415.

46 Sanguo zhi, 62.1415.

47 Sanguo zhi, 62.1415.

48 This letter contains the only mention of Zhou Guang in Sanguo zhi and there are no references to him in Hou Han shu or Jin shu.

49 Sanguo zhi, 62.1416.

50 Sanguo zhi, 62.1417.

51 Xu You was known to have been friendly with both Yuan Shao and Cao Cao when he was young and was one of several major strategists of Yuan Shao before he came to serve Cao Cao. See Pei Songzhi’s commentary to Sanguo zhi, 6.195, 12.373. There are other, brief mentions of Xu You in Pei Songzhi’s commentary, e.g., Sanguo zhi, 6.188, 10. 322. Yan Fu and Zhao Ji are mentioned only in this fabricated letter in Sanguo zhi.

52 Sanguo zhi, 62.1417.

53 Sanguo zhi, 62.1417.

54 See Meow Hui Goh, “The Art of Wartime Propaganda: Chen Lin’s Xi Written on Behalf of Yuan Shao and Cao Cao,” Early Medieval China 23 (2017): 45–46.

55 The different responses of Deng Ai 鄧艾 (195–264) and Huo Yi 霍弋 (d. ca. 271) to the issue of qujiu provides a snapshot of this debate: Deng Ai argued that men of talent should submit to “the true sovereign [chosen by Heaven],” adding that “this is Heaven mandating the way for ‘leaving [one lord] to approach [another]’” (see Pei Songzhi’s commentary to Sanguo zhi, 33.901); Huo Yi, by contrast, emphasized loyalty, prioritizing the safety and dignity of one’s sovereign (see Pei Songzhi’s commentary to Sanguo zhi, 41.1008).

56 Sanguo zhi, 62.1416.

57 The portions of the two rivers Huai and Si in question here, as well as Xiapi, were all in modern Jiangsu province. Jingzhou was in modern He’nan province and Hubei province. Yangzhou stretched across modern Anhui, Jiangsu, Jiangxi, Zhejiang, Fujian, Hubei, and He’nan.

58 Hebei, as the name indicates, was just north of the Yellow River in modern Shanxi province.

59 Sanguo zhi, 62.1416.

60 Sanguo zhi, 62.1417.

61 Pei Songzhi’s commentary to Sanguo zhi, 21.610. The reason that Wu Zhi was posthumously titled Shameful Marquis was allegedly because he had “relied on [borrowed] authority to act without restraint” 怙威肆行 while alive (Pei Songzhi’s commentary to Sanguo zhi, 21.610); what the allegation exactly refers to is not known but some accounts suggest that Wu Zhi had had conflicts with several of his fellow officials (see Pei Songzhi’s commentary to Sanguo zhi, 21.609–10). That Wu Zhi was a politically controversial figure in Cao Rui’s court can perhaps be attested by the brevity of his Sanguo zhi biographical note (in only one line; see Sanguo zhi, 21.607), which is particularly jarring when we consider his influence and status during Cao Pi’s time.

62 Sanguo zhi, 62.1416.

63 Sanguo zhi, 62.1416.

64 Lu Yun, “Yu xiong Pingyuan shu” 與兄平原書, in Lu Shilong wenji jiaozhu 陸士龍文集校注, annot. Liu Yunhao 劉運好 (Nanjing: Fenghuang chubanshe, 2010), 1044.

65 On Chen Lin and Hu Zong’s literary reputation in their own day, see note 81 below; for the case of Liu Fang, see note 82 below. It is worth noting that Wu Zhi was also said to be favored by Cao Pi due to his literary talent. Sanguo zhi, 21.607.

66 Sanguo zhi, 21.600.

67 Sanguo zhi, 21.600. The translation of the official title “junmou jijiu” 軍謀祭酒, which is not found in Hucker’s A Dictionary of Official Titles in Imperial China, is my own. Ruan Yu, like Chen Lin, was recognized by Cao Pi as one of the “Seven Masters” (“Qizi” 七子) for his literary talent. See Cao Pi, “Lun wen” 論文, in his Dian lun 典論, in Yan Kejun, “Quan Sanguo wen,” 8.1098a.

68 Sanguo zhi, 62.1418.

69 Sanguo zhi, 14.457.

70 Robert Ashmore in a 2018 study makes the following observation about the “proxy texts” produced in the ninth century: “[m]astering of the demanding art of drafting proxy texts was held in the highest social and cultural esteem, and the ideal of the ci chen 詞臣, or ‘rhetorical minister,’ who possessed such attainments, was a model of governmental service linked to the highest and most elite postings at court and in the regional administrations.” Ashmore, “The Mastering Voice: Text and Aurality in the Ninth-Century Mediascape,” in Memory in Medieval China: Text, Ritual, and Community, ed. Wendy Swartz and Robert Ford Campany (Leiden: Brill, 2018), 243–66. There is no evidence that, more than six centuries earlier, there was the master-disciple model of learning to write elegant “proxy texts” or the collection, circulation, and transmission of outstanding examples of this type of work through copying and anthologization, which have been observed for the late Tang period; but, in the three cases studied here, we clearly see earlier examples of the precursor of the Tang proxy-writing culture. It is worth considering that the practice of employing talented writers to compose official documents had always existed in Chinese political culture.

71 For example, the Caos were the precursors of a type of poetry written in the voice of a contemporaneous woman—the wife of an acquaintance, the daughter of a famous public figure, or a sister. Examples include Cao Pi’s “Guafu shi” 寡婦詩, Ding Yi’s 丁廙 (d. 220) “Cai Bojie nü fu” 蔡伯喈女賦, and Cao Zhi’s 曹植 (192–232) “Xu chou fu” 敘愁賦, and so on.

72 Wells, To Die and Not Decay: Autobiography and the Pursuit of Immortality in Early China (Ann Arbor: Association for Asian Studies, 2009), 7.

73 Genette, Palimpsests: Literature in the Second Degree, trans. Channa Newman and Claude Doubinsky (Lincoln: University of Nebraska Press, 1997), 161. It should be noted that by “imitation,” Genette is fundamentally dealing with the relationship between or among texts, as his definition of transtextuality makes clear (Palimpsests, 1). The same can be said about scholars of early medieval Chinese literature who have examined the practice of “imitation,” whether called ni 擬, fang 仿, dai 代, or xiao 效; see, for example, Stephen Owen, The Making of Early Chinese Classical Poetry (Cambridge, MA: Harvard University Asia Center, 2006), 260–97; Brigitta Ann Lee, “Imitation, Remembrance and the Formation of the Poetic Past in Early Medieval China” (Ph.D. dissertation, Princeton University, 2007); Nicholas Morrow Williams, Imitations of the Self: Jiang Yan and Chinese Poetics (Leiden: Brill, 2015). A full discussion to compare and contrast “imitation” and “writing in deception” will require much more space than the present article can afford, but, as I have demonstrated thus far, “writing in deception” and, more broadly, “writing in the voice of another” did not concern textual imitation but rather the usurpation or embodiment, to various degrees, of the voice, and identity, of another person.

74 Genette, Palimpsests, 161.

75 Genette, Palimpsests, 161.

76 Sanguo zhi, 14.457.

77 Sanguo zhi, 62.1414–17. Here, Chen Shou offers praise for Hu Zong’s “covenant,” saying “the language and implied meaning are exceedingly beautiful”文義甚美 (Sanguo zhi, 62.1414), while noting that the text of the piece is in Sun Quan’s biography (Sanguo zhi, 47.1134–35).

78 Sanguo zhi, 62.1414.

79 There is no hint of any unease toward fabrication by writing in either the account about Qiao Mao’s fabrication of a “Three Dukes’ proclamation” or that about Chen Lin’s alteration of Gongsun Zan’s letter.

80 Shangshu, in Shisanjing zhushu, 2:18.487–88.

81 Chen Lin and Hu Zong are two cases in point. Better known is the case of Chen Lin, who, when he was serving Cao Cao’s competitor Yuan Shao, had composed a proclamation against Cao Cao on Yuan Shao’s behalf, in which he not only makes numerous charges against Cao Cao himself, but also makes negative insinuation against Cao Cao’s father and grandfather. Reportedly, though he confronted Chen Lin about it, Cao Cao did not punish him for it, as he “loved his talent” 愛其才 (Sanguo zhi, 21.600). Though less well known, there is a similar story about Hu Zong. He was said to indulge in drinking, and while drinking he would often behave in an unrestrained manner, cheering loudly at will, pushing around the drinking cups, or even grabbing and hitting those around him. Sun Quan reportedly never criticized him for such behavior because he “loved his talent” (Sanguo zhi, 62.1418).

82 Sanguo zhi, 14.461. Chen Shou commented on Liu Fang and his fellow courtier Sun Zi 孫資 (d. 251) together, opining that Liu Fang exceeded Sun Zi in talent but lagged behind Sun Zi in self-cultivation. Liu Fang’s career in the Cao regime began with a letter that he wrote on behalf of Wang Song 王松 (fl. 3^rd c.). The letter, written in response to Cao Cao’s letter to Wang Song to recruit him, was “rather lavish” and greatly impressed Cao Cao. Sanguo zhi, 14.456.

83 Sanguo zhi, 58.1352.

84 Sanguo zhi, 58.1352.

85 Sanguo zhi, 14.457.

86 Pei Songzhi’s commentary to Sanguo zhi, 58.1352.

87 Zhan Ying 詹鍈, Wenxin diaolong yi zheng 文心雕龍義證 (Shanghai: Shanghai guji chubanshe, 1999), 780.

88 Liu Xie, Wenxin diaolong yizheng, 1567.

89 Liu Xie, Wenxin diaolong yizheng, 1569–70.

90 Owen, The Making of Early Chinese Classical Poetry (Cambridge, MA: Harvard University Asia Center, 2006), 220–22.

91 Owen, The Making, 215.

92 Owen, The Making, 222.

93 Owen, The Making, 218.