引 言
（Ⅰ）待时而举：从公元前6世纪到公元前3世纪的晏子形象
首先出现在历史中的，当然是晏婴这个人。在他去世多年以后，出现了“晏子”——《晏子春秋》中的同名主人公。作为公元前6世纪生活于齐国的一个历史人物，有关于他的生平的可靠史料非常稀少并且难以考证。尽管如此，从公元前5世纪到公元前4世纪，晏婴的个人特征逐渐合并汇聚成一个独特的形象。最初，晏婴只是在《论语》中被简单提及，《论语·公冶长》中记载：子曰：“晏平仲善与人交，久而敬之。”接着，晏婴在《左传》中首先作为历史人物登场。《左传》中记载了晏婴的几个故事，这些故事主要发生在公元前6世纪下半叶齐国公室衰微、田氏即将取而代之这一历史背景之下。所有这些故事集中在一起塑造出一个极具吸引力的形象，作为齐国的大臣，晏婴善于巧妙进谏，同时，在侍奉三位不称职的君主的时候，他坚持原则、不畏权势。
接着，在公元前4世纪下半叶，晏子的形象开始在文学中产生影响，其形象毁誉参半。稷下学宫的成员之一淳于髡（生活于公元前340年左右），在向齐威王（公元前378年—前320年）进谏时曾经充满赞美地称引晏婴。然而，孟子（公元前372年—前289年）的弟子，来自齐国的公孙丑，坚持将孟子与管仲和晏子比较的时候，却使孟子觉得受到冒犯。孟子傲慢地回应弟子的提问，否认管仲和晏子作为两位伟大的齐国大臣的重要性，他把他们看作是严格受限于他们所处的时代与地位的两位普通官员。稍后于孟子，在荀子（公元前310年—前235年，一说公元前314年—前217年）的著作中，同样也记录了对待晏子和管仲的轻蔑态度。《荀子·大略》中云：“子谓子家驹续然大夫，不如晏子；晏子功用之臣也，不如子产；子产惠人也，不如管仲；管仲之为人，力功不力义，力知不力仁，野人也，不可以为天子大夫。”管仲尚是“野人”而“不可以为天子大夫”，晏婴的成就还不如管仲，那么他对待晏婴的态度就可想而知了。
（Ⅱ）粉墨登场：从公元前3世纪到公元前2世纪的晏子形象
在公元前300年到公元前200年之间，晏子的形象逐渐发生了变化。无论是在包含有晏婴形象的故事中，还是在冠上他的名字的文献记载里，晏婴都变成一个越来越中心化的人物。直到公元前3世纪中期，晏婴的所作所为，无论是被赞许还是被批评，一直被文本中他所处的历史环境所遮蔽，晏婴的形象超越了自我言说所展现的个性特征，服务于更广泛的目的。然而在公元前250年左右，在叙述中处于边缘地位的晏子形象逐渐占据了中心位置。随后，晏婴这一人物被广泛引用并受到越来越多的关注。他的形象也越来越复杂——有时候具有示范意义，有时候又暧昧不清——并且最终形成一系列极具特色的轶事，这些轶事不仅聚焦于他的所作所为，同时也关涉到他具有独特思维模式的个性特征。这些主角化的变化反映在当时三部重要的著作中：《墨子》、《吕氏春秋》和《韩非子》。
在《墨子》的第三十九篇《非儒》中，晏子不再是一个附属的人物，而是成为墨家学派中“非儒”的一名主将。他被描绘为齐君的密友与顾问，他对于儒家的世界观和过于现实的做法的一系列敌对观点促成了齐君站在墨家立场上反对儒家。晏子的态度非常激烈；他不仅谴责孔子本身是一个谋逆者甚至是一个凶手，还批判儒家的核心价值，指出儒家的鼓吹者是一群进行无用表演的人。从此以后，晏子不仅成为一个墨家学者，而且变成了一位反对孔子和儒家学派的领头人，晏子的这一身份在其他一些文献和注释中同样也有所反映。
在《吕氏春秋》中，晏子作为一个道德高尚的、善于自省的、无畏的独特模范而发挥作用。书中有三段记载，描写了晏子在绝望和生命受到威胁的情况下的举动。在第一段记载中，晏子的仁慈与道德使得他人愿意为他牺牲自己的生命。在第二段记载中，他的仁慈的行为描画出他极为可贵的自省与自律的能力，一个完美主义者可以从他的全部行为中意识到其中所蕴含的意义。在第三段记载中，当晏子面对迫在眉睫的死亡威胁的时候，他所展现出的伟大的真正的勇气足以作为典范。
在《韩非子》中有三段非常尖锐地批评了晏子对君主关切的问题所做的分析、他仁慈的救济主张以及他有关于严刑与缓刑的心理上的考虑。看起来《韩非子》似乎把晏子当作是一种具有威胁性的世界观的体现，从法家的立场上看，必须对其进行挑战与批驳。因而，晏子在文本中受到激烈的哲学批判，甚至他的一些论点从某种程度上被看作是“不忠诚”的体现。这三段中的第一段记载抨击了晏子著名的节俭与节制的特点。文章不仅论证了这两种价值的无意义，因为他们在某种程度上导致大臣们对获得成功缺乏动力，更宣称孔子曾经专门批评过晏子。第二段记载则认为晏子对于官僚系统和百姓的仁慈与慷慨反映出他缺乏摆脱困境的能力，从而会使他的君主面临更加严重的灾难。最后，在第三段中则抨击了晏子的缓刑主张，认为晏子自己也不相信自己的论点。对于《韩非子》来说，晏子的宽缓态度表明晏子并未清楚地了解治理的意义。
（Ⅲ）独奏者：汉代前期的晏子形象与《晏子》文本
（Ⅲ-1）《淮南子》中的“晏子谏”
所有这些文本的例子表明在公元前200年左右，晏子的形象不仅在相关的叙述中扮演着越来越重要的角色，而且很有可能促成了某些独特文本的形成，这些文本仅关注晏子的历史，有些甚至把他的名字作为标题。公元前139年，晏子形象的演进变得非常明显。在这一年，汉宗室刘安向汉武帝进献了一部百科全书式的哲学著作《淮南子》。《淮南子》第21篇《要略》，目的在于总结分析之前二十篇的内容，最后又通过回顾和解释那些在过去扮演着重要角色的杰出人物之所以开创新说的原因为全书的结束。《要略》中说：
首先，《淮南子》在这里交代了“晏子之谏”这一文本之所以产生的背景。其次，晏子对齐景公的“谏”，不仅仅是今传本《晏子春秋》的核心内容，而且还是刘向（公元前77年—前6年）整理的《晏子》的前两章的标题。最后，《淮南子》所列举引起晏子之“谏”的齐景公的乱政，与今传本《晏子春秋》中近一半的内容相符合。我们据此或许可以得出这样的认识，在公元前139年左右，存在着一部题名与“晏子”相关的书，这部书的内容与今传本《晏子春秋》的部分内容非常接近。实际上，大约五十年之后，晏子已经被看作是一位直言敢谏、充满勇气的大臣，并且他还成为当时流行于世的《晏子春秋》的作者。晏子的形象，最终被司马迁（约公元前145年—前86年）《史记》中那些令人难以忘怀的传记记载所定型。
（Ⅲ-2）《史记》中的晏子形象与《晏子春秋》
晏子的传记记载于《史记》的第六十二卷，司马迁把管仲的传记放在晏子之前写在同一章《管晏列传》中。在晏子的传记中，前两段专门叙述晏子，而后面两段则是将管仲和晏子放在一起讨论。在《史记》卷三十二《齐太公世家》中，也记载了来自《左传》的有关于管仲和晏子的很多传记资料。读者应该注意到，《齐太公世家》中有关于管子的内容，被概括进了《管晏列传》，而有关于晏子的内容则没有被司马迁记录到《管晏列传》中。另外我也希望读者在开始阅读《管晏列传》之前，先关注一下司马迁在《管晏列传》的结尾所做的总结。司马迁在其中表达了自己对于晏子极深的敬意，他说：“假令晏子而在，余虽为之执鞭，所忻慕焉。”
司马迁对于晏子的敬意，在整篇《列传》中得到彻底的体现，也影响到其内容的选择。司马迁在叙述完管仲的事迹之后说：“管仲卒，齐国遵其政，常强于诸侯。后百余年而有晏子焉。”司马迁用这句话将管仲和晏子联系在一起，使二者的传记构成了一个完整的历史序列。虽然管仲与晏子在历史上所发挥的作用不可同日而语，但是读者却不得不对二人做出比较：管仲，一位杰出的政治家，建立了齐国的霸业、塑造了齐国一百年的历史；而晏子，一位杰出的谏臣，却无法挽救齐国公室被取代的命运。在晏子传记的开头，司马迁写道：
读完晏子传记的第一部分，再回想管仲的传记，读者可以看到管仲在齐国历史上的重要时刻发挥着关键作用，并塑造了齐国后来的历史。然而，在晏子的生平中却没有类似的叙述。晏子传记的开头部分，仅仅是简单地介绍了晏子的名字、籍贯以及侍奉的君主。读者从中看不出晏子对于齐国历史有任何贡献，这里既没有涉及任何重要的历史事件，也没有可以让人留下深刻印象的人生事业。相反，这些记述倒是比较详细地描述了晏子的个性和道德行为的特点，仿佛他是一部缺少故事主线的书中的主人公。因此，当读者读完晏子传记的开头部分，会觉得晏子在齐国历史的地位远逊于管仲。此外，除去粗疏的大背景，读者对于晏子生平的主要事迹几乎毫无线索。这也就是说，晏子之所以在当时出名，是因为他节俭、勇毅、正直、诚实与大胆等个性特征，而这也成为吸引后人注意的主要原因。
从这一点上来看，读者会对司马迁将二者合并为一个列传的做法感到困惑。然而，根据列传最后的总结，读者或许可以多少理解这一令人费解的处理方法。司马迁对晏子的为人推崇备至，他竭力缩小管仲与晏子这两位齐国大臣在历史影响力上的巨大差距。为了达到这一目的，司马迁将管仲的历史成就与晏子的高尚德行相并列，从而避免了对二者历史作用的直接比较。
晏子传记的第二个部分包含了有关于晏子的两个故事，这两个故事也出现在今传本的《晏子春秋》中。第一个故事相当于《晏子春秋》第五卷第24篇“晏子之晋睹齐累越石父解左骖赎之与归”的简写。这个故事讲述晏子用自己的一匹马赎回了贤人越石父，之后晏子用一般性的礼节对待他，这引起了越石父的不满与抗议，晏子立刻忏悔自己的过失并且待之如上宾。第二个故事则与《晏子春秋》第五卷第25篇“晏子之御感妻言而自抑损晏子荐以为大夫”完全相同。在这段故事中，晏子的车夫为晏子驾车时得意扬扬，车夫的妻子看到后要求与他离婚，并且说，晏子身不满六尺，相齐国，名显诸侯，然而态度谦卑，而你身长八尺，为人驾车，反而志得意满。这番话令车夫悔过，并从此改变了自己的态度，晏子觉察到车夫的变化，便推荐他担任了齐国的大夫。
这两个故事非常微妙地提出一个隐含的问题：一个人的真正身份到底由什么构成？从这一角度来看，这两段故事既可以从哲学的层面来分析，也可以从心理学层面来讨论。而在这两段故事中，除了暗示晏子是一个小个子之外，再无有关于他生活的任何其他信息。看起来，这两段故事同样也是通过描述晏子的美德来向读者展现晏子两方面的人格特点：他不遗余力地解救处于困境中的人，并且他善于自省从而有能力调整自己在复杂情况中的态度；另外，他有一种巨大的超越表象和探寻内在品质的能力。
司马迁在晏子的传记中记录了这两个故事，到底想要表达什么呢？有鉴于他在列传最后的总结中对晏子所表达的“忻慕”之情，相比于对管仲的记载，关于晏子的记载无论在字数上还是在内容上都显得比较单薄。
在列传的最后部分，司马迁说：“吾读管氏《牧民》、《山高》、《乘马》、《轻重》、《九府》及《晏子春秋》，详哉其言之也。既见其著书，欲观其行事，故次其传。至其书，世多有之，是以不论，论其轶事。”
这段话在晏子文本的历史上是非常重要的一段记载。司马迁非常明确地指出他曾经读过管仲的五篇论述以及《晏子春秋》这部书。
司马迁表示《晏子春秋》为晏子所作。他同时还指出，这部书写得很详细，在当时很流行。司马迁的这段话充分证明在大约公元前100年左右，《晏子春秋》已经被看作是有关于晏子的真实传记。《史记》中将其题名为“晏子春秋”，并且这个名字多次出现在汉代学者的记述中，“春秋”本来是史书的名字，既然这部书题名为“春秋”，也证明其书写了有关于晏子的历史。
最后，司马迁对管仲和晏子分别做出评价。对于管子，司马迁含蓄地表达出严厉的批评态度：
所谓“孔子小之”，是指《论语》中记载孔子曾经说过“管仲之器小哉”。司马迁通过引用孔子的观点，将之前所记载的管子的所有成就进行了贬损。相反，对于晏子，司马迁却表达出强烈的赞美以及将其看作是个人偶像的崇敬：
《管晏列传》至此结束，司马迁在最后表达出自己渴望侍奉晏子的崇敬之情，这使读者反思这部《列传》中的两位主人公，到底哪一位才是齐国更加杰出的人物。司马迁在《史记》中非常聪明地将二人联系到一起，而他的目的也很快达到，《史记》成书后不久，晏子已经成为与管子相并列并且在此基础上去评价和衡量的重要人物。
公元前1世纪，晏子的形象继续在汉代学者的著述中留下痕迹。桓宽在《盐铁论》中不止一次提到晏子，晏子不仅作为历史人物被提及，他说过的话也被当作是引用的对象。更重要的是，在公元前1世纪末，刘向整理校订出一部书题名为“晏子”并且为该书写了一篇前言，晏子的形象在这部书中更加令人难以忘怀。
（Ⅲ-3）刘向整理的《晏子》
公元前26年，汉成帝（公元前33年—前7年在位）命令刘向校理群书。作为结果，刘向搜集了超过八百个有关于晏子的生平以及其生活时代的故事，并且努力将它们编联成一个由215章组成的整体，这215章展现出比较明显的主题，大致按照年代顺序排列，最终形成了晏子独特的个人形象。这是一个身材矮小但是具有强烈使命感的杰出人物形象；是一个节俭与慷慨双重性格的缩影；一位心理操纵与喜剧表演的大师；一位具有渊博的知识、语用推理能力并且强烈反对盲目崇拜超自然现象的杰出学者；一位试图挽救无可救药的君主、像希绪弗斯一样努力的勇气与道德的模范；一位与管仲比肩、与孔子竞争的人物。然而更重要的是，刘向所整理的这215章内容，是中国古代有关于晏子这个独特的人物形象的第一本文学传记，描画了这位“社稷之臣”，这位没有门生的大师，这位生活中真正的英雄的一生。刘向几乎为他所整理的所有图书都写有“叙录”，“叙录”中的内容包括本书的内容简介、版本来源、作者生平以及对于本书的意义和真实性的讨论等等。在刘向的《晏子叙录》中，同样也涉及了这些内容。《晏子叙录》中先列举了《晏子》八篇的内容。前六篇被称作“内篇”，并分别题名为“谏上”“谏下”“问上”“问下”“杂上”“杂下”。第七篇和第八篇被称作“外篇”。第七篇题名为“重而异者”，第八篇题名为“不合经术者”。在列举了《晏子》八篇之后，刘向交代了校订《晏子》的各种版本及来源。之后，他比较清楚简洁地记录了晏子的生平事迹。最后，他对《晏子》这部书做出评价，对不同部分的真实性做出说明。刘向在《晏子叙录》中说：
显然，《叙录》中的第一部分是刘向根据自己校理《晏子》时的实际经验所写。只有他才可以真正告诉未来的读者，《晏子》的定本是如何形成的。在第三部分，同样也展现出刘向是本书的直接整理者，他认为前六章是关于晏子言语的真实记录，而后两章的真实性值得怀疑，尽管如此，刘向依然审慎地将其附录在书末。在记录晏子生平的第二部分，如果读者与《史记》中的晏子传记相比较就会发现，刘向像司马迁一样表达出对晏子的崇敬之情。然而，尽管司马迁说他曾经读过《晏子春秋》，但是他在写作晏子传记的时候依然非常小心地避免直接从中选取材料来组织晏子的传记；相反，刘向却从自己所编订的《晏子》中借用了大量材料来构成自己对晏子生平的描述。
从此以后，晏子的历史形象主要被刘向所整理的《晏子》以及其《晏子叙录》所确定，而其他文献中所记录的晏子在历史的记忆中则越来越模糊。
（Ⅳ）流传中逐渐定型——从东汉到唐代《晏子》与《晏子春秋》的演变
刘向去世以后到唐代，晏子的文本，或者题名为《晏子》，或者题名为《晏子春秋》，在流传过程中被以下的15部著作所列举、引用或者评价：
据张守节《史记正义》中的记载，刘歆曾经在《七略》中儒家著作的部分记录有“《晏子春秋》七篇”。
班固（公元32年—92年）《汉书·艺文志》著录有“《晏子》八篇”。
王充（公元27年—100年）在《论衡》中称“管仲、晏婴，功书并作”，并且屡次称引晏子所说的话，这些话几乎都见于今传本《晏子春秋》。
应劭（大约死于公元204年左右）《风俗通义》引用了今传本《晏子春秋·内篇·杂下》“景公病水瞢与日斗晏子教占瞢者以对”章并直接表明引自《晏子春秋》。
在王肃（公元195年—256年）年轻时写的《孔丛子》中，曾提到“晏子之书亦曰春秋”，王肃在书中还猛烈批评了把晏子看作是反对儒家的墨家信徒的观点。
刘勰（公元465年—522年）在《文心雕龙·诸子》中说道：“管、晏属篇，事核而言练。”
郦道元（死于公元527年）在《水经注》中提到“刘向叙《晏子春秋》”，并称引了今传本《晏子春秋·谏下》“景公养勇士三人无君之义晏子谏”章的内容。
虞世南（公元558年—638年）在唐朝建立之前编辑的类书《北堂书钞》中多次称引《晏子》与《晏子春秋》，所有的内容均见于今传本《晏子春秋》。
欧阳询（公元557年—641年）在公元622年至624年之间编成的类书《艺文类聚》中也多次称引《晏子》和《晏子春秋》的内容，所有的内容亦均见于今传本《晏子春秋》。
魏征（公元580年—643年）在公元631年编成的《群书治要》中，包含了38条称引自《晏子》的内容，这些内容也全部见于今传本《晏子春秋》。这38条引用均比较完整，有7千余字，相当于今传本《晏子春秋》百分之十七的内容。魏征在书中将这38条引用重新排列，尽管它们都被标明为来自内篇六篇的相应篇章，但是实际上其中也包括了现在属于外篇两篇的内容。
同样有魏征参与撰写的《隋书·经籍志》子部儒家类中著录有“《晏子春秋》七卷”，作者被标为“齐大夫晏婴”。《隋书·经籍志》是从汉代到唐代唯一保留下来的完整的书目著作，其中还保存了现在已经失传的唐代之前的一些书目——如阮孝绪（公元479年—549年）《七录》——的内容与特点。
在唐代的正史，如后晋刘昫（公元887年—946年）所编撰的《旧唐书》以及宋代学者欧阳修（公元1007年—1072年）和宋祁（公元998年—1061年）编撰的《新唐书》中，《晏子春秋》均著录于儒家类，七卷，作者为晏婴。
李善（死于公元689年）在《文选注》中引用《晏子春秋》40次。所有的引用均见于今传本《晏子春秋》的八篇中。
徐坚（公元649年—729年）所编撰的类书《初学记》中，多次引用《晏子》与《晏子春秋》的内容。所有的引用也都见于今传本《晏子春秋》。
马总（死于公元823年）的《意林》中引用了16条题名为《晏子》的内容，这部《晏子》共八篇，所有的内容均见于今传本《晏子春秋》。
总而言之，在这一时期：
1. 著录、称引、流传的晏子文本有两个题名：《晏子》和《晏子春秋》。
2. 刘向的《晏子叙录》广泛流传并且直接被引用。
3. 晏子文本中“内篇”六篇的标题在当时已经存在并且被引用。
4. 晏婴被看作是晏子文本的作者。
5. 晏子文本被归类为儒家文献。
6. 晏子文本的文学意义被高度赞扬。
7. 大量的晏子文本出现在类书或类似的书中。
8. 晏子文本的“外篇”两篇有时被合并为一篇。偶尔地还有将外篇内容归类到内篇的情况。因此，我们有理由推测，外篇的内容在分卷或分篇时，它们的归属比较灵活多变。
（Ⅴ）身份突变：柳宗元（公元773年—819年）对《晏子春秋》墨家性质的论辩
公元8世纪末，晏子文本被高度重视并且稳步流传的进程被打断。那曾经被看作是晏子所作，冠以刘向的《叙录》，在各主要书目中被归类为儒家文献，在各种著述中被广泛引用并被当作历史典故的晏子文本，其身分认同在柳宗元的质疑后迅速瓦解。柳宗元在《辩晏子春秋》中说：
柳宗元总结了墨家学说的特点，并且指出《晏子春秋》的内容与墨家学说非常接近。柳宗元接着说：
柳宗元将晏子文本看作是冒名晏婴的墨家弟子的著作，并且切断了其与并非墨家的历史人物晏婴的关联，这一结论貌似公允，实际却极具颠覆性。而且，柳宗元还否定了晏子文本为晏子后人所创作的可能性，这就相当于将该书看作是一部“伪书”了。柳宗元的论辩的结果，推翻了对晏子文本原来的认识，使其成为齐国墨家暗中宣扬自己主张的工具。
总之，刘向已经指出，在他所整理出的《晏子》八篇中，包含有一些后世创作的并非晏子所作的内容。接着，王肃在《孔丛子》中竭力辩驳晏子并非反对儒家，尽管在《晏子春秋》中有一些墨家的观点。然而柳宗元直言不讳的批评彻底改变了《晏子春秋》的流传轨迹，从此之后，有关于这部书的真实性、成书时间、作者以及其中的哲学观点都成为争论的焦点，而后世所有相关的争论无不回应着柳宗元破除古代文本偶像的观点与精神。
（Ⅵ）混乱的著录：宋代的《晏子春秋》
到了宋代，《晏子春秋》的流传出现了两条不同的线索：一条是在类书的引用中，似乎晏子的文本依然稳定；另一条则在各种书目的著录中，著录中的晏子文本似乎并不稳定。
（Ⅵ-1）宋代类书中的《晏子春秋》
当我们检查宋代类书中直接或间接引用的《晏子春秋》的时候，我们似乎可以得出这样的印象，在这300多年的历史中，《晏子春秋》似乎只有一个版本，而且这个版本非常稳定，被反复引用。
我统计了122条直接引用的晏子文本，从中可以看出晏子文本的流行趋势以及其在宋代作为古代典故所具有的稳定性。
如果我们将宋代类书中所引用的《晏子春秋》的内容与唐代类书和各种选集中所引用的《晏子春秋》的内容合并成一个综合数据库，我们根据两个时代的重复引用率可以看出《晏子春秋》的文本从唐初到宋末稳定流传的线索。在这个数据库中有一条内容特别值得注意，那就是今传本《晏子春秋》的最后一条内容（第215条），这条内容不仅首先在两部唐代类书《群书治要》（公元631年）和《意林》（约823年）中被直接引用，而且随后又在宋代的类书《太平御览》（公元938年）、最后在明代的类书《天中记》（公元1569年）中被引用。而这一条内容甚至还有更古老的源头，在银雀山汉简本《晏子》中第624至630简上，我们也发现了相同的内容。这七根竹简，从银雀山1号墓（公元前140年）穿越到唐、宋、明时期的类书中，最后又出现在今传本《晏子春秋》各种各样的印本中，我们因而可以认为《晏子春秋》的整个文本具有超越时代的稳定形态。
（Ⅵ- 2）宋代书目中的《晏子春秋》
从公元1世纪到隋唐时期，晏子的文本在流传过程中往往被题名为《晏子》或者《晏子春秋》，或者是8篇，或者7篇，或者是7卷。在此时期，晏子文本的作者一直被认定为晏婴本人，除了柳宗元（公元773年—819年），他认为晏子的文本是齐国墨家学派的作品。到了宋代，很多记载列出的晏子的文本变成了12卷。宋代的一些学者认为汉代出现的八篇的晏子文本已经失传，一些主要的书目拒绝承认晏婴为晏子文本的作者。
最早记录12卷本《晏子春秋》的，大概是王尧臣（公元1001年—1056年）。他在《崇文总目》儒家类中记录有《晏子春秋》12卷，并且指出汉代八篇的晏子文本已经亡佚，同时，王尧臣还认为将晏子文本的作者看作晏婴是错误的，实际上这一个文本应该是后人所作。
在王尧臣之后，晁公武（公元1105年—1180年）在他的《郡斋读书志》中也记录了12卷的《晏子春秋》。晁公武大段引述柳宗元的观点，并且将《晏子春秋》归类到“墨家类”著作中。
之后，陈振孙（约1211年—1249年）在《直斋书录解题》中也著录了《晏子春秋》12卷，陈振孙依然将其归类为“儒家类”著作，并且说：“《晏子春秋》十二卷，齐大夫平仲晏婴撰。《汉志》八篇，但曰《晏子》。《隋》《唐》七卷，始号《晏子春秋》。今卷数不同，未知果本书否？”
南宋末年，王应麟（公元1223年—1296年）在《汉书艺文志考证》中亦采用了柳宗元的观点，认为此书实乃墨家学派的著作，王应麟还指出晏子的文本之所以从八篇变为七卷乃至十二卷，是因为“后人采婴行事为书，故卷颇多于前《志》”。
清末，刘师培（公元1884年—1919年）也曾经对《晏子春秋》从八篇变为七卷再到十二卷的过程有比较详细的讨论。他在《晏子春秋篇目考》中说：“隋唐《志》皆七卷，盖合《杂》上下二篇为一。”至于宋代出现的12卷本，他认为是将七卷本中除“外篇”上下二篇之外的五篇各分为二篇，这十卷内容与“外篇”两篇共同构成了12卷本的《晏子春秋》。
相比于刘师培的推论，一百多年前的《四库提要》对《晏子春秋》卷数的说明似乎更为严谨。《四库提要》中说：“《汉志》《隋志》皆作八篇，至陈氏、晁氏《书目》乃皆作十二卷，盖篇帙已多有更改矣。”在这短短的几句话中，四库馆臣承认历来书目著录上的差异，并且将其看作是文本流传过程中技术上的必然现象，无须过分解读。这一总结，将所有不确定性归结为文本流传的漫长历史。四库馆臣在这一问题的处理上比刘师培强作解人的做法更为可取。
（Ⅶ）印本流传：从宋代到清代的《晏子春秋》
（Ⅶ-1）《晏子春秋》的早期印本：元刻本与明活字本
我们首先必须指出，《晏子春秋》的木刻本应当在宋代就已经出现。然而，宋刻本既无明确记载亦无实物留存，因此相关讨论均属推测。而我们现在能够找到明确记载的两部最早的印本《晏子春秋》分别是：
（1）江苏常熟张金吾（公元1787年—1829年）《爱日精庐藏书志》著录有元刻本《晏子春秋》八卷，并云：“凡内篇六卷，外篇二卷，合八卷。卷首有吴岫印记、刘向序。”元刻本今亦不传，相关记载亦可参见吴寿旸（公元1771年—1835）《拜经楼藏书题跋》。
（2）浙江仁和丁松生（公元1832年—1899年）八千卷楼藏明活字本。这一版本亦八篇，书前有刘向的叙录，每篇前有本篇的目录。这一版本在20世纪被影印，收录入《四部丛刊初编》，然后成为“先秦两汉古籍逐字索引丛刊”史部第四种《晏子春秋逐字索引》所依据的底本。
（Ⅶ-2）《晏子春秋》的后期版本：1780年《四库全书》本与1788年孙星衍的《晏子春秋音义》
四库馆臣将《晏子春秋》归入史部“传记类”，如此分类，避免了正面回答《晏子春秋》到底是属于儒家学说还是墨家学说的各种疑问。根据《四库提要》的记录，《四库全书》所录《晏子春秋》为“明李氏绵眇阁刻本”，这一版本不像明活字本一样有目录和叙录，《晏子春秋提要》列在书前被当作序言。该提要浓缩了从汉代到宋代千余年的目录记载历史，将《晏子春秋》与唐代魏征的《谏录》和李绛的《论事集》并列。提要还指出，《晏子春秋》的编辑者姓名佚失，题名“晏婴”乃“依托也”，书中又有后人窜入内容，已非原本。针对这部书到底是儒家著作还是墨家著作的问题，提要认为，《吕氏春秋·仲春纪·当染篇》中记载鲁惠公时留止周桓王使者史角，史角的后人居住鲁国，墨子曾经从史角的后人问学，晏婴生活的时代稍早于墨子，自然也有可能先闻其说，因此书中有近似于墨家学说的内容并不奇怪。提要最后还讨论了晏子文本《晏子》与《晏子春秋》“二名兼行”的问题以及之所以选择“明李氏绵眇阁刻本”的原因。
孙星衍校订的七卷本《晏子春秋》包括了晏子八篇的内容。通过篇卷混编，孙星衍希望能够解决著录中汉代八篇与隋唐七卷的文本差异问题。孙星衍的校本，依据两个明代版本，这两个明代版本既无前言亦无目录，孙星衍将刘向的叙录放在书前，还加上自己的一篇序言。书后，附录两卷考证内容，题名为“音义”，是孙星衍据《艺文类聚》、《初学记》、《太平御览》以及《文选》注等文献对《晏子春秋》的异文和部分内容所作的研究和分析。孙星衍在《晏子春秋》校本的序言中作出如下结论：
1. 根据文本内部的证据和书目中的记载，流传下来的晏子文本绝非伪书。
2. 流传下来的晏子文本成书于战国时期，材料来源于齐国的史书，由晏子的宾客在他死后搜集成书。
3. 明代的某些刊本变乱次序、删改内容，讹谬甚矣。
4. 晏子文本的内六篇符合六经之义，属于儒家内容。
5. 《玉海》引《崇文总目》著录《晏子春秋》十四卷，认为是后人采晏婴故事成书，因此导致卷数增多。《玉海》的说法不足信。
6. 柳宗元等人认为晏子文本的内容为墨家学派的作品，属于不学无术的无知之论。
7. 古人书外篇多为依托，刘向也已经怀疑晏子文本的外篇为后世辩士所为，外篇之依托不足以损害内篇记录晏子言行之真实。
（Ⅶ- 3）晚清民国时的《晏子春秋》
晚清民国时期，对于《晏子春秋》的文本研究并没有更多突破。此时比较著名的研究者是梁启超（公元1873年—1929年）和刘师培（公元1884年—1919年）。梁启超对于《晏子春秋》文本的看法，实际上整合了过去两千年来中国学者的各种意见。梁启超提出了一个大胆的推测，他认为《汉书·艺文志》中所著录的《晏子》，不仅即是司马迁在《史记》中所提到的《晏子春秋》，而且很可能与刘安在《淮南子》中所提到的“晏子之谏”为同一文本。接着，梁启超遵循柳宗元的观点，认为《晏子春秋》是齐国的晏子宾客所作，他同时还强调，按照四库馆臣的意见，这些齐国的晏子宾客并不知道墨子（这样就使得《晏子春秋》中即使有一些类似于墨家的思想，但是也与墨子没有关联）。梁启超最后总结说，《晏子春秋》以“晏子”为书名是比较晚的事情，这部书的实际成书时代可能从战国时期一直延续到汉代初年。对于流传下来的晏子文本，梁启超怀疑其不一定与司马迁和刘安所提到的文本相一致，但是他同时论证，流传下来的晏子文本很可能就是刘向所编辑整理的本子，而不太可能是后世的作品。
清代末年，在上海出现了最早的西方人对《晏子春秋》的评价。英国传教士伟烈亚力（Alexander Wylie）在1867年按照《四库全书》的分类编成了《中国文献录》，他在其中的传记类也列举了《晏子春秋》，并且介绍这本书完成于公元前若干世纪，这是一本有关于晏婴的个人传记，而晏婴是墨子的知名门徒。
总而言之：清末民初，有关于《晏子春秋》的真实性与思想特征等决定性问题依然悬而未决。持续进行的有关于《晏子春秋》文本的学术研究需要一个更广阔的研究者和读者圈子，以便能够创造出一个可以包容新观点从而解决围绕着文本的众多不确定问题的空间。实际上，这个更广阔的圈子在20世纪出现了，对于《晏子春秋》的研究已经大大扩展到西方世界。
（Ⅷ）考古发现与数字人文：《晏子春秋》研究在20世纪和21世纪的拓展
（Ⅷ-1）1923年至1972年的《晏子春秋》
1972年，在银雀山发现的102枚“晏子”竹简及残片，震惊了学术界，对《晏子春秋》的文本研究造成了巨大的影响。在此之前，不少中外学者努力使《晏子春秋》更容易为研究中国哲学的学生所利用，并且在其作者、时代以及真实性等方面都提出了一些新的意见。这些学者包括：德国汉学家伏尔克（A lfred Forke,1867—1944）；法国汉学家马伯乐（Henri Maspero,1883—1945）；美国华裔学者高克毅（George Kao,1912—2008）；美国学者吴克（Richard L. Walker,1922—2003）；美国学者伯顿·沃森（Burton Watson,1925—2017）。而中国学者则包括：张纯一（1871—1955）；吴则虞（1913—1977）；王更生（1928—2010）。
1923年，伏尔克（A lfred Forke）在西方学界发表了具有开创性的关于《晏子春秋》的论文，至今依然有重要的影响。伏尔克坚决主张《晏子春秋》是一部真实的古代文献，产生于公元前5世纪。他认为，《晏子春秋》的主角晏子，之所以将他的一生致力于服务他的君主和人民，是因为一系列的实用主义的价值观。对于伏尔克来说，晏子是一位政治家，而不是在抽象的知识与艰深的学习世界中开拓道路的儒家理论家。在伏尔克眼中，晏子也达不到管仲那样伟大的政治家的地位，然而，作为一个“角色”，晏子明显比管仲更出色。伏尔克的论文由多部分组成，其中包括从不同文献中搜集并重新编排的完整的晏子生平传记，有关于晏子文本历史的详细讨论，对于不同主题如伦理、节俭、人际关系、君民关系等所作的长篇的哲学分析。颇具突破意义的是，伏尔克提出，相比较于孔子和墨子的宗教信仰，晏子文本中所表现出来的对于某种“自然宗教”的支持显示出晏子思想更具理性主义的特征。最后，伏尔克在文章中表达出一个相当有见地的意见，他认为《晏子春秋》第八篇开头的几条猛烈抨击孔子和儒家学说的内容应该也是真实可信的。
1927年，马伯乐（Henri Maspero）在《古代中国》（La Chine antique）中讨论《晏子春秋》的内容还不到两页，然而这两页内容却包含了对晏子整个文本的精彩描述。马伯乐认为《晏子春秋》成书于公元前4世纪中叶，相比于同时代的其他文本，《晏子春秋》是一部历史浪漫主义的作品，也就是说，作者将真实的历史事件与自己的想象混合在一起从而创作了这样一部作品。马伯乐是第一位指出晏子文本中除了倾向墨家、反对儒家之外同时还反对管子的学者，但他也认为《晏子春秋》中的哲学思想无聊乏味。然而就文学风格而言，马伯乐赞赏文本中很多场景的生动描写。在有关于书目的大段注释中，他厘清了从汉代到唐代原本《晏子春秋》的不同的流传线索，即使其在宋代和元代屡被改编，但是原本《晏子春秋》文本中的大部分内容依然被保留在今传本中，马伯乐的这一分析极具说服力。
1946年，高克毅（George Kao）在《中国智慧与幽默》（Chinese Wit & Humor）的选集中收录了《晏子春秋》中十二段相对较长的内容的英文翻译。在选集中，高克毅不仅向西方读者介绍了机智的晏子，而且在智慧角度上，将晏子看作是与孔子、孟子、庄子、列子和韩非子相并列的不朽形象。
1952年，吴克（Richard L. Walker）有关于《晏子春秋》的极具分量的分析研究探索了中国学者长期争论的很多复杂问题，从而将《晏子春秋》介绍到英语世界的学术圈中。时至今日，对于那些希望研究像《晏子春秋》这样的文本的人来说，吴克的论文依然如同一个路标值得重视。吴克最主要的目标是提升始终徘徊在主流研究之外的《晏子春秋》的地位，并且检查伏尔克1923年的论文在发表二十五年之后如何经受时间的考验。为了达到这一目的，吴克建议将“晏子春秋”这一比较古典而庄严的书名改换成比较平易近人的书名，如“晏子说”“晏子格言”等。同时，他根据文本中所展现出来的晏子的特点，将其仅仅定性为善于“通过语言技巧摆脱困境”的人。至此，读者对吴克的真正意图困惑不解。显然，人们不明白如何通过贬低书名以及将主人公贬低为只会耍嘴皮子的人来提升其地位。而实际上，在他的论文的开场白以及所有的介绍性的评论与结论中，吴克始终瞄准着他的真正目标：有关伏尔克的1923年的论文中，他在文章开始阶段，先是表达了几句对于伏尔克在文本历史以及相关学术领域所做大量工作的赞美之词，接着便开始批评伏尔克对于文本的过度的哲学解读，并且列举了其在汉学研究方法论上的缺陷，吴克用多少有些蔑视的态度说：“因此，虽然我们可能会同意伏尔克教授关于《晏子春秋》成书于更早时代的结论，但是并非因为他给出的理由。”接着，吴克批判性地考察了《晏子春秋》成书时的政治背景，以记载了类似内容的其他文献为参考，从语法上分析了《晏子春秋》的语言习惯，并且对多层次文本中包括什么和没包括什么进行了出色的评估，从而得出以下结论：
1962年，伯顿·沃森（Burton Watson）在《早期中国文学》（Early Chinese Literature）中有关于《晏子春秋》的一篇很短但是非常精彩的章节。这一章节为理解和探索晏子文本，特别是在心理学层面上，指明了新的方向。伯顿·沃森在他对早期中国文学的考察中，将《晏子春秋》归类于“哲学”部分，并且将其主人公晏子描绘为一个提倡节俭并且因为他无可躲避的唠叨而使得他放纵的君主生活痛苦的严厉墨家。对于两千年来所形成的一个直言敢谏的崇高榜样，伯顿·沃森提出了一个新的理解，而在这一理解中，伯顿·沃森站在了自我放纵的齐君这边。他的新思路颠覆了千年以来《晏子春秋》文本崇高的道德核心，将其转变为一出现代戏剧，同时邀请观众对其自由解读。此外，沃森将这部戏剧的未来读者吸引到一幅既不能进步也不能突破的静止世界的画像前。这个世界和《晏子春秋》中的主要角色在他们原本的位置上固定不动，实际上，按照沃森所描述的：“几乎总是齐君承认他的愚蠢并且承诺改革，而在接下来的故事中，我们发现他不停地重蹈覆辙、故技重施。”毫无疑问，沃森对于《晏子春秋》的观点使这部书充满戈多式的荒诞——主人公啥事也没干。
从1930年到1962年，中国大陆出现了两部《晏子春秋》的注释本，这为读者提供了极大的方便。
1936年，张纯一出版了《晏子春秋校注》，这是一部内容极其丰富和详细的注本。张纯一首先向读者提到孙星衍1788年对《晏子春秋》所做的工作，然后他仔细比较了其他经典文献中的类似内容，对《晏子春秋》的各种异文作出细致的梳理与考证。
1962年，一部应当被看作是最全面详细的《晏子春秋》注本出版，这就是吴则虞的《晏子春秋集释》。在这部书中，吴则虞对于《晏子春秋》本身的各种有价值的资料以及其成书和流传的各种线索做了百科全书式的搜集与整理工作。两千年来，对于《晏子春秋》文本的每一条引用、每一条注释几乎都被吴则虞搜罗殆尽。吴则虞不仅对每一个难以理解的问题都给出了自己的意见，这些意见广征博引、极具启发，同时他还附录了大量的参考文献供读者了解整个学术史的发展。在前言中，吴则虞提出五条意见论证《晏子春秋》的作者是类似于淳于越那样的来自齐国的学者，在秦统一天下后所作。其中前三条意见属于文本之外的外部证据，说服力不大。但是第四条和第五条意见则为讨论《晏子春秋》的作者提供了一个新的思考维度。在这两条意见中，吴则虞指出淳于越所提出的“师古长久”与《晏子春秋》中的“毋变尔俗”“重变古常”的思想相一致，而《晏子春秋》中的谏议，带有托古讽今的意味，又与李斯所说的“各以其学议之”的“议”相一致，因此吴则虞认为《晏子春秋》的成书，“极有可能就是淳于越之类的齐人，在秦国编写的”。可惜，吴则虞可以用来比较的淳于越与晏子的材料数量太少，导致他的结论很难成为定论。尽管如此，这些有关于文章体裁的争论应该已经激起汉学界足够兴趣，促使汉学家建立一个相似文体的数据库，搜集各种线索的马赛克，它们或许会给我们指出一个方向，从此我们可以辨认出创造《晏子春秋》文本原型的那只“隐形的手”。总而言之，吴则虞作为一个注释者、目录学家和一个文本比较研究的创新者，其所取得的成果有目共睹，今后任何有关于《晏子春秋》的研究都必须以他的著作作为出发点。
1966年，台湾师范大学王更生完成了他的博士论文《晏子春秋研究》。这是王更生研究《晏子春秋》的第一个贡献，随后，他又出版了两部《晏子春秋》的白话翻译以及很多其他优秀的成果，这使他成为《晏子春秋》研究领域的专家。在其博士论文以及之后其编辑的《晏子春秋》读本中，王更生将已经出现在吴则虞书中的那些传统中国学者所关注的作者、真实性以及成书年代等问题做了简明扼要的总结。在他编辑的《晏子春秋》读本中，包括有完整详细的编年体的晏子传记，还有从诸如形而上学、精神、生死、伦理、政治哲学、财政、外交、自我修养、文学体裁、语法和修辞等各种角度对《晏子春秋》文本所作的分析与描述。
毫无疑问，在1972年之前的数十年间，无论是对于中国读者还是西方读者来说，《晏子春秋》都变得越来越容易接近和阅读。这不能不归功于众多学者在注释、翻译以及研究方面持续不断的努力。然而，1972年银雀山竹简的发现，迫使具有漫长历史的《晏子春秋》文献研究不得不面对这102根竹简重新评价已有的各种结论。
（Ⅷ-2）银雀山汉墓竹简
1972年，在山东临沂东南的银雀山，发现了两座汉代墓葬，出土4 942枚竹简和大量残片。这两座墓葬的埋葬时间大概是公元前140年至134年和公元前118年，但是竹简的书写年代应当更早。银雀山汉简整理小组的成员将银雀山1号墓出土的4 942枚竹简分成两组。第一组竹简所记录的内容见于传世文献，而第二组竹简所记录的内容则早已佚失。在第一组竹简中，有102枚竹简组成了一个连续的序列（编号528—630），其中包括16段比较清楚的文本，这些文本包含了今传本《晏子春秋》八篇中每一篇的部分内容。值得注意的是，这102枚竹简共有2 970个字，这相当于今传本《晏子春秋》7.2%的内容。
此外，1972年至1977年，在甘肃破城子、河北定县、安徽阜阳的考古发现中，都发现了《晏子》文本的一些残片，这些残片的时代大约是公元前179年至公元前49年。可惜，所有这些残片都太过残破，以至于没有办法复原出哪怕是一段完整的内容。
总之，在银雀山汉简与今传本《晏子春秋》间，存在着惊人的相似文本。根据1972年至1977年其他考古挖掘中所发现的类似《晏子》文本的残片，以及《淮南子》中所提到的“晏子之谏”，我们推测在大约公元前200年或者更早，有若干种竹简本《晏子》在知识阶层广泛流传。据此，我们可以得出一个更加令人兴奋的假设，那就是《晏子》文本的原型很可能在更早的时候就已经开始形成。
（Ⅷ-3）淳于髡（约生活于公元前340年）、滑稽与稷下学宫
1936年，罗焌（公元1874年—1932年）在研究过《晏子春秋》的学术史后得出了一个崭新的、惊人的结论。他认为：《晏子春秋》中所记录的故事和谏言，与齐国的淳于髡、楚国的优孟和秦国的优旃的事迹非常近似。很明显，这个想法太过唐突而且毫无根据。然而，它给我们提供了一个新的视角，它剥去了晏子文本中独特的历史和哲学身份，将其主角晏子从一个渊博的学者和政治家替换为一位修辞大师或者一名俳优。
四十年后，在1976年，王更生也提出了一个类似的观点。王更生指出，在晏子的文本中多次出现了没有意义的调笑段落，这暗示着作者或许是像淳于髡一样的俳优，晏子的文本反映出齐国稷下学宫中辩士的特点。
1986年，“淳于髡假设”被吕斌再次强调。吕斌提出五条证据来论证淳于髡的言行和生平与《晏子春秋》中的内容有密切的联系，因此或许淳于髡即《晏子春秋》的作者。特别是，他指出淳于髡的劝谏就是建立在对晏子的崇拜的基础上；他可以直接进入齐国档案馆并从中获取有关于晏婴的材料；他就像晏子一样并不严格属于哪一学派；而且在整部《晏子春秋》中到处都展现出他的主张与滑稽风格。最后，吕斌把“淳于髡假设”又推进一步，他指出，无论是淳于髡还是晏子，都被描述成极为矮小的人。这一身材上惊人的相似，与其他证据一起，构成了吕斌的假说，一位与淳于髡生平非常近似的作者将自己的个人经验、哲学思想以及身体特征投射到他所编写的文本上，这一文本借用了历史上的名人晏子的名字，《晏子春秋》就此诞生了。
渐渐地，“淳于髡假设”成为《晏子春秋》作者问题上一个重要的意见。1998年，陶梅生在《新译晏子春秋》中提出淳于髡的追随者编写了《晏子春秋》的看法。2000年，林心欣在她的硕士论文《晏子春秋研究》中讨论了将《晏子春秋》归类为“儒家”、“墨家”和“传记类”的各种问题，她提出应将其归类为“俳优”，并进而讨论了淳于髡与《晏子春秋》的作者之间的关系。2005年，赵逵夫在论文《〈晏子春秋〉为齐人淳于髡编成考》中提出了八条证据论证《晏子春秋》为淳于髡所编。赵逵夫的意见重复了吕斌在1986年所作《淳于髡著〈晏子春秋〉考》一文中的观点并且更加详细地论证了这一假说，他指出身材矮小的淳于髡编写了《晏子春秋》这样一部有关于身材矮小的晏子的故事集，通过这部书，淳于髡为自己赢得尊重。2011年，Andrew Meyer发表了一篇具有启发意义的综合性的论文，在论文中，他令人信服地指出《晏子春秋》是在田齐（公元前379年—221年）统治者支持下所创作的，无论是《管子》还是《晏子春秋》，它们最初被编成的目的是让齐国的知识阶层将自己当作“齐国学派”，而非“儒家”或者“墨家”。
如今，在汉学研究领域以及相关的互联网上，《晏子春秋》为淳于髡或者其追随者所编的意见越来越成为一种共识。Andrew Meyer的论文所讨论的内容以及深入的分析为未来的汉学研究铺平了道路，也许今后可以将他的观点与其他学者对稷下学宫、淳于髡的生平和中国古代俳优传统的研究结合在一起得出更进一步的结论。到那个时候，《晏子春秋》最初是由谁创作或者编写的，我们或许会得到一个确定的答复。
（Ⅷ-4）近年来《晏子春秋》的研究以及未来数码人文和计算机思维时代的《晏子春秋》研究
自从1974年以后银雀山汉简整理出版，《晏子春秋》变成了新的汉学研究领域的焦点。许多学者分享他们的见解与专业知识，探索与《晏子春秋》相关的文本考察方面与历史问题方面研究的新路径。在这些学者中，值得注意的是Rainer Holzer，他在1983年将《晏子春秋》内六篇翻译成德语，他在译文中加入一篇研究性的导言、详细的文本考证和许多宝贵的注释。1993年，Stephen Durrant在M ichael Loewe的《早期中国文献书目指南》中为《晏子春秋》的文本和历史贡献了一份精练的学术提要，从而为研究《晏子春秋》的学生提供了极其宝贵的帮助。在中国，除了众多白话翻译和注释之外，从1974年到2010年，有不少于280篇与《晏子春秋》相关的学术论文在学术期刊上发表。此外，从1995年到2010年，在不同的大学和学院有30篇相关学位论文提交。
如此看来，除非在未来的考古挖掘中再次发现新的有关于晏子的竹简或相关文献，《晏子春秋》最初成书问题的研究将很难在过去研究的基础上再有突破。未来对于《晏子春秋》起源的研究，恐怕需要依靠最近被数码人文学学者发展出来的计算和统计工具进行。在谷歌搜索引擎中，输入“晏子春秋”，大约会看到400 000条结果。网站，比如中国古代文献数据库（CHANT）、中文文献计划（Ctext）、台湾汉学数字典藏资源等等，提供了数以百万计的可搜索字符，保存了浩瀚如海洋的中文文献，无论古代还是现代，无论是经典还是非经典，也无论是广泛流传还是极其罕见。文本挖掘方法一定能够分析这些数以百万计字符的大型数据库，并提出首次出现某些名称、词汇或成语的时间以及之后它们被如何使用等各种建议，从而使研究人员能够成功地回答有关文本形成时间的问题。而且，某些网站如DID-ACTE和费正清中心已经开始支持主要依靠计量文献学（stylometric analysis）和数据提取的汉学研究新技术。事实上，计量文献学程序可以学习某位作者的写作模式，并计算出这位作者写出一个作者身份不明的已知文本的可能性，从而解决关于某些特定文本的作者问题。根据最近的报道，来自得克萨斯大学奥斯汀分校的两位研究人员，用程序学习了某位作者的54个戏剧作品的模式，并计算出该作者 撰写出一个已知文本的可能性，而这一已知文本在之前被看作是他人的作品。他们的测量分析确定了《将错就错》（Double Falsehood）这部从前被看作是莎士比亚学者刘易斯·西奥博尔德（Lew is Theobald，1688—1744）的作品，实际上是威廉·莎士比亚本人的作品。这一发现，震动了莎士比亚学界，也为我们提供了充分的理由去想象，在不久的将来在《晏子春秋》研究领域很可能也会发生类似的颠覆现象。
（欧永福 文，吴洋 译）
Introduction
The Yanzi Figure and the Yanzi Text in History
(Ⅰ) Waiting in the W ings: The Yanzi Figure From the 6th through the 3rd Century BCE
First came Yan Ying[1] 晏婴—the person. Years later,after he had long passed away,[2] came Yanzi 晏子—the eponymous protagonist of The Spring and Autumn Annals of Master Yan 晏子春秋. Reliable evidence about the life and times of Yan Ying as a historical figure who lived in Qi 齐 during the 6th century BCE is scarce and cannot be accurately assessed.[3] Through the fifth and fourth centuries BCE his identity,nevertheless,gradually coalesced into a distinctive portrait. Initially,Yan Ying was briefly mentioned in the Analects 论语,in which he was succinctly praised by Confucius as a person who excelled at social relationships.[4] Next,he made his debut as an historical figure in the Zuozhuan 左传.[5] This text contains several anecdotes about him,centering on the pivotal events of the waning of the ruling house in Qi and the rise of the Tian 田 clan to supremacy during the second half of the sixth century BCE. These anecdotes,taken together,form an appealing personal portrait of Yan Ying as a minister who both delivered eloquent remonstrations and played a principled and courageous role in the service of three failing rulers.[6]
Subsequently,during the second half of the fourth century BCE,the presence of Yanzi’s figure in the literature began to make a mark,as he received both praise and critique. Chunyu Kun 淳于髡 (c. 340 BCE),a member of the Jixia 稷下 patronage community,[7] cited him with admiration in the remonstrations he delivered to Duke Wei of Qi 齐威王 (c. 378–320 BCE).[8] M encius 孟子 (c. 372–c. 289 BCE),on the other hand,was offended when his disciple,Gongsun Chou 公孙丑 from Qi,persisted in weighing the abilities of his master Mencius against those of Guan Zhong管仲[9] and Yanzi.[10] Mencius responded to the notion condescendingly by denying the importance of both Guan Zhong and Yanzi as two great m inisters in Qi,characterizing them instead as two state functionaries whose value was strictly limited to their own immediate time and place.[11] The same dismissive attitude towards Yanzi and Guan Zhong was recorded a bit later in the preserved philosophical w ritings of Xunzi 荀子 (ca. 310–235 BCE; alt. ca. 314–ca. 217 BCE). A certain “Master,” points that both Yan Ying and Guan Zhong are presented as ministers of major accomplishments; he then indicates that the accomplishments of Guan Zhong were superior to those of Yan Ying; finally,he concludes the section by calling Guan Zhong “a rustic boor” who did not properly fulfill the post of minister to his ruler,the Son of Heaven.[12] By implication,what is one to think of Yan Ying,whose achievements were claimed to be inferior to those of Guan Zhong?
(Ⅱ) Center Stage: The Yanzi Figure from the 3rd through the 2nd Century BCE
Between the years 300 BCE and 200 BCE the Yanzi figure gradually underwent a change. His character evolved into a more central figure,both in narrated episodes in which he played a role,and in textual records that bore his name. Until the middle of the third century BCE his deeds,admired or dismissed,were always overshadowed by the general historical circumstances that framed his place in the text,and his character served a broader purpose than the limited scope of his pronouncements. However,around 250 BCE,the Yanzi figure moved from the periphery to occupy center stage in the narrated episodes. In the follow ing years,his figure became a source for full references and attracted increasing critical attention. His image became more complex—sometimes exemplary,sometimes dubious—and eventually gave rise to a body of characteristic anecdotes that focused not only on his public deeds but also on his personality as an individual w ith a unique thought process. This shift towards the center is reflected in the follow ing three major texts of the era: In Mozi 墨子;[13] Lüshi chunqiu 吕氏春秋,[14] and Hanfeizi 韩非子.[15]
In Mozi’s 墨子 chapter thirty-nine,“Against the Confucians” 非儒,Yanzi no longer plays a subsidiary character role,but rather emerges as a dom inant figure on the “anti-Confucian,” Mohist stage. He is portrayed as a close confidant and advisor to the Duke of Qi,whose Mohist agenda against the Confucians elicits from Yanzi a series of hostile view s against the Confucians’ worldview and excessive practices.[16] His attitude is harsh; he not only denounces Confucius personally as a rebellious and even murderous figure,but also strips Confucianism of its core values,presenting its advocates as practitioners of useless showmanship. From this point in the text onwards,the Yanzi figure becomes identified in several other texts and references as well,not only as a Mohist but also as a torchbearer against Confucius and Confucians alike.[17]
In the Lüshi chunqiu 吕氏春秋,the Yanzi figure functions authentically as a unique paradigm of a moral,self-critical,and courageous individual. The text contains three episodes that portray him acting in desperate and life-threatening situations. In the first,his benevolence and moral authenticity is so exemplary and inspirational for others that those he encounters devotedly sacrifice their lives for him.[18] In the second,his act of benevolence demonstrates his rare ability to be self-critical and selfdemanding—a perfectionist’s awareness of the implications of the entire scope of his actions and conduct.[19] In the third,he is presented as a model of great authentic courage,as he faces imm inent execution.[20]
The Hanfeizi 韩非子 contains three sections on Yanzi pointedly directed against his reasoning regarding governmental concerns,against his humane approach to relief efforts,and against his psychological considerations in regard to harsh and lenient punishments. It seems as if the Hanfeizi considered the Yanzi figure a threatening embodiment of a world view that had to be challenged and refuted on Legalist grounds. The Yanzi figure is therefore met in the text w ith sharp philosophical criticism,and at one point he is even branded “insincere,” w ith respect to his own argument. The first among these three sections attacks Yanzi’s famous traits of frugality and abstemiousness. The text not only argues against the senselessness of these two values,insofar as they can result in a complete lack of motivation amongst ministers to succeed,but also claims that Confucius himself had previously voiced this particular criticism against Yanzi.[21] The second section claims that Yanzi’s humane and generous approach towards officialdom and the people shows that he lacked the ability to remove troubles and thereby exposed his ruler to grave disaster.[22] Finally,in the third episode,the text attacks Yanzi’s lenient approach to punishment,claiming that Yanzi himself did not believe his own argument. For the Hanfeizi,Yanzi’s lenient approach show s that Yanzi failed to understand clearly the meaning of governance.[23]
(Ⅲ) The Soloist: The Yanzi Figure and the Yanzi Text during the Former Han
(Ⅲ-1) The Rem onstrations of Yanzi 晏子谏 in the Huainanzi 淮南子
A ll these textual examples suggest that around 200 BCE,the Yanzi figure not only played a more central role in narrated episodes that bear his name,but also most likely constituted the inspiration for several distinctive texts that focused entirely on his legacy—some of which included his name in the title. In 139 BCE,this stage in the evolution of the Yanzi figure became evident. In that year,the imperial kinsman Liu An 刘安 presented to Emperor Wu 武帝 an encyclopedic philosophical text known as the Huainanzi. The text’s twenty-first chapter,entitled “An Overview of the Essentials 要略,”[24] aims to summarize and analyze the entire preceding twenty chapters and concludes by review ing and explaining the genesis of important texts w ritten by significant advisors and thinkers who played an influential role in the past. The text reads:
Duke Jing of Qi
enjoyed music and sex while inside his palace
and enjoyed dogs and horses while outside his palace.[25]
When hunting and shooting,he would forget to return home.[26]
When enjoying sex,he did so indiscriminately. [27]
He built a terrace over the Road Bedchamber[28]
and cast a grand bell.
When it was struck in the audience hall,
the sound [was so thunderous that] all the pheasants outside the city
walls cried out.[29]
In a single morning [i.e.,one court session] he distributed three thousand bushels of grain as largesse.[30] Liangqiu Ju and Zijia Kuai[31] led him about from the left and the right. Therefore,The Remonstrations of Yanzi 晏子谏 were born.[32]
To sum up: First,the Huainanzi discusses the circumstances that gave rise to a text that it refers to as The Remonstrations of Yanzi; second,“remonstrations” voiced by Yanzi against Duke Jing are not only one of the core subjects of the entire bulk of the received version of the YZCQ but also the title name that Liu Xiang 刘向 (77–6 BCE) gave to the first two chapters of his definitive text of Yanzi;[33] third,the Huainanzi’s description of the seven circumstances that gave rise to this The Remonstrations of Yanzi match their corresponding items in four of the eight chapters of the received version of the YZCQ. We may thus conclude w ith relative confidence that around 139 BCE a text existed whose title identified it w ith Yanzi and whose contents shared a great similarity to the received version of the YZCQ. And indeed,some fifty years later,this stage of textual development was clearly defined when Yanzi’s legacy was sealed as an outspoken and courageous m inister and as the author of a popular,circulating text,entitled The Spring and Autumn Annals of Master Yan 晏子春秋. This legacy was secured among other unforgettable biographical narratives in Sima Qian’s (c. 145–c. 86 BCE) Shiji 史记.
(Ⅲ-2) The Yanzi figure and the YZCQ in the Shiji
Yanzi’s biography appears in chapter 62 of the Shiji,follow ing that of Guan Zhong,which constitutes the first part of the Chapter.[34] It consists of four parts: the first two deal exclusively w ith Yanzi,while the third and fourth focus on him and Guan Zhong together. A number of biographical sketches of both Yanzi and Guan Zhong,mostly drawn from the Zuozhuan,are scattered throughout the preceding chapter 32 of the Shiji.[35] Readers should note,however,that Sima Qian does not include in chapter 62 any of those biographical sketches previously related to Yanzi—only those of Guan Zhong’s are recapitulated in the latter’s biography. They should also note at the outset the concluding line of Yanzi’s biography,before commencing to read this piece regularly from beginning to end. In this line,Sima Qian expresses his own profound admiration for Yanzi’s personality in the following emotive statement:
If Yanzi were still alive,though I were only holding the whip for him,I would be pleased and longing to serve him.[36]
Now,as the readers return to the opening lines of the biography,they may notice that this concluding statement of admiration actually resonates throughout the entire memoir,shaping its choices regarding contents. Sima Qian first weaves the two memoirs together by establishing an historical link between Guan Zhong and Yanzi w ith the follow ing words:
A fter Guan Zhong died,the state of Qi followed his governmental policies and Qi was regularly stronger than the other feudal states. Over one-hundred years later,Yanzi was there.[37]
Both the convergence and divergence between the biographies of Guan Zhong and Yanzi are thus established,bringing together chapter 62 as one historical unit. Readers are now compelled to examine and evaluate from a comparative perspective two figures of unequal historical magnitude: Guan Zhong,a minister whose brilliant policies are now pronounced to have shaped over 100 momentous years in Qi history; and Yanzi,a brilliant and outspoken minister,but a complete failure in saving the ruling house of Qi from its eventual demise. The first part of Yanzi’s biography reads:[38]
Yan Pingzhong 晏平仲,Ying 婴,was a native of Yiwei 夷维 in Lai 莱.[39] He served Duke Ling 灵 (r. 581–554),Duke Zhuang 庄 (r. 553–548),and Duke Jing 景 (r. 547–490) and because of his frugality[40] and vigor,he was esteemed in Qi. Even after he became Prime M inister of Qi,he did not have two servings of meat,and his concubines did not wear silk. At court,if the Duke addressed him,he would speak audaciously; if the Duke did not address him,he would act audaciously.[41] If the Way prevailed in the country,he would follow orders. If it did not,he would weigh the consequences of the orders. For these reasons,he was renowned among the princes in his service during three reigns.[42]
A fter reading this first part of Yanzi’s biography bearing in m ind the preceding biography of Guan Zhong,readers see that Guan Zhong is presented in chapter 62 as a remarkable figure who helped shape the history of Qi and who was personally involved in major turning points of its history. However,no similar narration is provided in the presentation of Yanzi’s life. The first part of Yanzi’s biography contains only a succinct record of names, place of residence,rank,and an approximate count of the years he spent in his official service to three dukes. Readers are offered no information regarding any major historical achievement of Yanzi in Qi; the text describes neither any important incidents nor any significant developments in his personal life by which readers might form a judgment. In contrast,the text itself does provide judgment in the form of a detailed list of observations regarding Yanzi’s traits,virtuous conduct,and personality,as if he were a protagonist in a book whose story-line was somehow missing. Thus,after finishing reading the first part of Yanzi’s biography,readers might well conclude that Yanzi played a much lesser role in the history of Qi than did Guan Zhong. Moreover,except for general background,readers receive no substantial information about the actual events and major developments of Yanzi’s life. That said,they would most likely be attracted to his personality as a protagonist whose frugality,vigor,courage,honesty,integrity,and audacity w ere evident throughout his life,having m ade him em inently famous among the people of his time.
At this point,readers may find themselves perplexed by Sima Qian’s selection of different types of biographical data for the two parts of the chapter. However,in light of the concluding line,they may resolve this discrepancy in view of Sima Qian’s admiration of Yanzi’s personality and his attempt to minim ize the significant gap between the historical impacts of these two ministers on the fate of Qi. Sima Qian achieved this by posing Guan Zhong’s historical achievements against Yanzi’s finest virtues,thereby avoiding a direct comparison between the former,whose historical significance was great,and the latter,who bore almost no historical status but instead was a unique individual w ith rare qualities.
The second part of the biography[43] consists of two consecutive episodes about Yanzi,which appear in the received text of the YZCQ.[44] The first among the two narrates in concise form an episode strikingly similar to Item 5.24 [134]. It tells the story of Yanzi’s ransoming a certain Yue Shifu from captivity by selling one of his four horses,but failing,afterwards,to treat Yue Shifu w ith the proper respect. The ransomed prisoner reacts in protest against Yanzi’s neglect of the proper rites,which brings Yanzi to immediately repent and treat him as an honored guest. The second episode is identical,word for word,w ith Item 5.25 [135]. It narrates the episode of the wife of Yanzi’s eight-foot tall charioteer who is seeking to divorce her husband on the grounds that he drives Yanzi’s chariot in a pompous manner. Yanzi,she argues,is not even six-feet tall[45] and acts modestly,even though he is a renowned prime minister. Her argument brings the charioteer to immediately repent and change his manner driving and Yanzi,who notices the change,recommends him for a high governmental post.
The two episodes are subtle,focusing on one implicit question: what does authentic identity consist of? In that regard,the two episodes can be analyzed from both philosophical and psychological perspectives. However,regarding Yanzi’s biography,they contain no information about his life except for giving indirectly the detail that he was a person of diminutive stature. It seems that this part of Yanzi’s biography again merely illustrates his set of virtues by narrating two accounts that lead the readers to realize two aspects of Yanzi’s character: that he spared no effort to save a fellow person from a life-threatening predicament; that he was self-critical and thus able to change attitude in a complex personal situation; and that he had a profound ability to go beyond the outward appearances and fathom the inner reality of the people who surrounded him.
What was Sima Qian in incorporating these two episodes into Yanzi’s biography? Again,in light of the “admiring” concluding epilogue,and given that Guan Zhong’s biography in chapter 62 consists of 525 words while Yanzi’s (at this point) consists of only 75,it appears that Sima Qian needed to lengthen Yanzi’s biography in order to create a more balanced proportion between his discussions of Yanzi and Guan Zhong.
The third part of the biography[46] is m ainly a literary assessm ent focusing on the respective w ritings of Guan Zhong and Yanzi. It reads:
“I have read th rough M r. Guan Zhong’s [1] ‘Shepherding the People,’[2] ‘Mountains on High,’ [3] ‘Chariots and Horses,’ [4] ‘Light and Heavy,’ and [5] ‘Nine Bureaus’[47] as well as the [6] ‘Spring and Autumn Annals of Master Yan.’ Their words are expressed w ith such detail that having seen the books they w rote I wanted to examine their actions and therefore have attached their biographies. As for their books,many people today have copies of them and therefore I have not selected from them but instead have selected some neglected stories.”[48]
The third part of the biography entails a major stage in the history of the Yanzi’s text. Sima Qian clearly states that he read five titles attributed to Guan Zhong as well as an additional sixth text,entitled The Spring and Autumn Annals of Master Yan 晏子春秋 attributed to Yanzi. He also states that these texts were finely w ritten and very popular in his time. This statement provides solid evidence that by around 100 BCE,Yanzi’s text had assumed its textual identity as an authentic biography. Further confirmation for this status is offered by its metonymic title (春秋)[49] given in the Shiji,and also by its broad circulation in the Han intellectual milieu.[50]
The last,fourth part of chapter 62 is the epilogue of the two combined biographies of Guan Zhong and Yanzi.[51] Sima Qian presents the hidden tension between the historical memories of the two Qi ministers that he combined in chapter 62 to its culmination by issuing an implied harsh judgment against Guan Zhong:
Guan Zhong was what the world refers to as a worthy official,but Confucius belittled him. Could it be because he considered the Way of the Zhou to be in decline and [because] Duke Huan was worthy,yet Guan Zhong did not exhort him to become king,but rather proclaimed himself Overlord?
W ith these w ords,all o f Guan Zhong’s previously narrated achievements as a figure who shaped the history of Qi are reduced to mediocrity,w ith the metaphor Confucius had coined about him: “Guan Zhong was a vessel of sm all capacity” 管仲之器小哉.[52] As for Yanzi,however,Sima Qian brings the epilogue to its end w ith words of strong praise that show his great admiration for him as a personal idol:
“When Yanzi fell down upon the corpse of Duke Zhuang,he would not leave until he completed his ritual duties for his lord.[53] Could we say that he was ‘one who regards seeing what was right but not doing it as cowardice’?[54] When it came to remonstrating,he was not afraid to brave his ruler’s displeasure,and he was one whom we might refer to as ‘exhausting his utmost in how to be loyal as he advances,and in how to correct his ruler’s faults as he withdraws.’[55] If Yanzi were still alive,though I were only holding the whip for him,I would be pleased and longing to serve him.”[56]
These final words w ith which Sima Qian imagines himself playing the role of Yanzi’s old driver during the dangerous time surrounding the assassination of Duke Zhuang[57] conclude Chapter 62 of the Shiji. Sima Qian expresses his longing to serve Yanzi and admiration for him,leaving the reader to reflect upon the two memoirs and ponder who among their two heroes ascended to a greater dominance in the history of Qi. However,the link between the two memoirs has been cleverly achieved and,indeed,in the years immediately follow ing the composition of the Shiji,Guan Zhong and Yanzi were referred to jointly and evaluated in comparative judgment as equals.[58]
The Yanzi figure continued to make a mark in the Han intellectual sphere during the first century BCE. The Yantie lun 盐铁论 by Huan Kuan桓宽 (fl. 81–60 BCE) refers to him several times not only as a historical personage[59] but also as the source of a quoted text.[60] Moreover,toward the end of the century,his figure took on unforgettable proportions in a definitive text that Liu Xiang collated,prefaced,and entitled “Yanzi.”
(Ⅲ-3) The Form ation of Liu Xiang’s Yanzi
In 26 BCE,Emperor Cheng 成帝 (r. 33–7 BCE) ordered Liu Xiang刘向,the great Han bibliographer,to co llect and collate canonical scriptures,traditional narratives,and philosophical and literary works of diverse kinds from all over the empire,along w ith those stored in the repositories of the palace.[61] Subsequently,Liu Xiang collected over eight hundred anecdotes that focused on the life and time of Yanzi and w ith much effort turned them into a set of 215 glued items which exhibit an apparent thematic and faint chronological order,that form the definitive text from which Yanzi’s distinctive personal legacy was fashioned. This was a legacy of an exceptional personality of self-defined vocation and of diminutive body stature; an epitome of thriftiness and generosity alike; a master of psychological manipulation and com ic,theatrical gestures; a profound scholar of vast know ledge and pragmatic reasoning,and a strong advocate against slavishly follow ing the supernatural; a moral authority and courageous role model for the people of his era,who made Sisyphean efforts to lead the absurdly corrupt rulers he served towards just rulership; an equal to Guan Zhong and a match—even a rival—to Confucius.[62] Above all,however,the definitive text of 215 items that Liu Xiang produced was the first literary biography in ancient China that focused exclusively on one leading protagonist—Yanzi—a rare individual,a minister of the Altars of Soil and Grain;[63] a master w ithout a disciple,who was a true hero of his life.[64] For almost each of his collated works,Liu Xiang presented a memorial to the Emperor consisting of a table of contents; an account of the textual material that was used in forming the collated text; a short narrative of the life,time,and works of the eponymous author; and a discussion of the value,authenticity,and authorship of the presented text.[65] Liu Xiang’s report on Yanzi 晏子,[66] follows the exact same pattern: It opens w ith a table of contents which consists of eight pian. The first six are named “Inner Chapters” (内篇) and are respectively entitled “Remonstrations” 1 and 2(谏上下),“Queries” 1 and 2 (问上下),and “M iscellany” 1 and 2 (杂上下). The seventh and eighth pian are designated “Outer chapters” (外篇) and are respectively entitled “Repetition cum Difference” (重而异者) and “Incompatible w ith Classical Learning” (不合经术者).[67] A fter the table of contents,it opens w ith the project he set himself to accomplish,in which he recounts the main points surrounding the process involved in collating the different texts of Yanzi,and acknow ledges the people who,under his editorial leadership,contributed to bringing the definitive text into existence. In the second part,he draw s a lucid biographical sketch documenting recurring themes in Yanzi’s legacy. Finally,in the third and last part of his report,he critically evaluates the literary and philosophical value of Yanzi,providing some answers regarding the authenticity and the authorship of the text and justifying his bold editorial decision to include anti-canonical items in the text. The Report reads:
The Comm issioner of Water M anagement of the Eastern Part of the Metropolitan Area and Counsellor of the Palace,Your Servant [Liu] X iang says: The royal copies of Yanzi 晏子,which Your Servant has collated,consisted of eleven pian 篇. Together w ith [Fu] Can 富参,[68] the Commandant of Changshe 长社,Your Servant has carefully collated and compared them w ith the five-pian text used by the Grand Astrologer,with Your Servant’s own one-pian text,and with the thirteen-pian text used by [Fu] Can. The entire corpus of royal copies and private copies of Yanzi formed a set of thirty pian,comprising 838 items 章. I discarded twenty-two pian that duplicate the same material comprising 638 items,and established a definitive text of eight pian comprised of 215 items. Thirty-six items in the definitive text were absent from the private copies,while seventy-one items were absent from the royal copies. The royal and private copies were incorporated into the definitive text after com paring each of the variants and fixing the text on this basis. The royal copies had the character yao 夭for fang 芳; the character you 又 for bei 备; the character xian 先 for niu 牛; and the character zhang 章 for zhang 长—and many other occurrences of this kind.[69] I have combined and pruned them w ith great care and after having transcribed everything into the definitive text,I inscribed it on dried bamboo slips cured over fire so that fine copies may be made from them.
Yanzi’s personal name was Ying,his posthumous nam e Pingzhong 平仲. He was a native of Lai—today’s Donglai 东莱. He was a person of broad learning and retentive memory,and possessed a thorough understanding of ancient and modern learning. He served Dukes Ling,Zhuang,and Jing of Qi,and carried out his activities w ith parsimony.[70] He exhibited supreme loyalty and remonstrated relentlessly in leading the state of Qi. On the strength of his efforts,these rulers were able to rectify their conduct and the people were able to keep their family together. When he was not in government service,he retired and plowed in the countryside; when he was in government office,he never violated his principles. He could not be intim idated by evil; even when a naked blade was placed upon his chest,he did not relent to Cui Zhu’s 崔杼 intimidation. He presented his remonstrations to the dukes of Qi; his words were elusive but telling,rhetorically pleasing but cutting. When he served as an envoy to the state sovereigns,no one could contravene his arguments. Such was the extent of his broad know ledge and thorough understanding. He was virtually equal to Guan Zhong 管仲. Within his family,he treated his relatives w ith proper affection; towards others,he treated the worthy generously. He served in the position of Prime M inister and received an emolument of tenthousand zhong 钟.[71] Therefore,there were more than five hundred fam ily members and relatives that depended upon his emolument in order to buy clothing and food,and the number of scholar-recluses who depended upon
1 Duke Zhuang was invested as the ruler of Qi by the powerful Cui Zhu 崔杼,who later murdered him for having an extra-marital affair w ith his wife. For a detailed description of Duke Zhuang’s assassination and for an exposition of Yanzi’s moral dilemma over serving the new ruler of Qi,Duke Jing 景公,who was also enthroned by Cui Zhu,see YZCQ,Item 5.3 [113]; Zuozhuan,B9.25.2/282. his emolument in order to receive their daily bread was great. Yanzi wore coarse clothing and covered himself in deer pelts; he drove a worn out nag and a broken down chariot. He exhausted his entire emolument on family,relatives,and friends. Because of all this,the people of Qi treated him with honor. Apparently,Yanzi was a person of diminutive height.
A ll the text’s first six pian are loyal remonstrations presented to his rulers; their literary style is remarkable,their moral principles can serve as paradigms,and all of them accord w ith the principles of the Six Classics. In addition to these six pian,there were duplicate items of somewhat different words and phrases that I dared not omit but rather rearranged them into one additional pian. There were also items that were somewhat not in accord w ith the learning of the Classics. They do not seem to be the words of Yanzi. I suspect that they were the product of m asters of rhetoric of a later age. Once more,I did not dare omit them and have arranged them in one pian. A ltogether the text consists of eight pian; the first six should always be available at your majesty’s side for your royal perusal. I have carefully arranged the text according to its proper sequence. Your Servant,[Liu] Xiang,hereby presents them to your Majesty,at the risk of my own life.[72]
Certainly,the first part of Liu Xiang’s Preface is exclusively based on a body of various pericopae laid before him as the chief editor,and collator of the text. Only he could authentically tell future readers how the process of producing the definitive text of Yanzi had evolved. His personal involvement w ith this process is also reflected in the third and last part of the Preface,in which he expresses not only his scholarly view s regarding the value of the text but also his m isgivings about setting off the items of the six “Inner Chapters”—which he thought authentically recorded Yanzi’s own words—from the spurious items cited by masters of rhetoric of later times,in the eighth,“Outer Chapter,” that he decided to retain. As for the second, biographical part of the Preface—readers who examine it in comparison to the biography of Yanzi in the Shiji w ill notice that Liu Xiang shares w ith Sima Qian the same admiration for Yanzi. Nevertheless,while Sima Qian was very careful not to draw extremely from the text that he,according to his own statement,had carefully read—The Spring and Autumn Annals of Master Yan,[73] Liu Xiang borrowed heavily from his collated Yanzi in draw ing up his version of Yanzi’s biography.[74]
From that moment onward,the historical legacy of Yanzi has been primarily represented and judged by the text (and preface) that Liu Xiang produced and introduced to his peers. A lmost all other recorded echoes of Yanzi that have been found to date in other texts have been dimmed w ithin the historical recollection.
(Ⅳ) Taking Hold in the Process of Transm ission: The Yanzi and the YZCQ from the Later Han through the Tang
A fter the death of Liu Xiang (d. 8 BCE),from the first century CE through the Tang,Yanzi’s text,entitled either Yanzi or YZCQ,was listed,quoted and evaluated in the follow ing 15 stages in its textual transmission:
It was probably listed under the title YZCQ in the section devoted to Confucian w ritings in Liu Xin’s 刘歆 (d. CE 23) bibliographical catalogue — Qilue 七略 (Seven Summaries),as a text of 7 pian.[75]
It was listed under the title “Yanzi” in the section devoted to Confucian w ritings in Ban Gu’s 班固 (32–92 CE) bibliographical chapter of the Hanshu 汉书 —Yiwenzhi’s 艺文志 (Treatise on Literature),as a text of 8 pian.[76]
The Lunheng 论衡,by Wang Chong 王充 (27–100 CE),referred to both Yanzi and Guan Zhong as “successful w riters;”[77] it also quoted various statements pronounced by the Yanzi figure,almost all of which appear in parallel in the received text of the YZCQ.
The Fengsu tongyi 风俗通义,by Ying Shao 应劭 (d. ca. 204 CE),quoted the entire item 6.6 [146] of the YZCQ,identifying it as drawn from the YZCQ.[78]
The Kongcongzi 孔丛子,a work w ritten by Wang Su (195–256 CE) in his youth,[79] noted that the title Chunqiu 春秋 was included in the title of the “Book of Yanzi.”[80] It also strongly attacked the notion that Yanzi was an anti-Confucian who subscribed to Mohist views.[81]
Liu Xie 刘勰 (465–522) stated in his The Literary M ind and the Carving of Dragons 文心雕龙 that Yanzi’s and Guan Zhong’s w ritings are characterized by clarity in their factual accounts through their refined language style.[82]
The Shuijing zhu 水经注,by Li Daoyuan 郦道元 (d. 527),quoted a section from Item 2.24 [49] of the YZCQ,specifying Liu Xiang’s prefaced text of the YZCQ as the source of the citation.[83]
The Tang encyclopedia (类书),Beitang shuchao 北堂书钞,completed by Yu Shinan 虞世南 (558–638) before 618,quotes the Yanzi and the YZCQ dozens of times,[84] all of which appear throughout the eight pian of the received text of the YZCQ.[85]
The Tang encyclopedia,Yiwen leiju 艺文类聚,completed by Ouyang Xun 欧阳询 (557–641) between the years 622 and 624,quoted the Yanzi and the YZCQ dozens of times,all of which appear scattered throughout the eight pian of the received text of the YZCQ.[86]
The Qunshu zhiyao 群书治要,an anthology of canonical and early historical statecraft,completed in 631 by Wei Zheng 魏征 (580–643),contains,listed under the authorship of Yan Ying,a collection of 38 full items entitled Yanzi,[87] all of which appear throughout the eight pian of the received text of the YZCQ.[88] These 38 intact items consisted of 7,047 characters,which amount to no less than 17% of the entire bulk of 41,324 characters of the received text. The items are rearranged,sometimes in slightly different sequence,to form a Yanzi anthology of six parts. A lthough they are titled identically as the six “Inner Chapters” of Liu Xiang’s Yanzi: “Remonstrations,” (1 and 2),“Queries” (1 and 2),and “M iscellany” (1 and 2),they also include several items from the 7th and the 8th “Outer Chapters” of the text as well.
The bibliographical monograph of the Suishu Jingji zhi 隋书经籍志from 656 listed the YZCQ in the section devoted to Confucian w ritings as a seven-juan 卷 text,authored by Yan Ying.[89] This bibliographical monograph is the only one compiled between the Han and the Tang,which is still preserved intact,and includes features from several now lost pre-Tang catalogues including Ruan Xiaoxu’s 阮孝绪 (479–549) Qilu 七录.
The official dynastic history of Tang,the Jiu Tangshu 旧唐书 by Liu Xu 刘昫 (887–946),was later recompiled by the Song scholars Ouyang Xiu欧阳修 (1007–1072) and Song Qi 宋祁 (998–1061) as Xin Tangshu 新唐书. The two “histories” listed the YZCQ in the section devoted to Confucian w ritings as a seven-juan 卷 text,authored by Yan Ying.[90]
The Li Shan 李善 (d. 689) commentary to the Wen Xuan 文选注quoted the YZCQ about 40 times. All these references appear in the eight pian of the received text of the YZCQ.
The Tang encyclopedia Chuxueji 初学记 by Xu Jian 徐坚 (659–729) quoted the Yanzi and the YZCQ dozens of times,all of which appear throughout the received text of the YZCQ.[91]
The philosophical anthology Yilin 意林,authored by Ma Zong 马总 (d. 823),listed the text under the title Yanzi,as an eight-pian text,comprising 16 items,all of which appear in the received text of the YZCQ.[92]
To sum up,during the period under discussion:
1. The Yanzi text was listed,recognized,and transm itted under two titles: Yanzi and YZCQ.
2. Liu Xiang’s prefaced Yanzi was circulated and directly referred to as a source of citation.
3. A ll the text’s Inner Chapters’ titles were known and referred to by their names.
4. Yan Ying was ascribed with the authorship of the text.
5. The text was classified as a Confucian text.
6. The literary value of the text was highly appreciated.
7. A considerable part of the text appeared in encyclopedias and compendia alike.
8. The text’s two Outer Chapters were sometimes reduced to one “Outer Chapter.” Occasionally,in the reduction of these two chapters,items from them were scattered under the names of the six “Inner Chapters.” Thus,we may reasonably assume that the outer chapters’ text subdivision to either pian or juan was somewhat flexible.
(Ⅴ) Identity M utation: Liu Zongyuan’s 柳宗元 (773–819) M ohist View of the YZCQ
This stable transm ission of a highly regarded text,ascribed to its ancient author and bearing a preface by its historical imperial collator,was suddenly disrupted. The fixed status that the text enjoyed until the end of the eighth century—its repeated listings in all dynastic bibliographies and major catalogues alike as a Confucian text,as well as the extensive quotations as a source of traditional lore in the most celebrated encyclopedias—ended rapidly. It was the great poet and statesman of the Late Tang,Liu Zongyuan,whose short essay “Discussing YZCQ” placed the exalted status of the text into doubt. His essay reads:
Sima Qian read the YZCQ and highly valued it,but he did not know how this text came to be. Some say: “Yanzi w rote it and others carried it forward.” Others say: “It was w ritten by the descendants of Yanzi.” A ll of them are w rong. I suspect that it was someone among the followers of Mozi in Qi who w rote it. M ozi loved frugality and Yanzi was famous for this trait in his generation. Therefore,the followers of Mozi w rote his deeds w ith great respect in order to promote their own methods.
A t this point,Liu Zongyuan summarizes basic features in Mohist philosophical agendas,concluding that the YZCQ shares basic traits w ith them.[93] He then continues:
Ever since Liu Xiang and Liu Xin,Ban Biao (3–54) and Ban Gu,fathers and sons all classified the text among Confucian works. It is extremely regrettable that all these scholars did not examine it more closely. Had the author not been a man from Qi,he could not have recounted Yanzi’s deeds; had the author not been a follower of Mozi,he could not have rendered it the way he did. Henceforth,when classifying the text amongst philosophical books,it should be classified Mohist. It is not that Yanzi was a Mohist,but those who w rote the book were followers of the way of Mohism.[94]
This concluding line is a bombshell in disguise—it disassociates the historical,“non-M ohist” Yan Ying from the M ohist-oriented text that bears his name. Furthermore,given the previous statement that Yanzi’s descendants likew ise did not author the text,it im plies that the text was not authentic. The upshot of this contention is that it did not represent authentically what it presumes to represent but serves as a vehicle to advance a hidden agenda of some Mohists from Qi to promote their own ideas.
To summarize: Liu Xiang had stated previously that the 8th pian of the Yanzi,which he created,contained problematic material that did not reflect the words of Yanzi but was probably the product of masters of rhetoric of a later age. Subsequently,the young Wang Su made a stupendous effort throughout the whole 18th chapter of the Kongcongzi to persuade his readers that Yanzi was not an anti-Confucian who subscribed to Mohist views identical to those he himself pronounced in the 8th pian of the YZCQ. But it was rather Liu Zongyuan’s blunt and unsparing criticism that changed the course of the entire text of YZCQ’s trajectory—its authenticity,date,authorship,and philosophical vision henceforth always would be subject to debate[95]—and almost each of these debates w ill always echo Liu Zongyuan’s iconoclastic views of the text,at least in spirit.
(Ⅵ) A Bibliographical Confusion: The YZCQ during the Song
The transm ission of the YZCQ during the Song splits into two inconsistent lines of transmission: the encyclopedic,which manifests the impression of a stable transmission of the text; and the bibliographical,which seems to undermine this impression.
(Ⅵ-1) The YZCQ in Song encyclopedias
The incorporation of many direct and indirect quotations from the YZCQ into the major encyclopedic compilations of the Song establishes an image of stable transm ission of a single text over a period of 300 consecutive years throughout the Song.[96] The follow ing list of 122 directly quoted items demonstrates the scope of the text’s popularity and its unchanging transm ission-status as a source of ancient lore during the Song:
When all the Song encyclopedia quotations of the YZCQ are digitally assembled to form an integrated database together w ith all the YZCQ quotations that appear in the Tang encyclopedias and anthologies,the appearance of identical items repeatedly quoted from the two eras suggests a continuous line of stable transmission of the YZCQ text that starches from the early Tang to the end of the Song.[97] One specific item in this database,however,deserves a privileged status in evaluating the transmission of the text. The last,215th item in the received text of the YZCQ is not only first directly quoted in two Tang anthologies,Qun shu zhi yao (631) and Yilin (ca. 823),but is then subsequently quoted in the Song encyclopedia Taiping yulan (938),and finally is quoted in the M ing encyclopedia Tian zhong ji 天中记 (1569). However,this passage can also be traced back to its ancient origin as bamboo strips no. 624-630,which comprises the sixteenth item in the Yinqueshan Yanzi.[98] The stages that these seven bamboo strips have traversed in their journey from Tomb #1 in Yinqueshan (140 BCE),through the Tang anthologies and the Song and M ing encyclopedias,up to the formation of the received text from the various printed editions,can therefore be specified,and can enhance the impression of the stable transm ission of the entire text of the YZCQ throughout the ages.[99]
(Ⅵ-2) The YZCQ in Song catalogues and bibliographical lists
From the 1st century CE through the Sui and the Tang,the Yanzi text was persistently transm itted under the titles Yanzi or YZCQ as either an 8 pian,7 pian or 7 juan text. Throughout this period,all ascribed the authorship of the text to Yan Ying himself,except for Liu Zongyuan (773–819),who argued in his “Mohist” essay that the text was a product of a follower of Mozi from Qi.[100] In the Song,however,many sources began to list the text as a 12 juan one.[101] Some scholars argued that Yanzi’s 8 pian text of the Han was no longer extant,and several major bibliographers denied to Yan Ying the authorship of the text.[102]
Wang Yaochen 王尧臣 (1001–1056) was probably the first to record in his catalogue,Chongwen zongmu 崇文总目,a 12 juan text of the YZCQ. He placed the text in the section devoted to Confucian w ritings,stated that the 8 pian text of the Han was no longer extant,and determ ined that it would be w rong to ascribe the authorship of the text to Yan Ying because,as he argued,the text was w ritten by later authors.[103]
Follow ing Wang Yaochen,Chao Gongwu 晁公武 1105–1180 also listed the YZCQ as a 12 juan text in his private library catalogue Junzhai dushu zhi 郡斋读书志. However,he incorporated a large quotation from Liu Zongyuan’s “Mohist” essay,explicitly adapted Liu’s argument,and placed the text in the section devoted to M ohist w ritings.[104]
Somewhat later,Chen Zhensun 陈振孙 (fl. 1211–1249) added in his Zhizhai shulu jieti 直斋书录解题 a detailed reference to the YZCQ,in which he tried to incorporate all possible bibliographical worlds. He placed the text back in its traditional section devoted to Confucian w ritings and then made the following comment:
YZCQ in 12 juan. W ritten by Yan Ying Ping Zhong,a High Officer from Qi. The Hanshu states that the text consists of 8 pian but the reference is to a text entitled Yanzi. In the Sui and Tang,the text was listed as 7 juan text and began to be entitled YZCQ.[105] Now the number of juan differs so I do not know if this text is indeed the original text or not.[106]
A t the end of the Song era,Wang Yinglin 王应麟 (1223–1296),commenting on the section devoted to Confucian w ritings in the Hanshu Yiwenzhi,adapted Liu Zongyuan’s “Mohist” view of the text. He also proffered an awkward explanation concerning the discrepancy regarding whether the text involves 8 pian or 12 juan. He followed Wang Yaochen’s argument according to which later authors compiled the text and thereby arrived at the conclusion that later textual accretion of material altered the inner division of the text from 8 pian to 12 juan.[107]
It is worth noting that,as late as the end of the Qing,Liu Shipei 刘师培 (1884–1919) added another convoluted explanation in order to deal w ith the conflicting bibliographical references to the text of the YZCQ as either 8 pian or 7 or 12 juan text. He suggested that during the Sui and the Tang,the two “M iscellany” (杂) chapters (pian 5 and 6) were combined to form one juan and that afterwards,in the Song,the first 5 juan of this “new” 7 juan edition were divided into two juan,forming a set of 10 juan that together w ith the “Outer” 2 juan (which remained intact) gave rise to the 12 juan edition of the YZCQ.[108]
It seem s,however,that the elegant w ay in w hich the editors of Siku quanshu (1782) dealt w ith this puzzle of divergent bibliographical references,approximately 100 years before Liu Shipei,should have resolved this confusion. Toward the end of the “Synopsis” (提要) of their textual version of the YZCQ,they added the follow ing short section in which they discussed the bibliographical discrepancies of the text:
The Hanzhi [Hanshu Yiwenzhi] and the Suizhi [Suishu Jingjizhi] listed the text as an 8 pian text. Then,in the bibliographies of Chen Zhensun and Chao Gongwu it was listed as a 12 juan text. It turned out that by then,the arrangements of the text’s chapters have already been considerably altered.[109]
In these few short lines,the Siku Quanshu’s editors first acknow ledge the existence of the bibliographical discrepancies recorded from the Han throughout the Song. They then suggest the obvious. Namely,they claimed that these discrepancies were the result of the considerable changes in the arrangement of chapters that the text had undergone over the decades; that is,that all these discrepancies were purely technical in nature,and nothing more. Simply stated,the entire uncertainty boils down to the confusion surrounding the transmission and editing of the text over the years.
(Ⅶ) In Print: The YZCQ from Song through Qing
(Ⅶ-1) Early Editions of the YZCQ: The Yuan woodb lock and the M ing m ovab le type editions
It is entirely reasonable to infer that a woodblock edition of the YZCQ already existed in the Song; however,evidence to that effect is quite paltry,alm ost nonexistent. Therefore,any such assum ptions rem ain a m atter of conjecture.[110] In any event,two of the notably earliest woodblock and moveable type editions of the YZCQ whose existence has been verified are:
[1) A woodblock edition stored in the fam ily collection of Zhang Jinwu’s 张金吾 (1787–1829) in Zhao-wen 昭文—today’s Changshu City (常熟) in Jiangsu Province (江苏).[111] It was comprised of 8 pian and included a table of contents for each chapter and explanatory headings that were inserted at the beginning of all 215 items of the text.[112]
[2) A movable type edition produced in the M ing,and preserved in Ding Songsheng’s 丁松生 (1832–1899) Library of 8,000 juan of Movable Type Editions in Renhe County (仁和) in Zhejiang Province (浙江省).[113]
This edition also consisted of 8 pian and included a table of contents for each chapter and exp lanatory headings that w ere inserted at the beginning of all the 215 items of the text. Years later,this M ing movable type edition was reproduced in the 20th century in the SBCK series and then became the text on which the text of the YZCQ ICS Ancient Chinese Text Concordance Series was based.[114]
(Ⅶ-2) Later editions of the YZCQ: The Siku Quanshu edition of 1780 and Sun Xingyan’s commentary edition of 1788
The Siku Quanshu editors classified the YZCQ in the sub-category of“Biographical Works” (传记) in the “History Branch” of their monumental collection of texts. In doing so,they avoided the need to determine whether the YZCQ comprised a Confucian or a Mohist text.[115] Their textual version,as previously noted,was based on the Li fam ily M ianmiaoge woodblock edition of the late M ing and therefore included neither a table of contents nor item headings,as did the late M ing movable type edition that was based on the Yuan woodblock edition. Their Synopsis (提要),which is printed as a preface to the text,begins by condensing into a few lines over 1,000 years of bibliographical history,from the late Han through the Song. They then characterize the contents of the text as similar to those of two famous Tang admonishment texts,[116] and then render their judgment concerning the text’s authorship problem. They hold that the name of the text’s compiler was m issing from the original text and that therefore Yan Ying was named as its author in order to give the text credibility. The editors then point to an absurd item that could not have possibly been included in the original text.[117] They continue to present the historical debate between those scholars who considered the YZCQ a Confucian text and those who thought the text was Mohist. They conclude that Yan Ying,the person,lived a bit before Mozi,at the time the wandering scholar Shi Jue (史角),whose disciples eventually became Mozi’s teachers,was staying in Lu,[118] so Yan Ying might have known him and been introduced to “Mohist” ideas even before Mozi himself. The Synopsis then refers to the fact that the text was listed under two titles—Yanzi and YZCQ—and determ ines that the use of the title YZCQ was appropriate because some texts such as Lüshi chunqiu and others were not chronologically arranged by years and months and nevertheless bore in their title the metonymy Chunqiu 春秋 (“spring and autumn”). The Synopsis ends by presenting the divergent bibliographical references to the text and explains their editorial choice for the adapted M ing edition.
Sun Xingyan’s 7-juan commentary edition contained 8 pian.[119] By interw eaving pian and juan,Sun X ingyan m anaged to resolve the historical problem of the bibliographical discrepancy that arose from the bibliographical references to the Han 8-pian text and the Tang 7-juan text. His edition,which was based on two M ing editions,included neither a table of contents nor any item headings. He appended a long set of philological notes,entitled Yinyi 音义,[120] of textual variations and clarifications,many of which were based on extracts from major encyclopedias such as Yiwen leiju,Chuxue ji,and Taiping yulan,as well as from the commentary to the literary anthology Wen Xuan. His edition included Liu Xiang’s Preface to the text,as well as his own Preface,[121] in which he made the follow ing important statements:
1. Based on internal textual and external bibliographical evidence —the text could not possibly have been considered a forgery.[122]
2. The text was created during the Warring States period after Yanzi’s death by the hands of guest retainers who collected stories about his deeds and activities from the official annals of the state of Qi.[123]
3. The censorship performed by some M ing publishers against the anti-Confucian items that appear in the last Outer Chapter of the text is unjustified.
4. The textual material in the six Inner Chapters of the text justifies its traditional inclusion under the section devoted to Confucian w ritings.
5. The explanation provided in the Yuhai for the discrepancies in the number of juan in the text is “preposterous.”
6. Scholars,such as Liu Zongyuan,who considered the YZCQ a Mohist text,are ignorant and lacking in understanding.[124]
7. The material contained in the last Outer Chapter of the text in no way harms the “great Way” of Confucianism. Liu Xiang likew ise referred to the last Outer Chapter of the YZCQ as containing inauthentic material w ritten by later people,and that in contrast to the authentic “inner chapters,”half of all ancient textual “outer chapters” always served as a platform for inauthentic w riters seeking credibility.
(Ⅶ-3) The YZCQ during last decades of the Qing
During the last several decades of the Qing,the continuing textual scholarship on the YZCQ did not register any major breakthrough providing new insights into the text. Notable during this period is the celebrated late Qing and early Republican scholar Liang Qichao 梁启超 (1873–1929),who,together w ith scholars such as the aforementioned Liu Shipei 刘师培 (1884–1919),focused on the central questions related to the YZCQ. Liang’s view of the text is a composite of many of the previous insights that Chinese scholarship had produced over the preceding two millennia.[125] Liang first offered the daring bibliographical observation according to which the Yanzi text listed in the Hanshu was not only the same text to which Sima Qian referred in the Shiji,[126] but probably the same text Liu An referenced in his Huainanzi; that is,the Remonstrations of Yanzi.[127] He then followed Liu Zongyuan’s view that the YZCQ was w ritten by followers of Yanzi in Qi but nevertheless simultaneously stressed,in light of the Siku Quanshu editors’ perspective,that these followers from Qi did not know Mozi (alluding to the fact that although the YZCQ m ight have a M ohist flavor,it had nothing to do w ith Mozi the person). Liang concluded his short piece by stating that the YZCQ acquired the celebrated name of “Yanzi”in its title only in later times,and that the actual date of its composition might be stretched beyond the Warring States period into the beginning of the Han dynasty. As for the received version,Liang expressed doubts as to whether the version is identical to the text referred to by Sima Qian and Liu An,but simultaneously averred that the received version is most probably the same text Liu Xiang compiled and is thus by no means a later product.
Towards the end of the Qing,one of the earliest scholarly references to the YZCQ in a Western language appeared in Shanghai. The British m issionary A lexander Wy lie in 1867 fo llowed the Siku Quanshu classification and listed the text in the section devoted to biographical works of his Notes on Chinese Literature. He then added a short comment saying that the YZCQ was w ritten some centuries before the Christian era and that it was “a personal narrative” of Yan Ying,“a reputed disciple of” Mozi.[128]
In summary: at the end of the Qing and during the early Republic years,the decisive questions about the authenticity and philosophical characteristic of the YZCQ remained suspended in air—the ongoing textual scholarship of the YZCQ was in need of a broader circle of contributors and committed readers of the text in order to create a space in which new alternative voices could cope w ith the unsettled questions surrounding the text. Indeed,as it turned out,during the 20th century,the YZCQ’s scholarly circles of committed readers and contributors did expand significantly —westwards.
(Ⅷ) An A rcheological Break through and Digital Hum anities: Expanded Accessibility to the YZCQ during the 20th Century and into the 21st
(Ⅷ-1) The YZCQ from 1923 to 1972
The discovery of 102 “Yanzi” bamboo strips and fragments,excavated at Yinqueshan in 1972,shook up the long history of textual scholarship of the YZCQ.[129] Prior to that,during the first seven decades of the 20th century,several Western and Chinese scholars made the YZCQ more accessible to students of Chinese philosophy and provided new perspectives regarding the problem of the text’s authorship,date,and authenticity. These scholars include A lfred Forke (1867–1944); Henri Maspero (1883–1945); George Kao (1912–2008); Richard L. Walker (1922–2003); and Burton Watson (1925–2017). In China the most prom inent YZCQ scholars were,among others,Zhang Chunyi 张纯一 (1871–1955); Wu Zeyu 吴则虞 (1913–1977); and Wang Gengsheng 王更生 (1928–2010).
A lfred Forke’s pioneering 1923 essay on the YZCQ still resonates to this day w ith important insights.[130] Forke argued that the YZCQ was an authentic—that is,an ancient—text,which he dated to the fifth century BCE.[131] He claimed that the text’s chief protagonist,Yanzi,dedicated his life to the service of his ruler and his people based on a set of pragmatic values. For Forke,Yanzi was a politician and not a pathbreaking theoretician immersed as Confucians had been—in his view—in a world of abstract know ledge and profound learning. Nor was he,in Forke’s eyes,a great statesman of Guan Zhong’s caliber; however,as a “charakter,” Forke argued,Yanzi outstripped the latter. Forke’s essay was comprised of several parts including a fully reconstructed biography of Yanzi drawn from various traditional sources; a detailed discussion of the history of the text; and a lengthy analysis of the text’s philosophy regarding topics such as ethics,thrifty,human relationships,and the interaction between the ruler and the people. Notable among these points is Forke’s groundbreaking analysis of the text’s advocacy of a kind of “natural religion” that makes it “ganz rationalistisch” in comparison to Confucius’ and Mozi’s religious beliefs. Finally,the essay presented an insightful examination of the authenticity of the first few items of the eighth chapter of the YZCQ,which contain aggressive attacks on Confucius and Confucianism.
Henri Maspero’s 1927 discussion of the YZCQ in his La Chine antique was less than two pages long; yet it comprised a masterful characterization of the entire text.[132] Maspero dated the YZCQ to the middle of the fourth century BCE and identified it,among other texts of that era,as a “historical romance”; that is,as a text whose w riter mixed genuine historical facts w ith his own imagination. Maspero was the first to observe that the text exhibits in addition to its Mohist inclination a contrarian attitude not only towards Confucius but also towards the Guanzi. He nevertheless maintained that the YZCQ’s philosophical ideas,for all their intellectual value,were uninteresting and banal. Regarding literary style,however,Maspero expressed his admiration for the liveliness w ith which many of the scenes were staged. His long bibliographical note[133] in which he traced references to the authentic YZCQ in various sources from Han through the Tang—that is,references to the text before it underwent editorial changes during the Song and Yuan—still serves as a persuasive argument that a great part of the original YZCQ text was authentically transmitted through the received version.
George Kao’s 1946 anthology Chinese Wit & Humor[134] included English translation of twelve relatively long items of the YZCQ. The anthology not only introduced the witty Yanzi to the Western reader but also granted Yanzi equal footing,as far as “w it” was concerned,w ith such monumental figures in Chinese history as Confucius,Mencius,Zhuangzi,Liezi and Hanfeizi.
Richard Walker’s 1952 weighty sinological analysis of the YZCQ introduced the text to academ ic circles in the English-speaking world and explored many of the text’s complexities long discussed by Chinese scholars.[135] To this day,Walker’s essay still provides a roadmap that sinologists should take into account when constructing their own critical studies of a text such as the YZCQ. Walker’s primary goal was to elevate the status of the YZCQ,“which has remained outside the pale of the extensive research,” and to see how Forke’s 1923 essay “stood the test of time” 25 years after it was published. In doing so,Walker suggests to replace the dignified title “Annals of the Philosopher Yen” w ith a more casual title such as “Speeches” or Aphorisms” of Yanzi. A t the same time,he characterizes Yanzi’s ability as it emerges from the text as merely “to extricate himself from ticklish situations by verbal gym nastics.” Thus,the reader is left perplexed as to the essayist Walker’s real intention. Obviously,one cannot elevate a long-ignored text like the YZCQ to the status of the most celebrated text of ancient China by deflating its title and by characterizing its protagonist as proficient in nothing more than the mediocre art of “verbal gymnastics.” And indeed,as the opening statements of the essay and all its introductory comments conclude,Walker actually aims at the heart of his real target: Forke’s 1923 essay. He begins by recording a few agreeable words concerning Forke’s “great amount of spadework” on the history of the text and its surrounding scholarship,and then launches an attack on Forke’s philosophical over-reading of the text and on his sinological methodology by listing flaws,culm inating in the follow ing somewhat disparaging statement: “Thus,while we may agree with Professor Forke on an early date for the YTCC [YZCQ],it is not alone for the reasons which he gives”.[136] Walker then critically examines the political m ilieu surrounding the composition of the text,grammatically analyzes the language of the text in comparison to other texts that exhibits similar contents,and brilliantly evaluates what is included and what is “not included” in the multi-layer text to reach the follow ing conclusion:
The YTCC [YZCQ] is an authentic text most probably pre-dating the Tso-chuan [Zuozhuan]. A likely date of composition is sometime two or three generations after the death of Yen-tzu [Yanzi] when stories about him had become legendary,and not so long after his death for impossible fabrications to have entered in or for the text to have been formalized,philosophically or institutionally. This would probably place it some time before 400 B.C.[137]
In 1962,Burton Watson’s excellent short subchapter on the YZCQ paved new directions for understanding and exploring the text,especially on psychological levels.[138] Watson placed the YZCQ in the “Philosophy” section of his survey of early Chinese literature and portrayed its chief protagonist,Yanzi,as a dour Mohist who advocated frugality and made the life of his indulgent ruler very difficult as a result of his “ubiquitous carpings.”Offering a fresh perspective on a story that for almost two millennia was championed as a rare example of a minister of great integrity who delivered honest remonstrations against his self-indulgent ruler,Watson took the Duke’s “side.” Watson’s new approach contravened the pious moralizing the text had as its core for m illennia,and transform ed it into a m odern theatrical drama that the audience is invited to interpret freely. Furthermore,Watson drew the attention of future readers of this drama to its portrait of a static world that allows neither progress nor breakthroughs. The world and the main protagonists of the YZCQ remain unmoved from their original positions; indeed,in Watson’s words: “A lmost invariably the Duke admits his folly and agrees to reform. Yet in the follow ing anecdote we find him back at his old mischief.” No doubt,Watson’s view of the YZCQ gave the text a flare of Godot absurdity—“Nothing to be done.”[139]
Between 1930 and 1962,two new annotated editions appeared in China that made the YZCQ much more accessible to potential readers everywhere. In 1936,Zhang Chunyi 张纯一 published an annotated edition entitled: Yanzi chunqiu jiaozhu 晏子春秋校注 (preface,1930).[140] This extremely rich and detailed commentary dealt w ith the textual difficulties of the text by first referring the reader to Sun Xingyan’s commentary of 1788 and then by m aking extensive comparative references to commentaries dealing w ith similar textual difficulties in other classical texts.
Then,in 1962,what still ought to be regarded as the most comprehensive edition of the YZCQ,was published in China. Wu Zeyu’s 吴则虞 Yanzi chunqiu jishi 晏子春秋集释[141] offered an encyclopedic work containing everything worth know ing about the text itself and the surrounding circumstances of its composition and transmission. Nearly all of significance that had been stated about the text—every quote and every interpretation over a period of almost two millennia—can be found in Wu Zeyu’s edition. Not only is his own commentary instructive and altogether enlightening regarding almost every textually difficult passage—to which he attached a m yriad of textual references from many other commentaries[142] —but he also provided appendices that gave the reader an extensive array of references to historical scholarship that had accumulated throughout the ages.[143] In his “Preface”[144] he developed a five-point argument in which he traced the authorship of the text to such Academ icians as Chunyu Yue 淳于越 (ca. 214 BCE) from Qi,who relocated to Qin 秦 and served the First Emperor after the unification of China.[145] The first three of Wu Zeyu’s five arguments were contextually external and therefore did not carry much weight; however,his fourth and fifth arguments provided a new dimension for the discussion of the authorship of the text. They pointed to some striking value-based ideas and stylistic sim ilarities between Chunyu Yue’s recorded remonstrations and those pronounced by Yanzi in the YZCQ. Unfortunately,the num ber of items in this list of com pared sim ilarities presented by Wu Zeyu was too small to derive a definitive conclusion that Chunyu Yue was behind the compilation of the text. Nevertheless,until now,these stylistic arguments should stir enough sinological interest to prompt the formation of a digital database of such similarities,a mosaic of “clues,”which could point at the direction of the identity of the “hands” which weaved the proto-text of the YZCQ. To conclude,Wu Zeyu’s monumental achievement as commentator,as bibliographer and as innovator of textual comparative research,is such that any future study of the YZCQ will have to regard his edition as the proper point of departure.
In 1966,Wang Gengsheng 王更生 completed his PhD dissertation “Yanzi chunqiu yanjiu 晏子春秋研究” (“Research on the YZCQ”) at the Taiwan National Normal University.[146] This was his first contribution to the study of YZCQ,which was subsequently followed by several other excellent works and two baihua translations[147] of the text,which garnered for him a central place in the historical scholarly study of YZCQ. The dissertation and its subsequently revised book edition constitute a concise summary of material that had already appeared in Wu Zeyu’s work of traditional Chinese scholarship concerning the authorship,authenticity,and date of composition of the YZCQ. The revised book also included a fully detailed biography and chronology of Yanzi’s life,as well as a thorough exposition and analysis of the text’s views on both philosophical and linguistic matters such as metaphysics,the spirits,life and death,ethics,political philosophy,finance,diplomatic affairs,self-cultivation,literary style,syntax,and the use of analogies.
Indeed,during the several decades prior to 1972,the YZCQ became much more accessible to readers and scholars in China and in the West alike. It was due to the continuous scholarly efforts in the study,commentary and partial translation of the text. In 1972,however,the discovery of bamboo strip fragments excavated at Yinqueshan forced the long history of textual scholarship of the YZCQ to reevaluate its own convictions in light of 102 bamboo strips.
(Ⅷ-2) The Yanzi Text in the Yinqueshan Han bam boo strips 银雀山汉墓竹简[148]
In 1972,in the course of a construction project,4,942 bamboo strips and fragments were excavated at Tomb #1 and Tomb #2 at the Yinqueshan银雀山,a small hill southeast of the city of Linyi 临沂 in Shandong Province,located,by some estimates,w ithin the southern reaches of the state of Qi. The period of burial for both tombs has been dated to approximately 140/134 BCE and 118 BCE,but the 21,728 characters[149] w ritten in Clerical Script (隶) on the 4,942 unearthed bamboo strips were probably w ritten somewhat earlier.[150] Members of The Committee for the Reconstruction of Yinqueshan Han Dynasty Bamboo Strips divided the 4942 bamboo strips excavated at Tomb #1 into two groups. The first group consisted of bamboo strips that bore textual material comparable to extant traditional texts,and the second group consisted of textual materials that otherw ise have been lost. Within the first group,102 strips arranged in a continuous sequence (numbers 528-630) formed sixteen distinct textual sets,which comprise at least one item from each and every one of the eight pian of the received version of YZCQ,corresponding in content and phraseology,with very similar strings of wording. Notably,the 102 Yinqueshan Bamboo Strips of Yanzi consisted of 2,970 characters,which amount to no less than 7.2% of the entire bulk of 41,324 characters of the received text.[151]
Two tables are presented below. Table 1 demonstrates the nearly exact textual correspondence between the first set of the Yanzi bamboo strips (No. 528-531) and the corresponding Item 3 in the received version of the YZCQ. Table 2 lists all the 102 Bamboo Strips,grouped in sixteen sets arranged in continuous order,alongside their corresponding items and chapter numbers in the parallel Items in the received version of the YZCQ.
Table 1: Textual correspondence between the first set of the Yanzi bamboo strips
[No. 528-531) and the corresponding Item 3 in the received version of the YZCQ.
1st row: Yinqueshan zhujian 银雀山汉墓竹简 Yanzi[152]
2nd row: Transcribed copy[153]
3rd row: YZCQ,Chapter 1.3 [Item 3]
Table 2: The bamboo scripts and their corresponding items and chapter numbers in the parallel items in the received version of the YZCQ
Yinqueshan zhujian Yanzi YZCQ’s Received version
Furthermore,between 1972 and 1977,other archeological findings such as bits of fragments of the Yanzi text dated from 179–49 BCE were excavated at Pochengzi 破城子 in Gansu 甘肃 (1972–1974),Dingxian 定县 in Hebei 河北 (1973),and Fuyang 阜阳 in Anhui 安徽 (1977). All these scripts,however,were too fragmented to allow a reconstruction of even one full coherent passage of the earliest text.[154]
To conclude: Striking textual similarities exist between the sixteen “Yanzi” sets of the 102 Yinqueshan Han Bam boo Strips and their corresponding items in the YZCQ’s received version. In light of all the other archeological findings of bits of fragments of the Yanzi text that were excavated in China between 1972 and 1977,and the internal textual evidence of the Huainanzi (according to which a text identified as The Remonstrations of Yanzi was known in 139 BCE,whose contents were notably similar to those of the received version of the YZCQ),[155] it would be quite reasonable to establish that around 200 BCE or maybe even earlier,several bamboo strip versions of a certain Yanzi text were circulating throughout the intellectual world of the early Han dynasty. And,in light of this conclusion,we may now turn to examine one exciting hypothesis positing the composition of a proto-Yanzi text took place at an even earlier date.
(Ⅷ-3) Chunyu Kun 淳于髡 (c. 340 BCE),[156] the “Jesters” 滑稽,and the Jixia 稷下 Patronage Comm unity Connection
In 1936,as Luo Jun 罗焌 (1874–1932) was concluding his survey on the historical scholarship of the YZCQ,[157] he introduced a,new,staggering idea. Luo proposed his own personal view of the text: that the recorded deeds and remonstrations in the YZCQ show great affinity to those recorded about jesters like Chunyu Kun from Qi,Jester Meng 优 孟 from Chu,and Jester Zhan 优 旃 from Qin.[158] Clearly,this idea was too abrupt and unsubstantiated. Nevertheless,it introduced a new perspective on the longheld view of the text in that it stripped the text of its distinctive historical and philosophical status and replaced the underlying identity of its chief protagonist,Yanzi,from that of a profound scholar-statesman to that of a mere master of rhetoric or jester.[159]
Forty years later,in 1976,Wang Gengsheng tossed a sim ilar idea into the arena of YZCQ scholarship. Wang Gengsheng suggested that such nonsense appears several items in the text,[160] indicating that it must have been w ritten by jesters,like Chunyu Kun and his likes,and it must reflect the imprint of the sophists of the Jixia patronage community.[161] Then,in 1986,the “Chunyu Kun hypothesis” was reinforced by Lü Bin 吕斌.[162] Lü presented a five-point argument show ing that various elements in Chunyu Kun’s recorded sayings and life history suggest a strong affiliation w ith the YZCQ and hence an authorial connection. Specifically,he argued that Chunyu Kun’s remonstrations were based on admiration for Yanzi;[163] that he had direct access to the Qi archive from which he could derive information concerning Yan Ying; that he,like Yanzi,did not strictly follow any school of philosophy; and that echoes of his ideas and his facetious style resonate throughout the YZCQ. Finally,Lü pushed the “Chunyu Kun hypothesis”a step further by pointing out that both Chunyu Kun and Yanzi were portrayed as extremely short people. This striking physical resemblance,Lü argued,in conjunction w ith all the other evidence,suggests that an author w ith a biographical profile akin to Chunyu Kun projected his own personal experiences,philosophical ideas,and body image into a text that bore the name of an honored figure—Yanzi—and thus the proto-YZCQ was born.[164] Gradually,the “Chunyu Kun hypothesis” became one of the dom inant theories of YZCQ authorship. Tao Meisheng 陶梅生,in his 1998 Baihua translation of the YZCQ,raised the possibility that the followers of Chunyu Kun compiled the text.[165] Lin Xinxin 林心欣,in her master’s thesis of 2000,after discussing questions of classifying the text as Confucian,or Mohist,or biographical,she added a whole new classification to the YZCQ entitled “Jesters” (俳优 ),in which she discussed the connection between Chunyu Kun and the authorship of the YZCQ.[166] Then,in 2005,Zhao Kuifu 赵逵夫 presented an eight-point,detailed and annotated argument largely replicating that of Lü Bin in 1986,in support of the hypothesis that the dim inutive Chunyu Kun compiled the YZCQ,an account of the diminutive Yanzi,in order to gain respect for himself despite his short stature.[167] Finally,in 2011,Andrew Meyer published a stimulating,comprehensive paper in which he persuasively showed,inter alia,that the YZCQ was a product of Qi state patronage of the Tian 田rulers (379–221 BCE) and that the “proto-Guanzi and YZCQ served (among other things) as extended pleas to intellectuals to think of themselves not as ‘Confucians’ or ‘Mohists’ but as ‘Qi-ists.’”[168] The notion that Chunyu Kun or his followers were the compilers of the YZCQ became quite commonplace in recent sinological w ritings,as well as on several internet sites.[169] However,the scope and the in-depth analysis of Meyer’s paper paved the way for future sinological research that may bring together his insights w ith those revealed by other researchers regarding the Jixia patronage community,the life and time of Chunyu Kun,and the jesting tradition in ancient China.
Such research possibly yield a concrete image of the first author or compiler of the proto-YZCQ.
(Ⅷ-4) The YZCQ scholarship in recent years and its future in the age of Digital Humanities and Computational Thinking
Ever since the Yinqueshan Han Dynasty bam boo strips w ere reconstructed,analyzed and circulated after 1974,the YZCQ became the focus of renewed sinological interest. Many scholars shared their insights and expertise,exploring new avenues to investigate textual aspects and historical issues connected w ith the study of the YZCQ.[170] Notable among these scholars are Rainer Holzer who,in 1983,translated the six Inner Chapters of the YZCQ into German,to which he added a learned introduction,detailed textual comments,and many other valuable notes.[171] In 1993,Stephen Durrant contributed an elegant roadmap to the text and its history in M ichael Loewe’s bibliographical guide on early Chinese texts,thereby creating an invaluable aid for any student of the YZCQ. [172] In China, in addition to the appearance of dozens of new baihua annotated translations of the text,no less than 280 scholarly articles have been published about the YZCQ in academic journals between 1974 and 2010. Moreover,30 theses and dissertations have been presented at various Chinese universities and colleges between 1995 and 2010.[173]
It seems,however,that unless a “new” set of bamboo strips of Yanzi or any other relevant text[174] is uncovered from some future archaeological site,the study of the birth of the proto-text of the YZCQ may have exhausted itself and cannot proceed beyond the achievements of the scholarly work of the past decades. Future study of the origins of the YZCQ must therefore be conducted “algorithm ically” w ith tools recently developed by scholars of Digital Humanities. A t present the title “晏子春秋” yields a massive approximately 400,000 results in Google search.[175] Websites such as the Chinese Ancient Texts Database (CHANT); the Chinese Text Project (Ctext),the Digital Resources of Sinology of the Academia Sinica,and many others,provide millions upon m illions of searchable characters representing the vast ocean of Chinese texts from ancient times to the present,including canonic,marginal,broadly circulated,and rare texts. Text mining methods w ill definitely enable analysis of the big database of these m illions of characters and suggest when certain names,words or idioms appear for the first time,and how they are utilized afterwards,allow ing researchers to successfully answer questions concerning the time of composition of texts.[176] Furthermore,websites like Harvard’s China Biographical Database Project (CBDB) and Fairbank Center’s Digital China already bear potential for an academic undertaking in which stylometric analysis and data extraction techniques play a major sinological role. Indeed,stylometric programs w ill learn the patterns of a certain author and calculate the probability that this author w rote a given text whose authorship is yet unknown,thereby solving questions regarding the authorship of particular documents. As recently reported,[177] two researchers from the University of Texas at Austin learned the patterns of 54 dramatic works of a certain author and calculated the probability that this author w rote a given text previously attributed to someone else. Their stylometric analysis identified Double Falsehood,a play previously attributed to the Shakespearean scholar Lew is Theobald (1688–1744),as being the work of William Shakespeare himself. This discovery,which has shaken the Shakespearean scholarly community,provides every reason to imagine that a similar disruption in the near future is possible in the world of scholarship of the YZCQ.
注释
[1]Follow ing the first line of Yanzi’s biography in the Shiji 史记 (Shiji,2134),Sima Zhen’s (fl. 730) 司马贞,the author of the Shiji suoyin 史记索隐,points out that Yanzi’s personal name was “Ying” (婴),posthumous name “Ping” (平),and courtesy name “Zhong” (仲). According to Liu X iang 刘向 (79–8 BCE),“Ping Zhong 平仲”w as his posthumous name. See,Section (III-3). On the basis of his posthumous name “Ping 平,” he is often referred to as Yan Pingzhong 晏平仲.
[2]Yan Ying is traditionally believed to have lived from 589 to 500 BCE. While the Shiji 史记,1505,gives the year of Yanzi’s death (500 BCE),his birth year remains uncertain and disputed. For a detailed discussion of the sources for these dates,see Wang Gengsheng 王更生,Yanzi chunqiu yanjiu 晏子春秋研究. Wenshi 文史,Taibei (1976): 18-34.
[3]It would be,nevertheless,unreasonable to cast doubt on the traditional view of the historicity of Yan Ying who lived in this period. See the observation Richard L. Walker made decades ago in his “Some Notes on the Yen-tzu ch’un-ch’iu.” Journal of the American Oriental Society 73.3 (1953): 156.
[4]Analects 论语,5.17/10/21. For alternative datings of the Analects,see M ichael Hunter. Confucius beyond the Analects. Studies in the History of Chinese Texts. Leiden and Boston: Brill,2017; John Makeham,“The Formation of Lunyu as a Book.” Monumenta Serica 44 (1996); Tae Hyun Kim and Mark Csikszentm ihalyi,“History and Formation of the Analects,” in Am y O lberding,ed.,Dao Companion to the Analects. Dao Com panions to Chinese Philosophy 4. Dordrecht,Netherlands: Springer,(2013): 21-36.
[5]Zuozhuan 左传,A9.1/226-B10.32/406,passim. For a reconstruction of Yanzi’s biography elicited from the Zuozhuan and other traditional sources,see Zheng Qiao 郑樵 (1104–1162),Tongzhi 通志,卷九二,Siku Quanshu 四库全书,15A8-24B7; http://ctext.org/library.pl?if=en&_file=10374&_page=30 (2006-2017); Wang Gengsheng,ibid.,9-18; A lfred Forke,“Yen Ying,Staatsmann und Philosoph,und das Yen-tse tsch’untch’iu.” Asia Major (first series) Introductory Volume (Hirth Anniversary Volume) (1923): 107-124; Rainer Holzer,Yen-tzu und das Yen-tzu ch’unch’iu. Peter Lang,Frankfurt,(1983): 2-7.
[6]The three dukes whom Yanzi served were Duke Ling 灵 (r. 581-554 BCE),Duke Zhuang 庄 (r. 553–548 BCE),and Duke Jing 景 (r. 547–490 BCE). For a vivid portrait of Yan Ying from the Spring and Autumn period (722–453 BCE) based on several anecdotes of the Zuozhuan,see Yuri Pines,“From Teachers to Subjects: M inisters Speaking to the Rulers,from Yan Ying 晏婴 to Li Si 李斯,” in Garrett P. S. Olberding,ed. Facing the Monarch: Modes of Advice in the Early Chinese Court,Harvard East Asian Monographs 359,Cambridge,Mass.,and London (2013): 70-80. For a literary-poetic analysis of two dialogues in the Zuozhuan—one betw een Yan Ying and Shuxiang 叔向,and another between Yan Ying and Duke Jing 景公—see Waiyee Li,The Readability of the Past in Early Chinese Historiography. Harvard East Asian Monographs 253. Cambridge,Mass.,(2007): 349-355. However,(见下页)(接上页) regarding the historicity of the portrait of Yan Ying in the Zuozhuan,we should also note w ith caution Pines’ remark about the ongoing “heated dispute”regarding the “several datings” of the Zuozhuan. See Yuri Pines,“Rethinking the Origins of Chinese Historiography: The Zuo Zhuan Revisited” Journal of Chinese Studies,The Chinese University of Hong Kong,49 (2009): 429-442. See also the “Introduction,” in Stephen Durrant,et al.,trs. Zuo Tradition / Zuozhuan 左传: Commentary on the “Spring and Autumn Annals”. 3 vols. Classics of Chinese Thought. Seattle: University of Washington Press,2016; Barry B. Blakeley,“On the Authenticity and Nature of the Zuo zhuan Revisited.” Early China 29 (2004): 217-67.
[7]I borrow the term “the Jixia patronage community” from Andrew Meyer,“‘The A ltars of the Soil and Grain are Closer than Kin’ 社稷戚于亲: The Qi 齐 Model of Intellectual Participation and the Jixia 稷下 Patronage Community.” Early China 33-34 (2010-11): 37-99.
[8]Shiji,2347.
[9]Guan Zhong (c. 720–645 BCE) or Guanzi 管子 is the famous seventh-century prom inent m inister in the State of Qi who effectively served Duke Huan 桓公 (d. 643 BCE) and was chief instrumental in making him the first Overlord (霸) in the Spring and Autumn Period. The philosophical collection entitled Guanzi 管子 is traditionally ascribed to him.
[10]For Andrew Meyer’s penetrating analysis of the rivalry betw een Mencius,representing the “textual ideal,” and Yanzi,representing “social reality,” see Andrew Meyer,ibid.,86-95.
[11]Mencius 孟子,3.1/14/6-15. Notably,however,Mencius 2.4/9/9-17 contains also a favorable reference to Yanzi in a poetic form concerning the remarkable persuasive power of Yanzi to bring about a fundamental change in Duke Jing’s attitude towards his people.
[12]Xunzi 荀子,27/131/16-17. It is not clear who the “Master” is; it can be either Confucius or Xunzi.The Xunzi also contains an episode (27/134/1-4) in which Yanzi bids farewell to Zengzi 曾子 with some wise words.
[13]For a discussion of the dating of Mozi’s Chapter 39 to the late Warring States Period,see Sixin Ding,“A Study on the Dating of the Mozi Dialogues and the M ohist View of Ghosts and Spirits.” Contemporary Chinese Thought,vol. 42/4 (2011): 51-53; Carine Defoort and Nicolas Standaert,eds. The Mozi as an Evolving Text: Different Voices in Early Chinese Thought. Studies in the History of Chinese Texts 4. Leiden and Boston: Brill (2013): 5; John Knoblock and Jeffrey Riegel,trs. Mozi墨子: A Study and Translation of the Ethical and Political Writings. China Research M onograph 68. Berkeley: Institute of East Asian Studies,University of California (2013): 40.
[14]The Lüshi chunqiu compiled in 239 BCE by a group of scholars retained by Lü Buwei,aimed to explore all the know ledge of the world in one monumental encyclopedia.
[15]Han Fei 韩非 (ca. 280–233 BCE) was the name of a multifaceted Chinese philosopher who,w ith several other philosophers and statesmen,developed the main doctrines of Legalism (Fajia 法家) in the end of the Warring States period.
[16]Mozi 墨子,9.7/65/24-66/14.
[17]E.g.,Kongcongzi 孔丛子,6.1/60/5-62/28 w here Kung Fu 孔鲋 (子鱼),an eighth generation descendant of Confucius who went to great lengths to refute a set of Mohist claims that Yanzi was an anti-Confucian who criticized Confucius on the basis of a Mohist agenda. See also Zhang Zhan’s 张湛 (fl. ca. 370 CE) commentary on the Liezi 列子 (冲虚真经),SBCK,7/2B14; Liu Zongyuan 柳宗元 (773–819),Bian Yanzi chunqiu 辩晏子春秋,http://ctext.org/w_iki.pl?if=gb&_chapter=493447 (2006–2017): No. 16.
[18]Lüshi chunqiu,12.2/59/1-14.
[19]Ibid.,16.2/91/19-26.
[20]Ibid.,20.3/130/27-29.
[21]Hanfeizi,33/90/19-23.
[22]Ibid.,34/97/20-24; 34/98/22-99/9.
[23]Ibid.,37/117/26-118/4.
[24]Huainanzi,21/223/19-228/31.
[25]See,e.g.,YZCQ,Item–1.5 [5]; Item–1.25 [25].
[26]See,e.g.,YZCQ,Item–1.23 [23].
[27]See,e.g.,YZCQ,Item–3.5 [55]; Item–4.2 [82].
[28]See,e.g.,YZCQ,Item–2.7 [32].
[29]See,e.g.,YZCQ,Item–8.9 [206]. According to Item 8.9 [206] in the received version,the huge bell indeed em itted a thunderous sound; however,the received version does not mention pheasants crying out outside the city walls when the bell was struck. In view of the “m issing pheasants” from the received version of the YZCQ,Rainer Holzer raised doubts regarding the affiliation betw een the text referred by the Huainanzi as The Remonstration of Yanzi and the received version. See Holzer’s comment,ibid. 9.
[30]See,e.g.,YZCQ,Item–4.1 [81].
[31]See,e.g.,YZCQ,Item–1.12 [12].
[32]Huainanzi,21/228/13-14. Except for two m inor changes,the translation is adapted from John S. Major,et al.,trs. The Huainanzi: A Guide to the Theory and Practice of Government in Early Han China. Translations from the Asian Classics. New York: Columbia University Press (2010): 865.
[33]For Liu Xiang’s definitive text of Yanzi,see,Section (Ⅲ-3).
[34]Shiji,2131–2137. Translated in William H. Nienhauser,Jr. Ed. The Grand Scribe’s Records. Memoirs o f Pre-Han China. Bloom ington and Indianapolis: Indiana University Press (1994): vol. 7,14-17. For Nienhauser’s analysis of the biographies of Guan Zhong and Yan Ying in the Shiji,see “The Implied Reader and Translation: The Shih Chi as Example,” in Eugene Eoyang and Lin Yao-fu eds. Translating Chinese Literature. Indiana University Press,Bloom ington,(1995): 15-40. See also Nienhauser’s “Sima Qian and the Shiji,” in Andrew Feldherr and Grant Hardy,The Oxford History of Historical Writing. New York,Ox ford University Press,vol.1,(2011): 472.
[35]Shiji,1477–1513. Translated in William H. Nienhauser,Jr. Ed. The Grand Scribe’s Records. The Hereditary Houses o f Pre-Han China,Part I. Bloom ington and Indianapolis: Indiana University Press (2006): vol. V.1,31-130.
[36]Shiji,2137.
[37]Shiji,2134. Except for a few changes,this translation of Yanzi’s biography is adapted from W illiam H. Nienhauser,Jr.,Memoirs,ibid. For the calculation of over a hundred years’ gap between Guan Zhong and Yanzi,see Memoirs,ibid.,p. 14,n. 39.
[38]Shiji,2134.
[39]For a discussion of the location of Yiwei in Lai—Donglai 东莱,see Memoirs,ibid.,p. 14. n. 40.
[40]Notably,this particular reference to Yanzi’s frugality is one of the rare cases that indicates a direct textual affiliation between the biography of Yanzi in the Shiji and the YZCQ. References to Yanzi’s lifelong “frugality” w ith one term “俭” appear in several Han texts; how ever,the Shiji uses here the com pound expression “节俭,”w hich appears in connection to Yanzi only in the YZCQ,(2.14/15/27; 4.14/34/27) and not in any other pre-Han or Han texts.
[41]Cf. Analects,14.3/37/10.
[42]Shiji,2134.
[43]Shiji,2135.
[44]YZCQ,Items 5.24 [134]; 5.25 [135],47/18-48/14. In the third part of the biography,however,Sima Qian specifically states that these two items were just “neglected stories” that were not selected from the text of The Spring and Autumn Annals of Master Yan that he read.
[45]Since the ancient Chinese foot,chi 尺,is said to be about 23 centimeters,according to the narrated episode in Yanzi’s biography in the Shiji and Item 5.25 [135],Yanzi’s height would be less than 138 centimeters and his eight-feet charioteer would be184 centimeters. According to Item 7.1 [171],however,Yanzi’s height was no more than five chi (115 centimeters).
[46]Shiji,2136.
[47]For a discussion of these five titles,some of which appear in the received text of Guanzi,see A llyn W. Rickett,tr. Guanzi: Political,Economic,and Philosophical Essays from Early China. Princeton Library of Asian Translations,Princeton (1985): vol. 1,p 6,n. 12-16.
[48]These “neglected stories” (轶事) are most probably mentioned here in reference to the two episodes previously recounted in the second part of Yanzi’s biography. They cannot be related to the major events narrated in Guan Zhong’s biography,which do not fall into the category of “neglected stories.”
[49]Later the editors of the Siku Quanshu would comment that like the YZCQ,some other texts,such as Lüshi chunqiu,were not arranged in strict chronological order by years and months and nevertheless bore in their title the metonymy Chunqiu 春秋.(“Spring and Autumn”).
[50]Notably,the present third part of the biography m ight be somewhat garbled. The title The Spring and Autumn Annals of Master Yan 晏子春秋,which appears at this point in the received text of the Shiji,as the sixth in a series of titles of Guan Zhong,w as not included in the text of the Shiji that Liu X iang quoted from in his Preface to the Guanzi. The latter reads: “His Honor the Grand Scribe said: ‘I have read through M r. Guan Zhong’s [1] “Shepherding the People,” [2] “Mountains on High,”[3] “Chariots and Horses,” [4] “Light and Heavy,” and [5] “Nine Bureaus.” His words are expressed w ith such detail.’” (SBCK,1B12-2A1). Indeed,The Spring and Autumn Annals of Master Yan 晏子春秋 was not included as the sixth title in the text of the Shiji that Liu X iang quoted from. In addition,the fact that Liu Xiang him self,about 75 years later,entitled his collated text of Yanzi “Yanzi 晏子” and not “Yanzi chunqiu 晏子春秋” m ight show that he was not aware of the latter title,although he was quite familiar w ith the biography of Yanzi in the Shiji,as his Preface to his Yanzi晏子 indicates. See,in the present Section (III-3). Since the next reference to Yanzi’s text as The Spring and Autumn Annals of Master Yan 晏子春秋 occurred about 300 years later,in Ying Shao’s 应劭 (d. ca. 204 CE) Fengsu tongyi 风俗通义 (9.2/65/19) and the Kongcongzi 孔丛子 (5.3/58/13),the title,The Spring and Autumn Annals of Master Yan 晏子春秋 in the received text of the Shiji may have been a later insertion. It is quite possible,on the other hand,that the quoted passage from Liu Xiang’s text of the Shiji,in his Preface to the Guanzi,was incomplete and that the parallel passage in the receive text of the Shiji is the text that genuinely records the comp lete set of the six texts that Sima Qian read,including The Spring and Autumn Annals of Master Yan 晏子春秋.
[51]Ibid.,2137.
[52]Cf. Analects,3.22/6/18.
[53]Cf. Item 5.2 [112]; Zuozhuan,B9.25/283/12; Shiji,1502. For a refined analysis of the episode that narrates the reaction of Yanzi to Cui Zhu’s assassination of Duke Zhuang,see Wai-yee Li,The Readability of the Past in Early Chinese Historiography. Harvard East Asian Monographs 253. Cambridge,Mass.,(2007): 322-326.
[54]Cf. ibid.,2.24/4/21.
[55]Cf. Zuozhuan,B7.12.5/176/7; Xiaojing 孝经,17/4/18.
[56]This adm iring statement,w hereby Sima Qian assumes the role of Yanzi’s driver,is made in reference to Yanzi’s chariot driver during the time before and after the assassination of Duke Zhuang. See,Items 5.2 [112]; 5.3 [113].
[57]See Items 5.2 [112]; 5.3 [113].
[58]See Liu Xiang Preface to the Yanzi text where Yanzi is said to be the equal of Guan Zhong. See also,Lunheng 论衡,82/357/12-13: “Guan Zhong and Yan Ying were both successful statesmen and w riters.”
[59]Yantie lun 盐铁论,5.4/ 32/8; 6.2/44/20; 7.1/50/20-23.
[60]Ibid.,5.5/32/32.
[61]For several excellent discussions and references to Liu Xiang’s bibliographical enterprises,see Anne Behnke Kinney,tr. Exemplary Women of Early China: The Lienü zhuan of Liu Xiang. New York: Columbia University Press: (2014): xvxvii; Bret Hinsch,“The Com position of Lienüzhuan: Was Liu X iang the Author or Editor?” Asia Major (third series) 20.1 (2007): 1-23; David R. Knechtges and Taiping Chang eds. Ancient and Early Medieval Chinese Literature. Leiden: Brill (2010): vol. 1,560-565; Haun Saussy,“The Unreliable Anthologies: Liu Xiang and Liu Xin in Historical perspective.” AAS Annual Meeting,Chicago,March 1997; Michael Loewe,Biographical Dictionary of the Qin,Former Han and Xin Periods (221 BC–AD 24). Handbuch der Orientalistik IV.16. Leiden: Brill (2000): 372-375; Piet Van Der Loon. “On the Transm ission of the Kuan-tzu.” T’oung Pao 41 (1952): 358-93; Deng Junjie 邓骏捷,Liu xiang xiao shu kao lun 刘向校书考论. Ren m in chu ban she,Beijing,2012.
[62]For an excellent discussion entitled “The ‘anti-Confucius’ of Qi: The Yanzi chunqiu and the Defense of the Qi Model,” see Andrew Meyer,“‘The A ltars of the Soil and Grain are C loser than K in’ 社稷戚于亲: The Qi 齐 M odel of Intellectual Participation and the Jixia 稷下 Patronage Community.” Early China 33-34 (2010-11): 59-66.
[63]She 社 w as the spirit of the five colors of soil; Ji 稷 was the spirit of the five major grains. The altars of these two spirits (Sheji 社稷) were located at the center of each capital of the various territorial domains in ancient China and were used as a common political symbol for the state or the nation. Regarding the concept of “loyalty to the altars,” see Yuri Pines,“Friends or Foes: Changing Concepts of Ruler-Minister Relations and the Notion of Loyalty in Pre-Imperial China.” Monumenta Serica 50 (2002): 35-74.
[64]Paraphrasing Charles Dickens’ opening of David Copperfield: “W hether I shall turn out to be the hero of my own life,or whether that station w ill be held by anyone else,these pages must show.” For different,more reserved analyses of the narrative license of the YZCQ,see Burton Watson,Early Chinese Literature. New York: Columbia University Press (1962): 185; Brian Moloughney,“History and Biography in M odern China” (PhD diss.,Australian National University1994): 12-13. See also Henri Maspero,La Chine antique. Presses Universitaires de France - PUF; Édition: Nouv. Éd,Paris (1985): 486-7.
[65]See Piet Van Der Loon,ibid.,360.
[66]Liu Xiang’s report on Yanzi is included as a “Preface” in many early and later editions of the YZCQ.
[67]SBCK,1A2-9.
[68]For a discussion of the identity of Fu Can,see Piet Van Der Loon,ibid.,p. 361,n. 2.
[69]For a detailed analysis of the interchangeability of these characters,see JS,51.
[70]Duke Zhuang was invested as the ruler of Qi by the powerful Cui Zhu 崔杼,who later murdered him for having an extra-marital affair with his wife. For a detailed description of Duke Zhuang’s assassination and for an exposition of Yanzi’s moral dilemma over serving the new ruler of Qi,Duke Jing 景公,who was also enthroned by Cui Zhu,see YZCQ,Item 5.3 [113]; Zuozhuan,B9.25.2/282.
[71]Zhong 钟,a measure of grain in the state of Qi,consisted of six-hundred and forty liters.
[72]SBCK,1B1-2B7.
[73]For a rare case in which Sima Qian borrows a term directly from one of Yanzi’s texts,see n. 42,above. Moreover,Sima Qian’s reference to Yanzi’s conduct in the events surrounding the assassination of Duke Zhuang is recorded in both the YZCQ the Zuozhuan. See above,n. 60.
[74]Liu Xiang biographical portrait of Yanzi derives the basic facts about the time and places in which Yanzi lived from the biography of Yanzi in the Shiji. However,all other personal references in this portrait,such as to Yanzi’s emolument of tenthousand zhong; to his support of more than five hundred fam ily members,relatives,and scholar-recluses; to his coarse clothing; and to his driving a w orn out nag and broken down chariot,do not appear in any other source and therefore are most probably drawn entirely from Liu Xiang’s collated text.
[75]See Zhang Shoujie’s 张守节 (fl. 725–735),Shiji zhengyi 史记正义,Shiji,2136. Years later,Liu Shipei 刘师培 (1884–1919),claimed that Zhang Shoujie was mistaken in replacing “eight” (八) pian w ith “seven” (七) pian,and in replacing the title of Ruan Xiaoxu’s (479–549) lost catalogue Qilu 七录 with Liu Xin’s Qilue 七略. See “Yanzi chunqiu pianmu kao” 晏子春秋篇目考 in his Zuoan ji 左盦集,VII2A; http://ctext.org/library.pl?if=gb& file=81813& page=4 (2006-2017). Liu Xin’s Qilue was based on the previous work of his father Liu Xiang—Bielu 别录 (Separate Records). Both the Bielu and the Qilue were lost at the end of the Tang dynasty.
[76]Hanshu,1724. Ban Gu’s Yiwenzhi 艺文志 (Treatise on Literature) preserved selected and abridged extracts from Liu Xing’s Qilue.
[77]Lunheng,82/357/12-13.
[78]Fengsu tongyi,9.2/65/19.
[79]For a discussion of the authenticity,date,and authorship of the Kongcongzi,see Yoav Ariel,K’ung-ts’ung-tzu: The K’ung Family Master’s Anthology. Princeton Library of Asian Translations. Princeton,(1989) 56-69.
[80]Kongcongzi,5.3/58/13.
[81]Ibid.,6.1/60/5-62/28.
[82]Wenxin diaolong 文心雕龙,SBCK,6A9.
[83]Shuijing zhu 水经注,4.1 河水; SBCK,35B3.
[84]A ll the similar determinations that w ill follow were arrived at by digitally comparing and locating parallels between the received text of the YZCQ and the encyclopedic compilations or anthologies under consideration in_http://ctext.org (2006–2017).
[85]O f these instances,the YZCQ is explicitly cited in Bei-tang shu-chao ten times. Some of the quoted item s in the Tang encyclopedias cite the YZCQ for their textual source,but many others are introduced w ith “晏子曰” (“Yanzi said”),without direct reference to the YZCQ as the specified source. A ll these quotations,however,appear in the received text of the YZCQ.
[86]Of these instances,the YZCQ is explicitly cited 11 times.
[87]The Qunshu zhiyao also contains in its section of the Shiji a full quotation from Yanzi’s biography,which refers to Yanzi’s text not as Yanzi but as YZCQ. See,http:// ctext.org/text.pl?node=278855& if=en [3] (2006-2017).
[88]See,http://ctext.org/text.pl?node=278901& if=en (2006-2017).
[89]Suishu,34,997. According to the Siku quanshu tiyao (YZCQ) 四库全书提要 (晏子春秋),2A6,the Suishu Jingji zhi listed the YZCQ as an “eight [pian]” text,not “seven [pian].”
[90]Jiu Tangshu,2023; Xin Tangshu,1509.
[91]Of these instances,the YZCQ is explicitly cited 9 times.
[92]See Yilin 意林,卷 1,1/5B1; http://ctext.org/library.pl?if=en&_file=33691&_page=32 (2006–2017).
[93]Liu Zongyuan’s list of the Mohist’s dom inant features,which he claims the YZCQ shares,is problematic. While the six “Inner Chapters” of the YZCQ bear notable Mohist features such as “Moderate Consumption” (节用) and an enthusiastic attitude in favor of friendly and humane interaction among people (an attitude Liu Zongyuan m ight have incorrectly,in my opinion,interpreted as the famous Mohist trait of “Im partial Love—兼爱”),his reference to the “condemnation of Confucius” (非孔子) and “condemnation of Confucianism” (非儒) as Mohist features are prevalent only in the eighth pian,the outer-heretical chapter of the received text. As for the Mohist trait of fondness for discussing the affairs of the ghosts (好言鬼事)—ghosts do not play any part in the entire text and there is not a single passage that speaks to “condemnation of music” (非乐). Yanzi indeed expresses in Item 1.6 [6] his revulsion of popular music,but only because he is piously inclined,as he him self proclaim s,only to the music of antiquity. Furthermore,in Item 2.22 [47],Yanzi pronounces his objection to the lavish burial (非厚葬) Duke Jing is about to give (见下页)(接上页) Liangqiu Ju; yet his objection is not to lavish burials,but rather to the lavish burial of a manipulative m inister such as Liangqiu Ju. Liu Zongyuan may have come to realize the possible weakness of his “Mohist” argument and therefore sealed the argument by pointing out a true textual fact that garners for the YZCQ a M ohist flavor. On tw o different occasions in the text,M ozi him self proclaim s consecutively his exaltation of Yanzi’s Way (道). See Item s 3.5 [55]; 5.5 [115].
[94]辩晏子春秋,in 增广注释音辩唐柳先生集,SBCK,http://ctex_t.org/library. pl?if=gb&_file=78042&_page=126 (2006-2017).
[95]This debate has spanned a thousand years. Scholars like Chao Gongwu 晁公武(1105–1180) and Zhang Xuecheng 章学诚 (1738–1801) adapted Liu Zongyuan’s “M ohist” view of the text (JS,605); but others,like Sun Xingyan 孙星衍 (1753–1818),rejected it completely. Much later,Zhang Chunyi 张纯一,in his preface to his commentary edition of the YZCQ of 1930,suggested that the text was made of m ixed material—60–70% Mohist,and 30–40% Confucian. See his Preface in Yanzi chunqiu jiaozhu 晏子春秋校註 in the Xinbian Zhuzi jicheng 新编诸子集成series,Vol. 6. Taibei,Shijie shuju 台北,世界书局 (1972): 1. http://ctext.org/library. pl?if=gb&_file=94142&_page=513 (2006–2017).
[96]All of the follow ing Song encyclopedia quotations appear in the received text of the YZCQ. However,as is the case w ith the Tang encyclopedias,some of the quoted items in the Song encyclopedias cite the YZCQ as their textual source,but many others are introduced w ith “晏子曰” (“Yanzi said”),w ithout direct reference to the YZCQ as the specified source.
[97]See,e.g.,Item 1.25/10/27 [25] in the received text,which appears in parallel and is referred to d irectly to the YZCQ in bo th the Tang antho logy Qunshu zhiyao,群书治要,卷三十三,晏子,杂上,and,almost 600 years later,in the Song encyclopedia Jizuan yuanhai 记纂渊海 (卷四十八). See http://ctex_t. org/tex_t.p_l?node=417989&_if=en [5] (2006-2017),and http://ctex_t.org/w_iki. pl?if=gb&_chapter=433368 [90] (2006-2017),respectively.
[98]For a discussion of the discovery of 102 Yanzi bamboo strips and fragments,excavated in 1972 at Yinqueshan,see,Section (VIII-2).
[99]See YZCQ,8.18a/75/3-7 [215]; Yinqueshan Han mu zhujian (yi),[释文; 注释],104-105; 群书治要,卷 三十三,晏子,杂下,http://ctext.org/text.pl?node=417998&_if=en [4] (2006-2017); 意林,卷一,晏子八卷,http://ctext.org/text.pl?node=566572&_if=en [16] (2006-2017); 太平御览,鳞介部七,鱼上,http://c_tex_t.o_rg/tex_t. pl?node=409716& if=en [44] (2006-2017); and Tian-zhong ji 天中记,卷二十七,http://ctext.org/w iki.pl?if=en& chapter=73792[6] (2006-2017). For a full discussion of the Yinqueshan Han bamboo strips 银雀山汉墓竹简,see,Section (VIII-2). The fact that all sixteen items of the Yinqueshan Han bamboo strips,from ca. 140 BCE,com prise 7.2% of the entire received text of YZCQ,reflects conclusively on the text’s authentic stable transm ission from the Han onwards up to the present. However,the fact that at least one of these items can be traced to its various transmitted Tang,Song and Ming stages throughout the ages gives the impression of this authentic transm ission of the YZCQ text more weight.
[100]See,Section (V).
[101]The Songshi Yiwenzhi 宋史艺文志 (Songshi,5171),compiled after the Song by Tuotuo 脱脱 (1313–1355),situates the YZCQ in the section devoted to Confucian w ritings also as a 12 juan text.
[102]An exception to this Song attitude can be found in the Song encyclopedia Cefu Yuangui 冊府元龟 (1013),w hich identifies “Yan Ying,the Prime M inster of Qi” as the author of the YZCQ. Nevertheless,the encyclopedia does not refer to the number of the pian or juan in the text,while almost all other entries for texts offer specified number of either pian or juan. This absence gives the impression that the compiler of the encyclopedia,Wang Qinruo 王钦若 (962–1025),may have been aware of the conflicting information regarding the number of pian or juan,and sought to avoid the problem altogether by ignoring it completely. See Cefu Yuangui,卷八百五十四,http://ctext.org/wiki.pl?if=en&_chapter=64829 [5] (2006-2017).
[103]Chongwen zongmu 崇文总目,http://ctext.org/wiki.pl?if=en&_chapter=415282 [72] (2006-2017).
[104]Junzhai dushu zhi 郡斋读书志,卷十一道家类法家类墨家类纵横家类 [95; 96]; http://ctext.org/wiki.pl?if=gb&_chapter=564340 [95; 96] (2006-2017). The entry’s inclusion of the directly referred long quotation from Liu Zongyuan’s “Mohist”essay pronounced some 300 years earlier testifies to the formative status of Liu Zongyuan’s essay in the history of the YZCQ text.
[105]Obviously,Chen Zhensun was not very familiar with the biography of Yanzi in the Shiji,in which Sima Qian explicitly states that he read a text entitled YZCQ. See,Section (III-2).
[106]Zhizhai shulu jieti,卷九,儒家类,http://ctext.org/wiki.pl?if=gb&_chapter=811528#儒家类 (2006-2017).
[107]H an Yiw enzhi kaozheng 汉艺文志考证,卷五,h ttp://c tex t.o rg/w ik i. pl?if=en&_chapter=556508 [5; 6] (2006-2017). See also,Wang Yinglin’s Yuhai,玉海,卷四十一. http://ctext.org/w_iki.pl?if=gb&_chapter=933116&_remap=gb [11; 12] (2006-2017). About 500 years later,Sun Xingyan 孙星衍 (1753–1818),in his introduction to his commentary edition of the YZCQ,argued that in the Song YZCQ text,each of the 7 juan was split into two,form ing a 14 juan text. He argued that the Yuhai reference to the YZCQ as a 12 juan text was an error and characterized Wang Yinglin’s explanation of the discrepancies in the number of juan of the text as “preposterous” (妄言). See,Yanzi Chunqiu 晏子春秋,Sibu beiyao,Shanghai,Zhonghua shuju (1989): Vol. 53,1A5; 1B15.
[108]See “Yanzi chunqiu pianmu kao” 晏子春秋篇目考 in his Zuoan ji 左盦集, V II2A-B; http://ctext.org/library.pl?if=gb&_file=81813&_page=4 (2006-2017).
[109]Siku quanshu tiyao (YZCQ) 四库全书提要 (晏子春秋),2A7-8.
[110]The Li fam ily M ianm iaoge w oodblock edition of the M ing 明李氏绵眇阁刻本,in Feng M engzhen’s 冯梦桢 (1546–1605) Xian Qin zhuzi hebian 先秦诸子合编,was said to be based on a Song woodblock edition,according to one of its two added postscripts (JS,43). In contrast,however,the catalogue of Ju Yong 瞿镛(1794–1846) records it as based on a Yuan woodblock edition. See,http://ctext.org/ library.pl?if=gb&_file=29296&_page=6 (2006-2017). Furthermore,Wu Zeyu 吴则虞(1913–1977),in his critical discussion of the Li Fam ily edition of the late M ing (JS,ibid.) points out that the fact that the last item of the received version (YZCQ,8.18 [215]) is m issing from this edition indicates that it could not have been produced in the Song. Notably,moreover,the Li Family edition of the late M ing was the edition upon which the editors of the Siku quanshu 四库全书 choose to base their edition,but their “Synopsis” refers to neither the Song nor the Yuan editions,upon which their choice of this late M ing edition w as based. See,http://ctext.org/library.pl?if=gb &_file=62378&_page=3(2006-2017).
[111]昭文张氏所藏元刻本.
[112]This Yuan woodblock edition is no longer extant. For a detailed description of it,see Wu Shouyang’s 吴寿旸 (1771–1835),Bai jing lou cangshu tiba ji 拜经楼藏书题跋,http://ctext.org/w iki.pl?if=gb& chapter=105945#晏子春秋 (2006-2017).
[113]仁和丁松生八千卷楼藏明活字本.
[114]For an encyclopedic list arranged in chronological order of extant and non longer extant editions,commentaries,and other major referential works regarding the YZCQ,see Yan Lingfeng 严灵峰,Zhou Qin Han Wei zhuzi zhijian shumu 周秦汉魏诸子知见书目,Taibei,正中书局 (1975–1977): vol. 3,1-29. For a sim ilar list and discussion,see Wang Gengsheng 王更生,Xinbian Yanzi chunqiu 新篇晏子春秋,Taibei,台湾古籍 (2001): 745-774. See also Lin X inxin’s 林欣心 discussion of major printed editions of the YZCQ in,Yanzi chunqiu yanjiu 晏子春秋研究 (MA Thesis,Sun Yat-Sen University,2000): 12-17.
[115]Stephen Durrant,however,has argued that this “biographical” classification gave rise to a new problem because the term “Biography” should not be applied strictly to a text such as the YZCQ. See his “Yen tzu ch’un ch’iu” in M ichael Loewe,ed. Early Chinese Texts: A Bibliographical Guide. Early China Special M onograph Series 2. Berkeley (1993): 487.
[116]The two Tang texts are Wei Zheng’s 魏征 Jianlu 谏录 and Li Jiang’s 李绛 (764–830)论事集 Lunshiji.
[117]The Synopsis refers here to Item 8.12 [209] in the received version. This item carries explicit homosexual implications and was therefore absent from several editions,including from the Siku quanshu version. How ever, not only did this “homosexual”item remained intact in the received version,but it was also anthologized in Mark Stevenson and Cuncun Wu,eds.,Homoeroticism in Imperial China: A Sourcebook. Routledge,New York (2013): 10.
[118]See,Lüshi chunqiu,2/4/10/10.
[119]See,Yanzi Chunqiu 晏子春秋,Sibu beiyao 四部备要,Beijing 北京,Zhonghua shuju 中华书局,(1989): Vol. 53,3-49.
[120]Ibid.,51A-88B. The Sibu beiyao edition also appends an additional set of notes to the YZCQ by Huang Yizhou 黄以周 (1828–99),see ibid.,91A-121A.
[121]Ibid.,1A-2A.
[122]These decisive anti-forgery arguments that Sun Xingyan incorporated throughout his Preface were made in response,inter alia,to Qing scholars such as Wu Dexuan 吴德旋 (1767–1840) and Guan Tong 管同 (1780–1831),who believed that the YZCQ w as a forgery of the Six Dynasties period (220–589 CE) or even later. See,JS,630-631.
[123]These state annals of Qi are mentioned in Mozi,8.3/52/14.
[124]Sun X ingyan’s use of such harsh rhetoric against Liu Zongyuan because of Liu’s “M ohist” theory,was later approved and reinforced by Liu Shipei (1884–1919). See his “Yanzi fei Mojia bian” 晏子非墨家辨,in his Zuoan ji 左盦集,VII1A; http:// ctext.org/library.pl?if=gb&_file=81813&_page=2 (2006-2017).
[125]See JS,632.
[126]See Section (III-2).
[127]See Section (III-1).
[128]See,Alexander Wylie,Notes on Chinese Literature. Shanghai: Presbyterian M ission Press,1867,rpt.,Taipei: Bookcase Shop Lim ited (1970): 34.
[129]See,Section (VIII-2).
[130]A lfred Forke,“Yen Ying,Staatsmann und Philosoph,und das Yen-tse tsch’untch’iu.”Asia Major (first series) Introductory Volume (Hirth Anniversary Volume) (1923): 101-144. Forke abridged this article in his Geschichte der Alten Chinesischen Philosophie. Hamburg: L. Friederichsen and Co. (1927),Rep.,Cram,De Gruyter & CO (1964): 82-92. For tw o critical review s of Forke’s essay,see Paul Pelliot,“Un nouveau périodique oriental: Asia Major,” T’oung Pao 22,(1923): 354-5; Richard L. Walker,“Some Notes on the Yen-tzu ch’un-ch’iu.” Journal of the American Oriental Society 73.3 (1953): 156-8.
[131]A year earlier,the great modern Chinese scholar and reformer Hu Shi 胡适 “rejected”(in his ow n w ords) the YZCQ from his book on the grounds of the text’s “doubtful authenticity.” See his The Development of the Logical Method in Ancient China. The Oriental Book Company,Shanghai (1922): “Preface,” 1.
[132]Henri Maspero,La Chine antique. Presses Universitaires de France - PUF; Édition: Nouv. Éd,Paris (1985): 486-7.
[133]Ibid.,486.
[134]George Kao,Chinese Wit & Humor. Coward-M cCann,Inc. New York (1946): 37-46. For further discussions of the YZCQ’s strong elements of humor,see Christoph Harbsmeier,“Humor in Ancient Chinese Philosophy.” Philosophy East and West,39.3 (1989): 295-6; David R. Knechtges,“Wit,Humor,and Satire in Early Chinese Literature (to A.D.220).” Monumenta Serica 29 (1970-71): 88.
[135]Richard Walker,“Some Notes on the Yen-tzu ch’un-ch’iu.” Journal of the American Oriental Society 73.3 (1953): 156-163. Walker’s PhD dissertation w hich he completed at Yale (1950) was entitled,“The Multi-State System of Ancient China.”It was published under the same title in 1953 by Shoe String Press,Hamden CT. For his view of Yanzi’s achievements as a minister in Qi,see ibid.,p. 69-71; 77-78.”
[136]Richard Walker,“Some Notes on the Yen-tzu ch’un-ch’iu.” Journal of the American Oriental Society 73.3 (1953): 159.
[137]Ibid.,p. 162.
[138]Burton Watson,Early Chinese Literature. New York: Columbia University Press (1962): 183-6.
[139]See Samuel Beckett,Waiting for Godot. Grove Press,New York (1982): 2.
[140]Yanzi chunqiu jiaozhu 晏子春秋校注 in the Xinbian Zhuzi jicheng 新编诸子集成series,Vol. 6. Taibei,Shijie shuju 台北,世界书局,1972. http://ctext.org/library. pl?if=gb& file=94142& page=513 (2006-2017).
[141]Wu Ze-yu’s 吴则虞,Yanzi chunqiu jishi 晏子春秋集释,2 vols. Beijing,Zhonghua shuju 中华书局,1962. A revised,extended edition w as published in 2011 under the title (Zeng ding ben) Yanzi chunqiu jishi (增订本) 晏子春秋集释. Beijing tushuguan chubanshe,北京图书馆出版社,2011.
[142]JS,1-523.
[143]JS,525-657.
[144]The “Preface” appears in JS,21-23,preceding p. 1 of the beginning of vol. 1 of the commentary.
[145]Wu Zeyu was very cautious not to attribute the authorship of the YZCQ directly to Chunyu Yue as the single person who compiled the text. He nevertheless pronounced his verdict on the authorship in such an ambiguous w ay (淳于越之类的齐人,在齐国编写的) that scholars interpreted Wu Zeyu’s position as saying that he considered Chunyu Yue to be the final compiler of the text. See e.g.,M asayuki Sato,The Confucian Quest for Order: The Origin and Formation of the Political Thought of Xun Zi. Sinica Leidensia 58. Leiden,Brill: (2003): 211,n. 79.
[146]The dissertation w as published later under the same title. See Wang Gengsheng Wang 王更生,in Yanzi chunqiu yanjiu 晏子春秋研究. Wenshi 文史,Taibei,1976.
[147]Wang Gengsheng 王更生,Yanzi chunqiu jinzhu jinyi 晏子春秋今注今译. Taiwan shang wu yin shu guan 台湾商务印书馆,1987; Xinbian Yanzi chunqiu 新编晏子春秋. Taiwan guji chuban 台湾古籍出版,Taibei 台北,2001.
[148]For the discovery of the Yinqueshan bamboo strips of Yanzi 银雀山竹简,晏子,see Wen wu 文物 (1974-6): 17-19,37; Pian Yuqian 骈宇骞,Yinqueshan zhujian Yanzi chunqiu jiaoshi 银雀山竹简 晏子春秋校释,Wenjuan,Taibei,(2000) 10-19. See also Yinqueshan Han mu zhujian (yi) 银雀山汉墓竹简 (壹). Wenw u chubanshe,Beijing,1985; Wu Jiulong 吴九龙,Yinqueshan Han jian shiwen 银雀山汉简释文. Wenwu chubanshe,Beijing 1985; “Background to the Excavation at Yinqueshan,”in D. C. Lau and Roger T. Ames,Sun Bin: The Art of Warfare. A Translation of the Classic Chinese Work of Philosophy and Strategy. SUNY Series in Chinese Philosophy and Culture. A lbany (2003) 187-196.
[149]For the numbers and distribution of the characters of the various texts w ritten on the Yinqueshan Han Bamboo Strips excavated from Tomb #1,see “A Com puterized Database of Excavated Wood/Bamboo and Silk Scripts of China (Jianbo),” CHANT (Chinese Ancient Texts Database),last modified 2005,http://www.chant.org/jianbo1/ stat.aspx.
[150]The terminus post quem of the Yinqueshan Han Bamboo Strips cannot be dated much earlier than approximately 200 BCE because all the characters on the strips from Tomb # 1 belong to an early period in the standardization of the clerical script; this script had become the standard style after the unification of Chinese states under the state of Qin in 221 BCE.
[151]For the list of all the 41,324 characters that appear in the received version of the YZCQ,see D. C. Lau and Chen Fong Ching,A Concordance to the Yanzi chunqiu. The Commercial Press,Hong Kong (1993): 427-432.
[152]For a reproduced copy the original 102 Yinqueshan Yanzi Bamboo strips,see Yinqueshan Han mu zhujian (yi) 银雀山汉墓竹简 (壹). Wenwu chubanshe,Beijing (1985): [摹本] 81-89.
[153]For a fully edited and annotated transcription of the clerical script text w ritten on the 102 Yinqueshan Yanzi Bamboo strips,see ibid.,Yinqueshan Han mu zhujian (yi) [释文] 87-106.
[154]For an account of these archeological findings,see Pian Yuqian 骈宇骞,ibid.,pp. 10-11.
[155]See,Section (III-1).
[156]Chunyu Kun w as an official and envoy,a famous rhetorician and a dwarf jester w ho,like Yanzi in his time,delivered remonstrations and persuasions to a drunken and debouched duke—in this case Duke Wei of Qi 齐威王 (c. 378–320 BCE ). For an excellent and detailed discussion of Chunyu Kun’s activities as a member of the Jixia patronage community,see Giulia Baccini,“The Forest of Laughs (Xiaolin): Mapping the Offspring of Self-aware Literature in Ancient China.” PhD Diss.,Universita Ca’Foscari,Venezia (2011): passim. See also,Oliver Weingarten. “Chunyu Kun: Motifs,Narratives,and Personas in Early Chinese Anecdotal Literature.” Journal of the Royal Asiatic Society 27.3 (2017): 501-521.
[157]JS,617.
[158]For the account in the Shiji of these three “Jesters” (滑稽),see Shiji,3197; 3200; 3202 respectively.
[159]Luo Jun substantiated his argument by referring his readers to the Hanshi waizhuan韩诗外传,10.17/76/15,which portrays Yanzi as the greatest master of rhetoric in the w orld (晏子,天下之辩士也). Cf. YZCQ,Item 6.10 [150],w here the K ing of Chu’s entourage portrays Yanzi as a master of the art of rhetoric (习辞者).
[160]E.g.,items 6.24 [164]; 6.30 [170]; 6.6 [146].
[161]Wang Gengsheng 王更生,Yanzi chunqiu yanjiu,72-3. Gao Heng 高亨 suggested even earlier (in 1961) the idea that both the YZCQ and the Guanzi 管子 were com piled by members of the Jixia patronage community in Qi. See Gao Heng’s “Yanzi chunqiu de xiezuo shidai” 晏子春秋的写作时代,in Wen shi shu lin 文史述林,Beijing,Zhong hua shu ju,北京: 中华书局 (1980): 397-398.
[162]Lü Bin 吕斌,“Chunyu Kun zhu Yanzi chunqiu kao 淳于髡著《晏子春秋》考.”Qilu Xuekan 齐鲁学刊,(1985; 1): 73-76.
[163]Shiji,2347.
[164]Lü supported his argument regarding the identical dim inutive body image of Chunyu Kun and Yanzi by pointing at a possible interchangeability betw een Item 6.9 [149] in the YZCQ and Xinxu 新序,11.24/65/8. In the YZCQ,Yanzi is sent on a diplomatic mission to Chu and because of his diminutive stature the people of Chu unsuccessfully attempt to humiliate him. A somewhat parallel episode is found in the Xinxu; however,this time the short envoy is Chunyu Kun,who arrives at Chu from Qi and manages to deal w ith the hum iliation the King of Chu rains dow n upon him for his short height.
[165]Tao Meisheng 陶梅生,Xinyi Yanzi chunqiu 新译晏子春秋. Taibei Shi,San min shu ju gu fen you xian gong si 台北市,三民书局股份有限公司 (1998): 30.
[166]Lin X inxin 林心欣,Yanzi chunqiu yanjiu 晏子春秋研究. National Sun Yat-Sen University,MA Chinese Literature (2000): 41.
[167]Zhao Kuifu 赵逵夫,“Yanzi chunqiu wei qiren Chunyu Kun biancheng kao 晏子春秋为齐人淳于髡编成考.”Guangming ribao 光明日报 (2005; 1; 28).
[168]Andrew Meyer,ibid.,66. As noted,already in 1961 Gao Heng claimed that both the Guanzi and the YZCQ were the product of the Jixia patronage community in Qi.
[169]E.g. Masayuki Sato,ibid.,p. 211; Fu Junlian 伏俊琏,“Chunyu Kun jiqi lunbian ti za fu 淳于髡及其论辩体杂赋.” (“Chunyu Kun and His Controversial Mixed Fu.”). Qilu xuekan 齐鲁学刊,(2010; 2) 105-108; Giulia Baccini,ibid.,87,n. 379; https:// en.w_ikipedia.org/wiki/Chunyu_Kun (Last modified on 24 May 2015).
[170]In addition to the aforementioned contributions made by Yuri Pines,Wai-yee Li,and Andrew M eyer,see also M asayuki Sato’s “The Concept of Li in the Yanzi Chunqiu”in his The Confucian Quest for Order: The Origin and Formation of the Political Thought of Xun Zi. Sinica Leidensia 58. Leiden,Brill,(2003): 211-217; David Schaberg,“Platitude and Persona: Junzi Comments in Zuozhuan and Beyond,” in Helw ig Schm idt-Glintzer et al.,eds. Historical Truth,Historical Criticism,and Ideology: Chinese Historiography and Historical Culture from a New Comparative Perspective. Leiden Studies in Comparative Historiography 1. Leiden and Boston: Brill (2005): 188-192; R. Smart,“How Yanzi Fulfills his Responsibilities as M inister in the Rhetorical Techniques within the jian (Remonstrance) of the Yanzi chun qiu.”MA Thesis,University of Canterbury,2008; Scott Cook,“The Changing Role of the Minister in the Warring States: Evidence from the Yanzi chunqiu 晏子春秋,” in Yuri Pines,et al.,eds. Ideology of Power and Power of Ideology in Early China. Sinica Leidensia 124. Leiden and Boston,Brill (2015): 181-2010.
[171]Rainer Holzer,Yen-tzu und das Yen-tzu ch’un-ch’iu. Würzburger Sino-Japonica 10. Frankfurt and Berne,Peter Lang,1983.
[172]Stephen Durrant,“Yen tzu ch’un ch’iu,” In M ichael Loewe,ed. Early Chinese Texts: A Bibliographical Guide. Early China Special Monograph Series 2. Berkeley (1993): 483-489. See also Shih Hsiang-lin [SHL],“Yanzi chunqiu,” in David Knechtges and Taiping Chang eds. Ancient and Early Medieval Chinese Literature: A Reference Guide. Leiden [Netherlands],Brill,Boston (2010-2014): 1868-1873.
[173]For a detailed list of all of these item s,see the recently published massive reference guide on the history of research on the YZCQ by Liu Wenbin 刘文斌,Yanzichunqiu yanjiu shi 晏子春秋研究史. Renm in wenxue chubanshe 人民文学出版社,Beijing (2015): 228-247.
[174]For an example,some traditional sources quote Chunyu Kun as having authored a text entitled Wangduji 王度记. The text is long lost,surviving in a m iniscule number of passages. A conceivable future archeological discovery of the text may finally settle the case of the true authorial connection between Chunyu Kun and the YZCQ.
[175]For the sake of comparison,title “孟子” yields at present about 1,880,000 hits in Google search.
[176]I have engaged in such internet-based database analysis in the present study by conducting queries on the combined database of Tang-Song-M ing encyclopedias,and the Yanzi Text in the Yinqueshan Han Bamboo Strips,in order to establish a line of stable transm ission of the YZCQ text that stretches from the early Han through the Tang,Song,until the end of the M ing. See above p. 29,n.87; p. 36.
[177]See Melissa Healy,“Study finds a disputed Shakespeare play bears the master’s mark.” The LA Times,April 10,2015.