從「放鄭聲」看孔子刪詩説的可靠性

The Reliability of Confucius’ Deletion of Poems in the Book of Song As Shown from His "Deletion of the Music of Zheng

Abstract

LANGUAGE NOTE | Document text in Chinese; abstract also in English.

司馬遷《史記•孔子世家》記載了古代《詩經》有詩三千餘篇，孔子去其重復，為三百零五篇。歷代學者對此有贊成的，也有反對的，到現在還未成定讞，成為了兩千多年來的學術公案。本文嘗試從孔子自己説的「放鄭聲」、「鄭聲淫」這個標準來看孔子删詩説的可靠性。文中首先提出古代聲（樂）、詩不分的現象，這在《荀子》、《墨子》、《儀禮》、《禮記》及後人顧頡剛的研究中都有説明。由於聲、詩不分，孔子既説「放鄭聲」，則鄭詩也應在被放之列，因為我們不能想像只有歌詞（詩），沒有音樂（聲）的「樂歌」。但現在我們看毛詩，鄭詩歷歷在目，則孔子何嘗删過詩？至於逸詩問題，本文作者認為可能在《詩經》四家之外，尚有其他版本家派，至今失傳，只留下一些佚文佚句，為今本毛詩所無，而由先秦學者引用，得以留存。1977年安徽阜陽出土的漢簡《詩經》，或者可以證明這點。阜詩與四家詩都不同，不禁使人懷疑它是否四家詩以外的另一版本。假如是的話，則逸詩的問題自可迎刃而解。而最主要的，阜陽詩經中收錄有鄭風八首，可見不論目前所見任何版本流派，都收錄了鄭風。由此，本文作者認為孔子删詩説是不可靠的，孔子其實並未删過詩。

The Shi Ji (Historical Records) by Sima Qian of the Han Dynasty recorded that in ancient times, there were more than 3000 poems comprising the Shi Jing (Book of Songs). Later, according to tradition, Confucius edited the classic and reduced the number of poems to 305. Ever since Sima Qian’s record, there have been scholarly disputes supporting or rejecting the question of whether or not Confucius’ deletion of poems really took place. This dispute has been on-going for nearly 2000 years and has not been resolved. The present paper attempts to solve this problem by examining Confucius’ criteria and attitude in dealing with the Book of Songs. Confucius had once said that he would like to delete the music from the State of Zheng, because the music of that State was yin 淫 (salacious or exceeded the limit). This incident was first recorded in the Analects. The present paper suggests that in the first place music and lyrics from ancient times in the Book of Songs were inseparable. This point was mentioned in Xunzi, Mozi, Yi Li, and the Li Ji etc., and studies by Gu Jie-gang also supported this view. Since it was supposed that Confucius deleted the music of Zheng, it was unavoidable that the words (poems) that accompanied it would also be deleted, as the two were closely intertwined. Moreover, within that tradition, it was unthinkable to talk about a “song”, and have only words and no music. It thus can be argued that Confucius actually had never deleted these poems from the Book of Songs, as the Zheng poems still exist in the Mao edition of the classic.

As to the question that other poems do exist in pre-Qin writings and cannot be found in the Mao edition, which scholars term yi-shi (lost poems), the present paper suggests that there may have been other editions of the Book of Songs in ancient times which are now mostly lost. Thus those poems that appear in pre-Qin writings actually come from other lost editions. The 1977 excavation of a Han Dynasty bamboo copy of the Shi Jing from Fu Yang, Anhui proves this point. The Fu Yang Shi Jing is different from the Si-Jia-Shi (the exiting four editions of the Book of Songs), and therefore it may have come from a different edition that was lost. If this is true, this is evidence to support the hypothesis of this paper as to why yi-shi were found in pre-Qin writings. Most importantly is that the Fu Yang Shi Jing recorded eight poems from the State of Zheng. Since all editions of the Shi Jing that exist today contain poems from Zheng, the present paper concludes that Confucius had never deleted poems from the Book of Songs. Therefore, Sima Qian’s statement in the Shi Ji is unreliable.