翻译不是把现代塞回古代Translation Is Not Smuggling Modernity Into Antiquity
这封读者来信值得认真对待的地方,不是它替吉尔伽美什史诗辩护,而是它点出了一件事:翻译本身就是一次解释权的争夺。Philip Terry 在评论 Armitage 的译本时用了 sex worker 这个词,Simon Dawson 反驳说,沙姆哈特(Shamhat)在美索不达米亚文化里是伊什塔尔女神的女祭司,性在那里可以是通向超越与转化的神圣行为。Stephen Mitchell 说得更直白:阿卡德语里的那个词不是我们意义上的“妓女”。
问题不在于 sex worker 这个词今天是否体面。问题在于,把一个有名字、有神职、有仪式功能的人,压缩成一个现代职业标签,等于宣布我们这套范畴是普适的、可以随身携带到任何时代的。这是第三章那个机制:夺取话语权,夺取解释权,夺取“什么是事实”的制造权。翻译者不必恶意,他只需要相信自己的词汇表足够用。
Dawson 还提到一件事:基督教圣经习惯把希伯来语的 qedeshim(“被奉献者”)译成“鸡奸者”或“男性神庙娼妓”。这条线索比它看上去更重要。一个词被反复译错,几百年下来,错的那个版本就成了原文的“原意”,后来的人再读,读到的是翻译者的道德秩序,不是美索不达米亚的。文化暴力最经济的地方就在这儿:它不需要审查,只需要一个大家都觉得顺手的词。
第二封信从另一个方向切进来。Fiona Collins 说,Armitage 的这本书不该叫 translation,除非他自学了楔形文字——她自己试过,做不到。它应该叫“从某个译本改编”。而那个真正懂苏美尔语或阿卡德语的译者,应该得到同等的承认。这一条我完全同意,而且它和第一封信是同一件事的两面:署名和命名都是定价权。谁被写成“译者”,谁被写成“素材来源”,决定了读者以为意义是从谁那里发出来的。
但我要指出这封信里没有出现的东西。Dawson 引用了 Stephen Mitchell 的导言,Collins 提到了一位“技能巨大”的译者,两封信都没有写出这位译者的名字。一篇讨论“谁该得到承认”的来信,自己漏掉了那个该被承认的人。这不是刻薄,这是材料:解释权的垄断往往不靠压制,靠的是省略,省略到连抗议省略的人也跟着省略。
至于“same-sex physical love”这个说法,Dawson 的反驳是对的但还不够狠。吉尔伽美什蒙上亡友的脸“像新娘一样”,然后走进荒野,这份悲痛把他推向求不死。把它读成性,是把一段关于丧失与必死性的叙事,降级成一个身份标签。用现代范畴去认领古代,看起来是宽容,实际是把对方收编成自己的注脚。
所以这两封信合起来讲了一件事:古代文本从来不是被“发现”的,是被重新命名的。每一次命名都在分配谁有资格说话。这一次,两位读者把命名的手往回按了一下——但那个译者的名字,还是没有人写出来。
What deserves attention in these letters is not that they defend the Gilgamesh epic. It is that they name something translation usually hides: every translation is a fight over who gets to interpret. Philip Terry, reviewing Armitage's version, used the words sex worker. Simon Dawson objects that Shamhat is a priestess of Ishtar, that in Mesopotamian culture the sexual could be a sacred act leading to transcendence and transformation. Stephen Mitchell says it plainly: the Akkadian word does not mean "prostitute" in our sense.
The issue is not whether sex worker is a respectable term today. The issue is that a named woman with a temple role and a ritual function gets compressed into a modern occupational label, and the compression announces that our categories are universal, portable into any century. This is the mechanism from chapter three: seize the discourse, seize the interpretation, seize the power to manufacture what counts as fact. The translator need not be malicious. He only needs to trust his own vocabulary.
Dawson adds a detail that matters more than it looks. Christian Bibles routinely render qedeshim, "consecrated ones" in Hebrew, as "sodomites" or "male cult prostitutes." A word mistranslated for centuries becomes the text's "original meaning" to everyone downstream, and what they read is the translator's moral order, not Mesopotamia's. This is where cultural violence is cheapest: it needs no censorship, only a word that feels convenient.
The second letter cuts from another side. Fiona Collins argues the book should not be called a translation unless Armitage taught himself cuneiform, which she tried and found beyond her. It is an adaptation from a translation, and the translator, whose skill with a language as complex and forgotten as Sumerian or Akkadian is enormous, should get equal recognition. I agree, and it is the same argument as the first: bylines and labels are pricing power. Who is written as "translator" and who as "source material" decides where readers believe the meaning comes from.
But look at what is missing. Dawson quotes Mitchell's introduction; Collins refers to a translator of huge skill. Neither letter names that translator. A letter demanding recognition for someone omits the someone. This is not a cheap shot, it is the material: monopoly over interpretation usually works not by suppression but by omission, so effective that even the people protesting the omission omit too.
On "a celebration of same-sex physical love," Dawson's correction is right but too gentle. Gilgamesh covers his dead friend's face "like a bride" and wanders the wilderness, and the grief drives him to seek immortality. Reading that as sex downgrades a poem about loss and mortality into an identity label. Claiming antiquity with modern categories looks like generosity; it is annexation.
Both letters together say one thing: ancient texts are never simply discovered, they are renamed. Every renaming allocates who is allowed to speak. This time two readers pushed the naming hand back a little. The translator's name still did not appear.