Ancient Greek thought on prosody: through time and Latin eyes

拉丁作家写的拉丁单词重音听起来往往像希腊作家写的希腊口音。例如，拉丁作家告诉我们，如果一个拉丁重音音节在单词的倒数第二个音节，并且有一个长元音，只要最后一个音节的元音短，那么这个音节就有一个“旋音”。在希腊语中也是如此。现代学者要么(i)评论说，拉丁语法学家在重音学问题上模仿希腊语法学家，好像他们没有注意到拉丁语和希腊语是不同的语言，要么（ii）争辩说，拉丁语的口音确实与希腊语非常相似，有尖音和绕音，分布规则也非常像希腊语。但这两种观点都忽略了重要的一点。当我们的拉丁语法学家在写作的时候，在希腊语中听不到尖锐和回旋重音的区别。因此，如果我们怀疑某些拉丁单词具有无意义的弯曲，那么，至少在拉丁语法学家遇到这种说法的时候，它几乎不比某些希腊单词在任何可听意义上具有弯曲的说法更有道理。此外，在希腊语中，长时间以来，回旋和尖锐的口音听起来彼此不再不同，希腊语法学家继续在他们对语言的描述中使用这种区别。在其他概念中，长元音和短元音的区别也是如此。这些特征不仅仅是过去的遗物，而是在希腊语法思想中获得了不同的地位。它们变成了抽象的特征——换句话说，是帮助语言整体描述的技术手段。例如，一些元音在理论上是长而另一些在理论上是短的，这一概念使得继续非常简洁地描述希腊口音的可能位置成为可能（重音可能落在最后三个音节中的任何一个，除非最后一个包含“长元音”）。从这个角度来看，当拉丁语法学家从希腊语法学家那里接受一个概念，比如锐音和旋音的区别时，他们是在把这个概念直接带入他们自己的描述系统的抽象层面。对于拉丁语来说，这个概念作为一种技术手段几乎没有实际用途，但它的抽象性质使它很容易被接受，而不会对实际发音产生任何影响。相反，当拉丁语法学家所陈述的内容在希腊语中是完全可以听清的（例如重音所在的音节）时，这些应该被严肃地视为关于拉丁语发音的陈述。这里的一个例子是拉丁语法学家的学说，即后缀-que， -ve， -ne和-ce使重音在前一个音节上，而不管该音节的重量。现代学者经常怀疑这是否总是正确的，并怀疑希腊语法学家在希腊语中写类似的序列时受到了不当的影响。但是，关于哪个音节重音的陈述不能被解释为只属于抽象层面：它们应该被认真对待，就像关于拉丁语发音的陈述一样。拟议的项目将收集和展示希腊语法传统中关于重音和元音长度的抽象思维的证据（两者都属于古老的“韵律”学科），并将提出希腊语法描述的抽象特征——包括一些随着语言变化而变得抽象的特征——作为抽象特征被拉丁语法思维所接受的假设。该项目旨在鼓励在世界其他语法传统中对类似问题进行更多的研究。此外，它的目的是丰富学校的语言教学和学习，通过提请注意历史和现代的技术来描述一种语言从另一种语言引进的类别，从而使外语不那么陌生。

Latin authors writing about the Latin word accent often sound like Greek authors writing about the Greek accent. For example, Latin authors tell us that a Latin accented syllable has a 'circumflex' if it is the next-to-last syllable in the word and has a long vowel, provided that the vowel of the last syllable is short. Exactly the same happens to be true in Greek.Modern scholars either (i) comment that Latin grammarians 'ape' Greek grammarians on accentological matters, as if they fail to notice that Latin and Greek are different languages, or (ii) argue that Latin really had an accent remarkably similar to that of Greek, with acutes and circumflexes and remarkably Greek-like rules for their distribution.But both views miss an important point. By the time our Latin grammarians are writing, the distinction between acute and circumflex accent is very unlikely to have been audible in Greek. Thus if we suspect the claim that certain Latin words had a circumflex of being nonsensical, it is hardly more so than the claim that certain Greek words had a circumflex in any audible sense, at least at the time when the Latin grammarians encountered this claim.Moreover, long after the circumflex and acute accents had ceased to sound different from each other in Greek, Greek grammarians continued to make use of this distinction in their descriptions of the language. The same is true for the distinction between long and short vowels, among other concepts. Rather than being merely relics of the past, these features had acquired a different status in Greek grammatical thought. They had become abstract features--in other words technical devices that helped in the overall description of the language. For example, the concept that some vowels were theoretically long while some were theoretically short made it feasible to continue describing the possible positions for the Greek accent very succinctly (the accent may fall on any of the last three syllables, unless the last contains a 'long vowel').In this light, when Latin grammarians take over from Greek grammarians a concept such as the distinction between acute and circumflex, they are taking the concept directly into an abstract level of their own descriptive system. For Latin the concept is of little practical use as a technical device, but its abstract nature makes it easy to take over without any consequences for actual pronunciation. Conversely, when Latin grammarians make statements about matters whose Greek counterparts would have been perfectly audible (for example the actual syllable on which the accent falls), these should be taken seriously as statements about the sound of Latin.An example here is the Latin grammarians' doctrine that the enclitics -que, -ve, -ne and -ce cause the accent to go on the preceding syllable, regardless of the weight of that syllable. Modern scholars have frequently doubted that this can always be true, and have suspected undue influence from Greek grammarians writing about similar sequences in Greek. But statements about which syllable is accented are not interpretable as belonging only to an abstract level: they should be taken seriously as statements about the sound of Latin.The proposed project will gather and present evidence for abstract thinking in the Greek grammatical tradition on accentuation and vowel lengths (both coming under the ancient discipline of 'Prosody'), and will develop the hypothesis that abstract features of Greek grammatical description--including some that became abstract as the language changed--were taken over into Latin grammatical thinking as abstract features.The project aims to encourage more work on similar questions in other grammatical traditions of the world. Furthermore, it aims to enrich language teaching and learning in schools by drawing attention to historical and modern techniques for describing one language in categories imported from another, thus making a foreign language less alien.