VO 型語言中的名前關係子句

Prenominal Relative Clauses in Verb-Object Languages

VO 語序與名前關係子句的組合是相當少見的；Dryer (2005e) 在他的七百五十六個語言的樣本中只找到五個語言；其中三個是漢語系語言，一個是漢藏語（可能反映有共同的來源或與漢語有所接觸），第五個則是台灣南島語阿美語 (Amis)。雖然我們可以把阿美語的現象歸因於和漢語的接觸，但是深入的調查分析並不支持這個假設；因為在一些其他的台灣南島語中發現的相同現象與漢語的影響程度並無相關，而可能是南島語系中古老的現象。漢語系語言和阿美語間的相似性可能只是偶然（不過也不能完全排除兩者在遠古時代有親屬關係或是彼此有密切接觸之可能性）。有些學者認為VO 語序與名前關係子句之所以不容易組合在一起是因為這會使語句處理相當困難。不過，雖然罕見，這種組合在某些語言（如漢語系語言）似乎非常穩定，最近的兒童語言習得研究也發現這種組合其實有一些處理上的優勢。

The combination of verb-object order with prenominal relative clauses is extremely rare; Dryer (2005e) finds it in only 5 languages in his sample of 756; of these, 3 are Sinitic, one is Sino-Tibetan (probably reflecting either shared ancestry or contact with Chinese), and the fifth is the Formosan language Amis. While it is tempting to suggest contact with Chinese as the explanation for this pattern in Amis, further investigation goes against this hypothesis, since the same pattern is found in some other Formosan languages, does not correlate with degree of Chinese influence, and may be old in the Austronesian family. This similarity between Sinitic and Amis may be due to chance, though it is also conceivable that it could reflect genealogical or contact relations in the more distant past. The rarity of the combination of verb-object order with prenominal relative clauses has led some linguists to suggest a processing explanation that would mitigate against this combination. However, although rare, the combination seems stable in at least some of the languages where it occurs (e.g. Sinitic), and recent work in child language acquisition suggests that the combination may actually provide a processing advantage to the child acquiring the grammatical relations in a relative clause relativizing an object.