With massively parallel corpora of hundreds or thousands of translations of the same text, it is possible to automatically perform typological studies of language structure using very large language samples. We investigate the domain of word order using multilingual word alignment and high-precision annotation transfer in a corpus with 1144 translations in 986 languages of the New Testament. Re...

In this work, we study the impact of the word order decoding direction for statistical machine translation (SMT). Both phrase-based and hierarchical phrasebased SMT systems are investigated by reversing the word order of the source and/or target language and comparing the translation results with the normal direction. Analysis are done on several components such as alignment model, language mod...

arc also 'out of position', since you would normally expect VP-modi(ying PPs of this kind to appear immediately to the right of the modified VP (so that the canonical versions of these sentences would have been '1 met Ar thur in the park' and 'I believed that site loved me with all my heart. ') . There arc various reasons for moving things around in this waymoving 'who' to the left; in (1) prov...

This paper advances a general account of word order in resultatives. My conclusions derive from consideration of several languages, both VO and OV. Featured from the first group are English, Igbo, Edo, Ambae, Paamese, Mandarin, Shanghainese, Thai, and Vietnamese; included in the second are German, Ijọ, Japanese, Malayalam, Kannada, Mizo, and Yi. Resultatives are single clause constructions comp...