许家金--公开课件.pptVIP

In Search of a Measure of Grammaticality: A Probabilistic Approach Liang Maocheng In Search of a Measure of Orality: A Probabilistic Approach Xu Jiajin Register Variation: A SP-WR case Xu Jiajin Beijing Foreign Studies University Written language bias For decades Have we gone astray in linguistic inquiry? In the least, equal attention should be given to spoken language. What are the defining features of spoken language? Written/literate features Long, embedded complete sentences Relative clauses Long, nominalised abstract words Passive voice etc Spoken features Incomplete sentences, ellipsis Questions, interrogative sentences Tag questions Inversion, post-posed elements Contractions Small, easy everyday words Interactive, interpersonal, expressive Hedges, discourse marker use Fillers, pause markers Negation (Yes). No. Why do they differ? According to Wallace Chafe (1982), two factors explain the differences between written discourse and verbal interaction: 1) Writing takes longer than speaking; 2) Writers do not contact with readers. The first factor is responsible for the very many fragments in SD; and the second factor is responsible for the detachment from the readers as opposed to the high involvement in verbal communication. Distinctive features b/t speech writing Think about typical written discourse and typical spoken discourse Academic writing (research articles) Everyday casual conversation (small talk) Paper reading and replicate Writer/reader visibility in EFL written discourse Stephanie Petch-Tyson Spoken features of learner writing Writer reader visibility Petch-Tyson（1998） Questions for reading Research aims Data Method Conclusion Critique Biber’s MF/MD model 67 features (Biber 1988) 125 features (Biber 2007) 141 features (Xiao 2009) Features collected in literature and according to intuition List of spoken features will never be exhausted. BFSU_MF_MD tool Two overarching approaches Rule-based: Sth that we are sure about Pro and