Formulaic sequences with ideational functions in L1 student and expert academic writing in English

Corpus studies have revealed that formulaic sequences are prevalent in academic discourse in English. The predominant trend in this research area is to take a frequency-based approach (e.g., lexical bundles, n-grams), relying on the computer to retrieve continuous word sequences that occur frequently in a given corpus. Such an approach has helped bring to light a rich repertoire of FSs with textual or interpersonal functions (e.g., on the other hand, it is possible to) that characterises successful academic writing. However, the use of formulaic language that is central to the construction of disciplinary knowledge has received relatively little attention partly due to the limitations of the identification method. Through manual identification and annotation of FSs in context, the present study examines successful L1 student and expert writing. The results reveal that both are highly formulaic in quantitative terms, and ideational FSs account for approximately 70% of all FSs identified. However, each has its own distinct features in terms of the variety of FSs used. In general, the student corpus employs more everyday FSs which are often highly idiomatic, whereas the expert counterpart yields more FSs associated with research and reasoning processes. It is also argued that knowledge of conventional usage patterns for what seem to semantically transparent and syntactically flexible FSs in academic discourse is not necessarily an inherent part of native speakers’ linguistic competence, but needs to be acquired incrementally through formal instruction and training by non-native and native students alike.