Coding all clauses in L2 data: A call for consistency

Abstract

Linguistic complexity is considered a hallmark of language proficiency (Norris & Ortega, 2009), and it is often a focus of second language (L2) research (Pallotti, 2015). Within the construct of complexity, syntactic complexity considers the combination of words and clauses. The term clause, however, is underspecified in the field, and L2 researchers only occasionally define how clause is operationalized in the coding. After a review of the importance of consistent clause coding, this commentary argues that coordinated verb phrases, nonfinite clauses, and verbal small clauses can reach the complexity of finite clauses by the production of another element, making these constructions commensurate with subordinate finite clauses. Since these constructions are structurally similar to accepted clauses (e.g., imperatives), their exclusion from the clausal complexity calculation threatens the validity and reliability of the coding. Because a narrow definition requires an arbitrary division along a continuum of complex verbal constructions, the field should adopt this broader definition of clause. Further, researchers should clarify their methodology choices and acknowledge coding challenges, which are common in L2 data, especially L2 speech data. The increased transparency in research methodology supports the interpretation of research results and facilitates comparisons across L2 studies.