The recent "TESOL Quarterly" article by Biber, Gray, and Poonpon (2011) raises important considerations with respect to the use of syntactic complexity (SC) measures in second language (L2) studies. The article draws the field's attention to one particular measure--complexity of noun phrases (NP) (i.e., noun phrases with modifiers, such as premodifying adjectives and nouns and postmodifying prepositional phrases)--to which many L2 researchers have not paid due attention. On the basis of their findings, the authors critique the T-unit-based SC measures that have been commonly used in L2 writing studies, specifically mean length of T-unit (MLTU) and clauses per T-unit (C/TU), and urge the use of measures for NP complexity in writing studies. Their article seems to claim that because of the register differences observed in the SC features, different SC measures should be used in studying L2 developmental samples in the two registers--conversation and formal writing, as the title of the article indicates. However, what troubles the author and maybe other readers is that the study reported in the article is not itself a developmental study, but rather a corpus-based discourse study of SC features of two different registers based on language samples from "proficient" speakers and writers. The question is, then, how valid is it to make claims about developmental trends in SC based on findings of register differences found in proficient users' language production?