The difference between ‘language’ and ‘dialect’ is known not to be straightforward, yet it is commonly assumed that linguistic structure plays a direct role in the distinction. This chapter presents the case that the notion ‘a language’ is a matter of social construction, with linguistic structure playing only an indirect role. Beginning with Kloss’s abstand and ausbau languages, several cases of the social construction of languages with little abstand are examined in detail. Ebonics is scrutinized as a possible case of a language consisting of ‘abstand dialects.’ It is argued that the notion ‘a language’ cannot be established as a natural category. If so, there is no way linguistic science can determine that African American Language is either a language or a dialect. If so, there may be rhetorical reasons to speak of it as a language, if we want to convince nonlinguists that AAL is a respectable linguistic system. John R. Rickford has always been a staunch defender of African American Language 1 as a legitimate linguistic system. I routinely assign ‘A Suite for Ebony and Phonics’ (Rickford 1997a), and his book, Spoken Soul: The Story of Black English (2000a), which he coauthored with Russell J. Rickford, when I teach a course that includes AAL as a topic. He has contributed mightily to the scientific analysis of syntactic features of AAL over his entire career (e.g., Rickford 1975; Rickford et al. 1996b; Rickford and Rafal 1996f). On my list, he is one of the two or three best linguists in the world at this. John has not fully addressed the degree of separation between AAL and English, although he has discussed the issue (e.g., Fasold et al. 1987; Rickford 2006e). 2 Nevertheless, I suspect he will have a substantial interest in whether or not AAL should be constructed as a language, not a dialect of English, and what various implicit social constructions of the system reveal about their advocates’ opinion about speakers of AAL, which I am going to address here.