No mixed grammars, no phonological disjunction: A new perspective on intra-sentential code-switching

•MacSwan's minimalist model of CS violates the fundamental tenets of the MP.•MacSwan's conception of CS as union of two Gs is minimalist equivalent of the EC.•Phonological Form Disjunction Theorem is minimalist version of Poplack's FMC.•CS should be considered 'mixing' of language-specific Ls (not Gs) through the CHL.•Each mixed sentence, like an unmixed one, is an expression of one and only one G.

The study aims to challenge MacSwan's (2000, 2005, 2010) Minimalist model of intra-sentential CS on theoretical grounds. It argues that his conception of CS as a 'union' of, at least, two lexically-encoded grammars (Gs) constrained by the requirements of 'mixed Gs' is a Minimalist version of the Equivalence Constraint. It argues that the logical consequence of employing the Minimalist Program as theoretical framework is to view CS, instead, as 'mixing' of, at least, two language-specific halves (Lexicons) through the language-independent Computational System of Human Language (CHL) to produce a 'well-formed' grammatical structure which is externally counted as an expression of one and only one G; hence, no 'hybrid' expressions, no 'mixed Gs'. Likewise, MacSwan's Phonological Form Disjunction Theorem is conceptually redundant as the default design of the CHL itself restricts CS within X0, and, consequently, turns out to be as much an unmotivated CS-specific grammatical postulate as is the Free Morpheme Constraint a CS-specific constraint. With its elimination, the differences between monolingual and bilingual linguistic competence are logically reduced only to an additional L, enabling a bilingual speaker to produce an infinite number of well-formed sentences which are counted as expressions of either Gx or Gy.