Characterising compensation. (Commentary on Ullman and Pierpont, "Specific language impairment is not specific to language: the procedural deficit hypothesis").

This article considers Ullman and Pierpont's Procedural Deficit theory of Specific Language Impairment (SLI). The theory represents an innovative attempt to fill the gap between brain and cognition in SLI, and has the potential to explain the non-linguistic as well as linguistic deficits seen in this disorder. The theory is reviewed with regard to: (1) the claims it makes on the domain-specificity of language structures; (2) the falsifiability conditions of the theory; (3) the level of detail at which compensatory processes are specified; and (4) from a computational perspective, whether the inferences that the theory draws from uneven behavioural impairments to underlying structural deficits are necessary ones.