based principles of word order. Other parts of language, such as details of phonology, phrase structure, and in
particular the inflectional system, can be severely impaired in late language learning, yielding Basic Variety and pidgins.
Similarly, morphology above all is vulnerable to agrammatic aphasia and to Specific Language Impairment (at least as
currently described).
Wecanlocalizethese problemsthroughviewingUniversalGrammar as builtoutof layeredsubcomponents. UG isnot
simply on or off in these abnormal situations. Rather, some of its subcomponents are particularly
impaired—significantly, the same ones in case after case. The robust remnant is protolanguage.
The overall conclusion is that grammar is not a single unified system, but a collection of simpler systems. Many of
these systems are built up as refinements of pre-existing interfaces between components. Hence the evolution of the
language capacity can be seen as deeply incremental, adding more and more little tricks to the cognitive repertoire
available to the child faced with acquiring a language.
We should also observe that these subsystems are added specializations for language. For instance, a syste mof
grammatical relations and a system of morphological agreement make a lot of sense as refinements of a
syntax–semantics mapping. But they are totally useless to any other cognitive capacity; they are exquisitely specialized.
This should occasion no surprise: the successive refinements of the structure of the eye—and the visual parts of the
brain such as stereopsis—are useless for anything else too. In other words, there is no need to appeal to changes in
general-purpose capacities, àlaPiattelli-Palmerini and Toulmin, to explain the incremental development of the
language capacity.
What is also new here is the hypothesis that certain design features of modern language resemble“fossils”of earlier
evolutionary stages. To some degree, then, the examination of the structure of language can come to resemble the
examination of the physical structure of present-day organisms for the traces of“archaic”features.