On pidgins and other restricted linguistic systems
Conclusion
A number of different forms of linguistic communication have been grouped together as ‘‘degraded forms of language’’ (JackendoV 1999) or restricted linguistic systems (Botha 2003b, 2005, 2005/6), most of all child language, late untutored second language aquisition, pidgins and related contact varieties, language use of agrammatic aphasics, homesigns, twins’ languages, historically early non-grammaticalized language, and language use by isolated children like Kaspar Hauser, Genie, or Chelsea. With reference to language evolution, it would seem that these systems can be divided into two groups. The first comprises non-grammaticalized systems, such as those of homesigners, twins’ language speakers, and isolated children, but also those of some trained animals (see “Some cognitive abilities of animals Discussion”); in the second group are the weakly grammaticalized systems such as (extended) pidgins, and presumably also the Nicaraguan Sign Language. Systems of the first group are characterized by the presence of many or all of the properties listed in (42): They lack salient properties of human languages, in particular grammaticalization, subordination, and recursive syntactic structures, while the latter have such properties, at least in some basic form.
The reasons for this typological distinction are probably complex, but there is one reason that is probably crucial—one that is sociolinguistic in nature: As we observed in “Some cognitive abilities of animals Discussion”, grammaticalization requires a linguistic system that is used regularly and frequently within a community of speakers and is passed on from one group of speakers to another. This condition is somehow met in the case of the second group, but not in the case of the first group: Homesigners, isolated children, twins’ language speakers, and trained animals do not dispose of an appropriate sociolinguistic environment that would have made grammaticalization possible. This leaves us with the question of why systems of the first group also lack subordination and recursive syntactic structures. The answer has to be deferred, where we will argue that grammaticalization is causally responsible for these two properties to arise.
