In evolutionary biology, it is useful and customary to draw a line between
original function and current utility (see B ateson and Laland 2013), the impor-
tance of which was already recognized by Darwin himself in his discussion of
preadaptation (exaptation will be more appropriate). An oft-quoted example is
bird feathers, which originally had an adaptive value in relation with thermo-
regulation, sexual dimorphism, locomotion and the like before they became
useful for flight (see C larke 2013 for new findings on this issue).
In the case of human language, there is no denying that its current utility
includes communicative function among many others, though whether com-
munication is a major function of language or whether language is a major tool
of communication even today remains to be seen (think of the vast area of our
daily communication where nonverbal means are far more efficient), in part due
to the vague nature of communication itself. The question to be asked in the
context of language evolution is, then, whether communication was the original
function or rather an extended, later co-opted function.
There is one well-known piece of argument against the view that communica-
tion was the original function, which goes as follows. The first individual who
happened to acquire language by mutation would have had no one else around
to use this new faculty to communicate with. This “lone mutant” puzzle can-
not be solved by kin selection (F itch 2005) either, because it would have taken
at least two generations (parent-offspring) before kin selection could work to
make communication by language adaptive. The first mutant had to remain all
alone in any event, no matter how short the period was. The alternative view,
that the original function of language was thought, internal to the individual,
therefore deserves credit.
Within the minimalist framework of generative grammar, this latter view
has been corroborated by the (theory-internal) observation that the com-
putational system of human language (syntax) is optimally designed for
internalization, not for externalization (C homsky 2013, B erwick et al. 2013;
see also Bolhuis et al. 2014, 2015). That is to say, mapping from syntax
to the Conceptual-Intentional (CI) interface for thought takes place in a
uniform and straightforward manner across languages, while mapping to the
Sensory-Motor (SM) interface for communication requires many language-
particular adjustments, including specification of diverse morphophonological
realization, linearization of hierarchical structure and deletion of copies both
created by the recursive combinatorial operation Merge, itself generating
only unordered sets, all of which would be unnecessary if human language
were exclusively an instrument of thought. Language as an instrument of
communication has to be more complicated and therefore is very likely to
be a later innovation, a matter of cultural evolution, to speak metaphorically
(see Figure 9.1).
The problem with this minimalist consideration to support the view that com-
munication initially had nothing to do with language evolution is, of course,
that it is totally theory-dependent; a different way of looking at language will
necessarily lead to a different picture of language evolution. This is where
142 Koji Fujita