Perhaps surprisingly, what is implied here is a syntax-first model of language
evolution. Contrary to what is often taken for granted, in this model, syntax
was not the finishing touch of language evolution but rather its point of entry.
So syntax must have evolved where there was no corresponding (compositional)
semantics, for example, and there is some indication that something similar indeed
takes place in the evolution and development of birdsong (phonological) syntax
(Berwick et al. 201 1). The birth of the human lexicon may also be dependent
on Merge; as proponents of Distributed Morphology argue (Marantz 1997
et se q.), words are syntactically derived objects and not something prestored in
the lexicon before syntactic computation takes place.
Let me stress that the virtue of these hypotheses reviewed here lies not in their
novelty or radicalism but in that they show the possibility and necessity of con-
sidering a wider range of conceivable scenarios, without any bias towards hitherto
proposed popular hypotheses. Pluralism is the essence of evolutionary studies.
7 Conclusion
This chapter has pointed out four major fallacies commonly found in evolu-
tionary linguistics (sections 2–5) and briefl y presented a conceivable scenario
concerning the evolution of Merge and human language (section 6), which I
believe can carefully avoid falling victim to these fallacies.
It is important to note that the Merge-only hypothesis is not an example
of the fallacy of a single origin: It says that the evolution of human language
required only Merge as a new ingredient together with all other preexisting
capacities, which Merge allowed to evolve into distinct components of language.
This is totally different from saying that language evolved from Merge alone.
Merge itself evolved from its own precursor (Action Merge) through the process
of descent with modification and is therefore no exception to the orthodoxy
of Darwinian evolution. To complete this picture, it is also necessary to clarify
the adaptive value(s) of Merge per se, because Merge must have evolved first
before it became useful as a component of human language, which I hope to
address in future work.
In addition to those discussed in this chapter, one can easily detect a variety
of fallacies which today’s evolutionary linguistics suffers from, including strong
adaptationism (hyper-selectionism). Adaptation and selection continue to be a
major factor in biological evolution, including the evolution of Merge and lan-
guage, but they can no longer be the only factor. Modern biological thinking
reflects this understanding in the formation of the Extended Synthesis, which adds
many new ideas like evo-devo, epigenetic modification and niche construction
to the Modern Synthesis (Pigliucci and Müller 2010). Pluralism has started to
prevail. The same should be the case with evolutionary (bio)linguistics, too.
References
Balari, Sergio, and Guillermo Lorenzo. 2010. Communication: Where evolutionary
linguistics went wrong. Biological Theory 5:228–239.
150 Koji Fujita