6 Conclusion
This chapter reviews and discusses recent research on hierarchical structure in
sentence processing. First, Frank et al.’s (2012) argument against hierarchical
structure in language use was not justifi ed, supporting the “one-system view”
(Lewis and Philips 2015) that the cognitive system responsible for linguistic
knowledge is employed in language use. Second, evidence was presented to
support the claim that structure-dependence is neurologically real, strengthen-
ing the claim that the principle is a part of language design. Finally, it was
suggested that domain-general Merge and domain-specifi c Merge might be
implicated in BA 44 and BA 45, respectively.^8 Regarding the evolution of
language, what evolves is I-language of human beings. Therefore, research on
the evolution of language necessarily implicates the research on the evolution
of the capacity to acquire and use language. This chapter showed that how
languages are processed in the brain may deepen our understanding of the
evolution of language.
Notes
∗ This is a revised version of Yusa (2012b), written in Japanese, and Yusa (2012a),
incorporating the recent fi ndings of language processing, although the basic claim
regarding the functional specifi city of Broca’s area remains the same. I would like
to thank Neal Snape and Manami Sato for helpful comments on the earlier ver-
sion of this paper. This study was supported in part by the JSPS Grant-in-Aid for
Scientifi c Research (A) (23242025, Principal Investigator: Koji Fujita, Kyoto
University) and Challenging Exploratory Research (25580133, Principal Investiga-
tor: Noriaki Yusa, Miyagi Gakuin Women’s University).
1 See Pallier et al. (2011), who identify brain regions correlative to the number of
syntactic constituents as opposed to linear distance, contrary to Frank et al’s
(2012) claim.
2 See Yusa (2015) for the signifi cance of bilingual research in terms of the study
on human language faculty and language evolution. Section 4 is in part an excerpt
from Yusa et al. (2011).
3 BA 44 has been reported to be consistently activated in the processing of linguistic
hierarchical structure-dependencies (see Tettamanti and Perani 2012 for a review),
but the data do not seem to be crystal-clear. For example, Rogalsky and Hickok
(2011) claim that phonological short-term memory can explain the involvement
of BA 44 in sentence comprehension.
4 Makuuchi et al. (2009) show that the main effect of hierarchy involves BA 44,
but see Rogalsky and Hickok (2010) for a different claim, as mentioned in
footnote 3.
5 Fujita (2014) claims that Action Grammar is a precursor to Recursive Merge in
faculty of language in a narrow sense ( Hauser et al. 2002).
6 Initial activation in BA 45 for invented languages in Musso et al. (2003)
might suggest that BA 45 was sensitive to real words used in the experiments
of the acquisition of invented languages. This might provide neuroimaging
data that words can be constructed in syntax as Distributed Morphology
claims.
7 See Friederici (2009) for the claim that there are in fact two dorsal pathways and
two ventral pathways.
8 See Boeckx et al. (2014) for relevant discussions regarding BA 44 and 45.
226 Noriaki Yusa