Íslenskt mál og almenn málfræði - 01.01.2023, Blaðsíða 207

misha becker
Review and commentary
on Nowenstein’s dissertation

Iris Nowenstein’s dissertation, Building yourself a variable case system: The acquisi-
tion of Icelandic datives, presents a careful investigation of the regularities and
irregularities of the Icelandic case system, with a focus on children’s acquisition
of the system. Because of the variability in how case corresponds to the syntactic
and semantic roles of NPs in Icelandic, the acquisition question is central to
building a complete analysis of the system itself. Obviously, the system is learn-
able (children acquire the system and grow up to be adult speakers of Icelandic,
perpetuating both the regular and the variable components), but elucidating how
children learn it on the basis of input and what must be known a priori, is a non-
trivial task.
I’ll begin this review by spelling out what I see as the primary strengths of
this work, and then I’ll discuss how it contributes in important ways to current
theories of language acquisition.
One great strength of this dissertation is its wide empirical base: it includes
careful experimental work targeting both comprehension and production, with
existing and novel verbs, involving a large number of participants covering an
impressive range of ages as well as parent-child dyads, plus a careful and thorough
corpus analysis of alignment between syntactic role, thematic role, and case
marking. The experimental and corpus work build upon one another to provide
a fuller picture of both the nature of the language input children receive (corpus),
how children interpret that input (experiment), and what they themselves pro-
duce (both corpus and experiment).
The well-designed and thorough empirical component of the dissertation is
built upon a solid theoretical foundation that is interesting and insightful: it tack-
les not only the important question of how children generalize certain morpho-
logical patterns, but also how they deal with variation in the input and learn rules
that are not fully productive (in the sense of applying in every case), not to men-
tion in a system that involves syncretism in case marking. The dissertation
addresses the question of how these rules and patterns are learned, using both
theoretical models based on the Tolerance Principle (Yang 2016) and empirical
data based on children’s comprehension and production. But it goes beyond that
to extend to sociolinguistic language variation and change above the individual
level. There is even a suggestion of how the Tolerance Principle could have come
about through evolutionary processes. This is a remarkable range of theoretical
subareas of linguistics to cover within a single dissertation.