Íslenskt mál og almenn málfræði - 01.01.2011, Síða 170

i68
Elisabet Engdahl
which semantic or discourse-related categories are represented as distinct projec-
tions in the syntactic structure (cf. Rizzi 1997). He summarizes this in terms of
hypotheses concerning the number of functional projections and the setting of IP-
parameters (p. 52 ff.). Asgrímur follows Bobaljik and Thráinsson (1998) in assum-
ing that if a language has rich agreement, then the finite verb will move out of the
VP to a position in the so called Inflection Phrase, IP. This is known as V-to-I-
movement. Icelandic has the required tense/person agreement and V-to-I-move-
ment is thus predicted to apply in all embedded clauses. This chapter also contains
an interesting discussion of what methods are most useful for eliciting and inves-
tigating syntactic variation. The author here reports on an interesting triangula-
tion of methods (written questionnaires, filling in blanks and responses to audio-
recorded and stimuli) which shows that the written questionnaires give reliable
results on the whole, but that there are constructions where the availability of a
recorded stimulus made a difference (p. 5of.). Chapter 3 is devoted to variation
with respect to the order of adverbs and finite verbs in subject-initial embedded
clauses and chapter 4 deals with so-called Embedded Topicalization, i.e. when a
subordinate clause begins with a constituent other than the subject. Judgments on
Stylistic Fronting and Expletive Insertion, as well as quantitative overviews, are
presented and discussed in chapter 5. The extent to which the different syntactic
constructions can be shown to be connected is assessed in chapter 6. In chapter 7,
a detailed proposal for the syntactic structure of embedded clauses is put forward
and finally the author summarises the results in chapter 8.
Comments and questions
At first, the reader may be worried that the large amount of data reported in the
thesis will make it hard to follow. However, Asgrímur presents the data in well
chosen overview tables which makes it easy to see the variation in judgments and
how this correlates with the age of the informants. A good example is Table2y on
p. 102 which provides an overview of the acceptability judgments concerning V2
(= V precedes adverb) and V3 (= V follows adverb) order in embedded clauses.
Although most speakers of Icelandic prefer the V2 order, the youngest Icelandic
speakers are more likely to accept the V3 order in ad-clauses than the oldest speak-
ers, whereas the oldest speakers are more likely to accept the V3 order in relative
clauses. In this clause type, the informants are quite uncertain about the accept-
ability of V3 order, as in (ía). Asgrímur shows that the type of subject also mat-
ters. If the subject of the relative clause is a pronoun, as in (ía), more people accept
the V3 order than if it is a lexical NP, as in (íb). However, examples like (íc),
where the subject has been relativized, are not included in this discussion.5
5 Formatting of the examples as presented here follows the conventions of íslenskt
mál rather than the format used in the thesis when there is a difference. The editors.