| AUTHOR: Julien, Marit
TITLE: Nominal Phrases from a Scandinavian Perspective
PUBLISHER: John Benjamins
Michael T. Putnam, Carson-Newman College
This monograph presents a new model of the internal syntax of nominal phrases,
with a particular focus on Scandinavian languages. The degree of variation
within Scandinavian languages is quite remarkable and makes this closely-related
language family an excellent source for such a case study. The largest variation
with regard to the syntax of nominal phrases in this language family is found in
definite nominal phrases and in the realization of possessors. Although these
two topics and structures receive the most attention in this monograph, other
topics related to nominal phrases – such as predicate nominals – are also
addressed. Although the framework/formalism in which Julien's treatment of the
internal syntax of nominal phrases in Scandinavian languages is couched in the
most recent version of the Principles-and-Parameters framework, the Minimalist
Program (Chomsky 1995), a handful of Julien's proposals challenge recent
advances in the standard theory (e.g., his treatment of predicate nominals in
Chapter 7). As such, Julien's work simultaneously advances and challenges core
conceptual claims pertaining to the internal syntax of nominal phrases (in
Scandinavian) in the Minimalist Program.
In Chapter 1, Julien sketches out in detail his assumptions for the general
structure of the DP, outlined in (1) below (the structure in one is slightly
amended in later chapters).
[DP [CardP [αP [nP [NumP [NP N]]]]]]
Of particular interest for Scandinavian DPs, Julien devotes exacting attention
to the projection below DP, a projection which he labels nP. The N head always
moves to n, and it is in n that the suffixed article is generated, with the
exception of Danish. Accordingly, with the exception of indefinite singular DPs
in Icelandic, Julien argues that in referential DPs the D head must be visible
if the reference of the DP as a whole is dependent on D. D is visible if there
is phonologically overt material in D or in Spec,DP (p. 24). Julien also deals
with indefinite nominal phrases, which according to his analysis do not display
any movement in Scandinavian except for the obligatory movement of N to Num and
n. Thus, the only variation found in Scandinavian indefinite DPs has to do with
the realization of D: in indefinite plural DPs, and in indefinite singular DPs
based on mass nouns, D can be phonologically null. In all other cases where the
reference of indefinite DPs is dependent on D, D must be spelled out as an
indefinite determiner, except in Icelandic, where there is no indefinite
singular determiner. Julien purports that it is the morphological case that
enables an indefinite DP to be interpreted as referential in Icelandic (p. 25).
Chapter 2 takes a closer look at the semantic contributions of n and D (event if
their feature content is identical). Following this assertion, an overtly
definite n head gives the DP a specific reading, while an overtly definite D
head supplies the inclusiveness (or uniqueness) that is characteristic of
definite expressions (p. 75). In his analysis of nouns that resist nominal
inflection (section 2.8), Julien advocates that nP (and subsequently also DP) is
a strong phase (similar to clausal-level vP) in the sense of Chomsky (2001 and
later work). Departing from recent assumptions of phase-based minimalist
desiderata, Julien follows Svenonius (2000, 2001) in assuming that a (strong)
phase goes to Spell-Out immediately upon its completion, and critically not at
the completion of the immediately preceding (strong) phase. Thus, as soon as nP
is completed, it is spelled out. In this chapter Julien also addresses varieties
of ''double definiteness'' of Scandinavian (e.g., Norwegian, Swedish and Faroese),
the syntax and morphology of adjectival phrases in the light of ''blocking
effects'' created by the movement of nP to Spec,DP (see also Roehrs 2006), and
certain types of Scandinavian nouns that do not take nominal inflection.
The discussion in Chapter 3 directs its attention to the connection between
determiners and relative clause types. As noted by Julien, in those Scandinavian
varieties where a pronominal determiner regularly shows up in definite nominal
phrases containing adjectives or weak quantifiers – namely, in Danish, Faroese,
Norwegian and Swedish – the very same pronominal determiner is also often used
in absence of pronominal modifiers if the noun is followed by a restrictive
relative clause (p. 108). Although some variation does exist amongst speakers,
restrictive relative clauses appear to be able to trigger the presence of a
pronominal determiner. In the previous chapter (Chapter 2), Julien advances the
claim that a pronominal determiner appears when there is no phonologically
realized element in Spec,CP. If this claim is correct, restrictive and
non-restrictive relative clause constructions differ not only in their covert,
but also in their overt syntactic structure (contra Kayne 1994). Under this
analysis, in non-restrictive relative clause constructions, the correlate is a
full DP generated in the highest specifier of the relative clause. This DP moves
to the specifier of a DP projection that takes the relative clause as its
complement. In contrast, the element that appears in the highest specifier
position in restrictive relative clause constructions is an nP. Another ''welcome
result'' of this proposed model is that the relative correlate is taken to be
separate from the relative operator. As a consequence of the separation of these
two elements, the observation that the correlate and operator can differ with
regard to definiteness and case receives a straightforward explanation.
Demonstratives and strong quantifiers are the primary topic of discussion in
Chapter 4. The fact that demonstratives can precede pronominal determiners in
Scandinavian suggests that demonstratives are generated above D, in a projection
Julien labels DemP. Demonstratives can (apparently) specify the reference of the
DP, thus explicating why we (normally) see no pronominal determiner in
Scandinavian when a demonstrative is present, ''regardless of whether the DP
contains adjectives or numerals'' (p. 137). Julien argues that strong quantifiers
are generated even higher up than demonstratives, and take a DemP or a DP as
their complement. Since strong quantifiers have nominal category features, in
Mainland Scandinavian the DP projection can be empty following a strong
quantifier in cases where the DP has a discourse anaphoric function.
Chapters 5 and 6 present an analysis of postnominal and pronominal possessors
respectively. The common trait of all Scandinavian postnominal possessors is
that they are licensed in Spec,NP, the position where they are base-generated.
Historically (and in some variants of modern Icelandic) the possessor would
appear in the genitive case and be licensed directly by the noun; however, in
modern Mainland Scandinavian varieties postnominal possessors are licensed
through agreement with a possession feature [POSS] in n. Similar to postnominal
possessors, pronominal possessor are also base-generated in Spec,NP and interact
with n in much the same way as their postnominal counterparts (i.e., through the
[POSS] feature). Contrastively the pronominal possessors end up in Spec,DP (via
Spec,PossP). If the possessor is pronominal, neither D nor Poss is normally
spelled out (although there are exceptions), but if the possessor is
non-pronominal, the Poss head gets spelled out as –s (-sar(a) in Faroese), which
only realizes the [POSS] feature of Poss, or as a pronominal element, which
spells out a set of phi-features in addition to the [POSS] feature. Julien also
discusses parametric variation within Scandinavian languages regarding the
distribution of possessor phrases, including dialects of Swedish and Danish
where the –s inflection suffixed to a possessor is not the phonological
realization of Poss (but rather a [POSS] feature belonging to the possessor
itself), Icelandic where the possessors move to the front of the noun only if
they are focused, and many varieties of Norwegian, which show a pattern that
falls between that of the Swedish/Danish and Icelandic variants.
In Chapter 7, Julien addresses the claim that nominal phrases that function as
predicates are structurally distinct from nominal arguments. Julien argues here
that nominal predicates differ from nominal arguments semantically, ''such that a
nominal phrase can be a predicate only if it can be assigned a purely
intensional interpretation'' (p. 296). In contrast, a nominal argument can have
an intensional or an extensional interpretation. Such a stance goes against
recent theorizing in syntax; however, it does return to claims made by Williams
(1983), Partee (1987) and Jackendoff (2002). Chapter 8 concludes this work by
presenting empirical evidence from non-Scandinavian languages that further
support the structures and claims championed by Julien in this text.
This is volume represents scholarship par excellence in the generative treatment
of nominal phrases. Both the empirical and conceptual coverage of natural data
in Scandinavian languages make this work a must-see reference for scholars
involving in any serious study of DP-syntax and/or Scandinavian linguistics.
Following Roehrs' (2006) adaptation, review and eventual critique of some of the
core arguments of Julien's work, I concur with Roehrs (2006) that there are two
issues in particular that Julien may wish to revisit in her research (for a more
in depth discussion of Roehrs' treatment of these matters, the reader is
referred to Roehrs 2006: Chapter 2). First, recall from the DP structure in (1),
that CardP and αP receive separate functional projections that are filled,
according to Julien, if lexically filled by a numeral and adjective(s),
respectively. In order to account for the different patterns exhibited in
Scandinavian DPs, Julien assumes for unmodified DPs that it is a lexical feature
that determines whether Art is overtly realized (in languages such as Norwegian,
Swedish, Faroese, and Icelandic) or D (in Danish and literary Icelandic). In all
cases, ArtP moves to Spec,DP to license the DP. The determiner in Art is
supported by N-raising and the one in D by ArtP, containing the head noun.
Although Roehrs makes use of the same underlying structure for DPs in German
that Julien does, he departs significantly from her in two ways. One of the
major differences between Julien's and Roehrs' treatment of determiners lies in
their respective treatment of the Icelandic data. Similar to Julien, Roehrs
argues that modifiers induce a blocking effect; however, Roehrs conjectures that
the long-distance agreement relation between art and D is ''blocked'' by Agree
(probe-goal), and hence does not rule out movement due to the presence of an
adjective. Furthermore, allowing phrasal movement across the adjective gives a
straightforward account of the Old Icelandic pattern and allows movement of a
phrase across a numeral in common Modern Icelandic without further assumptions.
Quoting Roehrs (2006:114), ''The main differences between Old and (common) Modern
Icelandic are that (i) movement of NP was replaced by that of AgrP over time,
such that the adjective now moves along to Spec,CP and (ii) the determiner does
not move to D overtly anymore.'' Secondly, while Julien states which determiner
position is overtly realized in which language (Art or D), Roehrs claims that
determiners can be split up in only some languages (if an adjective is present).
Such an analysis has the advantage of achieving ''a homogenous account of the
unmodified DP'' by connecting ''the property of splitting up determiners to the
corresponding different semantic interpretations of the DPs, overtly manifested
in some languages but not others'' (Roehrs 2006:114). In future work, it will be
interesting to see how Julien can address Roehrs' assessment of her research.
To recapitulate, aside from these aforementioned issues that Julien must address
in future research, this work represents an amazingly thorough treatment of
nominals in Scandinavian and will serve as a benchmark reference work for many
years to come.
Chomsky, N. 1995. _The minimalist program_. Cambridge, MA: MIT Press.
Chomsky, N. 2001. Derivation by phase. In _Ken Hale: A life in language_, M.
Kenstowicz (ed), 1-52. Cambridge, MA: MIT Press.
Jackendoff, R. 2002. _Foundations of language_. New York: OUP.
Kayne, R. 1994. _The antisymmetry of syntax_. Cambridge, MA: MIT Press.
Partee, B. 1987. Noun phrase interpretation and type-shifting principles. In
_Studies in discourse representation theory and the theory of generalized
quantifiers_, J. Groenendijk, D. de Jongh and M. Stokhof (eds), 115-143.
Roehrs, D. 2006. _The Morpho-Syntax of the Germanic Noun Phrase: Determiners
MOVE into the Determiner Phrase_. PhD dissertation, Indiana University.
Svenonius, P. 2000. Impersonal passives: A phase-based analysis. In _Proceedings
of the 18th Scandinavian Conference of Linguistics_, vol. 2, Arthur Holmer,
Jan-Olof Svantesson and Åke Viberg (eds), 109-125. Lund: Linguistics Department,
Svenonius, P. 2001. On object shift, scrambling and the PIC. _MIT Working Papers
in Linguistics_ 39: 267-289.
Williams, E. 1983. Semantics vs. syntactic categories. _Linguistics and
Philosophy_ 6: 423-446.
ABOUT THE REVIEWER
Michael T. Putnam is Assistant Professor of German and Linguistics at
Carson-Newman College. His research interests include (but are not limited to),
biolinguistic theory, i.e., specifically on language as an ''organ'' of the human
brain, syntactic theory, the syntax-semantics interface, and Germanic linguistics.