Example 32b is grammatical and suitable for further application of Merge; 32c is unusable as a dependent in a larger syntactic structure. The structure that then arises is given in 34b.
One question remains to be addressed. Although 31a,b are ungrammatical as they stand, they become perfectly fine when the pronominal possessor is dropped, as in The remarkable thing for 31a,b is that here dropping the pronoun has an effect on grammaticality.
A Restriction on Recursion - den Dikken - - Syntax - Wiley Online Library
Why is this? Our suggestion for 35 will therefore be that it has a structure different from those in 34 —a simpler structure, not containing IP2 at all:. The idiomatic expressions in i are essentially epithets, expressing that a discourse referent can be characterized by the adjective involved in the epithet. As Bartos suggests, these expressions are formally possessive, but there is no possessor represented in the syntactic structure. We herald this as support for the recursion restriction in 1.
The contrast in 37 shows that demonstratives can serve as possessors of complex noun phrases if they bear dative case, but not if they are caseless. Bartos a mentions this contrast but has no account for it. The improvement of i as compared with 37a may also suggest a link to the 31a — 31c contrast presented in section 3, although it remains unclear to us at this time how such a link could be profitably exploited in the analysis.
We assume that Hungarian freestanding demonstratives are dominated by a DP see Kenesei :sect. Because independently used demonstratives always have D projected, the recursion restriction in 1 prevents them from being used as caseless possessors. The ungrammaticality of 37a is thus accounted for.
When demonstratives are used adnominally rather than independently in Hungarian, two patterns present themselves. In one, shown in 38a , the demonstrative precedes the definite determiner and shows case concord with the head noun. This is illustrated in 39a. The two patterns are discrete; they cannot be mixed.
Directory of Open Access Books
To account for these patterns, we need to be explicit regarding the functional sequence of the complex noun phrase. Following Den Dikken's discussion of the parallels between the extended projections of N, P, and V, we assume that a functional head for Aspect is projected close to the lexical root. This projection harbors event aspect or Aktionsart in the clausal domain and number in the nominal domain. On top of this aspectual projection, a head for Deixis customarily called T in the clause is merged. The extended projection is topped off in the clausal case by C and in the nominal case by D.
Demonstratives systematically belong to the DxP portion of the structure of the extended noun phrase. So whenever an extended noun phrase contains a demonstrative, it must be as large as a DP. The D head of that DP may very well be silent; but it must be structurally present, because without it, DxP would not survive. Against this background, we can explain the ungrammaticality of 38c as follows. We just hypothesized that a D head must be present outside DxP. But there is no overt definite determiner in 38c —and quite independently, we know about occupancy of Spec,DP in Hungarian that it requires the presence of an overt determiner in the D head in order to be legitimate:.
These never occur in predeterminer positions and do not have freestanding, independent uses. They have no phrasal distribution. All the evidence suggests that they are exponents of the Dx head in the extended projection of the noun, lower than D cf. They can, however, raise to D and legitimate the absence of a lexical determiner. One last comment is due in this section about the two different types of demonstratives in Hungarian: the concordial and nonconcordial ones. The distribution of these two types of demonstratives suggests, as we argued in the previous subsection, that the former are phrasal originating in Spec,DxP whereas the latter are exponents of the Dx head.
Citations per year
We assume that in the complex noun phrase, only the nominal core and the D head are specified for case. Any other element inside the complex noun phrase that bears a case morpheme formally matching that of D and N must have obtained its case in a relationship of concord with something that inherently bears a case feature. Predication relations routinely give rise to case concord. With the Dx head as a relator Den Dikken of a predication relation between the terms in its complement and specifier positions, as in 42 , the phrasal demonstrative in Spec,DxP serves as a predicate of the NumP in the complement of the Dx, which is in possession of a case feature.
The fact that the phrasal demonstrative in 42 shows the same case as its subject i.
Case concord is the reflex of a relation between two phrases; heads show no case concord. For Hungarian, this is perhaps particularly evident: although Hungarian has plenty of cases and marks cases profusely in its morphophonology, the definite and indefinite articles of the language exhibit no case concord, because they are heads. That, in turn, sets a chain of events in motion that explains why 38a is grammatical but 38b and 38c are not.
Our discussion so far has focused mostly on Hungarian. So by the logic of the discussion in this article recall especially sect. The nonconstituency of the man predicts its nonextractability. There can be no doubt that strings like the man can subextract from complex noun phrases, even definite ones:. So the possessor should logically be able to strand the genitival marker. Radford shows, giving 47b as an attested example, that I think can squeeze itself in between who and the genitival marker. This is what we think applies to the cases in 47 and If so, these sentences are irrelevant to the question of whether the man can subextract from the man's coat.
Even if they do actually involve extraction with stranding of the genitival marker, however, 47 and 48 remain irrelevant to the question of whether the and man in 43 form a constituent.
Remarks on Semantic Information Description by Noun Phrases
If these meet with native speaker approval, they can be analyzed in terms of resumption. If they are bad, they pattern with The fact that these examples are irredeemable thus confirms—entirely independently of what might be going on in 47 and 48 —that the man in 43 is not a constituent see 44 , as predicted by the recursion restriction in 1. The placement in the outer D head of the definite article immediately preceding the possessor also accounts for the fact that the man's coat in 43 is outwardly definite, hence cannot serve as the associate of there , whereas a man's coat can.
In this respect, 50a patterns just like 50b. With the in the earlier example 43 occupying the D head of the possessive noun phrase as a whole, there can be no question that the man's coat distributes externally like a noun phrase headed by the. But absent a proper understanding of how percolation works, such mechanisms will not be explanatorily adequate. An analysis that treats the in 43 as the head of the possessive DP explains 50a in just the same way as we customarily explain 50b : there cannot take a definite DP as its associate.
We have argued that the fact that Hungarian interrogative, distributive, relative, and demonstrative pronouns can be dative possessors but cannot fill the caseless possessor position can be directly derived from the recursion restriction. On the assumption that DP is a phase supported here if the argumentation in this article is on target , caseless possessors are thus subject to the recursion restriction: they must be smaller than DP.
Interrogative, distributive, relative, and demonstrative pronouns are excluded from the caseless possessor position because they all project up to DP. Together, the restriction in 1 and its counterpart, the familiar condition on copy deletion in 2 , form a tandem that robustly and reliably constrains a fundamental property of natural language: recursion. Volume 21 , Issue 1. The full text of this article hosted at iucr. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account.
- Gábor Alberti, Tibor Laczko (eds);
- Audacity meaning in gujarati.
- Verschenen: Syntax of Dutch. Verbs and Verb Phrases, Volume 1 &2.
- Absolute Returns: The Risk and Opportunities of Hedge Fund Investing.
If the address matches an existing account you will receive an email with instructions to retrieve your username. Syntax Volume 21, Issue 1. But it also makes it possible for you to recognize a foreigner and to detect in what ways his sentences are ungrammatical. Scientific grammars constitute attempts at describing in a precise and enlightening way what this hidden knowledge in your brain is, how it is structured, why certain grammatical rules are the way they are.
The past fifty years or so have seen a veritable explosion of scientific work on grammar. Most of this work is in specialized articles that aim to argue specific claims about the structure of grammar in general, and that concentrate on a specific language only to put it in perspective against data from other languages.
Furthermore, ways of talking about grammar, ways of presenting formal structures, ways of interpreting data have changed drastically in the course of a mere five decades. This makes things hard for those who look for serious answers to questions about the structure of a given language. At the same time the enormous explosion of the scientific work on an ever increasing number of languages makes presenting -the scientific grammar of language X' a daunting task. But the time is ripe to do it, and as Syntax of Dutch shows it can be done and done well.
What we present in this work, the Syntax of Dutch, is the very first such attempt. In view of the massive size of the material, the work concentrates on one central aspect of grammar, viz. He has published with colleagues many volumes in the Syntax of Dutch series, and has co-edited Broekhuis, H. Main objective ix[-]3. Intended readership ix[-]4. Object of description x[-]5. Organization of the material xv[-]6.
History of the project and future prospects xix[-]7. Acknowledgments xxi[-]Introduction 1[-]Chapter 1 [-]Nouns: characterization and classification 3[-]1. Characterization 5[-]1. Classification 16[-]1.