Adjectives in layers

b. .pjat'opredelennyxbalerin.

fivecertain.GENballerinas.GEN

`In the Mariinsky Theatre danced five ballerinas of a certain kind.' m

Moreover, even the following phrase is acceptable in an appropriate context:

(i) pervyjepjat'pervyxucitelej

first.NOMfivefirst.GENteachers.GEN

`the first five first teachers'

In (25a), the nominal receives a specific interpretation and is understood to refer to a particular set of ballerinas. The adjective opredelennye `certain, particular' affects the referentiality status of the whole nominal, making sure that it refers to a specific group of individuals. In turn, as pointed out by Pereltsvaig, (25b) relates to five ballerinas of a certain kind. Once the adjective occupies a lower position, it applies to the property denoted by the head noun.

Demonstratives provide another interesting case to consider:

(26) a. etipjat'masin

these.NOMfivecars

`these five cars'

b. pjat'etixmasin

fivethese.GENcars

`five of these cars'

`five cars of this kind'

The A-initial order illustrated in (26a) is semantically unmarked, in the sense that here we receive the typical meaning associated with a demonstrative: the phrase refers to a group of individuals that are familiar from either linguistic or physical context. In turn, (26b) is interpreted differently. One interpretation that it may have is a partitive interpretation `five of these cars'. To distinguish the demonstrative and the partitive reading, imagine a manager of a car dealership giving instructions to a salesperson to sell (26a) or (26b): in the former case, the manager has to point out the specific five cars in the lot and the salesperson is required to sell the five cars pointed out by the manager. But if the instruction are to sell (26b), the manager must point out a larger number of cars on the lot (i.e., more than five), and the salesperson is required to see some five cars out of that larger set.

But (26b) may also have a different meaning, according to which the demonstrative applies to some property of cars involved, rather than to the specific set of cars. This meaning is comparable to that of `five such cars' or `five cars of this type'. The relevant type of cars must be familiar from the context, but not necessarily a set of the particular cars involved. In fact, the phrase as a whole need not be interpreted as definite or even specific. The latter use of a demonstrative is illustrated in (27):

(27) Vsego	za	vojnu	VVS	SSA	poterjali
all.in.all	during	war	Air.Force	USA	lost
pjat'	etix	masin...
five	these.GEN	vehicles.GEN



7. All in all, during the war, US Air Force lost five vehicles of this kind

The generalization appears to be the following: under the Num-initial pattern, demonstratives and specificity markers apply to the property denoted by the NP. But under the A-initial pattern, these items relate to specific individuals that instantiate that property.

Also, a similar kind of behavior is exhibited by a number of indefinite pronouns. For instance, consider the pronoun kakoj-to `some'. As a rule, such pronouns appear phrase-initially, mark the nominal as indefinite and provide information regarding the speaker's knowledge about the referent (namely, that the referent cannot be identified by the speaker). Thus, the phrase in (28a) can be used to relate to five ballerinas that are not known to the speaker. But an alternative order, as in (28b), is possible too. This phrase is most likely to be interpreted as `five ballerinas of some type', where the speaker does not know which type of ballerinas exactly the women instantiate.

(28) a. kakie-topjat'

balerin ballerinas.GEN

some.NOMfive

`some five ballerinas'

b. pjat'kakix-to

balerin ballerinas.GEN

fivesome.GEN

`five ballerinas of some kind'

Our next example, appropriately, involves the adjective sledujuscij `next'. As with the adjectives considered above, sledujuscij `next' may occur either before or after a numeral, with the different case marking.

(29) a. sledujusciepjat'knig

next.NOMfivebooks.GEN

`the following five books'

b. pjat'sledujuscixknig

fivenext.GENbooks.GEN

`five of the/some following books'

This example is somewhat different from what we have discussed above. Both orders are acceptable. As expected from our discussion of similar examples so far, the adjective in (29b) can be understood as propertymodifying, in the sense that the entities instantiate the property of being a book that comes later in some ordering than a certain contextually specified book (e.g., was written later or stands on the shelf after some contextually specified book). The phrase is definitely not exhaustive; see discussion of exhaustivity in the next subsection. Also as expected, the phrase with the A-initial pattern in (29a) can be interpreted as meaning `the next five books', but it can also have the interpretation in which the list of books is to be provided after the phrase. In such cases, the adjective is cataphoric, and the nominal has to be interpreted referentially.

(30) Interespredstavljajutsledujusciepjat'knig:

interestpresent.3.PLfollowing.NOMfivebooks.GEN:

“Emma”,“Gordost' i predubezdenije”,“Oliver Twist”,

Emma,Pride and Prejudice,Oliver Twist,

“Mol' Flanders”i“Alice v strane cudes”.

Moll FlandersandAlice in Wonderland.

`The following five books are of interest: Emma, Pride and Prejudice, Oliver Twist, Moll Flanders and Alice in Wonderland.'

Of course, under the cataphoric use, the adjective cannot be interpreted as applying to the property. Rather, its use is purely referential. Non- surprisingly, such an interpretation is possible only under the A-initial order.

The facts are rather similar with the adjective takoj `such', which can also receive a cataphoric reading, but only if it appears in the higher position. Both (31a) and (31b) may mean `two such textbooks', although under this interpretation, (31b) is somewhat more natural.

(31) a. takie such.NOM

dva

two

takix

such.GEN

ucebnika textbooks.GEN

ucebnika textbooks.GEN

b. dva two

Interestingly, even under the `such' reading, the two phrases do differ. The phrase in (31b) may relate either to the subject of the textbooks (e.g. such textbooks = textbooks of physics) or to less inherent properties (e.g. such textbooks = old and dirty textbooks), but the phrase in (31a) cannot refer to the subject of the textbook. For example, (31b) but not (31a) can serve as a natural continuation for Nam ocen' nuzny ucebniki po semantike. My kupili... `We really need textbooks in semantics. We bought...'. Our explanation for this fact is as follows: in (31a) takie `such' is merged outside the NP too high to be interpretable as the argument of ucebnik `textbook', while in (31b) takie `such' is low enough to be interpretable as referring to the argument of `textbook'.

Furthermore, (31a) is much more appropriate than (31b) in cataphoric cases like (32), where the nominal is followed by the list of textbooks, and takie is interpreted as `the following':

Bylikuplenytakiedvaucebnika: wereboughtsuch.NOMtwotextbooks.GEN

“Vvedenie v semantiku” i “Osnovy sintaksiceskogo analiza” Introduction to Semantics and Basics Syntactic Analysis `The following two textbooks were bought: Introduction to Semantics and The Basics of Syntactic Analysis!

Again, the cataphoric meaning, which forces a referential interpretation of the nominal, is only available under the A-initial pattern.

Our final example involves the adjective redkij `rare' and its interpretational properties. As a rule, this adjective modifies the property denoted by the noun and, thus, appears under the Num-initial pattern (33 a). However, the alternative A-Num order is possible as well, which results in an interesting shift in the interpretation:

(32) a.Pjat' redkixzivotnyxpereplyvutetureku.

five rare.GENanimals.GENwill-swim-acrossthisriver

`Five rare animals can swim across this river.'

b.Redkie pjat'zivotnyxpereplyvutetureku.

rare.NOM fiveanimals.GENwill-swim-acrossthisriver

8. It's difficult to find five animals that are able to swim across this river

The sentence in (33a) asserts that five rare animals, i.e. five animals of rare species, will cross the river, whereas the sentence in (33b) involves a totally different claim. Roughly, the latter sentence asserts that there exist few (if any) groups of five animals that are able to cross the river. Such animals may not exist at all; if they exist, they need not be of rare kinds.¹⁰ Thus, while in (33a) the adjective modifies the kind, in (33b), it relates to instances of the kind (we may say that it essentially quantifies over five-member sets of such instances, specifying that there are few such sets).

On the basis of the data discussed above, the following conclusions can be drawn. First, we have shown that a range of adjectives can appear in two distinct positions, above or below NumP. Second, the interpretation of such adjectives depends on the structural position they occupy. Third and final, the position located above NumP is associated with the referentiality status of the nominal; adjectives that appear in this position describe not the property but rather the individual instantiations that the nominal refers to.

The last point is especially important. It reveals that in Russian, there is a structural position within nominal expressions that appears above NumP

10 Compare to the quote from Nikolai Gogol's Vecera na xutore bliz Dikan'ki, praising the width of Dnieper:

(i) Редкая	птица	долетит	до	середины	Днепра.
Redkaja	ptica	doletit	do	serediny	Dnepra.
rare	bird	will-fly.PERF	to	middle	Dnieper.GEN
Пышный!	Ему	нет	равной	реки в	мире.
Pysnyj!	Emu	net	ravnoj	rekiv	mire.
Magnificent	to-it	there-is-no	equal	river in	world

`Hardly any bird could fly to the middle of the Dnieper. Magnificent! It has no equal river in the world” and is responsible for referential interpretation. This position makes sure that the nominal is not interpreted as a property (type <e, t>), but rather as an individual (<e>) or a quantifier (<<e, t>, t>). Of course, this is exactly what characterizes the DP projection. Thus, the data provided above constitute evidence for the existence of the DP projection in Russian.

9. Exhaustivity

We claim above that even an article-less language like Russian has a DP projection and that this is the projection crucial to the referential interpretation of poslednie-type adjectives, which are merged in an aP immediately above the DP. These claims are further buttressed by the data involving exhaustivity inferences that depend on the position of possessive phrases. Like other examples of poslednie-type adjectives discussed in the preceding subsection, possessive adjectives (e.g., diminy `Dima's') and possessive pronouns (e.g., moi `my') too can appear either above or below a numeral. If they appear above a numeral, the phrase receives an exhaustive interpretation, which is associated with definiteness and with the DP projection (cf. [Zamparelli, 2000]). In contrast, when a possessor appears below NumP, exhaustivity is absent. These facts constitute further evidence that the high position in which adjectives can appear is located in the DP field.



To illustrate, consider the following contrast:

(34) a. pjat'	diminyx	knig
five	Dima's.GEN	books
b. diminy	pjat'	knig
Dima's.NOM five	books

both: `Dima's five books'

Normally, possessive adjectives such as diminy `Dima's' appear to the right of the numeral, as in (34a), and in such cases, exhaustivity is absent. However, they may also appear phrase-initially, to the left of the numeral, in which case the exhaustivity interpretation emerges. Thus, (34b) presupposes that Dima has exactly five books, whereas (34a) does not carry such a presupposition.¹¹

The facts concerning demonstratives are actually similar:

(35) a. etipjat'masin

these.NOMfivecars2

`these five cars'

(35) b. pjat'etixmasin

fivethese.GENcars

`five of these cars'

`five cars of this kind'

The phrase in (35a), in which the demonstrative appears above NumP, is exhaustive: it presupposes that there exist exactly five relevant cars. This, we propose, results from the fact that the demonstrative appears in the DP area. In contrast, (35b) does not involve exhaustivity, since here, the demonstrative appears below NumP (and is thus not in, or immediately above, the DP).

10. Adjectives in Small Nominals

If poslednie-type adjectives are merged in the functional projection above the DP, we predict that they should not be grammatical in nominals that lack the higher levels of the functional architecture, specifically the DP. In other words, we predict that high adjectives considered in this section are ungrammatical in Small Nominals [Pereltsvaig, 2006a; Kagan, Pereltsvaig, 2011]. The prediction is borne out. For instance, Kagan and Pereltsvaig (2011) argue that genitive complements of intensive reflexive verbs in Russian (verbs that contain the prefix na- and the suffix -sja) are Small Nominals which lack the DP and even the NumP projections. For instance, this view is supported by the fact that the nominals in question cannot contain numerals and other quantifying expressions.

(36) *Ja najelas'pjati/ djuzinykotlet.

Ina-ate-sjafive.GENdozen.GENburgers.GEN

intended: `I ate my fill of five / a dozen burgers.' [Kagan, Pereltsvaig, 2011, p. 223]

Crucially for our purposes, genitive complements of intensive reflexives cannot contain high adjectives. Within these genitive nominals, such adjectives are either ruled out completely or, if accepted, receive the propertymodifying interpretation that is associated with lower positions.

(37) a. Janacitalas'takix

ucebnikov.

textbooks.GEN

Ina-read-sjasuch.GEN

`I have read my fill of such textbooks.'

b. Masanasmotrelas'etixfil'mov.

Mashana-watched-sjasuch.GENfilms.GEN

`Masha has watched her fill of such films.'

c. Lenanajelas'redkixkonfet.

Lenana-ate-sjarare.GENsweets.GEN

`Lena has eaten her fill of a rare type of sweets.'

Specifically, takix in (37a) can only relate to the type of textbooks involved. For instance, the sentence may mean, depending on the context, that the speaker has had enough of reading textbooks written by nonprofessionals or textbooks that are not reader-friendly. Furthermore, takix `such' may be interpreted as referring to the subject-matter of the textbooks the speaker has had enough of: for example, the speaker has had enough of reading textbooks in physics. Recall from our earlier discussion that this interpretation is possible only if takix `such' is merged low enough to be interpretable as an argument of ucebnikov `textbooks.' However, takix `such' in (37a) cannot receive a cataphoric reading, which is associated with a referential interpretation and signals the presence of a DP projection:

(38) *Janacitalas'

Ina-read-sja

“Vojna i mir”,

War and Peace intended: `I have read my fill of such books as the following: War and Peace, Idiot, etc.'

takix	knig:
such.GEN	books.GEN
“Idiot”	i t.d.
Idiot	etc.

Analogously, the demonstrative etix `these' in (37b) relates to the kind of movies involved. The object is interpreted as `such movies' or `movies of this type'. It cannot be used to refer to a particular set of movies that has been previously mentioned in the discourse. Finally, (37c) asserts that Lena has eaten a fair amount of a rare type of sweets. It cannot mean that it is difficult to find sweets that Lena has eaten in a sufficient quantity. In other words, the adjective redkix modifies the property-denoting noun, rather than quantifying over objects.

The sentences in (37) above show that even poslednie-type adjectives are forced to receive property-related meanings that have been shown above to arise in lower positions. This is to be expected if, as we argue, the referential and quantificational meanings of such adjectives only arise when they are merged above the DP projection, which is absent in Small Nominals such as the genitive complements of intensive reflexives. If a given adjective can only receive the “high”, referentiality-oriented interpretation, due to its lexical properties or to the context in which it appears, it is incompatible with Small Nominals:

(39) *Janaelas'{ostal'nyx/sledujuscix/pervyx/dannyx}

Ina-ate-sja{remaining/following/first/given}

kotlet.

burgers

intended: `I ate my fill of the {remaining/following/first/given}

burgers.' [Kagan, Pereltsvaig, 2011, p. 223]

To sum up, the fact that poslednie-type adjectives are attached at the DP-level is supported by their incompatibility with Small Nominals, in which this projection is absent.

Of course, these facts also leave open the possibility that the adjectives appear lower than the DP and are merged immediately above the NumP projection, which is absent from these nominals as well. However, there are two reasons to reject this alternative. Firstly, the referentiality-oriented interpretation of numerous poslednie-type adjectives makes them more naturally associated with DP than with NumP. Secondly and most importantly, the alternative analysis is ruled out by the fact that there is another type of adjectives (dobryx-type) that are merged between the DP and NumP projections (cf. position aP-2 in our tree in (1) above). These adjectives are discussed in the following section.

11. Dobryx-type adjectives

As we show in the preceding section, adjectives do not always modify the noun itself (contra [Rijkhof, 2002], who places adjectives in the innermost (Quality) layer). Quite the contrary, adjectives may modify not only the property denoted by the noun itself or its projection NP, but also provide additional referential information about the individual denoted by the DP as a whole. In this section, we consider yet another type of adjectives in Russian, which modify (or express speaker's evaluation) of the quantity denoted by the NumP. This type of adjective has been identified by Babby (1987); following Babby's work and Pereltsvaig (2011), we will refer to these adjectives as the dobryx-type adjectives. In addition to dobryx `good', this relatively small class of adjectives includes celyx `whole', dolgix `long', kakix-nibud' `some/any', nepolnyx `incomplete' and a few others.

In terms of their position, such adjectives appear before numerals (and other quantity expressions), as shown in (40) below.

(40) a. celyxtridcat'svobodnyxdnej

wholethirtyfreedays

`a whole thirty free days' [Babby, 1987, p. 121]

b....otnositel'nonedavnootkrytocelyxdofigapescer

relativelyrecentlydiscoveredwholetofigcaves

`Relatively recently, a whole lot of caves has been discovered (there).'

Note that those adjectives among the dobryx-type that can occur both above and below numerals have different meaning, depending on the position:

(41) a. celyxdesjat'celyxbutylok

wholetenwholebottles

`a whopping ten unbroken bottles'

b. Onsoversildobryxdesjat'dobryxdel.

hecommittedgoodtengooddeeds

`He committed a whopping ten kind deeds.'

It can be seen that when the adjectives appear in the higher position (according to our analysis, aP-2; cf. (1) above), they relate to the quantity denoted by the quantifier. In particular, they express the speaker's evaluation of this quantity, generally either as impressively high or as relatively low. Importantly, the adjectives, when they appear to the left of the numeral, do not modify the noun and contribute no information about the property denoted by the NP. Thus, both desjat' butylok `ten bottles' and celyxdesjat' butylok `a whopping ten bottles' contribute exactly the same property of being a bottle. In contrast, when the same adjective appears in a lower position, it does affect the property contributed by the nominal (but provides no evaluation of the quantity). Thus, the phrase desjat' celyx butylok `ten unbroken bottles' involves the property of being an unbroken bottle. Once again, we see that the interpretation of the adjective and its domain of modification depends in a crucial way on the syntactic position it occupies. In this section, we concentrate on the higher position available to dobryx-type adjectives, which is located above the numeral and in which, a quantity-related evaluative meaning is triggered.

12. Dobryx-type adjectives: Syntactic analysis

Our proposal is that these adjectives are merged in aP-2, that is above NumP but below the level of DP (and consequently, below the level of the poslednie-type adjectives, discussed in the preceding section). That the dobryx -type adjectives are merged below the poslednie-type is confirmed by their relative ordering:

(42) posledniecelyxsem'letotdany

lastwholesevenyearsgiven

polnometraznomyxudozestvennomufil'mu

[feature-lengthfictionfilm].DAT

`The last whole seven years have been dedicated to a feature-length fiction film.'

Note that dobryx-type adjectives do not affect the referentiality status of the DP (unlike the poslednie-type adjectives), nor do they modify the property contributed by the nominal (unlike the lower adjectives that appear to the left of the numeral; cf. Section 3). Instead, they apply to the quantificational meaning component. This is captured under our analysis by the fact that they occupy a special syntactic position, which differs from those of all the other adjectives. The dobryx-type adjectives appear below the DP level, and thus too low to affect referentiality. Rather, they are merged in the aP immediately above NumP, which results in their quantity-related interpretation.

A further support for our analysis comes from a negative demonstration: we show in the next subsection that an alternative analysis that places dobryx-type adjectives closer to the numeral that they modify is not validated by the data.

13. Dobryx-type adjectives: An alternative analysis (to be rejected)

According to Babby's (1987, p. 122) original analysis of dobryx-type adjectives, they must occur closer to the numeral that they modify. Unfortunately, his analysis cannot be easily restated in contemporary X'-theoretic terms, but one way to implement his general idea would be to place dobryx-type adjectives in the specifier of the functional projection in which the numeral itself appears (or in the specifier of the quantityexpressing PP, such as do figa `lots' in (40b) above). In what follows we provide evidence against this alternative analysis and ultimately reject it in favor of the analysis outlined in (1) above.

But before we proceed, it is crucial to consider the question of where the numeral itself is. According to the dual analysis of Bailyn (2004) and Pereltsvaig (2006b), the position of the numeral depends on the case marking pattern: in Babby's (1987) heterogeneous case pattern, that is if the noun phrase as a whole appears in a structural case position (i.e., nominative or accusative case), the numeral is argued to appear in the specifier of NumP, whereas in Babby's homogeneous case pattern, that is if the noun phrase as a whole appears in an oblique case position (e.g., dative, genitive, instrumental or prepositional case), the numeral is argued 2to appear in the head of NumP. One piece of evidence for this dual analysis

Iinvolves the distribution of phrasal quantity expressions such as the PPm do figa `lots': it can occur only in the heterogeneous case pattern illustrated in (43 a) and not in the homogeneous case pattern, illustrated in (43b). Since this quantity expression is phrasal it cannot appear in those structures where the corresponding numeral would occur in the head position.

(43) a.Bondvypil{pjat'/dofiga}koktejlej.

Bonddrank-upfive. ACC/tofig}cocktails. GEN

`Bond drank up {5 / a lot of} cocktails.'

b.Bondnapilsja{pjat'ju/*dofiga}koktejljami.

Bondgot-drunkfive.INSTR/*tofig }cocktails.INSTR

`Bond got drunk from {5 / a lot of} cocktails.'

As shown below, dobryx-type adjectives are possible in both case patterns:

(44) a. Artisty polucili celyxdesjat' nagrad.

rewards.GEN

desjatju

ten.INSTR

Actors received whole.GEN ten.ACC

`The actors received a whole ten rewards.'

b. Artistybylipremirovanycelymi

Actorswereprizedwhole.INSTR

nagradami. rewards.INSTR

`The actors received a whole ten rewards.'¹⁴

On the alternative analysis, which we will ultimately reject below, the examples in (44) will be analyzed as follows: in the homogeneous case pattern in (44b), the numeral is in the Num° and celyx `whole' can be taken to be in the Spec-NumP. The structure for (44a) is a bit more complicated: here, the adjective is in the specifier of a phrasal category QP headed by the numeral and the QP is in the specifier of NumP; the Num° itself is empty (as shown by Pereltsvaig 2006b, this empty Num° serves as an intermediate landing site for Approximative Inversion, which is possible in the heterogeneous case pattern but not in the homogeneous case pattern). These alternatives are schematized below:



Although the structures in (45) seem to represent better Babby's original insight that celyx `whole' modifies the numeral only, we argue that the structure we proposed in (1) above - the relevant portion of this structure is repeated in (46) below - is the correct one.

The argument in favor of (46) over the alternatives in (45) comes from the data involving the so-called Approximative Inversion and the optional numeral classifiers (such as stuk `items'; cf. Section 2 above). The Approximative Inversion is a process which creates an approximative meaning by inverting the highest nominal element around the numeral; this highest nominal element can be any element with nominal morphology: a numeral classifier, as in the examples below; a measure or container noun; or a lexical head of the noun phrase itself (see Pereltsvaig 2006b for a more detailed discussion of Approximative Inversion in Russian). The simplest case of Approximative Inversion inverts the noun around the numeral:

(47) a. sto karandasej

100 pencils.GEN

`a hundred pencils'

b. karandasej sto

pencils.GEN 100

`approximately a hundred pencils'

If an optional numeral classifier like stuk `items' is present, the Approximative Inversion will invert the classifier rather than the noun around the numeral.

(48) a. sto(stuk)

100items.GEN

`a hundred pencils'

b. stuksto

items.GEN100

karandasej

pencils.GEN

karandasej

pencils.GEN

`approximately a hundred pencils'

(48) c. *karandasej sto stuk

pencils.GEN 100 items.GEN

Given the three-way distinction we adopt between numeral, sortal and noun classifiers (see Section 2 above), we take the classifier stuk `items' to be merged in UnitP (tucked between NumP and ClP). Also, we follow Pereltsvaig's (2006b) analysis of Approximative Inversion as Head Movement into Evid° (i.e., the head of the EvidP, an optional projection merged inside DP, immediately above NumP). Now we have all the pieces in place and are ready to consider what happens in cases of Approximative Inversion of the classifier stuk `items' in the presence of a dobryx-type adjective. Since the numeral classifier is the highest nominal element (as discussed immediately above), it will be the nominal element to invert around the numeral. But where does it land? It turns out that the landing site for stuk `items' is between the dobryx-type adjective and the numeral rather than above the adjective:

(49) a. *stuk	dobryx	sto	karandasej
items.GEN	good.GEN	100	pencils.GEN
b. dobryx	stuk	sto	karandasej
good.GEN	items.GEN	100	pencils.GEN

Additional naturally occurring examples are provided below:

(50) a. dobryx	stuk	dvadcat'	pisem
good.GEN	items.GEN	20	letters.GEN
ot	svoej	pervoj	nastojascej
from	self's	first	true
i	neznoj	ljubvi
and	tender	love

`approximately a good 20 letters from my first true and tender love'

b. dobryxstukdesjat'opernyxteatrov good.GENitems.GEN10opera(A).GENtheaters.GEN

`approximately a good 10 opera theaters'

c. dobryxstuktridcat' drugixkanalov good.GENitems.GEN30 other.GENchannels.GEN

`approximately a good 30 other channels'

The alternative analyses schematized in (45) provide no room for the classifier stuk `items' to land: since on these analyses the dobryx-type adjective is merged in the specifier of the projection headed by the numeral, there is no room between the adjective and the numeral where another functional projection (i.e., EvidP) can be tucked in. According to the analysis we propose in (46), the EvidP must be merged between aP-2 and NumP.

Conclusion

To sum up, in this paper, we have investigated syntactic and semantic properties of prenominal adjectival modifiers, focusing on Russian facts. We have argued that prenominal adjectives may appear in six distinct structural positions, and that each position correlates with certain semantic (and in some cases pragmatic) properties. The internal functional structure of a DP that we have assumed for this purpose consists of projections that are, crucially, independently motivated and used to account for additional phenomena in different languages. We have demonstrated that the syntaxsemantics interface plays an important part in the behavior of prenominal adjectives. The semantics of an adjective correlates with the structural position it occupies; further, numerous adjectives can appear in more than one position, in which case the syntactic position of an adjective determines the way in which it gets interpreted. In other words, the semantic contribution of an adjective is often determined not only by its lexical meaning but also on the basis of the syntactic position it occupies.

The present investigation has consequences for a number of additional phenomena, independently discussed in the linguistic literature. Firstly, the syntactic-semantic properties of poslednije-type adjectives provide evidence in favor of the existence of the DP projection in an article-less language like Russian, an issue that has received a considerable attention in the recent years (see [Zlatic, 1997; Progovac, 1998; Willim, 1998, 2000; Leko, 1999; Rappaport, 2001; Franks, Pereltsvaig, 2004; Trenkic, 2004; Boskovic, 2005, 2008, 2009, 2010; Pereltsvaig, 2006a, 2007, 2008; Lutikova, 2010; Boskovic, Gajewski, 2011], inter alia). A second and partially related fact is that possessive expressions in Russian do contribute an exhaustivity presupposition, contrary to what has been believed previously, although the presupposition only arises in a certain syntactic configuration. These facts suggest that Russian is much more similar to a language with articles like English than it may superficially seem to be. Thirdly, we argued, contrary to Aikhenvald (2000), that Russian has numeral classifiers, which occupy the same position as numeral classifiers in other languages do. While the use of classifiers in Russian is relatively restricted and the classifier system of this language is not very rich, the facts discussed in Section 2 point to certain similarities between Russian and the more classical classifier languages. This way, the investigation of adjectival syntax and semantics reveals a number of cross-linguistic patterns that plausibly point to universal principles governing languages with superficially different properties.

syntax semantic language adjectives



References

1. Abney, 1987 - Abney S.P. The English noun phrase and its sentential aspect. Ph.D. diss. MIT, 1987

2. Academy Grammar, 1980 - Академическая грамматика русского языка / Под ред. Н.Ю. Шведовой. М., 1980. [Akademicheskaya grammatika russkogo yazyka [Academy Grammar of the Russian language]. N.Yu. Shvedova (ed.). Moscow, 1980.]

3. Aikhenvald, 2000 - Aikhenvald A. Classifiers: A typology of noun categorization devices. Oxford, 2000.

4. Babby, 1987 - Babby L.H. Case, prequantifiers, and discontinuous agreement in Russian. Natural Language and Linguistic Theory. 1987. № 5(1). Рр. 91-138.

5. Bailyn, 1994 - Bailyn J.F. The syntax and semantics of Russian long and short adjectives: An X'-theoretic account. Annual Workshop on Formal Approaches to Slavic Linguistics. The Ann Arbor Meeting. J. Toman (ed.). Ann Arbor, 1994. Pp. 1-30.

6. Bailyn, 2004 - Bailyn J.F. The case of Q. Proceedings of FASL 12. The Ottawa Meeting 2003. O. Arnaudova, W. Browne, M.L. Rivero, D. Stojanovic (eds.). Ann Arbor, MI, 2004. Pp. 1-36.

7. Baker, 2003 - Baker M.C. Lexical Categories. Verbs, Nouns, and Adjectives. Cambridge, UK, 2003.

8. Barcott, 2011 - Barcott B. Land of the Spirit Bear. National Geographic. 2011. August. Рр. 34-53.

9. Borer, 2005 - Borer H. Structuring Sense. An exo-skeletal trilogy. Vol. 2: The Normal Course of Events. Oxford, 2005.

10. Boskovis, 2005 - Boskovis Z. On the locality of left branch extraction and the Structure of NP. Studia Linguistica. 2005. № 59 (1). Рр. 1-45.

11. Boskovis, 2008 - Boskovis Z. What will you have, DP or NP? Proceedings of the 37th Annual Meeting of the North East Linguistic Society. Elfner E., M. Walkow (eds.). Book Surge Publishing, 2008.

12. Boskovis, 2009 -Boskovis Z. More on the no-DP analysis of article-less languages. Studia Linguistica. 2009. № 63 (2). Рр. 187-203.

13. Boskovis, 2012 - Boskovic Z. On NPs and clauses. Discourse and grammar: From sentence types to lexical categories. G. Grewendorf, T.E. Zimmermann (eds.). Berlin, 2012. Pp. 179-245.

14. Boskovis, Gajewski, 2011 - Boskovic Z., Gajewski J. Semantic correlates of the DP/NP parameter. Proceedings of NELS 39. S. Lima, K. Mullin, B. Smith (eds.). Amherst, MA, 2011. Pp. 121-134.

15. Chierchia, 1995 - Chierchia G. A Note on the Contrast Individual Level vs. Stage Level Predicates in German. Temporal Reference, Aspect and Actionality. P.M. Bertinetto, V. Bianchi, J. Higginbotham, M. Squartini (eds.). Vol. 1: Semantic and Syntactic Perspectives. Rosenberg and Sellier, 1995.

16. Chierchia, 1998 - Chierchia G. Reference to Kinds Across Languages. Natural Language Semantics. 1998. № 6. Рр. 339-405.

17. Cinque, 1994 - Cinque G. Evidence for Partial N-Movement in the Romance DP. Paths Towards Universal Grammar. Studies in Honor of Richard S. Kayne. G. Cinque, J. Koster, J.-Y. Pollock, L. Rizzi, R. Zanuttini (eds.). Washington, DC, Pp. 85-110.

18. Cinque, 1995 - Cinque G. Italian syntax and Universal Grammar. Cambridge, UK, Pp. 287-309.

19. Cinque, 1999 - Cinque G. Adverbs and Functional Heads: A Cross-Linguistic Perspective. Oxford, 1999.

20. Cinque, Krapova, 2007 - Cinque G., Krapova I. A note on bulgarian numeral classifiers. Pitar Mo$: A Building With a View. Papers in Honour of Alexandra Conilescu. G. Alboiu, A.A. Avram, and D. Isac (eds.). Bucharest, 2007. Pp. 45-51.

21. Delsing, 1993 - Delsing L.-O. On attributive adjectives in Scandinavian and other languages. Studia Linguistica. 1993. № 47(2). Рр. 105-125.

22. Demonte, 2008 - Demonte V. Meaning-form correlations and adjective position in Spanish. Adjectives and Adverbs: Syntax, Semantics, and Discourse. L. McNally, C. Kennedy (eds.). Oxford, 2008. Pp. 71-100.

23. Engelhardt, Trugman, 1998 - Engelhardt M., Trugman H. D as a Source of Adnominal Genitive in Russian. Formal Approaches to Slavic Linguistics. The Connecticut Meeting 1997. Z. Boskovic, S. Franks, W. Snyder (eds.). Ann Arbor, MI, 1998. Pp. 114-133.

24. Franks, Pereltsvaig, 2004 - Franks S., Pereltsvaig А. Functional categories in the Nominal Domain. Proceedings of FASL 12. The Ottawa Meeting 2003. O. Arnaudova, W. Browne, M.L. Rivero, D. Stojanovic (eds.). Ann Arbor, MI, 2004. Pp. 109-128.

25. Geist, 2010 - Geist L. The argument structure of predicate adjectives in Russian. Russian Linguistics. 2010. № 34 (3). Рр. 239-260.

26. Giusti, 2002 - Giusti G. The functional structure of Noun Phrases: A bare phrase structure approach. Functional Structure in DP and IP. The Cartography of Syntactic Structures. G. Cinque (ed.). Vol. 1. Oxford, 2002. Pp. 54-90.

27. Gutierrez-Rexach, Mallen, 2002 - Gutierrez-Rexach J., Mallen E. Toward a unified minimalist analysis of pronominal adjectives. Structure Meaning and Acquisition in Spanish. Papers from the 4^th Hispanic Linguistics Symposium. C. Clements (ed.). Somerville, MA, 2002. Pp. 178-192.

28. Kagan, 2011 - Kagan O. On Speaker Identifiability. Journal of Slavic Linguistics. 2011. № 19 (1). Рр. 47-84.

29. Kagan, Alexeyenko, 2011 - Kagan O., Alexeyenko S. Degree modification in Russian morphology: The case of the suffix -ovat. Proceedings of IATL 26. N. Boneh (ed.). Jerusalem, 2011.

30. Kagan, Alexeyenko, in progress - Kagan O., Alexeyenko S. On the semantics of long and short adjectives in Russian. Ms. Ben-Gurion University of the Negev and University of Osnabruck. In progress.

31. Kagan, Pereltsvaig, 2011 - Kagan O., Pereltsvaig А. Syntax and semantics of bare NPs: Objects of intensive reflexive verbs in Russian. Empirical Issues in Syntax and Semantics 8. Bonami O., Cabredo Hofherr P. (eds.). Paris, 2011. Pp. 221-238.

32. Kratzer, 1989 - Kratzer A. Stage-level and individual-level predicates. Papers on Quantification. E. Bach, A. Kratzer, B. Partee (eds.). University of Massachusetts, Amherst, 1989.

33. Kratzer, 1995 - Kratzer A. Stage-level and individual-level predicates. The Generic Book. G.N. Carlson, F.J. Pelletier (eds.). Chicago, 1995. Pp. 125-175.

34. Kratzer, 1996 - Kratzer A. Severing the external argument from its verb. Phrase Structure and the Lexicon. J. Rooryck, L. Zaring (eds.). Dordrecht, 1996. Pp. 109-138.

35. Kuipers, 1967 - Kuipers A.H. The Squamish language: Grammar, texts, dictionary. The Hague, 1967. (Janua linguarum. Series practica, 73.)

36. Leko, 1999 - Leko N. Functional categories and the structure of the DP in Bosnian. Topics in South Slavic Syntax and Semantics. M. Dimitrova-Vulchanova, L. Hellan (eds.). Amsterdam, 1999. Pp. 229-252.

37. Lyutikova, 2010 - Лютикова Е.А. К вопросу о категориальном статусе именных групп в русском языке // Вестник Московского университета. Серия 9: Филология. № 6. С. 36-76. [Lyutikova Е.А. Towards the Question of Categorial Status of Noun Groups in Russian. Bulletin of Moscow University. Series 9: Philology. № 6. Рр. 36-76.]

38. McCoy, 1998 - McCoy S. Individual-level predicates and pronoun doubling in colloquial Russian. Formal Approaches to Slavic Linguistics. The Connecticut Meeting 1997. Z. Boskovic, S. Franks, W. Snyder (eds.). Ann Arbor, MI, 1998. Pp. 231--251.

39. McNally, 1993 - McNally L. Adjunct predicates and the individual/stage distinction. The Proceedings of the Twelfth West Coast Conference on Formal Linguistics. Е. Duncan, D. Farkas, P. Spaelti (eds.). Stanford, CA, 1993. Pp. 501-576.

40. Mel'chuk, 1985 - Mel'chuk I.A. Surface syntax of Russian numeral expressions. Wiener Slawischer Almanach. 1985. Sonderband 16.

41. Partee, 2006 - Partee B.H. A note on Mandarin possessives, demonstratives, and definiteness. Drawing the Boundaries of Meaning: Neo-Gricean Studies in Pragmatics and Semantics in Honor of Laurence R. Horn. B.J. Bimer, G. Ward (eds.). Amsterdam, 2006. Pp. 263-280.

42. Pereltsvaig, 2001 - Pereltsvaig A. On the nature of intra-clausal relations: A Study of copular sentences in Russian and Italian. Ph.D. diss. McGill University, 2001.

43. Pereltsvaig, 2006а - Pereltsvaig A. Small nominals. Natural Language and Linguistic Theory. 2006. № 24. Рр. 433-500.

44. Pereltsvaig, 2006b - Pereltsvaig A. Passing by Cardinals: In support of head movement in nominals. Formal Approaches to Slavic Linguistics 14: The Princeton Meeting. J.E. Lavine, S. Franks, H. Filip (eds.). Ann Arbor, MI, 2006. Pp. 277-292.

45. Pereltsvaig, 2007 - Pereltsvaig A. On the universality of DP: A view from Russian. Studia Linguistica. 2007. № 61(1). Рр. 59-94.

46. Pereltsvaig, 2008 - Pereltsvaig A. Split phrases in colloquial Russian. Studia Linguistica. 2008. Special volume on spoken language. № 62 (1). Рр. 5-38.

47. Pereltsvaig, 2011a - Pereltsvaig A. On number and number-neutrality in languages with and without articles. Paper presented at the Workshop on languages with and without articles, Paris, March 2011.

48. Pereltsvaig, 2011b - Pereltsvaig A. As easy as two, three, four? Formal Approaches to Slavic Linguistics 18: The Second Cornell Meeting. W. Browne, A. Cooper, A. Fisher, E. Kesici, N. Predolac, D. Zec (eds.). Ann Arbor, MI, 2011. Pp. 417-434.

49. Progovac, 1998 - Progovac L. Determiner phrase in a language without determiners. Journal of Linguistics. 1998. № 34. Рр. 165-179.

50. Rapoport, 1991 - Rapoport T.R. Adjunct-predicate licensing and D-structure. Syntax and Semantics. 1991. № 25. Рр. 159-187.

51. Rappaport, 1992 - Rappaport G.C. On the adnominal genitive and the structure of noun phrases in Russian and Polish. Melanges Paul Garde. M. Guiraud-Weber (ed.). Aix-en-Provence, 1992. Pp. 239-262.

52. Rappaport, 2001 - Rappaport G.C. Extraction from nominal phrases in Polish and the theory of determiners. Journal of Slavic Linguistics. 2001. № 8(3). Рр. 159-198.

53. Rutkowski, 2002a - Rutkowski P. Noun/pronoun asymmetries: Evidence in support of the DP hypothesis in Polish. Jezikoslovlje. 2002. № 3. Рр. 159-170.

54. Rutkowski, 2002b - Rutkowski P. The syntax of quantifier phrases and the inherent vs structural case distinction. Linguistic Research. 2002. № 7. Рр. 43-74.

55. Rutkowski, 2006a - Rutkowski P. Why Polish numerals should not be analyzed as nouns. Minimalist Views on Language Design. C. Yim (ed.). Seoul, 2006. Pp. 249-263.

56. Rutkowski, 2006b - Rutkowski P. The syntax of floating intensifiers in Polish and its implications for the Determiner Phrase hypothesis. Proceedings of the BLS-32. Z. Antic et al. (eds.). UC Berkeley, 2006. Pp. 321-333.

57. Rutkowski, 2007 - Rutkowski P. The Determiner Phrase hypothesis as a tool of syntactic analysis of Polish nominal phrases. Ph.D. diss. Warsaw University, 2007.

58. Rutkowski, Maliszewska, 2007 - Rutkowski P., Maliszewska H. On prepositional phrases inside numeral expressions in Polish. Lingua. 2007. № 117 (5). Рр. 784-813.

59. Sadler, Arnold, 1994 - Sadler L., Arnold D.J. Prenominal adjectives and the Phrasal/Lexical Distinction. Journal of Linguistics. 1994. № 30. Рр. 187-226.

60. Scott, 2002 - Scott G.-J. Stacked adjectival modification and the structure of nominal phrases. Functional Structure in DP and IP. The Cartography of Syntactic Structures. G. Cinque (ed.). Vol. 1. Oxford, 2002. Pp. 91-122.

61. Soschen, 2001 - Soschen A. On the distribution of Copular Elements in Hebrew, Russian, and Spanish. Ms. University of Southern California. 2001.

62. Sproat, Shih, 1988 - Sproat R., Shih C. The cross-linguistic distribution of adjective ordering restrictions. Interdisciplinary Approaches to Language. C. Georgopoulos, R. Ishihara (eds.). Dordrecht, 1988. Pp. 565-593.

63. Sproat, Shih, 1991 - Sproat R., Shih C. Prenominal adjectival ordering in English and Mandarin. NELS. 1991. № 18.Рр. 465-489.

64. Svenonius, 1993 - Svenonius P. The structural location of the attributive adjective. The Proceedings of the Twelfth West Coast Conference on Formal Linguistics. E. Duncan, D. Farkas, P. Spaelti (eds.). Stanford, CA, 1993. Pp. 439-454.

65. Svenonius, 2008 - Svenonius P. The position of adjectives and other phrasal modifiers in the decomposition of DP. Adjectives and Adverbs: Syntax, Semantics, and Discourse. L. McNally, C. Kennedy (eds.). Oxford, 2008. Pp. 16-42.

66. Timberlake, 2004 - Timberlake A. A Reference Grammar of Russian. Cambridge, UK, 2004.

67. Trenkic, 2004 - Trenkic D. Definiteness in Serbian/Croatian/Bosnian and some implications for the general structure of the nominal phrase. Lingua. 2004. № 114. Рр. 1401-1427.

68. Truswell, 2004 - Truswell R. Attributive Adjectives and the Nominals they Modify. Master's thesis. Oxford University, 2004.

69. Willim, 1998 - Willim E. On the DP-hypothesis in Polish, an articless language. Projections and Mapping: Studies in Syntax. P. Stalmaszczyk (ed.). Lublin, 1998. Pp. 137-158.

70. Willim, 2000 - Willim E. On the grammar of Polish nominals. Step by Step: Essays on Minimalist Syntax in Honor of Howard Lasnik. R. Martin, D. Michaels, J. Uriagereka (eds.). Cambridge, MA, 2000. Pp. 319-346.

71. Yadroff, 1999 - Yadroff M. Formal Properties of Functional Categories: The minimalist syntax of Russian Nominal and Prepositional Expressions. Ph.D. diss. Indiana University Bloominton, 1999.

72. Zamparelli, 2000 - Zamparelli R. Layers in the Determiner Phrase. New York, 2000.

73. Zavala, 2000 - Zavala R. Multiple classifier systems in Akatek (Mayan). Systems of Nominal Classification. G. Senft (ed.). Cambridge, 2000. Pp. 114-146.

74. Zlatic, 1997 - Zlatic L. The Structure of the Serbian Noun Phrase. Ph.D. diss. University of Texas at Austin, 1997.

Каган Ольга Александровна - Ph.D. (лингвистика); старший преподаватель кафедры иностранной литературы и лингвистики, Университет имени Давида Бен-Гуриона в Негеве, г. Беэр-Шева, Израиль

Kagan Olga А. - Ph.D. (Linguistics); senior lecturer at the Department of Foreign Languages and Linguistics, Ben-Gurion University of the Negev, Beer-Sheva, Israel

Перельцвайг Ася Михайловна - Ph.D. (лингвистика); преподаватель Института непрерывного обучения им. Ошера, Университет Санта-Клары, Соединенные Штаты Америки

Pereltsvaig Asya М. - Ph.D. (Linguistics); lecturer at the Osher Lifelong Learning Institute, Santa Clara University, CA, USA

Размещено на Allbest.ru

...