From HAVE to HAVE-verbs: relations and incorporation Bert Le Bruyn (Utrecht Institute of Linguistics) [email protected] Joint work with: Henriëtte de Swart, Joost Zwarts, Kirsten Schutter & Xiaoli Dong (Utrecht Institute of Linguistics) Le Bruyn, B., De Swart, H., & Zwarts, J. (2016). From HAVE to HAVE-verbs: Relations and incorporation. Lingua, 182, 49-68. Le Bruyn, B., Schutter, K. (2017). Existential verbs, article drop and non-presuppositional definites. [manuscript] 1. Pseudo-Incorporation with HAVE-verbs 1.1. Introduction Borthen (2003) was the first to point out that verbs like HAVE, BUY, WEAR, … come with the peculiarity of allowing for bare nominal (count) objects in a number of languages that normally impose the use of articles in argument position (unlike, e.g., Mandarin) and don’t have a generalized incorporation option (unlike, e.g., Hungarian). (1)

Han hadde rød he had red ‘He had a red coat.’

ytterfrakk. coat

(Norwegian, Borthen 2003)

(2)

Forouse frako. s/he-wears frock coat ‘He was wearing a frock coat.’

(Greek, Alexandropoulou 2013)

(3)

Ion are copil. Ion has child ‘Ion has a child.’

(Romanian, Dobrovie-Sorin et al. 2006)

(4)

Lleva sombrero. s/he-wears hat ‘S/he wears a hat.’

(Spanish, Espinal & McNally 2011)

(5)

M’acabo de comprar REFL.finish.1SG of buy ‘I’ve just bought myself a car. ’

cotxe. (Catalan, Espinal & McNally 2011) car

The literature seems to have converged on dubbing the verbs in question ‘HAVE-verbs’ because of the dimension of possession they all seem to come with. For the time being, we will ignore issues of pragmatics and productivity. They will however return in the course of the talk. Previous analyses have focused on formalizing the semantic and syntactic effects HAVE-verbs give rise to, but haven’t tried to explain why HAVE-verbs lend themselves to (pseudo) incorporation in contrast 1

to other verb classes. For illustration purposes, we briefly present the analysis of Espinal & McNally (2011) but the same holds for those presented in Borthen (2003) and Dobrovie-Sorin et al. (2006). Espinal & McNally (2011) propose that HAVE-verbs are just like regular transitive verbs except for the fact that they can undergo a special shift that allows them to function as incorporation verbs (henceforth IVs): (6)

Input

yxe[V(e)&Agent(e)=x&(e)=y&e’[depend(e,e’)&have(e’)&havee(e’)=y]]

Output Pxe[V(e)&Agent(e)=x&P((e))&e’[depend(e,e’)&have(e’)&havee(e’)=(e)]] where (e) represents the theme of event e The input in (6) is a regular transitive verb meaning that comes with a dimension of possession, made explicit in the requirement that the event e depends on an event e’ that is explicitly linked to a haverelation. In the output, the theme argument variable y has been suppressed. It only surfaces indirectly as (e). If we eliminate the technical complication of theme suppression as well as the event notation, and separate the actual input from the restriction that was built into it, we obtain the basic version of the shifting rule proposed by Espinal & McNally (2011): (7)

Input

yx(V(y)(x))

Output

Pxy(V(y)(x)&P(y))

Restriction

This rule can only be applied to HAVE-verbs

The lexical rule in (7) takes a transitive verb and turns it – through existential closure of its internal argument – into a verb that selects predicates rather than arguments. The restriction accounts for the empirical generalization that HAVE-verbs can function as IVs, but does not explain why HAVE-verbs and not verbs like to like or to see behave this way. It’s this why question that we will try to answer. 1.2. The gist of our analysis: a first foray into the semantics of HAVE There’s a long-standing intuition about (at least one version of) HAVE, viz. that it has little or no content of its own: (8)

John has a child.

In their analyses of (8), semanticists have mainly focused on two challenges: (i) making John into the first argument of child in a compositional fashion, (ii) deriving the ban on definite objects HAVE comes with (see (9)). (9)

*John has the child.

Here are some of the analyses that have been proposed: > Landman & Partee (1987) come up with a non-standard analysis of a that passes on the relational argument and make HAVE semantically empty except for the introduction of an existential restriction accounting for the unacceptability of (9). > Partee (1999) repeats the analysis but gets rid of the non-standard version of a that she replaces by quantifying-in the subject. 2

> Saebo (2009) works the existential restriction into a small clause following the relational noun and turns have into an expression that takes care of abstracting over the relational argument of child. > Landman (2004) assumes – in line with his adjectival analysis of indefinites – that a is semantically empty. For HAVE he assumes it incorporates the meaning of the relational noun while inverting its argument structure and adding existential quantification to its sortal argument (a process that Landman relates to passivization). Let’s zoom in on the analysis of Landman (2004), simplifying it a bit: (10)

Ryx(R(y)(x))

What happens is that HAVE takes the relation R as its argument, existentially closes off the relation’s sortal argument x and makes the relation’s relational argument y available for the subject to bind. Whether or not we follow Landman in adopting an adjectival analysis of a doesn’t matter. If we do, we immediately make it semantically void. If we adopt a standard quantifier semantics instead, we can add a BE type-shift to obtain the same effect. By building existential closure into HAVE we also create an existential environment that can be exploited to account for the unacceptability of (9). The reason HAVE needs to be as in (10) is that it has no semantic content of its own and needs the relational predicative input of its object. It’s consequently the very nature of HAVE that makes it into a verb that prefers predicates over arguments. If a language like English imposes the use of an indefinite article, this is nothing more than a syntaxsemantics mismatch where Ds are generalized in argument position without the semantics requiring it. Incorporation languages are then ‘cleaner’ in the sense that the syntax can follow the semantics more closely: (11)

Ion are copil. Ion has child ‘Ion has a child.’

(Romanian, Dobrovie-Sorin et al. 2006)

We further note that building existential closure into HAVE also immediately accounts for a previously unnoticed peculiarity of HAVE, viz. that its object cannot take wide scope. (12) – on the relational reading of HAVE – can only mean that John is childless: (12)

John doesn’t have a child.

This is strongly reminiscent of one of the basic properties of incorporated nominals, viz. their obligatory narrow scope. 1.3. Plan for the remainder of the talk We have proposed two intuitions: > HAVE-verbs are special in that they require the predicative input of their objects; > Languages may vary at the surface (enforcing the presence of an indefinite article or not) but they actually all allow for incorporation with HAVE-verbs. We have argued for this intuition for: > HAVE combined with relational nouns.

3

The remainder of the talk will complete the argumentation by focusing on three follow-up questions: (i) Why is incorporation with HAVE not limited to classical relational nouns? >> section 2: The dynamic semantics of HAVE (ii) How does the basic intuition according to which HAVE-verbs need the predicative relational input of their objects work out for other verbs than HAVE? >> section 3: The semantics and pragmatics of HAVE-verbs (iii) How universal is the semantics we propose for HAVE-verbs? >> section 4: HAVE-verbs beyond incorporation languages

2. The dynamic semantics of HAVE 2.1. Non-relational nouns and implicit relational arguments One immediate complication for our intuition that HAVE needs the predicative relational input of its object noun – and that this is why it allows for incorporation – is that HAVE also allows for incorporation with seemingly non-relational nouns: (13)

Ion are blog.

ROMANIAN

Ion has blog To remedy, we follow Qualia Theory and assume nouns can come with implicit arguments next to their explicit ones. A noun like ‘blog’ e.g. comes with a ‘creator’ argument. There’s an extensive literature on implicit arguments (survey in Bhatt & Pancheva 2006, follow-up work in Francez 2010, Landau 2010, Merchant 2013) but little work on implicit arguments of noneventive nouns and even less on how to formalize these in compositional semantics. We propose to formalize them in Dynamic Predicate Logic as dynamically existentially closed arguments. This means that they are saturated but that they can – in principle – still be activated (unlike statically existentially closed arguments). (14)

[[blogdynamic]]= xEdi(blog-written-by(di)(x)) [short for: xEdi(blog(x); written-by(di)(x))]

We dub the operation of activation explicitation (EXPL): (15)

[[EXPL(P)]]=

xy(P(y);dnx) for any one-place predicate P including the implicit argument dn where n ranges over i, ii, iii, iv, …

EXPL consists in equating the existentially closed argument with another variable that is abstracted over. The combination of EXPL with blog gives rise to (16): (16)

[[EXPL(blogdynamic)]]= yxEdi(blog-written-by(di)(x); diy)

This EXPL operation can be built into HAVE and this would give us the following semantics for HAVE: (17)

Pz(Ed1(EXPL(P))(z)(d1)) 4

(17) is a dynamic version of (10) that builds the EXPL operation into HAVE and thus allows HAVE to combine with implicitly relational nouns and probe their relational content. For (13), this would mean that we can combine blog with are and Ion and that it will mean that Ion is the creator of a blog. This is worked out in (18): (18)

[[Ion are blog]]

ROMANIAN

[[Ion have blog]]= [[blog]]=

xEdi(blog-written-by(di)(x))

[[have blog]]=

Pz(Ed1(EXPL(P))(z)(d1))

(-application)

z(Ed1(EXPL(xEdi(blog-written-by(di)(x))))(z)(d1))

(explicitation)

z(Ed1(vw(Edi(blog-written-by(di)(w));div))(z)(d1))

(-conversion)

z(Ed1(Edi(blog-written-by(di)(d1));diz))

[[Ion]] =

Ion

[[Ion have blog]] =

z(Ed1(Edi(blog-written-by(di)(d1));diz)) (Ion)

(-application)

Ed1(Edi(blog-written-by(di)(d1));diIon))

(statically)

x(blog-written-by(Ion)(x))

(xEdi(blog-written-by(di)(x)))

2.2. HAVE and relational nouns Whether or not we need to propose a separate variant of HAVE for relational nouns, depends on whether or not we assume relational nouns come with explicit or implicit relational arguments. In the former case, we’re dealing with expressions of type >, in the latter with expressions of type . Nothing crucial hinges on this choice. This being said, we defend that relational nouns – in the languages we are looking at – are implicitly and not explicitly relational. > It is difficult – if not impossible – to design a linguistic test to distinguish between classical relational nouns and nouns with an implicit relational argument: (19)

the child of John

(20)

*the day of John

(21)

the blog of John (Watson)

> Most relational nouns don’t impose the realization of their arguments. This is different in languages like Daakaka (von Prince 2012) where relational nouns do impose the realization of their relational arguments. (22)

I saw the branch (of the tree).

(23)

*… beke

DAAKAKA

Branch Intended ‘… the branch’ (24)

for *(John’s) sake

5

> Maintaining that relational nouns are of type > forces us to assume that type-shifting is less constrained than we normally assume it is. In particular, we would have to assume that non-meaningpreserving type-shifts apply freely to account for the intended reading of mother in (25) as ‘mother that Jane is taking care of’: (25)

[Context: Mary and Jane work in a team hired to guide young mothers through the first steps of being a mother] Mary to Jane: Have you already spoken to your mother today?

To account for the relevant reading of mother in (25) on an > analysis of relational nouns, one would have to assume that mother is spontaneously shifted to a non-relational interpretation and then shifted back to a relational > interpretation. No such free non-meaning-preserving type-shifts would be required if we assume mother to have an implicitly relational interpretation. The possessive your would then trigger either an application of EXPL (leading to the ‘standard’ mother reading) or an application of an operation of relationalization (REL) introducing the pragmatically defined ‘mother that Jane is taking care of’ reading. > Maintaining that relational nouns are of type > would also complicate the derivation of (26): (26)

John has the only sweet brother.

Only has to apply to an version of sweet brother in order to have a shot at getting the default reading of (26) according to which the only sweet brother refers to the only sweet brother of anyone. In order to be able to access the relational argument of brother afterwards, one would however have to quantify-in John, leading inevitably to an interpretation in which the only sweet brother is interpreted as the only sweet brother of John. An implicitly relational analysis of brother would avoid this problem in the way sketched below. We start by spelling out a dynamic version of only that combines with one-place predicates: (27)

[[only]] = Px(P(x);A d2(P(d2)d2x))

We now combine this with the semantics we assume for sweet brother: (28)

[[sweet brother]] =

yEdi(sweet-brother-of(di)(y))

[[only sweet brother]] = Px(P(x);A d2(P(d2)d2x))(yEdi(sweet-brother-of(di)(y))) (-conversion)

x(Edi(sweet-brother-of(di)(x));A d2(Edii(sweet-brother-of(dii)(d2))d2x))

(statically)

x(y(sweet-brother-of(y)(x))&z(v(sweet-brother-of(v)(z))zx))

What we have obtained now is an expression that denotes the singleton set containing the only sweet brother in the model. We now continue to let HAVE add existential closure over x and to equate di with John. The semantic contribution of the is spelled out in (iii) but we don’t include it in the full derivation as it would trigger a further (standard) type-shift without adding anything beyond the uniqueness contribution of only. PQEd1(P(d1);A d2(P(d2)d2d1);Q(d1))

(29)

[[thenon-presuppositional]]=

(30)

[[have only sweet brother]]= Pz(Ed1(EXPL(P))(z)(d1)) (x(Edi(sweet-brother-of(di)(x));A d2(Edii(sweet-brother-of(dii)(d2))d2x)))

6

 (-conversion) z(Ed1(EXPL(x(Edi(sweet-brother-of(di)(x));A d2(Edii(sweet-brother-of(dii)(d2))d2x))))(z)(d1))

 (explicitation) z(Ed1(xy(Edi(sweet-brother-of(di)(y));Ad2(Edii(sweet-brother-of(dii)(d2))d2y)));dix)(z)(d1))

 (-conversion) z(Ed1(Edi(sweet-brother-of(di)(d1));Ad2(Edii(sweet-brother-of(dii)(d2))d2d1);diz))

[[John have only sweet brother]]= z(Ed1(Edi(sweet-brother-of(di)(d1));Ad2(Edii(sweet-brother-of(dii)(d2))d2d1);diz)) (John)

 (-conversion) Ed1(Edi(sweet-brother-of(di)(d1));Ad2(Edii(sweet-brother-of(dii)(d2))d2d1);diJohn))

 (statically) x(sweet-brother-of(John)(x))&y(z(sweet-brother-of(z)(y))y=x)) Dynamically, the meaning of John has the only sweet brother is that there is an individual d1 who stands in the sweet brother relation to another individual who is identical to John. We furthermore know that there is no individual who is different from d1 and also stands to someone in the sweet brother relation. Statically, this is equivalent to John’s sweet brother being the only one in the model, the desired interpretation. One aspect of the derivation deserves closer attention, viz. the explicitation step in which x is equated with di. At first sight it would seem that x could be equated with di or dii. If this were the case, one might call into question the analysis as it’s only the equation with di that leads to the correct interpretation. Fortunately, the first impression is deceiving as dii appears in the scope of a universal operator and is consequently not available for dynamic pickup. This reduces the relevant candidates to just one – di – and makes sure our analysis gets the right result. 3. The semantics and pragmatics of HAVE-verbs 3.1. Preliminaries: Qualia theory and implicit relational arguments Qualia Theory is one of the constructs of Generative Lexicon Theory. The latter takes nouns in the lexicon to come with an Argument Structure – their regular entry – and a Qualia Structure. We illustrated this before in (14) for blog: xEdi(blog(x); written-by(di)(x))] _______ _________________ Argument Structure

Qualia Structure

The idea behind Qualia Structure is that the lexicon should provide information about nouns that extends beyond classical entries and that this information conforms to a general format consisting of four perspectives on objects, known as roles. Here we provide what we think is the most common way of presenting roles with – wherever available – the implicit relational arguments they generate.

7

- The constitutive role specifies what the objects denoted by the noun consist of. A book can, e.g., be said to consist of pages and a cover. The implicit arguments that can be derived from this role are those that stand in the part-of relation to the noun, the part-of argument. - The formal role specifies the position of a noun within a taxonomy. A noun like book could, e.g., be classified as an artifact. No implicit arguments come with this role. - The telic role specifies what the objects denoted by the noun are designed for. A book can, e.g., be said to be designed to be read by someone. The implicit argument this role generates is the individual that will put the object denoted by the noun to its intended use, the user argument. - The agentive role specifies the creator of the objects denoted by the noun. A book can, e.g., be said to have been written by someone. The implicit argument to be derived from this role is the individual that has created the object denoted by the noun, the creator argument. Qualia Theory holds that these roles should be included in the lexical semantics of the noun. Empirical support comes from the observations in (31) and (32). Qualia Theory was originally designed on the basis of English, but the facts that motivate it carry over to other languages: (31)

(32)

a.

John started a new book.

b.

Ion

a

început o

nouă

carte.

Ion

has

started a

new

book

ROMANIAN

a.

I really want to buy this house but I don’t like the front door.

b.

Eu

chiar

vreau să

cumpăr această casă,

I

really

want

to

buy

place

ușa

din

față.

of

front

pleases door-the

this

dar

house, but

nu-mi NEG-me ROMANIAN

By including both the agentive and the telic role, Qualia Theory can derive the systematic ambiguity of (31a) and (31b), which can mean that John/Ion started a new book as an author/creator or as a reader/user. (32a) and (32b) demonstrate that the definite the front door/ușa din față can be felicitous without previous mention. This follows if we assume front door/ușa din față is included in the constitutive role of house/casă. Qualia Theory hasn’t been specifically designed to deal with the domain of implicit relational arguments, so it shouldn’t come as a surprise that – as it stands – it has been claimed to be too powerful and too weak (see, e.g., Asher & Denis 2005). We use HAVE and prenominal genitives as our baseline for identifying relations included in nouns. For prenominal genitives, we stick to relations that 8

can be identified out of the blue, thus making sure that we are not talking about contextually supplied relations. The oddness of (33) an (34) illustrates the apparent overgeneration of Qualia Theory: (33)

?The door has a house.

(34)

?the door’s house

Under the assumption that the constitutive role generates an implicit door argument for house, we would expect (33) and (34) to be felicitous – contrary to fact. We however don’t take these examples to be an argument against Qualia Theory as a valid theory for implicit relational arguments. Instead, we conjecture that the relative unacceptability of (33) and (34) is similar to the one we find in the prepositional domain for (35): (35)

?the tree under the ant

The unacceptability of (35) is not semantic but pragmatic: in principle we locate smaller things with respect to bigger things and not the other way around. The unacceptability of (33)/(34) and (35) is not different: relating a house to a door is as pragmatically odd as locating trees with respect to ants. We conclude that (33) and (34) don’t show that Qualia Theory overgenerates as a theory of implicit relational arguments but that the constitutive role simply doesn’t give rise to relations that can easily be exploited by HAVE or prenominal genitives. Examples where Qualia Theory would appear to undergenerate are the following: (36)

John’s keys

(37)

the tree’s leaves

Qualia Theory – as it stands – does not provide the necessary means to interpret (36) as being about the keys that belong to John, nor to interpret (37) as being about the leaves that are part of the tree. One could of course argue that these are meaning components that should be included in the possessive marker but there’s also good reason to locate them in the noun and therefore add two extra roles to Qualia Structure, the possessor role and the holistic role, giving rise to the corresponding implicit arguments. The rationale behind enriching Qualia Structure with the possessor and the holistic role is that both generate relations that are more basic than the use relation the telic role generates. Keys are supposed to be used to open doors, but in order for this to be possible one first has to (temporarily) possess them. In a similar fashion, leaves are put to work by trees to take care of photosynthesis but this is only possible if they are actually part of the tree. The fact that the holistic and possessor role can be seen as underlying the telic role argues in favor of their inclusion in Qualia Structure. A full overview of the roles we assume is given in (38). The ones giving rise to relations that we expect to be exploited in relational constructions are marked by an asterisk. (38)

formal constitutive telic* agentive* holistic* possessor* 9

A small caveat is in place for relations probed by HAVE, viz. that HAVE comes with a built-in restriction to static relations. This means that a sentence like John has a blog can refer to John having created a blog but not to John creating a blog nor to John reading a blog, unlike John’s blog. We leave out this complication in the rest of the paper as we see no deep reason to expect this restriction to carry over to HAVE-verbs in general. 3.2. The gist of the analysis In section 2, we developed an analysis of HAVE that motivated its incorporation behavior on the basis of its lexical semantics. The question we turn to now is whether and how this analysis extends to verbs like to wear, to buy, to want, which have been classified as HAVE verbs in the incorporation literature. The question has to be qualified. Under the reasonable assumption that languages are economic in that they avoid having redundant lexical items, we do not expect to find more than one lexical item that instantiates HAVE and recycles the relation included in its object noun. The more interesting question is therefore which verbs are such that the relations they encode can (at least in part) be reconstructed based on the relational content of their object nouns. If such verbs exist, we could expect them to come with: - a regular version in which they take arguments and encode a fully specified relation and - a light version in which they take predicates and depend on the latter’s lexical semantics to specify (part of) the relation. One could e.g. think of a verb like WRITE whose regular entry can be formalized as in (39) while its light entry can be formalized as in (40): (39)

[[WRITEregular]]= xy(write(x)(y))

(40)

[[WRITEIV]]=

Pz(Ed1(EXPLcreate-by-writing(P))(z)(d1))

The operation of explicitation in (40) is sorted in the sense that it will not pick out just any implicit argument but only a creator argument that was involved in a create-relationship that involves writing. In (41), we show how the combination of (40) with blog gives us the same information as (39) would give us: (41)

[[WRITEIV]]=

Pz(Ed1(EXPLcreate-by-writing(P))(z)(d1))

[[WRITEIV blogdynamic]]= Pz(Ed1(EXPLcreate-by-writing(P))(z)(d1)) xEdi(blog-written-by(di)(x)) (-app/conversion)

z(Ed1(Edi(blog-written-by(di)(d1));diz))

(statically)

zx(blog-written-by(z)(x))

Entries like the one in (40) are the type of entries we expect HAVE-verbs to come with: in the same way as HAVE, they depend on the predicative relational information included in their object nouns. As should be clear from the above examples, the different qualia are predicted to be a good source of relational information that can be exploited by light versions of incorporation verbs. The same is unlikely to hold for the relational information included in classical relational nouns: most relations expressed by classical relational nouns are very specific and it would probably be hard to find a verb denoting a relation that can be reconstructed based on the relations included in these nouns. The lack of any real semantic content of HAVE seems to make it into the one verb that can specialize in combining with these relations. 10

We consequently predict incorporation verbs to express relations that can be reconstructed based on the relational information included in Qualia Structure. 3.3. HAVE-verbs: their identity and semantic behavior In section 3.2 we argued that the class of HAVE-verbs is limited to those verbs that express relations that can be reconstructed based on the relational information included in Qualia Structure. (42)

Relations we predict HAVE-verbs to be able to express - use (
As we indicated before, we assume all HAVE-verbs except HAVE come with some lexical content. This means they don’t have to function as IVs and are also likely to have regular entries next to their incorporation entries. Whether or not verbs that fall in these classes behave as IVs consequently remains a lexical matter and cross-linguistic variation in which verbs turn out to behave as IVs cannot be excluded. The list in (42) consequently functions as an upper bound. An overview of the different classes of HAVE-verbs that have been reported in the literature shows that the list in (42) makes the right predictions. Borthen (2003) points out that Norwegian possession/ownership verbs (e.g., have), usage verbs (e.g., wear), transfer of possession or ownership verbs (e.g., receive, give, buy) and a subclass of intensional verbs based on possession (need, want) allow for bare singular count objects. Dobrovie-Sorin et al. (2006) look into Romanian, Catalan and Spanish and mention have and acquisition verbs as well as some intensional and usage verbs like to look for and to use. Espinal & McNally (2011) and Borik et al. (2012) zoom in on Spanish/Catalan and Brazilian Portuguese respectively and mention the same verb classes as Dobrovie-Sorin et al. (2006). Finally, Lazaridou-Chatzigoga (2011) turns to Greek and reports that the same verbs as the ones found for Norwegian and Romance languages allow for bare objects in Greek as well. Two extra classes of verbs that have been reported to show IV behavior in Norwegian and Greek but that – up till now – haven’t been categorized as HAVE-verbs are creation and consumption verbs (like build and eat) (cf. Borthen 2003, Lazaridou-Chatzigoga 2011). Some of the verb classes reported were already exemplified in (1) to (5). Here, we provide examples of the other ones: intensional verbs (43-45), creation verbs (46-48), and consumption verbs (49). (43)

(44)

(45)

(46)

Juan

busca

secretaria.

Juan

looks-for

secretary

Per

a

aquest espectacle

necessitareu

for

to

this

event

you-will-need skirt

Ion

doreşte

nevastă

tânără.

Ion

wants

wife

young

stin

Costa Brava.

Htizi

spiti

(Spanish, Dobrovie-Sorin et al. 2006)

She-is-building house in-the Costa Brava 11

faldilla llarga. (Catalan, Espinal 2010) long

(Romanian, Dobrovie-Sorin et al. 2006)

(Greek, Lazaridou-Chatizgoga 2011)

(47)

Graphi

ghrama

sti

She-is-writing letter (48)

(49)

Han

striker genser.

He-is

knitting sweater

Idha

mia

afisa

ghrafomihani. (Greek, Alexandropoulou 2013)

at-the typewriter. (Norwegian, Borthen 2003)

me

ena

emvrio pu

kapnizi

tsigharo.

a

fetus

smokes

cigarette

(Greek, Alexandropoulou 2013)

I-saw

a

poster with

that

It is easy to verify that intensional verbs like NEED and WANT as well as creation and consumption verbs nicely fall into the classes of verbs we expect to be able to function in the same way as HAVE, each of them entailing or being entailed by a relation that is generated by Qualia structure. As we indicated before, we don’t expect every language to have both types of entries for each verb. This is a language-specific lexical matter. What we can provide is a way to test which entries a given verb has. To establish that a verb has an incorporation entry, one can simply look at whether it combines with bare nouns or not. The realization of OWN in Romanian, e.g., allows for bare nouns and consequently has to be assumed to have an incorporation entry: (50)

Ion

posedă buletin.

Ion

owns

ID

To test whether a given verb also has a regular entry, one should check whether it allows its object to take wide scope over negation. The realization of OWN in Romanian once more qualifies: (51)

Ion

nu

posedă un

tablou. [

A

Ion

not

own

painting.

He-has bought

a

cumpărat

mai multe several

anul

trecut, dar

unul

anume

un

l-a

putut

găsi].

year-the

past,

one

particular

not

it-has could

find.

but

‘Ion doesn’t own a painting. He bought several last year, but one in particular he couldn’t find.’ (51) allows for an interpretation according to which there is a particular painting Ion doesn’t own, despite the fact that he does have other paintings. This is a wide scope reading of un tablou over nu and indicates that poseda also has a regular entry. 3.4. HAVE-verbs: their pragmatics The type of incorporation associated with HAVE-verbs has been reported to come with a number of properties next to obligatory narrow scope of the verb’s object. We briefly outline these properties, based on the work by Espinal & McNally and discuss their status. (i) Incorporated nominals have a reduced ability to allow for anaphoric pickup of the referent of their sortal argument: (52)

Avui porta faldillai. # today she-wears skirt l’any passat. the year last

Lai it

hi her

vam we-have

regular given

(Catalan, Espinal & McNally 2011)

12

‘Today she’s wearing a skirt. We gave it to her as a present last year.’ (ii) Incorporated nominals have general number: they allow both for singular and plural interpretations. The possibility of a plural interpretation is illustrated in (53): (53)

Busco pis. Un a Barcelona i un a I-look-for apartment one in Barcelona and one in Girona. (Catalan, Espinal & McNally 2011) Girona ‘I’m looking for apartments. One in Barcelona and one in Girona.’

(iii) Incorporated nominals have restricted modification possibilities. We illustrate for individual-level adjectives (as opposed to kind-level adjectives): (54)

*

Té parella he-has partner ‘He has a tall/ill partner.’

alta/malalta. tall/ill

(Catalan, Espinal 2010)

Our analysis – as it stands – predicts none of these properties. Rather than being a shortcoming, this is actually an advantage of our analysis, as properties (i) to (iii) are not stable across languages. Work by Lazaridou-Chatzigoga and Alexandropoulou (in changing configurations) reveals that – at least for Greek – properties (i) to (iii) are not clearly attested. For (i) they note that the exact counterpart of (52) is indeed not very felicitous but that slight variations are perfectly acceptable: (55)

Foruse pukamisoi htes. Toi ihe aghorasi sti he-wore shirt yesterday it he-had bought in-the varkeloni. (Greek, Lazaridou-Chatzigoga 2011) Barcelona ‘Yesterday he had a shirt on. He had bought it in Barcelona.’

For (ii), they refer to Alexopoulou & Folli (2010) who note that (56) is infelicitous, despite the fact that it seems a simple variant of (53): (56)

Psahno aftokinito; # ena mikro ya tin poli ki ena I-look-for car one small for the city and one fortighaki ya ekdhromes. (Greek, Alexopoulou & Folli 2010) van for trips ‘I’m looking for a car; a small one for the city and a van for trips.’

For (iii) finally, they find that Greek bare nominal objects combining with IVs show no modification restrictions: (57)

Ehi psilo gomeno. S/he-has tall boyfriend ‘S/he has a tall boyfriend.’

(Greek, Alexopoulou & Folli 2010)

The comparison between (52) to (54) and (55) to (57) shows that there is cross-linguistic variation in incorporation with HAVE-verbs. One move would then be to give up the assumption that a unified analysis is possible. We however don’t think that this is a particularly insightful move and rather take the comparison to indicate that a unified analysis is possible but should not make any hard predictions about properties (i) to (iii). Our account conforms to this requirement and even comes with an extra 13

perk, viz. that it allows us to understand why properties (i) to (iii) may pop up with HAVE-verbs. This is due to the fact that we predict truth-conditional equivalence between the article and articleless variants of HAVE-verbs. This state-of-affairs is known to trigger pragmatic differentiation (Givón 1981, 1984; Blutner 2000; de Swart & Zwarts 2009) that may but need not end up being entrenched in some languages. Under the standard assumption that the indefinite article flags the introduction of a new discourse referent, we expect the full DP variant to be the preferred option in all cases in which the individual corresponding to the object is pragmatically relevant. This is obviously the case if anaphoric pickup occurs (property (i)) and in all cases individual-level modification is added to the object (property (iii)). Property (ii) follows under the assumption that the indefinite article is an explicit spell-out of number that is absent from the bare noun. We have argued that our analysis is faring well with the predictions it makes about incorporation with HAVE-verbs. In particular, we have seen that it makes semantic predictions about the stable property of narrow scope and pragmatic predictions about those properties that seem more variable. If these properties turn out to be stable in a given language, we can assume that some (or all) of the pragmatic properties have been entrenched and proceed to fine-tune our analysis with theme suppression as in Espinal & McNally (2011). We have thus succeeded in proposing an analysis that brings out what is common to a number of verb classes across languages without precluding variation.

4. HAVE-verbs beyond incorporation languages In section 3, we looked into HAVE-verbs in Romanian-type languages. What we want to do now is to take the discussion one last step further and argue that HAVE-verbs in general are not only special in Romanian-type languages but also beyond. This is the last step in arguing that HAVE-verbs have a lexical semantics that sets them apart from other verbs. 4.1. English In section 1.2. we showed that HAVE comes with a ban on definites that presuppose the existence of their referent. Even though (58) seems to be a counterexample, the interpretation of its negative counterpart in (59) tells us that this is not the case. (58)

John has the only sweet brother.

(=26)

(59)

John doesn’t have the only sweet brother.

To our understanding, (58) states that there is a single sweet brother in the model and that it’s John’s. If the only sweet brother were presuppositional we would expect (59) to mean that there is a single sweet brother in the model but that it’s not John’s. This is however not the interpretation we get. Rather, (59) means that there are other people than John who also have a sweet brother. The interesting thing about (58) is consequently not that it allows for a definite after HAVE but rather that it allows for the definite to be non-presuppositional, in contrast to the definites we saw before: (60)

*John has the child.

(=9)

We propose the following explanation (see also Coppock & Beaver 2015). We follow Partee (1987) in assuming a presuppositional and a non-presuppositional version of the. By Maximize Presupposition we expect the stronger presuppositional version to be the default. This explains why we don’t find sentences like (60): by Maximize Presupposition we expect the child and the blog to be interpreted 14

presuppositionally, which clashes with the existential environment HAVE creates. The context in (58) is special though in the sense that only overtly marks uniqueness of the object and is therefore incompatible with the indefinite. We conjecture that this is why Maximize Presupposition – a pragmatic principle – can be overruled and the non-presuppositional version of the definite article becomes available. The availability of a non-presuppositional version of the definite in the object position of HAVE is consequently due to the interplay between Maximize Presupposition, the ban on presuppositional definites HAVE comes with and the ban on (non-maximal) indefinites imposed by only. The crucial thing to take from this is that verbs that – like HAVE – come with built-in existential quantification are expected to allow for definites to get a non-presuppositional interpretation when they combine with only. If our analysis of HAVE-verbs is on the right track, we expect this to be one of their traceable properties. For English HAVE-verbs, we thus predict that they differ from other verbs in allowing for nonpresuppositional readings of definite DPs when combined with only. This prediction is borne out. (61) to (64) give examples of verbs that belong to the class of HAVE-verbs, expressing a use, creation, whole or possession relation: (61) (62) (63) (64)

John didn’t make the only pie. (‘other people also made a pie so there’s no unique pie’) This year didn’t contain the only happy day of my life. (‘other years also contained happy days so there’s no unique happy day of my life’) John didn’t hold the only glass of champagne. (‘other people were also holding glasses of champagne so there’s no unique glass of champagne’) John isn’t smoking the only cigarette. (‘other people are also smoking cigaretess so there’s no unique cigarette’)

The above verbs can be opposed to verbs like to like or to see that don’t probe relations in their object’s Qualia Structure and are consequently expected not to allow for non-presuppositional readings of the in combination with only. (65) and (66) show that this prediction is borne out: (65) (66)

#John doesn’t like the only fountain in the city. (‘other people also like fountains in the city so there’s no unique fountain’) #John didn’t see the only house. (‘other people also saw houses so there’s no unique house’)

On the basis of the opposition between the examples in (61) to (64) and (65)/(66) we conclude that we have positive evidence showing that HAVE-verbs have a special lexical semantics involving built-in existential quantification, not only in Romanian-type languages but also in English-type languages. This is strong support for our strategy to link their incorporation potential to their lexical semantics. 4.2. Chinese (L2 English) 4.2.1. Indefinite article drop If HAVE-verbs are cognitively different from regular verbs in that their default semantics requires predicates rather than arguments in their ‘object’ position, we predict L2 learners of English with an articleless L1 like Chinese not to immediately generalize the need of an indefinite article in object position and to display a higher article drop rate with HAVE-verbs than with regular verbs. The way we tested this is by tracking indefinite article drop for 10 regular and 11 existential verbs in a Chinese subsection of the error/correction-coded version of the Cambridge Learner Corpus with a focus on learners displaying article drop. Verb selection was based on criteria discussed above and the overall frequency of the verbs. The descriptive results are presented in Table 1 and show that our prediction is borne out: article drop is more frequent with HAVE-verbs than with regular ones. A generalized 15

mixed effects analysis on ordinal data with verb type and frequency as fixed effects, and random intercepts for learner and verb, confirms the significance of the influence of verb type (β = -4.05, p < .01), while controlling for frequency. The negative coefficient for regular verbs indicates that, compared to existential verbs, a higher indefinite article drop rate is less likely.

Indefinite article drop

Existential verbs

Regular verbs

1722 out of 6089 (28%)

226 out of 2952 (7,5%)

Table 1 Underproduction of articles is a well-known phenomenon in the literature focusing on the L2 acquisition of English by native speakers of an articleless language (e.g. Master 1987). We know underproduction is the norm for beginners and gradually disappears, though it remains a problem even for the most advanced learners. Beyond the obvious transfer effect, no other factors have been found to underlie underproduction and it is still unclear why it persists even for the most advanced learners. The data in Table 1 show that differentiating between verb types is a basic new desideratum for L2 acquisition research and the cognitive bias we identify might further hold the key to understanding why article drop persists. 4.2.2. Definite article overproduction The L2 literature focusing on the acquisition of English definite articles by learners with an articleless L1 has frequently noted that these learners overproduce definites (e.g. Thomas 1989, Ionin et al. 2009). The existence of two types of definites provides a straightforward explanation for this if we assume – with Barner et al. 2011 – that lexical scales like the one involving non-presuppositional and presuppositional the have to be learned: L2 learners use both types of definites interchangeably because they lack the (native) knowledge that the non-presuppositional version competes with the non-presuppositional one. We refer to this hypothesis on L2 definite article overproduction as the Non-Presuppositional Hypothesis (NPH). Where existential verbs team up with focus-sensitive only to unblock the use of non-presuppositional definites for native speakers, we expect them to have an important impact on the definites L2 learners produce as well. If the NPH is on the right track and if existential verbs are different from regular ones in the way discussed in section 1, we expect L2 learners to use non-presuppositional and presuppositional the interchangeably except in the object position of existential verbs given that this position favours non-presuppositional expressions. We then predict L2 learners to incorrectly use nonpresuppositional definites across the board but to do so more frequently with existential than with regular verbs. We tested this prediction by tracking overproduction of the definite article with the same verbs as before. Given the overall lower rate of overproduction of definites we increased our corpus size and looked at the full Chinese subsection of the Cambridge Learner Corpus with a focus on learners displaying definite article overproduction. Table 2 presents the descriptive results and shows that our predictions are borne out: the incorrect choice of a definite instead of an indefinite occurs both with regular and with existential verbs but is more frequent with the latter. We ran the same type of model as for indefinite article drop and found that the contrasts in Table 2 are significant (β = -1.999, p = .024). The negative coefficient for regular verbs indicates that, compared to existential verbs, a higher definite article overproduction rate is less likely.

16

Definite article overproduction

Existential verbs

Regular verbs

1402 out of 2244 (62%)

179 out of 608 (29%)

Table 2 The NPH combined with the existential vs. regular verb distinction straightforwardly explains the data in Table 2. The same does not apply to other hypotheses about the overproduction of the: the interpretation of the as a marker of noteworthiness (Ionin et al. 2004, Ionin et al. 2009), the role of egocentricity (Deprez et al. 2011), the as a marker of objective identifiability (Trenkic 2008), the as a marker of membership of a previously introduced set (Ko et al. 2010), etc. The data in Table 2 confirms that verb type differentiation is a basic new desideratum for L2 acquisition and supports the NPH as a valuable new hypothesis about the source of L2 definite overproduction.

5. Conclusion > HAVE-verbs constitute a special class of verbs that can best be described as light in the sense that they depend on their object nouns to provide them with relational input. > Semantically speaking, HAVE-verbs are best analysed as involving a built-in existential quantifier. > Empirically speaking, the following phenomena can be related to the discussion on HAVE-verbs: (i) pseudo-incorporation, (ii) the availability of non-presuppositional readings of definites in argument position, (iii) L2 indefinite article drop, (iv) L2 definite article overproduction.

References Alexandropoulou, S. (2013). The lexical restrictions on verbs with bare nominal complements in Greek and an analysis of such constructions with have-verbs (Master’s thesis, Utrecht University). Alexopoulou, T., & Folli, R. (2010). Indefinite topics and the syntax of nominals in Italian and Greek. Paper presented at the 3rd Mediterranean Syntax Meeting, Athens, Greece. Asher, N., & Denis, P. (2005). Lexical ambiguity as type disjunction. In Third international Workshop on Generative Approaches to the Lexicon. Barker, C. (1995). Possessive descriptions. Stanford: CSLI publications. Barker, C. (2011). Possessives and relational nouns. Semantics: An international handbook of natural language meaning. Barker, C., & Dowty, D. (1993). Non-verbal thematic proto-roles. In Proceedings of NELS (Vol. 23, No. 1, pp. 49-62). Barner, D., Brooks, N., & Bale, A. (2011). Accessing the unsaid: The role of scalar alternatives in children’s pragmatic inference. Cognition, 118(1), 84-93. Bhatt, R., & Pancheva, R. (2006). Implicit arguments. The Blackwell companion to syntax, 2, 554-584. Blutner, R. (2000). Some aspects of optimality in natural language interpretation. Journal of semantics, 17, 189-216. Boneh, N., & Sichel, I. (2010). Deconstructing possession. Natural Language & Linguistic Theory, 28, 140. 17

Borik, O., Cyrino, S., & Espinal, M. (2012). On determiners in languages with and without articles. Paper presented at the workshop 'Languages with and without articles', Paris 8, Paris. Borthen, K. (2003). Norwegian bare singulars (Doctoral dissertation, Norwegian University of Science and Technology). Boskovic, Z. (2008). What will you have, DP or NP?. In PROCEEDINGS-NELS(Vol. 37, No. 1, p. 101). Buck, C.D. (1949). A dictionary of selected synonyms in the principal Indo-European languages. University of Chicago Press. Chierchia, G. (1998). Reference to kinds across language. Natural language semantics, 6, 339-405. Chung, S., & Ladusaw, W. A. (2004). Restriction and saturation. MIT press. Coppock, E., & Beaver, D. (2012). Exclusivity, uniqueness, and definiteness. Empirical Issues in Syntax and Semantics, 9, 59-66. Coppock, E., & Beaver, D. (2015). Definiteness and determinacy. Linguistics and Philosophy, 377-435. Cyrino, S. and Espinal, M. (2011). Object BNs in Brazilian Portuguese. More on the NP/DP analysis. Paper presented at the CSSP 2011, Paris. Dayal, V. (2011). Hindi pseudo-incorporation. Natural Language & Linguistic Theory, 29, 123-167. De Bruin, J., & Scha, R. (1988). The interpretation of relational nouns. In Proceedings of the 26th annual meeting on Association for Computational Linguistics (pp. 25-32). Association for Computational Linguistics. De Swart, H. (2001). Weak readings of indefinites: type-shifting and closure.The Linguistic Review, 18, 69-96. Dekker, P. (1993). Existential disclosure. Linguistics and philosophy, 16, 561-587. Deprez, V., Sleeman, P., & Guella, H. (2011). Specificity effects in L2 determiner acquisition: UG or pragmatic egocentrism. In Selected Proceedings of the 4th Conference on Generative Approaches to Language Acquisition North America (GALANA 2010) (pp. 27-36). Dobrovie-Sorin, C., Bleam, T., & Espinal, M. T. (2006). Bare nouns, number and types of incorporation. Non-definiteness and plurality, 51-79. Espinal, M. (2010). Bare nominals in Catalan and Spanish. Their structure and meaning. Lingua, 120:984-1009. Espinal, M. T., & McNally, L. (2011). Bare nominals and incorporating verbs in Spanish and Catalan. Journal of Linguistics, 47, 87-128. Farkas, D., & De Swart, H. (2003). The semantics of incorporation. CSLI Publications, Stanford. Francez, I. (2010). Context dependence and implicit arguments in existentials.Linguistics and Philosophy, 33, 11-30. Givón, T. (1981). On the development of the numeral ‘one’as an indefinite marker. Folia Linguistica Historica, 15(Historica vol. 2, 1), 35-54. Givón, T. (1984). Syntax: A Functional-Typological Introduction. Amsterdam: John Benjamins. Groenendijk, J. A., & Stokhof, M. J. B. (1990). Dynamic Montague Grammar. [available at http://dare.uva.nl/document/2/27376]

18

Heim, I. (1982). The semantics of definite and indefinite noun phrases (Doctoral dissertation, University of Massachusetts). Heine, B. (1997). Possession: Cognitive sources, forces, and grammaticalization (Vol. 83). Cambridge: Cambridge University Press. Hellan, L. (1980). Toward an integrated theory of noun phrases. University of Trondheim. Hoop, H. de (2014). Case configuration and noun phrase interpretation. Garland. Iatridou, S. (1996). To have and have not: On the deconstruction approach. In Proceedings of WCCFL (Vol. 14, pp. 185-201). Ionin, T., Ko, H., & Wexler, K. (2004). Article semantics in L2 acquisition: The role of specificity. Language Acquisition, 12(1), 3-69. Ionin, T., Zubizarreta, M. L., & Philippov, V. (2009). Acquisition of article semantics by child and adult L2-English learners. Bilingualism: Language and Cognition, 12, 337-361. Kamp, H. (1981). A theory of truth and semantic representation. Formal semantics-the essential readings, 189-222. Ko, H., Ionin, T., & Wexler, K. (2010). The role of presuppositionality in the second language acquisition of English articles. Linguistic inquiry, 41(2), 213-254. Laca, B. (1999). Presencia y ausencia de determinante. In Gramática descriptiva de la lengua española (pp. 891-928). Espasa Calpe. Landau, I. (2010). The explicit syntax of implicit arguments. Linguistic Inquiry,41, 357-388. Landman, F. (2004). Indefinites and the Type of Sets. Blackwell. Lazaridou-Chatzigoga, D. (2011). The distribution and interpretation of bare singular count nouns in Greek. Paper presented at the workshop ‘Weak Referentiality’, UiL-OTS, Utrecht, The Netherlands. Lazaridou-Chatzigoga, D., & Alexandropoulou, S. (2013). A corpus study of Greek bare singulars: implications for an analysis. Revista da ABRALIN, 12, 233-251. Le Bruyn, B. (2010). Indefinite articles and beyond. LOT. Le Bruyn, B., de Swart, H., & Zwarts, J. (2013a). Establishing relations. [available at https://sites.google.com/site/blebruyn] Le Bruyn, B., de Swart, H., & Zwarts, J. (2013b). 'Have','with' and 'without'. In Proceedings of SALT (Vol. 23, pp. 535-548). Löbner, S. (1985). Definites. Journal of semantics, 4, 279-326. Martí, L. (2006). Unarticulated constituents revisited. Linguistics and Philosophy, 29, 135-166. Master, P. A. (1987). A cross-linguistic interlanguage analysis of the acquisition of the English article system (Doctoral dissertation, UCLA). Merchant, J. (2013). Voice and ellipsis. Linguistic Inquiry, 44, 77-108. Partee, B. (1987). Noun phrase interpretation and type-shifting principles.Studies in discourse representation theory and the theory of generalized quantifiers, 8, 115-143.

19

Partee, B. (1999). Weak NPs in have sentences. In JFAK [a Liber Amicorum for Johan van Benthem on the occasion of his 50th birthday], ed. J. Gerbrandy et al., CD-Rom. Amsterdam: University of Amsterdam. http://www.illc.uva.nl/j50/ Partee, B., & Borschev, V. (2003). Genitives, relational nouns, and argument-modifier ambiguity. Modifying adjuncts, 4, 67-112. Prince, K. von (2012). Nominal possession in Daakaka: Transitivizing vs. linking. In Proceedings of AFLA (Vol. 18). Pustejovsky, J. (1995). The generative lexicon. Cambridge: MIT Press. Ritter, E., & Rosen, S. T. (1997). The function of have. Lingua, 101, 295-321. Rosenbach, A. (2002). Genitive variation in English: conceptual factors in synchronic and diachronic studies (Vol. 42). Walter de Gruyter. Sæbø, Kjell Johan (2009). Possession and pertinence: the meaning of have. Natural language semantics, 17, 369-397. Stassen, L. (2009). Predicative possession. Oxford: Oxford University Press. Swart, H. de, & Zwarts, J. (2009). Less form–more meaning: Why bare singular nouns are special. Lingua, 119, 280-295. Tantos, A. (2008). Computing Events in Discourse: A Case Study Involving Light" have"(Doctoral dissertation, University of Konstanz). Thomas, M. (1989). The acquisition of English articles by first-and second-language learners. Applied psycholinguistics, 10(03), 335-355. Trenkic, D. (2008). The representation of English articles in second language grammars: Determiners or adjectives?. Bilingualism: Language and Cognition, 11, 1-18. Van Geenhoven, V. (1998). Semantic incorporation and indefinite descriptions: Semantic and syntactic aspects of noun incorporation in West Greenlandic. CSLI publications. Van Peteghem, M. (1993). La détermination de l'attribut nominal: étude comparative de quatre langues romanes (français, espagnol, italien, roumain). Brepols. Vikner, C., & Jensen, P. A. (2002). A semantic analysis of the English genitive. Interaction of lexical and formal semantics. Studia Linguistica, 56, 191-226. Xie, Z. (2014). The degree use of the possessive verb yǒu in Mandarin Chinese: a unified analysis and its theoretical implications. Journal of East Asian Linguistics, 23, 113-156.

20

relations and incorporation Bert Le Bruyn

relation. In the output, the theme argument variable y has been suppressed. ..... We now continue to let HAVE add existential closure over x and to equate di.

819KB Sizes 7 Downloads 200 Views

Recommend Documents

Selective Incorporation Reading and questions.pdf
There was a problem previewing this document. Retrying... Download. Connect more apps... Try one of the apps below to open or edit this item. Selective ...

Certificate of Incorporation -
Registrar of Companies. National Capital Territory of Delhi and Haryana. 2014-2015. ANJALI POKHRIYAL. Assistant Registrar of Companies. Digitally signed by ...

incorporation by law
humans simply because they are humans, has room for the law. How can morality ..... court may, in its discretion, make an order for security for costs. . . .” In that ...

BERT - ERNA by Lalylala.pdf
Page 1 of 15. Anh's Arigurumi. https://www.facebook.com/AnhsArigurumi. 1. BERT - ERNA. (Blackberry) - (Strawberry). by Lalylala. Page 1 of 15 ...

1 Incorporation Of Company And Matters Incidental Thereto.pdf ...
1 Incorporation Of Company And Matters Incidental Thereto.pdf. 1 Incorporation Of Company And Matters Incidental Thereto.pdf. Open. Extract. Open with.

Preverbal coordination and the semantics of noun incorporation in ...
In this talk I will consider certain semantic properties of noun incorporation in Hungarian on the basis of a similar construction, which I will refer to as preverbal ...

VERB INCORPORATION IN GUARANÍ CAUSATIVE ... - Paraguay
defines each lexical category (noun, verb and adjective- prepositions are considered to be a functional category in ..... is that of adjectives. Where nouns and verbs both have inherent properties which define their distribution and ...... Categorias

ARTICLES OF INCORPORATION OF PEMBROOK ...
In compliance with the requirements of the Corporations and Associations Article, ... Capitalized terms not defined herein shall have the meaning assigned to.

VERB INCORPORATION IN GUARANÍ CAUSATIVE ... - Paraguay
Object-prefix incorporation in proto Tupí-Guaraní verbs. Language. Sciences, 9(1), 45-55. Jensen, Cheryl. 1990. Cross-referencing changes in some Tupí-Guaraní languages. In D. Payne (Ed.), Amazonian linguistics: studies in lowland South American

Metropolitan Decentralization through Incorporation ...
Jul 16, 2007 - A second dimension affecting the ease or difficulty in forming political govern- ... erally represented by a locally oriented community club, which ...

Incorporation of Disposable Screen-Printed ... -
experiments reveal excellent reproducibility. The relative standard deviation for ip measured was 4.65% and 1.67% for. 11 different SPEs and 50 continuous injections using the same SPE, respectively. Working electrodes are known to foul during CE run

Balinese Pseudo-Noun Incorporation: Licensing under Morphological ...
Page 1 of 2. Balinese Pseudo-Noun Incorporation: Licensing under Morphological Merger. Keywords: pseudo-noun incorporation, head movement, adjacency, morphological merger. Head Movement of a nominal into a verb (Noun Incorporation) has two consequenc

Linker Histones Incorporation Maintains Chromatin ...
the experimental data, and use this value to model the linear part of the ..... in topology after nucleosome removal is consistent with the topological deformation of А1 turn per nucleosome. (c) Model of the nucleosome ..... electronics and software

5. HUMAN RESOURCE MANAGEMENT INCORPORATION OF ...
DR. AJAY KUMAR. 281-290. Page 3 of 21. 5. HUMAN RESOURCE MANAGEMENT INCORPORATI ... ORK ENVIRONMENT AND JOB SATISFACTION.pdf. 5.

THE POWER OF APPETITE CORRECTION eBook Bert Herring PDF ...
Download Free AC: THE POWER OF. APPETITE CORRECTION PDF ... A working appetite center means:you recover control over foodno calorie counting is ...

pdf-0969\sport-magazine-bert-jones-baltimore-colts-john-riggins ...
Whoops! There was a problem loading this page. pdf-0969\sport-magazine-bert-jones-baltimore-colts-john-riggins-yankees76-october-1976-from-sport.pdf.

SCJP SCJD 1.6 Kathy Sierra & Bert Bates.pdf
Page 2 of 890. Praise for the Sun Certified Programmer & Developer for Java 2 Study Guide. "Kathy Sierra is one of the few people in the world who can make ...

THE POWER OF APPETITE CORRECTION eBook Bert Herring PDF ...
money and time! (see below for details)You work with your body instead of fighting it!Hundreds of. AC/Fast-5 2016 survey respondents reported: average ...

Shaun de Bruyn - South Australian Tourism Industry Council
Mar 1, 2016 - For further information contact Bianca Borrett, Manager Communications, South Australian Tourism Industry Council on. 0421 355 252 or email ...

conte-asie-le-petit-chacal-et-le-chameau.pdf
entre les deux bosses, et le chameau. traversa la rivière à la nage. Quand. ils furent sur le bord, le petit chacal. sauta à terre, indiqua au chameau le. champ de ...