Navigation – Plan du site

Statives and nominalization1

Artemis Alexiadou
p. 25-52

Résumés

Dans cet article, j’examine les propriétés morpho-syntaxiques des verbes statifs et des nominalisations correspondantes. Je distingue, tout d’abord, deux classes de verbes statifs en grec: (i) les verbes ambigus entre une interprétation stative et une interprétation événementielle, et (ii) les verbes qui se prêtent, sans ambiguïté, à une interprétation uniquement stative. J’entreprends ensuite un examen des nominalisations formées à partir de ces deux classes de verbes. En prenant comme critère le type de modification qu’elles admettent, je montre que les nominalisations de statifs reposent sur une structure interne dépourvue du nœud exprimant la Quantité, responsable de l’héritage de la structure argumentale des verbes correspondants. Pour finir, je propose une structure articulée pour rendre compte à la fois des propriétés des verbes statifs et de celles des nominalisations statives.

Haut de page

Texte intégral

1. Introduction

  • 1 Many thanks to two anonymous reviewers and the editors of this issue for helpful comments and sugge (...)

1With a few notable exceptions (mainly in the context of psych verbs), nominalizations of stative predicates have not really received a lot of attention in the literature. One reason for this is the fact that it is not at all clear that such nominalizations support argument structure (AS), a point that relates to the general debate concerning the presence of event structure with stative verbs which is arguably responsible for the presence of AS. In this paper, I will survey some of the issues that arise in dealing with these nominalizations. Before turning to an investigation of statives, in the next sub-sections, I will briefly introduce Grimshaw’s analysis and the challenges this faces, as stative verbs and their nominalizations have to be seen in this context.

1.1 Grimshaw’s tripartite distiction

2As is well known, Grimshaw (1990) introduced a tripartite distinction of derived nominals: complex event, simple event and result ones. From her perspective, nouns such as examination in (1) are three-way ambiguous between a complex event reading that supports AS, and a simple event and a result reading that do not.

(1)

a.

the examination of the patients took a long time.

b.

the examination took a long time.

c.

the examination was on the table.

Simple event nominals are like complex event nominals in that they include an event reading, but like result nominals in that they lack obligatory arguments, as they lack a complex event structure, which, for Grimshaw, is the source of AS.

3For instance, an accomplishment verb like x constructs y is analyzed as an activity in which x engages in construction plus a resulting state in which existence is predicated of y (Grimshaw 1990: 26). This can be represented as in (2):

AS includes an aspectual dimension in that argumentrelations are jointly determined by the thematic properties of the predicate (i.e. the thematic hierarchy) and by the aspectual properties of the predicate, its event structure. The argument that participates in the first sub-event (‘activity’) is more prominent than the argument which participates in the second sub-event (‘result’). From this perspective, if a predicate lacks a complex event structure as the one in (2), it will also lack AS and will never take any grammatical arguments at all.

4Concerning derived nouns, Grimshaw proposes that the real distinction responsible for their multiple readings is that between nouns that have an associated complex event structure (her so-called complex event nominals) and nouns that lack such an event structure. Complex event nominals are amenable to an event structure analysis of the type in (2) and hence are capable of licensing arguments. Simple event and results ones are not amenable to such an analysis, and as a result they are incapable of licensing arguments.

1.2 Challenges for Grimshaw

5In the recent literature, this particular three-way distinction has been challenged. Specifically, Borer (2003), Alexiadou (2009), Harley (2009), and Alexiadou & Schäfer (2010a) all argue that one should distinguish between AS-nominals and R (eferential) nominals. These authors pointed out that there is no link between the presence of a complex event structure and AS in nominals (and perhaps in general). For the purposes of what follows, all nominals that lack AS qualify as R-nominals.

6First of all, and crucially for the present paper, AS-nominals need not be eventive and may be stative. The first case includes nominals derived from adjectives. As noted in Borer (2003), whenever the adjective in question can occur with arguments, these are directly carried over to the nominal derived from it:

(3)

a.

The court is aware of the problem.

b.

Pat is conscious of my presence.

c.

Jill is fond of classical music.

d.

 Robin is ready to leave.

e.

Marcia is close to her parents.

f.

The party is satisfied with the counting results.

(4)

a.

The court’s awareness of the problem

b.

Pat’s consciousness of my presence

c.

Jill’s fondness of classical music

d.

Robin’s readiness to leave

e.

Marcia’s closeness to her parents

f.

The party’s satisfaction with the counting results

7This is also the case for nominals derived from proto-typical stative verbs. In (5), as was the case in (3) and (4), the arguments of the verb are carried over to the nominal derived from it:

(5)

a.

John knows algebra.

b.

John’s knowledge of algebra

These nominalizations lack a complex event structure, and this is reflected in the fact that they do not form passive constructions (as first noted in Chomsky 1970 and references therein). Still, they license AS.

(6)

a.

*the fact’s knowledge

b.

knowledge of that fact

c.

the book’s destruction

d.

the destruction of the book

This restriction is captured by the Affectedness constraint, formulated by Anderson (1977):

(7)

If the head noun does not express an action which affects, i.e. modifies, the state of the object, the latter cannot occur in the prenominal position.

In more recent literature, this is taken to be a reflex of the event complexity of the nominal, see Doron and Rappaport Hovav (1991) and Sichel (2010) among others for discussion:

(8)

y is an affected argument of V (x, y) iff the event-structure of V contains a sub-eventuality e such that y, but not x, is an argument in e.

Destroy is such a predicate type, i.e. one can distinguish between a Cause and a Become part. Know is not such a predicate type, as one cannot distinguish between sub-eventualities.

8Second, as already pointed out, simple event nominals are eventive but still lack AS. Among these are nouns such as event, metamorphosis, journey, trip, etc. as illustrated by (9), Borer’s (2003) (25), and see the discussion based on Grimshaw (1990) concerning (2) above. The nominals in (9) are thus R-nominals:

(9)

a.

*The constant race to the mountains

b.

*The event in three hours

c.

*John’s deliberate trip to the mountains

d.

*A race from the station by Mary

e.

*The metamorphosis of the town in order to win a medal

Moreover, Alexiadou (2009), Harley (2009), Alexiadou & Schäfer (2010a), Roßdeutscher & Kamp (2010) pointed out that derived nominals may contain morpho-syntactic layers, which bring about eventivity, but still lack AS. In several cases, the presence of event structure is signaled by a verbal head, which can have a number of morphological realizations. In e.g. English, -ify, -en, -ize, -ate all qualify asverbalizing and eventive morphology.

The attachment of nominalizing morphology takes place outside the verbalizing head, as in (11).

  • 2 There is a further argument from interpretation : certain modifiers target directly the event intro (...)

However, the noun decoration does notalways have AS.2

(12)

a.

the decoration of the Christmas tree (AS + E-reading) took a long time.

b.

the decoration was expensive.

(R-reading)

c.

the decor was expensive.

(R-reading)

1.3 The scope of this paper

9Assuming then that (a complex) event structure is not a prerequisite for the licensing of AS in nominals and perhaps in general, in this paper I will examine the properties of stative nominalizations, focusing on deverbal stative nominals in (mainly) Greek.

10The following two issues will be touched upon: i) since there are several classes of stative verbs, do they all give equally good nominalizations? ii) What is the structure of stative verbs and their corresponding nominalizations?

11On the basis of Greek, I show that we find productive stative nominalizations of experiencer verbs, and verbs of existence. This raises the question what is special about these two stative verb classes. These two classes of stative verbs that allow nominalizations will be shown to lack a Davidsonian argument on the basis of adverbial manner modification tests. However, they allow modification via for x-time adverbials. On the other hand, those stative verbs that do not produce AS nominals will be shown to share one important property, namely they cannot be modified via for x-time. From this perspective, they behave like copula verbs. The consequences of this distinction, however, will have to await further research.

12The paper is structured as follows. In section 2, I offer a brief overview of the classification of stative verbs and discuss certain sub-classes in some more detail. In section 3, I turn to the nominalizations of stative verbs in Greek. In section 4, I discuss the syntax of stative verbs, and in section 5 the syntax of the corresponding nominalizations.

2. Classes of stative verbs3

2.1 How to define a stative verb

13A note of clarification is in order here. In the literature, the term event is usually used to refer to all non-statives, but it also may be used more narrowly to refer to events with a delimitation (see e.g. Parsons 1990). In this paper, I will use the term event to refer to all non-statives, regardless of termination. I will use the term eventuality for both events and states (Bach 1986). In other words, the term eventuality includes states and events, whereas the term event only refers to the eventive type of eventuality. Following Borer (2005), I will further distinguish between quantity vs. non-quantity eventualities, taking the latter to be homegenous predicates. From this perspective, statives are then non-quantity eventualities.

  • 4 From this perspective, processes are seen as part of the eventive part of eventualities. Note here (...)

14In Vendler’s (1967) four-way classification of verbs (which corresponds to what is referred to as inner Aspect/Aktionsart classification), state is defined as an eventuality in which there is no perceptible change4.

(13)

a. activities:

Mary danced for an hour.

b. accomplishments:

Mary built three houses in a year.

c. achievements:

The window broke.

d. states:

Mary knows the answer.

While it is generally agreed upon that all non-statives contain an event/Davidsonian argument, the presence of such an argument with stative verbs is disputed. Three main directions have been proposed. According to one view, all stative verbs lack a Davidsonian argument (Katz 2000). According to a second view, there are two classes of stative verbs, one with (14a) and one without (14b) a Davidsonian argument, e.g. Bach (1986), Kratzer (1995), Maienborn (2005). Bach refers to the former type as dynamic statives and to the latter as static statives. Maienborn uses the term Davidsonian states for the former group, and Kimian states for the latter group. Finally, the third view holds that all stative verbs have a Davidsonian argument, e.g. Parsons (1990), Mittwoch (2005), Martin (2006). In this paper, I side with the second view.

(14)

a.

sit, stand, lie + LOC

b.

be drunk, be in New York, own x, love x, resemble x

What kind of criteria can be used to determine the presence of an event argument within a sub-group of stative verbs? Maienborn (2005) as well as Borer (2005) pointed out that certain modifiers distinguish between the two classes of statives; these include manner and degree modifiers and are exemplified briefly below.

a) manner-modification: manner modification is out with static/Kimian-statives, cf. (15) vs. (16), which contains a dynamic/Davidsonian stative predicate:

(15)

*Jo owned the house quickly. (Borer 2005: 234)

(16)

Carol sat motionless at the table. (Maienborn 2005: 31a)

15b) degree-modification: Maienborn (2005) points out that in e.g. German the modifier ein bisschen ‘a little’ is ambiguous in two ways. It can be interpreted either as a degree modifier or as modifying the temporal extension of an event. Both readings are possible with an eventive verb like schwitzen ‘sweat’ as in (17).

(17)

degree and eventive reading

Carol hat gestern ein bisschen geschwitzt.

‘Carol has yesterday a little bit sweated.’

The time-span reading is only available for verbs that bear a Davidsonian argument, since ein bisschen specifies that the event didn’t last very long (18a). In verbs that lack such an argument, static/Kimian-statives, a time-span reading is excluded. Therefore either only the degree reading is available, as in (18b), or the modifier is completely ungrammatical, as in (18c).

(18)

a.

eventive reading

Paul hat ein

bisschen

im Garten gesessen.

‘Paul has a

little bit

in the garden sat.’

b.

only degree reading

Carol

ähnelte

ein bisschen

ihrer

Großmutter.

‘Carol

resembled

a little bit

her

grandmother.’

c.

*Nach ihrer 5. Heirat

hieß Liz ein bisschen Burton.

‘After her 5th marriage

was-named Liz a little bit.’

Both criteria can be used in Greek as well and give identical results: manner modification is out with static/Kimian-statives (19a), and the modifier ‘a little’ gives rise only, if at all, to the degree reading with such predicates (19b):

(19)

a.

*O Janis iksere Galika grigora.

the John knew French quickly

‘Johns knew French quickly.’

b.

only degree reading

I Maria

miazi

ligo

sti giagia tis.

the Mary-nom

resembles

a little

to grandmother hers

‘Mary resembles her grandmother a little.’

Statives, however, form an eventuality type. This can be shown on the basis of modification by for x-time, which distinguishes between quantity vs. non-quantity eventualities (Borer 2005). This criterion also applies to those statives that lack a Davidsonian argument as well, see Husband (2010) for some recent discussion. Borer (2005: 233f.) points out that for x-time modifiers that are a well-established test for the presence of atelic predicates, are not strictly speaking predicate modifiers. This is so because unlike in x-time modifiers, which cannot affect the event structure of a predicate, for x-time modifiers are operators of sorts, which turn non-quantity predicates to quantity predicates.

16As shown in (20), both activities and Kimian stative predicates are compatible with for x-time. Crucially, for x-time is excluded in the presence of quantity. In x-time, on the other hand, is a modifier of quantity, i.e. it always requires a quantity predicate (20b):

(20)

a.

Kim loved Robin for three years.

b.

Kim ran for 2 hours/*in 2 hours/to the store in 2 hours.

c.

#Kim loved Robin in three months.

17This criterion works for Greek as well:

(21)

a.

I Maria

agapuse

to Jani

ja tria hronia.

Mary

loved

John

for three years

b.

#I Maria

agapise

to Jani

se tris mines.

Mary

loved-perf

John

in three months.

Borer’s (2005) point is then that for x-time modifiers make reference to non-quantity and are equally possible with activities and statives. In this system, non-quantity predicates are homogenous, while quantity predicates are non-homogenous. From this perspective then, Kimian statives are also non-quantity, as they are homogenous.

18Borer further claims that for x-time is an operator on outer Aspect, while in x-time modifies inner Aspect, i.e. the domain of the VP. In other words, the domain of licensing of for-x time is aspectually distinct from the output of VPs, where inner Aspect is encoded.

19Table 1 summarizes the modification criteria to be used to classify stative predicates into two types, which are, however, both homogenous:

Table 1

manner

degree

for x-time (non-quantity)

Davidsonian stative

+

+

+

Kimian stative

-

-

+

20In the next section, I first offer a discussion of different classes of stative verbs in Greek and their behavior with respect to the above criteria. Two classes of verbs will be discussed: verbs that are ambiguous between a stative and an eventive reading (in section 2.2), and verbs that are unambiguously stative (in section 2.3). In section 3, I then examine the nominalizations of these two classes of stative verbs. As in Borer’s work, I will take a) the licensing of for x-time as evidence that there is non-quantity structure present in stative nominalizations (see also Markantonatou 1992) and their corresponding verbs; b) the absence of manner modification/event reading for ‘a little’ to classify the predicate as a Kimian state.

2.2 Verbs ambiguous between a stative and an eventive reading

21All the verbs discussed here fail the manner modification test and only have (if at all) the degree reading of ‘a little’ in their stative use.

i) Obstruct verbs (surround, cover, decorate, fill, block): Kratzer (2000) observes that verbs like obstruct may either have an agentive reading, which is eventive, and a stative reading. In the first reading, an agent is performing an action intentionally which results in a certain state brought about (22a). The stative reading, on the other hand, expresses only this resultant state. The reason for the stative/eventive ambiguity is that the verbs contain a cause-operator that relates two situations or two events to one another (the causer and the causee). A stative interpretation arises when both sub-eventualities are stative (22b):

(22)

a.

O stratos perikiklose to horio (grigora).

‘The army surrounded the village (quickly).’

b.

To tihos perikloni to horio (*omorfa).

‘The wall surrounds the village (*beautifully).’

(ii) Object-experiencer verbs assigning accusative case (e.g. anisiho ‘worry’, provlimatizo ‘puzzle’, enohlo ‘bother’, diaskedazo ‘amuse’, fovizo ‘frighten’, endiafero ‘interest’) (Anagnostopoulou 1999). In Greek, these are ambiguous between an eventive reading, which is necessarily agentive (see Anagnostopoulou 1999), and a causative reading, which several researchers classify as stative aspectually (see Arad 1998, Landau 2010 and references therein):

(23)

a.

O Janis

enohlise

ti Maria.

agentive/eventive

‘John

bothered

Maria.’

b.

I sokolata

enohli

ti Maria

(*ashima)

stative

‘The chocolate

bothers

Mary

(*ugly).’

The stative reading holds as long as the perception trigger is exposed to the experiencer (Arad 1998). Stative verbs of this class qualify as unaccusatives, according to Landau (2010), cf. Anagnostopoulou (1999).

22Note here that several verbs of this class have subject experiencer intransitive counterparts that are stative (Pesetsky 1995, Reinhart 2002, Landau 2010; Markantonatou 1992, Anagnostopoulou 1999 for Greek); these fall into two groups: Group I surfaces with active morphology (24a), while group II with non-active morphology (24b):

(24)

a.

I Maria

aidiase

me ti sokolata.

the Mary

disgust-3sg

with chocolate

‘Mary is disgusted with chocolate.’

b.

I Maria

endiaferete

ja ti musiki.

the Mary-nom

interest-non-active 3sg

for the music

‘Mary is interested in music.’

  • 5 There is an issue as to whether these verbs are unaccusative or unergative. In view of the fact tha (...)

What corresponds to the causer of the transitive verb surfaces in a PP, often distinct from the one found with Greek anticausatives/passives (ja ‘for’ (24b) vs. apo ‘from’/me ‘with’ in anticausatives/passives), but sometimes similar to that, (24a). This morphological pattern is generally found with the anticausative alternation, where two types of anticausative verbs can also be identified, one bearing active morphology, and one bearing non-active morphology (see Alexiadou, Anagnostopoulou & Schäfer 2006 for details). I take it thus that the verbs in (24) are unaccusatives5.

2.3 Non-ambiguous stative verbs

23Again all the classes discussed here fail the manner modification test and only have (if at all) the degree reading of the modifier ‘a little’.

i) Measure verbs are verbs that express the degree of a certain property such as length or duration.

(25)

To vivlio

kostizi

5 euro.

‘The book

costs

5 Euros.’

ii) Verbs of existence/presence and absence:

(26)

iparhun

pori.

exist-3pl

resources

‘Resources exist.’

iii) Verbs of modal state: mostly these have the form copula + noun but there are a few verbs of modal state.

(27)

To sholio

hriazete kalus daskalus.

‘The school

needs good teachers.’

iv) Verbs of spatial configuration (verbs which take PP complements):

(28)

I Galia sinorevi me ti Germania.

‘France borders with Germany.’

iv) Psych verbs, Anagnostopoulou (1999), Markantonatou (1992): here we have two sub-groups. The first group contains subject experiencer verbs like agapo ‘love’, miso ‘hate’, thavmazo ‘admire’.

(29)

a.

o Janis

agapai

ti Maria.

the-John-nom

loves

the-Mary-acc

‘John loves Mary.’

Several subject experiencer verbs in Greek e.g. fear bear non-active morphology, but license accusative Case and have causative counterparts that are object experiencer verbs (29c).

(29)

b.

o Petros

fovate

tis kategides

the-Peter-nom

fears-non active-3sg

the-storms-acc

‘Peter is afraid of storms.’

c.

I kategides

fovizun

ton Petro

the-storms-nom

frighten-3PL

the Peter-acc

‘Storms frighten Peter.’

The second group includes experiencer predicates with a dative (30b)/PP (30a) experiencer e.g. aresi ‘like’, ftei ‘bothers/matters’that qualify as unaccusatives (Anagnostopoulou 1999):

(30)

a.

To

krasi

aresi

ston

Petro

the

wine-nom

like-3sg

to-the

Peter

‘Peter likes the wine.’

b.

To

krasi

 tu

aresi

tu

Petru

the

wine-nom

cl-gen

like-3sg

the

Peter-gen

‘Peter likes the wine.’

3. Nominalizations of stative verbs in Greek6

  • 6 See Spencer & Zaretskaya (2003) for Russian, and Jaque (2010) for Spanish.

24Let us now consider the nominalizations derived from the verb classes introduced in the previous sections.

3.1 Nominalizations derived from ambiguous verbs

25In the case of obstruct verbs only the eventive reading can survive in the nominalization. This is shown by the fact that only the agentive PP is permissable in the nominalization:

(31)

to

perikikloma

tu

horiu

(apo to strato/*apo ton tiho)

the

surrounding

the

village-gen

(by the army/by the wall)

  • 7 As is well known, English nominalizations can only be related to the non-causative forms of the pre (...)

Object experiencer verbs also give agentive-eventive nominalizations, again manifested by the presence of the agentive PP:7

(32)

i tapinosi

tis Marias

(apo to Jani)

the humiliation

the Mary-gen

by John

Stative nominalscan, however, bederived from the stative unaccusative experiencer verbs, as the ones in (24). Observe that aspectual modification for-x-time is permitted:

(33)

a.

i aidia

tis Marias

ja ti sokolota

epi tosa hronia

the disgust

the-Mary-gen

for chocolate

for so many years

‘Mary’s disgust for chocolate for so many years.’

b.

to endiaferon

tu Jani

ja ta fita

epi dekaeties

the interest

the John-gen

for the plants

for decades

‘John’s interest in plants for decades.’

3.2. Non-ambiguous stative verbs

26In the case of measure verbs we do find a nominal form but aspectual modification is never allowed. This suggests that the nominal form in this case is basic, in the sense that the verb is derived from the noun (see Alexiadou 2009). As shown in (25), such verbs contain an overt verbalizer that takes an entity root as its input, e.g. kostos ‘cost’- > kost-iz-o ‘cost-verbalizer-1sg’:

(34)

to

kostos

tu

vivliu

(*epi

mia vdomada)

the

cost

the

book-gen

for

a week

‘the book’s cost’

Verbs of existence/presence and absence give productive nominalizations, and, as shown in (35), aspectual for-x-time modification is acceptable:

(35)

i iparksi

poron

epi tosa hronia

the existence

resources-gen

for many years

‘the existence of resources for many years’

Verbs of modal state do not give nominalizations at all.

Verbs of spatial configuration (PP complements) seem to have a nominal form related to them, but again this form does not license aspectual modification. In turn, this suggests that the nominal form is basic here. As was the case with measure verbs, there is a transparent morphological relationship between the noun and the verb, e.g. sinoro ‘border’- > sinor-ev-o ‘border-verbalizer-1sg’.

(36)

ta sinora

tis Germanis

me ti Gallia (*epi tosa hronia)

the borders

of Germany

with France for so many years

Finally, transitive subject experiencer verbs give productive nominalizations and allow for x-time aspectual modifiers. This is so irrespectively of the morphological form of the source verb, i.e. active vs. non-active:

(37)

i agapi

tis Marias

ja to Jani

epi tosa hronia

the love

the-Mary-gen

for John

for many years

‘Mary’s love for John for so many years’

(38)

a.

o

fovos

tu Jani

ja tis kategides

epi

tosa

hronia

the

fear

the John-gen

for storms

for

many

years

‘John’s fear of storms for many years’

b.

i lipi

tis Marias

ja ta gegonota

epi mia vdomada

the sadness

the Mary-gen

for the events

for one week

‘Mary’s sadness because of the events for one week’

Note that object-dative experiencer verbs do not seem to be able to nominalize productively. To the extent that a nominal form is available, it is only tenuously related to the verbal form:

(39)

O Jianis

eksefrase

tin areskia

tu.

the John-nom

expressed

the satisfaction

his

‘John expressed his satisfaction.’

(Markantonatou 1992: 90)

  • 8 Knowledge here seems to be the exception, gnosi tis algevras ‘knowledge of algebra’. But importantl (...)

As a final remark, let me point out that Markantonatou (1992: 89) showed that in Greek state AS nominals the genitive always encodes the experiencer argument, irrespectively of its grammatical function.8

27In conclusion, we find productive stative nominalizations of all unacussative stative verbs (experiencer verbs as well as verbs of existence), but also of the transitive subject experiencer verbs. While we do find nominal forms related to verbs of measurement as well as verbs of spatial configuration, these do not pass the aspectual criterion as opposed to the other three classes.

3.3 The morphology of stative nominalizations in Greek

28Kolliakou (1995) argued that -m- affixation is a productive process to build nouns out of verbs, but it is sensitive to the type of verbal predicate it attaches to. It namely requires a durative verb. Kolliakou noted that -m- does not combine with stative predicates (41); activities are compatible with -m- Accomplishments are also compatible with -m- and receive a durative/atelic interpretation:

(40)

a.

agapo

(love-1sg)

b.

pistevo

(believe-1sg)

(41)

a.

*agapima

(loving)

b.

*pistepsimo

(believing)

(42)

a.

perpatao

b.

to perpatima

walk-1sg

the walking

c.

htizo ena spiti

d.

to htisimo enos spitiu

build-1sg a house

the building of a house

e.

zografizo ena kiklo

f.

to zografisma enos kiklu

draw-1sg a cricle

the drawing of a circle

However, there are several nominalization of psych stative verbs with -m-:

(43)

a.

sevome

sevasm-os

respect-1sg

respect-ms

b.

thavmazo

thavmazm-os

admire-1sg

admiration

This forces us to seek a different explanation for the ungrammaticality of (41).

29If we look at the morphological form of pure stative nominals, these seem to be bare, i.e. they do not contain a particular nominalization affix. (44) shows that if even the corresponding verb contains verbalizing morphology (marked with bold), the nominalization lacks such a morphology:

(44)

a.

mis-o

mis-os

b.

pist-ev-o

pist-i

hate-1sg

hate

believe-1sg

belief

c.

agap-a-o

agap-i

love

love

d.

gnor-iz-o

gnos-i

e.

aidi-az-o

aidi-a

know

knowledge

disgust

disgust

This pattern is relatively systematic across stative verbs.

This also holds for the cases where the verb contains non-active morphology:

(45)

a.

endiafer-om-e

endiafer-on

am interested

interest-neut

b.

lip-ame

lip-i

am sorry non-active

grief-fem

c.

fov-ame

fov-os

am afraid-non-active

fear-masc

However, the picture is a bit more complicated than that. (46) shows that sometimes both bare and -m- nouns can be formed on the basis of the same verb:

(46)

a.

aidi-az-o

aidi-a

aidi-as-m-a/-os

disgust-1sg

disgust-fem

disgust-m-neut/ms

b.

sevome

sevas

sevas-m-os

respect-nonactive

respect-neut

respect-ms

(46a) seems to suggest that the two nominal forms relate to the verb in different ways. The -m- noun is derived from the verb, as it contains similarly to the verb an overt verbalizer, namely -az-. This is consistent with the generalization proposed in Alexiadou (2009) that -m- surfaces outside the verbalizing morphology. On the other hand, the bare noun lacks an overt verbalizer, suggesting that it is root derived. I come back to this issue in section 4.

30Crucially, however, when doublets exist, -m- and non -m- forms are not identical: -m- forms seem to be eventive, since the agentive PP is licensed:

(47)

a.

i adia

tu Jani

ja ti sokolota

the disgust

the John-gen

for chocolote

b.

to adiasma

tu Jani

apo tus politikus

the disgust

the John-gen

by the politicians

The above picture suggests that different affixes select for different nominal structural configurations.

4. The syntax of stative verbs

31My analysis of the syntax of stative verbs and their nominalizations relies on the following three assumptions:

Assumption 1:

statives are eventualities

Assumption 2:

a characterization of statives must be possible in terms of syntactic structure (Hale & Keyser 2002, Borer 2005)

Assumption 3:

the external argument of a stative verb is a holder (Kratzer 1996, Husband 2010)

32Kratzer (1996), in particular, argued that stative verbs contain only a single (structural) argument which is the holder of the state. In (48) the verb is stative and the external argument denotes, according to Kratzer, the person who is the holder of the state consisting in owning the dog. In Kratzer’s work only external arguments are severed from the verb. Internal arguments are still taken to be part of the VP, see Husband (2010) for an elaboration.

(48)

Mittie owns the dog.

33Under the assumption that external arguments are introduced by ‘light’ heads, to compute the meaning of (48), we need a Voice head adding the external argument, which is different from that introducing the argument of an eventive predicate. In Kratzer’s system one cannot combine the holder function with the denotation of an action predicate or the agent function with the denotation of a stative predicate. This explains for Kratzer why there is a connection between the Aktionsart of a verb and the thematic role of its external argument. Thus, subjects of stative verbs can be understood as possessors, experiencers and not as true causers.

  • 9 But see Arad (1998) and Landau (2010) for arguments that subject experiencers are mental locations, (...)

34Kratzer proposes that the external argument of stative verbs is introduced at the level of Voicestative.9

35Which stative verbs contain Voice? Subject experiencer verbs (know, love, hate) are the ones that can be represented on the basis of (49) above, i.e. they contain Voicestative. These are transitive and assign accusative. The other stative verbs all lack Voice (as they are unaccusative), see Landau (2010).

36From a semantic point of view,stativity is characterized as lack of change. The question is whether this has a direct reflex in the syntax of stative verbs, pretty much as presence of change-of-state is argued to be represented in terms of little v heads (see Alexiadou, Anagnostopoulou & Schäfer 2006, Ramchand 2008). In other words, what are the building blocks of stative verbs?

  • 10 For Hale & Keyser, categories that are inherently stative in this sense are: i) the heads that defi (...)

37As mentioned, Voice introduces the holder of the state. But what is the shape of the stative VP? To answer this, I will follow Hale & Keyser (2002), who proposed that stativity is the result of an inclusion of a PP relation, namely a PP of central coincidence, which is responsible for the lack of change interpretation. The PP introduces an/the second argument (in the case of transitives) as in (50), in a manner similar to non-core transitive verbs, see Marantz (2005).10

  • 11 This suggests that v in Greek does not always introduce event implications. Voicestative is the pro (...)

38(50) suggests that a verbalizer can be present with stative verbs, whose main function is to verbalize the root. There is some evidence for the presence of v in Greek: at least some stative verbs in this language, as discussed in section 3, have a morpho-syntax similar to that of eventive verbs, i.e. they contain some verbalizer, e.g. -az-, see Alexiadou (2009) and references therein, see also (46a) above.11

39A PP is generally assumed to be present in the representation of verbs of existence. Thus (52a) is the structure corresponding to (52b), see Alexiadou & Schäfer (2010b), for a recent discussion and references.

(52)

a.

[v [PPtheme in place]]

b.

iparhun pori.

exist-3pl resources

‘Resources exist.’

40A version of (52) is also involved in the case of unaccusative statives, i.e. those stative verbs that lack Voice, e.g. (53b).

41A note is in order here on experiencer verbs that bear non-active morphology. We should distinguish between two groups. The first group comprises verbs that assign accusative, as in (54); since they are transitive, we can assume that they include Voicestative. As to the locus of the morphology, two options are available to us: i) root-encoded, i.e. certain roots carry a diacritic that at Spell-out they require the presence of non-active morphology (see Embick 2000 for an explicit such proposal) or ii) Voice-encoded, which still preserves the availability of licensing accusative. In other words, Voice does surface with non-active morphology, but it still maintains the capacity to license accusative on the internal argument.

(54)

o Petros

fovate

tis kategides

Peter-nom

fears-nonactive

the storms

(54) is traditionally classified as a deponent verb. The counterpart of (54) in e.g. German is an inherently reflexive predicate:

(55)

Er fürchtet sich

vor

dem Sturm.

He fears reflexive

from

the storm

‘He fears the storm.’

(56)

*Er fürchtet mich vor dem Sturm.

‘He fears me from the storm’

One could argue here that non-active morphology in (54) surfaces on the basis of a more general mechanism that is also involved in inherent and naturally reflexives. From this perspective, such predicates are thus comparable to middles, see especially Kemmer (1993), Kaufmann (2001), Alexiadou & Doron (2010). These also bear non-active morphology:

(56)

O Janis

plithike.

John

washed-non-active-3sg

‘John washed himself.’

The second group includes verbs that are intransitive (i.e. unaccusative). I assume here that these contain an expletive Voicestative, Schäfer (2008):

(57)

O Janis

endiaferete

ja ti musiki.

the-John-nom

interest-non-active

for music

‘John is interested in music.’

5. The syntax of stative nominalizations

42Recall that, building on Borer (2005), the licensing of for x-time PPs is seen as evidence that event structure is present in stative nominalizations.

43The explicit claim here is that the event type (state) results from the combination of v and a PP (and in the case of transitive verbs a Voicestative). The for x-time adverbials modify/are licensed in Outer Aspect. -m- is a nominalizer that embeds AspectP, as in (58).

Recall also that state AS nominals appear to have the experiencer argument encoded by the genitive. I argued above that the experiencer argument is introduced in Voicestative in the case of transitive statives.

(59)

a.

o sevasmos

tu Jani

ja

tus

gonis

tu

the respect

the John

for

the

parents

his

‘John’s respect for his parents’

b.

I agapi

tis Marias

ja to Jani

the love

the Mary-gen

for John

‘Mary’s love for John’

As we see in (59), in Greek, but also in other languages, only one genitive is allowed within the DP. When a second argument is realized, this surfaces as a PP, as is the case in (59). In (59) then, the central coincidence preposition is overtly realized. As there is only one structural Case in the Greek DP (Alexiadou 2001 for details), then the highest argument is realized as genitive, which is the unmarked Case in the DP, following Marantz (1991).

44It has been argued for change-of-state verbs that their nominalizations are passive (of sorts), Grimshaw (1990), Alexiadou (2001), and this explains why only one DP can surface bearing structural genitive Case. State AS nominalizations cannot be passive, as there are no psych stative passives in general in Greek, see the discussion in Landau (2010).

45If passivization were involved in (59), it would have to be a kind of a pseudo-passive, since the theme is an argument of the PP. This option is generally not available in Greek.

46There is, however, an issue concerning the form of state nominals. As already mentioned, while the corresponding verbs have verbalizing morphology, this is absent in the nominals, see also (60). The question then is why the v morphology disappears (sometimes).

(60)

pist-ev-o

pist-i

;

agap-a-o

agap-i

believe-1sg

belief

love

love

If the presence of morphology relates to the presence of v, then the absence of morphology should signal the absence of v. In fact that would mean that such nominalizations are nominalizations of pure roots or that the nominal form is basic, i.e. non-verb derived. If this is correct, then we need to understand where the arguments of the nominalization come from (cf. Roy 2010). That the theme argument is introduced via a PP is clear. This could be included in the lower structure that gets nominalized. But what about the experiencer argument? One would need to argue that the experiencer is not introduced by Voice, but it is rather a possessor within the DP, probably introduced in nP.

47A second option here would be to suggest that actually these nominalizations are a kind of antipassive (AP) formation. The morphological pattern of AP is one in which the second argument is expressed as an oblique, while the first argument surfaces in the unmarked Case, here the genitive. On some analyses, antipassive is derived from the transitive verb through an affixation process; however, sometimes, but not always, it involves truncation of a transitivizing morpheme of the transitive verb. The AP affix can be zero with some verb stems (see especially work within Relational Grammar; Fortescue 1984). From this perspective, one could argue that the process of deriving a nominalization via AP ‘causes the verbalizing affix to disappear’. Clearly, further research is necessary in order to determine which of the two options is the most viable one.

48In the case of verbs of existence (and other unaccusative verbs),the single argument surfaces with genitive. These nominalizations are argued to have a structural representation as in (61):

49Turning now to the question of nominal layers, in a structure like the one in (61), the nP is responsible for the assignment of inflectional class, as all nouns in Greek belong to one inflection class. But are other higher nominal functional layers present?

50As far as I can tell, stative nominalizations do not pluralize (another property that they share with activities); hence NumberP should be excluded.

(62)

*I agapes tu Jani

ja ti Maria

the loves the John-gen for the Mary

If statives are non-quantities, this is expected. They behave similar to nominalizations derived from atelic/activity verbs, which also do not pluralize:

(63)

*The repeated observations of the stock prices induced their further decline.

(64)

a.

die Beobachtung

von Vögeln

the observe. ung

of birds

‘the observation

of the birds’

b.

*die Beobachtungen

von Vögeln

the observe.ung.pl

of birds

Alexiadou, Iordăchioaia & Soare (2010) argued that the reason why these nominalizations do not pluralize is their inner aspect inherited from the verb. The quantity pattern realizes plural, the non-quantity one does not, thus confirming previous observations in the literature, according to which bounded entities/eventualities can be pluralized, while unbounded ones cannot (see also Borer 2005).

6. Conclusions

51In this paper, I demonstrated that in Greek, productive stative nominalizations can be found for experiencer verbs, and verbs of existence. This raises the question what is special about these two classes.

52I showed that all stative verbs that form AS nominals lack a Davidsonian argument (manner modification test). The nominals qualify as AS nominals on the basis of the aspectual criterion, i.e. modification via for x-time. The stative nominals that cannot be classified as AS nominals share one important property, namely they cannot be modified via for x-time. Note here that this holds also for their verbal counterparts, see e.g. (65).

(65)

*i Galia

sinorevi

me ti Germania

ja hronia

France

borders

with Germany

for years

53There are several statives that behave like those in (65). Next to spatial stative verbs, we find measure verbs but also copula constructions involving individual level predicates, e.g. *Mary is fat for 2 hours. This behavior supports the idea that measure and spatial stative verbs are similar to the copula, in that they lack a temporal dimension. In turn, this suggests that the presence of a temporal dimension is a pre-condition for nominalization and licensing of AS for reasons that have to be clarified in future work.

Haut de page

Bibliographie

Alexiadou, Artemis (2001). Functional Structure in Nominals: Nominalization, and Ergativity. Amsterdam: John Benjamins.

Alexiadou, Artemis (2009). On the role of syntactic locality in morphological processes: the case of (Greek) derived nominals. In Giannakidou & Rathert (eds.): 253-280.

Alexiadou, Artemis; Anagnostopoulou, Elena; Schäfer, Florian (2006). The properties of anticausatives crosslinguistically. In Frascarelli (ed.): 187-212.

Alexiadou, Artemis; Doron, Edit (2010). The syntactic construction of two non-active voices: passive and middle. Ms. Universität Stuttgart and Hebrew University of Jerusalem.

Alexiadou, Artemis; Horrocks, Geoffrey; Stavrou, Melita (eds.) (1999). Studies in Greek syntax. Dordrecht: Kluwer.

Alexiadou, Artemis; Iordăchioaia, Gianina; Soare, Elena (2010). Number/Aspect interactions in the syntax of nominalizations: a Distributed Morphology approach. Journal of Linguistics 46: 537-574.

Alexiadou, Artemis; Rathert, Monika (eds.) (2010). The Syntax of Nominalizations across Languages and Frameworks. Berlin: Mouton de Gruyter.

Alexiadou, Artemis; Schäfer, Florian (2010a). On the syntax of episodic vs. dispositional -er nominals. In Alexiadou & Rathert (eds.). 9-38.

Alexiadou, Artemis; Schäfer, Florian (2010b).There-insertion: an unaccusativity mismatch at the syntax-semantics interface. Poster presentation at West Coast Conference on Formal Linguistics 28, University of Southern California.

Anagnostopoulou, Elena. (1999). On experiencers. In Alexiadou, Horrocks & Stavrou (eds.). 67-93.

Anderson, Mona (1977). NP Pre-posing in Noun Phrases, Proceedings of North East Linguistics Society (NELS) Meeting 8: 12-21.

Arad, Maya (1998). VP-structure and the syntax-lexicon interface. Ph.D. dissertation, University College London.

Bach, Emmon (1986). The algebra of events. Linguistics and Philosophy 9: 5-16.

Borer, Hagit (2003). Exo-skeletal vs. endo-skeletal explanations. In Moore & Polinsky (eds.). 31-67.

Borer, Hagit (2005). The Normal Course of Events. Oxford: Oxford University Press.

Carlson, Greg; Pelletier, Francis (eds.) (1995). The Generic Book. Chicago: University of Chicago Press.

Chomsky, Noam (1970). Remarks on Nominalization. In Jacobs & Rosenbaum (eds.): 184-221.

Doron, Edit; RappaportHovav, Malka (1991). Affectedness and externalization. Proceedings of North East Linguistics Society (NELS) Meeting 21: 81-94.

Dowty, David (1979). Word Meaning and Montague Grammar. Dordrecht: Reidel.

Embick, David (2000). Features, Syntax, and Categories in the Latin Perfect, Linguistic Inquiry 31: 185-230.

Giannakidou, Anastasia; Rathert, Monika (2009). Quantification, Definiteness and Nominalization. Oxford: Oxford University Press.

Grimshaw, Jane (1990). Argument structure. Cambridge, Mass.: MIT Press.

Fortescue, Michael (1984). West Greenlandic. London: Croom Helm.

Frascarelli, Mara (ed.) (2006). Phases of interpretation. Berlin: Mouton de Gruyter.

Hale, Ken; Keyer, Jay (2002). Prolegomenon to a theory of argument structure. Cambridge, Mass: MIT Press.

Harley, Heidi (2009). The morphology of nominalizations and the syntax of vP. In Giannakidou & Rathert (eds.). 321 – 343.

Husband, Matthew (2010). Compositional states. Paper presented at Semantics and Linguistic Theory (SALT) 20.

Jacobs, Roderick; Rosenbaum, Peter (eds.) (1970). Readings in English Transformational Grammar. Waltham, Mass.: Ginn and Company.

Jaque, Matias (2010). Nominalizaciones de verbos estativos: una primera approximación. Ms, Universidad Autonoma de Madrid.

Katz, Graham (2000). Anti neo-Davidsonianism: against a Davidsonian semantics for state sentences. In Tenny & Pustejosvky (eds.): 393-416.

Kaufmann, Ingrid (2001). Medium: Eine Studie zur Verbsemantik. Habilitationsschrift, University of Düsseldorf.

Kemmer, Suzanne (1993). The Middle Voice. Amsterdam: John Benjamins.

Kolliakou, Dimitra (1995). Definites and Possessives in Modern Greek: an HPSG Syntax for Noun Phrases. Ph.D. dissertation, University of Edinburgh.

Kratzer, Angelika (1995). Stage-level and Individual-level Predicates. In Carlson & Pelletier (eds.). 125 – 175.

Kratzer, Angelika (1996). Severing the External Argument from its Verb. In Rooryck & Zaring. (eds.). 109-137.

Kratzer, Angelika (2000). Building statives. Berkeley Linguistic Society 26.

Landau, Idan (2010). The Locative Syntax of Experiencers. Cambridge, Mass.: MIT Press.

Larson, Richard (1998). Event modification in nominals. Proceedings from Semantics and Linguistic Theory (SALT) 8. Cornell University, Ithaca, NY: 145-168.

Maienborn, Claudia (2005). On the limits of the Davidsonian approach: the case of copula sentences. Theoretical Linguistics 31-3: 275-316.

Maienborn, Claudia; Wöllstein, Angelika (eds.) (2005). Event Arguments: Foundations and Applications. Tübingen: Niemeyer.

Marantz, Alec (1991). Case and Licensing. Proceedings of Escol: 234-253.

Marantz, Alec (2005). Objects out of the lexicon: objects as event. Paper presented at the University of Vienna.

Markantonatou, Stella (1992). The Syntax of Modern Greek Noun Phrases with a Derived Nominal Head. Ph.D. dissertation, University of Essex.

Martin, Fabienne (2006).Prédicats statifs, causatifs et résultatifs en discours. Sémantique des adjectifs évaluatifs et des prédicats psychologiques [Stative, causative and resultative predicates in discourse.] Ph.D. dissertation, Université libre de Bruxelles.

Mittwoch, Anita (2005). Do States have davidsonian arguments? Some empirical considerations. In Maienborn & Wöllstein (eds.). 69 – 89.

Moore, John; Polinsky, Maria (eds) (2003). The Nature of Explanation in Linguistic Theory. Chicago: CSLI and University of Chicago Press.

Parsons, Terence (1990). Events in the semantics of English: a study in subatomic semantics. Cambridge, Mass.: MIT Press.

Pesetsky, David (1995). Zero syntax. Cambridge, Mass.: MIT Press.

Ramchand, Gillian (2008). Verb Meaning and the Lexicon: a First Phase Syntax. Cambridge Studies in Linguistics. Cambridge: University of Cambridge Press.

Rathert, Monika; Alexiadou, Artemis (eds.) (2010). The Semantics of Nominalizations across Languages and Frameworks. Berlin: Mouton de Gruyter.

Reinhart, Tanya (2002). The theta-system: an overview. Theoretical Linguistics 28: 229-290.

Rooryck, Johan; Zaring, Laurie (eds.) (1996). Phrase Structure and the Lexicon. Dordrecht: Kluwer.

Roßdeutscher, Antje; Kamp, Hans (2010). Syntactic and semantic constraints in the formation and interpretation of -ung- nominals. In Rathert & Alexiadou (eds.): 169-214.

Rothmayr, Antonia (2009). The Structure of Stative Verbs. Amsterdam: John Benjamins.

Roy, Isabelle (2010). Deadjectival nominalizations and the structure of the adjective. In Alexiadou & Rathert (eds.): 129-158.

Rozwadowska, Bozena (1997). Towards a Unified Theory of Nominalizations. Acta Universitatis Wratislaviensis.

Schäfer, Florian (2008). The Syntax of (Anti-) causatives. External Arguments in Change-of-state Contexts. Amsterdam: John Benjamins.

Sichel, Ivy (2010). Event-structure constraints on nominalization. In Alexiadou & Rathert (eds.): 159-198.

Spencer, Andrew; Zaretskaya, Marina (2003). Stative predicates in Russian and their nominalizations. Ms.

Tenny, Carol; Pustejosvky, James (eds.) (2000). Events as Grammatical Objects. Stanford: CSLI Publications.

Vendler, Zeno (1967). Linguistics in Philosophy. Ithaca: Cornell University Press.

Haut de page

Notes

1 Many thanks to two anonymous reviewers and the editors of this issue for helpful comments and suggestions. My research has been supported by a DFG [German Research Foundation] grant to the project B1, The Formation and Interpretation of Derived Nominals, as part of the Collaborative Research Center 732, Incremental Specification in Context, at the Universität Stuttgart.

2 There is a further argument from interpretation : certain modifiers target directly the event introduced by this verbalizer head inside the nominal (Larson 1998), see Roßdeutscher & Kamp (2010), Alexiadou & Schäfer (2010a):

Image4

The observation here is that in these cases the adjective fast and/or rough modifies the event of measuring or delivering respectively even if the nouns themselves have a result/object interpretation. Assuming that event modification makes necessarily reference to the presence of v, this means that both AS and non-AS nouns, as the nominals in (i), contain v.

3 See Martin (2006), Rothmayr (2009), Spencer & Zaretskaya (2003). In terms of event stucture, Dowty (1979) argues that stative predicates are the smallest and simplest building-blocks of event structure.

4 From this perspective, processes are seen as part of the eventive part of eventualities. Note here that Grimshaw’s concept of complex event nominals corresponds to Parson’s use of the term event. Here both complex event nominals and event nominals are eventive eventualities in opposition to stative eventualities.

5 There is an issue as to whether these verbs are unaccusative or unergative. In view of the fact that they can form an adjectival participle, I consider them unaccusative:

Image5

6 See Spencer & Zaretskaya (2003) for Russian, and Jaque (2010) for Spanish.

7 As is well known, English nominalizations can only be related to the non-causative forms of the predicates that distinguish between causative and non-causative readings (cf. Chomsky 1970, Grimshaw 1990, Pesetsky 1995:72, Markantonatou 1992: 90f.). Note here that Grimshaw (1990) argued that causative nominalizations are blocked as the corresponding verbs lack an external argument. According to Grimshaw, the kind of nominalizations that are excluded are non-agentive process ones. Result nominals and agentive process nominals, on the other hand, are licit. Greek is different, as it allows causative nominalizations, see (ii). It has to be determined, however, whether (ii) has a stative in addition to an inchoative reading:

Image6

8 Knowledge here seems to be the exception, gnosi tis algevras ‘knowledge of algebra’. But importantly in Greek the counterpart of (i) does not give a nominalization, suggesting that Greek gnorizo is not the counterpart of know but rather has the meaning ‘come to know’.

Image7

9 But see Arad (1998) and Landau (2010) for arguments that subject experiencers are mental locations, and thus should receive an analysis as in (52).

10 For Hale & Keyser, categories that are inherently stative in this sense are: i) the heads that define the external projection of A; ii) certain prepositions (e.g. in) and iii) the copula. Measure verbs are like copulas. (See also section 6.) Note that they cannot passivize and the measure phrase does not qualify as a referential argument.

Image9

11 This suggests that v in Greek does not always introduce event implications. Voicestative is the projection that introduces the external argument of a stative predicate, in case these have one, and does not come in different flavors.

Haut de page

Pour citer cet article

Référence électronique

Artemis Alexiadou, « Statives and nominalization », Recherches linguistiques de Vincennes [En ligne], 40 | 2011, mis en ligne le 01 novembre 2013, consulté le 25 mai 2017. URL : http://rlv.revues.org/1991 ; DOI : 10.4000/rlv.1991

Haut de page

Auteur

Artemis Alexiadou

Universität Stuttgart

Haut de page

Droits d’auteur

© Presses universitaires de Vincennes

Haut de page