�ݺ�ߣ

Diego Gabriel Krivochen (UNLP, Universität Potsdam)

Previous approaches
Pesetsky & Torrego (2001, 2004, 2007)

 Vehicle Requirement on Merge (2007: 1):
If α and β merge, some feature F of α must probe F on β.

 Conditions for Probing:
C-command
Feature identity

A model of Agree
 Agree (Feature sharing version)
i) An unvalued feature F (a probe) on a head H at
syntactic location α (Fα) scans its c-command
domain for another instance of F (a goal) at location
β (Fβ) with which to agree.

ii) Replace Fα with Fβ, so that the same feature is
present in both locations.

Problems
 C-command is stipulatively defined upon 2-D tree-like
representations.
 Restrictions on Merge: the operation does not follow
from conceptual necessity but from intra-theoretical
needs.
 Substantive complications: features, values, operations
(Agree = Search + Copy Value)

Case in P&T’s model
 Relation between Case in NPs and T finiteness (P&T,
2000: 5):

“The nature of nominative case
Nominative case is uT on D.”

uT on D is valued by non-defective T

The concept is extended to ACC case, depending on the
T node that values the uninterpretable feature on D.

Case as a subtype of Agree
 NOM = [TS [u-φ]…DP[u-T]]

 ACC = [TO [u-φ] …DP [u-T]]

Case in Radical Minimalism
 Case is a configurational relation, not a substantive
element (Cf. P&T, Chomsky)

 Relation between Theta-Theory and Case Theory:

 The optimal scenario would be that in which there is
a univoque relation between theta positions and
theta roles, and a Case interpretation for each
theta-position.

DeLancey (2001)
 “(…) suppose we could demonstrate that there are, say,
exactly x universal semantic roles which can occur as
core arguments in a clause in human language. The
most obvious language design would have x case
markers, one for each underlying role; every
argument would simply be marked for its semantic
role, which could then be read directly off the
surface morphosyntax (…)” (our highlighting)

vP

θ1 v

[CAUSE] VP

[GO/BE] PP

θ2 P

[WITH] / [TO] θ3

 Initiator will be assigned to the position θ1,
Theme (for Figure) and Location (for Ground)
vary between θ2 and θ3 depending on the
nature of the P head, central or terminal
coincidence. If P is central, then the Ground
will θ2 and θ3, Figure, while if P is terminal, it
will be the other way around.
 Force is within the Initiator sphere
Percept ("perceived entity") in the area
of Theme
Source, Path, Goal, Experiencer (see De Lancey,
2001), are interpreted within the Location
sphere.

 Case is the result of the local relationship between an
argument and a functional-procedural node.
 We will work with a system of three cases, also structured
as "spheres", in close relation to thematic roles. These
underlying and universal cases would be Nominative,
Accusative and Dative, names that we maintain as a matter
of convenience and expository purposes, but we really
should talk about an Initiator Case (Agent / Force), a
Theme Case and a Location Case (comprising variants of
terminal / central coincidence), regardless possibilities of
morphological realization in different languages.

 • Nominative - Initiator
• Accusative - Theme
• Dative - Location

 Nominative: read off from a {Time, {D}} local relation,
and interpreted thematically (in the explicature
building process, see Sperber & Wilson, 2003) as Agent
/ Force

 Accusative: read off from a {Cause, {D}} local relation,
and interpreted thematically as Theme, the object
(Figure) located in / moving towards, etc. a Ground

 Dative: read off from a {P, {D}} local relation, and
interpreted thematically as Location, the Ground in
Talmy’s terms.

An essential claim is that the spheres are not «far apart», but in
interaction, and there are points of contact. There are elements,
uses of the VI corresponding canonically to one «Case» that
appear in unusual configurations: of these, we will say they are
«intersective uses» of the Cases (blue circles):

NOM

ACC DAT

Examples:
Accusative > Dative: ACC expresses location in space, time
or property (i, ii and iii), or goal in movement (iv).
 Direction Accusative:
Latin: Eo RomamACC (Lit. Go Rome “I go towards/to Rome”)
Sanskrit: grāmamACC gacchami (Lit. Town go “I go to the
town”)
Polish: idę na pocztęACC ‘I am going to the post office’, idź po
niegoACC ‘go for him (go to pick him up)’
 Temporal Accusative:
Latin: quamdiu, quamdudum, quanto tempore complements.

•“Double accusative” in ditransitive constructions: traditionally
referred to as “thing ACC” and “person ACC”, they are within the
P domain, as Theme and Goal.

Latin: Verres MilesiosACC navemACC poposcit. (Lit. Verres the
milesians ship asked “Verres asked a ship to the milesians”),
Quid meACC istudACC rogas? (Lit. Why to-me this ask? “Why do
you ask me this?”)
Greek: διδάςκω τούσ παîδασ τήν γραμματικήν [didáskō toús
paîdasACC tḗn grammatikḗnACC] (Lit. I teach to-the children
grammar “I teach grammar to the children”)
Sanskrit: rājānamACC vacanamACC abravīt (Lit. King some words
said “(He/She) said some words to the king”)

Accusative > Nominative: ACC is the overt subject in non-finite clauses.

•ECM (Exceptional Case Marking):
Inglés: I want [themACC to come]
Latín: Video [teACC venire] (Lit. See you to-come, “I see you coming”)
Griego: λέγει ςε ἐλθεîν [légei [seACC eltheînINF]] (Lit. dice te haber venido –active
aorist infinitive- “(He/She) says you have come”)
Polish: Widzę cięACC spiącego. (‘I see you sleeping.’)

Dative > Nominative: Dative appears in subject position in the Spelled-Out
form.

•Quirky case:
Icelandic: HenniDAT leiddust ϸ NOM (Lit. Her bored they “They bored her”)
eir
MerDAT ϸ [MariaNOM vera garfuđ] (Lit. To me seems Mary be intelligent
ótti
“Mary seems to me to be intelligent”)
Polish: MariiNOM znudziła się pracaNOM. (‘Mary got bored with work’)
JanowiDAT podoba się we Włoszech. (‘John likes it in Italy.’)
Spanish: A JuanDAT leDAT basta con eso. (cf. Fernandez Soriano, 1999)
A MaríaDAT leDAT gusta el jazz

Values of the Genitive (DAT Sphere)
 Posession (material or metaphorical):
Sanskrit: Damayantyās svayaṃbaras (Lit. Damayanti’s choice)
Latin: id est domusNOM regisGEN (Lit. This is house of the king “This is the house
of the king”)
Greek: ἡ οἰκία τοû πατρόσ [hē oikía toû patrósGEN] (Lit. The house of the father)
Polish: sprawaNOM MariiGEN (‘Mary’s business/matter’)
 Origin:
Latin: AlteriusGEN partisGEN periculum (Lit. Somewhere else danger “The danger
[coming] from somewhere else”)
Greek (either concrete –cities- or abstract locations –families-): Σωκράτησ ὁ
ςωφρονίςκου [Sōkrátēs ho sōfronískou] (Lit. Socrates SofroniscoGEN
“Socrates, son of Sofronisco”), Ἀθεναίων [Athenáiōn] (“Socrates from
Athens”).
Polish: hałasNOM ulicyGEN ‘noise of the street’
 Quality (Only for individual level predicates, for stage level predicates ablative
case is used):
Latin: Vir magnaeGEN auctoritatisGEN (Lit. Man of great authority)
Polish: człowiekNOM honoruGEN (a person/man of honour)

 Location “through”:
Greek: θέουςαι πεδίοιο [théousai pedíoioGEN] (Lit. Running
meadowGEN “Running through the meadow”)
 Goal:
Sanskrit: PrājapaterGEN ātmānam paridadāmi (Lit. Prajapati myself
give “I give myself to Prajapati”)
 Source:
Greek: ή πολύσ oὐ πολύ απέχει τήσ ἥπειρυ [hḗ polýs ú polý apékhei
tḗs hḗpeiru] (Lit. The city not far away from the coast “The city is
not far away from the coast / departing from the coast”)
Polish: mlekoNOM od krowyGEN (Lit. milk from the cow, ‘cow milk’)
 Part-whole: “partitive Genitive”
Latin: Magna pars ItaliaeGEN (“A big part of Italy”)
Greek: πότεροσ τών αδελφών [póteros tṓn adelfṓnGEN] (“Which of
the brothers?”)
Polish: butelkaNOM wodyGEN (‘a bottle of water’)

 However, all uses locative establish a locative relation
between nominals, either concrete or abstract: a
nominal "belongs" to another in a concrete or abstract,
metaphorical way. Belonging is a type of location
(either central or terminal coincidence). Therefore, we
propose to subsume all of these uses the "sphere of
Dative".

 E.g.: John’s bag
[V BE [P John [[WITH CENTRAL] bag]]

 Is English [‘s] a CL (Radeva-Bork, 2012) or a case mark?
Abney (1987) supports the idea it is a CL in D0 raising to
Spec-DP, but:

 It is not an argument, rather, a locative predicate
 It only appears within full DP nominals, not with
pronouns, which inflect for Case:

i) A friend of [DP John’s]
ii) A friend of mine
iii) *A friend of [DP [[D mine’s] √Ø]] (see Panagiotidis,
2002 for an analysis of pronouns as D with empty root
complements)

 There is, however, another type of "genitive”: the
"Subjective" and “Objective", which appear in
nominalizations as a light PP. These kinds of genitive
also establish a locative relation between nominals: a
conceptual event, syntactically categorized as a
nominal, and a participant, either agent / initiator or
theme. This conceptual relation must be syntactically
realized without losing information (CP). This does
not imply that there is a transformational relation
between the nominalization and the tensed sentence
form, but the explicature building process is
analogous. These “genitives” are common features of
Romance and Germanic languages, so they can not be
considered an isolated anomaly.

Relational Semantic Structure
(Mateu, 2000; Jackendoff, 1987)
T

[event] r

Figure r

P Ground
E.g.: [T arrive [r Mary [r AT] the house]]

Possible Syntactic Realizations:
non-transparent interface
{time} {D}

[time] {event} [D] {event}

[event] {P} [event] {P}

{D} {P} {D} {P}

P {D} P {D}

Mary arrived (at the house) Mary´s arrival (at the house)

Causative constructions
R

Italia R

[cause] T

[GO] r

Albania r

[WITH] invasión

Syntactic instantiations
{D} {time}

La {cause} [time] {cause}

{P} {cause} {D} {cause}

P √Italia D √Italia
[cause] {event} [cause] {event}
Ø Ø

{P} {event} {event}
{D}

de {D} [event] √invasión D √Albania [event] √invasión

D √Albania

�ݺ�ߣ

Prepositions and Case Theory

More Related Content

Prepositions and Case Theory