User:Kaitlynkuhn/sandbox
In linguistics, dative shift refers to a pattern in which the subcategorization of a verb can take on two alternating forms, the oblique dative form or the double object construction form. In the oblique dative (OD) form, the verb takes a noun phrase (NP) and a dative prepositional phrase(PP), the second of which is not a core argument.
- John gave [NP a book] [PP.DATIVE to Mary].
In the double object construction (DOC) form the verb takes two noun phrases, both of which are core arguments, with the dative argument preceding the other argument.
- John gave [NP.DATIVE Mary] [NP a book].
more info here on what the shift actually is.
zCross Linguistic Data
[edit]Dutch
[edit]Similar to English, Dutch has many verbs of possessional transfer which display the phenomenon of dative alternation.
The double object construction has an unmarked NP theme and recipient object.
1a) Dutch: Jan heeft [NP.DATIVE zijn broer] [NP een boek] gegeven.
English Gloss: John has his brother a book given
English Translation: John has given his brother a book
The prepositional dative construction has only the theme encoded as a bare NP object and the recipient is marked by a preposition
1b) Dutch: Jan heeft [NP een boek] [PP.DATIVE aan zijn broer] gegeven.
English Gloss: John has a book to his brother given.
English Translation: John has given a book to his brother
Dutch has to main approaches to the analysis of the preposition aan.
Caused Motion Analysis
[edit]The dutch prepositional phrase aan is also typically analyzed in spatial terms. Van Belle and Van Lagendonck (1996) suggest that one of the major semantic determinants of the dative alternation in dutch is [+/- material transfer]. The construction in 1a highlights the involvement of the recipient, in this case, zijn broer, as an interested party involved in the event whereas the construction in 1b highlights the literal physical transfer of the book. (Van Belle and Van Lagendonck). This idea of material transfer is further highlighted in the following sentences:
2a) Vader bood oma zijn arm aan. ‘Father offered grandma his arm.’
2b) Vader bood zijn arm aan oma aan. ‘Father offered his arm to grandma.’
In these examples sentence 2a involving the oblique dative construction would be interpreted as a father offering his grandma his arm to hold in order to help her walk. In contrast when the double object construction is used, as in 2b, this sentence is interpreted as an unpleasant physical cutting off of the arm to give to his grandma because it is implying that a material transfer is involved.
Supporters of the caused motion analysis the [+/- material transfer] is only one major semantic determinants used in the analysis of the Dutch alternation. (cite this)
Dative Analysis
[edit]A Counter Argument
De Schutter et al. (1974, 1993) argues against the analysis of the aan-dative in spatial-directional terms outlined in the previous subsection. De Schutter (1974: 199) signals the existence of sen- tence pairs such as (8) below, in which the aan-variant is the unmarked option despite the obvious lack of a spatial transfer. Such examples illus- trate that the semantic distinction between the Dutch DOC and the aan- dative cannot be described in terms of ‘caused possession’ versus ‘caused motion’. Rather than an oblique locative phrase, the aan-phrase of the aan-dative is a prepositional object according to De Schutter (1974, 1993), which can encode a variety of semantic roles, just like the nominal indirect object of the DOC (recipient, possessor, addressee, . . .).
(8) a.
b. ? Ik bracht mijn geboortedorp een laatste groet.
Ik bracht een laatste groet aan mijn geboortedorp. ‘I paid a last salute to the village of my birth.
It is easy to see how this ‘dative’ approach is at odds with the ‘caused mo- tion’ approach sketched in the previous paragraph. If to is not an allative marker (anymore) in clauses like (9a) below, but simply marks Cindy as a recipient, then it cannot be maintained that there is a semantic di¤erence between (9a) and (9b) which can be described in terms of ‘caused motion’ versus ‘caused possession’. Rather, in this view, both constructions denote ‘caused possession’. The same applies, mutatis mutandis, to the Dutch sentence pair in (10): if aan does not denote a spatial relation in (10a), it can hardly be claimed that the aan-dative di¤ers from the DOC in presenting Cindy as a spatial goal rather than as a possessor.
(9) a. b. (10) a.
b.
John gave an apple to Cindy. John gave Cindy an apple.
Jan gaf een appel aan Cindy. ‘John gave an apple to Cindy.’ Jan gaf Cindy een appel.
‘John gave Cindy an apple.’
Of course, this does not preclude the existence of other fine-grained se- mantic distinctions between the constructions at stake: see, e.g., Wierz- bicka (1988) and Davidse (1996) for semantic hypotheses about the (English) dative alternation which do not involve a discussion of the prepositional variant in ‘caused motion’ terms but refer to notions such as the relative degree of affectedness of the theme and recipient participants or the ‘‘direction’’ of the resulting possession relation (i.e., ‘recipient has theme’ or ‘theme belongs to recipient’). Such hypotheses are not incompatible with an analysis of the preposition as a grammaticalized recipient marker.’
"To be sure, not all the studies mentioned above treat the dative alterna- tion as a simple opposition between a prepositional ‘caused motion’ and a double object ‘caused possession’ construction. In most cases, this opposition is part of a more intricate account of the semantic relation between the alternating constructions. In Langacker’s work, for instance, the above hypothesis about the path semantics of the to-dative is usually mentioned in the same breath with the hypothesis that the alternation is a matter of different selections of secondary figure: in the to-dative construction, the theme argument is the secondary figure and is accordingly encoded as the (direct) object, while the DOC construction selects the other downstream participant, the recipient, as secondary figure and hence direct object (see e.g., Langacker 1991b: 359–360).
Similarly, in Van Belle and Van Langendonck’s (1996) analysis of the Dutch alternation, [+/- material transfer] is presented as only one of the major semantic determinants, of equal importance to another factor, [þ/ involvement]. Still, it is fair to say that in many semantic studies of the dative alternation in English and/or Dutch, a discussion of the prepositional dative in ‘caused motion’ terms constitutes a crucial ingredient of the overall analysis." (you need to figure out how to properly explain this section and condense it with the existing information you have)
Similar to English, Dutch has many verbs of possessional transfer which display the phenomenon of dative alternation.
The double object construction has an unmarked NP theme and recipient objects
- Dutch: Jan heeft [NP.DATIVE zijn broer] [NP een boek] gegeven.
English Gloss: John has his brother a book given
English Translation: John has given his brother a book
The prepositional dative construction has only the theme encoded as a bare NP object and the recipient is marked by a preposition
2. Dutch: Jan heeft [NP een boek] [PP.DATIVE aan zijn broer] gegeven.
English Gloss: John has a book to his brother given.
English Translation: John has given a book to his brother
Other addition ideas
[edit]add tree with the sentence talking about VP raising under Chomsky
Kayne 1983
[edit]In his book, Connectedness and Binary Branching, Richard Kayne proposed an empty preposition being the source of the double-object construction. In his analysis, English prepositions have the ability to assign objective case. Kayne argued an empty preposition was responsible for allowing a double object construction.
- (8) V [PP Pe NP]NP
- (8) gave [pp Pe Mary] the book (is that what he means?, put it in?)
He continued with the notion that an empty preposition (Pe)cannot be the source of case, and inferred that instead the empty preposition could transfer the Case assigned by the verb. He further stipulated Case may only be transferred via prepositions that normally assign object case. Therefore, languages that do not assign object case via prepositions (such as French) cannot take the double-object form.
Barss and Lasnik 1986
[edit]In their 1986 paper "A Note on Anaphora and Double Objects," Barss and Lasnik point out a number of asymmetries in the behavior of the two NPs in double object construction. All of their findings point to the same conclusion: in constructions involving a verb phrase of the form V-NP-NP, the first NP c-commands the second, but not vice versa. Specifically, NP1 can bind NP2, but the opposite cannot occur. The paper provides significant evidence for rejecting linear phrase structure trees (Figure 1).
Maybe add in their tree suggestions and see what figure one is?