Predicative Possession

The best topics from Languages & Linguistics, kept on a permanent basis.
Post Reply
chris_notts
Avisaru
Avisaru
Posts: 275
Joined: Wed Dec 15, 2004 9:05 am
Location: Nottingham, England
Contact:

Predicative Possession

Post by chris_notts »

I've just been adding a section to my wiki briefly summarising "Predicative Possession" by Stassen. I thought people might be interested so I posted it here. You can also see it on the wiki:

http://www.chrisdb.me.uk/wiki/doku.php? ... on_schemas

For some background you can see the same wiki page.

I'd be interested in other people's thoughts about this typology. I personally find the (statistical) universals fairly convincing, since he goes into a lot of detail in the book to justify them. What I find less convincing is his explanation for the typology, since it involves positing that possessive constructions are/were underlyingly or originally biclausal. I'm personally not sure there's much evidence for this.

-------------------------------------------------------------------

Typological Correlations of Possession Schemas

Stassen has proposed a relationship between the possessive schema a language employs and other features of the language.

Foundation of the Typology

The possible types of possessive construction he considers in his study are:

1. Locational Possessive

The possessor is encoded as a locative oblique, and the possessee is the intransitive subject of a locative or existential verb. This type covers the location, goal and source schemas of Heine.

2. With Possessive

The possessor is encoded as the intransitive subject of a locative or existential verb. The possessee is encoded as an oblique, typically with a comitative preposition or case marker. The type is equivalent to the companion schema of Heine.

3. Topic Possessive

The possessor is encoded as the sentence topic, and the possessee is the intransitive subject of a locative or existential verb. This type covers Heine's topic schema.

4. Have Possessive

The possessor and possessee are encoded as subject/agent and object/patient of a transitive verb. This type covers Heine's action schema.

Typological Correlates

Stassen looks for correlates between the type of possessive a language uses for expressing predicative alienable possession and other features of the languages in his sample. He find relationships with:

1. The major strategy that a language uses for coordinating clauses expressing simultaneous events which involve different subjects.

This is classified in terms of balancing and deranking. If a language primarily uses balancing then both clauses could occur independently as individual main clauses without modification, whereas if it primarily deranks the one of the two clauses is in a form that cannot stand alone. Deranking strategies typically involve the use of non-finite verb forms.

2. Whether languages encode predicate nominal and locational clauses identically or not

English is an example of a language which codes both types of clause in the same manner, using the verb 'to be'. Other languages, such as Japanese, code the two types of clause differently:

Nominal

John wa usotuki da
J. TOP liar COP
John is a liar

Locational

Tukue no ue ni hon ga aru
desk GEN top LOC book SUBJ be.there.NONPAST
There is a book on the desk

The examples show that the verb used in nominal and locative predications are not the same.

Proposed Typological Implications

Stassen proposes the following typological implications:

1. If a language has a locational or with possessive, it deranks simultaneous different subject clauses

2. If a language has a topic possessive, it has balanced simultaneous different subject clauses and it splits nominal and locational clauses

3. If a language has a have possessive, it has balanced simultaneous different subject clauses and it encodes nominal and locational clauses in the same way

Explanation

Stassen explains his implications by proposing that the underlying structure of alienable possession is a sequence of simultaneous existential predications. In other words, possession is expressing the existence of the possessor and the possessee in the same space at the same time. This explains why three of the four major possessive types are modelled on locative/existential constructions. The assumed underlying structure is as follows:

[Possessor BE] [Possessee BE]

This structure would be represented in languages with deranking as one of the following, depending on whether the first or second clause is deranked:

[Possessor BE.DERANKED] [Possessee BE]

[Possessor BE] [Possessee BE.DERANKED]

There is at least one language which has this kind of expression for possessive encoding:

DAFLĀ

Lok nyi ak da-tla ka anyiga da-tleya
one man one be-CONV.PAST son two be-3DU.PAST
A man had two sons

Stassen assumes that typically the following occurs:

1. Ellipsis of the be predicate in the deranked clause
2. 'Inheritance' of the deranking marker by the subject of the deranked clause, by the subject of that clause, in the form of an oblique marker.

This means that languages with the first clause deranked arrive at a locative possessive, and languages with the second clause deranked arrive at a with possessive. Since verb final languages typically derank the first clause, and verb initial languages typically derank the second clause, this means that locative possessives are more common in SOV languages and with possessives are more common in VSO languages.

If a language is primarily balancing, the initial form of the possessive construction would be two balanced, coordinated clauses. This occurs in Ixtlan Zapotec:

Lèyetsì kyá δoá tù jrù-δí δoá tù βɛ́kù tò kyè
village mine exist one gentleman exist one dog small of.him
In my village there was a gentleman who had a little dog (lit: in my village, there was a gentleman, there was a little dog of his)

Stassen claims that the typical development from this structure is that the first existential verb is elided, to give the structure:

[Possessor] [Possessee BE]

This is the topic possessive construction. However, if nominal and locative predications are coded in the same way, this structure may be ambiguous between a possessive construction and a nominal predication. In some cases, other features of the language, such as possessor agreement on nouns, may help remove the ambiguity. Temporal adverbs or conjunctions can also be inserted to disambiguate between possessive and nominal predications. A few languages even tolerate the ambiguity.

Finally, some languages solve the problem by abandoning the 'natural' encoding for permanent alienable possession, and extending a strategy used to express other kinds of possession. This alternative would be the have possessive. Stassen claims that the 'natural habitat' of the have possessive is temporary possession, which makes sense given the typical origin of the strategy in verbs like 'hold'. This leads to another typological universal:

If a language uses a have possessive for alienable possession, it will use a have possessive for temporary possession.
Try the online version of the HaSC sound change applier: http://chrisdb.dyndns-at-home.com/HaSC

TomHChappell
Avisaru
Avisaru
Posts: 807
Joined: Wed Dec 28, 2005 2:58 pm

Post by TomHChappell »

That's interesting! 8) Thanks!

Post Reply