When is Possession not Possession?
- Rebecca Jackson, University of Salford
- Old College Lorimer Room, University of Edinburgh
When is possession not possession? A relevance-theoretic solution to the problem of the
possessive in English.
In English, it is possible to use possessive determiners in an NP even when the referent cannot
be thought of as being possessed by the person picked out by the pronoun. Compare (1) with (2 -3):
(1) My jacket is in the hallway. [A jacket that the speaker owns.]
(2) My programme is on T.V. [A T.V. programme that the speaker owns???]
(3) Your vase is on the table. [A vase that the hearer has given to the speaker and cannot ‘own’.]
The speaker of (2) might be understood to be referring to a television programme he wants to
watch or likes to watch. But interpreted in a context in which the speaker is a television producer, it
will be understood to be referring to a programme he has made. However, it is not clear what, if
anything, either interpretation has to do with a ‘standard’ notion of possession. What is clear is that
any attempt to solve this should account for as many usages of possessives as possible, regardless
of whether they are considered to be ‘standard’ or ‘non-standard’.
This paper begins by setting out how this problem might be tackled on the assumption that a
possessive determiner encodes a concept. First, I consider the view that possessives are
polysemous, but show that there are both empirical and theoretical arguments against this
approach. I then examine the idea that non-possessive uses of possessives are explained in terms
of Lakoff and Johnson’s (1981) notion of conceptual metaphor, and show that this fails for reasons
similar to the polysemy hypothesis. Finally in this section I consider the idea that a possessive
determiner encodes a concept of POSSESSION which is adjusted in context by processes of
pragmatic enrichment (Carston 2002). However, as I will show, the notion of grammatical
possession should not be confused with the concept of possession that plays a role in our thoughts
and inferences - any more than the (grammatical) gender of a word should be confused with our
concept of biological gender. Indeed, it seems that possessive pronouns do not encode conceptual
information at all.
In the second part of the paper, I argue that these expressions can be analysed within the
framework of Sperber & Wilson’s (1986) relevance theoretic pragmatics as semantically encoded
constraints on the recovery of explicit content (Blakemore 1987, 2002). According to this analysis, a
possessive determiner does not itself encode a constituent of a conceptual representation (or
proposition) which is interpreted for relevance, but activates a procedure for assigning reference.
Using naturally-occurring data from native speakers of English, I show how this analysis not only
provides a natural explanation for the variation in interpretation noted above, but also explains why a
possessive determiner cannot be used in any context at all.
References:
Blakemore, D (1987). Semantic Constraints on Relevance. Oxford: Blackwell
Blakemore, D (2002). Relevance and Linguistic Meaning. Cambridge: Cambridge University Press.
Carston, R (2002) Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford:
Blackwell.
Lakoff, G & Johnson, M (1981). Metaphors We Live By. Chicago: University of Chicago
University Press.
Sperber, D & Wilson, D, (1995) Relevance: Communication and Cognition. Oxford: Blackwell
possessive in English.
In English, it is possible to use possessive determiners in an NP even when the referent cannot
be thought of as being possessed by the person picked out by the pronoun. Compare (1) with (2 -3):
(1) My jacket is in the hallway. [A jacket that the speaker owns.]
(2) My programme is on T.V. [A T.V. programme that the speaker owns???]
(3) Your vase is on the table. [A vase that the hearer has given to the speaker and cannot ‘own’.]
The speaker of (2) might be understood to be referring to a television programme he wants to
watch or likes to watch. But interpreted in a context in which the speaker is a television producer, it
will be understood to be referring to a programme he has made. However, it is not clear what, if
anything, either interpretation has to do with a ‘standard’ notion of possession. What is clear is that
any attempt to solve this should account for as many usages of possessives as possible, regardless
of whether they are considered to be ‘standard’ or ‘non-standard’.
This paper begins by setting out how this problem might be tackled on the assumption that a
possessive determiner encodes a concept. First, I consider the view that possessives are
polysemous, but show that there are both empirical and theoretical arguments against this
approach. I then examine the idea that non-possessive uses of possessives are explained in terms
of Lakoff and Johnson’s (1981) notion of conceptual metaphor, and show that this fails for reasons
similar to the polysemy hypothesis. Finally in this section I consider the idea that a possessive
determiner encodes a concept of POSSESSION which is adjusted in context by processes of
pragmatic enrichment (Carston 2002). However, as I will show, the notion of grammatical
possession should not be confused with the concept of possession that plays a role in our thoughts
and inferences - any more than the (grammatical) gender of a word should be confused with our
concept of biological gender. Indeed, it seems that possessive pronouns do not encode conceptual
information at all.
In the second part of the paper, I argue that these expressions can be analysed within the
framework of Sperber & Wilson’s (1986) relevance theoretic pragmatics as semantically encoded
constraints on the recovery of explicit content (Blakemore 1987, 2002). According to this analysis, a
possessive determiner does not itself encode a constituent of a conceptual representation (or
proposition) which is interpreted for relevance, but activates a procedure for assigning reference.
Using naturally-occurring data from native speakers of English, I show how this analysis not only
provides a natural explanation for the variation in interpretation noted above, but also explains why a
possessive determiner cannot be used in any context at all.
References:
Blakemore, D (1987). Semantic Constraints on Relevance. Oxford: Blackwell
Blakemore, D (2002). Relevance and Linguistic Meaning. Cambridge: Cambridge University Press.
Carston, R (2002) Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford:
Blackwell.
Lakoff, G & Johnson, M (1981). Metaphors We Live By. Chicago: University of Chicago
University Press.
Sperber, D & Wilson, D, (1995) Relevance: Communication and Cognition. Oxford: Blackwell