LLCC Seminar - Elitzur Bar-Asher Siegal (Linguistics,HUJI)

Date: 

Monday, June 1, 2015, 2:30pm to 4:00pm

See also: LLCC Seminar

Location: 

room 200, The Australian Research and Graduate Studies Complex, School of Education, Mount Scopus, Jerusalem

Title: The NP-Strategy for Expressing Reciprocity: History and Semantics

Abstract: My paper focuses on the NP-strategt for expressing reciprocity. Constructions are included under this category based on two criteria: 1) They share the same range of uses (to be demonstrated); 2) The encoding is non-verbal, i.e., verbs in the relevant constructions are transitive (unlike verbal encoding of reciprocity). Thus, (1a), a reciprocal sentence, which denotes a symmetric relation between its participants has the same predicate and argument structure as (1b):

(1) a. James and Beth love each other
b. James loves Beth.

It has been repeatedly noted that cross-linguistically the same NP-expressions that encode symmetric relations (e.g., English each other) express other relations where strong reciprocity is impossible (Fiengo & Lasnik 1973, Dougherty 1974, Lichtenberk 1985, Dalrymple et al. 1998, Williams 1991, Beck 2001, Haas 2010, Evans et al. 2011). For example, the following sentence does not express a symmetric relation:

(2) They were hiding behind each other.

The situation of having different semantic functions for different sentences raises the following questions, phrased by Dougherty (1974: 18-19):  “How is a specific input lined to a specific output? That is, what is the rule of semantic interpretation for each other sentences?… how specific interpretation (or range of interpretations) is assigned to an arbitrary sentence”. 
Dalrymple et al. (1998) propose that the meaning of the reciprocal sentences varies from one sentence to another and is taken from a small inventory of meanings. According to them it is possible to predict a context-sensitive meaning of every reciprocal sentence and that in a given context a sentence takes the strongest meaning that is consistent with known facts about the antecedent and the scope in the specific context. If the context is inconsistent with a high requirement then a weaker meaning is assigned to the sentence.
In this paper I will I argue for an opposite theory, that the NP-expressions do not encode reciprocity at all, and that in a given context they must be strengthened and interpreted as denoting symmetric relations. In order to support this claim I will provide ambiguities that were not discussed in the past, and also consider the historical origin of the constructions among the Semitic languages (discussed at length in Bar-Asher Siegal 2014) which can provide further support for my proposal.