TY - JOUR
T1 - Psychological explanations of deep and surface anaphora
AU - Murphy, Gregory L.
N1 - Funding Information:
* Some of the discussion presented here is based on a Ph.D. thesis submitted to Stanford University. I am glad to acknowledge the helpful comments of Barbara Malt, Ivan Sag, Thomas Wasow and especially Patricia Clancy and Herbert Clark on the thesis and/or earlier versions of this article. Preparation of this article was supported by NSF grant 83-15145.
PY - 1985/12
Y1 - 1985/12
N2 - A distinction has been made between two classes of anaphora. One class can take as antecedents not only linguistic constituents, but also objects and events in the extra-linguistic context. The other class accepts only certain linguistic elements as antecedents. It is not well understood why there should be two such classes, nor why a given form of anaphora is in one class rather than the other. This article attempts to explain the existence of the two classes by analyzing the difficulty a listener would have in recovering the antecedents of various forms of anaphora. This analysis suggests that it is intrinsically more difficult to discover the antecedents of some forms than of others, and that it is just these forms (with one exception) that have restrictions on acceptable antecedents. Therefore, the grammatical distinction between these two classes is not arbitrary, but subserves an important communicative function - that of ensuring that antecedents for anaphors are always recoverable.
AB - A distinction has been made between two classes of anaphora. One class can take as antecedents not only linguistic constituents, but also objects and events in the extra-linguistic context. The other class accepts only certain linguistic elements as antecedents. It is not well understood why there should be two such classes, nor why a given form of anaphora is in one class rather than the other. This article attempts to explain the existence of the two classes by analyzing the difficulty a listener would have in recovering the antecedents of various forms of anaphora. This analysis suggests that it is intrinsically more difficult to discover the antecedents of some forms than of others, and that it is just these forms (with one exception) that have restrictions on acceptable antecedents. Therefore, the grammatical distinction between these two classes is not arbitrary, but subserves an important communicative function - that of ensuring that antecedents for anaphors are always recoverable.
UR - http://www.scopus.com/inward/record.url?scp=33745871335&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33745871335&partnerID=8YFLogxK
U2 - 10.1016/0378-2166(85)90004-9
DO - 10.1016/0378-2166(85)90004-9
M3 - Article
AN - SCOPUS:33745871335
SN - 0378-2166
VL - 9
SP - 785
EP - 813
JO - Journal of Pragmatics
JF - Journal of Pragmatics
IS - 6
ER -