Divergent conditioning among unexpressed subjects in English:

The role of referentiality

​

Amy Lindstrom

Old Dominion University

​

Abstract

​

Ample research exists on variable subject expression in null subject languages; however, the conditioning of unexpressed subjects in English has been underexplored, primarily due to the idea that expressed subjects are obligatory.

Existing research on subject expression shows that referential selection is largely dependent on continuity, where more activated topics and accessible participants are more likely unexpressed (cf. Givón 1983). What is not known, however, is whether this holds with nonreferential subjects (4-5).

 

This study examines the factors conditioning referential and abstract null subjects in English. 1500 tokens of 3rd person unexpressed subjects and their competing pronominal variants were extracted from corpora representing interactive conversation (SBCSAE) and monologic narrative (Pear Stories).

 

Results show that 3rd person referential subjects are unexpressed where prosodic linking is tightest (1) and decrease proportionally to lesser cohesion    (2-3).

 

1)...she was happy with that and Ø hung up.                                                                (91%)

 

2) he comes over there,    continuing intonation (comma)

    and Ø is talking with that woman.                                                                               (46%)  

3) Because they didn't understand.   final intonation (period)

    ..And they were no longer in control.                                        (23%)

 

 

Other significant variables include a chronological relationship between verbs and the oft-cited cross-linguistic constraint of switch reference. In essence, referential unexpressed subjects are conditioned by higher degrees of cohesion, measured by prosodic linking, sequential events and referential maintenance.

 

On the other hand, abstract it expression is unaffected by cohesion and is instead sensitive to the semantic class of the accompanying verb. Patterns of grammaticalization emerge from these findings, indicating that in the case of abstract it, collocations indicating speaker subjectivity lack an expressed subject (4), while those occurring with be (5) are nearly always expressed.

 

4) Ø Seems like any time I've seen a wrecker out here,
     there's always two guys in there.                                    

 

5) ... (H) I don't know what it was.
    But it was like I went [to]
    Bahia,... last Sunday,                    

 

These results support what Erker & Guy (2012) refer to as an autonomy effect, whereby frequent form pairings, whether with overt or null pronouns, become entrenched in these associations.

 

The distinction between the discourse/pragmatic factors affecting referential subject expression and the grammaticalizing effects of abstract subject expression illustrates a division among the ‘category’ of 3rd person, where continuity in discourse is contingent on referentiality.

​

References

 

Chafe, Wallace, ed. 1980. The Pear Stories: Cognitive, Cultural and Linguistic         Aspects of Narrative Production. Norwood, NJ: Ablex.

 

Du Bois, John W., Chafe, Wallace L., Meyer, Charles, and Thompson, Sandra A.

2000.     Santa Barbara corpus of spoken American English (SBCSAE),

Parts 1-4. Philadelphia: Linguistic Data Consortium.

 

Erker & Guy. 2012. The role of lexical frequency in syntactic variability: Variable subject personal pronoun expression in Spanish. Language. 88(3): 526-              557.

Givón, T. 1983. Topic continuity in discourse: An introduction. In T. Givón (ed),    Topic     continuity in discourse: A quantitative cross-linguistic study, 1-41.                 Amsterdam: John Benjamins.