GB522 - grambank/grambank GitHub Wiki

Can the S or A argument be omitted from a pragmatically unmarked clause when the referent is inferable from context (‘pro-drop’)?

Summary

This feature identifies languages that do not require a phonologically overt S or A argument expressed by a noun phrase or phonologically independent pronoun when the referent can be inferred from context. These languages are often called ‘pro-drop’ languages, and the phenomenon is generally referred to as ‘null anaphora’ or ‘zero anaphora’.

Here ‘omitted’ means phonologically unexpressed (we do not intend for this feature to make any distinctions regarding the theoretical status of a null argument). To count as 1, pro-drop should occur where the S or A argument might otherwise be expressed by an independent pronoun or full noun phrase. Specifically, pro-drop should occur in simple, declarative main clauses.

Pro-drop is not mutually exclusive with indexing of the S or A argument on the verb. If, in the absence of an independent pronoun or nominal S or A, there is indexing of this argument on the verb, you may still code 1.

Procedure

  1. If the S or A argument may be omitted in a pragmatically unmarked main clause when its referent can be inferred from context, code 1.
  2. If the S or A argument is always phonologically expressed by a noun phrase or phonologically independent pronoun in a pragmatically unmarked main clause, code 0.
  3. If the S or A argument may be omitted in a pragmatically unmarked main clause when its referent can be inferred from context, but only if that referent represents a certain person/number category, code 1 and provide a comment that describes under what condition the S or A argument may be omitted.

Examples

Czech (ISO 639-3: ces, Glottolog: czec1258)

Czech allows S and A arguments to be expressed by pronouns when the referent is understood from context, but it also allows these subject arguments to be phonologically unexpressed.

a. Jan  včera      navštív-il
   Jan  yesterday  visit-PST.M
   ‘Jan visited yesterday.’

b. (on)     říd-il        auto
   (3SG.M)  drive-PST.M   car
   ‘He drove a car.’

c. (on)      m-ěl         tři    kufr-y
   (3SG.M)   have-PST.M   three  suitcase-PL
   ‘He had three suitcases.’

The pronouns in parenthesis in the examples above can be omitted, and in fact would be in most natural contexts. The resulting constructions with no phonologically overt subjects where a pronoun would otherwise occur would lead to a 1 code for Czech.

Aneityum (ISO 639-3: aty, Glottolog: anei1239)

Subjects are always expressed in Aneityum by either a lexical noun phrase or a phonologically independent pronoun (or both, in the case of the subject topicalisation construction) (Lynch 2000: 114–116).

a. et       apam  plen
   3SG.AOR  come  plane
   ‘The plane is coming.’ (Lynch 2000: 115)

b. et       itiyi  hag  aan  a     nuyaleg
   3SG.AOR  NEG    eat  3SG  TEMP  morning
   ‘He/she didn’t eat in the morning/this morning.’ (Lynch 2000: 123)

Though it is not relevant to the simple declarative clauses targeted by this feature, it is interesting to note that this language even overtly expresses subjects with phonologically independent pronouns in imperative clauses:

c. adia     aak
   go.away  you.SG
   ‘Go away!’ (Lynch 2000: 137)

Aneityum does not allow pro-drop and is thus coded 0.

Further reading

Chomsky, Noam. 1981. Lectures on government and binding: The Pisa lectures. Holland: Foris Publications.

Gilligan, Gary Martin. 1987. A cross-linguistic approach to the pro-drop parameter. Los Angeles: University of Southern California. (Doctoral dissertation.)

References

Lynch, John. 2000. A grammar of Anejom̃. (Pacific Linguistics, 507.) Canberra: Research School of Pacific and Asian Studies, Australian National University.

Related Features

Patron

Hannah J. Haynie