Workshop description

Backgrounded Reports
Commitment and negation in parenthetical reports and reportative evidentials

Workshop, Nijmegen, January 15, 2016

There are various forms of reported speech in which the aspect of reporting is somehow backgrounded, or not-at-issue. and the content of the report itself, i.e., the complement or quoted phrase, serves as the main point. This backgrounding may be achieved syntactically, as in parenthetically framed direct or (free) indirect speech.

(1) "I'll be late again," she said, "so don't wait up."

(2) She would be late again, she said, so we shouldn't wait up.

Backgrounded reporting may also be realized morphologically, as in reportative evidential marking in languages like Cuzco Quechua (Faller 2002) or Cheyenne (Murray 2011), and in the reportative subjunctive or modals in German (Schenner 2007, Fabricius-Hansen&Saebo 2004) or ancient Greek (Bary&Maier 2014):

(3) Para-sha-n-si [Cuzco Quechua, Faller 2002]

     It is raining, I am told

(4) Er sagte, sie sei schön. Sie habe grüne Augen [German, Jäger 1971]

Finally, the backgrounding of the reporting may be entirely unmarked and left to pragmatics. This phenomenon is known as semantically parenthetical reporting (Simons 2007, Hunter et al. 2006, Hunter 2015). To illustrate, consider the dialogue in (5).

(5) A: Why was John not at the meeting?

     B: Oh, Mary said he's ill

The fact that Mary said something is no reason for John to miss a meeting. Hence, the at issue component of the report, i.e. what answers the question under discussion, is the complement, that John is ill.

A test often used to detect backgroundedness is based on the idea that parenthetical/not-at-issue content cannot be easily disputed or denied: saying "no, that's not true" typically targets only the at issue component of the previous speech act. Another test is based on the idea that not-at-issue material projects when embedded under negation or other operators. It remains to be seen to what extent the various phenomena above pass these tests and others, and what explains the differences and exceptions.

One of the principal challenges that comes up in the analyses of these phenomena is that there's a mismatch between what is at issue and what the speaker is committed to. In evidentiality research this is known as reportative exceptionality (AnderBois 2014, Koev forthcoming):

        (6) Dadating daw siya sa isang oras, pero hindo talaga. [Tatalog, Schwager 2010]

             will.come REP he in one hour but not really

            He says he will come in an hour, but in fact he won’t.

In recent years various semantic analyses have been proposed to analyze reportative exceptionality, but it remains to be seen if any of these can be extended to some of the other phenomena discussed above.

In this workshop we want to bring together the more or less disjoint strands of research on the different classes of backgrounded reporting to explore the possibilities of a unified approach of all these phenomena that takes into account the relevant facts about commitment, denial, and projection.

Invited speakers:
Regine Eckardt
Martina Faller
Julie Hunter
Todor Koev


Corien Bary & Emar Maier

ERC Projects PERSPECTIVE (Bary) & BLENDS (Maier)