This document explains how to fill out the Evidentials page of the Grammar Matrix Customization questionnaire. General instructions on using the questionnaire can be found here.

The standard reference for the Evidentials Library and its implementations is Haeger 2017. The full reference and .bib entry can be found here.

Evidentiality refers to the grammatical expression of information source. This may appear in a number of forms, such as in verbal inflection or in verbal auxiliaries. Within a language, evidential semantics can be organized into an inventory of terms. Examples of cross-linguistically common terms include a reported term, indicating that the speaker knows what s/he is saying because someone told him/her, and a firsthand term, indicating that the speaker perceived the event being discussed him/herself.

This version of the Grammar Matrix supports verbal inflection and auxiliary verbs as means of expressing evidentiality.

The Evidentials page allows you to define the semantic inventory of evidentials in your language. There are two options for how to do this: Select among common hierarchy elements and Specify your own hierarchy.

If you choose to select from common evidential terms, you will be able to choose among firsthand, non-firsthand, visual, non-visual sensory, inferential, reported, quotative, and "everything else". Note that the reported term is often called hearsay as well.

If you need terms not included in the pre-provided list, you need to specify your whole evidential inventory manually. You can specify any number of evidential terms by clicking the Add an evidential term button and entering the name of the term in the textbox that appears.

When an evidential inventory is defined on the Evidentials page, it allows a pseudo-feature to be specified on lexical rule types on the Morphology page or on auxiliary verbs on the Lexicon page. This is called a pseudo-feature because, while it is specified as a feature in the questionnaire, in the grammar produced by the Grammar Matrix evidential semantics are represented by elementary predications, not features on events.

In order to specify a verbal affix that expresses evidentiality, add a lexical rule type to a position class on the Morphology page. In the Features section, click the Add a feature button. Select evidential in the Name field and select the evidential term name in the Value field, and make sure the feature is specified on the verb.

In order to specify a verbal auxiliary that introduces evidential semantics, go to the Lexicon page and click the Add an Auxiliary Type button. Select the No predicate option under This auxiliary type contributes. Click the Add an auxiliary feature button and, as with verbal inflection, choose evidential and the name of the term in the Name and Value fields, respectively, and make sure the feature is specified on The auxiliary.

When a lexical rule type or auxiliary verb specifies a value for the evidential feature, rather than adding a feature to events, as is done with tense, aspect, and mood, an independent elementary predication is introduced. The PRED value of this predication is of the form ev_<evidential-name>_rel. The EP for the main event being communicated is linked to the ARG1 of the evidential predicaiton. As an example, below is a snippet of the code of a grammar that expresses a firsthand evidential term by verbal inflection, found in your_language.tdl:

evidential-lex-rule := cont-change-only-lex-rule & same-spr-lex-rule & same-spec-lex-rule &
  [ C-CONT [ RELS <! event-relation &
                     [ LBL #ltop,
                       ARG0 event,
                       ARG1 #harg ] !>,
             HCONS <! qeq &
                      [ HARG #harg,
                        LARG #larg ] !>,
             HOOK [ LTOP #ltop,
                    INDEX #mainev,
                    XARG #mainagent ] ],
    DTR.SYNSEM.LOCAL.CONT.HOOK [ LTOP #larg,
                                 XARG #mainagent,
                                 INDEX #mainev ] ].

firsthand-evidential-lex-rule := evidential-lex-rule &
  [ C-CONT.RELS <! [ PRED "ev_firsthand_rel" ] !> ].

Auxiliary verbs expressing evidentiality leverage the same tdl as auxiliaries expressing other elementary predications. Below is a snippet from the your_language.tdl file of a grammar with a directive evidential auxiliary verb:

arg-comp-aux-with-pred := arg-comp-aux & hcons-lex-item &
  [ SYNSEM [ LOCAL.CONT.HCONS <! qeq &
                                 [ HARG #harg,
                                   LARG #larg ] !>,
             LKEYS.KEYREL event-relation &
                          [ ARG1 #harg ] ],
    ARG-ST < [ ],
             [ LOCAL.CONT.HOOK.LTOP #larg ] > ].

dir-aux-lex := arg-comp-aux-with-pred &
  [ SYNSEM.LOCAL.CAT.VAL.COMPS.FIRST.LOCAL.CAT.HEAD.FORM finite ].

And below is the lexical entry for that auxiliary in lexicon.tdl:

dir := dir-aux-lex &
  [ STEM < "dir" >,
    SYNSEM.LKEYS.KEYREL.PRED "ev_directive_rel" ].

One current limit on the evidentials library is related to limits on word order in the auxiliary system. Currently, only one word order for auxiliaries and their complements is allowed per language. However, some languages, such as Basque, have different word orders for their evidential and tense/aspect auxiliaries; thus, as of this writing, only one of the two auxiliary systems can be produced by the Grammar Matrix in a single grammar.

Aikhenvald, A. (2006). Evidentiality. Oxford linguistics. Oxford University Press.

• bibtex:

  @book{Aikhenvald:06,
  title={Evidentiality},
  author={Aikhenvald, A.Y.},
  series={Oxford linguistics},
  year={2006},
  publisher={Oxford University Press}
  }

Haeger, Michael. (2017). An Evidentiality Library for the LinGO Grammar Matrix.

• bibtex:

  @mastersthesis{Haeger:17,
  author = {Michael Haeger},
  year = {2017},
  title = {An Evidentiality Library for the LinGO Grammar Matrix},
  school = {University of Washington}
  }

Murray, S. E. (2017). The Semantics of Evidentials. Oxford University Press.

• bibtex:

  @book{Murray:17,
  title={The Semantics of Evidentials},
  author={Sarah E. Murray},
  year={2017},
  publisher={Oxford University Press}
  }