Mood, Modality and Force in GOLD

[Dunstan]

Dear All,

We've been considering GOLD features for inclusion in a database we
are creating. Looking at the list again, we would like to suggest the
following changes to the definitions of Mood, Modality and Force. This
is because the Mood and Force definitions appear to be too similar and
they don't capture the relationship between Mood and Modality, where
Mood is one of the morphological expressions of Modality, with the
latter being a semantic notion. Both ModalityProperty and
ForceProperty are not grammatical features, and should not be included
under Morphosyntactic Property but rather under Semantic Property.


SUGGESTED CHANGES
“MoodProperty is the class of properties that concern the grammatical
encoding of modality or force of a proposition. It is typically a
morphological verbal category and may be closely related to/interact
with the tense system of a language.”

“ForceProperty is the class of properties that concern the status of
the speech act associated with a proposition. Among the traditionally
defined values are Declarative, Imperative and Interrogative. Force is
sometimes grammatically encoded in the mood system of a language or
may be encoded through syntactic means.”

“ModalityProperty is the class of properties that concern the status
of the proposition itself, and not the event to which the proposition
refers [Palmer 2001, 1]. It typically concerns situations that need
not be real [Portner 2009, 1]. It represents those dimensions in the
domains of knowledge (epistemic modality), social relations (deontic
modality), ability (abilitative modality), and others. Modality is
sometimes grammatically encoded in the mood or evidential systems of a
language, and sometimes lexically encoded in modal verbs, auxiliaries,
or other parts of speech.”

ADDITIONAL SUGGESTIONS
*Conditional needs to be included under MoodProperty as well, since
some languages have a morphological conditional. It should probably
stay under ModalProperty too – I have just been looking at Portner’s
(2009) book Modality, which discusses how semanticists generally treat
the English conditional as containing a null modal. Thus, there would
be two types of Conditional (ConditionalMood and
ConditionalModality).
*Hortatory should be changed to Hortative, to be consistent with the
other –ives in ForceProperty, and also consitent with the example
given.
*EvidentialFeature should be EvidentialProperty for consistency. Or is
there a distinction made between Feature and Property?

These suggestions have arisen in discussion between:
Dunstan Brown
Marina Chumakina
Grev Corbett
Anna Kibort
Claire Turner

This is a particularly tricky area to deal with, which is another
reason why we have to strive to get it right.
Best,
Dunstan

[scott farrar]

Hi Dunstan, I don't think anyone has touched these concepts since some
Arizona students entered them into GOLD. I agree they need some major
work. Thanks for these concrete suggestions. Let me look into this and
I'll try to get you a new version with the changes.

Anyone else have any ideas on what needs to be added to Mood Modality Force ?

best
Scott