6
ought to be considered separate. When
MOOD
is distinguished from
MODALITY
, it is often
described as the category that refers to the objective evaluation of the truth of the statement by
the speaker (see Aronson 1991). As such,
MOOD
can be divided into two types:
REALIS
,
which
presents the contents of an utterance as a fact and corresponds fundamentally to the
INDICATIVE
;
and
IRREALIS
, which presents the contents of the utterance as non-factual and encompasses
CONDITIONAL
,
OPTATIVE
,
DESIDERATIVE
, and other hypothetical moods. Because
MOOD
and
SPEECH ACT
/
SENTENCE TYPE
are closely related, the grammatical correlates of
SPEECH ACT
(i.e.
SENTENCE TYPES
- see Sadock and Zwicky 1985) are sometimes referred to as
MOOD
(
INTERROGATIVE
mood,
IMPERATIVE
mood). Some authors (e.g. Cinque 1999)
refer to
EVIDENTIALITY
as a sub-class of
MOOD
,
and in traditional descriptions of Balkan languages,
evidentiality and related phenomena are also often referred to as types of mood (e.g. Bulgarian
preizkazvano naklonenie ‘reported mood’ and Albanian
mënyre habitore ‘admirative mood’).
Palmer (1986: 6, 8) treats
EVIDENTIALITY
as a closely related to
EPISTEMIC
MODALITY
, but
considers
EVIDENTIALITY
,
MODAL
verbs, and
MOOD
to be part of a broader category of
MODALITY
.
While it can useful to differentiate
MOOD
and
MODALITY
in
terms of an
objective/subjective distinction, some authors offer grounds for not making that distinction.
Bybee and Fleischman (1995, 2), for example, state that “
mood refers to a formally
grammaticalized category of the verb which has a
modal function”, while modality “is the
semantic domain pertaining to elements of meaning that languages express.” And Akatsuka
(1985) refers to the realis and irrealis moods, and combines
MOOD
and certain aspects of
MODALITY
by proposing an epistemic scale bounded by realis and irrealis. Much of the debate
over
the separate status of
MOOD
and
MODALITY
stems from the similarity of the two terms, both
7
of which originate in Latin
modus, and authors working in different traditions often mean very
different things when they refer to mood, modal verbs, and modality.
Scholars who do consider
MODALITY
to be a distinct category typically divide its
semantic domain into at least two subtypes:
DEONTIC
and
EPISTEMIC
.
DEONTIC MODALITY
involves “an indication of the moral desirability of the state of affairs expressed in the utterance”
and is typified by such English modal verbs as
should or
must (Nuyts 2005, 9). E
PISTEMIC
MODALITY
, then, “concerns an indication of the estimation…of the chances that the state of
affairs expressed in the clause applies in the world or not” and is typified by such English
adverbs as
maybe or
certainly,
and the modal verb might (Nuyts 2005, 10). Nuyts (2005, 7) also
includes the subtype
DYNAMIC MODALITY
,
which is “an ascription of a capacity/ability to the
subject-participant in the clause,” as is typified by the English constructions
can or
be able.
These disparate meanings can be traced back to the philosophical concept of
POSSIBLE WORLDS
,
with modals expressing the relationship between possible worlds and the real one.
Evidential meaning has been most closely associated with
EPISTEMIC MODALITY
,
which,
under some theories, necessarily expresses a speaker’s subjective opinion, since objective
opinion is typically considered to be
MOOD
. A number of scholars have noted, however, that
utterances that fairly
clearly express
EPISTEMIC MODALITY
may have both objective and
subjective interpretations, and that these objective interpretations are clearly not
MOOD
(Lyons
1977).
(3)
John may be in Indianapolis by now.
Theories that allow for modals to have two interpretations ascribe to (3) the following two
interpretations:
8
(3a)
O
BJECTIVE
:
Given John’s typical driving speed, and given the distance from here to
Indianapolis, it is possible that John is in Indianapolis.
(3b) S
UBJECTIVE
:
I believe that it is possible that John has reached Indianapolis, but I cannot
be sure.
The concept of
NON
-
CONFIRMATIVITY
that is found throughout this work would be considered a
subjective type of
EPISTEMIC MODALITY
, in as much it relates to the speaker’s personal
assessment of the contents of the proposition, and not to the speaker’s statement of some “good,
mathematically or formally reliable evidence” (Nuyts 2001, 393).
Although the term
status is not widely used, it is perhaps more suitable for describing
non-confirmativity, as
STATUS
has been specifically defined as a subjective category (Aronson
1991) and because non-confirmativity was specifically
formulated as a variety of
STATUS
(Aronson 1967; Friedman 1978; 1980). Nevertheless, it is also useful to refer to
MODALITY
, as a
number of recent works have discussed the properties of subjective
MODALITY
, particularly in
relation to their compatibility with certain types of questions and conditionals (Nuyts 2001;
Papafragou 2006). Because non-confirmative morphemes in Uzbek and Kazakh behave in
peculiar ways in the environment of questions and conditionals, reference to these works that
refer to
MODALITY
is necessary as well. In referring to the category that encompasses non-
confirmativity, I often employ the compound term
status/modality,
as the use of the term status
ties this work to previous scholarship on evidentiality in Eurasia (Jakobson 1957/1971; Aronson
1967; Friedman 1978; 1980), and the use of the term
modality allows for reference to a number
of relevant formalist works (DeHaan 1999; Nuyts 2001; Matthewson et al. 2007; McCready and
Ogata 2007).