Mood in Morphology
Summary and Keywords
The category of mood is closely related to modality, though specifically involves grammatical (inflectional) means for expressing core modal meanings (most notably, those of possibility and necessity). In other words, mood is defined as modality that is grammaticalized in the verbal system as an inflectional category. The category of mood is found in nearly all full-fledged inflectional verbal systems, along with the categories of aspect and tense. The typical opposition expected within the system of moods is the division into “indicative” and “non-indicative” moods, dependent on the real vs. irreal (or, more precisely, asserted vs. non-asserted) status of the proposition.
There is no “preferable” morphological device for the expression of mood in the world’s languages—all the existing grammatical means are in demand, both synthetic and periphrastic. Among the segmental markers of mood affixal marking prevails, involving both prefixes and suffixes and various combinations thereof (yielding circumfixal marking). Non-segmental and suprasegmental marking of mood is less frequent, but also quite common. Another strategy for mood marking in the languages of the world is suppletion, when inflectional modal meanings require a different stem feeding into the verbal paradigm.
Along with dedicated morphological markers of mood, there exists a plethora of cumulative types of marking, when mood is expressed simultaneously with other verbal categories, such as tense, aspect, voice, person, number, and possibly some others.
The structure of mood as a grammatical category poses a challenge for universal typological descriptions, as the diversity of all its guises in the world’s languages is notoriously high. Imperative and subjunctive are regarded as the two core non-indicative members of mood domain attested cross-linguistically. A kind of terminological complication may arise with respect to the terms indicative vs. subjunctive and realis vs. irrealis. Still, there exist some points that reveal the differences between subjunctive and irrealis, syntactic distribution being one of the most essential (given that subjunctive is to be considered primarily as a morphological device for expressing syntactic subordination).
Of course, the systems of mood in the world’s languages often display a greater diversity within the domain of non-indicative moods, and specifically epistemic and volitive values grammaticalize to separate inflectional forms, comprising various epistemic and optative moods respectively.
1. Mood as a Grammatical Category
This article provides an overview of the category of mood and its morphological expression in the world’s languages. The category of mood is closely related to modality (see Nuyts, 2005 for a concise presentation of the whole semantic domain and related issues), though specifically involves grammatical (inflectional) means for expressing core modal meanings of possibility and necessity. In other words, mood is defined as modality that is grammaticalized in the verbal system as an inflectional category.
A similar view on mood is shared, among many others, by Lyons (1977), Bybee, Perkins, and Pagliuca (1994), Palmer (2001), and Thieroff (2010). Notably, Bybee et al. (1994, p. 181) suggest the following way of distinguishing mood from modality: modality is regarded as a conceptual domain and mood as its inflectional expression. Among the grams that are traditionally associated with mood, one can name, for instance, subjunctive, optative, and imperative (although the list can be much broader). These categories are united on the basis of expression of “speaker-oriented” (the term suggested in Bybee et al., 1994, p. 177) modal meanings, which tend to grammaticalize from “agent-oriented” modalities (corresponding to “subject-oriented modality” in van der Auwera & Plungian, 1998). The latter are rarely expressed by inflectional categories, while the former frequently take on inflectional expression (see Bybee, 1985). Accordingly, this article is aimed mostly at the morphological expression of mood.
The category of mood is found in nearly all full-fledged inflectional verbal systems, along with the categories of aspect and tense (the well-known TAM triad) (see Dahl, 1985). The typical opposition expected within the system of moods is the division into “indicative” and “non-indicative” moods, dependent on the real vs. irreal (or, more precisely, asserted vs. non-asserted) status of the proposition. Terminological conventions accepted in this article are generally in line with Palmer (2001) and Nuyts and van der Auwera (2016), although some issues are discussed along the way (cf. the notion of subjunctive in Section 3.2).
The frequently cited study by Bhat (1999) offers the division of languages into tense-, aspect-, and mood-prominent types, where the notion of prominence presupposes a higher degree of grammaticalization of one of the TAM categories, resulting in its more obligatory, systematic, and ramified expression. Among the mood-prominent language types, Bhat mentions, for instance, Kayardild (an Australian Tangkic language of south Wellesley Island), where the final inflections of verbal forms express primarily modal meanings. In another language, Caddo (Caddoan language family, United States), there are two different sets of pronominal prefixes, which are attached to the verb to indicate its real or irreal status, so that both members of the realis/irrealis opposition become marked. This survey proves that the category of mood is not only present in various language systems, but is able to play a dominant part in them.
This article is organized as follows: Section 2 is dedicated to the morphological means of mood expression, considering both segmental and non-segmental ones. In Section 3, the categories included under the heading “mood” are described, with a special focus on imperative and subjunctive as the two core non-indicative members of mood domain attested cross-linguistically. Section 4 provides concluding remarks.
2. Morphological Expression of Mood
There is no “preferable” morphological device for the expression of mood in the world’s languages—all the existing grammatical means are possible, both synthetic (different types of affixes) and periphrastic, though the latter are not our primary concern here.
2.1 Segmental Markers of Mood
One of the widespread ways of encoding mood is affixal marking, involving both prefixes and suffixes and various combinations thereof. Thus, a prefix be-, which is confined to marking imperative and subjunctive, is found in Persian (see Perry, 2007). The imperative is formed by adding the prefix to Stem I (traditionally called a “present stem”) with singular (-ø) or plural (-id) endings:
The imperative form in singular is used “only to familiars and social inferiors; the plural may be used additionally to address one person in polite usage” (Perry, 2007, p. 1001). The prefix be- also serves to form present subjunctive, which is observed in sentential complements of verbs expressing command, desire, hope, fear, and other irreal modalities. Such complements are usually introduced with the help of the broad complementizer ke (2)–(3).
Turkic languages exhibit a wide range of mood-marking suffixes. Some of the suffixes have undergone peculiar shifts in their meaning within the domain of modality (the diachronic development of these markers is analyzed in Rentzsch, 2015, among others). The imperative paradigm varies among different Turkic languages with regard to the singular imperative form. It can be expressed by the plain verb stem in many modern Turkic languages, although there are cases, such as Modern Uyghur, where singular imperative is ascribed its own suffix -(X)ŋ, with -(X)ŋlar as its plural counterpart (4).
According to Rentzsch (2015, p. 182), the plain stem (e.g., kör) is confined to familiar or derogatory settings with one addressee. Nevertheless, Turkic languages are quite consistent in plural imperative suffixal marking, unlike, for example, Manchu and Classical Mongolian, where the plain verb stem is readily applicable to both singular and plural addressees.
Turkic optative is predictably characterized by suffixal marking as well, where a widely disseminated optative suffix has developed from the Old Turkic prospective in -GAy (for the semantic development of this suffix and some innovative optative markers, see Rentzsch, 2015, pp. 187–194). An illustration comes from Modern Qazaq (5) and Uyghur (6).
It is also possible to encounter circumfixal marking of mood categories in the world’s languages. One of the examples is the desiderative mood (glossed as desid) in Telqep Chukchi (described in Dunn, 1999), which is formed by a circumfixal marker (a prefix re-/ ce and a suffix -ŋ).
Along with specialized morphological markers of mood, there exists a plethora of cumulative types of marking, when mood is expressed simultaneously with other verbal categories, such as person, number, tense, and so forth. Indo-European languages provide numerous examples of cumulation of subjunctive with grams of person and number, where fusion may reach a considerable degree. Compare, for example, French present and perfect subjunctive forms in (8a–b), which differ in the aspectual interpretation (quoted after De Mulder, 2010, p. 166).
The Telqep variety of Chukchi has already been mentioned with regard to the circumfixal desiderative marker. Prefixal marking of other mood categories is also present in Telqep Chukchi: the intentional and conditional moods are marked by prefixes fused with person and number markers (see Dunn, 1999, p. 188). The use of the intentional mood with its prefixal marking is not confined to intended or hypothetical actions only (9), but is also compatible with imperative contexts (10).
Returning to the Indo-European languages, one can mention Latin as a familiar illustration of cumulative expression of mood and tense. They are found in Latin subjunctive in all its temporal variations: present and imperfect (with the synthetic forms of both active and passive voice), perfect and pluperfect (with passive voice formed periphrastically). The verb agere ‘drive; act, do’ in present and imperfect subjunctive is displayed in (11)–(12), where the independent use of subjunctive refers to its “deliberative” function: the speaker asks for a possible or necessary way of acting (see Palmer, 2001, p. 128).
The present subjunctive in Latin is formed with the help of a vowel suffix: -e- for the verbs of the first conjugation (ornare type) and -a- for the three remaining conjugation types. This system is largely preserved in modern Romance languages: for a similar vowel alternation in Portuguese, see Becker (2010, p. 180), and in Spanish, see Laca (2010, p. 200).
Cumulative exponents of mood and other verb categories are expected in fusional (or inflected) languages, which tend to use a single inflectional morpheme to denote multiple grammatical features. Fusion is considered as a characteristic feature of Semitic and Indo-European (Slavic, Baltic, Romance, Iranian, etc.) languages, but is not limited to them. On the other hand, agglutinative languages, with their tendency to express one grammatical category per affix, usually possess independent markers for the categories of mood. Several examples may be given from the Uralic language family, which is a famous hotbed of agglutinative morphology.
Tommola (2010) describes three morphological moods marked in the Finnish verbs, apart from the Indicative—namely, Imperative, Conditional, and Potential. The formation of Imperative will be discussed further with respect to the stem alternations it triggers; the Conditional forms are derived by adding the suffix -isi- to the verbal stem, and the Potential forms—with the help of the suffix -ne-, both getting the personal endings after the markers of mood (so that any kind of cumulation is avoided). The Finnish Conditional is used to refer to situations that are considered counterfactual, hypothetical, unreal, or desirable by the speaker (Tommola, 2010, p. 519). In (13) the Past Conditional (formed periphrastically with the help of the verb ‘to be’) is used to encode a situation that was possible in the past, but was not realized.
The Potential suffix -ne- is diachronically older than the Conditional suffix, and has cognates in many Finno-Ugric languages (see Tommola, 2010, p. 522). Some of them use the reflex of this suffix a Conditional marker (cf. the affix -n- of the Conditional mood in Hungarian in De Groot, 2010, p. 563; see also Majtinskaja, 1974, pp. 312–313). In Mari (both Meadow and Hill varieties), however, the same suffix -ne- has lost its conditional meaning and denotes the desiderative mood (cf. (14) from Hill Mari).i
2.2 Non-Segmental and Suprasegmental Markers of Mood
So far only segmental morphemes denoting different mood categories have been discussed. Nonetheless, as well known, one may encounter occasional examples of verb forms that bear a grammatical feature without a particular segmental marker of this feature. The formation of Russian imperative can serve as a good example of this sort. The singular imperative form is derived from the so-called present tense stem with the help of either a segmental marker (suffix -i) or a non-segmental marker, sometimes accounted for as a “zero suffix,” which causes a consonantal alternation: hard (non-palatalized) consonants in the present tense stem, ̴ soft (palatalized) consonants in the imperative (15).
Turning back to the Finnish imperative, the second person singular form is “a minimal form of the verb, expressed by the simple present stem plus a certain type of sandhi” (Tommola, 2010, p. 514). This modification is preserved as a reminder of the old imperative morpheme *-k, going back to the Uralic protolanguage.
Another example is ablaut or umlaut vowel alternations, which can be found in German subjunctive, cf. ich hätte ‘I have.sbjv.1’.
Apart from the segmental and non-segmental markers of mood, the possibility of suprasegmental marking should be mentioned. Nurse (2003) reports on a combination of segmental and tonal means that are used to encode Subjunctive in Lega (Bantu). The form of Subjunctive has a high toned subject marker and a final high tone suffix -e while all syllables in between are low. This tonal pattern was reconstructed for Bantu Subjunctive in Meeussen (1962, p. 74), and only the Lega language is considered to retain it.
Another possible strategy for the languages of the world is to provide a special form (which is derived from a different stem) in a verb paradigm with a certain grammatical function. Suppletion in verb paradigms frequently tends to be category-specific in that it usually involves the tense and aspect categories or the imperative mood.
Suppletion with regard to the imperative mood (which concerns both imperatives and hortatives, specific directive forms including the first person) seems to be correlated with specific lexemic groups and geographical areas (cf. Veselinova, 2013). The dominant meanings that have a tendency to be marked with special imperative forms are ‘to come / to go’ (cf. the imperative forms of the verb ‘to come’ in Egyptian Arabic in (18)) and ‘to give’. Veselinova (2017) argues that this distribution may serve as an argument in describing suppletion as a functionally motivated phenomenon (as one of the most common uses of imperatives is the verbs of motion).
Suppletion within the imperative mood is widespread among the language families of the African continent (and its eastern parts in particular), as well as in the languages of the Balkans and the Caucasus (Veselinova, 2013).
Another example of suppletion within the domain of mood comes from Latin, where the verb esse ‘to be’ has a suppletive paradigm in subjunctive: sim (present subjunctive), fuerim (perfect subjunctive), and fuissem (plusquamperfect subjunctive). Although some of these stems are in fact etymologically related, they can be synchronically analyzed as suppletive and are inherited and preserved by Modern Romance languages (for instance, the stem fu- is found in Spanish and French, fo- in Italian and Portuguese). Similarly, Icelandic subjunctive forms (sé, sért, etc.) of the verb vera ‘to be’ are highly irregular due to their suppletive formation.
2.4 Analytic Markers
The main focus of this article is the synthetic morphological marking of mood; nevertheless, the world’s languages often have a wide range of analytic markers “at hand,” such as imperative-hortative particles or special optative constructions. The Russian construction with davaj(te) ‘let’s’ (this particle comes from the imperative form of the verb davat’ ‘to give’) may serve as an illustration of an analytic hortative.
A hortative construction with synonymous particles ajda(te) / äl(dä) + verb in 1pl of the non-past tense is found in Hill Mari.
The English hortative with let’s itself provides a good example of an analytic construction, where a full-fledged verb let has conflated with a personal pronoun and grammaticalized into a hortative particle.
Sometimes particles are used as intensifiers of the categorial meaning. The example in (21) from Hill Mari (see Mordashova, 2017) may be easily interpreted as optative (expressing the wish of the speaker) without the particle tek ‘let’, given a single jussive form of the verb liäš ‘to become’ (jussive stands for a third person directive form).
Numerous extensions of the topic related to the periphrastic marking of mood fall outside the scope of this article.
3. Structure of the Category
The structure of mood as a grammatical category poses a challenge for universal typological descriptions, as the diversity of all its guises in the world’s languages is notoriously high. Recall that mood is considered a language-specific category, expressed through inflection in distinct sets of verbal paradigms, while modality can be regarded as a typologically universal domain with a broad semantic scope of meanings, not necessarily expressed through inflection. According to Dahl (1985, p. 26), “one generalization that can be made is that the typical opposition is between indicative and non-indicative moods, where the indicative is always the more ‘real’ or ‘asserted’ member of the opposition.”
Among the categories of mood, imperative is the most likely to claim the status of a universal category, which is found in all or almost all languages, as opposed to other moods. Language speakers simply need basic means to express a statement, a question, or a command; the latter is commonly expressed by imperative mood, although there may be other devices such as subjunctive or future verb forms and indirect speech acts (such as It’s too cold here, hinting to close the window or bring a warm blanket to the speaker).
Imperatives worldwide have been a topic for many typological studies (cf. Aikhenvald, 2010; Gusev, 2013), where some common myths about the apparent “simplicity” of imperatives are being debunked. It is true that in many languages imperatives are rather short forms that may often have no overt marking at all (a bare verbal stem is used). Imperatives tend to have a meager system of TAM distinctions, as opposed to declarative or interrogative sentences. However, if we divert our attention from the familiar Indo-European languages, it will become obvious that “elsewhere in the world, imperatives have what other words do not have” (Aikhenvald, 2010, p. 5). In particular, imperatives may possess categories that are specified for the imperative mood only: notably, distance in space and time markers come to play here. Aikhenvald (2010) provides an astonishing example from Tariana (northwest Amazonia), where immediate, postponed, proximal, and distal imperatives are distinguished.
Imperatives are also interesting from both phonetical and syntactical points of view. The same content within an imperative speech act may sound as a harsh command or a mild request, depending on the intonation contour and syntactic shape. Though the subject of a classical imperative is invariably the addressee (and the overt subject “you” is often avoided), the explicit expression of a subject within an imperative sentence may lead to some curious consequences. In English, for instance, it may contribute “to a somewhat impatient, irritated, aggressive, or hectoring effect” (Huddleston, 2002, p. 926), or, on the contrary, may have “very much the opposite effect of soothing reassurance, encouragement, support.”
Apart from commands directed at the second person, one could also think of commands oriented at “me” or “us” (usually called “hortatives,” “exhortatives,” or “adhortatives”) or at a third person (known as “jussives” or “injunctives”). Aikhenvald (2010) treats them as “non-canonical” imperatives, whereas canonical ones are those directed at the addressee. Moreover, for such languages as Evenki (Tungusic) or Una (Papuan) it is the most appropriate decision to claim that imperatives form one paradigm for all the persons there. In Evenki, for instance, there are two imperative paradigms, related to as “near future” and “remote future” imperatives. Both paradigms include all person–number combinations with special suffixes, where first person plural inclusive and exclusive are distinguished.
The issue of morphological expression of imperatives-hortatives in the languages of the world has been addressed in van der Auwera, Dobrushina, and Goussev (2013). This paper (and the related WALS map) distinguishes between four types of imperative-hortative systems, which rely on the parameter of formal homogeneity (maximal vs. minimal) with the imperative second singular. More precisely, two imperative-hortative forms are called homogeneous if they are formed using the same kind of morphological (affixation, affix ordering) or syntactic (pronouns, particles) means. Homogeneity is accounted for as maximal if the second singular imperative is formally homogeneous with the other second persons, with the third persons, and with at least the inclusive first person plural. On the other hand, if a language has a system with an imperative second person singular that is not formally homogeneous with any of the other forms, then the language is said to have a “minimal system.” Not surprisingly, many languages find themselves beyond the scope of simply maximal or minimal systems, and this issue (as well as statistical representation and geographical distribution) is also referred to in van der Auwera et al. (2013).
Finally, it is worth mentioning that a language may have more than one imperative form. According to Aikhenvald (2010, p. 7), Tucanoan languages boast up to 11 imperatives. This number is achieved due to verbal affixes, encoding special directive meanings (invitation, recommendation, request, etc.). Here is an example from Tuyuca, where a neutral imperative is marked with the suffix -ya (23a), whereas (23b–d) represent invitation, permission, and warning, each marked with its own suffix. The most intriguing use is that of (24), “urging the addressee to confirm the result, or to see for themselves” (Aikhenvald, 2010, p. 201).
The use of imperatives is not restricted to the main clauses only. In some languages imperative forms are accepted in dependent clauses (especially conditional and concessive; see Aikhenvald, 2010, pp. 234–255 for further details), where they lose their original meaning of commands. Sentence (25) is a relevant illustration from Russian, where the imperative acquires a conditional meaning. This topic will not be elaborated on in this article.
Another category frequently observed in the world’s languages is subjunctive mood, which is opposed to indicative mood on the grounds of the “assertion” and “non-assertion” dichotomy. Indeed, the use of subjunctive does not depend on what is factual or non-factual, but rather on what is asserted or not. This is made clear with the case of Spanish subjunctive, described by Lunn (1995) and cited by Palmer (2001, p. 3), where subjunctive is used in several types of subordinate clauses that are “unworthy of assertion.”
A kind of a terminological complication may arise with respect to the terms indicative vs. subjunctive and realis vs. irrealis (see, e.g., van der Auwera & Schalley, 2004). As Palmer (2001, p. 5) puts it, there is no clear-cut typological difference between these categories, as both can be regarded as instances of the typological categories of Realis and Irrealis. In the grammatical descriptions of European languages the indicative/subjunctive opposition has been accepted, whereas studies of some other languages (e.g., Native American or Papua New Guinea languages) have been made in terms of realis and irrealis.
Nevertheless, there still exist some points that reveal the differences between subjunctive and irrealis. One can name their syntactic distribution: subjunctive is used mainly in subordinate clauses and the semantic foundations for such syntactic behavior on the synchronic level are often blurred, whereas irrealis markers are common within both main and subordinate clauses. When it comes to the categorial semantics, both subjunctive and irrealis may be called “notionally redundant” (see Palmer, 2001, p. 186). This refers to language systems where the appearance of subjunctive is determined by a certain type of complementizer, or where irrealis is required by certain grammatical markers within a sentence (e.g., future, negation). This is observed in so-called “joint” systems, such as the one in Amele (Papuan) illustrated in (29), where the marker of irrealis is inevitably triggered by the future suffix. “Non-joint” markers of irrealis, on the contrary, occur in isolation and serve themselves as indicators of certain notional categories (for a broader discussion, see Palmer, 2001, pp. 145–167; see also Plungian, 2005 for an overview of irrealis and modality in Russian).
This section is concentrated mostly on subjunctive mood and its functions within main and subordinate clauses. As Jespersen (1924, p. 314) puts it, one of the functions of the subjunctive is simply that of being subordinate, as it is the mood typically used in subordinate clauses. However, some languages (Latin and its Romance descendants among them) provide a range of examples with subjunctive used in the main clause. Let us consider this option first.
3.2.1 Subjunctive in Independent Clauses
Most of the occurrences of subjunctive in independent clauses acquire directive or optative interpretation. Directives expressed by a subjunctive can be oriented at a third person (jussives), at the first person plural (hortatives), or even at the prototypical for imperatives second person—in such cases subjunctive usually expresses a polite request, although the verbal form is normally third person singular. These possibilities are shown in (30)–(32) from French and Catalan.
(30) French (De Mulder, 2010, p. 167)
(31) Catalan (Quer, 2010, p. 225)
(32) Catalan (Quer, 2010, p. 225)
(33) Italian (Squartini, 2010, p. 247)
(34) French (De Mulder, 2010, p. 167)
(35) French (De Mulder, 2010, p. 167)
(36) Portuguese (Becker, 2010, p. 182)
Finally, subjunctive is widely accepted in epistemic contexts, when the speaker has doubts about the veracity of the proposition. In Spanish and Portuguese it appears under the scope of an epistemic modal (and not vice versa; cf. (37a–b)), and in Italian—in interrogative clauses with mandatory complementizer che (38).
(37) Spanish (Laca, 2010, p. 217)
(38) Italian (Squartini, 2010, p. 247)
All these uses concur in displaying non-assertive status of the proposition. The speaker, choosing the subjunctive form, signals that the proposition is not a part of what is to be considered a true representation of reality, as the situation encoded by subjunctive is, roughly, either hypothetical or located in the future.
3.2.2 Subjunctive in Dependent Clauses
Let us now turn to the (synchronically) primary function of subjunctive, traditionally described as a mood of subordinate clauses. The types of clauses where subjunctive may occur include complement, adverbial, and relative ones. In complement clauses subjunctive is usually required by linguistic expressions conveying will, orders, or demands (39a), feeling or subjective appreciation (39b), and doubt (39c).
(39) French (De Mulder, 2010, pp. 168–169)
Subjunctive is often used to indicate indirect speech. This can be viewed as a peculiar feature of Germanic languages (cf. (40) from Icelandic), in which they differ from Modern Romance languages that commonly have indicative in verba dicendi complements. According to Sigurðsson (2010, p. 50), in Modern Icelandic, the most important factor that triggers subjunctive marking in these complements is that the speaker does not take responsibility for their truthfulness. The same takes place in German, though colloquial style may violate the “rule” of subjunctive marking of reported speech (41).
Among adverbial clauses where subjunctive is appropriate one can mention temporal, purposive, and concessive clauses. The latter type is somewhat curious, because the content of concessive clauses normally does not presuppose any irrealis components. Thus, it seems that the occurrence of subjunctive after although-complementizers is motivated syntactically rather than semantically. However, De Mulder (2010, pp. 172–173) suggests a possible explanation for subjunctive in concessive contexts.
Although the content of the clause may be presented as already part of the speaker’s and the interlocutor’s representation of reality, these subordinate clauses are not incompatible with the subjunctive’s meaning as defined before, since the consequential relation that should normally have been instantiated, is absent, suggesting that the event in the subordinate clause did not have the effects foreseen and that other, alternative causes were more efficient. This use of the subjunctive thus is compatible with the idea that the speaker does not consider the event to be really part of his or her representation of reality.
This, nonetheless, does not seem to withstand the contradictory examples from Italian reported by Squartini (2010, p. 246). Although subjunctive is required by the conjunction sebbene (42a), anche se (42b) triggers the Indicative, despite the semantic similarity of the two conjunctions.
In relative clauses the use of subjunctive is restricted to non-specific contexts, as in (43a) from Spanish, where the existence of an object verifying the description in the relative clause is not entailed. (43b), on the other hand, presupposes the existence of a certain employee, whom I am looking for and who can, in addition, speak English.
Diachronically, subjunctives used in subordinate clauses in the languages of the world usually start as forms encoding agent-oriented modalities. Subordinating uses (with verbs of desire or will) tend to be semantically motivated at first, as the dependent proposition acquires an undoubtedly irreal, non-asserted status (formal semanticists would suggest that it has to be interpreted with respect to the worlds different from the real world). Later, the use of subjunctive is being generalized on a wider range of contexts—on other types of subordinate clauses, where it would not have been treated as appropriate originally. Thus, when the semantic contribution of a subjunctive becomes weaker, the syntactic dependencies strengthen, subjunctive becoming just a formal marker of a certain clause type. The development of subordinating moods and their further destiny is addressed in Bybee et al. (1994).
3.3 Other Moods?
Imperative and subjunctive tend to be the most widespread categories of mood in the world’s languages. However, there exist rather rich systems where there are many more than one non-indicative mood. First, the grammaticalization of volitive markers results in the emergence of optative moods, which are present in many languages, including Ancient Greek and Sanskrit, as well as Caucasian and Turkic languages. According to Dobrushina, van der Auwera, and Goussev (2013), morphological optative (an inflected verb form dedicated to the expression of the wish of the speaker) is a fairly infrequent phenomenon, and other grammatical means such as modal verbs or elliptical conditionals are used across the languages to encode this meaning. An example of the morphological optative from Bagvalal (Daghestanian) is provided in Dobrushina (2001, pp. 328–329), cited here in (44).
Epistemic moods appear on the basis of modal markers of possibility and necessity and signal different types of epistemic evaluation: low epistemic probability (dubitative), epistemic probability (probabilitive), and epistemic certainty (assertive). Epistemic moods are found in Turkic, Samoyedic, and Daghestanian languages.
Tundra Nenets (Samoyedic, Uralic) has 15 inflectionally formed non-indicative moods (Nikolaeva, 2014). Among them there are at least three moods expressing epistemic meanings. Probabilitative (glossed as prob) has forms in present, past, and future tenses. The present (45a) is formed with the complex suffix -n(ʹ)aqxiə or -t(ʹ)aqxiə, and the past (45b) with -meəqxiə. The future probabilitative (45c) is formed by the future affixes -tə- or -ŋko-, preceding the present probabilitative. Probabilitative expresses a high degree of uncertainty about the described situation: the speaker makes a guess, often based on inference.
(45) Nikolaeva (2014, p. 100)
Another epistemic mood is called approximative (glossed as aprx), and it expresses a meaning similar to ‘it seems, it looks like’. The denoted degree of certainty is higher than that of the probabilitative, because the described situation is not based on inference, but is rather witnessed by the speaker directly or is familiar from her previous experience. The forms of approximative historically go back to the combination of the non-finite (participial) forms and the affix -rəxa. Present, past, and future forms are demonstrated in (46a–c).
(46) Nikolaeva (2014, pp. 102–103)
The highest degree of certainty is expressed by the mood with a suffix -wan°ŋkəbʹa, termed as “dubitative” by Nikolaeva (2014) and “superprobabilitative” by Salminen (1997) and Burkova (2010). The latter labels seem far more satisfactory, as “dubitative” usually serves to denote low probability. Present and future forms of the superprobabilitative (glossed as sprob) are synthetic (47a,c), while the past is formed periphrastically, with the help of the auxiliary ŋǣ- ‘to be’ (47b).
(47) Nikolaeva (2014, pp. 98–99)
4. Concluding Remarks
This article provided a brief description of mood as a category that expresses modality by formal grammatical means. It is commonly marked by verbal inflection and coexists with other grammatical categories within a single cumulative exponent. However, there are languages where mood categories are encoded by special affixes (prefixal, suffixal, or circumfixal) or by non-segmental means.
The core of mood is covered by the indicative vs. subjunctive opposition, as well as the imperative, which claims a status of a universal mood category, being present in (almost) all languages. The systems of mood in the world’s languages often do not confine themselves to a single non-indicative mood, and epistemic modal meanings or volitive expressions grammaticalize to separate inflectional forms, comprising epistemic and optative moods respectively.
Aikhenvald, A. Y., & Dixon, R. M. W. (Eds.). (2017). Commands: A crosslinguistic typology. Oxford, UK: Oxford University Press.Find this resource:
Elliott, J. R. (2000). Realis and irrealis: Forms and concepts of the grammaticalisation of reality. Linguistic Typology, 4(1), 55–90.Find this resource:
Frawley, W. (Ed.). (2006). The expression of modality. Berlin, Germany: Mouton de Gruyter.Find this resource:
Rothstein, B., & Thieroff, R. (Eds.). (2010). Mood in the languages of Europe. Amsterdam, The Netherlands: Benjamins.Find this resource:
Aikhenvald, A. (2010). Imperatives and commands. Oxford, UK: Oxford University Press.Find this resource:
Becker, M. (2010). Mood in Portuguese. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 179–198). Amsterdam, The Netherlands: Benjamins.Find this resource:
Bhat, D. N. S. (1999). The prominence of tense, aspect and mood. Amsterdam, The Netherlands: Benjamins.Find this resource:
Burkova, S. (2010). Kratkij očerk grammatiki tundrovogo dialekta neneckogo jazyka [A brief sketch of Tundra Nenets grammar]. In N. B. Koškarëva (Ed.), Dialektologičeskij slovar’ neneckogo jazyka (pp. 80–349). Ekaterinburg, Russia: Basko.Find this resource:
Bybee, J. (1985). Morphology: A study of the relation between meaning and form. Amsterdam, The Netherlands: Benjamins.Find this resource:
Bybee, J., Perkins, R., & Pagliuca, W. (1994). The evolution of grammar: Tense, aspect and modality in the languages of the world. Chicago, IL: University of Chicago Press.Find this resource:
Dahl, Ö. (1985). Tense and aspect systems. Oxford, UK: Blackwell.Find this resource:
De Groot, C. (2010). Mood in Hungarian. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 551–571). Amsterdam, The Netherlands: Benjamins.Find this resource:
De Mulder, W. (2010). Mood in French. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 157–179). Amsterdam, The Netherlands: Benjamins.Find this resource:
Dobrushina, N. (2001). Formy imperativnoj serii [Forms of imperative series]. In A. E. Kibrik (Ed.), Bagvalinskij jazyk: grammatika, teksty, slovari (pp. 319–332). Moscow, Russia: Nasledie.Find this resource:
Dobrushina, N., van der Auwera, J., & Goussev, V. (2013). The optative. In M. S. Dryer & M. Haspelmath (Eds.), The world atlas of language structures online. Leipzig, Germany: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Dunn, M. J. (1999). A grammar of Chukchi. Ph.D. dissertation, Australian National University, Canberra.Find this resource:
Gusev, V. (2013). Tipologija imperativa. Moscow, Russia: Jazyki slavjanskoj kul’tury.Find this resource:
Hill Mari text corpus – a corpus of oral narratives, collected during the field trips of the Department of Fundamental and Applied Linguistics (Moscow State Lomonosov University) to Gornomari district, Mari El, Russia, 2016–2018.
Huddleston, R. (2002). Clause type and illocutionary force. In R. Huddleston & G. K. Pullum (Eds.), The Cambridge grammar of the English language (pp. 851–945). Cambridge, UK: Cambridge University Press.Find this resource:
Jespersen, O. (1924). The philosophy of grammar. London, UK: Allen and Unwin.Find this resource:
Laca, B. (2010). Mood in Spanish. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 198–221). Amsterdam, The Netherlands: Benjamins.Find this resource:
Lunn, P. V. (1995). The evaluative function of the Spanish subjunctive. In J. Bybee & S. Fleischman (Eds.), Modality and grammar in discourse (pp. 419–449). Amsterdam, The Netherlands: Benjamins.Find this resource:
Lyons, J. (1977). Semantics. 2 vols. Cambridge, UK: Cambridge University Press.Find this resource:
Majtinskaja, K. E. (1974). Sravnitel’naja morfologija finno-ugorskix jazykov. In V. I. Lytkin & K. E. Majtinskaja (Eds.), Osnovy finno-ugorskogo jazykoznanija (voprosy proisxoždenija i razvitija finno-ugorskix jazykov) (pp. 214–382). Moscow, Russia: Nauka.Find this resource:
Meeussen, A. E. (1962). De tonen van subjunktief en imperatief in het Bantoe (Annales Sciences Humaines 42). Tervuren, Belgium: Musée Royal de l’Afrique Centrale.Find this resource:
Mordashova, D. (2017). Kategorija naklonenija v gornomarijskov jazyke [Category of mood in Hill Mari]. Ms, Lomonosov Moscow State University.Find this resource:
Nikolaeva, I. (2014). A grammar of Tundra Nenets. Berlin, Germany: Mouton de Gruyter.Find this resource:
Nurse, D. (2003). Aspect and tense in Bantu languages. In D. Nurse & G. Philippson (Eds.), The Bantu languages (pp. 90–102). London, UK: Routledge.Find this resource:
Nuyts, J. (2005). Modality: Overview and linguistic issues. In W. Frawley (Ed.), The expression of modality (pp. 1–27). Berlin, Germany: Mouton de Gruyter.Find this resource:
Nuyts, J., & van der Auwera, J. (Eds.). (2016). The Oxford handbook of modality and mood. New York, US: Oxford University Press.Find this resource:
Palmer, F. R. (2001). Mood and modality (2nd edition). Cambridge, UK: Cambridge University Press.Find this resource:
Perry, J. R. (2007). Persian morphology. In A. S. Kaye (Ed.), Morphologies of Asia and Africa (pp. 975–1019). Winona Lake, Indiana, US: Eisenbrauns.Find this resource:
Plungian, V. (2005). Irrealis and modality in Russian. In B. Hansen & P. Karlík (Eds.), Modality in Slavonic languages: New perspectives (pp. 187–198). Munich, Germany: Sagner.Find this resource:
Quer, J. (2010). Mood in Catalan. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 221–237). Amsterdam, The Netherlands: Benjamins.Find this resource:
Rentzsch, J. (2015). Modality in the Turkic languages. Berlin, Germany: Klaus Schwarz Verlag.Find this resource:
Salminen, T. (1997). Tundra Nenets inflection. Helsinki, Finland: Mémoires de la Société Finno-Ougrienne 227.Find this resource:
Sigurðsson, H. Á. (2010). Mood in Icelandic. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 33–56). Amsterdam, The Netherlands: Benjamins.Find this resource:
Squartini, M. (2010). Mood in Italian. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 237–251). Amsterdam, The Netherlands: Benjamins.Find this resource:
Thieroff, R. (2010). Moods, moods, moods … In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 1–33). Amsterdam, The Netherlands: Benjamins.Find this resource:
Tommola, H. (2010). Mood in Finnish. In B. Rothstein & R. Thieroff (Eds.), Mood in the languages of Europe (pp. 511–528). Amsterdam, The Netherlands: Benjamins.Find this resource:
van der Auwera, J., N. Dobrushina, & V. Goussev. (2013). Imperative-hortative systems. In M. S. Dryer & M. Haspelmath (Eds.), The world atlas of language structures online. Leipzig, Germany: Max Planck Institute for Evolutionary Anthropology.Find this resource:
van der Auwera, J., & Plungian, V. (1998). Modality’s semantic map. Linguistic Typology, 2(1), 79–124.Find this resource:
van der Auwera, J., & Schalley, E. (2004). From optative and subjunctive to irrealis. In F. Brisard, M. Meeuwis, & B. Vandenabeele (Eds.), Seduction, community, speech: A Festschrift for Herman Parret (pp. 87–96). Amsterdam, The Netherlands: Benjamins.Find this resource:
Veselinova, L. (2013). Suppletion according to tense and aspect. In M. S. Dryer & M. Haspelmath (Eds.), The world atlas of language structures online. Leipzig, Germany: Max Planck Institute for Evolutionary Anthropology.Find this resource:
Veselinova, L. (2017). Suppletion. In Oxford research encyclopedia of linguistics.Find this resource: