Skip to content

Knowledge base · Part VI — Word Classes and Nominal Morphology · this topic as a network

36. Verbal Subclasses: A Valence-Based Word-Class Taxonomy

What are the major verb subclasses, and how are they defined by argument count and roles?

56 sentences from 23 sources · every sentence names the pages it rests on

Verbs that by themselves correspond to a full sentence containing both subject and verb elements are termed complete verbs (kanzen dōshi). (Nakagawa 2024: 129) One-argument inkar ('to look, cast one's gaze') is used when the object seen is not predetermined, whereas two-argument nukar ('to see, find') is used when the seen entity is specified. (Nakagawa 2024: 131) One-argument ikka is used when the item being stolen is not specified, whereas two-argument eikka is used when the stolen object is overtly specified. (Nakagawa 2024: 132) Ainu verbs are classified into three types based on their aspectual properties: activity verbs (action verbs), change-of-state verbs (change verbs), and stative verbs. (Nakagawa 2024: 438) The verb rikin denotes upward movement, but it is also used with the meaning of piling up or increasing in height. (Satō 2008: 337)

A valence-based taxonomy

What formal evidence supports dividing verbs into valence-based subclasses in the language?

Ainu verbs are classified into four major classes: complete verbs, intransitive verbs, transitive verbs, and the copula. (Nakagawa 2024: 117, Refsing 1986, Satō 2008: 21) Transitive and intransitive verbs in Ainu are distinguished both syntactically and morphologically. (Bugaeva 2006: 185, Bugaeva 2008) In Ainu, there is no clear grammatical distinction between states and processes, and most verbs combine both meanings. (Bugaeva 2014) Verbs are classified into plain verbs (classified by number of objects), existential verbs (expressing 'exist'), and designator verbs (copula). (Murasaki 2025: 90) Verbs in Ainu can be classified from 0-argument to 4-argument verbs according to the number of noun phrases they require. (Nakagawa 2024: 117) Basic verbs have positive arities of +1 for one-argument verbs, +2 for two-argument verbs, and +3 for three-argument verbs. (Nakagawa 2024: 137)

The person-marking diagnostic

How does personal affixation reveal which argument roles a verb takes?

Intransitive and transitive verbs employ different person indexing for S and A in the first-person plural and the indefinite person. (Bugaeva 2012: 474–475) Ainu verbs are divided into three types according to their conjugation patterns. (Kindaichi & Chiri 1936) Class I verbs are those that prefix the Classical first-person subjective affix (a-- type), and verbs requiring an object or complement (transitive verbs and incomplete intransitive verbs) belong to this class. (Kindaichi & Chiri 1936) Class II verbs are those that suffix the Classical first-person subjective affix (--an type), and verbs not requiring an object or complement (complete intransitive verbs) belong to this class. (Kindaichi & Chiri 1936) Class III verbs are those in which the Classical first-person subjective affix infixes (--a-- type), and certain compound verbs belong to this class. (Kindaichi & Chiri 1936) When combined with a nominative pronominal affix that has an alternation between suffixed and prefixed forms (1st person plural and indefinite person), an intransitive verb always takes the suffixed form. (Refsing 1986: 140)

Intransitive verbs

How does the language handle verbs with a single argument, and what semantic types are included?

Intransitive verbs cannot combine with objective pronominal affixes or take any object. (Refsing 1986: 77, 140) Class II verbs whose stem does not change between singular and plural are classified as division i (第 i 種). (Kindaichi & Chiri 1936) Verbs such as ak 'shoot an arrow', arka 'hurt', isu 'live', ipe 'eat', kira 'flee', rai 'die', and hotke 'sleep' belong to Class II division i. (Kindaichi & Chiri 1936) Class II division ii verbs are those that have different stems for singular and plural, where the plural stem is formed by removing the final vowel of the singular stem and adding -pa. (Kindaichi & Chiri 1936) In Class II division ii verbs, the plural stem (e.g. hoshippa) is used in all plural forms as well as in the second person and third person honorific singular forms (hoshippa-an, hoshippa). (Kindaichi & Chiri 1936) All verbs belonging to Class II division ii are compound verbs whose root is a transitive verb of the type listed in § 121 (1). (Kindaichi & Chiri 1936) The verb siknu 'be alive' can also mean 'come back to life' (become alive). (Nakagawa 2024: 120) In Ainu, the single argument of a one-argument verb is a grammatical subject and is not necessarily a semantic agent, sometimes representing an undergoer. (Nakagawa 2024: 129) Stative verbs in Ainu can express static states by themselves, but they can also convey inchoative change of state ('become [that state]') and can take personal affixes. (Yoshikawa 2022: 19.2)

kim ta arpa=an akusu
mountain-ward LOC go=4.S when
‘When (I) went to the mountains…’
toan nupuri ri
DEM mountain be.tall
‘That mountain is high.’

Transitive verbs

How are transitive verbs with two arguments marked, especially when both are non-third-person?

Transitive verbs can combine with pronominal affixes in both the nominative and the objective cases, and can take noun expressions as objects. (Refsing 1986: 77) Transitive verbs include two-place verbs that take a single object as well as three-place verbs that take both an indirect object and a direct object. (Satō 2008: 21)

kani ku=respa p anakne
1SG 1SG.A=rear.raise NMLZ TOP
‘As for those (children) I raised…’

Three-place verbs

How are verbs with three arguments, such as give or tell, formed and inflected?

As far as currently known, the locative object verb o ('to put X into Y') is the only underived basic stem that functions as a three-argument verb. (Nakagawa 2024: 134) In principle, four-argument verbs can be formed by adding a causative suffix to a three-argument verb, though actual attested examples are extremely rare. (Nakagawa 2024: 135) In their basic forms, verbs range from one-argument to three-argument verbs; o 'to put into a place' is the only basic three-argument verb, and almost all other basic verbs are either one-argument or two-argument. (Nakagawa 2024: 136) The causative verb ta-re 'X causes Y to draw Z (water)' is a three-argument verb. (Satō 2023)

topenpe hekattar ku=kore
sweets children 1SG.A=give
‘I gave the children sweets.’

The copula and the existential verbs

What special properties do copula, existential, and negative existential verbs have?

wen menoko ku=ne
poor woman 1SG.A=COP
‘I am a poor woman.’
· constructed

Avalent predicates

Are there predicates that take no arguments, and how are they formally identified?

Zero-argument (complete) verbs consist of meteorological predicates that commonly contain an incorporated subject sir 'appearance/land' and cannot take any external or dummy subject. (Bugaeva 2012: 470–471) Verbs taking zero noun phrase arguments, such as 'onuuman 'the day grows dark', form a distinct valence class of complete verbs (v0). (Murasaki 2025) Many zero-argument verbs are formed by prefixing sir- ('around, appearance') to a one-argument verb. (Nakagawa 2024: 129) In the Saru dialect, temperature verbs form pairs of zero-argument and one-argument verbs, where zero-argument verbs express weather or climate temperature, while one-argument verbs express the subjective sensation felt by an experiencer. (Nakagawa 2024: 130) The zero-argument verb hawas ('there is word/sound that...') takes no subject and is used when the speaker is unspecified, whereas the one-argument verb hawean ('to say') requires an overt or specified subject. (Nakagawa 2024: 131) Refsing (1986) classifies forms such as sirsesek, aptoas, asuras, and mean as 0-place 'Closed Verbs', whereas Tamura's Saru dialect dictionary treats apto as and asur as as two words (subject + 1-place verb). (reported in Nakagawa 2024: 493) In Tamura's Saru dialect dictionary, of the 68 verbs classified as 0-place verbs, 56 begin with sir-, two with me-, two with haw-, two with hum-, and six with other elements. (reported in Nakagawa 2024: 493) Because 0-place verbs by definition take no personal affixes, there is no morphological criterion based on affix placement to determine their word boundary. (Nakagawa 2024: 493) The element me in mean 'be cold' is an unautonomous nominal root and cannot form an independent noun phrase separated from an by adverbial particles (*me anak an, *me ka an). (Nakagawa 2024: 494) The element sir cannot be separated from the predicate by the adverbial particles anak or ka (*sir anak pirka, *sir ka pirka). (Nakagawa 2024: 494) Because apto can occur as an independent noun separated by particles (e.g. apto ka isam), there is no ground to treat apto as as a single 0-place verb rather than a two-word clause. (Nakagawa 2024: 494) Closed verbs are a small group of verbs that for semantic reasons never combine with pronominal affixes, never take any other subject or object, and do not accept verb prefixes or suffixes. (Refsing 1986: 77) Verbs that never combine with a pronominal affix form a distinct subgroup called closed or non-affixing verbs. (Refsing 1986: 140) Closed verbs do not combine with verb prefixes or suffixes either, and appear only with some of the auxiliaries. (Refsing 1986: 140) Closed verbs are characteristically terms for natural phenomena or for acts with no specific agent, but a few stative verbs also belong to this group. (Refsing 1986: 140) Closed verbs have an incorporated subject, so that they never appear with any independent subject either. (Refsing 1986: 140) Closed or non-affixing verbs never combine with pronominal affixes, verbal prefixes, or verbal suffixes, and they only appear with some auxiliaries. (Refsing 1986: 140) Closed verbs have an incorporated subject, never appear with any independent subject, and do not distinguish singular and plural forms. (Refsing 1986: 140) Zero-argument verbs (complete verbs) take neither a subject nor an object and do not undergo person inflection. (Satō 2008: 3) There exists an atransitive (zero-place) verb type in Ainu. (Satō 2025)

Locative-object verbs and ambitransitives

How are locative-object verbs marked, and what defines an ambitransitive verb?

Because kohemesu ('to climb [onto/up]') is a two-argument verb whose climbed location is a direct object, a case particle such as ta cannot mark that location. (Nakagawa 2024: 132) The form o exhibits three homophonous verbs: a two-argument locative object verb serving as the plural of oma ('be in / enter [a place]'), a two-argument non-locative object verb meaning 'to ride', and a three-argument locative object verb meaning 'to put X into [a place]'. (Nakagawa 2024: 134) Locative object verbs are defined as a small set of verbs that take a grammatically locative argument as their direct object. (Nakagawa 2024: 212) In traditional Ainu linguistics, combinations where an object and verb appear separately but in a fixed, idiomatically combined phrase with no intervening elements are called 'third group verbs' (e.g., i-par a-oyki 'people support me'). (Satō 2022: 16.4)