Skip to content

Chapter 113Phasal, Iterative, and Habitual Aspect

Inceptive and terminative phasal aspect via auxiliary verbs, plus iterative and habitual marking.

113.1 Phasal and iterative strategies in the auxiliary system

Three aspectual domains occupy the periphery of the core existential-aspect system. Completive aspect — the marking that an event has been carried through to its endpoint — is expressed by the pair verbs wa isam and wa okere, which draw on the lexical content of isam ‘not exist, be absent’ and okere ‘finish’ respectively. Iterative aspect, marking that an event or state recurs, is expressed by the reduplicated perfective auxiliary a … a and by the dedicated auxiliary ranke ‘do repeatedly, again and again’. Habitual and generic readings arise from kor an in appropriate temporal or frequentative contexts. The recent-perfect auxiliary nisa ‘just completed’ marks an event as already finished and still current at the moment of speech. Nakagawa (2024: 450); Satō (2006: 47)

All these strategies recruit the auxiliary and pair-verb frames whose architecture is laid out in Chapter 108 (Tense, Aspect, Mood, and Modality: Overview). The progressive kor an and the resultative wa an are treated as the core existential-aspect system in Chapter 110 (The Existential-Aspect System: kor an Progressive and wa an Resultative-Perfect); the perfective auxiliary a and its admirative counterpart aan are treated in Chapter 111 (The Perfective/Anterior Particle a (a, a…a, aan)). Nakagawa (2024: §15.1)

aspect domainformvalueframe type
inceptivekor an (extension)change underway, endpoint in viewpair verb
completive (disappearance)wa isamaction completes with subject or outcome gonepair verb
completive (throughput)wa okereaction carried through to its endpair verb
iterative (reduplication)a … acompleted event occurs repeatedlyauxiliary reduplication
iterative (auxiliary)rankeaction done again and againauxiliary
habitualkor an + frequentative framecustomary or repeated actionpair verb
recent perfectnisaevent just finished, result currentauxiliary

113.2 Completive aspect: wa isam and wa okere

Both wa isam and wa okere belong to the pair-verb frame introduced in Chapter 108 (Tense, Aspect, Mood, and Modality: Overview): the second member remains a fully finite verb, and both predicates in V wa V₂ carry their own subject affixes. The bracketing is therefore [V]VP wa [V₂]VP, formally two clauses joined by the sequential conjunctive wa, in contrast to the bare auxiliaries that attach directly without a conjunction Nakagawa (2024: 450–451); Bugaeva (2012: 495). The table below situates the two completives alongside wa anu, which carries a related but distinct value.

formsecond verbvalueJapanese glosssource
wa isamisam 'not exist, be absent'completion with disappearance or adverse outcome〜てしまうNakagawa (2024: 452); Bugaeva (2012: 495)
wa okereokere 'finish'action carried through to completion〜てしまう / 〜おえるNakagawa (2024: 452–453)
wa anuanu 'put, place'leave the result in place (concrete objects)〜ておくNakagawa (2024: 453)

wa isam marks completion of the event with the added sense that the subject or its outcome disappears or turns out badly; the second verb isam 'not exist, be absent' retains its concrete lexical force while grammaticalising into the completion marker. Nakagawa's Japanese gloss is 〜てしまう, and Bugaeva gives it the label "completive" Nakagawa (2024: 452); Bugaeva (2012: 495).

(1)
tane now
rikin ascend
wa CONJ
isam be.gone
pe NMLZ
okake after
ta LOC

‘Now, after (the god) had ascended and vanished,’

ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 7 [aa-irc/005#251]

rikin wa isam: the subject rises and then is absent — isam retains its lexical sense 'be absent'. pe okake ta nominalizes the wa isam clause and frames it temporally.

(2)
iyos then
onne grow.old
pa PL
wa CONJ
isam be.gone

‘Then they grew old and passed away.’

ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 8 (part 2) [aa-irc/007#438]

The postverbal plural clitic pa marks a plural subject; wa isam closes the sequence with the lexical force of disappearance.

(3)
caricari be.scattered
wa CONJ
isam be.gone
ru-we track-POSS
ne COP
kuni COMP
a=ramu 4.A=think
kor while

‘(while she was) thinking that (things) had been scattered and vanished,’

ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 9 [aa-irc/009#127]

wa isam is embedded inside a nominalized complement (ru-we ne) governed by ramu 'think'; kuni is the complementizer.

wa okere uses okere 'finish' as its second predicate and foregrounds thoroughness of completion rather than disappearance of the outcome. With stative predicates in Saru and Horobetsu, the construction shifts into an intensive-stative reading 'be extremely Adj', a transparent extension of 'finish being in a state' Nakagawa (2024: 452–453).

(4)
e=kar 2SG.A=make
wa CONJ
okere finish

‘You finished making it.’

constructed example

wa okere foregrounds completion of the action; there is no sense of loss or disappearance as in wa isam.

The pair-verb form wa okere contrasts with the bare auxiliary okere used without the conjunctive wa: the conjunction-bearing version carries a nuance of regret or dismay at the completion, while the bare auxiliary is unmarked for affect Takahashi (2022). Nakagawa's Table 74 lists bare okere as approximating wa okere in meaning, confirming that the conjunctive version is the emotionally marked option Nakagawa (2024: 453).

Bugaeva notes a structural parallel between the pair-verb set and Japanese aspectual periphrases: wa isam〜てしまう, wa okere〜おえる, wa anu〜ておく. Nakagawa resists a simple calque account, pointing to wa inu 'try by hearing/tasting' as a type that lacks a Japanese structural model (*〜て聞く), suggesting the pair-verb frame is native and certain Japanese-facing semantic expansions developed under contact Bugaeva (2012: 495); Nakagawa (2024: 455) ‹contested›.

113.3 Inceptive extensions of kor an

The progressive kor an, whose core function is described in Chapter 110 (The Existential-Aspect System: kor an Progressive and wa an Resultative-Perfect), extends into an inceptive domain when the predicate denotes a change of state and the temporal perspective falls near the onset of the change. Satō's examples make the extension concrete: sirpeker kor an 'it is getting light' describes the sky as in the process of lightening before full daylight is established; the change is under way and its endpoint is in view Satō (2008: 197). Bugaeva and Nakagawa both list the inceptive as one of the readings available to kor an alongside the progressive core Bugaeva (2012: 494); Nakagawa (2024: 441).

(5)
sirpeker dawn.break
kor while
an exist.SG

‘Dawn is breaking; it is getting light.’

Satō 2008: 197; Hokkaido (dialect not further specified)

kor an with a change verb in its inceptive extension: the transition to daylight is ongoing and approaching its endpoint.

(6)
epuykehe bud-POSS
tane now
sipuske open
kor while
oka exist.PL

‘The buds are now about to open.’

Satō 2008: 85; Hokkaido (dialect not further specified)

kor oka (plural existential) with a plural subject; the inceptive reading: the opening is under way with completion imminent.

No dedicated inceptive auxiliary exists in Saru or Chitose. The inceptive reading of kor an is driven by verb type and discourse context, not by a distinct grammatical form. Prospective readings tied to intention or imminence — etokus ‘be about to’, etokoyki ‘be getting ready to’, oasi ‘surely will’ — are treated in Chapter 119 (Epistemic Modality, Dubitative, Intentive, and Irrealis).

113.4 Iterative aspect

113.4.1 The reduplicated a … a construction

The perfective auxiliary a ‘perfective/anterior (SG)’ (plural rok ‘perfective/anterior (PL)’) reduplicates to mark iterative or emphatic-continuative aspect. The construction V a V a signals that the event occurred repeatedly; when the subject is fourth person, the person marker =an migrates to the second auxiliary rather than remaining solely on the lexical verb Nakagawa (2024: 307–308).

(7)
onkami=an worship=4.S
a PFV
=an =4.S
a PFV

‘(I) worshipped again and again.’

Nakagawa 2024: 307; Saru, ex. 453

Reduplicated a … a with a fourth-person subject: =an attaches to the second a. Only the person marker reduplicates with a fourth-person subject, not the full stem.

Bugaeva labels this pattern "emphatic continuative" Bugaeva (2012: 493). Ijäs, in her Lesson 9, treats prolonged continuity without connecting the pattern to the auxiliary a; the correspondence is established by Nakagawa Ijäs (2023); Nakagawa (2024: 307) ‹corpus-confirmed›. The full treatment of a — its morphological source, its contrast with aan, and its range of values — belongs to Chapter 111 (The Perfective/Anterior Particle a (a, a…a, aan)).

Lexical reduplication, a distinct phenomenon, operates inside roots and derivational suffixes: -rototke encodes a violent repeating sound or act, and -atki encodes repetition or persistence, both lexically restricted to certain phonological shapes Nakagawa (2024: 204–208). That root-bound reduplication is not a productive auxiliary strategy; it is treated alongside the other Aktionsart suffixes in the lexical-aspect section of Chapter 108 (Tense, Aspect, Mood, and Modality: Overview).

113.4.2 The iterative auxiliary ranke

The auxiliary ranke ‘do repeatedly, again and again’ attaches directly to the lexical verb and marks iterative or habitual repetition: 何度も〜する Nakagawa (2024: 312). Bugaeva lists it as item 5 in her Saru auxiliary inventory with the label "iterative/habitual" Bugaeva (2012: 493). Like the other auxiliaries in Nakagawa's Table 74, ranke takes no person marking; the lexical verb carries the subject index. The following example shows ranke inserted between the lexical verb and a progressive frame, stacking iterativity with a continuing situation:

(8)
arpa go
ranke ITR
kor while
oka=an exist.PL=4.S

‘(They) keep going again and again.’

Nakagawa 2024: 312; Saru, exx. 471–472

ranke follows arpa 'go' and precedes kor oka=an; ranke takes no person marking. The combination marks repeated habitual going.

The auxiliary tek (Saru/Chitose) covers adjacent ground — brief or momentary action ('do briefly/suddenly') — and is listed alongside ranke in Nakagawa's Table 74 Nakagawa (2024: 309–313). The iterative-durative overlap between ranke and the continuative forms kane an and kan an is addressed in Chapter 112 (The 'Appearance' Continuative: siran, siri…, and kane an).

113.5 Habitual and generic aspect

Hokkaido Ainu has no dedicated habitual morpheme. Habitual and generic readings arise from a combination of kor an, frequentative adverbs or temporal frames, and the event-plural marker pa. The progressive and habitual are formally identical; it is the temporal expression in the clause that licenses the habitual reading Nakagawa (2024: 441); Satō (2008: 197); Bugaeva (2012: 494).

(9)
kor while
pon little
seta dog
turano together
sinot=an play=4.S
kor while
an=an exist=4.S
ayne as

‘Every noon she would play with the puppy,’

ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 9 [aa-irc/009#14]

sittokap kor 'whenever noon (comes)' provides the frequentative temporal frame; sinot=an kor an=an is the habitual kor an construction. ayne marks the clause as background to what follows.

The postverbal plural clitic pa (the pa₂ of Chapter 58 (Verbal Number: Suppletive Stems and the Pluractional -pa)) contributes event-plural or pluractional force in habitual and iterative contexts. Following a predicate whose participant plurality is already marked by a suppletive stem, pa shifts the reading toward repetition across occasions rather than a single plural event. The interaction of pa with suppletive stems and the personal-agreement system is treated in Chapter 58 (Verbal Number: Suppletive Stems and the Pluractional -pa) and Chapter 70 (Person–Number Interactions and Plural Agreement).

The auxiliary ranke (§ above) marks the habitual more explicitly when the repeating character of the situation is foregrounded. Together these resources — kor an with frequentative context, ranke, and event-plural pa — occupy the space that a dedicated habitual morpheme fills in languages where habituality is morphologically distinct from aspect (Bybee et al. 1994); (Comrie 1976).

113.6 The recent perfect nisa

The auxiliary nisa ‘just completed; recent perfect’ marks an event as already finished and still current at the moment of speech: すでに起こってしまったことを表す Nakagawa (2024: 312). Bugaeva calls it "recent perfect aspect nisa (just finished)" and lists it as item 7 of the Saru auxiliary inventory Bugaeva (2012: 493). Nakagawa also groups nisa with the completion and past strategies — alongside a and wa isam — as optional devices for anchoring a situation before the moment of utterance Nakagawa (2024: 436).

Like the other auxiliaries in Table 74, nisa takes no person marking; the lexical verb before it carries the subject index. It may be followed by the evidential frame ruwe ne, where the recent-perfect value and the fact-asserting evidential combine naturally.

(10)
arwenkamuy truly.evil.kamuy
ne COP
an=an exist=4.S
nisa PRF.REC
ru-we track-POSS
ne COP
na FIN

‘(I) have now truly become an evil kamuy.’

Nakagawa 2024: 311; Saru, ex. 470

nisa follows the existential an=an and precedes the evidential ruwe ne; the recent-perfect value and the fact-evidential reinforce each other.

Within the completion field, three forms target different facets of the endpoint. nisa foregrounds recency: the event is recent and its result is present now. The auxiliary a foregrounds the event as a background to what follows and carries direct-evidential force (treated in Chapter 111 (The Perfective/Anterior Particle a (a, a…a, aan))). wa isam foregrounds disappearance and belongs to the pair-verb frame; it adds the affective colouring of loss or adverse outcome. Unlike wa isam, nisa is a bare auxiliary — no conjunctive particle, no second person-indexed predicate — so the construction belongs to the auxiliary frame, not to the pair-verb frame.

The surface token nisa returns 3,355 hits in the present corpus of 196,184 sentences (rank 71 by frequency), but the token is homophone-prone — bare nouns and other elements sharing this phonological shape are not systematically distinguished at the surface-string level — so the count is indicative of high frequency rather than an accurate auxiliary tally.

References cited in this chapter

Bugaeva (2012) ·Bybee et al. (1994) ·Comrie (1976) ·Ijäs (2023) ·ILCAA Ainu materials (1976) ·Nakagawa (2024) ·Satō (2006) ·Satō (2008) ·Takahashi (2022)