Skip to content

Chapter 132Sequential wa and Clause Chaining

The default connective wa 'and (then)', multi-clause event chains, and the culminative linker aine, disambiguated from the homophonous resultative/perfect wa.

132.1 Overview

The particle wa ‘and (then), so (connective)’ is the most versatile and most frequent of Hokkaido Ainu's conjunctive particles. Its core function, in Tamura's formulation as transmitted by Nakagawa, is to gather two events E1 and E2 into one coherent action and present them as a single sentence-unit Nakagawa (2024: 342). Where hine merely lists events in temporal order, wa asserts a closer bond: the first event supplies the result-state, the manner, or the cause that licenses the second. The three semantic readings share the surface form and overlap in natural speech; the context, the verb's aktionsart, and the presence of reinforcing adverbs (such as the sequence-marker orowa(no)) determine which reading is primary in any given clause pair Nakagawa (2024: 339–342); Satō (2008: 45). The system as a whole is laid out in Chapter 131 (Clause Linkage in Ainu: Typology and Inventory); the present chapter treats wa and its culminative companion ayne specifically.

132.2 Three semantic readings

Satō identifies three uses of wa: sequential continuation (often reinforced by orowa 'and then'), cause-to-effect (reinforced by kusu 'therefore'), and attendant circumstance or manner Satō (2008: 45). Nakagawa adds a unifying temporal criterion: the first VP must be at its result-state stage — the action has concluded and that result-state frames E2 — rather than ongoing (simultaneity, kor) or stative-negated (absence of change, no) Nakagawa (2024: 341).

particlephase of first VPcore readingminimal example
waresult-state after an actionsequence; manner; cause–effectipe wa soyne
'ate, then went out'
korduring the actionsimultaneityipe kor soyne
'went out while eating'
nostative or negated action'in the state of'; 'without doing'somo ipe no soyne
'went out without eating'

The minimal triad comes from Nakagawa's analysis of the Saru particle system Nakagawa (2024: 341); for kor and no see Chapter 134 (Simultaneous kor 'while' and Overlapping-Event Clauses). Nakagawa also notes that the result-state account alone does not predict every wa collocation: the fixed phrase apkas wa arpa 'walk-and-go' shows wa marking walking as the route or method of going, even though the walking partly overlaps the going. Tamura's integration criterion (via Nakagawa) accommodates this under a "state/method/reason" cluster that subsumes all three readings Nakagawa (2024: 342). The manner reading appears clearly in the following Saru corpus example:

(1)
corpokke beneath.it
a=eapkas 4.A=walk.along
wa CONJ
arpa=an go=4.S
ayne CONJ

‘Walking beneath it, I went along, and then…’

Nakagawa 2024: 339; Saru, Nakagawa ex. 605, after Kayano 1974a: 68

wa marks the walking as manner and route of the going; ayne is the culminative particle treated in §5 below.

The cause–effect reading is clearest when E1 directly occasions E2:

(2)
hokure INTERJ
pirkano properly
e=ipere 2SG.A=feed
wa CONJ
mimus get.fat
yakne COND

‘Now, feed her well so that, once she has fattened up…’

Nakagawa 2024: 339; Saru, Nakagawa ex. 604

wa links the feeding (E1, cause) to the fattening (E2, effect); the subsequent conditional yakne opens a premise on the result-state.

132.3 Subject continuity: statistical preference without a constraint

In Refsing's Shizunai material, approximately ninety percent of the clause pairs joined by wa share a subject Refsing (1986: 240). The preference reflects the particle's integration semantics: if E1 is the manner or prior stage of E2, the same actor typically performs both. Yet the preference is not a syntactic rule. Satō shows that the two clauses joined by wa may have different subjects — unlike Japanese 〜て, which requires coreference — and that fact bears on its analysis as a conjunctive particle rather than a same-subject converb Satō (2008: 45–46); Bugaeva (2022).

(3)
hekaci child
suma stone
osura throw
wa CONJ
ontaro barrel
hokus fall.over

‘The child threw a stone, and the barrel fell over.’

Satō 2008: 45; Saru

The thrower (child) is the A of the first clause; the barrel is the S of the second. Satō uses this example to show that wa carries no same-subject requirement.

Reference tracking across a wa chain rests on the obligatory person indices that each finite verb carries independently. The statistical same-subject preference, combined with the absence of a formal constraint, places wa in the configuration that the cosubordination typology of (Foley & Van Valin 1984) describes — finite clause-chaining without converb morphology and without a grammaticalized same/different-subject opposition. The comparison with hine and akusu, and the broader question of whether any Hokkaido particle encodes subject continuity in a switch-reference-like way, belong to Chapter 133 (The hine/akusu Contrast and Switch-Reference-Like Linkage).

132.4 Multi-clause chains in narrative

Hokkaido Ainu prose narrative strings several wa-linked clauses into a single chain. Each medial predicate closes on its own person indexing followed by wa, before the chain continues to the next step. The result presents a process as an integrated sequence of stages:

(4)
sikerpeni Phellodendron
kapu bark
a=sospa 4.A=peel
wa CONJ
siwnin yellow.bark
pe NMLZ
a=uyna 4.A=take
wa CONJ
kusuri medicine
ne COP
a=kar 4.A=make

‘She peeled the bark of the Phellodendron tree, took the yellow inner bark, and made it into medicine.’

Ainu Koraci 1991; Saru, 横山裕之, 'sikerpeni', Ainu Times no. 18/3, text 12; corpus ainu-times/018/3#12

A three-step wa-chain; each step is a completed result-state enabling the next. pe after siwnin nominalizes the NP; the fourth-person a= carries the narrating subject.

The free connective adverb orowa(no) 'and then, from there after' can follow a wa-marked clause to make the sequential boundary more prominent Nakagawa (2024: 516); Ijäs (2023). This stacking — connective particle before, connective adverb after — is one of the clearest diagnostics separating the two word classes. The treatment of orowa(no) as a discourse boundary marker belongs to Chapter 138 (Tail-Head Linkage and Narrative Clause Cohesion).

132.5 The culminative ayne

The particle ayne ‘and finally, at last (after a protracted action)’ closes a stretch of sustained effort and points to the eventual outcome. Where wa integrates events step by step, ayne marks the endpoint of a duration — a process that went on and finally yielded a result Nakagawa (2024: 346); Satō (2008: 44). Refsing notes a nuance of impatient delay ('and then at last') absent from the neutral chainer hine Refsing (1986: 247).

The most characteristic environment is the intensifier pattern V a V a ayne, in which the perfective auxiliary a reduplicates the verb to signal repeated iteration before the particle closes the stretch:

(5)
ku=cis 1SG.S=cry
a PFV
ku=cis 1SG.S=cry
a PFV
ayne CONJ

‘Crying and crying, and finally…’

Ainu Koraci 1991; Saru, 丸野和子, '"Kuma" turano k=an', Ainu Times no. 15/1, text 24 (excerpt)

The V a V a ayne intensifier pattern: the perfective a marks each iteration; ayne closes the protracted stretch and points to the outcome. The full sentence continues with the result clause.

The particle easir 'only then, at that point' frequently follows ayne to mark the delay before the outcome Satō (2008: 44); Ijäs (2023). The pattern wa…wa…ayne is common in narrative: a run of wa-chained steps leads up to a final ayne that marks the point of resolution, as in the manner example above (corpokke a=eapkas wa arpa=an ayne).

The etymology is transparent: durative auxiliary a plus the neutral chainer hine, grammaticalized into a separate particle (Chiri 1936; Tamura, via Refsing (1986: 247–248)). Tokachi uses ahinne (occasionally ainne), and some Tokachi varieties such as Honbetsu lack the form altogether, distributing its function across hine and wa Nakagawa (2024: 346–347). With a negated preceding clause, ayne shifts to a 'before having V-ed' reading (negated-protraction), attested in Chitose Nakagawa (2024: 347).

132.6 wa as a gateway to the auxiliary system

Wa is the conjunctive particle most recruited as the pivot of Ainu's periphrastic aspect constructions Refsing (1986: 240). The result-state reading extends naturally into a family of analytic forms in which the result-state of a closed event is overlaid by an existential or auxiliary verb, yielding a resultative, a terminative, a prospective, an attemptive, or a benefactive nuance. Nakagawa's §15 (pp. 452–455) lists the principal forms; the full aspectual analysis belongs to Chapter 110 (The Existential-Aspect System: kor an Progressive and wa an Resultative-Perfect).

constructionglossed structurecore value
V wa an / V wa okaV CONJ exist.SG / exist.PLresultative state: 'be in the state resulting from V-ing'
V wa isam / V wa okereV CONJ be.no.more / finishterminative/completive: 'have V-ed (to completion)'
V wa anuV CONJ put.asideprospective: 'do V in preparation, keep V-ed for future use'
V wa inkar / V wa inuV CONJ try.seeing / try.hearingattemptive: 'try V-ing'
V wa koreV CONJ givebenefactive: 'do V for someone'

The resultative wa an — where the preceding verb closes on its result-state and the existential an predicates that state as ongoing — produces a modifier-head construction interpretable as a resultative relative clause:

(6)
racitke hang
wa CONJ
an exist
tuki cup

‘a hanging cup’

Nakagawa 2024: 339; Saru, Nakagawa ex. 607

racitke wa an is the attributive resultative: 'the cup that hangs / is in the state of being hung'. The construction modifies tuki as a head noun.

The terminative wa isam foregrounds completion rather than persistence:

(7)
arpa go
wa CONJ
isam be.no.more

‘S/he has gone.’

constructed example

In all these constructions wa is the conjunctive particle, the same item that links finite clauses in the sequential reading; the boundary with the connective use is syntactic (an auxiliary follows immediately and predicates only aspect) rather than formal. The perfective auxiliary a that often precedes wa in scene-switch positions adds a further layer, completing the situation before the chain continues; that interaction is treated in Chapter 111 (The Perfective/Anterior Particle a (a, a…a, aan)).

132.7 The dialectal allomorph ma

When wa immediately follows a stem ending in -n or -m, the initial w- assimilates to the preceding nasal and the particle surfaces as [ma]. Refsing documents the alternation for Shizunai Refsing (1986: 240), and Nakagawa writes the allomorph phonetically in his transcriptions: the resultative an wa an appears as an ma an Nakagawa (2024: 339). The pattern is phonological and pan-dialectal at the segment level ‹corpus-confirmed›: the standard romanized orthography retains wa throughout to represent the underlying form, while katakana transcription writes マ in the relevant environment, reflecting the surface pronunciation (nukopoli, aynu-corpora Discord 2024; aynu-corpora Discord (2023–2026)).

The same assimilation applies before any stem ending in a nasal, including an 'be/exist', oman 'go (honorific)', and verbs whose final syllable carries -n. The allomorph is predictably conditioned; no separate particle ma exists alongside wa.

132.8 Disambiguation

The form wa covers at least three distinct grammatical entries, kept apart by position and syntactic host rather than by form Nakagawa (2024: 340); Ijäs (2023).

entrygrammatical classkey diagnosticexample
wa sequentialconjunctive particlefollows a finite VP; always opens a further clause; swappable for hineipe wa soyne
'ate and went out'
wa(no) ablativecase particle / postpositionfollows a NP or locative nominal; glossed ABL; extends to wanomosir wa
'from the land'
wa sentence-finalfinal particleclause-final; emphatic or exclamatory; no following clausepirka wa
'it's fine! (emph.)'

The connective wa always follows a finite verbal predicate and always opens a following clause; it never occurs clause-initially. The ablative wa(no) attaches to a nominal and typically co-occurs with a spatial or temporal noun (cf. orowa = or + wa 'from that place'). The sentence-final emphatic is isolable at the end of a complete utterance. In natural text the three are rarely ambiguous; glossing discipline requires keeping them distinct, since the particle counts for the sequential wa are inflated by ablative and final-particle tokens in any raw word-form search ‹corpus-suggested›.

References cited in this chapter

Ainu Koraci (1991) ·aynu-corpora Discord (2023–2026) ·Bugaeva (2022) ·Foley & Van Valin (1984) ·Ijäs (2023) ·Nakagawa (2024) ·Refsing (1986) ·Satō (2008)