Skip to content

Chapter 63First- and Second-Person Singular Affixes (ku=/en=, e=/e=)

The SAP-singular core: 1sg subjective ku= vs objective en=, and 2sg e= (syncretic across A/S/O), with their morphophonology and distribution.

63.1 Forms and alignment

In the Hokkaido Ainu person paradigm, the first- and second-person singular cells are distinguished by whether the language maintains or collapses the S/A/O contrast within the row. First singular keeps two distinct forms: ku= for the single argument of an intransitive verb (S) and for the transitive subject (A), and en= for the object (O). Second singular has one form, e=, in all three roles. In alignment terms, the 1SG row is nominative–accusative (S=A≠O) and the 2SG row is neutral (S=A=O) Nakagawa (2024: 165–169); Bugaeva (2012: 471–472); Satō (2008: 31–32); Tamura (1972: 27); Bugaeva (2008: 32). The architecture of the complete paradigm — including the side-encoding of the first plural exclusive and the fourth person, the two-index cap on the verb, and the boundary-notation convention — is set out in Chapter 62 (Architecture of the Personal-Affix System: The Four Persons and the S/A/O Paradigms). The present chapter treats the 1SG and 2SG rows in detail.

formal personfree pronoun (Saru)A: transitive subjectS: intransitive subjectO: objectalignment
1SGkaniku=; Saru/Chitose/Mukawa k= before vowel-initial stems (a, u, e, o)ku=; same allomorphyen=S=A≠O; nominative–accusative
2SGeanie=e=e=; 1→2 uses portmanteau eci= insteadS=A=O; neutral

The free pronouns kani and eani are optional emphatic or contrastive expressions; ordinary finite clauses carry the person information on the verb alone. Their etymology — kani from the 1SG prefix plus an existential base; eani invariant across the surveyed Hokkaido dialects — and their distribution relative to the verbal index are treated in Chapter 49 (Personal, Reflexive, and Emphatic Pronouns) Tamura (1971: 3–5); Nakagawa (2024: 250–253).

63.2 First-person singular subject: ku= and k=

The prefix ku= indexes 1SG in both the intransitive-subject and the transitive-subject roles; the two uses share a single form. This identity of S and A, set against the distinct object en=, is the formal basis for describing the 1SG row as nominative–accusative Nakagawa (2024: 166–169); Bugaeva (2012: 472).

(1)
orwano then
ku=cise 1SG.POSS=house
or place
ta LOC
ku=an 1SG.S=exist
kor while

‘Then, while I was in my house, …’

Nakagawa 2024: 166; Ishikari, Nakagawa ex. 104, after Sunazawa Kura 1983: 75

The intransitive existential ku=an indexes 1SG.S. The nominal ku=cise carries the homophonous first-person prefix in the possessive domain (glossed POSS here); nominal possession belongs to the chapter on the possessed form.

In Saru, Chitose, and Mukawa, ku= reduces to k= before stems beginning in a, u, e, or o: ek 'come' yields k=ek, and eramuan 'understand' and eunkeray 'receive' appear as k=eramuan and k=eunkeray. Nakagawa writes the Hidaka-western and Chitose convention as k(u)=, recording the same alternation; the notation reflects phonological context, not two morphologically separate forms Nakagawa (2024: 49, 166); Bugaeva (2012: 467–468).

(2)
aynu Ainu
itak language
k=eramuan 1SG.A=understand
wa CONJ
kusu because
icen money
k=eunkeray 1SG.A=receive
wa CONJ
pirka good
pirka good

‘Because I had learnt Ainu, I was able to receive money, which was very good.’

Nakagawa 2024: 168; Saru, Nakagawa ex. 111

Both transitive verbs show k=, the reduced allomorph of 1SG.A ku= before the vowel-initial stems eramuan and eunkeray.

Bugaeva classifies ku= as a full prefix: it is inseparable from the stem and shifts the accent kernel, properties that set it apart from the more autonomous intransitive-subject suffixes =an and =as, which can be separated from the verb by intervening words and form their own prosodic units Bugaeva (2012: 472–473). Nakagawa writes all personal indexes with = as a practical orthographic convention, noting the symbol carries no claim about clitic status Nakagawa (2024: 49). Satō and Tamura write these forms with a hyphen; the learner grammar of Ijäs does likewise, stating the hyphen is a teaching convention rather than a pronunciation symbol Ijäs (2023). Sakaguchi's explicit clitic/affix criteria — developed primarily for Sakhalin material, and applied to Hokkaido only by extension — place root-bound personal indexes in the prefixal class and rank =an/=as as the most autonomous subject markers; on these diagnostics, ku= would belong to the fully prefixal end Sakaguchi (2024: ix, 77–85, 163) ‹contested›. The ordering diagnostics are developed in Chapter 69 (The Personal-Affix Template: Position Classes and Affix Ordering).

Ijäs introduces ku= early as the obligatory subject marker on both transitive and intransitive verbs, presenting it before the optional independent pronoun and treating the pronoun as emphatic or clarifying rather than grammatically required Ijäs (2023). This is consistent with the head-marking account of the Ainu verb given by Satō Satō (2025: 361) and with the typological framework of (Nichols 1986). An idiosyncratic note in Ijäs's phonetic guidance specifies that initial i retains the fuller form and that careful Saru speech may realise the u vowel before other vowels; these are L2-oriented observations about pronunciation and do not affect the morphological alternation Ijäs (2023).

(3)
ku=hosipi 1SG.S=return

‘I go back.’

constructed exampleIjäs 2023; Saru

Constructed following Ijäs Lesson 2, which uses this verb type to introduce 1SG.S on an intransitive. Ijäs writes ku-hosipi.

63.3 First-person singular object: en=

The object counterpart of 1SG is en=. It indexes a first-person singular participant in the O role: the patient of a monotransitive verb or, with three-place verbs such as kore 'give' and epakasnu 'teach', the human recipient rather than the theme Nakagawa (2024: 169); Satō (2008: 147); Satō (2023: 47); Satō (2025: 368–371). The secundative object-indexing pattern — recipient indexed, theme left as a bare noun phrase — is developed in Chapter 66 (Object Indexing: Monotransitive and Ditransitive Clauses).

(4)
en=epunkine 1SG.O=protect
wa CONJ
en=kore 1SG.O=give
yan IMP

‘Please protect me.’

ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, menokoyukar 'Kamuy yukar 4' [aa-irc/011#5]

The imperative force supplies a second-person subject; both instances of en= mark the 1SG participant as the object/recipient.

(5)
nep what
kameaske monster
ene thus
ne COP
yakkay even
en=kasuy 1SG.O=help
kun should
pe NMLZ
an exist
cik if
pirka good
hike although

‘It would be good if any monster were to help me.’

Nakagawa 2024: 170; Tokachi, Nakagawa ex. 122, after Takahashi 2007: 63

The prefix en= indexes the speaker as the object of kasuy 'help'; the agent is left to context.

A syncretism affects en= in the Saru and Chitose combination tables. Across most Hokkaido dialects, a second-person subject acting on a 1SG object surfaces as a second-person A exponent followed by en=e=en= for 2SG→1SG or eci=en= for 2PL→1SG. In Saru and Chitose, the expected second-person A exponent is absent, and bare en= covers the 2SG→1SG cell; Chitose extends the same bare form to 2PL→1SG. The bare en= form is therefore shared between the 3→1SG and 2→1SG cells in these dialects, and the subject is recovered from discourse or from an overt pronoun Nakagawa (2024: 170, 173); Satō (2008: 146–147) ‹contested›. The full dialect-by-dialect combination table, and the parallel 1→2 portmanteau eci=, are treated in Chapter 65 (The eci= Portmanteau (1A→2O) and Second-Person Plural Syncretism).

Ijäs presents the bare-en= situation to learners as a predictable rule: when the 1SG participant is in O position and the subject is second or third person, the verb shows only en=, with the subject inferred from context Ijäs (2023). The constructed example below illustrates the Saru/Chitose 2SG→1SG reading.

(6)
topenpe sweets
en=kore 1SG.O=give

‘You gave me sweets.’

constructed exampleIjäs 2023; Saru

Constructed following Ijäs. The translation supplies a 2SG subject from context; in Saru/Chitose the verb shows only en=, which covers both 3→1SG and 2→1SG readings. The phenomenon is attested in Nakagawa 2024: 170, 173 and Satō 2008: 146–147.

In the literary and oral-narrative register, the 2→1SG configuration maps to i=, the fourth-person object index, with the subject again unexpressed Satō (2008: 261). That use belongs to the functional range of the fourth person and is treated in Chapter 67 (The Indefinite/Fourth Person: Forms, Reference, and Agent-Defocusing).

63.4 Second-person singular: e=

The 2SG index e= is neutral across argument roles: one form marks 2SG.S, 2SG.A, and 2SG.O alike. This pattern is shared by the 2PL form eci=, so neither the singular nor the plural second-person index distinguishes subject from object Tamura (1972: 27); Satō (2008: 32); Bugaeva (2012: 472). On an intransitive verb, e= marks 2SG.S; on a transitive verb with a third-person counterpart, e= may mark either 2SG.A or 2SG.O, and the role is supplied by the co-occurring indexes and by discourse.

(7)
sinen alone
ne COP
e=oman 2SG.S=go
wa CONJ
e=ray 2SG.S=die

‘If you go alone and die, …’

Nakagawa 2024: 167; Shizunai, Nakagawa ex. 106

Both intransitive verbs carry e= as 2SG.S; no object is present.

(8)
kunneywa morning
wakka water
e=ku 2SG.A=drink

‘You drink water in the morning.’

constructed exampleIjäs 2023; Saru

Constructed following Ijäs Lesson 3. The prefix e= indexes a 2SG transitive subject; the third-person object wakka receives no index.

The role ambiguity of e= on a transitive verb is inherent to the neutral row. In a clause such as kanpisos e=kore, the prefix may signal either 2SG.A (the addressee gives the book to someone) or 2SG.O (someone gives the book to the addressee), depending on which participant is third person; the gloss assigned in interlinear analysis follows what the discourse establishes. Ijäs flags this ambiguity explicitly in her fifth lesson as a diagnostic for learners Ijäs (2023).

When a 1SG or 1PL.EXCL subject acts on a second-person object, the portmanteau eci= replaces any sequence of separate first- and second-person indexes. All four 1→2 combinations — 1SG→2SG, 1SG→2PL, 1PL.EXCL→2SG, 1PL.EXCL→2PL — surface as eci=, so that forms such as *ku=e= and *ci=e= do not occur Tamura (1972: 27–28); Bugaeva (2008: 38); Bugaeva (2012: 474); Nakagawa (2024: 170, 173). The portmanteau's double role — 1→2 combination exponent and 2PL index for S, A, and O — is treated in Chapter 65 (The eci= Portmanteau (1A→2O) and Second-Person Plural Syncretism).

(9)
nep something
ka even
eci=kasuy 1SG.A>2SG.O=help
pe NMLZ
an exist
ya Q

‘Is there anything I can help you with?’

Nakagawa 2024: 173; Saru, Nakagawa ex. 128, after Honda 2001: 44

The single exponent eci= encodes the 1SG.A→2SG.O combination; no separate ku= or e= appears.

Second-person neutrality is the consensus descriptive position for Hokkaido Ainu Tamura (1972: 27); Satō (2008: 32); Bugaeva (2012: 472). Its diachronic status is more open: most treatments assume an earlier S/A≠O distinction was lost, while Satō's reported 2024 position treats neutrality as original — a strategy of empathy or deferred agency never preceded by a subject/object split Dal Corso (2025: 194–195) ‹contested›. The alignment-level debate and the competing system-wide accounts are taken up in Chapter 72 (Tripartite, Nominative-Accusative, and the Central Alignment Debate).

References cited in this chapter

Bugaeva (2008) ·Bugaeva (2012) ·Dal Corso (2025) ·Ijäs (2023) ·ILCAA Ainu materials (1976) ·Nakagawa (2024) ·Nichols (1986) ·Sakaguchi (2024) ·Satō (2008) ·Satō (2023) ·Satō (2025) ·Tamura (1971) ·Tamura (1972)