Chapter 2A Typological Profile of Hokkaido Ainu

2.1 The genetic isolate and its typological placement

Hokkaido Ainu has no demonstrated genetic relationship with any other language family Bugaeva (2022: Ch.1 §1); Shibatani (1990: §1). Proposals connecting Ainu to Japanese, Korean, Altaic branches (Tungusic, Mongolic, Turkic), Austronesian, Nivkh, or Chukotko-Kamchatkan have each failed to produce the systematic sound correspondences and shared morphology that genetic relationship requires. The fuller treatment of the isolate status and the history of macro-comparison controversies belongs to Chapter 5 (The Genetic Position of Ainu: Isolate Status and Macro-Comparison Controversies).

The isolate argument has a structural dimension that goes beyond the absence of lexical correspondences. Bugaeva's analysis frames Ainu as "a head-marking language of the Pacific Rim" — after the framework of Nichols (1986) — and treats the structural mismatch between Ainu's head-marking profile and the dependent-marking profile of all its geographical neighbours as a positive argument against relatedness Bugaeva (2022: Ch.1 §2, §7). Head-marking is "well attested in the Americas and Melanesia but vanishingly rare elsewhere" Nichols (1986) and characterizes the Greater Pacific Rim (GPR) macroarea — a population of languages whose distribution Nichols traces back more than ten millennia Bugaeva (2022: Ch.1 §2). Northeast Asia is predominantly dependent-marking: Japanese, Korean, Tungusic, Mongolic, and Turkic all mark syntactic relations on the dependent NP (nominal case, adnominal genitive, uninflected postpositions governing case). The head/dependent-marking parameter is among the most diachronically stable typological features known Nichols (1986); its consistent divergence between Ainu and every language proposed as its relative accordingly constitutes evidence independent of the lexical record Bugaeva (2022: Ch.1 §7).

In a single typological characterization: Hokkaido Ainu is SOV, agglutinating, polysynthetic, incorporating, and head-marking at both clause and phrase level Bugaeva (2022: Ch.1 §1); Shibatani (1990: §1). The sections that follow treat each dimension of that profile in turn, closing with a WALS-style feature summary.

2.2 Head-marking, indexing, and the absence of case

Nichols' parameter classifies languages by the locus of the morphological signal of a syntactic relation: on the head of the phrase, on the dependent, or on both. Head-marking correlates — verbal cross-reference of arguments, noun incorporation, instrumental/directional affixes on verbs, inflected adpositions, pronominal affixes on nouns — differ systematically from dependent-marking correlates — case on nouns, adnominal genitive marking, agreement suffixes on attributives, uninflected adpositions governing case Nichols (1986). Hokkaido Ainu shows the head-marking correlates and lacks all the dependent-marking ones across both clause and noun phrase Bugaeva (2022: Ch.1 §7); Satō (2025: 361).

The grammatical role of a noun phrase in the clause is signalled by person indexes on the finite verb, not by case suffixes on the NP itself. A transitive verb carries an A-index for the agent/transitive-subject and may carry an O-index for the patient; an intransitive verb carries an S-index. The NP argument is caseless. Third-person arguments are zero-indexed; first and second persons carry overt prefixal indexes for A/S and O roles Bugaeva (2022: Ch.1 §4.1); Nakagawa (2024: 163–169); Bugaeva (2012: 471).

The absence of morphological case in an SOV agglutinating language is cross-linguistically unusual. Japanese and Korean, Ainu's closest geographical neighbours, pair SOV order with elaborate nominal case systems; Ainu makes the same constituent order work without case, relying entirely on verbal person-indexing and semantics for argument tracking Bugaeva (2022: Ch.1 §4.1, §4.4); Satō (2025: 361). WALS-type surveys classify Hokkaido Ainu as "borderline" or "zero" in case inventory Bugaeva (2022: Ch.1 §4.4).

The full architecture of the S/A/O paradigm — four formal persons, the side-encoding of intransitive S versus transitive A, portmanteau exponents, and the mixed alignment that results — is set out in Chapter 62 (Architecture of the Personal-Affix System: The Four Persons and the S/A/O Paradigms).

2.3 Possession indexed on the head noun

In the noun phrase the same head-marking logic operates. A possessed nominal does not require a case-marked genitive possessor; instead the head noun itself carries two signals of the possessive relation — a personal prefix agreeing with the possessor, and an affiliative suffix marking the noun as standing in a possessed relation Bugaeva (2022: Ch.1 §4.4); Bugaeva (2021).

Within the nominal domain Ainu distinguishes obligatorily-possessed (inalienable) nouns — body parts, kin terms, and spatial-relational nouns that must carry a possessor — from nouns that are non-possessible Bugaeva (2022: Ch.1 §5.1). These possessive classes are absent from the surrounding Northeast Asian languages Bugaeva (2022: Ch.1 §7) and align Ainu with GPR-type possessive-class systems. The full classification, the morphophonology of the affiliative suffix, and the appositive possessive construction are treated in Chapter 43 (The Alienable/Inalienable Split and Possessive Classification), Chapter 45 (Morphophonology of the Affiliative Suffix and Echo/Copy Vowels), and Chapter 44 (Concept Form vs. Affiliative (Possessed) Form).

2.4 Polysynthesis and noun incorporation

Hokkaido Ainu is genuinely polysynthetic in Nichols' sense of open head-marking: there is no fixed closed paradigm of what categories may accumulate on the verb Bugaeva (2022: Ch.1 §4.7). A single verb word can stack person indexes, applicative prefixes, reflexive or reciprocal prefixes, incorporated nominals, and a number suffix while remaining one phonological word. The density of such accumulation is most visible in Classical Ainu, particularly in the heroic epic genre yukar Shibatani (1990: §1); Shibatani (1988):

In ordinary Hokkaido speech, a comparable degree of stacking is less frequent; a transitive verb routinely carries two person indexes and one or two derivational prefixes, and a single incorporated nominal already constitutes a polymorphemic complex. The register difference between Classical Ainu and everyday speech — and what it implies for the grammar/lexicon boundary — is a live research question Shibatani (1990: 5) ‹contested›.

Satō (2016; 2022) provides the fullest typology of Ainu noun incorporation, identifying four types that differ in their effect on verbal valency Satō (2016); Satō (2022):

The valency effects of each type — and in particular the A-incorporation type, unusual cross-linguistically — are treated in Chapter 88 (Object (Patient) Noun Incorporation), Chapter 89 (Subject Incorporation and Possessor-Stranding), and Chapter 90 (Oblique Incorporation and the Syntactic-vs-Lexicalist Debate).

How productive and how polysynthetic Ainu incorporation is in formal terms is contested. Shibatani (1988) analysed Ainu incorporation as a productive type-I lexical compounding process without semantic bleaching of the incorporated nominal, placing Ainu at the strong end of the polysynthesis scale Shibatani (1988). Bugaeva's (2022) characterization as "open head-marking" identifies why: the inventory of what may be incorporated or stacked is unconstrained by a fixed paradigm, unlike the closed agreement paradigms of dependent-marking languages Bugaeva (2022: Ch.1 §4.7) ‹contested›. The applicative system (e-, ko-, o-) further intersects with incorporation: an applicative argument can be simultaneously incorporated in ways that bear on the syntactic status of both operations Kaiser (1998); these interactions are treated in Chapter 78 (The Applicative System: Overview and the Preverb-vs-Applicative Debate).

2.5 SOV order and the head-final clause

The basic constituent order of Hokkaido Ainu clauses is SOV: subject precedes object, the finite verb closes the clause, and the clause is strictly head-final throughout. Modifiers precede their nominal heads; subordinate clauses precede the matrix verb; postpositional phrases — built on relational nouns and postpositions — follow the head noun with the adpositional element in final position Bugaeva (2022: Ch.1 §4.3); Satō (2025: 361); Nakagawa (2024: §4). The relational-noun system and the postpositional inventory are treated in Chapter 52 (Relational and Spatial (Locative) Nouns and Their Possessed Forms) and Chapter 53 (Postpositions, Local Case, Motion Events, Comitative, and Instrumental).

The SOV/head-final pattern sits in a typological tension with the head-marking type. GPR head-marking languages tend toward verb-initial order; Ainu's SOV gives it a surface resemblance to its dependent-marking neighbours (Japanese, Korean, Turkic) that does not extend to structural organization Bugaeva (2022: Ch.1 §4.3, §7). Bugaeva accounts for this via pragmatic right-dislocation of subject and object NPs: when arguments are shifted rightward for discourse reasons, the resulting surface order approaches verb-initial, aligning Ainu with the pragmatically common patterns of the GPR grouping Bugaeva (2022: Ch.1 §4.3). The full syntax of constituent order and argument realization is in Chapter 92 (Basic Constituent Order and Head-Final Syntax).

Community member @aomidori_cs_ (aynu-corpora Discord, 2024-01-29) has proposed that the SOV order is not original to Ainu but the result of contact with Northeast Asian SOV-dominant neighbours, citing the preverbal position of the negator somo ‘not (clausal negator)’ as a cross-linguistic diagnostic found in SVO languages aynu-corpora Discord (2023–2026) ‹speculative›. No comparative reconstruction of proto-Ainu word order is currently available to evaluate the proposal.

2.6 The alignment profile: mixed morphological type

The person-affix paradigm of Hokkaido Ainu shows a typologically mixed alignment. The first-singular row is nominative–accusative: the same prefix ku= marks both S (intransitive subject) and A (transitive subject), opposed to en= for O. The second-person rows are neutral: e= (2SG) and eci= (2PL) cover S, A, and O alike. The third person is zero in all three roles. The first-plural exclusive and the fourth person are tripartite: S is marked by a suffix (=as, =an) while A is marked by a prefix (ci=, a=) and O by a distinct prefix (un=, i=) Nakagawa (2024: 165–169); Bugaeva (2012: 472). This side-encoding — prefix for transitive A, suffix for intransitive S — is the morphological basis on which Bugaeva characterizes the overall system as "mixed, with elements of tripartite alignment" Bugaeva (2022: Ch.1 §1).

The detailed person paradigm, portmanteau exponents, and the full alignment debate — tripartite, hierarchical, inverse — are treated in Chapter 62 (Architecture of the Personal-Affix System: The Four Persons and the S/A/O Paradigms) and Chapter 72 (Tripartite, Nominative-Accusative, and the Central Alignment Debate). The split-intransitive question, which in Ainu surfaces in verb classes and valency morphology rather than in the agreement paradigm, is treated in Chapter 71 (The Split-Intransitive (Active–Stative) Core).

The fourth person (also termed the indefinite person) is a feature of the Ainu person system unusual enough to warrant mention here. A single form-set covers first-person-plural inclusive ('we including you'), indefinite/impersonal agent, second-person honorific address, and logophoric (protagonist) reference in oral-narrative direct speech Bugaeva (2022: Ch.1 §4.1); Nakagawa (2024: 163–164, 174); Bugaeva (2008: 40–43). Terminology for the set is unsettled across sources: Nakagawa uses "fourth person", Tamura lists the same forms as separate sub-persons, and Bugaeva analyses the quotation use as logophoric Tamura (1972: 18–20); Bugaeva (2012: 474–476) ‹contested›. The full referential range is catalogued in Chapter 67 (The Indefinite/Fourth Person: Forms, Reference, and Agent-Defocusing) and Chapter 68 (Honorific and Logophoric Uses of the Fourth Person).

2.7 Typologically unusual features in the regional context

2.8 The WALS-style feature bundle

The table below assembles the major typological parameters of Hokkaido Ainu. Sources for each feature are given in the preceding sections; primary reference is Bugaeva (2022: Ch.1) and Shibatani (1990: §1).

The head-marking profile aligns Ainu with the GPR typological cluster; the SOV order gives it a surface resemblance to Northeast Asian dependent-marking languages that does not carry into morphological organization. The combination of polysynthesis, a four-term analytical evidential system, and the incorporation-plus-applicative interaction makes the structural profile of Hokkaido Ainu distinctive even within the GPR grouping.