Where the sources disagree

Satō notes that specific descriptions cannot be generalized indiscriminately across dialects, while observing that basic grammatical structures remain largely consistent.

Chiri abandons vowel-harmonic ra-row variation in favor of invariant ル for coda -r.

Kayano reverts the standard modern トゥ back to the older ド spelling.

Small-kana coda orthography resolves the ambiguity by reserving ッ for -tC- and small kana for geminates.

Tamura resolves the Akor Itak ambiguity by writing coda -t as ト rather than ッ.

Analyses differ on whether glottal stop is recognized as a phoneme or considered subphonemic

Refsing describes coda stops as weak plosives whereas standard descriptions characterize them as unreleased applosives

Yakumo shifts the peak to the third syllable when both first and second syllables are open, departing from the pan-Hokkaido second-syllable rule.

Pitch accent is phonologically distinctive in Hokkaido varieties, but non-distinctive in Sakhalin Rachishka.

Nakagawa reports slight terminal rising in affirmative sentences across Ainu, whereas Refsing reports slightly falling declarative intonation in Shizunai.

Questions have falling intonation in Sakhalin Ainu but rising intonation in Shizunai.

Declaratives have rising intonation in Sakhalin Ainu but falling intonation in Shizunai.

Shizunai questions exhibit rising intonation, while Saru attests interrogatives without final rise.

Nakagawa analyzes reduplication as expressing lexical aspect rather than number or action plurality.

Nakagawa presents an inventory of seven parts of speech in his 2022 description and six major word classes in 2024.

Hirosawa asserts ruwe can mark direct evidence, contradicting Sato's claim reported by Hirosawa that it marks only indirect evidence.

Hirosawa reanalyzes discourse tokens previously treated as integration/summarization as new-information presentation.

The traditional claim that adjectives categorically lack imperatives contrasts with the stative-inchoative analysis, under which property verbs can form imperatives in their inchoative reading.

In many dialects the morphological distinction between diminutive okkay-po and compound okkayo-po is collapsed into polysemous okkaypo.

Nakagawa (p.110) presents Sakhalin -(a)hcin as a nominal plural forming device contrasting with Hokkaido, but later (p.261) argues that -(a)hcin is not an exponent of grammatical nominal number.

Murasaki notes that unlike Hokkaido utar, Sakhalin 'uta(h) is not restricted to humans and can be used with all non-affiliative nouns.

Bugaeva asserts that -utar inside the possessive suffix yields an associative plural and outside yields a regular plural, whereas Shibatani asserts the exact reverse.

Kindaichi & Chiri claim utar cannot attach to inanimate nouns like shuma or ni, whereas Nakagawa states utar can attach to inanimates, citing suma utar and ni utar.

Nakagawa analyzes affiliative marking as potentially open to all common nouns depending on construal, whereas Refsing analyzes it as a closed lexical class of about fifty items.

Nakagawa documents affiliative kuwehe in Saru trap-bow contexts, whereas Satō states ku never takes affiliative forms.

Chitose data shows non-specific possessors can co-occur with affiliative possessed nouns, contrasting with Tamura's observation for Saru.

Contrasts the house noun uni (which inflects for person) with cise (which rejects person affixes in Saru).

One claim says exactly one classifier; the other says one or two verbs

Refsing analyzes kor as a postposition; Bugaeva analyzes it as a finite verb with the possessor as its argument

Kindaichi & Chiri restrict kor to human possessors; Bugaeva's analysis covers alienables more broadly

Shizunai shows kor-dominant pattern; general Ainu description posits three construction types

Satō notes that 1sg-prefixed younger sibling terms cannot address the listener, while Nakagawa notes junior interlocutors can generally be addressed with kinship terms.

Nakagawa notes Saru uses matapa for brother's younger sister, alongside Chitose macirpe and general Hokkaido tures.

Refsing states mici never takes a belonging form in Shizunai, while Nakagawa reports ku=mici for 1sg in Saru.

Shibatani accounts for the observed sentence-initial tendency as an artifact of subject pronoun omission rather than syntactic fronting.

Bugaeva describes wh-words as occupying their base in-situ slots, contrasting with Kindaichi and Chiri's generalization of sentence-initial positioning.

Positionals forming adverbial adjuncts without case postpositions versus with case postpositions.

Asahikawa is the sole Hokkaido exception to the vigesimal base hot

Refsing finds no regular general prefix template in Shizunai, in contrast to Tamura's three prefix slots for Saru.

Sakhalin uses singular 'oman where Saru uses arpa, both sharing plural paye.

Refsing finds that Shizunai permits plural verbs with numerals under certain conditions, directly contradicting the absolute prohibition reported by Chiri and Tamura.

While numeral subjects usually take singular verb stems in Shizunai, the attachment of a pronominal affix obligatorily conditions the plural stem.

Shibatani analyzes verbal number as lexical selectional restriction rather than agreement, whereas Nakagawa describes intransitive stem alternations as grammatical agreement with nominative personal a

Nakagawa defines auxiliary verbs as the wa-connected construction where the auxiliary bears personal affixes, whereas Satō defines auxiliaries as directly following verbs without taking personal affix

1sg follows nominative-accusative alignment whereas 2nd person shows neutral alignment without distinction between A, S, and O

Inclusive versus exclusive tripartite agreement paradigms

Hokkaido description specifies human recipient selection on kore and nukare

Ishikari has split O-V-S prefix-suffix combinations, whereas Saru combined affixes are exclusively prefixed.

Describes the circumfixal exception to the general prefixation of both arguments.

Nakagawa rejects treating -ka as having a distinct regional dialect distribution, whereas Bugaeva identifies it as more frequent in Northeastern Hokkaido and Sakhalin.

Contrasts which semantic role is privileged for person indexing among three-argument verbs (recipient with kore vs theme with omare).

Satō's direct versus indirect reflexivity analysis directly competes with the traditional volitional versus non-volitional analysis of the yay-/si- distinction.

Tamura and Nakagawa analyze the construction as lacking object-to-subject promotion, whereas Shibatani analyzes it as an intransitive passive.

Shibatani claims transitive subjects are systematically excluded from noun incorporation, whereas Satō establishes that transitive subject incorporation exists in Ainu.

Irregular lexical forms exhibit the reverse order Appl-N-V instead of N-Appl-V.

Nakagawa analyzes the incorporated element as a bare verb rather than an adverb that has dropped -no.

Satō rejects productive adverbial incorporation in Ainu, whereas Shibatani asserts that adverbs incorporate rather freely.

Satō views verbal adverbial prefixes as virtually limited to ar-, whereas Nakagawa lists a broader inventory of six prefixes.

SV/AOV formulation adopts Dixonian syntactic primitive labels rather than traditional SOV.

Pronominal subject prefixes yield surface OSV, contrasting with nominal subject SOV.

Satō analyzes overt nouns as appositional adjuncts to a bound prefix, whereas Bugaeva treats third-person subject omission as null-argument ellipsis of an NP.

Both concern kor but one addresses its semantic use (form-function) while the other addresses its morphosyntactic incorporation analysis

Statement 4 classifies kotom/ruwe/hawe/hita as nominalizers under a broad ainu scope, while statements 5/7 classify kur/utar/pe/usi/uske/hita as 'concrete' nominalizers in Shizunai dialect, reflecting

While standard of comparison proper is inaccessible, an equivalent construction exists via degree postposition pak(-no) stranding.

Bugaeva provides counterexamples to Tamura's claim that first person is never used in quotes referring to another speaker.

Ishikari shares sekor with Saru and Chitose but replaces hawean with itak.

Central Hokkaido (Asahikawa) allows sekor to introduce indirect reported discourse, whereas Southern Hokkaido dialects do not.

Nakagawa identifies emphasis of the extraordinary in conversational dialects, whereas Satō identifies a softened nuance in oral literature.

Bugaeva documents that yak is also attested with current and future events, contradicting Tamura's claim that yak is restricted to past events.

Ishikari adds final particle ya to singular imperatives rather than relying solely on the bare verb.

Shizunai uses na whereas Saru uses ro for hortative expressions.

Both express inability in Sakhalin Ainu but contrast lack of capacity with circumstantial impediment.

Shizunai uses kusu ki for the intentional mood where other Hokkaido dialects use kus(u) ne.

Negative individual-level restriction complementary to stage-level predication

Bugaeva asserts that ka adds focus and unexpectedness, whereas Nakagawa states that ka is optional and does not add unexpectedness or additive meaning.

Satō views ka in negative expressions as lacking emotional/emphatic nuance, in contrast to Bugaeva's focus/unexpectedness analysis.

Sakhalin vs general Ainu description of the resultative disappearance construction with wa isam

The two claims partition the set of negative auxiliaries by their syntactic distribution: some allow ka and independent use, others do not.

Disagreement on whether conjunction markers prosodically group with the preceding predicate before a pause or with the following clause after sentence-final intonation.

unexpectedness and discovery account vs non-causal resultative account

Horobetsu prefers awa whereas Shizunai and Tokachi prefer akusu

Nakagawa characterizes Sakhalin kusu as expressing continuation and resulting state (like Hokkaido kor and wa) rather than cause/purpose, whereas Murasaki describes it as expressing purpose, reason, o

Contrasts the scalar 'not even' reading in negative contexts with the additive 'also' reading in positive contexts.

Refsing treats post-consonantal [a] as an allomorph of ya in Shizunai, whereas Nakagawa shows a is an independent particle in Chitose and Ishikari.

Sakhalin Ainu possesses an age-graded register split that is reported absent in Hokkaido dialects.

In Chitose popo functions in a request to an older brother, whereas in Saru it is an affectionate vocative suffix for children generally.

Nakagawa claims interjections never enter grammatical relations with other elements, while Refsing observes them initiating or interpolating into full sentences.

Refsing describes ononno in Shizunai as expressing happiness and gratitude, whereas Kindaichi & Chiri gloss it as praise or triumph ('hurrah').

The prototypical protagonist of human prose tales differs between southwestern Hokkaido (unnamed ordinary human) and central/eastern Hokkaido (Otasut un kur).

Prose tales lack melody in Hokkaido, whereas Sakhalin tuytah tales frequently embed sung melodic passages.

Nakagawa proposes that fourth-person narration was original with first-person plural inserted for poetic effect, contradicting Kubodera's proposal that first-person plural is the archaic stratum.

Nakagawa criticizes the traditional terminology 'first-person narrative' affirmed by Satō, arguing the grammatical person is not an ordinary first person.

Both describe plural verbal agreement with a singular narrative protagonist in non-southwestern dialects, formulated respectively within a logophoric vs. fourth-person narrator framework.

Attests an isolated counterexample to the strict genre split between ci- and a- series forms in Horobetsu.

Tamura describes unshifted non-first persons in quote clauses, which Bugaeva characterizes as the unshifted second person defining semi-direct discourse.

Both describe singular verb stems used for singular protagonist reference in southwestern dialects, framed as logophoricity vs. fourth-person narrator agreement.

Both involve toy but in different couplet functions: pun vs. intensification of badness

Eastern Hidaka substitutes kaykuma for rabbit, whereas Shiraoi substitutes tukisarus.

Dispute over whether Proto-Ainu possessed only distinctive vowel length or both vowel length and pitch accent.

Kindaichi & Chiri confine arpa to Hidaka, whereas Nakagawa & Fukazawa report arpa in Chitose as well.

Contrasts Kindaichi & Chiri's Hidaka dialect boundary with Nakagawa's Saru dialect group.

Notes the use of totto for mother in the Mukawa interior alongside its occurrence in Ishikari and Chitose.

Shows the regional divergence in the meaning of ihunke between 'to curse' and 'lullaby'.

Demonstrates the reciprocal penetration of pa- and ca- across the mouth isogloss boundary.

Attributes the nasal before voiced obstruents to donor dialect prenasalization in Tōhoku Japanese rather than an Ainu-internal insertion rule.

Both involve Ainu i- but in different grammatical functions (indefinite object vs. nominal prefix)

Both involve EOJ i- but sourced from different Ainu morphemes (e- vs. i-) with different functions

Both EOJ pa forms share the same Ainu source form pa but with different meanings

Kurils have only pet, while Hokkaido/Tohoku have both elements