Chapter 123hawe ne — the Reportative / Hearsay Evidential

123.1 Place in the evidential paradigm

hawe ne is the reportative member of the Hokkaido Ainu evidential paradigm. Like its three siblings it consists of a formal noun in the possessed form followed by the copula ne ‘be, copula’: where ruwe ne signals a fact recognized through knowledge or reasoning, siri ne a directly visible situation, and humi ne a non-visual sensation, hawe ne signals that the information was acquired verbally — through hearing someone speak, receiving a report, or encountering hearsay. The architecture common to all four is set out in Chapter 120 (The Nominalization-plus-Copula Evidential Schema); the other three evidentials are treated in the inferential ruwe ne, the situational siri ne, and the sensory humi ne.

In Dal Corso's eight-form analysis, hawe ne is the direct evidential of hearing (DIR.HRN), the lowest-reliability direct form in the paradigm; its indirect counterpart is hawe as (IND.HRN), treated in §hawe as below Dal Corso (2018: 304–306, 337). In Dal Corso's reference texts, hawe ne accounts for 298 of 2232 evidential tokens (13.35%), and hawe as for a further 94 (4.21%), making the two hawe-forms together the second most frequent pair after ruwe ne Dal Corso (2018: 72).

123.2 Morphology: hawe is haw-e, 'the voice/word of —'

The source noun is haw ‘voice’. Tamura's Saru dictionary defines haw as '(人間・動物の)声、(バイオリンなどの)音' — the voice of a human or animal and the sound of certain instruments; general inanimate sounds such as a drum are expressed instead with the separate noun hum ‘sound, feeling’ Tamura (1996: s.v. haw, hawe). The ordinary possessed form is hawehe 'the voice of —'. The evidential formal noun is hawe = haw-e, the voice noun plus its possessive suffix, but like the other formal nouns of the evidential paradigm it blocks the additional possessive extension: the form with a further -he is ungrammatical as the evidential (*hawehe) Tamura (1996: s.v. hawe); Nakagawa (2024: 258). Glossing it as voice-POSS tracks this morphological origin.

The lexical noun haw ‘voice’ remains productive alongside the formal noun. The community notes the prohibitive construction iteki e=hawe aste 'do not raise your voice' — where e=hawe is the second-person possessed form of haw and aste ‘raise, cause to stand’ is the lexical verb — as evidence that the person-marked lexical noun is still active, distinct from the evidential formal noun which takes no person prefix aynu-corpora Discord (2023–2026) (nukopoli, aynu-corpora Discord, 2025-12-03) ‹contested›. A related observation: sekor haw=an '(one) says thus' and sekor isoytak '(one) narrates thus' are both attested as quotative expressions, differing in whether a general act of speaking or more formal narrating is meant aynu-corpora Discord (2023–2026) (thegodofneet, aynu-corpora Discord, 2025-06-03) ‹corpus-suggested›. In sekor haw=an the verb haw=an is the lexical 'to speak / give voice', not the formal noun; the two functions of the root coexist.

Chiri described the diachronic shift as running from 'there is a voice/rumor of X' (「云々の声がある」「云々の噂がある」) to the hearsay assertive Kindaichi & Chiri (1936) (cited via Hirosawa (2026: 66)), following the same cline documented for all four evidentials: sensory noun > possessed formal noun > clause-final evidential Takahashi (2018: 111).

123.3 Meaning: verbal source and reportative function

Nakagawa defines hawe ne as indicating that the speaker judges from information received verbally (言葉で聞いた情報から判断していること) Nakagawa (2024: 392). Three overlapping functions are documented in the sources.

The core use is the direct report: the speaker heard a specific person say something, and hawe ne encodes that hearing as the information source. Tamura describes this as '誰かが言ったことに基づいて述べる; 他人の発言の紹介、自分の発言の さらなる説明' — presenting what someone said, or elaborating on one's own earlier statement Tamura (1996: s.v. hawe) (cited via Hirosawa (2026: 67)).

The second function is hearsay: the information reaches the speaker through general social knowledge or rumor rather than from a named interlocutor. The example in Hirosawa (2026: 71, ex. 3), where a narrator attributes a plague to a divine figure's anger on the basis of circulating report rather than personal reasoning, illustrates this function; the example is drawn from Tamura 1988 audio materials (narrator Wateke, Saru) Hirosawa (2026: 71).

Third, Nakagawa notes for Chitose that hawe ne may mark the speaker's own statement, as in the framing expression hawe ne yakun 'if that is the case / given those words' Nakagawa (1995) (cited via Hirosawa (2026: 68)) ‹corpus-suggested›. In this use, the form makes explicit the verbal nature of what was just said.

Dal Corso places hawe ne at the lowest-reliability position of the four direct evidentials, reflecting the sense hierarchy sight > touch/taste/smell > hearing. Source reliability in his framework is a property of the acquisition channel, not a claim about epistemic certainty — hawe ne specifies how the information was obtained without committing the speaker to a degree of confidence in its truth Dal Corso (2018: 305–306, 327, 339) ‹consensus›.

Hirosawa (2026: 81) observes that near-identical sentences can alternate hawe ne and ruwe ne — one version presenting the proposition as heard speech, the other as assimilated knowledge — without a sharp semantic break Hirosawa (2026: 81); Dal Corso (2018: 306, 339). The boundary between the two forms is gradient rather than categorical ‹contested›.

123.4 The sekor … hawe ne quotative frame

The most prominent environment for hawe ne in the Saru and Chitose corpus is the quotative frame built with the complementizer sekor ‘(quotative)’. A clause of quoted content is followed by sekor and a speaking verb, with hawe ne at the clause edge:

[quoted content] sekor [verb of saying] hawe ne

sekor marks the boundary of the quoted content; the speaking verb specifies the act of utterance; hawe ne then asserts that the speaker's information derives from hearing. The full analysis of sekor and direct speech framing is in Chapter 105 (Quotative Complementation and Reported Speech (sekor)). The use of reported-speech frames in oral literature is described further in Chapter 150 (First-Person Narration, the Logophoric Fourth Person, and Reported Discourse).

The community has proposed that Saru sekor is analytically se + kor 'saying while holding', and that ari / ani — the equivalent in other dialects — preserves an older quotative form, with sekor a Saru-centred innovation aynu-corpora Discord (2023–2026) (nukopoli, aynu-corpora Discord, 2024-11-09; 2024-12-06) ‹contested›. These etymological questions belong to Chapter 105 (Quotative Complementation and Reported Speech (sekor)); the relevant point here is that the sekor … hawe ne combination is the canonical reported-speech construction in the Saru/Chitose corpus ‹corpus-confirmed›.

123.5 hawe as: the indirect reportative

The indirect counterpart of hawe ne is hawe as (phonologically hawas), formed by replacing the copula ne with the posture verb as ‘stand’, the same auxiliary used in humi as. The table below summarizes the two hawe-forms along Dal Corso's organizing parameters Dal Corso (2018: 304–330).

hawe as has specialized so thoroughly as a reportative that it is the only indirect form with a robust hearsay reading: humi as is consistently inferential and never reportative, while hawe as handles both the quotative use (overt speaker named) and the hearsay use (source diffuse or absent) Dal Corso (2018: 328–330) ‹corpus-confirmed›.

As a lexical item, hawe as means 'a voice is heard; someone speaks; a sound resounds' — the compositional reading of 'voice stands' Dal Corso (2018: 318). The evidential is an extension of this lexical meaning, and in some corpus occurrences the boundary between a literal sound event and a reportative attribution is genuinely ambiguous ‹speculative›.

123.6 Clause-final frames and illocution

The four formal nouns share a paradigm of clause-final frames; Nakagawa arranges them as: NMLZ + ne declarative, NMLZ + an content question (with a question word), bare NMLZ polar question and exclamation, and NMLZ + un affirmative answer Nakagawa (2024: 392–396). The hawe column is as follows.

The polar reduction hawe! follows the insubordination pattern documented for the other bare-nominalizer forms: a once-dependent nominalized clause used as a main clause to express surprise or strong emotion at the reported content Nakagawa (2024: 395); Takahashi (2013: 132). Interrogative frames in the wider system are treated in Chapter 144 (Interrogative Strategies, Question Particles, and Evidential Questions).

The hawe ta an? variant with an inserted ta is dialectally restricted, attested in Shizunai and Horobetsu and apparently absent from Saru/Chitose Nakagawa (2024: 394). Whether ta here is the locative/temporal postposition or a separate interrogative element is not settled in the accessible sources ‹speculative›.

123.7 Scope over TAM and the embedded clause

The clause dominated by hawe ne is a fully finite predication: it carries its own person marking, aspect, and perfective auxiliaries. hawe ne scopes over the entire TAM-marked clause, and negation belongs inside the evidential (… somo ki hawe ne 'it is said that [one] does not …'), not on hawe ne itself Nakagawa (2024: 393). Post-evidential material — the perfective a, final particles, conjunctions, and nominalizers — may follow hawe ne, exactly as for ruwe ne; the evidential sits above clause-internal aspect but below clause-linkers.

The ordering restriction — hawe ne kusu is grammatical while *kusu hawe ne is not in this meaning — confirms that the evidential sits at the outer edge of the VP and inside the scope of clause-linkers, the same layering found for ruwe ne Nakagawa (2024: 393) ‹consensus›. What accessible sources do not yet settle is the precise constituency status of hawe ne within the nominalized projection ‹speculative›.

123.8 Double evidentiality and narrative register

hawe as and hawe ne both enter double-evidential constructions with ruwe ne. The most frequent pattern is … hawe as ruwe ne, where a character's indirect verbal source is layered under the narrator's ruwe ne, which presents the event as traditional shared knowledge Dal Corso (2018: 331–333). A second pattern, … hawe ne nankor ruwe ne, uses ruwe ne as an epistemic reinforcer on a lower-reliability direct form Dal Corso (2018: 332). In both, the outer ruwe ne can ratify an already evidentially marked proposition as established fact; it behaves as a knowledge operator that scopes over the other evidential. The wider picture of evidential stacking is in Chapter 125 (Mirativity, Evidential Scope, and Grammaticalization to Sentence-Final Particles).

In narrative texts hawe ne marks the narrator's explicit attribution of a proposition to a verbal source, contrasting with the narrative default ruwe ne, which presents events as traditionally known facts. The genre-level distribution of the evidential forms is treated in Chapter 155 (Narrative TAM and Evidential Patterning by Genre).

123.9 Diachrony and dialect microvariation

The shift from sensory noun to formal noun is transparent and is noted in the earliest grammars: haw 'voice' > haw-e (possessed) 'the voice of —' > hawe ne 'it is said that …', parallel to the evolution of sir, hum, and ru Tamura (1996: s.v. haw, hawe); Kindaichi & Chiri (1936); Takahashi (2018: 111). Dal Corso's general assessment applies: the grammaticalization is partial — the lexical source noun is still visible in the blocking of -he, and the lexical verb haw as 'a voice sounds' remains active — so the shift is a matter of degree rather than categorical completion Dal Corso (2018: 24–25) ‹consensus›.

Dialect realization of the reportative evidential varies in the same way as the other three forms:

• Saru, Chitose, Biratori — the possessed form hawe in the full eight-form direct/indirect system Dal Corso (2018: 304–337).
• Tokachi (and frequently Ishikari) — the bare concept form haw closes the clause rather than the possessed hawe; aw is an attested Tokachi allomorph; iruska haw ne 'was angry (reportedly)' exemplifies the pattern Nakagawa (2024: 110, 390); Takahashi (2013: 130–131).
• Horobetsu — the possessed-form nominalizer hawe grammaticalizes a step further into adverbial-clause and conjunctive-particle uses not reported for well-described Saru, extending the 'versatile nominalization' cline Takahashi (2018: 107–112).

The dialect geography of the quotative particle that co-occurs with hawe ne mirrors this variation: Saru and Chitose use sekor while other dialects use ari / ani, a distribution that community discussion has analyzed as a Saru-centred innovation against a more widely preserved older form aynu-corpora Discord (2023–2026) (nukopoli, aynu-corpora Discord, 2024-12-06) ‹contested›. The full dialect picture of the quotative is described in Chapter 105 (Quotative Complementation and Reported Speech (sekor)).