Chapter 124humi ne — the Non-Visual Sensory Evidential
humi ne ('it is the sound/feeling that'), marking evidence from non-visual senses — non-verbal sound and bodily/internal sensation — and inference from such cues.
124.1 Place in the evidential paradigm
Among the four direct evidential forms of Hokkaido Ainu, humi ne marks that the speaker's information derives from a non-visual sensory source: a physical sound, a smell or taste, a touch or bodily sensation, or an internal impression that is neither visible nor verbally reported. The construction — a nominalized clause headed by a formal noun and closed by the copula ne — is common to all four terms and is described in Chapter 120 (The Nominalization-plus-Copula Evidential Schema). The other three direct forms are treated in the inferential ruwe ne, the situational siri ne, and the reportative hawe ne.
Dal Corso's morphosyntactic account arranges the four direct forms along a source-reliability scale: ruwe ne > siri ne > humi ne > hawe ne. Source reliability in this account designates the psychophysical channel of acquisition, not a degree of epistemic certainty: a speaker using humi ne specifies non-visual sensory access, not a lower confidence in the proposition Dal Corso (2018: 304, 312–313, 339). The visual/non-visual macrodivision groups ruwe ne and siri ne on the visual side and humi ne and hawe ne on the non-visual Dal Corso (2018: 312). Mapped onto the typology of (Aikhenvald 2004: 63–65), the system is C1-type (visual, non-visual sensory, inferred, reported) Hirosawa (2026: 64).
humi ne is among the less frequent members of the paradigm. In Dal Corso's Hokkaido reference texts it supplies 45 of 2232 evidential tokens (2.02%), with its indirect counterpart humi as adding a further 37 (1.66%) Dal Corso (2018: 72). Both counts place the sensory evidential as the third-most frequent direct form and second-to-last in the full eight-form system ‹corpus-confirmed›.
124.2 Morphology: humi is hum-i, 'the sound/feeling of —'
The formal noun humi is the possessed (affiliative) form of the noun hum ‘sound, feeling’. Tamura's Saru dictionary records hum (ordinary possessed form humi(hi)) with the primary senses 音・感じ 'sound; feeling/sensation' Tamura (1996: s.v. hum, humi(hi)). The same source noun underlies the lexical predicate humas (hum-as 'to sound, there is a noise'), where as ‘stand’ is the posture-verb element that also appears in hawe as and in other transient-state predicates such as apto as 'it rains' and rera as 'the wind blows' Dal Corso (2018: 318, 329).
As a formal noun in the evidential function, humi blocks the additional possessive ending that the full lexical possessed form may carry: the regular possessed noun is humi(hi), but evidential *humihi is ungrammatical — the same blocking that disallows *ruwehe for the inferential Tamura (1996: s.v. humi, ruwe); Nakagawa (2024: 258). This restriction, shared by all four formal nouns and discussed in Chapter 35 (The Formal / Defective Noun Set (ruwe, hawe, siri, humi, hi, pe, kur)), is a diagnostic of partial grammaticalization: the form has shed ordinary head-noun syntax without yet becoming a bound morpheme Dal Corso (2018: 24–25) ‹consensus›. The standard gloss is sound/feeling-poss: the morpheme is hum-i (base plus possessive suffix), whose implicit possessor is the nominalized clause.
The scope of hum sets it apart from the adjacent source noun haw ‘voice’. Tamura is explicit: inanimate and non-vocal sounds — a drum, ambient noise, a sound with no identified speaker — take hum, while human and animal voice takes haw Tamura (1996: s.v. haw). This lexical distinction is grammaticalized in the evidential pair: humi ne for non-verbal sensory access, hawe ne for verbal report and hearsay. The construction is a copular predication over a nominalized clause — [clause] humi 'the sound/feeling that [clause]', closed by the copula ne (treated in Chapter 96 (The Copula ne and Predicate-Nominal Clauses)).
124.3 Meaning: the non-visual sensory domain
The traditional label for this form is Kindaichi's 感説法 'feeling-mood', characterizing it as expressing 'it feels like … / it seems like …' in the modality of uncertain sensory judgment Kindaichi & Chiri (1936); Hirosawa (2026: 65–66). The modern evidential framing retains the source semantics. Tamura specifies humi ne as appropriate for hearing, touch, and bodily or emotional feeling Hirosawa (2026: 67). Nakagawa's Chitose dictionary extends the domain to 音・匂い・味・体感・直感 — sounds, smells, tastes, bodily sensations, and intuitions obtained through any means except sight and verbal report Nakagawa (1995); Hirosawa (2026: 68, 71). His 2024 grammar gives the standard definition: humi ne marks that the speaker's judgment rests on sensation other than sight and language Nakagawa (2024: 392). Satō's question paradigm assigns 感覚・物音 'sensation, physical sound' to the humi domain Satō (2008: 135).
Dal Corso glosses the form dir.flt (direct evidential, feeling) and lists the covered channels as hearing, smell, touch, taste, and an internal 'sixth sense' Dal Corso (2018: 306). He documents a recurring sensor-stimulus mismatch: the evidential derives historically from the noun meaning 'sound, feeling', yet it licenses gustatory and other non-auditory stimuli Dal Corso (2018: 307, 311). The following example shows a gustatory stimulus under the sound-derived evidential — the embedded clause concerns being given food to eat, not hearing a sound:
‘Whether I was given all sorts of delicious things to eat…’
Dal Corso 2018: 306; Saru, Wateke (Saru), Tamura audio materials; Dal Corso siglum TMA: 42
Dal Corso ex. 242 (DIR.FLT). A gustatory stimulus (being given food) appears under the sound/feeling-derived humi ne, illustrating the sensor-stimulus mismatch discussed at pp. 307, 311. The verb complex a=i=y-e-re is a causative benefactive ('caused to eat for 4O'); ya is a question/dubitative particle.
The extension from auditory to other senses is not unexpected: the source noun hum ‘sound, feeling’ already covers both auditory and somatic senses lexically, and as it grammaticalized into a formal noun its scope broadened across all perceptual channels not taken by siri (sight) or hawe (verbal report). Chiri noted the expansion from sound to taste, colour, and smell in the earliest grammatical account Hirosawa (2026: 66). The following table shows the attested domains:
| domain | typical situation | attested source |
|---|---|---|
| physical (non-verbal) sound | ambient noise, an unidentified shot or movement, a sound heard without visual source | Hirosawa (2026: 71); ILCAA Ainu materials (1976) |
| taste / gustatory | evaluating food or drink; being given something to eat | Dal Corso (2018: 306–307); Nakagawa (2024: 395) |
| touch / temperature | a surface felt to be frozen solid, cold, or altered in texture | Kindaichi & Chiri (1936); Hirosawa (2026: 66) |
| internal emotional state | anger, fear, or emotion perceived as rising inside the speaker | ILCAA Ainu materials (1976) |
| internal bodily impression / intuition | a bodily transformation sensed from within; a vague somatic impression | Dal Corso (2018: 306); ILCAA Ainu materials (1976) |
Touch and temperature are attested from the early grammatical literature. Chiri's canonical example, cited through Hirosawa, encodes the tactile state of a surface frozen solid:
‘It had completely frozen solid.’
Hirosawa 2026: 66; Hokkaido (dialect not further specified)
Chiri 1936: 192, cited via Hirosawa 2026: 66. A bodily-tactile state (freezing solid) is encoded with humi ne; the admirative tail awan carries the surprised-retrospective value. The source noun hum 'feeling' grounds the tactile domain in the same formal noun that covers sound.
Internal emotional states use the same construction. In the following example, the narrator perceives anger welling up from within — an internal bodily impression with no external auditory source:
‘It felt as though anger was welling up.’
ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 'Folktale 13' (Saru); AA研 aa-irc/016, sent. 122
The fourth-person prefix a= marks the agent in the narrative frame. The periphrasis humi ne pekor yay-nu=an 'it felt as though' is a common collocation combining the sensory evidential with the simulative pekor and the reflexive-perception verb yaynu; see the section on frames for related variants.
A common periphrasis in the corpus combines humi ne with a simulative particle and a mental-state verb — humi ne kunak ku=ramu 'I thought it felt like …' or humi ne pekor yaynu=an 'it seemed as though …' — embedding the sensory evidential as a complement to a perception or cognition verb. The following example uses the kunak variant:
húmas humi ne kunak ku=ramu kor ku=soyne a p
‘Thinking that it sounded like there was a noise, I went outside — but (nothing was there).’
ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 'Folktale 8 (Part 1)' (Saru); AA研 aa-irc/006, sent. 358
The contracted form húmas (= hum-as 'a sound arises') provides the sensory stimulus; humi ne encodes the speaker's non-visual sensory access to that sound; kunak introduces the complement of the cognition verb ramu 'think/feel'. The nominalized tail p sets the clause as backgrounded against a disappointed expectation.
The boundary with siri ne is explicit in the definition: siri encodes a directly visible situation or appearance, while humi covers any percept not accessed through sight. The boundary with hawe ne runs along the verbal/non-verbal divide: hawe (from haw ‘voice’) encodes information obtained from words — spoken, reported, or overheard — whereas humi covers physical sounds and bodily perceptions that are not themselves linguistic utterances Tamura (1996: s.v. haw, hum).
124.4 Clause-final frames and illocution
humi participates in the shared clause-final frame paradigm documented by Nakagawa Nakagawa (2024: 392–396) and summarized in Chapter 120 (The Nominalization-plus-Copula Evidential Schema). The interrogative frames are treated in Chapter 144 (Interrogative Strategies, Question Particles, and Evidential Questions).
| frame | illocutionary value | note |
|---|---|---|
| humi ne | declarative, non-visual sensory direct evidential | formal noun + copula; may take final particles and clause-linkers after it |
| humi an? | content question / dubitative (内容疑問文) | existential replaces copula; Tamura records hnta ninpa wa arpa humi an? 'where might (she) have gone?' Tamura (1996: s.v. humi) |
| bare humi | polar question / exclamation | reduced, insubordinated frame; exclamatory force is common, especially for gustatory and thermal sensation |
| humi ka isam | 'not feel … at all' | total absence of the non-visual percept; uses the negative existential isam; recorded as a fixed expression Tamura (1996: s.v. humi) |
The bare exclamatory humi! is the most distinctive member. It packages an immediate sensory reaction — most frequently gustatory or thermal — as its own utterance, without the copula:
usey ku=ku wa, keraan humi!
‘I drank tea — how delicious!’
Nakagawa 2024: 396; Chitose, Nakagawa ex. 793, Chitose
The bare formal noun humi functions as an exclamative main clause, without the copula ne. The gustatory evaluation triggers the non-visual sensory evidential; the insubordinated frame conveys immediate spontaneous reaction.
Negation follows the pattern shared by all four evidentials: the proposition is denied within the embedded clause (the pattern …somo ki humi ne 'there is no feeling that [one does …]'), with humi ne remaining outside the scope of negation. The fixed expression humi ka isam expresses the complete absence of the relevant non-visual percept. The interaction of the evidential with the wider TAM and negation apparatus is treated in Chapter 100 (TAM, Evidentiality, and Negation on Nonverbal Predicates).
124.5 The indirect counterpart: humi as
The indirect sensory evidential humi as replaces the copula ne with the posture verb as ‘stand’, shifting the evidential reading from direct to more distanced sensory access: the speaker has sensory grounds but presents them as slightly removed or inferential Dal Corso (2018: 306, 328). Dal Corso glosses it ind.flt (indirect evidential, feeling). The opposition copula ne (internal, direct) vs posture/existential verb (external, indirect) runs across all four formal nouns and is set out in Chapter 120 (The Nominalization-plus-Copula Evidential Schema).
A functional asymmetry distinguishes humi as from its verbal-source counterpart. hawe as has largely specialized as a reportative — encoding hearsay when the source is absent and quotation when it is overt Dal Corso (2018: 328–330). humi as retains its sensory basis throughout: it is consistently inferential from a non-verbal sensory impression and is never reportative Dal Corso (2018: 328) ‹corpus-confirmed›. As a lexical predicate, humas (hum-as) means 'to make a sound, to resound' Dal Corso (2018: 318). The physical-sound use is the most frequent in the corpus:
‘The sound of someone arriving was heard.’
ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 'Folktale 8 (Part 2)' (Saru); AA研 aa-irc/007, sent. 200
The posture verb as replaces the copula ne: the narrator infers the person's arrival from an auditory impression rather than direct sight. The indirect humi as contrasts here with the direct humi ne that would encode the arrival as directly sensed.
124.6 Double evidentiality with humi
Like the other formal nouns, humi enters double-evidential constructions in which two evidentials stack within a single clause Dal Corso (2018: 331–333). The combination humi ne humi as stacks the direct and indirect sensory evidentials on the same non-visual channel:
‘It felt as though I had become a great fish, and then I swam after my elder brother.’
ILCAA Ainu materials 1976; Saru, Kawakami Matsuko, uwepeker 'Folktale 11' (Saru); AA研 aa-irc/013, sent. 32
Dal Corso 2018: 333 (ex. 275 context). The direct humi ne presents the bodily transformation as directly felt; the immediately following humi as then hedges it as also inferential, conveying that the same sensation was simultaneously vivid and uncertain. Both evidentials draw on the non-verbal sensory channel.
This pattern differs from the double-evidential constructions in Chapter 121 (ruwe ne — the Inferential / Visual-Trace Evidential), where ruwe ne ratifies an indirect form from a distinct channel as established knowledge. In humi ne + humi as, the same sensory channel provides both layers; the two evidentials together convey a closely felt percept reported with residual uncertainty Dal Corso (2018: 331–333).
124.7 Scope, mirativity, and the admirative tail
The constituent nominalized by humi is a full finite clause carrying its own person marking, aspect, and negation, over which the evidential then scopes. The scope properties are shared across all four formal nouns and are set out in Chapter 120 (The Nominalization-plus-Copula Evidential Schema); nothing in the structure of humi ne alters that general picture.
Mirativity is carried in Hokkaido Ainu by the auxiliaries aan / awan and by anan, not by the formal-noun evidentials themselves Hirosawa (2026: 90); Dal Corso (2018: 336). Chiri's freezing example above, rupus humi ne awan, shows the admirative tail awan occupying the post-evidential slot: the surprised-retrospective reading comes from awan, not from humi ne itself. The full treatment of mirativity and its interaction with the evidential system is in Chapter 125 (Mirativity, Evidential Scope, and Grammaticalization to Sentence-Final Particles).
124.8 Diachrony and dialect variation
The grammaticalization path parallels that of the three sibling formal nouns. The noun hum ‘sound, feeling’ grammaticalized first into the possessed nominal hum-i, and from there into a clause-final formal noun whose implicit possessor is the nominalized event rather than a personal possessor Tamura (1996: s.v. hum); Hirosawa (2026: 65–66). The overall cline — sensory noun > possessed formal noun > evidential — is shared by sir, haw, and ru and places the Ainu system within the nominalization-based evidential strategy discussed by Dal Corso (2018: 24–25); the path is set out for the full paradigm in Chapter 120 (The Nominalization-plus-Copula Evidential Schema). Dal Corso's account of the degree of grammaticalization applies to humi as to the others: partial, with residual nominal behaviour visible in the possessive blocking Dal Corso (2018: 24–25).
Dialect variation in the realization of humi follows the system-wide patterns:
- Saru, Chitose, Biratori — the core of Dal Corso's data — use the possessed form humi throughout, with the direct/indirect split humi ne / humi as Dal Corso (2018: 304–337).
- Tokachi (and frequently Ishikari) close the clause with the bare concept form hum rather than the possessed humi; the declarative frame attested by Takahashi appears as … ran hum ne awan Takahashi (2013: 129–131); Nakagawa (2024: 110).
- Horobetsu grammaticalizes the possessed-form nominalizers a step further into adverbial-clause markers and conjunctive-particle uses not attested in well-described Saru — an instance of what Takahashi calls 'versatile nominalization' Takahashi (2018: 107–112).
Ijäs's pedagogical grammar treats humi ne under the evidential framing without adjudicating the analytical disputes discussed in Chapter 120 (The Nominalization-plus-Copula Evidential Schema) Ijäs (2023). The genre patterning of humi ne within narrative texts and its distribution relative to ruwe ne across frame-text, reported inner experience, and direct speech are taken up in Chapter 155 (Narrative TAM and Evidential Patterning by Genre).
References cited in this chapter
Aikhenvald (2004) ·Dal Corso (2018) ·Hirosawa (2026) ·Ijäs (2023) ·ILCAA Ainu materials (1976) ·Kindaichi & Chiri (1936) ·Nakagawa (1995) ·Nakagawa (2024) ·Satō (2008) ·Takahashi (2013) ·Takahashi (2018) ·Tamura (1996)