Chapter 16The Consonant Inventory and Its Phonetic Realization
The consonant inventory, its positional realizations, and differences among speakers and descriptive analyses.
16.1 The consonant phonemes
The ordinary Latin transcription of Hokkaido Ainu uses eleven consonants: /p, t, k, c, s, h, m, n, r, y, w/. Nakagawa and Satō describe an inventory of eleven, while Bugaeva includes the glottal stop /ʔ/ and counts twelve Nakagawa (2024: 27); Satō (2008: 10); Bugaeva (2012: 465). The difference concerns the analysis of glottal closure, whose occurrence is recognized in both approaches; see Chapter 20 (Glottal Stops and the Distribution of /h/) ‹contested›.
| Manner | Labial | Coronal | Dorsal | Glottal |
|---|---|---|---|---|
| Stop | /p/ | /t/ | /k/ | /ʔ/ in analyses that count it as a phoneme |
| Affricate | /c/ | |||
| Fricative | /s/ | /h/ | ||
| Nasal | /m/ | /n/ | ||
| Rhotic | /r/ | |||
| Glide | /w/ | /y/ [j] |
The table gives broad places of articulation. In particular, /y/ is palatal and /w/ is labial-velar. Letter values follow Ainu romanization: c represents an affricate and y the glide written [j] in IPA Shiraishi (2022: §3).
16.2 Syllable-initial consonants
The stops /p, t, k/ have bilabial, dental, and velar articulations. A voiced pronunciation does not create a contrast with a separate series /b, d, g/. Nakagawa reports that /p/ is particularly liable to sound voiced, with /t/ and /k/ showing the tendency less strongly Nakagawa (2024: 28).
The affricate /c/ is commonly described as [tʃ] or [tɕ]. Bugaeva records speaker variation between [tʃ] and [ts] in cup “sun; moon” Bugaeva (2012: 466). Shiraishi, reporting Tamura, specifically notes a [ts] realization before /u/ Shiraishi (2022: §3). These descriptions concern variation within the affricate phoneme.
The two fricatives have different distributions of phonetic variation. The palatalization of /s/ is described in Chapter 18 (/s/ and the s ~ š Palatalization Alternation). For /h/, Nakagawa describes [h], [ç], and [ɸ] realizations according to the following vowel. Shiraishi reports Tamura's observation that this conditioning can be weak, especially in recordings of older speakers Nakagawa (2024: 28); Shiraishi (2022: §3). A single phonetic table would conceal some of this speaker variation.
16.3 Syllable-final consonants
Nine consonants occur as Hokkaido syllable codas: /p, t, k, s, m, n, r, y, w/. The affricate /c/ and fricative /h/ are excluded from this position. Analyses with /ʔ/ also exclude it from the coda Nakagawa (2024: 27); Bugaeva (2012: 467). The permitted combinations are described in Chapter 21 (Syllable Structure, Phonotactics, and Word-Edge Constraints).
Final stops are unreleased: the closure is made without a following audible burst. Bugaeva gives kap [kap̚] “skin; bark,” kut [kut̚] “girdle,” and sik [sik̚] “eye” Bugaeva (2012: 465). A final stop is therefore a consonant of the syllable even when its release is difficult to hear.
Sakhalin comparisons require a separate inventory of permitted codas. Tangiku describes the widespread replacement of historical final stops by /h/, while also identifying Sakhalin localities that preserve final stops Tangiku (2022: §§2.2–2.3). These dialect differences should be retained in comparisons; see Chapter 162 (Sakhalin and Northern Kuril Comparisons).
16.4 Nasals and place assimilation
The nasal phonemes are /m/ and /n/. Final /n/ is ordinarily articulated with the tongue tip at the teeth, but it takes the place of a following consonant: [m] before /p, m/ and [ŋ] before /k/. Nakagawa illustrates the latter with honkor [hoŋkor] “be pregnant” Nakagawa (2024: 30).
Nasals before labials are written differently across dictionaries. A spelling with n can preserve a morphological relationship or follow a convention for nasal neutralization; it need not indicate an alveolar pronunciation in that position Nakagawa (2024: 58–59). The alternatives are explained in Chapter 15 (Orthographic Standardization and the Word-Division Problem).
16.5 Voicing variation
Stops and the affricate can be voiced between vowels, after a nasal, and after /r/. Shiraishi gives hapo ~ habo “mother,” tampe ~ tambe “this one,” and unarpe ~ unarbe “aunt,” reporting the descriptive literature Shiraishi (2022: §3.1). Voicing is variable in these environments.
Shiraishi also reports Yamashita's study of recorded Saru speech, in which the post-nasal environment had the highest voicing rate, approximately 40%. That figure describes the study's sample. It is evidence of a tendency, with ample unvoiced realizations remaining Shiraishi (2022: §3.1).
16.6 The rhotic
The common realization of /r/ is a tap [ɾ], with trilled, lateral, and stop-like variants reported in the literature. In final position it often has a vowel-like release Nakagawa (2024: 28, 30); Shiraishi (2022: §3.2). The contrast between final /r/ and /r/ followed by a vowel is established by pronunciation and by morphological behavior. It is examined in Chapter 19 (The Rhotic /r/ and Coda-r).
16.7 Glides
The glides /y/ and /w/ occur at the beginning and end of syllables. In final position they have vowel-like realizations, but their morphological behavior supports grouping them with consonants. Shiraishi points to the selection of the consonant-following nominalizer in okay-pe “those who are” and siw-pe “bitter thing” Shiraishi (2022: §3). The evidence and the older diphthong terminology are discussed in Chapter 22 (Glides, Vowel Hiatus, and the Diphthong Question).
References cited in this chapter
Bugaeva (2012) ·Nakagawa (2024) ·Satō (2008) ·Shiraishi (2022) ·Tangiku (2022)