Owase

Shu Hirata

Jump to Content

Chapter 5 Owase

In: Word-Prosodic Systems of Japonic Languages

Author:

Shu Hirata

Shu Hirata
Search for other papers by Shu Hirata in
Current site
Google Scholar
PubMed

Type:: Chapter

Pages:: 156–171

DOI:: https://doi.org/10.1163/9789004755789_006

Access via:

Dar Hadith al Hassania

Download PDF

1 Introduction

1.1 Location

Owase is spoken in the Owase district of Owase City, which is situated in the southern region of Mie Prefecture. The prosodic system of Owase exhibits distinct features that set it apart from the varieties spoken in other districts of Owase City (Kindaichi 1975), that is, Kuki, Kita-Wauchi, Minami-Wauchi, and Sugari districts (Owase-shi-shi Hensan Junbikai ed. 1969: 652).

1.2 Speakers

As of April 2024, the total population of the Owase district is reported to be 13,695 (Owase City 2024). However, owing to ongoing changes in the prosodic systems among the younger generation, the number of Owase speakers who possess the prosodic system described in this chapter is estimated to be 10,000 or fewer. A previous survey of young people (teens at the time of 2014) conducted by the author documented a change in the prosodic system; the figure of 10,000 is due to the fact that about 3,000 people in their 30s or younger are recorded in the Owase district.

1.3 Classification

Tojo (1954) states that Owase is classified within the Kinki group (see § 2.1 of Chapter 1, this volume). However, Owase exhibits remarkably different prosodic features, both synchronically and diachronically, from well-known varieties of Kinki such as those found in Kyoto and Osaka.

1.4 Previous Studies

Kindaichi (1959) conducted the earliest study of the prosodic system of Owase. Subsequently, Uwano (1987: 31–32) identified diachronic changes that distinguish it from other Kinki varieties across several varieties in Owase City, including the Owase variety, specifically in the tonal mergers of two-mora words between Owase variety and Kinki varieties and the tonal change in the registers. Hirata (2020a) provides a comprehensive description of the prosodic system of Owase, while Hirata (2020b) details the vocative intonation patterns that appear in personal names when calling out to someone.

1.5 Data

All data presented herein are based on a female speaker born in 1954 and were collected through the author’s fieldwork primarily conducted at Owase City Central Community Center, Mie Prefecture. The data were collected by having the speaker pronounce a list of words or short sentences prepared by the author.

1.6 Overview of the Segmental Phonology

Owase has five vowel phonemes (1).

(1) Vowel phonemes

/i, e, a, u, o /

The phoneme /u/ is realized as [ɯ] and vowel length is distinctive. In this chapter, long vowels are transcribed as a sequence of identical vowels, for example /toori/ [toːɾi] ‘street’.

Owase has 15 (or 17) consonant phonemes (2) (Table 5.1).

(2) Consonant phonemes:

/p, t, k, b, d, ɡ, s, h, c, z, m, n, r, j, w, (ɸ, β)/

Based on fieldwork, the phoneme /c/ is realized as a voiceless alveolar affricate [ts]. The phoneme /z/ can be realized as either an affricate [dz] or a fricative [z]. Alveolo-palatal allophones are systematically observed in alveolar obstruents /c, z, s/, realized as [tɕ], [ʑ ~ dʑ], and [ɕ], respectively, before /i/ or /j/. The phonemes /ɸ/ and /β/, which are observed only in recent loanwords, are indicated in parentheses above.

Table 5.1

Consonant phoneme inventory and typical allophones. (Phonemes are indicated by slashes, whereas typical allophones are listed in parentheses. Symbols in square brackets are used for phonemic transcriptions where the phonetic realization may not be immediately apparent.)

	Bilabial		Alveolar		Alveolo-Palatal			Velar		Glottal
Plosive	/p/	/b/	/t/	/d/				/k/	/ɡ/
Affricate			/c/ [ts]	/z/ [dz]	(tɕ)	(dʑ)
Nasal		/m/		/n/
Tap				/r/ [ɾ]
Fricative	/ɸ/	/β/	/s/	(z)	(ɕ)	(ʑ)				/h/
Approximant							/j/		/w/

The syllable structure in Owase follows the pattern C₁ G V₁ V₂ C₂, with the nucleus vowel V₁ being obligatory. The other elements—the onset consonant C₁, glide G, second element of a diphthong or long vowel V₂, and coda consonant C₂—are optional. The coda slot C₂ can be filled by either a nasal or the same obstruent as the onset of the following syllable. The glide /j/ can occupy the G slot. The rhyme slots, namely V₁, V₂, and C₂, each have a length of one mora.

2 Word-Prosodic System

2.1 Classification Based on Uwano (1989, 1999)

The prosodic system of Owase is classified as Uwano’s (1989, 1999) multipattern system, which is defined as the system where the number of tonal contrasts increases proportionally with the word length (see § 2.4 of Chapter 1, this volume).

2.2 Number and Nature of Tonal Contrasts

Since the prosodic system of Owase is classified as a multipattern system, defining the number of contrastive tonal patterns poses challenges (see § 2.4 of Chapter 1, this volume).

Hirata (2020a) proposes that the prosodic system of Owase is characterized by a lowering accent kernel (see § 2.3 of Chapter 1, this volume) and three registers (see § 2.3.2 of Chapter 1, this volume). Such a system theoretically predicts that the number of contrastive tonal patterns for n-mora words is 3n + 2 (Hirata 2020a).¹ However, as Hirata (2020a) also notes, this prediction is not borne out due to numerous gaps in the proposed framework. For example, three-mora words exhibit only five contrastive patterns.

Based on Hirata (2020a), this chapter posits that Owase has a lowering accent kernel and three registers: α, β, and γ. The characteristics of the accent kernel and registers are elaborated upon in the following discussion. In the subsequent analysis, the tonal patterns of Owase are denoted as Types α0, α1, α2, β2, and γ3, for example. In these designations, the Greek letter indicates the register, and the accompanying number specifies the position of the accent kernel. The numbers “1”, “2”, “3”, and so on denote the first, second, and third morae, respectively, while “0” indicates the absence of an accent. (Throughout this chapter, the terms accent and accent kernel are used interchangeably.)

The tonal patterns of two- to four-mora nouns produced in isolation in Owase (Hirata 2020a) are illustrated in Table 5.2. The diacritics “⸢” and “⸣” indicate a pitch rise and fall, respectively, between morae, whereas “⸣⸣” denotes a pitch fall within a mora (see § 2.2 of Chapter 1, this volume). Owase has no monomoraic lexical words. Consequently, in monosyllabic words with a monophthong in their syllable nucleus, the vowel is lengthened, resulting in bimoraic words such as Type α0 ka⸢a ‘mosquito’, Type α1 ⸢ha⸣a ‘tooth’, and Type γ0 ⸢tee ‘hand’.

Table 5.2

Tonal patterns of Owase

		Unaccented	Initial-accented	Second-accented	Third-accented
2-μ	Register α	Type α0	Type α1
		u⸢si ‘cattle’	⸢i⸣si ‘stone’
	Register γ	Type γ0
		⸢ito ‘thread’
3-μ	Register α	Type α0	Type α1	Type α2	Type α3
		saka⸢na ‘fish’	⸢i⸣noci ‘life’	o⸢to⸣ko ‘man’	asi⸢ta ‘tomorrow’
	Register γ				Type γ3
					⸢kuzira ‘whale’
4-μ	Register α	Type α0	Type α1	Type α2	Type α3
		tomoda⸢ci ‘friend’	⸢ka⸣puseru ‘capsule’	e⸢ha⸣gaki ‘picture postcard’	kami⸢so⸣ri ‘razor’
	Register β			Type β2	Type β3
				⸢mezi⸣rusi ‘mark’	i⸢cizi⸣ku ‘fig’
	Register γ				Type γ3
					⸢murasa⸣ki ‘purple’

Hirata’s (2020a) evidence for a tonal contrast based on the location of the accent kernel in Owase can be observed, for example, in (3), which presents four-mora nouns with Register α.

(3)	a.	Type α1	⸢ka⸣puseru ‘capsule’
	b.	Type α2	e⸢ha⸣gaki ‘picture postcard’
	c.	Type α3	kami⸢so⸣ri ‘razor’

As discussed later, Register α is characterized by a tonal pattern with only one high-pitched mora within a domain. The high pitch is distributed to the first, second, and third morae in (3a), (3b), and (3c), respectively. Hirata (2020a) interprets this high pitch as the manifestation of a lowering kernel located on the first, second, and third morae, respectively, which are presented in bold in (3).

Evidence for a tonal contrast based on features other than accent kernel locations—specifically, contrasts that Hirata (2020a) attributes to the registers—is provided, for example, by the three four-mora words in (4).

(4)	a.	Type α3	kami⸢so⸣ri
	b.	Type β3	i⸢cizi⸣ku
	b.	Type γ3	⸢murasa⸣ki

The words in (4) equally have a pitch fall from the third to the fourth morae, which Hirata (2020a) interprets as the realization of a lowering kernel located on the third mora. Therefore, following Hirata’s (2020a) framework, the four-mora words in (4) share the common feature of having an accent kernel on the third mora, which is presented in bold. However, despite the identical location of the accent kernel, the three words in (4) are distinguished by the pitch patterns of the portion preceding the accented mora. Hirata (2020a) refers to such patterns contributing to the paradigmatic contrasts in this portion as Registers α, β, and γ, respectively. Register α is characterized by one high-pitched mora and is thus also called the “one-high register”. Register β features both the accented mora and its immediately preceding mora as high-pitched, hence the term “two-high register”. Register γ exhibits a level pitch from the beginning of the word to the accented mora and is thus referred to as the “level register”.

In patterns where the accent kernel is on the word-final mora, a pitch fall is not realized when the word is produced in isolation, thereby leading to neutralization between, for example, Type α0 and Type α3 in three-mora words (Table 5.2). However, a pitch fall is realized when a word with an accent on the word-final mora is accompanied by a particle such as the topic marker =wa, as shown in (5).

(5)	a.	Type α3	asi⸢ta ‘tomorrow’	asi⸢ta⸣=wa ‘tomorrow=top’
	b.	Type α0	saka⸢na ‘fish’	sakana=⸢wa ‘fish=top’
	c.	Type γ3	⸢kuzira ‘whale’	⸢kuzira⸣=wa ‘whale=top’

As illustrated in (5b), Type α0 words, that is, unaccented words with Register α, exhibit a high pitch on the final mora of the bunsetsu—a unit comprising a lexical word and any following particles (see § 2.6 of Chapter 1, this volume). Consequently, the high pitch observed on the final mora of the word produced in isolation shifts rightward to that of the bunsetsu when one or more particles are attached to the word. This high pitch consistently appears when the word or bunsetsu is produced in isolation. Notably, however, when followed by another bunsetsu, the high pitch is deleted depending on the tonal class of the subsequent words. This alternation in Type α0 is discussed in § 2.3.

Register γ words also exhibit an alternation. When produced in isolation, these words start with a high pitch, which continues until the accented mora if an accent kernel is present or extends to the end of the word if none is present. However, when preceded by another word, Register γ words may start with a low pitch, depending on the tonal class of the preceding word. As demonstrated in (6), which presents Type α1, Type β2, and Type γ3 words preceded by the Type α0 word kono ‘this’, both Type α1 and Type β2 words, despite initially having a high pitch on the surface, do not exhibit this alternation. By contrast, Register γ words do, thus indicating that this alternation is a unique characteristic of Register γ.

(6)	a.	Type α1	⸢ka⸣puseru ‘capsule’	→	kono ⸢ka⸣puseru ‘this capsule’
	b.	Type β2	⸢mezi⸣rusi ‘mark’	→	kono ⸢mezi⸣rusi ‘this mark’
	c.	Type γ3	⸢murasa⸣ki ‘purple’	→	ko⸢no murasa⸣ki ‘this purple color’
					~ ko⸢no⸣ murasa⸣ki

This alternation in Register γ words will be elaborated upon in § 3.1.

As previously noted, Hirata’s (2020a) framework has numerous gaps. For instance, Register β words have not been identified for words with three or fewer morae, nor have Type γ1 and Type γ2 words. However, as illustrated by the example in (4), the presence of tonal features that are not predictable solely from the location of the accent kernel, along with the paradigmatic contrasts among the three patterns they create, supports the theory of three distinct registers in Owase.

2.3 Domain of Lexical Tone Assignment

The domain of tone assignment is primarily the lexical word, although the tone of Type α0 is assigned to the bunsetsu—a unit comprising a lexical word and any following particles (see § 2.6 of Chapter 1, this volume). Additionally, a few particles possess lexical tones; with these particles, the domain for Type α0 becomes smaller than the bunsetsu.

As presented in (5b), Type α0 words exhibit a high pitch on the final mora of the bunsetsu. Consequently, the high pitch that appears on the final mora of a noun produced in isolation is shifted to the particle when the noun accompanies one. As demonstrated in (7), this shift does not occur for the high pitch of accented words, which remains within the noun, thereby indicating that the domain of the accent kernel is lexical words.

(7)	a.	Type α0	saka⸢na ‘fish’	sakana=⸢wa ‘fish=top’
	b.	Type α1	⸢i⸣noci ‘life’	⸢i⸣noci=wa ‘life=top’
	c.	Type α2	o⸢to⸣ko ‘man’	o⸢to⸣ko=wa ‘man=top’
	d.	Type α3	asi⸢ta ‘tomorrow’	asi⸢ta⸣=wa ‘tomorrow=top’

As noted in § 2.2, the high pitch of Type α0 alternates when followed by another bunsetsu. Specifically, the high pitch at the end of the bunsetsu is deleted except when followed by a bunsetsu with Register γ words (8). In these cases, Register γ words in the second bunsetsu begin with a low pitch instead of a high pitch due to the alternation of Register γ, as discussed in § 3.1.

(8)	a.	Type α0	saka⸢na ‘fish’	Type α0	ara⸢u ‘to wash’
		→ sakana ara⸢u ‘(I) wash the fish.’
	b.	Type α0	saka⸢na ‘fish’	Type α1	⸢i⸣tamu ‘to be damaged’
		→ sakana ⸢i⸣tamu ‘Fish are damaged.’
	c.	Type α0	kore=⸢wa ‘this=top’	Type α2	o⸢to⸣ko ‘man’
		→ kore=wa o⸢to⸣ko ‘This is a man.’
	d.	Type α0	kore=⸢wa ‘this=top’	Type α3	asi⸢ta ‘tomorrow’
		→ kore=wa asi⸢ta ‘This is tomorrow.’
	e.	Type α0	kore=⸢wa ‘this=top’	Type β2	⸢mezi⸣rusi ‘mark’
		→ kore=wa ⸢mezi⸣rusi ‘This is a mark.’
	f.	Type α0	kore=⸢wa ‘this=top’	Type β3	i⸢tizi⸣ku ‘fig’
		→ kore=wa i⸢tizi⸣ku ‘This is a fig.’
	g.	Type α0	kuru⸢ma ‘car’	Type γ0	⸢hairu ‘to enter’
		→ kuru⸢ma⸣ hairu ‘The car enters.’
	h.	Type α0	kore=⸢wa ‘this=top’	Type γ3	⸢dango ‘dumpling’
		→ kore=⸢wa⸣ dango ‘This is a dumpling.’

This could suggest that the high pitch of Type α0 is a phrase-level tone with a domain larger than the bunsetsu. However, this phenomenon here is interpreted as the deletion of a word-level tone, assigned to a bunsetsu, when adjacent to a bunsetsu of a particular class. In other words, a word-level tone is deleted by a phrase-level rule whose domain includes multiple bunsetsus.

The high pitch of Type α0 exhibits a specific behavior when followed by a particle with a lexically specified tone, such as =gurai ‘at least’, which belongs to Type α2. The accent kernel of =gurai is always realized regardless of the tonal class of the preceding noun (9).

(9)	a.	Type α0	saka⸢na ‘fish’	sakana=gu⸢ra⸣i ‘fish=at_least’
	b.	Type α1	⸢a⸣bura ‘oil’	⸢a⸣bura=gu⸢ra⸣i ‘oil=at_least’
	c.	Type α2	o⸢to⸣ko ‘man’	o⸢to⸣ko=gu⸢ra⸣i ‘man=at_least’
	d.	Type α3	asi⸢ta ‘tomorrow’	asi⸢ta⸣=gu⸢ra⸣i ‘tomorrow=at_least’

As presented in (9a), the high pitch of Type α0 is deleted when followed by a particle with a Type α2 tone, which is similar to the behavior observed when followed by another bunsetsu with a Type α2 word. This chapter interprets this as indicating that, when followed by a particle with a lexically specified tone, the right edge of the domain of high pitch of Type α0 aligns with the left edge of the particle. When the register of the particle is other than Register γ, the high pitch of Type α0 is deleted.

2.4 Tone-Bearing Units

Determining the tone-bearing in Owase is challenging. The “counting unit” for measuring phonological length or distance (McCawley 1978; see § 2.7 of Chapter 1, this volume) has yet to be fully investigated. While the mora serves as the unit bearing the accent kernel in Register α words, due to limited data, the corresponding unit in other registers is less certain.

The fact that the mora is the unit bearing the accent kernel in Register α words is demonstrated by the contrast between words with such kernel on the first mora and those with one on the second mora within a heavy syllable (10).

(10)	a.	Type α4	vs.	Type α3
		inza⸢i⸣si ‘Inzai City’		egu⸢za⸣iru ‘EXILE’ (A band’s name)
	b.	Type α2	vs.	Type α1
		to⸢o⸣ri ‘street’		⸢go⸣oru ‘goal’
	c.	Type α2	vs.	Type α1
		o⸢n⸣na ‘woman’		⸢ka⸣nna ‘plane (woodworking tool)’
	d.	Type α4	vs.	Type α3’
		senta⸢k⸣ki ‘washing machine’		suku⸢ra⸣ppu ‘scrap’

In (10a), the second syllable is heavy with a diphthong; in (10b), the first syllable is heavy with a long vowel; in (10c), the first syllable is heavy with a nasal in the coda; and in (10d), the penultimate syllable is heavy with an obstruent in the coda. The examples on the left have an accent kernel on the second mora of the heavy syllable, while those on the right have it on the first mora.

However, a contrast like (10) is not found in Register β and Register γ words. Notably, at least four Register γ words are observed where the accent kernel is placed in the first mora of the heavy syllable, as documented in (11). The word (11c), which has a derogatory meaning, is presented using a cv notation for consonants and short vowels and n for a moraic nasal.

(11)	a.	⸢mome⸣n ‘cotton’	b.	⸢jaro⸣o ‘guy’
	c.	⸢cvcv⸣ncv (meaning omitted)	d.	⸢musi⸣nkee ‘insensibility’

In these four words, the accent kernel is located on the second mora counting from the beginning of the word, thus making them appear to be Type γ2 words. However, given that no Register γ words have been found with the accent kernel on the second mora of a heavy syllable, Register γ words in Owase can be assumed to have a constraint prohibiting the accent kernel from being placed in such position. Since no Type γ2 words are found except for those with a heavy second syllable, while many Type γ3 words exist, the four words in (11) are likely to be Type γ3 in their underlying form. This constraint suggests a shift of the accent kernel to the first mora of a heavy syllable. Consequently, Type γ2 words were not defined in § 2.2 of this chapter. If this interpretation holds, the unit carrying the accent kernel in Register γ words may be considered the syllable rather than the mora.

The question of counting units in Owase remains an area for future research. Hirata (2020a) highlights that the Latin Stress Rule (Kubozono 1996) constitutes part of the loanword accentuation rules in Owase, just as it does in Tokyo Japanese. This rule dictates that an accent kernel is placed on the penultimate syllable if it is heavy; otherwise, on the antepenultimate syllable. Given that the aforementioned rule is sensitive to the weight of the penultimate syllable, some loanword accentuation patterns described in Hirata (2020a) may indicate that the counting unit in Owase is the syllable.

2.5 Directionality

In Owase, determining directionality, which is defined as the direction in which the position of word-level prominence is calculated (see § 2.8 of Chapter 1, this volume), is also challenging. While systematic gaps in the distribution of the accent kernel could hint at directionality, such an explanation is not evident.

Register α has the least restrictive distribution for accent kernels, where all morae except the word-final mora can carry an accent kernel; however, three-mora words are an exception where all morae can carry an accent kernel. Register β and Register γ words significantly restrict the distribution of accent kernels, but their implications for directionality are unclear.

However, if the Latin Stress Rule discussed in § 2.4 applies to loanword accentuation in Owase, the right boundary of the words is used to determine the prominence location, as this rule places the accent kernel on the penultimate or antepenultimate syllable. This may suggest a right-to-left directionality in loanword accentuation rules.

2.6 Obligatoriness and Culminativity

Obligatoriness requires at least one prominence within the word domain, whereas culminativity requires at most one prominence within the word domain (Hyman 2006; see § 2.5 of Chapter 1, this volume). The prosodic system of Owase meets the culminativity requirement but not the obligatoriness requirement.

As discussed in § 2.2, Owase does not exhibit tonal patterns with two or more metrical prominences within the word domain, thereby satisfying the former requirement. In other words, once a pitch fall occurs, another pitch fall does not occur in the same bunsetsu. In addition, Owase includes Type γ0, which lacks any metrical prominence, and thus fails to meet the obligatoriness requirement.

Although accent kernels provide prominence to words, not all tonal patterns that lack an accent kernel are without prominence. Tonal patterns lacking an accent kernel include Type α0 and Type γ0, with β0 being absent. However, Type α0 always has a high pitch at the end of the bunsetsu, which marks prominence within the word domain. This high pitch is not present in the underlying form but is introduced by a rule whose domain of application is the bunsetsu. In the framework of this volume, tones with a bunsetsu as their domain are considered part of the lexical representation. As discussed in § 2.3, this high pitch is not realized when a specific tonal pattern follows; however, this absence is interpreted as the deletion of a word-level tone by a phrase-level rule. In contrast, Type γ0, such as ⸢tee ‘hand’ and ⸢ito ‘thread’, maintains a high pitch from the beginning to the end of the word, lacking significant pitch changes. Similarly, the alternation form of Type γ0 discussed in § 3.1 exhibits a continuous low pitch throughout the word, thereby failing to meet the obligatoriness requirement.

2.7 Prosodic Compound Rules

The prosodic compound rules in Owase are right-dominant: the tonal pattern of the compounds is determined by the lexical information of the second member, with the first playing no role.

The tonal pattern of a compound is determined by its second member, referring to its lexical rather than prosodic information. For instance, although eki ‘station’, sjugi ‘principle’, and iro ‘color’ all belong to Type α1, the resulting tonal patterns of their compounds vary. Specifically, while all compounds exhibit Register α, their accentuation differs: eki ‘station’ leads to an antepenultimate accent (12a), sjugi ‘principle’ results in a penultimate accent (12b), and iro ‘color’ produces an unaccented pattern (12c). Certain second members cause compound words to adopt Register β. For example, Type α0 ha⸢ko ‘box’ results in a compound with Register β and a penultimate accent (12d).

(12)	a.	Type α1 ⸢e⸣ki ‘station’:
		Type γ3 ⸢owase ‘Owase’+⸢e⸣ki
		→ Type α3 owa⸢se⸣eki ‘Owase station’
		Type α0 jokoha⸢ma ‘Yokoyama’+⸢e⸣ki
		→ Type α4 jokoha⸢ma⸣eki ‘Yokohama station’
		Type α0 sappo⸢ro ‘Sapporo’+⸢e⸣ki
		→ Type α4 sappo⸢ro⸣eki ‘Sapporo station’
	b.	Type α1 ⸢sju⸣gi ‘principle’:
		Type α1 ⸢si⸣hon ‘capital’+⸢sju⸣gi
		→ Type α4 sihon⸢sju⸣gi ‘capitalism’
		Type α0 kyoosa⸢n ‘coproduction’+⸢sju⸣gi
		→ Type α5 kjoosan⸢sju⸣gi ‘communism’
		Type α0 kanemo⸢o⸣ke ‘money-making’+⸢sju⸣gi
		→ Type α6 kanemooke⸢sju⸣gi ‘profit-oriented mindset’
	c.	Type α1 ⸢i⸣ro ‘color’:
		Type α0 nezu⸢mi ‘mouse’+⸢i⸣ro
		→ Type α0 nezumii⸢ro ‘gray color’
		Type γ3 ⸢daida⸣i ‘bitter orange’+⸢i⸣ro
		→ Type α0 daidaii⸢ro ‘orange color’
		Type β3 a⸢kamu⸣rasaki ‘reddish purple’+⸢i⸣ro
		→ Type α0 akamurasakii⸢ro ‘reddish purple’
	d.	Type α0 ha⸢ko ‘box’:
		Type α1 ku⸢su⸣ri ‘medicine’+ha⸢ko
		→ Type β4 kusu⸢riba⸣ko ‘medicine box’
		Type α0 toohjo⸢o ‘vote’+ha⸢ko
		→ Type β5 toohyo⸢oba⸣ko ‘ballot box’
		Type α3 dan⸢bo⸣oru ‘cardboard’+ha⸢ko
		→ Type β6 danboo⸢ruba⸣ko ‘cardboard box’

2.8 Merger and Split of Historic Tonal Classes

The correspondences in tonal classes of one- and two-syllable nouns between Middle Japanese (MJ) (see § 2.10 of Chapter 1, this volume) and Owase are illustrated in Table 5.3.

Table 5.3

Correspondences in tonal classes between MJ and Owase

MJ tonal classes	Owase tonal classes
1.1	Type α0
1.2	Type α1
1.3	Type γ0
2.1	Type α0
2.2	Type α1
2.3
2.4	Type γ0
2.5

While Owase exhibits the merging of Classes 2.2 with 2.3 and Classes 2.4 with 2.5, similar to the Nairin and Churin types, it differs from these in that it retains the distinction between Classes 1.1, 1.2, and 1.3. Uwano (1987: 31) emphasizes that the prosodic system of Owase is a subcategory of the Chuo type, developed through tonal changes from the same system found in the Katsuura variety.

3 Detailed Description of the Prosodic System

3.1 Alternation in Register γ Words

As mentioned in § 2.2, Tonal Register γ words exhibit an alternation of tonal patterns. Register γ words typically have a high-beginning pattern but shift to a low-beginning pattern in certain environments.

This alternation occurs, for example, when the Type α0 adnominal pronoun kono ‘this’ precedes a Register γ word (13). When the pitch fall due to the accent kernel of Register γ words is realized, that is, when the accent kernel is located on the non-final mora of the bunsetsu, alternation is optional: in such cases, Register γ words can be realized in either a high- or low-beginning pattern (13d).

(13)	Pitch patterns of Register γ words preceded by Type α0 ko⸢no ‘this’
	a.	Type γ0 ⸢ame ‘rain’:	ko⸢no⸣ ame ‘this rain’
	b.	Type γ0 ⸢ame=wa ‘rain=top’:	ko⸢no⸣ ame=wa ‘this rain=top’
	c.	Type γ3 ⸢kuzira ‘whale’:	ko⸢no⸣ kuzira ‘this whale’
	d.	Type γ3 ⸢kuzira⸣=wa ‘whale=top’:	ko⸢no⸣ kuzira⸣=wa ‘this whale=
			top
			~ ko⸢no kuzira⸣=wa

The alternation of Register γ words to the low-beginning pattern is not confined to environments where they follow the Type α0 word kono ‘this’. When Register γ words are preceded by a bunsetsu containing a Type α0 word (14a, b) or a Type γ0 word (14c), they also exhibit a low-beginning pattern.

(14)	a.	Type α0 usi=⸢o ‘cow=acc’	Type γ0 ⸢kau ‘to feed’
		→ usi=⸢o⸣ kau ‘(I) feed cows.’
	b.	Type α0 kaze=⸢ga ‘wind=nom’	Type γ0 ⸢hui-ta ‘to blow-pst’
		→ kaze=⸢ga⸣ hui-ta ‘Wind blew.’
	c.	Type γ0 ⸢huru ‘to fall’	Type γ0 ⸢ame ‘rain’
		→ ⸢huru⸣ ame ‘the rain that falls’

There is no alternation when the pitch fall due to the accent kernel is realized on the preceding bunsetsu; in such cases, Register γ words exhibit a high-beginning pattern (15a, c). Notably, even if the preceding bunsetsu contains a lexically accented word, if the accented mora is located at the end of the bunsetsu, this hinders the pitch fall from being realized, Register γ words still undergo the alternation, thus exhibiting a low-beginning pattern (15b, d).

(15)	a.	Type α3 aso⸢ko⸣=ni ‘there=dat’	Type γ0 ⸢kuru ‘to come’
		→ aso⸢ko⸣=ni ⸢kuru ‘(I) come there.’
	b.	Type α3 aso⸢ko ‘there’	Type γ0 ⸢kuru ‘to come’
		→ aso⸢ko⸣ kuru ‘(I) come there.’
	c.	Type γ3 ⸢kabuto⸣=o ‘helmet=acc’	Type γ0 ⸢dasu ‘to let out’
		→ ⸢kabuto⸣=o ⸢dasu ‘(I) put out the helmet’.
	d.	Type γ3 ⸢kabuto ‘helmet’	Type γ0 ⸢dasu ‘to let out’
		→ ⸢kabuto⸣ dasu ‘(I) put out the helmet.’

The above discussion demonstrates that Register γ words exhibit an alternation when the pitch fall due to the accent kernel is not realized in the immediately preceding bunsetsu. In other words, the alternation occurs when the preceding bunsetsu ends with a high pitch. Under these environments, Register γ words display a low-beginning pattern; conversely, if the preceding bunsetsu ends with a low pitch, such words maintain a high-beginning pattern.

4 Phrase-Level Prosody

4.1 Interrogative Intonation

As of May 2024, research regarding interrogative intonation has not been conducted.

4.2 Vocative Intonation

In some varieties like traditional Koshikijima in Kagoshima Prefecture (Kubozono 2018) and Kobayashi (Hirata 2018), a vocative particle is obligatory in every vocative sentence; however, Owase lacks an overt morphological marker for vocative forms, instead relying on intonation to differentiate vocative forms from citation or declarative forms.

According to Hirata’s (2020b) study on the vocative form of personal names, vocative intonation depends on whether the fall due to an accent kernel is realized. For names with an accent kernel in a non-final mora, such as Type α1 ⸢ke⸣nziroo ‘Kenjiro’, Type α2 na⸢o⸣ziroo ‘Naojiro’, and Type α3 kiku⸢ta⸣roo ‘Kikutaro’, vocative intonation is characterized by an expanded pitch range, which results in an accentual fall with a larger excursion—that is, a larger distance between the highest and lowest points of the pitch contour—than in citation or declarative forms. Conversely, for names where a pitch fall is not realized—either unaccented names such as Type α0 haru⸢o ‘Haruo’ or those with an accent kernel on the final mora such as Type γ3 ⸢tanaka ‘Tanaka’—such a fall is added at the end of the word. The precise location of this additional fall depends on the structure of the final syllable. The fall occurs within the final mora if the final syllable is light, as in Type α0 haru⸢o ‘Haruo’ becoming haru⸢o⸣⸣. If the final syllable is heavy, the fall occurs between the first and second morae of that syllable or within the second mora, as in Type α0 kentaro⸢o ‘Kentaro’ becoming kentar⸢o⸣o or kentaro⸢o⸣⸣.

Acknowledgements

This work was supported by JSPS KAKENHI Grant Numbers 14J09547 and 16H06784. We would like to express our heartfelt gratitude to the speakers of Owase for their kind cooperation during the fieldwork. Their contributions were invaluable to this research. Any remaining errors or inaccuracies are, of course, solely the responsibility of the author.

The expected number of tonal patterns should be 3(n+1) assuming no restrictions on the location of the accent kernel within a word, a tonal class lacking an accent kernel (unaccented class), and three registers with no restrictions on their combinations with accentuations exist. However, since one of the three registers, specifically, Register β, is defined as having a high pitch on the accented mora and the mora immediately preceding it, there is theoretically no Register β with an accent kernel on the word-initial mora. Therefore, the number of tonal patterns predicted by theory is 3(n + 1) –1, or 3n + 2 (Hirata 2020a).

References

Hirata, Shu (2018) Kobayashi hōgen no yobikake intonēshon [Vocative intonation in Kobayashi Japanese]. Proceedings of the 157th Meeting of the Linguistic Society of Japan: 438–441.
- Search Google Scholar
- Export Citation
Hirata, Shu (2020a) Mie-ken Owase hōgen no akusento kenkyū [Accentology of the Owase dialect in Mie Prefecture]. Tokyo: Hituzi Syobo.
- Search Google Scholar
- Export Citation
Hirata, Shu (2020b) Mie-ken Owase hōgen no jinmei akusento to yobikake intonēshon [The accentuation of person’s names and vocative intonation in the Owase dialect in Mie Prefecture]. Proceedings of the 161st Meeting of the Linguistic Society of Japan: 368–373.
- Search Google Scholar
- Export Citation
Hyman, Larry M. (2006) Word-prosodic typology. Phonology 23(2): 225–257.
- Search Google Scholar
- Export Citation
Kindaichi, Haruhiko (1959) Kumanonada engan shohōgen no akusento—Kitamuro-gun no bu [The accents of dialects spoken in Kumano-nada coastal area: The part of Kitamuro-gun]. In Nagoya Daigaku Bungaku-bu jisshūnen kinen ronshū [Essays in celebration of the 10th anniversary of School of Letters, Nagoya University], 681–702. Nagoya: Nagoya Daigaku Bungaku-bu.
- Search Google Scholar
- Export Citation
Kindaichi, Haruhiko (1975) Nihon no hōgen: Akusento no hensen to sono jissō [Japanese dialects: The change of accent and its states]. Tokyo: Kyoiku Shuppan.
- Search Google Scholar
- Export Citation
Kubozono, Haruo (1996) Syllable and accent in Japanese: Evidence from loanword accentuation. The Bulletin of the Phonetic Society of Japan 211: 71–82.
- Search Google Scholar
- Export Citation
Kubozono, Haruo (2018) Kagoshima hōgen to koshikijima hōgen no yobikake intonēshon [Vocative intonation in Kagoshima and Koshikijima Japanese].Proceedings of the 157th Meeting of the Linguistic Society of Japan: 432–437.
- Search Google Scholar
- Export Citation
McCawley, James D. (1978) What is a tone language? In Victoria A. Fromkin (ed.) Tone: A linguistic survey, 113–131. New York: Academic Press.
- Search Google Scholar
- Export Citation
Owase City (2024) Nenrei-betsu danjo-betsu jinkō-chō [Age- and gender-specific population survey]. Retrieved 1 April 2004 from https://www.city.owase.lg.jp/cmsfiles/contents/0000004/4231/R6.5.pdf.
- Search Google Scholar
- Export Citation
Owase-shi-shi Hensan Junbikai (ed.) (1969) Owase-shi-shi (gekan) [The history of Owase City (last volume)]. Mie: Owase-shi.
Tojo, Misao (1954) Nihon hōgengaku [Japanese dialectology]. Tokyo: Yoshikawa Kobunkan.
- Search Google Scholar
- Export Citation
Uwano, Zendo (1987) Nihon hondo shohōgen akusento no keifu to bunpu (2) [Genealogical relationships and the geographical distribution of the accents in Mainland Japan (2)]. Nihon Gakushi-in Kiyo 42–41: 15–70.
- Search Google Scholar
- Export Citation
Uwano, Zendo (1989) Nihongo no akusento [Accent in Japanese]. In Miyoko Sugito (ed.) Kōza Nihongo to Nihongo kyōiku 2: Nihongo no onsei on’in jō [Series on Japanese language and Japanese language education 2: Japanese phonetics and phonology volume 1], 178–205. Tokyo: Meiji Shoin.
- Search Google Scholar
- Export Citation
Uwano, Zendo (1999) Classification of Japanese accent systems. In Shigeki Kaji (ed.) Proceedings of symposium cross-linguistic studies of tonal phenomena: Tonogenesis, typology, and related topics, 151–178. Tokyo: ILCAA.
- Search Google Scholar
- Export Citation

Citation Info

Save
Cite
Email this content

Share link with colleague or librarian

You can email a link to this page to a colleague or librarian:
Email this content
or copy the link directly:

https://brill.edhh.ma/display/book/9789004755789/BP000005.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Collapse
Expand

Word-Prosodic Systems of Japonic Languages

Series: Endangered and Lesser-Studied Languages and Dialects, Volume: 5

E-Book ISBN:: 9789004755789

Publisher:: Brill

Print Publication Date:: 24 Feb 2026

Subjects
- Asian Studies
  - Japan
- Languages and Linguistics
  - Language Documentation & Description (Grammars)
  - Phonetics & Phonology

Front Matter

Back Matter

ProCite

RefWorks

Reference Manager

BibTeX

Zotero

EndNote

Metrics

	All Time	Past 365 days	Past 30 Days
Abstract Views	0	0	0
Full Text Views	20	20	1
PDF Views & Downloads	27	27	1

African Studies	Education	Media Studies
American Studies	History	Middle East and Islamic Studies
Ancient Near East and Egypt	Human Rights and Humanitarian Law	Musicology
Art History	International Law	Philosophy
Asian Studies	International Relations	Religious Studies
Biblical Studies	Jewish Studies	Slavic and Eurasian Studies
Biology	Languages and Linguistics	Social Sciences
Book History and Cartography	Life Sciences	Theology and World Christianity
Classical Studies	Literature and Cultural Studies

Subjects

Authors

Open Access

Product Information

Company

Contact

Chapter 5 Owase

1 Introduction

1.1 Location

1.2 Speakers

1.3 Classification

1.4 Previous Studies

1.5 Data

1.6 Overview of the Segmental Phonology

Table 5.1

2 Word-Prosodic System

2.1 Classification Based on Uwano (1989, 1999)

2.2 Number and Nature of Tonal Contrasts

Table 5.2

2.3 Domain of Lexical Tone Assignment

2.4 Tone-Bearing Units

2.5 Directionality

2.6 Obligatoriness and Culminativity

2.7 Prosodic Compound Rules

2.8 Merger and Split of Historic Tonal Classes

Table 5.3

3 Detailed Description of the Prosodic System

3.1 Alternation in Register γ Words

4 Phrase-Level Prosody

4.1 Interrogative Intonation

4.2 Vocative Intonation

Acknowledgements

References

Citation Info

Share link with colleague or librarian

Word-Prosodic Systems of Japonic Languages

Table of Contents

References

Metrics

Metrics