You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
Languages
Download PDF
  • Article
  • Open Access

24 July 2023

‘Good’ Is ‘Possible’: A Case Study of the Modal Uses of ‘Good’ in Shaoxing

and
1
Institute of Corpus Studies and Applications, Shanghai International Studies University, Shanghai 201613, China
2
Institute of Linguistics, Shanghai International Studies University, Shanghai 201613, China
*
Author to whom correspondence should be addressed.
Languages2023, 8(3), 177;https://doi.org/10.3390/languages8030177
This article belongs to the Special Issue Typology of Chinese Languages: One Name, Many Languages
Version Notes
Order Reprints

Abstract

This paper sets out to investigate the modal uses of the lexeme hɒ3 ‘good’ in the Jidong Shaoxing variety of Wu and to reconstruct its grammaticalization pathway. Modal meanings of hɒ3 include circumstantial possibility, deontic possibility and necessity, and epistemic possibility. These meanings can be summarized as ‘can’, ‘may’, and ‘should’, respectively. The modal meanings of hɒ3 are derived from its meaning of ‘fit to’ rather than ‘good’. We propose here that hɒ3 first extended to express circumstantial possibility, and then further extended to denote deontic modality and participant-internal possibility in two separate directions: (i) circumstantial possibility > deontic modality, and (ii) circumstantial possibility > participant-internal possibility. The epistemic use of hɒ3 is proposed as the final stage of the lexeme’s modal extension.

1. Introduction

Modality is a semantic domain concerning how languages code possibility and necessity. Languages may adopt different strategies to express modality, such as auxiliary verbs, morphological devices of mood, modal affixes, lexical means, modal adverbs and adjectives, modal tags, modal particles, and modal case (). While Sinitic languages are well known for lacking morphological mood, modality is expressed by auxiliary verbs, adverbs, and potential constructions, as well as particles. Among these devices, auxiliary verbs are the most common means of expressing possibility and necessity in Sinitic languages ().
Modal auxiliaries in Standard Mandarin can be divided into necessity and possibility modal verbs. Possibility modal verbs include hui⁴ 会 ‘can < know, comprehend’, ke3yi3 可以 ‘can, may < fit so as to’, and neng2 能 ‘can < capable’, while necessity modal verbs include dei3 得 ‘must < obtain’, gai1 该 ‘should < owe’, and yao⁴ 要 ‘need, must’. According to a survey by (), these are common modal verbs in Sinitic languages. Different Sinitic languages or dialects have developed various lexical sources into modal auxiliaries, for example, kuan⁴⁴ 管 ‘can < manage, supply’ in Fangcheng Central Plains Mandarin (), iəɯ22fɐʔ⁵ 有法 ‘can < have methods’ and ua⁵1 话 ‘can < say’ in Liujiang Jinde Hakka (), tɕʰieʔ⁵⁻33teʔ⁵⁻33loʔ2⁻⁵3 吃得落 ‘can < can eat’ in Shaoxing Wu (), and hao 好 ‘can < good’ in Jieyang Min () and Hakka (; ; ). In fact, hao ‘good’ has been adapted as a modal auxiliary in a contiguous area around the Yangtze River Delta. This phenomenon might be considered as a micro-areal phenomenon.
This paper will offer a case study on the modal uses of hɒ3 好 ‘good’ in the Jidong variety of Shaoxing Wu. First, a detailed synchronic description of the modal uses of hɒ3 ‘good’ in Jidong Shaoxing is given. This is followed by a diachronic reconstruction of the pathway of grammaticalization from ‘good’ to ‘can, may, should’ of hɒ3. Observation of hɒ3′s distribution indicates that its modal uses are derived from the auxiliary use of its meaning ‘fit to’ rather than directly from its meaning ‘good’. We find that in a grammatical sentence containing hɒ3, the situation expressed by the matrix VP is often enabled by an external or an internal condition. It is from this circumstantial possibility that the participant-internal possibility of hɒ3 is derived. Our finding contributes some new evidence in support of ’s () findings that circumstantial possibility may develop into participant-internal possibility and deontic possibility may extend to deontic necessity. The bidirectional developments ‘circumstantial ↔ participant-internal’ and ‘deontic possibility ↔ deontic necessity’ hold (). We also propose that the epistemic use of hɒ3 is derived from the uses of both circumstantial and participant-internal possibility. The semantic connection between the deontic and epistemic uses of hɒ3 is not obvious in Jidong Shaoxing, unlike the well-known deontic–epistemic polysemy of English must ().
Section 2 discusses the modality types distinguished in this paper. Section 3 presents background information and basic linguistic features of the target language. Section 4 introduces the polysemy of the lexeme hɒ3. Section 5 sets out to illustrate the modal uses of hɒ3. Section 6 aims to reconstruct the grammaticalization pathway of hɒ3 and offers a comparison with the diachrony of hao3 ‘good’ in the history of Chinese. Section 7 is a general conclusion.

2. Modality Types and Terminology

The domain of modality can be organized in several ways. Deontic, dynamic, and epistemic are the most broadly accepted concepts. () contrasts root modality, which comprises both dynamic and deontic, and epistemic modality. () oppose epistemic and non-epistemic (later revised as situational ()), with the latter subdivided into participant-internal and participant-external. () proposes an opposition of event modality (deontic and dynamic) and propositional modality (epistemic and evidential). () suggests volitive modal categories opposed to non-volitive modal categories. Given that a reorganization of our understandings of modality is not an aim of the current paper, here we adopt subcategories of modality solely on the basis of their relevance in describing the polysemy of hɒ3. Relevant modal concepts include:
  • Circumstantial possibility
  • Circumstantial possibility refers to a proposition enabled by certain external circumstances (). It can also be interpreted as a possibility allowed by external conditions, as in (1).
(1)
You can get to University City by taking subway line 9.
  • Permission and weak obligation
  • Both permission and obligation belong to deontic modality, also known as deontic possibility and necessity. Deontic modality refers to possibility or necessity determined by certain social norms, expectations, or a speaker’s desire or command (see ; ). Permission denotes that a participant is allowed to complete an action (), as in (2), while weak obligation means that a participant was advised to complete an action (). In English, weak obligation is usually expressed by should and ought, as in (3).
(2)
You may leave now.
(3)
You should leave now.
  • Participant-internal possibility
  • Participant-internal possibility refers to “a kind of possibility […] internal to a participant engaged in the state of affairs” and it covers dynamic possibility, ability, and capacity (), as in (4).
(4)
She can lift that heavy stone.
  • Epistemic possibility
  • Epistemic possibility refers to a speaker’s degree of certainty about a proposition (); see (5). Epistemic modality is generally relevant to knowledge, belief, and related notions.
(5)
The bus may be late (due to the snow).

3. Jidong Shaoxing and Some Basic Features

3.1. Variety under Investigation

Shaoxing 绍兴话 is a Northern Wu dialect of Sinitic belonging to the Linshao subdivision 临绍小片 of the Taihu division 太湖片 (). The variety under investigation, Jidong 稽东, is spoken in Jidong Town in the southern suburb of Shaoxing Prefecture and is classified as belonging to the Southern Suburb variety of Shaoxing ().
The data presented in this paper were collected with four native speakers from 2022 to 2023. Our consultants represented three generations within the same family. They were Mr. Huang Tangfu 黄汤富 (born in 1955), Mrs. Huang Xingqin 黄杏琴 (born in 1957), Mr. Huang Yongjiang 黄永江 (born in 1977), and the second author Mr. Huang Xiao 黄骁 (born in 2002). Mr. Huang Yongjiang speaks an innovative variety of Jidong Shaoxing, while the other three speak a conservative variety.
Our corpus included both spontaneous and elicited data. Spontaneous data illustrated in this paper were either extracted from a corpus of six hours of audio material or taken from unrecorded daily conversations which were not part of the corpus. Elicited data comprised about 130 sentences. Elicited data presented in this paper will be indicated as ‘elicitation’, while the data with unmarked sources were either from our corpus or the unrecorded daily conversations. It is worth mentioning that when performing elicitation we did not simply ask for translations from Standard Mandarin to Jidong Shaoxing. Instead, taking semantic nuances of modals into consideration, we provided different contexts as stimuli for our consultants.

3.2. Basic Features

Shaoxing possesses eight tones: Tone 1/33/, Tone 2/13/, Tone 3/435/, Tone 4/213/, Tone 5/52/, Tone 6/231/, Tone 7/4/, and Tone 8/23/. Tones 7 and 8 are two checked tones. Any syllable bearing Tone 7 or 8 ends with a glottal stop/ʔ/. We will hereafter refer to the tones with a superscript of the number by which the tone is named. Note that tone sandhi is not represented in our transcription. Shaoxing is basically a VO language, especially in dependent clauses and when an object appears with a complex modifier, as shown in (6). Wu languages are characterized by topic prominence () and constructions with topicalized objects are commonly observed in Shaoxing, as in (7). A topicalized object can occupy either a sentence initial position (7) or can follow the subject and precede the verb. As an analytic language, Shaoxing grammatical relations are realized by prepositions or word order. See example (6) for an example of a dative argument marked by the dative preposition pə⁷ ‘to’.
(6)我拨侬讲讲□前头做生活种事体。
ŋo⁴pə⁷noŋ⁴kɒŋ3kɒŋ3ŋa⁴ʑjᴇ̃2də2tso⁵saŋ1ɦwo⁸tsoŋ3zɿ⁶tʰi3.
1sgdat2sgtell.dlm1plbeforetimedolifeclfthing
‘I’ll tell you something about how (I) made a living before.’
(7)榧子摘落来要囥得好。
fi3tsɿ3tsa⁷-lə⁸lᴇ2ʔjɒ⁵kʰɒŋ⁵-tə⁷-hɒ3.
Torreyapick-fall.comeneedstore-vcomp-good
‘After picking the Torreya nuts, (they) must be well stored.’
() provides a detailed description of the Shaoxing grammatical system, though it focuses on the Keqiao 柯桥 variety, i.e., the Western Suburb variety, which is slightly different from the variety under investigation here.

3.3. Modal Auxiliaries and Potential Constructions

We summarize the distributional features of auxiliary verbs in Jidong Shaoxing following () and (). In Jidong Shaoxing, auxiliary verbs form a closed class of words. They are free elements taking verbal or clausal complements in the form of [AUX VP] which feature reduced verbal behaviors. Auxiliary verbs in Jidong Shaoxing cannot take any aspectual marking as verbs do.
In an auxiliary construction, the strategy V (NEG) V is applied to the auxiliary verb to form a polar question. An auxiliary verb can stand alone to answer a polar question. See the example below for a polar question formed with the auxiliary ɦwᴇ⁶ 会 ‘can’ and a positive response.
(8)侬游水会会游{个□}？会{个□}。
noŋ⁴ljɤ2sɿ3ɦwᴇ⁶ɦwᴇ⁶ljɤ2ɡɒ?ɦwᴇ⁶ɡo⁸.
2sgswimcancanswimaff.prtcanaff.prt
‘Can you swim? (Yes, I) can.’ (Elicitation)
Some common Sinitic auxiliary verbs are used in Jidong Shaoxing. ɦwᴇ⁶ ‘can < know’ is used to express possibility, as in (8); ʔjɒ⁵ 要 ‘must, should < need’ is used to code necessity; and ɕjaŋ3 想 ‘want’ and kʰiŋ3 肯 ‘be willing’ are used to express willingness. While neng2 and ke3yi3 are used to denote possibility and permission in Standard Mandarin, these meanings are expressed by potential constructions and hɒ3 ‘can < good’ in Jidong Shaoxing.1 For this, see Table 1 and also ().
Table 1. Modal auxiliary verbs in Jidong Shaoxing.
Sinitic languages often adopt potential constructions to express possibility in the form of [V-POT-COMP], with [V-NEG-COMP] as the negated form (). In the negated form, the potential marker and the negator are considered as infixes (; ). In Jidong Shaoxing, potential constructions, V-tə⁷-COMP, are also commonly used to express possibility and are relevant to the modal uses of hɒ3 in Jidong Shaoxing. When denoting ‘cannot’, it is a negated potential construction that is used rather than the negated hɒ3 ‘can’, fə⁸ hɒ3 ‘not good’. The most generalized potential construction is V-tə⁷-lᴇ2 ‘V-pot-come’, with V-və⁸-lᴇ2 ‘V-neg-come’ as the negated form. Both (9a) and (9b) can denote ‘cannot run’, with dɒ2-və⁸-lᴇ2 ‘run-neg-come’ having an additional meaning of prohibition.
(9)a.逃不动b.逃不来
dɒ2-və⁸-doŋ⁴ dɒ2-və⁸-lᴇ2
run-neg-move run-neg-come
‘cannot run’ (Elicitation) ‘cannot run’ (Elicitation)
See Section 5 for more details.

4. Polysemy of hɒ3 in Jidong Shaoxing

In Jidong Shaoxing, the lexeme hɒ3 好 is a polysemous and multi-functional word. hɒ3 can serve as an adjective/adjectival verb ‘good, fitting, ready, done, ok’, as an auxiliary verb ‘fit to, easy to, can, may, should’, as a resultative verb complement denoting a completed action, as an adverb ‘quite’, and as a complementizer introducing a purposive clause. The mentioned uses of hɒ3 in Jidong Shaoxing are also attested in Standard Mandarin (), except that hɒ3 in Jidong Shaoxing possesses more modal uses than hao3 does in Standard Mandarin.
As an adjective, the basic meaning of hɒ3 is ‘good’. Adjectives in Jidong Shaoxing function like intransitive verbs in a way, a common feature for East and Southeast Asian languages. No copula is needed when forming an adjectival predicate, and some adjectives can bear aspectual marking as intransitives do. The two features that distinguish adjectives from intransitive verbs are that (i) adjectives can modify nouns directly and (ii) adverbs can be derived from them. Examples (10) and (11) illustrate the attributive and the predicative uses of hɒ3 ‘good’, respectively. Example (12) shows its derivational use as an adverb, achieved via reduplication and the use of the adverbializer tɕjɒ⁵.
(10)破剪刀覅用哉，用个把好剪刀。
pʰa⁵tɕjᴇ̃3tɒ¹fjɒ3ɦjoŋ⁶dzᴇ,ɦjoŋ⁶ɡə⁸po3hɒ3tɕjᴇ̃3tɒ¹.
brokenscissorneg.impusecrsuses_proxclfgoodscissor
‘Don’t use the broken scissors. Use this good pair.’ (Elicitation)
(11)□个香榧是好{个□}。
ŋa⁴ɡə⁸ɕiaŋ¹fi3zᴇ⁴hɒ3ɡo⁸.
1plpossTorreyaindeedgoodaff.prt
‘Our Torreya nuts are indeed good.’
(12)米好好□来□，侬弄渠作啥？
mi⁴hɒ3hɒ3tɕjɒ⁵lᴇ2doŋ,noŋ⁴loŋ⁶ɦi⁴tsə⁵so⁵?
ricewelladvbe.at 2sgdo 3sgdowhat
‘The rice was fine there. How come you messed it up?’
The meaning ‘good’ can imply the meaning ‘suitable’. In (13), hɒ3 denotes ‘suitable’ or ‘better’.
(13)渠脸盘大，还是留长头发好。
ɦi⁴ljᴇ̃⁴bə̃2do⁶,ɦwæ2zə⁸ljə2dzaŋ2də2fa⁷hɒ3.
3sgfacebigstillkeeplonghairgood
‘She has a big face and long hair suits her.’
In (14), marked by the currently relevant state marker dzᴇ, hɒ3 denotes ‘ready, done’. hɒ3 in (15) serves as a resultative complement turning the verb tɕʰiə⁷ ‘eat’ into a telic verb phrase expressing a completed action.
(14)饭好哉。
væ⁶hɒ3dzᴇ.
mealgoodcrs
‘The meal is ready.’
(15)饭吃好哉。
væ⁶tɕʰjə⁷-hɒ3dzᴇ.
mealeat-goodcrs
‘(I) finished (my) meal.’
As an auxiliary, hɒ3 takes verbal or clausal complements and can denote ‘fit to, be easy to, can, may, should’. hɒ3 in (16) denotes ‘be easy to’, while (17) is ambiguous and can be interpreted as either ‘fit to’ or as ‘can’. See Section 5 for more examples of the modal uses of hɒ3.
(16)个道题□坐好做□。
ɡə⁸dɒ⁴di2mə⁸zo⁴hɒ3tso⁵jə⁸.
s_proxclfquestiontopsit/extremelygooddoprt
‘This problem is extremely easy to solve.’
(17)雄花粉埲拢过个半个月好打药水□。
ɦjoŋ2hwo1fɘŋ1boŋ⁴-loŋ⁴ku⁵ɡə⁸pə̃⁵ɡə⁸ɦjo⁸hɒ3
male.flower.pollenshake-gatherpassclfhalfclfmonthgood
taŋ3ɦja⁸sɿ3ɦa.
beatpesticideprt
‘A half month after (female Torreya trees) are pollinated manually (with) male pollen,
i. it’s the right time for spraying pesticide.’
ii. one can spray pesticide.’
In terms of syntactic position, the auxiliary hɒ3 might sometimes be considered as an adverb. This is especially true in an example like (16) where the meaning ‘be easy to’ could be interpreted as ‘easily’ modifying the verb tso⁵ ‘do’. However, auxiliaries and adverbs are characterized by different syntactic behaviors. As mentioned in Section 3.3, when transforming a declarative sentence with an auxiliary into a polar question, the strategy V (NEG) V is applied to the auxiliary instead of to the main verb, as in (18).
(18)个道题好（不）好做□？
ɡə⁸dɒ⁴di2hɒ3(və⁸)hɒ3tso⁵ɦa?
s_proxclfquestiongoodneggooddoprt
‘Is this problem easy to solve?’ (Elicitation)
Compare (19) with (16) and (18). The adverb ŋᴇ2 ‘very’ in (19a) occupies the same preverbal position as hɒ3 does in (16). To transform (19a) into a polar question, one must apply the strategy V (NEG) V to the predicative adjective næ2 ‘difficult’, as in (19b). Yet, this strategy can never be used with an adverb. The form ŋᴇ2 (və⁸) ŋᴇ2 ‘lit: very not very’ in (19c) is ungrammatical.
(19)a.个道题呆难□。
ɡə⁸dɒ⁴di2ŋᴇ2næ2da.
s_proxclfquestionverydifficultprt
‘This problem is very difficult.’ (Elicitation)
b.个道题难（不）难□？
ɡə⁸dɒ⁴di2næ2(və⁸)næ2ɦa?
s_proxclfquestiondifficultnegdifficultprt
‘Is this problem difficult?’ (Elicitation)
c.*个道题呆（不）呆难□？
*ɡə⁸dɒ⁴di2ŋᴇ2(və⁸)ŋᴇ2næ2ɦa?
s_proxclfquestionverynegverydifficultprt
Attempted: ‘Is this problem very difficult?’ (Elicitation)
Beyond its uses as an auxiliary, hɒ3 also has more grammaticalized uses. Namely, hɒ3 is used as a complementizer to introduce a purposive clause equivalent to the English counterpart ‘so as to’. This function may be closely related to the auxiliary use of hɒ3. In the example below, the first clause ‘take away the quilt’ is uttered in a context in which this action will allow the speaker to put a sheet on the bed. The two clauses are linked by hɒ3.
(20)棉被驮还，好让我铺床单方便些。
mjᴇ̃2bi⁴do2-ɦwæ2,hɒ3ȵjaŋ⁶ŋo⁴pʰu1dzɒŋ2tæ1fɒŋ1bjᴇ̃⁶sə⁷.
quilttake-awaygoodlet1sglaysheetconvenienta.bit
‘Take away the quilt so as to let me easily put the sheet on.’ (Elicitation)
Finally, like Standard Mandarin, hɒ3 in Jidong Shaoxing can serve as an adverb meaning ‘quite’, i.e., a degree intensifier. This use of hɒ3 is quite limited. The most common case is that hɒ3 modifies the adjective ljaŋ⁴ ‘several’, as in the following example.
(21)头歇儿车路里有好两个人轧死哉□。
də⁴ɕjᴇ̃tɕʰjo1lu⁶li⁴ɦjə⁴hɒ3ljaŋ⁴ɡə⁸ȵiŋ2za⁸-sa⁷
just.nowroadinsidethere.begoodseveralclfpersoncrush-die
dzᴇmə⁸.
crsprt
‘There were easily several people that were crushed to death on the road.’
As can be observed from the examples given above, the modal uses of hɒ3 ‘good’ account for a number of its auxiliary uses. The nuances of its modal meanings will be elaborated in the next section.

6. Reconstruction

We have shown in Section 4 the polysemy of the lexeme hɒ3 in Jidong Shaoxing, but from which meaning is the modal hɒ3 derived? We propose that the modal meaning of hɒ3 is basically derived from the meaning ‘fit to’ but not from the primary meaning ‘good’. To be exact, the hɒ3 of ‘fit to’ has extended to denote first circumstantial and then deontic possibility and necessity. It is from circumstantial possibility that hɒ3 has extended to express participant-internal possibility. Finally, hɒ3 has extended to denote epistemic possibility. For this, see Figure 1.
Figure 1. Extension pathways of hɒ3 in Jidong Shaoxing.
We propose that the extension of hɒ3 from a non-modal auxiliary to a modal auxiliary is motivated by contextual reanalysis and is a result of grammaticalization. Even though the ‘fit’ hɒ3 and the modal hɒ3 are both auxiliaries, the modal hɒ3 is more desemanticized than the ‘fit’ hɒ3, with desemanticization being one of the four parameters for identifying grammaticalization ().
The following subsections will explain the semantic extension of hɒ3 stage by stage.

6.1. ‘Fit to’ > Circumstantial Possibility > Deontic Modality

We identify the auxiliary use of hɒ3 ‘fit to’ as the source meaning of its modal uses, since the auxiliary uses of hɒ3 provide the primary syntactic context for its further extension, or grammaticalization, to modal auxiliaries, that is, [AUX VP]. Note that the meaning ‘fit to’ derives from the meaning ‘good’, as ‘good’ can imply the meaning ‘suitable, fit to’. See example (13).
Ambiguity between the meaning ‘fit to’ and the modal meaning ‘can, may, should’ can be easily observed in Jidong Shaoxing. This kind of ambiguous context is labelled “bridging context”5 by () and “critical context” by (). We adopt ’s () context-induced grammaticalization model to illustrate the process from ‘fit to’ to ‘can, may, should’ for hɒ3 in Jidong Shaoxing. Ambiguous contexts play an important role in the process of semantic change and grammaticalization “giving rise to an inference in favor of a new meaning” (). They are the environments where the mechanism of reanalysis takes effect. That is to say, a bridging context of ‘fit to’-‘can’ provides a breeding environment where the modal meaning of hɒ3 can be inferred. A complete process for the emergence of a new meaning is proposed to comprise four stages: (i) initial stage, (ii) bridging context, (iii) switch context, and (iv) conventionalization ().
In the initial stage, ‘fit to’ is the only reading of hɒ3. Although in most cases hɒ3 ‘fit to’ can also be interpreted as ‘can, may, should’, especially in positive sentences, the exclusive meaning of ‘fit to’ is well preserved in the negated form of hɒ3, i.e., [NEG hɒ3 VP]. As illustrated in (44), the VP fə⁷ da hɒ3 tsʰə̃1 can only be interpreted as ‘not suitable to wear’. While a Mandarin native speaker or a speaker of other Wu dialects would probably not be convinced by our claim that the negated hɒ3 cannot be interpreted as ‘cannot, may not’, as we have mentioned above, the negated hɒ3 has not yet developed any modal meaning in Jidong Shaoxing. See (31)–(33) above. The meaning of ‘cannot, may not, should not’ can only be expressed by a negated potential construction.
Initial Stage
(44)是话下卯十二月里哉，侬个双鞋介薄横不大好穿哉。
zᴇ⁴ɦwo⁶ɦo⁴mɒ⁶zə⁸ȵi⁶ɦjo⁸li⁴dzᴇ,noŋ⁴ɡə⁸ɕjɒŋ1
iflaterDecemberinsidecrs2sgpossclf
ɦa2ka⁵bo⁸ʔwaŋ3fə⁷dahɒ3tsʰə̃1dzᴇ.
shoesothinafter.allnegverygoodwearcrs
‘Your shoes are so thin and by December they won’t be suitable to wear after all.’
The asymmetrical semantic extension of hɒ3 ‘can’ and fə⁷ hɒ3 ‘not suitable’ in Jidong Shaoxing helps us to locate the lexical source of the modal hɒ3. The asymmetry can be explained by the principle of persistence (), which refers to lexical traces being retained in a grammaticalized form in the process of grammaticalization.
In a bridging context, hɒ3 is ambiguous and can be interpreted either as ‘fit to’ or as ‘can’. It is in such contexts that the lexeme hɒ3 ‘fit to’ is reanalyzed as ‘can’. This reanalysis can be seen in example (45), where the clause ‘he’s not here’ provides a suitable condition for the speaker to say something, and for hɒ3 to be reanalyzed as ‘can’.
Bridging Context
(45)渠人{无有}□，我有两句说话好话哉。
ɦi⁴ȵiŋ2ʔȵjə3mə⁸,ŋo⁴ɦjə⁴liaŋ⁴tɕy⁵
3sgpersonneg.haveprt1sghaveseveralclf
ɕjo⁷ɦwo⁶hɒ3ɦwo⁶dzᴇ⁶.
speechgoodsaycrs
‘Since he’s not here, I have something that is suitable to say.’
‘Since he’s not here, I have something that can be uttered.’
In a switch context, the new modal meaning of hɒ3 is the only interpretation. However, as pointed out by (), in this stage the target meaning still needs to be supported by a context. In (46), in the context that ʔjɒ⁵ tso⁵ ɡə⁸ tu1 tso⁵-hɒ3 dzᴇ⁶ ‘(I) finish all that should be done’, the hɒ3 in the following clause liŋ⁶ŋa⁶ zɿ⁶tʰi3 ʔȵjə3 so⁵ hɒ3 tso⁵ ɡo⁸ can only be interpreted as a modal verb and the clause denotes ‘there’s nothing else that (I) can do’. Without this context, the clause liŋ⁶ŋa⁶ zɿ⁶tʰi3 ʔȵjə3 so⁵ hɒ3 tso⁵ ɡo⁸ can also be interpreted as ‘there’s nothing else that fits (me) to do’. Undoubtedly, it is the specific context that helps to rule out the source meaning ‘fit to’.
Switch Context
(46)要做个都做好哉，另外事体{无有}啥好做{个□}。
ʔjɒ⁵tso⁵ɡə⁸tu1tso⁵-hɒ3dzᴇ⁶,liŋ⁶ŋa⁶zɿ⁶tʰi3
needdonmlzalldo-goodcrsotherthing
ʔȵjə3so⁵hɒ3tso⁵ɡo⁸.
neg.havewhatgooddonmlz.prt
‘(I) finish all that should be done, there’s nothing else that (I) can do.’
At the stage of conventionalization, the modal meaning of hɒ3 becomes independent of the source meaning ‘fit to’ which means that its modal meaning does not need to be supported by a specific context. In (47), the ‘can’ meaning of hɒ3 is the only interpretation.
Conventionalization
(47)□头卯好走另外埭路个□。
ŋa⁴də⁴mɒ⁶hɒ3tsə3liŋ⁶ŋa⁶da⁶lu⁶ɡə⁸jæ.
1pljust.nowgoodwalkotherclfroadaffprt
‘We could take the other road just now.’
We must admit that, as a modal verb, hɒ3 has attained a certain degree of conventionalization, as demonstrated in (47). However, there are still constraints and restrictions closely related to the source uses of hɒ3 which can be explained by the principles of persistence and layering (). Ambiguity between ‘fit to’ and ‘can’ emerges when hɒ3 denotes circumstantial possibility. In addition to contextual information, the syntactic units and semantic components of a sentence are also important in interpretating the meaning of hɒ3. Compare examples (48) and (49) of circumstantial possibility below. Each component of the sentence adds to its interpretation. In (48), the verb ti3tsa⁵ ‘pay a debt in kind’ implies that the items used to pay a debt are of a certain value, thereby implying that items of a certain value ‘fit to’ and ‘can’ be used to pay a debt. In contrast, (49) is a simple statement that lettuce, a common vegetable, can be served after a simple preparation. The meaning ‘fit to’ is not compatible with this particular sentence.
(48)值铜钿个东西好抵债个□。
dzə⁸doŋ2djᴇ̃2ɡə⁸toŋ1ɕi1hɒ3ti3tsa⁵ɡə⁸jə⁸.
worthmoneyrelthinggoodrepayaffprt
’Anything of value is suitable for repaying the debt.
‘Anything of value can (be used to) repay my debt.’
(49)生菜水里汆一记就好吃哉。
saŋ1tsʰᴇ⁵sɿ3li⁴tsʰə̃1ʔjə⁷tɕi⁵ʑjə⁶hɒ3tɕʰjə⁷dzᴇ.
lettucewaterinsideblanchonevclfthengoodeatcrs
‘Just blanch in boiling water, and the lettuce can be eaten then.’
Unlike in cases of circumstantial possibility, when denoting deontic permission, weak obligation, participant-internal possibility, and epistemic possibility, hɒ3 can hardly be interpreted as ‘fit to’. One more example of permission (deontic possibility) is given below. Interpretating hɒ3 as ‘fit to’ in (50) is impossible. See (55) and (58) for examples of participant-internal possibility and epistemic possibility, respectively.
(50)小人□个好不懂礼貌□？
ɕjɒ3ȵiŋ2na⁸ɡə⁸hɒ3fə⁷toŋ3li⁴mɒ⁶ȵiŋ?
childhowgoodnegknowpolitenessprt
‘How could it be that children do not know about being polite?’
The ambiguity between ‘fit to’ and circumstantial ‘can’ is the major reason we have proposed in Figure 1 that, within the participant-external modality expressed by hɒ3, it is from circumstantial possibility that hɒ3 extends to express deontic modality. Our hypothesis conforms to general principles of grammaticalization. The fact that hɒ3 exhibits a high frequency of ambiguity when denoting circumstantial possibility suggests the hɒ3 of circumstantial possibility is less desemanticized and thus less grammaticalized. Cross-linguistically, it is also attested that circumstantial possibility can extend to express deontic possibility, such as ‘get to’ in English () and Chinese de2/dei3 得 ‘obtain’ (). See also hao3 ‘good’ in the history of Chinese, as discussed in Section 6.5.
Like circumstantial possibility, permission is a kind of possibility determined by external conditions. The example below gives a case that can be understood either as circumstantial possibility or as permission. On the one hand, kids are usually thought to have fewer obligations and more leisure time than adults do. Under such circumstances, kids can have fun and hang out as they wish. On the other hand, (51) can also be read as giving permission, in that kids may play at will since they are free from many social obligations.
(51)□大姑娘好随便搞{个□}，□大人随便搞不来个□。
na⁴do⁶ku1ȵjaŋ2hɒ3dzᴇ2bjᴇ̃⁶kɒ3ɡo⁸,ŋa⁴do⁶ȵiŋ2
2plgirlgoodat.willplayaff.prt1pladult
dzᴇ2bjᴇ̃⁶kɒ3-və⁸-lᴇ2ɡə⁸jæ.
at.willplay-neg-comeaffprt
‘You little girls can/may hang out and have fun as you wish, but as adults we can’t play at will.’ (Elicitation)
The stage of permission is probably an intermediate stage in hɒ3′s extension from circumstantial possibility to weak obligation (see for English must and German müssen) since we do not observe any ambiguous contexts of circumstantial possibility and weak obligation. Yet, ambiguity between permission and weak obligation is readily attested. Example (52) can be interpreted in two ways. If doing chores is the agreed daily routine prior to homework, hɒ3 denotes permission. However, if doing chores is the choice of the participant and there is still homework to do, hɒ3 is interpretable as weak obligation.
(52)是介{无有}事体哉，侬好做作业去哉。
zᴇ⁴ka⁵ʔȵjə3zɿ⁶tʰi3dzᴇ,noŋ⁴hɒ3tso⁵
apart.from.thisneg.havethingcrs 2sggooddo
tso⁷ȵjə⁸tɕʰi⁵dzᴇ.
homeworkgocrs
‘Apart from this, there are no chores. You may/should do your homework.’ (Elicitation)
In the example below, a father impatiently urges his child to do homework. The permission meaning of the clause hɒ3 tso⁵ tso⁷ȵjə⁸ is ruled out by the context and can only be understood as ‘(you) should do your homework’.
(53)有有搞撑□□？好做作业哉□！
ɦjə⁴ɦjə⁴kɒ3-tsʰaŋ⁵lᴇɒ?hɒ3tso⁵tso⁷ȵjə⁸dzᴇjə⁸!
havehaveplay-enoughprtprtgooddohomeworkcrsprs
‘Are you done with (the games)? (You) should do your homework.’

6.2. Circumstantial Possibility > Participant-Internal Possibility

Under the framework of context-induced grammaticalization, we propose that it is from circumstantial possibility that participant-internal possibility is derived. As claimed by (), “circumstantial possibility with animate agents usually presupposes ability”. As in (54), the action of crossing the ditch is enabled by two conditions. One is the width of the ditch, and the other is the physical ability of the participant. The former is the enabling circumstantial condition, while the latter is a determining inherent ability.
(54)□道沟{只有}一些末儿劳什，一记过之好跨过去个□。
haŋ⁵da⁶kjə1tɕjə⁵ʔjə⁷sə⁷ma⁸-ŋlɒ2zə⁸,ʔjə⁷
distclfditchonly.havea.little-dimthingone
tɕi⁵-ku⁵tsɿ1hɒ3kʰo1-ku⁵tɕʰi⁵ɡə⁸la.
vclf-dimgoodstride-pass.overaffprt
‘That ditch is such a little thing. (I) can cross over by taking just one jump.’
Example (55) is a have-construction used to express possibility. The possibility of earning money is enabled by the condition that the participant, my father, does woodworking. In fact, the have-construction tsoŋ3 ɦjə⁴ ljaŋ⁴ kʰwᴇ⁵ hɒ3 tsʰɘŋ⁵ can denote circumstantial possibility even if the context is not considered. Namely, there’s always some money that one can earn. Given that it is the same referent who does woodworking and earns money, the meaning of circumstantial possibility can be ruled out. Example (56) offers a case where participant-internal possibility is the only interpretation.
(55)□老爹做做木匠，总有两块钱好趁。
ŋa⁴lɒ⁴tja1tso⁵tso⁵mə⁸ɦjaŋ⁶,tsoŋ3ɦjə⁴ljaŋ⁴
1sg.poss.kindaddo.dlmcarpentersomehowhaveseveral
kʰwᴇ⁵hɒ3tsʰɘŋ⁵.
clfgoodearn
‘My father does some woodworking on and off and (he) can somehow make some money (out of it).’
(56)我眼睛是不好，介我好听个□。
ŋo⁴ŋjæ⁴tɕiŋ1zᴇ⁴fə⁷hɒ3,ka⁷ŋo⁴hɒ3tʰiŋ⁵ɡə⁸ja.
1sgeyeindeedneggoodbut 1sggoodhearaffprt
‘My eyes aren’t good, but (even so) I can hear.’
When denoting participant-internal possibility, there exist restrictions for hɒ3 that may be related to both circumstantial possibility and its lexical meaning.
In Jidong Shaoxing, potential constructions are commonly used to express participant-internal possibility (see also ). In this domain, the distribution of potential constructions and hɒ3 partially overlap. hɒ3 can be replaced by a potential construction in most cases, except for modal have-constructions, which exclusively use hɒ3. For example, the second clause of (56), reproduced below, can also be realized by a potential construction.
(57)介我听得见个□。
ka⁷ŋo⁴tʰiŋ⁵-tə⁷-ȵjᴇ̃⁶ɡə⁸ja.
but 1sghear-pot-seeaffprt
‘but (even so) I can hear.’
When denoting participant-internal possibility, potential constructions are more generalized and neutral, while hɒ3 is most often observed in one of two specific contexts. The first often involves an enabling condition, external or internal, as in (54). In the second, hɒ3 expresses a possible option. This is the case in (56), a sentence produced in the context of a concert. Here, the ability to hear provides an option for enjoying a concert, even though one’s eyesight is not good. These two types of contexts contain traces of hɒ3′s use denoting circumstantial possibility, i.e., possibility enabled by external circumstances. The example below shows a case where hɒ3 cannot be used to express inherent ability. To answer the question ‘Can you hear (me)?’, only the potential construction can be used, as in (58ai). hɒ3 can neither be used to form the question ‘Can you hear (me)?’ nor be used to answer the question, as in (58aii).
(58)q-侬耳朵好不好□？听不听得见□？
noŋ⁴ȵi⁴tohɒ3və⁸hɒ3lᴇ?tʰiŋ⁵və⁸tʰiŋ⁵-tə⁷-ȵjᴇ̃⁶lᴇ?
2sgeargoodneggoodprthearneghear-pot-seeprt
‘Did you get your hearing back? Can you hear (me)?’
ai-听得见{个□}。
tʰiŋ⁵-tə⁷-ȵjᴇ̃⁶ɡo⁸.
hear-pot-seeaff.prt
‘(Yes, I) can.’
aii-*好听{个□}。
*hɒ3tʰiŋ⁵ɡo⁸.
goodhearaff.prt
Attempted: ‘Yes, I can.’ (Elicitation)
In addition, we observe that hɒ3 is not compatible with the [V-tə⁷-COMP] potential construction. A pair of contrastive sentences is given in (59) to better illustrate this restriction on the use of hɒ3. The context of (59a) entirely suits hɒ3 in that the possibility of seeing clearly is allowed or enabled by wearing glasses. However, since the possibility of seeing clearly is expressed by the potential construction kʰə̃⁵-tə⁷-ȵjᴇ̃⁶ ‘can see clearly’, using hɒ3 here is ungrammatical. To produce a grammatical sentence with hɒ3 in such a context, the potential construction cannot be used, as in (59b).
(59)a.渠近视，要戴眼镜会（*好）看得见。
ɦi⁴dʑiŋ⁴zɿ⁶,ʔjɒ⁵ta⁵ŋjæ⁴tɕiŋ⁵ɦwᴇ⁶(*hɒ3)kʰə̃⁵-tə⁷-ȵjᴇ̃⁶.
3sgshort-sightedneedwearglassesonly.thengoodlook-pot-see
‘He’s short-sighted. (He) must wear glasses and only then he can see clearly.’ (Elicitation)
b.渠近视个□，要戴眼镜□，好看个□。
ɦi⁴dʑiŋ⁴zɿ⁶ɡə⁸jæ,ʔjɒ⁵ta⁵ŋjæ⁴tɕiŋ⁵mə⁸,
3sgshort-sightedaffprtneedwearglassesprt
hɒ3kʰə̃⁵ɡə⁸jə⁸.
goodlookaffprt
‘He’s short-sighted and he can see only if he wears glasses.’
This restriction might be related to the source meaning of hɒ3, ‘fit to’. As the meaning ‘fit to’ implies possibility, a possible explanation is that hɒ3 does not co-occur with a potential construction to avoid semantic redundancy. An analogy would be an awkward and redundant English construction, ‘fit to be able to’. This trace persists when hɒ3 is used as a modal auxiliary.6
The development of circumstantial possibility into participant-internal possibility was neglected in the early literature on modality, with the reverse pathway, participant-internal possibility > circumstantial possibility, generally being accepted by scholars (; ; ). With the addition of linguistic evidence from Southeast Asian languages, the proposed diachronic development from participant-internal possibility to circumstantial possibility was then revised (). () further confirms the pathway from circumstantial possibility to participant-internal possibility with the development of the Thai verb dây ‘emerge’ and the Japanese idek- ‘appear’. A view of bidirectional development between participant-internal and circumstantial possibility has now become mainstream ().

6.3. Circumstantial and Participant-Internal Possibility > Epistemic Possibility

We propose that in Jidong Shaoxing, the epistemic use of hɒ3 is the extension of both circumstantial and participant-internal possibility, contexts for both of which can be observed separately. Example (60) is understood as a case of epistemic possibility when the speaker makes a guess before fetching the clothes laid out in the sun. hɒ3 in this sentence can be interpreted as circumstantial ‘can’ if being in the sun long enough is considered as an enabling condition for drying the clothes.
Circumstantial–Epistemic
(60)晒起□两件衣裳好燥□哉。
sa⁵-tɕʰi3doŋljaŋ⁴dʑjᴇ̃⁶ʔi1ʑjɒŋ2hɒ3sɒ⁵doŋdzᴇ.
dry.in.the.sun-incdurseveralclfclothesgooddryprfcrs
‘The clothes in the sun may have dried.’
Example (61) is a case of participant-internal–epistemic polysemy. Example (62) gives an ambiguous case of circumstantial, participant-internal, as well as epistemic interpretations.
Participant-internal–Epistemic
(61)渠一百斤都挑得来□，八十斤咸般也好挑个□。
ɦi⁴ʔjə⁷pa⁷tɕiŋ1tu1tʰjɒ1-tə⁷-lᴇ2lᴇ,pa⁷zə⁸tɕiŋ1
3sgone.hundreda.half.kiloallcarry-pot-comeprteightya.half.kilo
ɦæ2pæ1ɦa⁴hɒ3tʰjɒ1ɡə⁸jə⁸.
certainlyalsogoodcarryaffprt
‘(Since) he can lift 50 kilos, he can/may certainly lift 40 kilos.’
Circumstantial–Participant-internal–Epistemic
(62)毕业哉个说话，渠好做个老师{个□}。
pjə⁷ȵjə⁸dzᴇɡə⁸ɕjo⁷ɦwo⁶,ɦi⁴hɒ3tso⁵ɡə⁸lɒ⁴sɿ1ɡo⁸.
graduationcrsif 3sggooddoclfteacheraff.prt
‘He can/may be a teacher after graduation.’
Cross-linguistically, deontic–epistemic polysemy is well attested and studied. English must is a well-known example (). Even though hɒ3 can be used to express both deontic and epistemic meanings, we do not posit an evolutional relation between the two meanings in Jidong Shaoxing, as polygrammaticalization () may also be possible. The main reason for this conclusion is that a bridging context of deontic–epistemic polysemy is rarely observed among conservative speakers. The three conservative speakers in this study considered the sentence in (63) to suggest permission or weak obligation, while only the innovative speaker involved in this study claimed that the sentence can express both deontic and epistemic meanings.
(63)五点钟哉，渠好去哉。
ŋ⁴tjᴇ̃3tɕjoŋ1 dzᴇ, ɦi⁴hɒ3tɕʰi⁵dzᴇ.
fiveo’clockcrs 3sggoodgocrs
‘It’s five o’clock (and time to get off). He may/should leave.’
‘It’s five o’clock (and time to get off). #He may probably be gone.’
For this study, we also tested quite a few deontic expressions formed by hɒ3 in epistemic contexts. The tests turned out to be failures with the three conservative speakers. One of the examples is given below. The clause hɒ3 tɕjə⁷hwɘŋ1 dzᴇ jə⁸ ‘(he) should get married’ expressed by hɒ3 in (64ii) is a speaker’s advice. Using it to answer the question ‘Is he married?’ in (65q) to express one’s presumption was ungrammatical for our conservative speakers but caused no problems for the innovative speaker, as shown in (65ai). Instead, the conservative speakers used ʔiŋ1kᴇ1 ‘should’ to form a presumption as an answer to the question, as in (65aii).
(64)渠年纪也不小□哉，好结婚哉□。
(i)ɦi⁴ȵjᴇ̃2tɕi⁵ɦa⁴fə⁷ɕjɒ3doŋdzᴇ.
3sgagealsonegsmallprfcrs
(ii)hɒ3tɕjə⁷hwɘŋ1dzᴇjə⁸.
goodmarrycrsprt
‘He’s not young. (He) should get married.’
(65)q-渠婚有有结□□？
ɦi⁴hwɘŋ1ɦjə⁴ɦjə⁴tɕjə⁷lᴇɦa?
3sgmarriagehavehavemarryprtprt
‘Is he married?’
ai-#渠好结婚□哉{个□}。
#ɦi⁴hɒ3tɕjə⁷hwɘŋ1doŋdzᴇɡo⁸.
3sggoodmarryprfcrsaff.prt
‘He may probably be married.’ (Elicitation)
aii-渠应该结婚□哉{个□}。
ɦi⁴ʔiŋ1kᴇ1tɕjə⁷hwɘŋ1doŋdzᴇɡo⁸.
3sgshouldmarryprfcrsaff.prt
‘He may probably be married.’
Epistemic possibility is probably the latest layer of the semantic and functional extension of hɒ3. There are two main pieces of evidence in support of this hypothesis.
First, when denoting epistemic possibility, the use of hɒ3 is still dependent not only on its circumstantial possibility use, but in many cases, on the source meaning ‘fit to’. This means that there needs to be a certain enabling condition for a grammatical proposition of epistemic possibility realized with hɒ3, whose surface meaning corresponds to ‘may’ but whose underlying meaning is ‘it is the right or proper moment for’. Example (66) is one such grammatical example where hɒ3 is used to express a presumption based on the speaker’s judgement and knowledge. Here, the sentence was produced during the airtime of a frequently watched television program. Similarly, example (60) can also be read ‘it’s the right time for the clothes to have dried’.
(66)介光折电视好开始□哉。
ka⁵kwɒŋ1tsə⁷djᴇ̃⁶zɿ⁶hɒ3kʰᴇ1sɿ3doŋdzᴇ.
nowclfTVgoodbeginprfcrs
‘The TV show may have been on.’
In comparison, although (67) is similarly an expression of probability based on one’s judgement, the use of hɒ3 would be ungrammatical. As mentioned above, hɒ3′s epistemic use is still restricted by its source uses. Looking awful is neither an enabling condition for falling ill, nor reflective of the moment for falling ill. Rather, it is a sign of being ill.
(67)渠人介难看，*好生毛病□哉□。
ɦi⁴ȵiŋ2ka⁵næ2kʰə̃⁵,*hɒ3saŋ1mɒ2biŋ⁶doŋdzᴇɒ.
3sgpersonsoout.of.sortsgoodhavesicknessprfcrsprt
‘He looks awful and may be sick.’ (Elicitation)
Second, different speakers show different degrees of tolerance for using hɒ3 to express epistemic possibility. The cases of (63) and (65) have already provided a glimpse into this situation. Examples (68) and (69a) were only accepted by our innovative consultant. Sometimes it is difficult to determine whether or not the epistemic meaning of hɒ3 can be accepted. We reproduce example (42) in (69b) to highlight cases of arbitrariness. The only difference between (69a) and (69b) is the subject. Our conservative consultants could only accept (69b), and when replacing the subject ‘car’ with the third person pronoun the sentence turns out to be ungrammatical.
(68)#渠今年好有五岁□哉。
#ɦi⁴tɕi⁴ȵjᴇ̃2hɒ3ɦjə⁴ŋ⁴sᴇ⁵doŋdzᴇ.
3sgthis.yeargoodhavefiveyearprfcrs
‘He may have been five years old this year.’ (Elicitation)
(69)a.#渠好来□哉。
#ɦi⁴hɒ3lᴇ2doŋdzᴇ.
3sggoodcomeprfcrs
‘He is probably coming.’ (Elicitation)
b.车好来□哉。
tɕʰjo1hɒ3lᴇ2doŋdzᴇ.
cargoodcomeprfcrs
‘The car/bus is probably coming.’
Restrictions on using the epistemic hɒ3 reflect the fact that hɒ3 is in the process of functional extension or grammaticalization. As can be observed from the examples above, the generalization of the epistemic hɒ3 varies among different speakers. Still, all of this information suggests that the epistemic use of hɒ3 is its latest layer of extension.

6.4. Accelerating Factors

We have identified the auxiliary hɒ3 ‘fit to’ as the source of its modal meanings including ‘can’, ‘may’, and ‘should’. Other factors may also accelerate or generalize the extension of hɒ3.
First, hɒ3 itself can be used independently to ask for agreement, which is probably an extension from the meaning ‘good’, as in (70). Since asking for agreement presupposes permission, this use of hɒ3 can definitely promote extension to permission.
(70)介我前拨侬换个，好？
ka⁷ŋo⁴ʑjᴇ̃2pə⁷noŋ⁴ɦwə̃⁶ɡə⁸,hɒ3?
so1sgfirstben2sgchangeclfgood
‘I’ll change it for another one for you, OK?’
Second, though rare, we do observe some contexts of ‘easy’-‘can’ polysemy, which means that ‘be easy to’ is also a possible source for ‘can’. Example (71) is a case where hɒ3 can either be interpreted as ‘be easy to’ or as the ‘can’ of circumstantial possibility, that is, either the thin and watery texture of corn porridge makes it easy to swallow, or one could drink the porridge (like drinking water). In addition, example (40) showed a case of ‘fit’-‘easy’-‘can’ polysemy, which can be interpreted as ‘(leaves) easily float on the surface of the water’.
(71)□辰光六谷糊煞煞薄{个□}，□□好呷个□。
haŋ⁵zɘŋ2kwɒŋ1lo⁸kwo⁷ɦu2sa⁷sa⁷bo⁸ɡo,do⁶do⁶hɒ3ha⁷
distmomentcornporridgethinaff.prtonogooddrink
ɡə⁸la.
affprt
‘The corn porridge (we used to eat before) was very thin.
a. [hɒ⁴3 haʔ⁵]: It was easy to drink (like drinking water).’ Elicitation
b. [hɒ⁴⁴ haʔ⁴]: (One) could drink (instead of chewing it, like drinking water).’
The reason the meaning ‘be easy to’ is not identified as the source for the modal hɒ3 is that contexts suggesting ‘easy’-‘can’ polysemy are less frequent than those suggesting ‘fit’-‘can’ polysemy. Moreover, the sandhi patterns of hɒ3 are different when denoting these two different meanings. As given in the translations of (71), hɒ⁴3 haʔ⁵ denotes ‘easy to drink’, a sandhi pattern of forming a compound word, while hɒ⁴⁴ haʔ⁴ signifies ‘can drink’.
Undoubtedly, these factors illustrated above contribute to the generalization of the modal uses of hɒ3. It is true that grammaticalization is unpredictable to a certain degree, but frequency of use still plays a role in expanding the possibilities a given form has for grammaticalization ().

6.5. Hao3 好 ‘Good’ in the History of Chinese

We have reconstructed the functional extension of hɒ3 in Jidong Shaoxing in Section 6.1, Section 6.2 and Section 6.3 by adopting the model of context-induced grammaticalization proposed by (). Although our reconstruction of hɒ3 cannot be directly supported due to a lack of diachronic records of Jidong Shaoxing,7 it conforms to the evolution of hao3 好 ‘good’ (the etymon of hɒ3) in the history of Chinese. Based on Li’s work (2017), and diachronic analyses proposed by () and (), the evolution of hao3 in the history of Chinese is reorganized and adapted in Table 2.
Table 2. Evolution of hao in the history of Chinese.
According to (), (), and (), modal uses of hao3 can be first observed in Early Medieval Chinese (3rd century–6th century).8 During this time, it was used to express circumstantial possibility and could be interpreted as ‘fit to’ or ‘can’, as shown in (72). Compare this example with the Jidong Shaoxing example (26), reproduced here in (73).
(72)羔有死者，皮好作裘褥，肉好做干腊，及作肉酱，味又甚美。
gao1you3si3zhe3,pi2hao3zuo⁴qiu2ru⁴rou⁴hao3
lambhavedienmlzskingooddofur.mattressmeatgood
zuo⁴gan1la⁴,ji2zuo⁴rou⁴jiang⁴,wei⁴you⁴shen⁴mei3.
docured.meatanddomeat.sauceflavoralsoverypretty
‘(If) there’s a dead lamb, the fur [can be]/[fits to be] made into a mattress and the meat [can be]/[fits to be] made into cured meat and meat sauce which is extremely delicious.’
Qi Min Yao Shu · Yang Yang 齐民要术·养羊 (544ad) [Essential techniques for the welfare of the people · Raising sheep]
(Cited from () and glossed and translated by S. Lü)
(73)Jidong Shaoxing
甲鱼背好做药个□。
ka⁷ŋ2pᴇ⁵hɒ3tso⁵ɦja⁸ɡə⁸jæ.
soft.shell.turtlebackgooddomedicineaffprt
‘Turtle shells can be made into (Chinese traditional) medicine.’
The ‘fit’-‘can’ polysemy of hao3 persisted until its circumstantial possibility use began to decline in Modern Chinese, specifically during the Qing Dynasty. In contemporary Standard Mandarin, only the fossilized zhi3hao3 只好 ‘can only’ is used to denote circumstantial possibility. As shown in (74), the deletion of the adverb zhi3 ‘only’ is ungrammatical in Standard Mandarin.
(74)Standard Mandarin
他腿断了，*（只）好在家休息。
ta1tui3duan⁴le,*(zhi3)hao3zai⁴ jia1xiu1xi⁴.
3sglegbreakcrsonlygoodathomerest
‘His leg is broken and he can only take a rest at home.’
As indicated in Table 2, towards the end period of Late Medieval Chinese, which corresponds to the Song Dynasty (960ad–1279), a significant new meaning of hao3 emerged—the deontic meaning ‘should’. However, this use only lasted to Pre-Modern Chinese. See (75).
(75)似这般汉，正好蓦头蓦面唾。
si⁴zhe⁴ban1han⁴,zheng⁴hao3mo⁴tou2mo⁴mian⁴tuo⁴.
resemblesomanjustgoodin.the.facespit
‘A person like this, (one) should spit on him in the face.’
Bi Yan Lu · 78 Ze 碧岩录·78则 (1125) [Blue Cliff Record · Verse 78]
(Cited from () and glossed and translated by S. Lü)
A bit later than the deontic use of hao3, the interpretation of participant-internal possibility appeared in Pre-Modern Chinese during the Yuan and Ming Dynasties (1271–1644), as shown below. Note that () does not single out the meaning of participant-internal possibility for hao3.
(76)您兄弟量窄，只好陪哥哥一小钟。
nin2xiong1di⁴liang⁴zhai3,
2sg.honsiblingcapacitynarrow
zhi3hao3pei2ge1ge1yi⁴xiao3zhong⁴.
onlygoodaccompanybrotheronesmallcup
‘I’m not good at drinking (alcohol) and I can only drink a small cup to accompany you.’
Yuan Qu Xuan · Zhusha Dan 元曲选·朱砂担 (1616) [Selected Yuan Theatre Plays · A Picul of Cinnabar]
(Cited from () and glossed and translated by S. Lü)
Like the deontic use, the participant-internal possibility use of hao3 did not last long and was not further generalized.
As for the meaning ‘be easy to’, the ‘easy’-‘can’ polysemy can also be observed for hao3, as in (77).
(77)嫂嫂，你如今真个不好过日子，不如跟着我一同回去住罢。
sao3saoni3ru2jin1zhen1ge⁴bu⁴hao3guo⁴ri⁴zi
sister-in-law2sgnowindeedneggoodlivelife
bu⁴ru2gen1-zhewo3yi⁴tong2hui2-qu⁴zhu⁴ba.
inferiorfollow-dur1sgtogetherreturn-goliveprt
‘Sister, you [aren’t easy to]/[can’t] make a living now. It would be better to come to live with me.’
Yuan Qu Xuan · Ren Fengzi 元曲选·任风子 (1616) [Selected Yuan Theatre Plays · Ren Fengzi]
(Cited from () and glossed and translated by S. Lü)
Nevertheless, as can be seen in Table 2, hao3′s meaning ‘be easy to’, considered as an evaluative meaning by (), emerged later than the meaning ‘fit to/can’, sometime between the Tang and the Five Dynasties (618–960ad) (see also ; ). This suggests that ‘be easy to’ is not the direct source for the modal uses of hao3. The meaning ‘be easy to’ for hao3 maintains an active status in Standard Mandarin.
The evolution of hao3 in the history of Chinese parallels the extension of hɒ3 in Jidong Shaoxing and supports our reconstruction of hɒ3. The fact that hao3 ‘good’ is used to denote weak obligation and participant-internal possibility in the history of Chinese sheds some light on the evolution of hɒ3 ‘good’ in Jidong Shaoxing. Both of these uses appeared much later than the circumstantial possibility use, suggesting that the chain ‘circumstantial possibility > participant-internal possibility’ for hɒ3 in Jidong Shaoxing is plausible. Furthermore, the emergence and generalization of hao3′s meaning ‘be easy to’ can also be mapped onto Jidong Shaoxing hɒ3. That ‘be easy to’ emerged later than the meaning of circumstantial possibility suggests the implausibility of identifying ‘be easy to’ as the source meaning for circumstantial ‘can’.

7. Conclusions

This paper has provided a case study on the modal uses of the lexeme hɒ3 ‘good’ in Jidong Shaoxing. The lexeme hɒ3 is a polysemous and multi-functional word. hɒ3 can serve as an adjective/adjectival verb ‘good, fitting, ready, done, ok’. Its modal uses include circumstantial ‘can’, participant-internal ‘can’, permission ‘may’, weak obligation ‘should’, and epistemic ‘can’. hɒ3 shows asymmetrical semantic extension in Jidong Shaoxing. While the positive form of hɒ3 possesses modal functions, the negated form fə⁷ hɒ3 only denotes ‘not good’ or ‘not suitable’.
We have proposed that it is from the meaning ‘fit to’ but not directly from the lexical meaning ‘good’ that the modal meanings of hɒ3 are derived. This pathway is different from the ‘good’ > deontic permission route found in some other languages (). Adopting the context-induced grammaticalization model, we have reconstructed the process of extension of hɒ3 in Jidong Shaoxing, proposing that the hɒ3 of ‘fit to’ has followed a multidirectional or polygrammaticalization pattern. This pattern contains an intermediate stage of circumstantial possibility in the development of both its deontic and participant-internal uses:
Chain 1: ‘fit to’ > circumstantial ‘can’ > deontic ‘may, should’.
Chain 2: ‘fit to’ > circumstantial ‘can’ > participant-internal ‘can’.
Within deontic modality, the hɒ3 of permission is reconstructed as the source for the hɒ3 of weak obligation, i.e., circumstantial ‘can’ > permission ‘may’ > weak obligation ‘should’.
Although the chain of circumstantial possibility > participant-internal possibility is cross-linguistically less common than the reverse, our proposition is based on linguistic facts in Jidong Shaoxing and conforms to the evolution of the etymon hao3 ‘good’ of hɒ3 in the history of Chinese. A similar pathway can be found in Chinese: ke3 可 ‘suitable’ > root possibility > ability (, ; see also ). Our findings contribute some new evidence for two proposed bidirectional developments in the modal domain: circumstantial ↔ participant-internal and deontic permission ↔ deontic necessity ().
The epistemic use of hɒ3 is proposed as the latest stage of extension in the current Jidong Shaoxing. It is from both circumstantial and participant-internal ‘can’ that the epistemic ‘can’ is developed:
Chain 3: circumstantial-participant-internal ‘can’ > epistemic ‘can, may’.
The epistemic use of hɒ3 is often restricted by the source meaning of circumstantial ‘can’ or ‘fit to’. Additionally, the degree of generalization of epistemic hɒ3 varies between different speakers.
The use of the lexeme hao ‘good’ as a modal verb can be considered a regional phenomenon. According to a preliminary cross-linguistic survey, this phenomenon is found across Shanghai, Jiangsu, and Zhejiang Provinces, an area usually known as the Yangtze River Delta, which covers the entire Wu speaking area and some Jianghuai Mandarin 江淮官话 speaking areas. For Wu dialects, hao ‘good’ as a modal verb is reported in Chongming 崇明 (), Shanghainese (), Suzhou 苏州 (), Hangzhou 杭州 (), Shaoxing, Yuyao 余姚 (), Ningbo (), Xianju, Wenzhou 温州 (), and Jinhua 金华 (). For Jianghuai Mandarin, the phenomenon is attested in Nantong 南通 (), Yangzhou 扬州 (), and Nanjing 南京 (). In addition, hao as a modal auxiliary is also attested in some discontinuous areas in Guangdong and Taiwan, that is, in Jieyang Southern Min and Hakka, as mentioned at the beginning of the paper. Certainly, the modal uses of hao vary in different languages and dialects. More work needs to be carried out to figure out how the modal functions of hao extend and to what extent its modal uses can be generalized in individual languages.

Author Contributions

Conceptualization, S.L.; methodology, S.L.; data analysis, S.L.; data collection, S.L. and X.H.; data transcription, X.H.; writing—original draft preparation, S.L.; writing—review and editing, S.L. and X.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Shanghai Planning Office of Philosophy and Social Science, grant number 2022BYY005.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

Our sincere gratitude goes to our consultants and to the five anonymous reviewers for their insightful and valuable comments. We would also like to thank Agnes Sheridan Conrad for proofreading the paper.

Conflicts of Interest

The authors declare no conflict of interest.

Abbreviations

* ungrammatical, # accepted only by innovative speakers, ( ) omissible, { } erosion, □ unknown etymon, 1 first person, 2 second person, 3 third person, adv adverbializer, aff affirmative, all allative, ben benefactive, c-poss circumstantial possibility, clf classifier, cop copula, comp complement, crs currently relevant state, d-nec deontic necessity, dim diminutive, dist distal demonstrative, dlm delimitative, dur durative, hon honorific, imp imperative, inc inchoative, kin kinship, lit literal meaning, mod modifier marker, neg negator, ono onomatopoeia, pass passive agent marker, pi-poss participant-internal possibility, pl plural, pot potential marker, prf perfect, prosp prospective, prt particle, q question, rel relativizer, s_prox speaker-based proximal demonstrative, sg singular, top topic marker, vclf verbal classifier, vcomp verb complementizer, viv vivid form of adjective.

Notes

1
A rare source for can modals observed in Jidong Shaoxing is the potential construction tɕʰjə⁷-tə⁷-lə⁸ 吃得落 ‘can < eat-pot-fall’. This construction can function like an auxiliary denoting possibility in the form of [tɕʰjə⁷-tə⁷-lə⁸ VP]. We do not identify tɕʰjə⁷-tə⁷-lə⁸ as an auxiliary verb because it is subject to a different negation strategy from that of negating an auxiliary. Instead of negating the whole potential construction in the form of [NEG tɕʰjə⁷-tə⁷-lə⁸ VP], it is the complement lə⁸ that is negated, i.e., [tɕʰjə⁷-və⁸-lə⁸ VP] 吃不落 VP ‘eat-neg-fall’ to signify ‘cannot VP’. See also ().
2
The have-construction might be considered to be a type of “modal possessive construction” () derived from a possessive-existential construction. However, unlike the constructions which are claimed to contain no overt modal elements by (), the overt modal verb hɒ3 must be used in the modal have-construction in Jidong Shaoxing. Such a modal have-construction is absent in Narrog’s findings. See also Note 3.
3
The construction with ɦjə⁴ ‘have’ in (30a-ii) is probably a bridging context (or an ambiguous context) between the possessive construction and the modal construction with ɦjə⁴ ‘have’. If the context of (30) is not taken into consideration, (30a-ii) can be interpreted either as ‘you have only one (candy) that you can eat’ or as ‘you can only eat one candy’. The covert subject ‘you’ of (30a-ii) can still be treated as the possessor of the candy, which is quite different from the case of (27). In (27), ‘three cups of rice’ is neither the possessor of ‘five bowls of (rice)’ nor the real subject of the sentence, but a topic whose semantic role is the material for ‘five bowls of (rice)’. From the perspective of semantic bleaching, example (27) is more generalized than (30a-ii). Since the development of possessive construction into a modal construction is not a focus of this paper, its grammaticalization pathway is not addressed.
4
Shanghainese belongs to the Shanghai subdivision 上海小片 of the Taihu division, while Xianju Wu belongs to the Taizhou division 台州片 ().
5
This term can be traced back to ().
6
One should note that this is not a general restriction of the lexeme ‘good’ in all the Wu dialects. For example, according to our analysis of one hour of video material, the modal use of hɔ3 ‘good’ in Shanghainese is more generalized and can co-occur with a potential construction in a sentence.
7
Historical documents compiled by missionaries concerning early Wu languages can be dated back to the mid-19th century, including Shanghainese, Ningbo, Wenzhou, Taizhou, Suzhou, and Jinhua ().
8
() propose that the meaning ‘fit to, easy to’ is derived from the meaning ‘love, like’ of hao⁴ rather than the meaning ‘good’. We do not address this proposition here because hɒ3 in Jidong Shaoxing does not possess the meaning ‘love’.

References

  1. Bao, Shijie 鲍士杰. 1998. Hangzhou Fangyan Cidian 杭州方言词典 [A Dictionary of Hangzhou]. Nanjing: Jiangsu Education Press. [Google Scholar]
  2. Bybee, Joan, Revere Perkins, and William Pagliuca. 1994. The Evolution of Grammar. Tense, Aspect and Modality in the Languages of the World. Chicago: The University of Chicago press. [Google Scholar]
  3. Cao, Zhiyun 曹志耘. 1996. Jinhua Fangyan Cidian 金华方言词典 [A Dictionary of Jinhua]. Nanjing: Jiangsu Education Press. [Google Scholar]
  4. Chao, Yuen Ren. 1968. A Grammar of Spoken Chinese. Berkely: University of California Press. [Google Scholar]
  5. Chappell, Hilary, and Alain Peyraube. 2016. Modality and mood in Sinitic. In The Oxford Handbook of Modality and Mood. Edited by Jan Nuyts and Johan van der Auwera. Oxford: Oxford University Press, pp. 296–329. [Google Scholar]
  6. Coates, Jenifer. 1983. The Semantics of the Modal Auxiliaries. London and Camberra: Croom Helm. [Google Scholar]
  7. Craig, Colette. 1991. Ways to Go in Rama: A Case Study in Polygrammaticalization. In Approaches to Grammaticalization. Edited by Elizabeth C. Traugott and Bernd Heine. Amsterdam and Philadelphia: John Benjamins, vol. 2, pp. 455–92. [Google Scholar]
  8. de Hann, Ferdinand. 2006. Typological approaches to modality. In The Expression of Modality. Edited by William Frawley. Berlin and New York: De Gruyter, pp. 27–69. [Google Scholar]
  9. Diewald, Gabriele. 2002. A model for relevant types of contexts in grammaticalization. In New Reflections on Grammaticalization. Edited by Ilse Wischer and Gabriel Diewald. Amsterdam and Philadelphia: John Benjamins, pp. 103–20. [Google Scholar]
  10. Evans, Nicholas, and David Wilkins. 2000. In the mind’s ear: The semantic extensions of perception verbs in Australian languages. Language 76: 546–92. [Google Scholar] [CrossRef]
  11. Fan, Xiaolei. 2014. Tense, Aspect and Modality in Chinese: A Typological Study. Ph.D. dissertation, Hong Kong University of Science and Technology, Hong Kong, China. [Google Scholar]
  12. Heine, Bernd. 1993. Auxiliaries: Cognitive Forces and Grammaticalization. New York and Oxford: Oxford University Press. [Google Scholar]
  13. Heine, Bernd. 2002. On the role of context in grammaticalization. In New Reflections on Grammaticalization. Edited by Ilse Wischer and Gabriel Diewald. Amsterdam and Philadelphia: John Benjamins, pp. 83–101. [Google Scholar]
  14. Heine, Bernd, and Heiko Narrog. 2010. Grammaticalization and linguistic analysis. In The Oxford Handbook of Linguistic Analysis. Edited by Bernd Heine and Heiko Narrog. Oxford: Oxford University Press, pp. 401–23. [Google Scholar]
  15. Hopper, Paul J. 1991. On some principles of grammaticization. In Approaches to Grammaticalization. Edited by Elizabeth C. Traugott and Bernd Heine. Amsterdam and Philadelphia: John Benjamins, vol. 1, pp. 17–35. [Google Scholar]
  16. Hopper, Paul J., and Elizabeth C. Traugott. 2003. Grammaticalization, 2nd ed. Cambridge: Cambridge University Press. [Google Scholar]
  17. Huang, Borong 黄伯荣, ed. 1996. Hanyu Fangyan Yufa Leibian 汉语方言语法类编 [The Grammar Handbook of Dialects in China]. Qingdao: Qingdao Publishing Group. [Google Scholar]
  18. Huang, Jilin 黄继林. 1996. Yangzhou Fangyan Cidian 扬州方言词典 [A Dictionary of Yangzhou]. Nanjing: Jiangsu Education Press. [Google Scholar]
  19. Huang, Xuezhen 黄雪贞. 1995. Meixian Fangyan Cidian 梅县方言词典 [A Dictionary of Meixian Hakka]. Nanjing: Jiangsu Education Press. [Google Scholar]
  20. Jiang, Meng 姜梦, and Zairong Cui 崔宰荣. 2017. Zhudongci ‘Hao’ de Yufahua Guocheng 助动词好的语法化过程 [On the Grammaticalization Process of the Auxiliary ‘Hao’]. Hanyu Xuexi 汉语学习 [Chinese Language Learning] 6: 21–30. [Google Scholar]
  21. Kuteva, Tania, Bernd Heine, Bo Hong, Haiping Long, Heiko Narrog, and Seongha Rhee. 2019. World Lexicon of Grammaticalization, 2nd ed. Cambridge: Cambridge University Press. [Google Scholar]
  22. Li, Jinxia 李晋霞. 2005. ‘Hao’ de Yufahua yu Zhuguanxing “好” 的语法化和主观性 [On the Grammaticalization and Subjectivity of ‘Hao’]. Shijie Hanyu Jiaoxue 世界汉语教学 [Chinese Teaching in the World] 44–49: 96. [Google Scholar]
  23. Li, Ming 李明. 2017. Hanyu Zhudongci De Lishi Yanbian Yanjiu 汉语助动词的历史演变研究 [A Diachronic Study on Chinese Auxiliary Verbs]. Beijing: The Commercial Press. [Google Scholar]
  24. Liu, Danqing 刘丹青. 1995. Nanjing Fangyan Cidian 南京方言词典 [A Dictionary of Nanjing]. Nanjing: Jiangsu Education Press. [Google Scholar]
  25. Liu, Danqing 刘丹青. 2001. Wuyu de Jufa Leixing Tedian 吴语的句法类型特点 [Syntactical Features of Wu Languages]. Fangyan 方言 [Dialects] 4: 332–43. [Google Scholar]
  26. Lü, Shuxiang 吕叔湘. 1999. Xiandai Hanyu Babai Ci 现代汉语八百词 [800 Grammatical Words in Contemporary Chinese]. Beijing: The Commercial Press. First published 1980. [Google Scholar]
  27. Meisterernst, Barbara. 2008a. Modal verbs in Han period Chinese Part I: The syntax and semantics of ke 可 and keyi 可以. Cahiers de Linguistique Asie Orientale 37: 85–120. [Google Scholar] [CrossRef]
  28. Meisterernst, Barbara. 2008b. Modal verbs in Han period Chinese Part II: Negative markers in combination with the modal auxiliary verbs ke可 and keyi 可以. Cahiers de Linguistique Asie Orientale 37: 197–222. [Google Scholar] [CrossRef]
  29. Narrog, Heiko. 2012. Modality, Subjectivity, and Semantic Change: A Cross-Linguistic Perspective. Oxford: Oxford University Press. [Google Scholar]
  30. Palmer, Frank R. 2001. Mood and Modality, 2nd ed. Cambridge: Cambridge University Press. [Google Scholar]
  31. Peyraube, Alain. 1988. Syntaxe Diachronique du Chinois: Evolution des Constructions Datives du XIVe Siècle Avant J.-C. au XVIIIe Siècle. Paris: Collège de France, Institut des Hautes Études Chinoises. [Google Scholar]
  32. Qian, Nairong 钱乃荣. 1997. Shanghaihua Yufa 上海话语法 [A Grammar of Shanghinese]. Shanghai: Shanghai People’s Press. [Google Scholar]
  33. Sheng, Yimin 盛益民. 2021. Wuyu Shaoxing (Keqiao) Fangyan Cankao Yufa 吴语绍兴（柯桥）方言参考语法 [A Reference Grammar of Shaoxing: The Keqiao Variety]. Beijing: The Commercial Press. [Google Scholar]
  34. Sun, Chaofen. 2006. Chinese: A Linguistic Introduction. Cambridge: Cambridge University Press. [Google Scholar]
  35. Tang, Zhenzhu 汤珍珠, Zhongmin Chen 陈忠敏, and Xinxian Wu 吴新贤. 1997. Ningbo Fangyan Cidian 宁波方言词典 [A Dictionary of Jingbo]. Nanjing: Jiangsu Education Press. [Google Scholar]
  36. Tao, Guoliang 陶国良. 2007. Nantong Fangyan Cidian 南通方言词典 [A Dictionary of Wenzhou]. Nanjing: Jiangsu People’s Press. [Google Scholar]
  37. Tao, Huan 陶寰, Chunhua Jin 金春华, and Yimin Sheng 盛益民. 2014. Wuyu Shaoxing Fangyan de Neibu Fenqu 吴语绍兴方言的内部分区 [Subdivision of Shaoxing Wu]. In Chengzetang Fangyan Luncong: Wang Futang Jiaoshou Ba Zhi Shouqing Lunwenji 承泽堂方言论丛—王福堂教授八秩寿庆论文集 [Chinese Dialect Essays for the Eightieth Birthday of Professor Wang Futang]. Edited by Li Xiaofan 李小凡 and Xiang Mengbing 项梦冰. Beijing: Language and Culture Press, pp. 361–89. [Google Scholar]
  38. Traugott, Elizabeth C., and Richard B. Dasher. 2001. Regularity in Semantic Change. Cambridge: Cambridge University Press. [Google Scholar]
  39. van der Auwera, Johan, and Andreas Ammann. 2013a. Situational Possibility. In WALS Online (v2020.3). Edited by Matthew S. Dryer and Martin Haspelmath. [Data set]. Zenodo: Available online: http://wals.info/chapter/74 (accessed on 2 December 2022). [CrossRef]
  40. van der Auwera, Johan, and Andreas Ammann. 2013b. Epistemic Possibility. In WALS Online (v2020.3). Edited by Matthew S. Dryer and Martin Haspelmath. [Data set]. Zenodo: Available online: http://wals.info/chapter/75 (accessed on 25 June 2022). [CrossRef]
  41. van der Auwera, Johan, and Vladimir A. Plungian. 1998. Modality’s semantic map. Linguistic Typology 2: 79–124. [Google Scholar] [CrossRef]
  42. van der Auwera, Johan, Petar Kehayov, and Alice Vittrant. 2009. Acquisitive modals. In Cross-Linguistic Semantics of Tense, Aspect and Modality. Edited by Lotte Hogeweg, Helen de Hoop and Andrej Malchukov. Amsterdam and Philadelphia: John Benjamins, pp. 271–301. [Google Scholar]
  43. Wang, Liyang 王黎阳. 2013. Fangchenghua zhong de Jige Qingtai Dongci 方城话中的几个情态动词 [Some modal verbs in Fangcheng]. Master’s dissertation, Henan University, Kaifeng, China. [Google Scholar]
  44. Wang, Ping 汪平, and Zhiyun Cao 曹志耘. 2012. Wuyu 吴语 [Wu]. In Zhongguo Yuyan Ditu Ji: Hanyu Fangyan Juan 中国语言地图集·汉语方言卷 [Language Atlas of China: Chinese Dialects]. Edited by 中国社会科学院语言研究所, 中国社会科学院民族学与人类学研究所 and 香港城市大学语言资讯科学研究中心. [Institute of Linguistics of the Chinese Academy of Social Sciences, Institute of Ethnology and Anthropology of the Chinese Academy of Social Sciences, and Language Information Sciences Research Centre of the City University of Hong Kong]. Beijing: The Commercial Press, pp. 103–9. [Google Scholar]
  45. Wang, Yingying 王盈盈. 2019. Liujiang Jinde Kejia Fangyan Qingtai Zhudongci Yanjiu 柳江进德客家方言情态助动词研究 [Modal Auxiliaries in Jinde Hakka]. Master’s dissertation, Guangxi University, Nanning, China. [Google Scholar]
  46. Xu, Huiling. 2007. Aspect of Chaozhou Grammar: A Synchronic Description of the Jieyang Variety. Journal of Chinese Linguistics Monograph Series No. 22; Hong Kong: Chinese University Press. [Google Scholar]
  47. Ye, Xiangling 叶祥苓. 1993. Suzhou Fangyan Cidian 苏州方言词典 [A Dictionary of Suzhou]. Nanjing: Jiangsu Education Press. [Google Scholar]
  48. Yeh, Jui-chuan 葉瑞娟. 2012. 論客家話‘好X’格式的語法化和主觀化 On the Grammaticalization and Subjectification of the ‘HO + X’ Construction in Hakka. Tsing Hua Journal of Chinese Studies 42–43: 527–65. [Google Scholar]
  49. You, Rujie 游汝杰. 2021. Xiyang Chuanjiaoshi Hanyu Fangyanxue Zhusuo Shumu Kaoshu 西洋传教士汉语方言学著作书目考述 [Historical Material Compiled by Foreign Missionaries on Chinese Dialects], revised ed. Shanghai: Shanghai Education Press. [Google Scholar]
  50. You, Rujie 游汝杰, and Qianming Yang 杨乾明. 1998. Wenzhou Fangyan Cidian 温州方言词典 [A Dictionary of Wenzhou]. Nanjing: Jiangsu Education Press. [Google Scholar]
  51. Zhang, Huiying 张惠英. 1993. Chongming Fangyan Cidian 崇明方言词典 [A Dictionary of Chongming]. Nanjing: Jiangsu Education Press. [Google Scholar]
  52. Zhou, Wei 周炜. 2019. Yuyao Fangyan Qingtai Dongci Yanjiu 余姚方言情态动词研究 [A Study on Modal Auxiliaries in Yuyao]. Master’s dissertation, Zhejiang University, Hangzhou, China. [Google Scholar]
  53. Zhu, Dexi 朱德熙. 1982. Yufa Jiangyi 语法讲义 [Lectures on Grammar]. Beijing: The Commercial Press. [Google Scholar]
  54. Zhuang, Chusheng 庄初生, and Tingting Huang 黄婷婷. 2014. Shijiu Shiji Xianggang Xinjie de Kejia Fangyan 19世纪香港新界的客家方言 [The Xinjie Dialect of Hakka in Hong Kong in 19th Century]. Guangzhou: Guangdong People’s Press. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
Ancient Connections of Sinitic
Previous
On Actuality Entailments, Causation, and Telicity in Balkar
Next