Skip to main content
  • Squib-A: Data and Databases
  • Open access
  • Published:

Interpreting 都 to 1 TO in earlier Southern Min texts


The paper explores the polyfunctional word 都 to 1 TO in early Southern Min texts dating back to the sixteenth century or even earlier. Unlike 都 dou 1 in Mandarin, which mainly functions as a maximality operator, 都 to 1 in Southern Min is chiefly used as a modal particle expressing a concessive meaning or meaning of unexpectedness as a type of conventional implicature rather than pragmatic inference. However, it can be identified as expressing maximality and exhaustivity in construction with wh-words and negation. In a nutshell, the concessive sense is taken as the default sense unless it is overridden by the interpretation of 都 to 1 as a maximality operator. I will examine the syntactic and semantic properties of 都 to 1 based on these texts. Its interpretation hinges on the structural position it occupies and the collocates it interacts with.

1 Introduction

As attested in early Southern Min texts,Footnote 1to 1 TO is mainly subject to two interpretations: (1) a concessive interpretation and (2) an interpretation of maximality and exhaustivity. The two interpretations are structure-driven. The concessive meaning is obtained by the conflict between the at-issue meaning of the sentence marked by 都 to 1 and the presupposition or inferred meaning driven by the linguistic context.Footnote 2 The universal and distributive meaning induced by 都 to 1 is only compatible with the plural meaning carried by noun phrases. Since the construction featuring 都 to 1 is ambiguous between concessive and distributive meaning, we need some linguistic cues to disambiguate it. Let us assume that the concessive meaning is the default, unless the noun phrase constrained by 都 to 1 is a determiner phrases with a plural quantifier. In this particular situation, 都 to 1 would take on the distributive function and become a universal operator. The concessive meaning would be overridden if there was no linguistic cue to bring about a conflict between presuppositions of the word meaning and the at-issue expression. Since 都 to 1 alone is potentially ambiguous between the concessive meaning and the interpretation of maximality and exhaustivity, my focus will be to pin down its exact meaning by examining the structural position it occupies in sentences. In addition to structural cues inside sentences, there is a need to draw on cross-sentential elements for pinning down the interpretation of 都 to 1. Our strategy is to first identify the sentence-internal elements that 都 to 1 co-occurs with. If there is a restrictor that 都 to 1 binds, then it takes on the function of a maximality operator (Giannakidou and Cheng 2006; Cheng 2009). If there is a contradiction or incompatibility between the at-issue expression and a certain assumption or the presupposition inferable from a preceding proposition over and above the sense of exhaustivity and maximality, then 都 to 1 will be construed as a modal particle featuring the concessive meaning. According to König (1988), concessive clauses involve incompatibility of two eventualities. Concessivity denotes conventional implicature arrived at based on the contradictory relation between the asserted part and presupposition in sentences.Footnote 3 Cross-linguistic evidence points to five types of origins of concessives: (1) elements expressing emotions such as spite, obstinacy, and contempt, (2) free-choice quantifiers, (3) temporal or conditional connectives, (4) elements emphasizing truth, and (5) elements implying the coexistence of two facts. 都 to 1 seems to belong to the second type of origin, as it features both a maximality operator and a concessive indicator. As will be argued in the following discussion, the interpretation of 都 to 1 will rest on the interface between sentence-internal and cross-sentential elements especially when the concessive sense is involved.

The structure of the paper is organized as follows. We will first explore the sequence of 都 to 1 + X (Section 2) where X may be the copula 是 si 7 (2.1), negatives (2.2), modals (2.3), agentive marker in passives (2.4), the aspect marker (2.5) or verbs (2.6). Then we will look into the sequence of X + 都 to 1 in Section 3 where X may be 盡 tsin 7 (3.1), 不 m 7 (3.2), or 是 si 7 (3.3).Footnote 4 Section 4 concludes the paper.

2 to 1 + X

The structural position that 都 to 1 occupies with respect to other functional words bears on the issue of its interpretation. 都 to 1 either (1) precedes or (2) follows other functional elements. Let’s consider the first 1 occurs before the focus marker 是 si 7, negative elements like 不 m 7, 無 bo 5, 未 be 7, and 袂 bue 7,Footnote 5 and modal verbs like 卜 berh 4, 著 tioh 8, 通 thang 1, 會 ue 7 , and 敢 kann 2. In this position, 都 to 1 is ambiguous between both the concessive and distributive interpretation.

2.1 to 1 + the focus marker 是 si 7

to 1 can be taken as an operator that binds its restrictor, whereas the copula 是 si 7 can be regarded as the focus marker that covers its nucleus scope based on the theory of tripartite structure (Partee 1991; Hajičová et al. 1998). While the nucleus scope always immediately follows the focus marker 是 si 7, the restrictor can precede or follow the maximality operator 都 to 1.Footnote 6 For example, the restrictor, indicated by being underlined below, precedes the operator in 1 and 2 and follows it in 3. The restrictor may remain implicit, as in 4.Footnote 7

2.1.1 都是 to 1 si 7

  1. (1)

    百般苦痛,都是為我 (26.074 嘉靖 Jiajing)

    i1__pah 4 puann 1__khoo 2 thiann 3__to1__si7__ui7__gua2


    The suffering she has gone through are all for me.

  2. (2)

    姻緣早晚,都是月老推付 (2.027 萬曆 Wanli)

    in 1 ien 5__tsa 2 buan 2__to1__si7__geh8lo2 __tui1hu3

    Predestined marriage sooner or later is arranged by the legendary go-between.

  3. (3)

    街上都是公子、王孫 (6.036 嘉靖Jiajing)

    kue1tsiunn7__to1__si7__kong 1 tsu 2__ong 5 sun 1


    Streets are all thronged with princelings and nobles.

  4. (4)

    我想起來都是五百年前註定 (9.017道光 Daoguang)


    I__think INCH__TO__COP__five__hundred__ago__predestine

    It dawns on me that it was predestined five hundred years ago.

2.2 to 1 + negatives

to 1 may occur before negatives such as 不 m 7 ‘not’, 未 ber 7 ‘not yet’, 無 bo 5 ‘not have’, and 袂 bue 7 ‘cannot, may not’. Let us take a short detour on diachronic aspects of negatives before embarking on discussing the collocation of 都 to 1 and the negatives. 不 m 7 is a negative element featuring a syllabic nasal, and the character 不 or its graph variant 伓 or 毌 adopted to represent the word is a semantic loan in Southern Min. In other words, m 7 and 不 *pK have their separate etymological origin. The pronunciation m 7 evolves from negatives featuring *m- initial, whereas不 comes from negatives featuring *p- initials shared by negatives such as 不 and 弗.Footnote 8 Depending on the kinds of predicates that it occurs with, 不 m 7 can function as a plain negative word or a negative coupled with a silent boulomaic modal, as in 伊不捌我 i 1 m 7 pat 4 gua 2 he not know I ‘he doesn’t know me’ and 伊不來 i 1 m 7 lai 5 he not come ‘He doesn’t want to come’. 未 ber 7 ‘not yet’ is a negative counterpart of the perfective aspect marker 了 liau 2 or the inchoative aspect marker 囉 loo 0 /啦 lah /啊 a0.Footnote 9bo 4 is indisputably taken to be a fusion of a negative m- and the verb 有 you 3 ‘have’. Old Chinese as a synthetic language boasts negatives with rich lexicalized sense of grammatical categories such as mood, conditionality, and aspect, among others. 無 bo 4 may be ambiguous between negation of existence and negation of deontic modality in Old Chinese. (Dobson 1966; Pulleyblank 1995: 84) 無 bo 4 in modern Southern Min features only negation of existence, but not the function of deontic modality, even though it develops a newly rising pluractional function, as in 伊有食薰 i 1 u 7 tsiah 8 hun 1 he have eat cigarette ‘he smokes’ taking on a habitual sense.Footnote 10bue 7 ‘cannot’, a phonetic loan character, is also a fusional word comprising a negative m- and a modal 解 ue 7 (often written as 會, a semantic loan character) ‘can’ denoting capability or possibility, as in 狗解吠 kau 2 ue 7 pui 7 dog can bark ‘Dogs can bark’ and 明仔載解落雨 bin 5 -a 2 -tsai 7 ue 7 loh 8 hoo 7 tomorrow can fall rain ‘It will rain tomorrow’.

Let us turn now to the collocation of 都 to 1 and negatives. 都 to 1 in this configuration carries a concessive meaning in that 都 to 1 as a modal particle can be used to mark the following at-issue expression in its scope which conflicts with the presupposition as the inferred meaning from what is expressed in a previous utterance.

2.2.1 都不 to1 m7

Take 都 to 1 followed by 不 m 7. For example, example 5 with the use of 都 to 1 yields the conventional implicature that the speaker dares not to make complaints, which is in conflict with what one is expected to do.

  1. (5)

    辛苦在心內,都不敢呾 (22.07 嘉靖 Jiajing)



    Even though hardship is in my mind, I dare not air it.

m 7 when immediately followed by activity verbs like 來 lai 5 ‘come’ will evoke a silent volitional modal WANT, as in 6.

  1. (6)

    伊心向痛,都不來見阮 (27.032萬曆 Wanli)


    he__heart__so__ache __TO__not__come__meet__I.PL

    Even though he is sad, he won’t come to see me

The meaning of WANT as a silent element is preempted by another modal like 願 guan 7 ‘be willing’ (7), or 畏 ui 3 ‘fear’ (8, 9) as well as 敢 kann 2 ‘dare’ (5) occupying the same position. 都 to 1 immediately followed by 不 m 7 is however construed to be preceded by a null subject recoverable based on the exophoric or endophoric contexts, as in 8 and 9. We can see that the inferred meaning from the expression before the comma is in conflict with the at-issue clause headed by 都 to 1 in 7, 8, and 9.

  1. (7)

    多少郎君卜求,我只心中都不願 (11.017 嘉靖 Jiajing)



    Though there are so many suitors, I will not accept their proposal.

  1. (8)

    潑伊人,都不畏了冷伊人 (22.111 嘉靖 Jiajing)



    Though I splash water on him, I don’t fear that it will make him catch cold.

  1. (9)

    瓜田李下,都不畏人疑 (30.065 道光 Daoguang)

    kua1__tien5__li2__ha7__to1__m7__ui3__lang5 __gi5


    Even though you are in the melon patch and under the plum tree, you do not fear being open to suspicion’

Following the suggestions made by an anonymous reviewer, we can motivate the concessive meaning conveyed by 都 to 1 as a kind of conventional implicature rather than conversational implicature. The concessive meaning conveyed by 都 to 1 in each of the above complex sentences is derivable due to the existence of background sentence. The meaning of the main clause in each of the above sentences is in conflict with the inference obtained from the antecedent clause featuring a silent concessive marker such as 著算 tioh 8 -sng 3 right count ‘even though’. For example, splashing water on someone will cause him to feel cold, as in 8. This is expressed in the subordinate clause, and its meaning conflicts with what is expressed in the main clause. The apparent noun phrase such as 瓜田李下 kua 1 tien 5 li 2 ha 7 in the melon plots and under the plum tree, a juxtaposition of two noun phrases, as in 9, may stand for a whole proposition when the missing predicate is restored.

2.2.2 都無 to1 bo5

to 1 followed by 無 bo 5 may yield a concessive meaning. As shown in 10, it is commonsensical that one will be displeased when scolded, but the sentence implies that it does not matter. When a man is getting old, he is expected to marry a woman. Contrary to this expectation, the speaker (v.i., 小七 sio 2 tshit 4 little seven ‘Little Seven’) has no wife at the mature and marriageable age, let alone at the old age, as in 11. Example 12 shows that aside from the concessive meaning, there is in addition a scalar sense in that there is a comparison between the lower and high caliber in a scale. That is, men’s capacity is higher than women’s capacity in the scale of decision-making. All of these sentences involve the scalar model where the two elements on a certain scale are compared. We can add, as insightfully suggested by an anonymous reviewer, the marker 連 lien 5 ‘even’ in the slot to signal the underlined explicit element being compared. There is an explicit underlined element compared with an implicit element in 10, and 11, whereas both compared elements are present in 12, namely the contrast between men and women in the capacity of offering an idea.

  1. (10)

    []罵簡都無打緊 (11.27 順治 Shunzhi)



    Even (you) scolding me will be okay.

  1. (11)

    小七[]老了都無厶 (30.124 順治 Shunzhi)



    I (Little Seven) am getting on in age, and yet I have no wife.

  1. (12)

    []恁乾埔人都無主意,阮姿娘人有乜主意? (34.009 道光 Daoguang)

    lin2__ta1poo1__lang5__to1 __bo5__tsu2i3__gun2__tsu1niu5__lang5__u7__mih4__tsu2i3


    You men don’t even have any idea, let alone we women, what idea can we women have?

Thus, both 都不 to 1 m 7 and 都無 to 1 bo 5 express a concessive meaning, but they differ at least on two counts. First, 都無 to 1 bo 5 rather than 都不 to 1 m 7 involves a scalar model. Second, the restrictor of 都 to 1 is the predicate that it precedes in the first case, whereas its restrictor in the latter case is the underlined element in the subject position.

2.2.3 都未 to 7 be 7

to 7 followed by 未 be 7 brings about a sense of unexpectedness. For example, it is a common practice to eat when one gets up in the morning, but 13 is just the opposite, viz., contrary to his routine practice, the speaker has not had his breakfast yet. Likewise, as shown in 14, 五娘 wu 3 niang 2 five lady ‘Wuniang’ (the female lead) expects 陳三chen 2 san 1 Chen three ‘Chen San’ (the male lead) to show concern for her, and yet he does not do so, contrary to her expectation.

  1. (13)

    我早起都未食 (22.371 嘉靖 Jiajing)



    I haven’t even eaten yet since I got up in the morning.

  1. (14)

    三哥自來寒舍都未八相動問 (23.317 光緒 Guangxu)



    You (i.e.,Third Brother) haven’t even asked after me yet since you came here.

2.2.4 都袂 to1 bue7

to 1 followed by 袂 bue 7 yields a concessive meaning. In a traditional male chauvinistic society, a man can scold his wife with impunity. Thus, what the speaker is driving at in 15 with the particle 都 to 1 is that the speaker cannot scold his wife given that he is the addressee’s wife, and this is in conflict with the assumed male chauvinistic belief. When the second clause is taken as a rhetoric question, the ultimate interpretation is that he cannot scold his wife. Likewise, in 16 the addressee (Chen San)’s reluctance to talk to the speaker (益春 yi 4 chun 1 benefit spring ‘Yichun’, the maid) runs counter to her expectation. In 17, the speaker (Yichun again) complains to Chen San that once his match with the lady Wuniang is secured he will forget all her contributions. This does not jibe well with her expectations. Example 18 shows that the tree even with its huge size cannot withstand the gold hatchet; similarly humans cannot either.

  1. (15)

    我是你翁仔,都袂罵得你? (11.65 順治 Shunzhi)



    I am your husband; couldn’t I even scold you?

  1. (16)

    你都袂割捨得共小妹呾一聲 (26.007 道光 Daoguang)



    You are even reluctant to talk to me.

  1. (17)

    姻緣成就,許時都袂記得小妹 (21.039 道光 Daoguang)



    If the marriage is fulfilled, you will not even remember me then.

  1. (18)

    大樹都袂受金刀斧,人都袂受得人千聲嘮嘈。 (21.038 道光 Daoguang)

    tua7 tshiu7__to1__bue7__siu7__kim1__to1poo2__lang5__to1__bue7__siu7__tit4__lang5 tshien1siann1__lo1tso1



    Even huge trees cannot bear the gold hatchet. People cannot even tolerate the sheer human din.

2.3 to 1 + modals

to 1 may occur before a range of modals such as the volitional modal 卜 berh 4 ‘want’, deontic modals like 著 tioh 8 ‘must’ and 通 thang 1 ‘can, may’, and dynamic modals like 會 ue 7 ‘can, be able to’ and 敢 kann 2 ‘dare’. 卜 berh 4 ( cf. 要 yao 4 ‘want’ M.) might be a modal verb indigenous to Min; seeing as its etymological origin is unknown, and the character used to identify it, 卜, was adopted based on its sound rather than meaning. 著 tioh 8 (cf. 得 dei 3 or 應該 ying 1 gai 1 ‘should’ M.) may be a result of grammaticalizing the causative verb 著 tioh 8 ‘have, ask, get (sb. to do sth.)’ (Hsu 2013). Whereas ue 7 written as a semantic loan character 會 (cf. 能 neng 2 ‘can’ or 會 hui 4 ‘can/may’ M.) can be traced back to 解 xie 4 ‘know, understand’ with confidence, the etymological origin of 通 thang 1 (cf. 可以 ke 3 yi 3 ‘can/be allowed to’ M.) remains still unsettled. There is no question about the etymology of 敢 kann 2.

All examples featuring this constructional type will take on the default concessive sense unless there are quantifier noun phrases as the restrictors bound by the maximality operator 都 to 1.

2.3.1 都卜 to 1 berh 4

Examples 19, 20, and 21 with the particle 都 to 1 coupled with the volitional modal卜berh 4 ‘want’ all show the stubborn propensity or behavior on the part of the subject in resistance to some common code of conduct or social mores. Thus, all the three examples show the unexpected behavior of the protagonist as well as a concessive sense. Such an interpretation is justified against the backdrop of the knowledge background provided for each case.

Chen San once eloped with his mistress Wuniang and her maid Yichun. Yichun accidentally tripped on the way. With no one lending her a hand she refused to go along. Wuniang uttered 19 to reproach her.

  1. (19)

    賊婢都卜來賴人 (29.074 光緒 Guangxu)



    You wretched maid nevertheless have the nerve to blame others.

Chen San asked the maid Yichun to persuade her mistress Wuniang to stop crying. The maid replied, uttering 20 with the hint that despite her pleading to the contrary, her mistress still cannot help crying.

  1. (20)

    伊都卜啼,罔乞伊啼 (34.067 道光 Daoguang)

    i1 to1 berh4 thi5 bong2 khit4 i1 thi5


    She wants to weep. We might just as well let her cry.

It is a routine practice befitting a decent lady to comb her hair every morning. Stubborn as she is, Wuniang is reluctant to do it even at the risk of being found out by her parents.

  1. (21)

    亞娘都卜不梳,亞公亞媽知了,不句愛罵 (22.003 道光 Daoguang)

    a1niu5__to1__berh4 __m7__sue1__a1kong1__a1ma2__tsai1__liau2__m7__ku3__ai3__ma7


    If you are not willing to have your hair combed, won’t grandpa and grandma scold you again when they get to know it?

2.3.2 都著 to 1 tioh 8

to 1, when followed by the deontic modal 著 tioh 8 ‘should’, may take on the function of the maximality operator, as in 22 with the quantifier noun phrase, or involve the comparison of two loci in a scale, as in 23.

  1. (22)

    三人手都著搥 (23.115 光緒 Guangxu)



    All three persons’ hands should be pounded.

  1. (23)

    我都著叫伊官,你夭不叫? (31.050 道光 Daoguang)k

    Footnote 11Immediately after 官 kuann 1 ‘official’ there may be a missing morpheme 儂 lang 5 ‘person, humans’. So 官儂 kuann 1 lang 5 means ‘master’ as opposed to 官儂 kuann 1 lang 0 ‘Mandarin’



    Even if I call him master. How can you not call (him)?

2.3.3 都會 to 1 e 7

to 1 combined with the dynamic modal 會 ue 7 ‘can’ is conducive to the quantificational interpretation of exhaustivity and maximality given the presence of the noun phrase denoting a list of things, as in 24, and the implicit universal time adverb such as 逐擺 tak 8 -pai 2 every time ‘every time, whenever’, as in 25.

  1. (24)

    琴、棋、書、畫我都會 (25.089 嘉靖 Jiajing)


    zither__chess__calligraphy__painting__I__TO__be.skilfful at

    I excel at zither, chess, calligraphy and painting.

  1. (25)

    恁師父磨鏡,都會唱歌 (9.136 順治 Shunzhi)



    Whenever your master polished mirrors, he would sing.

One may wonder whether the absence of a concessive meaning of 都會 to 1 e 7 may have to do with the sense of 會 ue 7 ‘can’. What we can observe at the present stage is that the concessive and quantitative senses of 都 to 1 are in competition. Whenever no inferential meaning is obtainable to be in conflict with the meaning of the main clause, then another meaning of 都 to 1 emerges. So its quantitative interpretation comes about by default.

2.3.4 都通 to 1 tang 1

to 1 in construction with the deontic modal 通 tang 1 ‘can, may’ yield the concessive meaning, as in 26 and 27. The male lead Chen San indented himself as a long-termed hired hand to redeem the cost of purposefully breaking a mirror. As a male servant, he still thinks that Yichun, the maid, does not enjoy the equal social status, as he is from a family of officials. In uttering 26, Chen San is disapproving of Yichun calling out his name. Both Chen San and Little Seven are male servants. They vie for currying favor with their master. So Little Seven teases at his workmate uttering 27. In his mind, both he and Chen San are equals. They are both in a scalar model competing with each other. Little Seven believes that if the addressee (viz. Chen San), as a less likely candidate for a certain task, can be allowed to do something, he surely can get the permission to do so.

  1. (26)

    只一簡仔,都通叫我名 (28.115 嘉靖 Jiajing)



    This maid can even call my name.

  1. (27)

    你都通磨鏡,我不通補鼎? (19.221 道光 Daoguang)



    If you can even polish mirrors, can’t I mend the wok?

2.3.5 都敢 to 1 kann 2

The same can be said of 28, 29, and 30 where the sequence of 都敢 to 1 kann 2 TO dare ‘even dare’ takes on the concessive meaning. As shown in 28, scolding is less desirable and therefore more unexpected than name calling in the scale of interpersonal verbal interaction. This scalar discrepancy justifies the use of 都 to 1. As shown in 28, Wuniang is indignant at Chen San (the male servant, a socially inferior person) taking the liberty of scolding her, an unexpected breach of the code of conduct. As in 28, the use of 都 to 1 is motivated in 30 since beating is less expected than scolding.

  1. (28)

    叫你名啞,你無正經,罵你都敢。 (17.1636-64 順治 Shunzhi)



    I call your name. Since you are not decent, I can even scold you without a qualm.

  1. (29)

    只賊奴乜大膽都敢罵我 (26.163 嘉靖 Jiajing)

    tsi2__tshat8loo5__mih4__tua7tann2__to1 kann2__ma7__gua2


    The rascal is so bold and even dares to scold me.

  1. (30)

    罵啞?打都敢 (4.05 順治 Shunzhi)



    Scold you? I even dare beat you.

2.4 to 1 + the agentive marker 乞 khit 4 in passives

khit 4 is a preposition heading a preposition (DP) in passives. It occurs both after 都 to 1 and the subject. We can see that 都 to 1 occupies a syntactic position higher than voice. It conveys a sense of unexpectedness over the propositional content denoted by the sentence, as in 31, 32, and 33.

  1. (31)

    亞娘,咱只處呾話,都乞陳三聽見勞 (22.081 道光 Daoguang)



    Our talks here have been overheard by Chen San, Madam.

  1. (32)

    汝心都乞狗食去 (8.014 光緒 Guangxu)



    Your heart has been eaten by dogs.

  1. (33)

    亞娘都乞小八攃去到泉州了 (32.055 光緒 Guangxu)



    The lady has been taken by Little Eight to Quanzhou.

Following Bowers (2010), I postulate a layer of VoiceP (VoiP for short) to account for the presence of the agent marker in passives. VoiP has the feature of [+/− act]. If the feature is positive, the agent (ag) selects DP in an active sentence. If the feature is negative, the ag selects a prepositional phrase headed by the preposition 乞 khit 4 in a passive.Footnote 12

2.5 to 1 + the aspect marker 在只 ti 7 tsi 2

to 1 occurs before a locative phrase 在只 ti 7 tsi 2 at-here ‘here’, which later developed into the progressive marker. Example 34 takes on the interpretation of maximality and exhaustiveness.

2.5.1 都在只 to 1 ti 7 tsi 2

  1. (34)

    簡都在只聽候啞娘 (15.095 順治 Shunzhi)

    kan1__to1__ti7 __tsi2 __thiann3hau7__a1-niu5


    I have all along been here waiting for you.

2.6 to 1 + verbs

to 1 may be immediately followed by verbs without being mediated by modals or other elements. Since verbs are an open class grammatical category, I will only focus on 有 u 7 ‘have’ as an example. The presence of 都 to 1 in 35 as a yes-no question casts into doubt the very proposition that men have periods, a totally impossible situation. Example 36 occurs in a situation in which Wuniang reminds Chen San that the lychee as a token of love is still in his place in response to his request for a keepsake. Chen San utters 37 complaining that Wuniang would not hold his hand. Yichun, the maid, says 38 calling her lady Wuniang’s attention to the situation that someone made a sound outside the wall. All these examples exhibit a concessive meaning and some measure of unexpectedness.

2.6.1 都有 to 1 u 7

  1. (35)

    乾埔人都有月經? (5.099 道光 Daoguang)

    ta1poo1__lang5__to1__u7 __geh8king1


    Do men even have periods?

  1. (36)

    阮都有荔枝值恁處 (23.400/401光緒 Guangxu)

    gun2__to1__u7__nai7tsi 1__ti7__lin2__ter3


    Why is the lychee that belongs to me at your place.

  1. (37)

    爾都有手不來牽 (23.303 光緒 Guangxu)



    You have hands, and you don’t even want to hold me.

  1. (38)

    牆外都有人做聲 (13.037 順治 Shunzhi)



    There is someone making a sound outside the wall, though.

All the cases of 都 to 1 featuring the sense of concession, preceding negatives (2.2), modals (2.3), the VoiP (2.4), the aspect marker (2.5) or verbs (2.6), occupy the specifier of the concessive project in complementizer CP in line with Rizzi’s fine structure of the left periphery. Negation, modality, voice, and aspect are the functional projection in TP.

3 X + 都 to 1

The second type of the structural position of 都 to 1 is its occurrence after other functional elements such as 盡 tsin 7, 不 m 7, and 是 si 1. Various types of constructions result from such combinations. 都 to 1, along with a preceding synonym, 盡 tsin 7, yields a compound maximality operator. When preceded by 不 m 7, 都 to 1 occurs in a tag question. 都 to 1 functions as an intensifier when followed by the focus marker 是 si 1.

3.1 盡都 tsin 7 to 1

盡都 tsin 7 to 1 entirely TO ‘all’ is a case of juxtaposed synonyms or rather hypercharacterization.Footnote 13 The compound is good for disambiguating the potential uncertain meaning of the monosyllable 都 to 1 and narrowing it down to the function of a maximality operator. There is either a quantifier noun phrase like 百姓人 peh 4 -sinn 3 lang 5 hundred name person ‘the folks’, as in 39, or a plural determiner noun phrase like 許賞燈人 hu 2 siunn 2 ting 1 lang 5 that enjoy lantern person ‘those lantern-viewing people’, as in 40, which serves as a restrictor bound by 都 to 1.

  1. (39)

    百姓人盡都歡喜 (38.015嘉靖 Jiajing)



    The folks are all in a merry mood.

  1. (40)

    看許賞燈人盡都歡喜 (7.029嘉靖 Jiajing)



    We can see those lantern-viewing throngs are all quite in a merry mood.

The juxtaposition of 盡都 tsin 7 to 1 is useful to disambiguate 都 to 1 and narrows down its meaning to ‘all’, as in 伊人琴棋書畫盡都曉通 i 1 lang 5 khim 5 ki 5 su 1 ui 7 tsin 7 to 1 hiau 2 thong 1 he person chess zither calligraphy painting exhaustively TO know through ‘He excels at all of the skills of music, chess, calligraphy and painting’ (26.010 嘉靖 Jiajing) where 盡都 tsin 7 to 1 functions as a maximality operator. Still another way to narrow down its meaning is for 都 to 1 to co-occur with a wh-word, as in 趁我興, 乜都度你 than 3 gua 2 hing 3 mih 4 to 1 thoo 7 li 2 follow I fancy what TO give you ‘Everything will be given to you if you suit my fancy’ where the universal quantifier ‘whatever, everything’ is built on the combination of the wh-word 乜 mih 4 ‘what’ and the operator 都 to 1 (16.061 順治 Shunzhi).

3.2 不都 m 7 to 1

According to the functional architecture of the languages proposed in Rizzi (1997, 2004), there are three layers: the VP layer (viz. the lexical layer headed by the verb), the IP (or rather TP) layer (viz. inflectional layers headed by functional categories such as agreement, modality, voice, tense, aspect, among others), and the CP layer (the complementizer layer hosting kinds of force such as interrogative, imperative, and exclamative mood).Footnote 14 There may be recursive topics but only a single focus between Force P and FinP. Later version of this approach entertains the idea of split-up Force. The function of 不 m 7 in the construction featuring 不都 m 7 to 1 not TO can be given an explicit and motivated account in terms of this theoretical framework.

When the negative word 不 m 7 occurs before 都 to 1, the sentence changes its mood from the indicative to the interrogative mood, realized as a tag question in this construction. 不 m 7 is higher than 都 to 1 in the hierarchy of functional categories as proposed in Rizzi (1997, 2004). That may be the reason why the sentence takes on the interrogative mood. 不都 m 7 to 1 seems to be similar to a negative tag question ‘isn’t it?, right?’, as in 41, each of which is interpretable given the cross-sentential elements as shown in previous or following sentences. What follows are three dialogs: (a) 41a–b, (b) 42a–b, and (c) 43a–d. This clause-initial 不 m 7 is on a par with one type of the tag question particle eh in Canadian English.Footnote 15 The common implication of these four dialogs conveyed by 都 to 1 involves the subjectivity on the speaker concerning the contradiction of two propositions.

In the first dialog, most probably Chen San hopes that his singing will be listened to but worries that there will be no one listening. He uses 41 as a tag question to convey his worry in the hope that the addressee will confirm it.

  1. (41)

    a. 生:不都無人聽 (9.167 順治 Shunzhi)

    Sing1: m7__to1__bo5__lang5__thiann1

    ML: not__TO__not.have__person__listen

    Male lead: Nobody will listen, right?

    b. 貼:唱那好,伊便有人聽 (9.168 順治 Shunzhi)

    Thiap4: tshiunn3__na7__ho2__i1__pien7__u7__lang5__thiann1

    SFC: sing__if__good__he__then__have__person__listen

    If you sing well, you will be listened to.

In the second dialog, the speaker takes for granted the proposition that the landlord has returned from a trip to collect rents, a proposition contradicting the latter’s pretension that he has not returned yet. He uses 42b as a tag question to secure confirmation from the landlord for his supposition.

  1. (42)

    a. 外:你去共伊說,叫: 我收租未返來 (37.025 道光 Daoguang)

    Footnote 16kio 3 is a quotative marker showing what follows is a direct quote. See Lien (2013) This quotative function of 叫 kio 3 is to some extent similar to like in English, except that the former only occurs alone immediately preceding the quoted speech. See, for example, Barbieri (2005) for the use of like as a colloquial quotative marker.

    Gua7: Lir2__khi3__kang1__i1__serh4__kio3__gua2__siu1__tsoo1__ber7__tng2__lai5


    Go and tell him, “ I have not returned from a trip to collect rents.”

    b. 淨:你不都返來勞? (37.026 道光 Daoguang)Footnote 17

    Loo 0勞 as well as liau 0了as a SFP denotes inchoative aspect.

    Tsing7: Lir2__m7__to1__tng2__lai5__loo0

    COM: you__not__TO__return__come__SFP

    Haven’t you return home?

In the third dialog, as in 43a–d, a buffoon (BFN) asks the maid Yichun whether Emperor Ming of Tang is a man or woman. She replies, ‘The emperor is a man. Him a woman?’ The original sentence for Him a woman? is 乜孜娘人? mih 4 tsu 1 niu 5 lang 5 what woman person, where 乜 mih 4 what is a wh_word functioning as a rhetoric marker. The sentence, therefore, can be taken to mean, what reasons can there be for the emperor to be a woman? The inference will be that surely enough, the emperor is not a woman. Based on the textual evidence of other versions of the Romance of the Lychee Mirror, such as in the 道光 Daoguang and 光緒 Guangxu editions, the missing link for 不都有月宮 pu 4 dou 1 you 3 yue 4 gong 1 not TO have moon.palace ‘Isn’t there Moon Palace?’ is the given fact that the emperor is a man. So we can rephrase it as follows: Granting that the emperor is a man, there should be a moon palace, right? This reading is made possible or plausible by the legend of the Emperor Ming of Tang’s visit to the moon palace, a common shared knowledge among the Chinese. 不都有月宮 can also be taken as a pun exploiting the homophony of 月宮 geh 8 -king 1 Moon Palace ‘Moon Palace’ and 月經 geh 8 -king 1 moon menstruation ‘periods’. Again, this interpretation is corroborated by the appearance of月經 geh 8 -king 1 in lieu of 月宮 geh 8 -king 1 in parallel passages in other versions. It befits the role of the BFN to make such a jocular utterance with double entendre. On the one hand, it is perfectly logical that the legend of the Emperor Ming of Tang is associated with the moon palace. On the other hand, one runs into an absurd assumption: 不都有月經? ‘Should the emperor be a male, could he be expected to have periods, eh?’ The speaker is concerned with the confirmation of his assumption that can be elicited from his addressee.

  1. (43)

    a. 丑:唐明皇是乾埔人啞是孜娘人? (4.392-3順治 Shunzhi)rFootnote 18

    ta 1 ‘dry’ in 43b is adopted in place of 幹 kan 3 ‘do’, an obvious typo, in the text. 乾 is admittedly not an etymologically viable character, in that 乾 kuann 1, as in 牛肉乾 gu 5 bah 4 kuann 1 cattle meat dry ‘beef jerky’, is a semantic loan character for 焦 ta 1 ‘dry’ in Southern Min. But the sense of dry is totally unrelated to the sense in the disyllabic word 乾埔 ta 1 poo 1 dry plain ‘male (person)’.

    Thiu2: tong5bing5__hong5__si7__ta1poo1__ang5__a7 si7__tsu1niu5__lang5

    BFN: Tang.ming__emperor __COP__man__person__or__woman__person

    Is Emperor Ming of Tang a man or a woman?

    b. 貼:唐明皇是乾埔人¸乜孜娘人? (4.394-5順治 Shunzhi)

    Thiap4: tong5bing5__hong5__si7__ta1poo1lang5__mih4__tsu1niu5__lang5

    SFC: Tang.ming__emperor__COP__male__person__what__woman__person

    Emperor Ming of Tang is a man. Is he a woman?

    c. 丑:不都有月宮 (4.396 順治 Shunzhi)

    Thiu2: m7__to1__u7__geh4king1

    BFN: not__TO__have__moon.palace/period

    There is a moon palace (periods), eh?

    d. 旦:正是月宮 (4.397 順治 Shunzhi)

    Tuann3:tsiann3 si7 geh4-king1

    FM: exactly__COP__moon.palace

    Exactly. There is a moon palace.

不都 m 7 to 1 in this section merits a comparison with 都不 to 1 m 7 in Section 2.2. The interpretation of each sequence can be arrived at by putting together the pieces and following the manner in which they are built into a whole. The former functions in a tag question, whereas the latter participates in a concessive sentence. The occurrences of 都 to 1 in both sequences are related. 都 to 1 in 不都 m 7 to 1 still retains its sense of exhaustivity and maximality in that 都 to 1 in 41a evokes the interpretation that the singer is worried that whenever he sings nobody will listen. He relies on 不 m 7 functioning as a tag question particle to ask his addressee to tell him whether his worry is valid. The latter gives a positive response to dispel his worry. Likewise, an interpretation prompted by 都 to 1 emerges in 43c that whenever Emperor Ming of Tang is mentioned, it inevitably conjures up an image of his visitation to the moon palace. The negative element 不 m 7 heading the clause functions as a tag question particle and is used by the speaker to seek confirmation of his supposition with the addressee. As shown in 43d, the addressee gives an emphatic positive answer where the copula 是 si 7 is preceded by tsiann 3 正 ‘exactly’.

3.3 是都 si7 to1

As shown in 44, 45, and 46, 是 si 7, when coupled with 都 to 1, functions as a focus marker making salient the truth of what has been stated in a previous sentence. The function of intensification of 都 to 1 seems to be derived from its original function of maximality and exhaustivity.

  1. (44)

    a. 貼:只牆外都親像乜人做聲 (24.079嘉靖 Jiajing)


    SFC: this__wall__outside__TO__look.like__what__person__make__sound

    Someone seems to make a sound outside the wall.

    b. 旦:益春,是都親像人做聲 (24.081嘉靖 Jiajing)

    Tuann3: Iah4 tshun1, si7 to1 tshin1-tshiunn7 lang5 tsoh4 siann1

    FL: Yichun__COP__TO__look.like__person__make__sound

    Indeed someone seems to make a sound, Yichun.

  1. (45)

    是都聽見了 (24.272嘉靖 Jiajing)



    You indeed heard it.

  1. (46)

    是都不打婆仔了 (5.343順治 Shunzhi)



    You sure don’t want to beat me.

4 Conclusions

In this paper, I propose that 都 to 1 as attested in earlier Southern Min texts features two basic functions, either as a maximality operator or a modal particle carrying concessive meaning or both. The maximality operator 都 to 1 binds its restrictor vis-à-vis the nucleus scope. The concessive meaning of the modal particle 都 to 1 is arrived at by asserting an at-issue proposition induced by the particle 都 to 1 in conflict with a presupposition. The sense of unexpectedness conveys the surprise on the part of the speaker at the occurrence of a situation. It can be viewed as resulting from incompatibility of two eventualities or a comparison of two propositions or two properties in a scalar situation. This is a case of conventional implicature rather than pragmatic inference. The paper also shows the relationship that the semantic functions of 都 to 1 have with its structural position in the hierarchy of functional categories and tries to pin down the theoretical importance of such a relation (Cinque 1999). I take the concessive meaning as the default interpretation in the sentences featuring 都 to 1, unless this interpretation is overridden by the interpretation of 都 to 1 as a maximality operator. The latter interpretation is made possible by the presence of the quantifier noun phrase that 都 to 1 binds as its restrictor vis-à-vis the nucleus scope, and no further additional conventional implicature such as concessivity is involved. A point worth mentioning is that the concessivity of 都 to 1 takes on the default function due to its robust existence despite its being the further development of the original sense of maximality and exhaustivity.

5 Endnotes


  1. Earlier Southern Min texts comprise five editions of playscript featuring the Romance of the Lychee Mirror (Wu, 2001a, b, c, d; Quanzhoushi 2010). Time frame of each edition is 嘉靖 Jiajing (1522–1566), 萬曆 Wanli (1573–1619), 順治 Shunzhi (1644–1661), 道光 Daoguang (1821–1850), and 光緒 Guangxu (1875–1908).

  2. I borrow the term at-issue meaning from Potts (2012). Unlike the conventional implicature, the at-issue meaning involves the current content that the speaker takes as primary in interaction with his addressee before it is accepted into the common ground.

  3. For the notion of conventional implicature see Karttunen and Peters (1979) and Potts (2012).

  4. The spelling of Southern Min in this paper is based on the Church Romanization given in Douglas (1873). Some modifications have been made. In particular, the diacritic tone marks have been replaced by numerical superscripts. No distinction is made between ch and ts or chh and tsh as they do not involve phonemic contrast. Open o and closed o are rewritten as oo and o, as in 古 koo 2 ‘old’ and 稿 ko 2 ‘draft’. Nasalization of vowels is signaled by a double n, as in 天 thinn 1 ‘sky’. // and /^/ are rendered as /er/ and /ir/ respectively.

  5. bue 7 is a fusional word of two morphemes: 毋 m 7 + 解 ue 7. Its positive counterpart is解 ue 7 (may, can), which is often rendered as 會, a loan character of 解. Note that the nasal onset *m- has to be denasalized when the final is not nasalized in modern Southern Min.

  6. The term maximality operator encompassing both maximality and exhaustivity is due to Giannakidou and Cheng (2006) and Cheng (2009). It was previously termed generalized distributive operator (Lin 1998). We can see that the function of quantification 都 to 1 in Southern Min is quite similar to 都 dou 1 in Mandarin in this respect.

  7. Abbreviations used in the paper: ACC accusative marker, BFN buffoon, male or female, CLF classifier, COM comic character, male or female, COP copula, EXCL exclusive, EXP experiential aspect Marker, EXT extent marker, FL female lead, INCH inchoative aspect marker, INCL inclusive, M Mandarin, ML male lead, MOD modal, PERF perfective aspect marker, PL plural, SECMC secondary male character, SFC secondary Female character, SFP sentence-final particle, SUBMC subsidiary male character, and SUF suffix The rendition of the names of various roles in the play is due to van der Loon (1992: 38).

  8. fou 3 and 不 bu 4 in modern Mandarin come from 不 *pK, a departing tone word, and 弗 *pKt, an entering tone word, in Old Chinese, respectively. See Yang (1971) for the insight on the etymological origin of 否 fou 3 (<不 *pK) and 不 bu 4 (<*pKt 弗) which is in turn due to Fangkuei Li. Note that Old Chinese reconstruction is based on Baxter and Sagart (2014: 330). See Norman (1995) for the idea that negatives with *p- initial and negative *m- initial are an important grammatical criterion on which northern and southern Chinese dialects can be distinguished. Note that apart from m 7, put 4 often written as 不 (<*pKt 弗) occurs only in the literary stratum in Southern Min, as in 伊真不孝 i 1 tsin 1 put 4 hau 3 he very not dutiful ‘He is not dutiful (to his parents)’. There is no reflex of不 *pK in the colloquial and indigenous layer of modern Southern Min, whereas it has developed into 否 fou 2 in modern literary Chinese. Admittedly, there are problems of textual mix-up of the reflexes of 不 *pK and 弗 *pKt due to obliteration of the latter character perhaps prompted by taboo considerations.

  9. Here, silent is taken in Kaynean sense (Kayne 2005). As I understand it, uncovering silent elements from a comparative perspective is an important means of establishing language universal patterns.

  10. It is noteworthy that 沒 mei 2 resulting from the convergence of 無 and 未 in Mandarin is ambiguous between negation of existence and negation of perfectives or inchoatives, as in 他沒腳 ta 1 mei 2 jiao 3 he not.have leg ‘he has no legs’ and 他沒繳 he not.yet pay ‘he has not paid (e.g., taxes) yet’. The two interpretations are realized as 伊無骹 i 1 bo 5 kha 1 he not.have leg and 伊 (猶) 未交 i 1 (iah 4 ) ber 7 kau 1 he not.yet pay in Southern Min.

  11. Immediately after 官 kuann 1 ‘official,’ there may be a missing morpheme 儂 lang 5 ‘person, humans’. So 官儂 kuann 1 lang 5 means ‘master’ as opposed to 官儂 kuann 1 lang 0 ‘Mandarin’

  12. Bowers (2010) posits three primary argument types, Ag(ent), Th(eme), and aff(ectee). An innovative element of his approach is that Ag merges first with a root verb in actives or passives, and the Ag occupies the lowest hierarchical position. VoiP is sandwiched by T, PrP, and various argument types. PrP or T and VoiP bearing EPP feature are responsible for assigning the structural cases, viz. nominative and accusative case, respectively. An exploration of passives in Southern Min along these lines will be quite fruitful. However, I will leave it for future endeavor, as it will go far afield from the present concern.

  13. Hypercharacterization involves co-occurrence of two synonyms redundant to each other. (Malkiel 1957; Lehmann 2005) The excessive morphological marking has the effect of mutual reinforcing. It bears on the notion of strengthening of informativeness (Traugott and König 1991) as well as layering and persistence as phenomena of grammaticalization (Hopper 1991).

  14. Cf. Wiltschko’s Universal Spine hypothesis where layers such as CP (linking), IP (anchoring), AspP (point of view), and vP (classification) are distinguished. (Wiltschko 2014: 75)

  15. See Avis (1972) for the function of the sentence-final particle eh? in Canadian English and Tubau (2014) for discussion of the confirmatory pragmatic particle innit in English. See also Columbus (2010) for various functions including confirmation check carried by invariant tags such as eh, yeah, no, and na in three varieties of English, viz. New Zealand, Indian, and British English.

  16. kio 3 is a quotative marker showing what follows is a direct quote (see Lien 2013). This quotative function of 叫 kio 3 is to some extent similar to like in English, except that the former only occurs alone immediately preceding the quoted speech. See, for example, Barbieri (2005) for the use of like as a colloquial quotative marker.

  17. Loo 0勞 as well as liau 0了as a SFP denotes inchoative aspect.

  18. ta 1 ‘dry’ in 43b is adopted in place of 幹 kan 3 ‘do,’ an obvious typo, in the text. 乾 is admittedly not an etymologically viable character, in that 乾 kuann 1, as in 牛肉乾 gu 5 bah 4 kuann 1 cattle meat dry ‘beef jerky’, is a semantic loan character for 焦 ta 1 ‘dry’ in Southern Min. But the sense of dry is totally unrelated to the sense in the disyllabic word 乾埔 ta 1 poo 1 dry plain ‘male (person)’.


  • Avis, Walter S. 1972. So eh? is Canadian, eh? Canadian Journal of Linguistics 17: 89–104.

    Google Scholar 

  • Barbieri, Federica. 2005. Quotative use in American English. A corpus-based, cross-register comparison. Journal of English Linguistics 33(3): 225–256.

    Article  Google Scholar 

  • Baxter, HWilliam, and Laurent Sagart. 2014. Old Chinese: A new reconstruction. New York: Oxford University Press.

  • Bowers, John. 2010. Arguments as relations. Cambridge: The MIT Press.

    Book  Google Scholar 

  • Cheng, Lisa L-S. 2009. On every type of quantificational expression in Chinese. In Quantification, definiteness, and nominalization, ed. Monika Rathert and Anastasia Giannakidou, 53–75. Oxford: Oxford University Press.

    Google Scholar 

  • Cinque, Guglielmo. 1999. Adverbs and functional heads. A cross-linguistic perspective. New York: Oxford University Press.

    Google Scholar 

  • Columbus, Georgie. 2010. Ah lovely stuff, eh?—invariant tag meanings and usage across three varieties of English. In Corpus-linguistic application: Current studies, new directions, ed. Stefan T Gries, Stefanie Wulff, and Mark Davies, 85–102. Amsterdam: NY: Rodopi.

    Google Scholar 

  • Dobson, William A C H. 1966. Negation in Archaic Chinese. Language 42(2): 278–294.

  • Douglas, Rev Cartairs. 1873. Chinese-English dictionary of the vernacular or spoken language of Amoy with the principal variations of the Chang-chew and Chin-chew dialects. London: Trubner and Co.

    Google Scholar 

  • Giannakidou, Anastasia, and Lisa L-S Cheng. 2006. (In)definiteness, polarity, and the role of wh-morphology in free choice. Journal of Semantics 23: 135–183.

    Article  Google Scholar 

  • Haegeman, Liliane. 2014. West Flemish verb-based discourse markers and the articulation of the speech act layer. Studia Linguistica 68(1): 116–139.

    Article  Google Scholar 

  • Hajičová, Eva, Barbara H Partee, and Petr Sgall. 1998. Topic-focus articulations, tripartite structures, and semantic content. Dordrecht: Kluwer Academic Publishers.

    Book  Google Scholar 

  • Hopper, Paul J. 1991. On some principles of grammaticalization. In Approaches to grammaticalization, vol. 1, ed. Elizabeth Closs Traugott and Bernd Heine, 17–36. Amsterdam: John Benjamins.

    Chapter  Google Scholar 

  • Hsu, Tingting Christina. 2013. Subjectification and the emergence of deontic modal verbs TIOH8, AI3, and E7 in Southern Min. Doctoral dissertation. National Tsing Hua University.

  • Karttunen, Lauri, and Stanley Peters. 1979. Conventional implicature. In Syntax and semantics, vol. 11: Presupposition, ed. Oh Choon-Kyu and David A Dinneen, 1–56. New York: Academic Press.

    Google Scholar 

  • Kayne, Richard S. 2005. Movement and silence. New York: Oxford University Press.

    Book  Google Scholar 

  • König, Ekkehard. 1988. Concessive connectives and concessive sentences: Cross-linguistic regularities and pragmatic principles. In Explaining language universals, ed. John A Hawkins, 145–166. Oxford: Basil Blackwell.

    Google Scholar 

  • Lehmann, Christian. 2005. Pleonasm and hypercharacterization. In Yearbook of morphology, ed. Geert Booij and Jaap van Marle, 119–154. Dordrecht: Springer.

    Google Scholar 

  • Lien, Chinfa. 2013. The polyfunctionality of Kio 3 叫in Niri 7 Kiann 3 Ki 3 荔鏡記: A framenet-based approach. Journal of Chinese Linguistics. 41(1): 170–196.

    Google Scholar 

  • Lin, Jo-wang. 1998. Distributivity in Chinese and its implications. Natural Language Semantics 6: 201–243.

    Article  Google Scholar 

  • Malkiel, Yakov. 1957. Diachronic hypercharacterization in romance, Archivum Linguisticum. 79–113.. 10: 1–36.

    Google Scholar 

  • Norman, Jerry 羅傑瑞. 1995. Origin of negatives in Jianyang dialect 建陽否定詞探源. Fangyan方言 1: 31–32.

  • Partee, Barbara. 1991. Topic, focus and quantification. In Proceedings of SALT 1, ed. Steven Moore and Adam Z Wyner, 159–188. Ithaca, New York: CLC Publications, Department of Linguistics, Cornell University.

    Google Scholar 

  • Potts, Christopher. 2012. Conventional implicature and expressive content. In Semantics: An international handbook of national language meaning, vol. 3, ed. Claudia Maienborn, Klaus von Heusinger, and Paul Portner, 2416–2536. Berlin: Mouton de Gruyter.

    Google Scholar 

  • Pulleyblank, Edwin G. 1995. Outline of classical Chinese grammar. Vancouver: University of British Columbia Press.

    Google Scholar 

  • Quanzhoushi Wenhuaju Quanzhou Difang Xiqu Yanjiushe 泉州市文化局泉州地方戲曲研究社. 2010. The Daoguang edition of Li Zhi Ji: Its photocopy and annotated edition 荔鏡記荔枝記四種第三種清代道光刊本荔枝記書影及校勘本. Beijing: Zhongguo Xiju Chubanshe.

    Google Scholar 

  • Rizzi, Luigi. 1997. The fine structure of the left periphery. In Elements of grammar, ed. L Haegeman, 281–337. Dordrecht: Kluwer.

    Chapter  Google Scholar 

  • Rizzi, Luigi. 2004. Locality and left periphery. In Structures and beyond, vol. 3: The cartography of syntactic structures, ed. Adriana Belletti, 223–251. Oxford: Oxford University Press.

    Google Scholar 

  • Traugott, Elizabeth Closs, and Ekkehard König. 1991. The semantics-pragmatics of grammaticalization revisited. In Approaches to grammaticalization, vol. 1: Theoretical and methodological issues, ed. Elizabeth Closs Taugott and Heine Bernd, 189–218. Amsterdam: John Benjamins Publishing Company.

    Chapter  Google Scholar 

  • Tubau, Susagna. 2014. The syntax of the confirmatory pragmatic particle innit. Atlantis: Journal of the Spanish Association of Anglo-American Studies 16(2): 53–72.

    Google Scholar 

  • van der Loon, Pier. 1992. The Chinese theatre and art song of south Fukien 明刊閩南戲曲絃管選本三種. Taipei: SMC Publishing, Inc.

    Google Scholar 

  • Wiltschko, Martina. 2014. The universal structure of categories: Towards a formal typology. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Wu, Shouli 吳守禮. 2001a. Annotated texts of the romance of Li Jing Ji of Ming Jiajing edition 明嘉靖刊荔鏡記戲文校理. Taipei: Congyi Workshop.

  • Wu, Shouli 吳守禮. 2001b. Annotated texts of the romance of Li Zhi Ji of Ming Wanli edition 明萬曆刊荔枝記戲文校理. Taipei: Congyi Workshop.

  • Wu, Shouli 吳守禮. 2001c. Annotated Texts of the Romance of Li Zhi Ji of Qing Shunzhi Edition 清順治刊荔枝記戲文校理. Taipei: Congyi Workshop.

  • Wu, Shouli 吳守禮. 2001d. Annotated texts of the romance of Li Zhi Ji of Qing Guangxu edition 清光緒刊荔枝記戲文校理. Taipei: Congyi Workshop.

  • Yang, Liensheng 楊聯陞. 1971. Notes on negatives 漢語否定詞雜談. Tsing Hua Journal of Chinese Studies 清華學報 9(1/2): 160–191.

Download references


An earlier version of this paper was presented at The 22nd Annual Conference of the International Association of Chinese Linguistics and The 26th North American Conference on Chinese Linguistics, School of Languages, Literatures and Cultures, College of Arts and Humanities, University of Maryland, College Park, MD, USA, May 2–4. I thank the audience for comments. The research that this paper is based on was partially supported by NSC 103-2923-H-007-001 and the Multicultural Studies in Monsoon Asia project, Research Center for Humanities and Social Sciences, National Tsing Hua University. I am debted to Liying Chen for sorting out the data and computerizing the paper. Thanks are also due to anonymous reviewers and Dingxu Shi for insightful comments and suggestions. All errors remain mine.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Chinfa Lien.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lien, C. Interpreting 都 to 1 TO in earlier Southern Min texts. lingua. sin. 1, 9 (2015).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: