Where Do Chinese Doublets Come From?—The Doublets from Prehistory to the Era of the Book of Poetry

Article Open Access

Where Do Chinese Doublets Come From?—The Doublets from Prehistory to the Era of the Book of Poetry

Author Information
Institute of Ethnology and Anthropology, Chinese Academy of Social Sciences, Beijing 100028, China
Authors to whom correspondence should be addressed.
Nature Anthropology 2023, 1 (1), 10005;  https://doi.org/10.35534/natanthropol.2023.10005

Received: 11 May 2023 Accepted: 25 June 2023 Published: 08 August 2023

Creative Commons

© 2024 by the authors; licensee SCIEPublish, SCISCAN co. Ltd. This article is an open access article distributed under the CC BY license (https://creativecommons.org/licenses/by/4.0/).

ABSTRACT: The earliest writing in China is the oracle bone inscriptions of the Shang Dynasty, which records early Chinese, also known as oracle bone Chinese, which are all monosyllabic-words (1300 BC). In the Bronze Inscriptions of the Western Zhou Dynasty and later handed down documents, doublets appear (beginning in 1046 BC). At present, the philological academy believes that the doublets recorded with two Chinese single-characters come from reduplication of two single-character symbols, but there is no complete argument and reliable evidence. This article, by using the opposite method of argument, reversely assumes that the single-characters (monosyllabic words) come from doublets and tries to demonstrate it. The article proves the truth of the origin of doublets based on the word distribution and semantic correspondence between doublets and single-characters in “the Book of Poetry”, that is, doublets are the source and single-characters are flows. Among them, 39.66% of the doublets have no corresponding single-characters, and they are the characters created to record doublets; 41.92% of the meanings of doublets have nothing to do with the meanings of single-characters, which proves that the doublets does not come from the combination of single-characters; 12.46% of the meanings of doublets are interpreted as the meanings of single-characters, which are the subjective errors of later generations of interpreters; the remaining 5.66% are only associated with proclitics and enclitics rather than single-characters. Finally, the article proposes that doublets originate from a unique mechanism of expressive morphology, which is a new type of etymological theory outside the morphological grammar system, and can create various polysyllabic ideophones, including the onomatopoeia or mimetic words. The article proves that a language begins with the creation of words. In the prehistoric period before the oracle bone inscriptions, Chinese ancestors had invented a large number of distinctive doublets (AA), couplets (AB) and other polysyllabic words (xA, or ABB, ABA’B), or ideophones. Due to the difficulty of writing, the doublets were hidden in spoken language for hundreds of years. It was not until the time of “Book of Poetry” and “Book of History” in the bronze inscriptions of the Western Zhou Dynasty that it entered the history and has continued to this day. Doublets are the earliest Chinese words and the beginning of Chinese civilization.
Keywords: Doublet; Single-character; Classification for character usage; Expressive morphology; Word creation; The Book of Poetry

1. Introduction

The origin of doublets in the Book of Poetry (shī jīng 1046–771BCE) is generally believed to be the overlapping or reduplication of single-characters or monosyllabic words. Why does a syllable reduplicate? There are many different opinions, such as the reasons of semantic enhancement [1] (p. 235), rhetorical technique [2] (pp. 57–60), baby talk [3] (pp. 366–369), reduplicated morphology [4] (pp. 428–429), [5] (p. 18) and enhancement of morphology [6] (p. 153), [7] (pp. 79–87), [8] (p. 6). Among them, Shi (2009) [9] put forward a new point of view and gave a seemingly rigorous argument. 1 Shi (2009) [9] suggests that AA-form doublets did not appear in the Shang Dynasty, and by the time of “the Book of Poetry” in the middle and late Western Zhou Dynasty, there were already a large number of them, revealing the logical development process of AA-form doublets from nothing to existence, from less to more. Shi also suggests that the function of the A-form is to describe the modality, and the AA-form to describe the modality as well. If the number of AA-form is more than that of A-forms, it shows that “the monosyllabic state adjective in the pre-Qin period has been very poor in its exclusive use, and its function of expressing the modality is weakened, so other means are needed to support and strengthen its descriptive function” [9]. Furthermore, Shi found that from the Book of Poetry to the Songs of Chu (chǔ cí), the number of monosyllabic state adjectives decreased significantly, while the number of enclitics increased (浩然、蕞爾), and the doublets were the most “ways to help monosyllabic state adjectives enhance their descriptiveness”, and thus asserted: “Therefore, we believe that the motivation of AA-form doublets is the weakening of the pragmatic function of monosyllabic state adjectives in the pre-Qin period [9]. 2 According to Shi’s argument, at least the following questions need to be clarified. 1. If the AA-forms come from the A-forms, the judgement should be proved by statistic data. 2. If the AA-forms come from the A-forms, the correspondence on the senses between AA-forms and A-forms should be perfectly achieved. 3. If the A-forms in the AA-forms are state adjectives, how could they be reduplicated in the age of the Book of Poetry when the reduplicative morphology did not appear? Let’s look at the first question. According to Shi’s statistics, there are 73 corresponding forms of monosyllabic state adjectives in all doublets (353) in the Book of Poetry [9], in other words, only 73 A-form monosyllabic adjectives constitute AA-forms. This means that the vast majority of doublets (280) do not appear to enhance the descriptivity of the A-forms, and to say the least, there is no evidence that most doublets come from A-forms, and the evidence for reduplication is seriously insufficient. According to Kennedy (1964) [10], among the 360 doublets in the Book of Poetry, 139 Chinese characters are only used in doublets and do not appear anywhere else in the Book of Poetry, that is, they do not appear alone. “They seem to have been created for this particular construction”. From this point of view, an AA-form from a A-form has been water without a source. These two facts clearly do not support the idea that doublets come from monosyllabic state adjectives. The second question. Shi (2010) [11] mainly discusses the purpose, function, and method of enhancing the descriptiveness of monosyllabic state words, but only a few cases are given in terms of the correspondence between AA-form and A-form meanings. In particular, Shi (2010) [11] presents the total number of doublets and single-characters, but does not list the table of doublets or single characters, which can easily confuse the different relationships between doublets and monosyllabic words (Chinese characters), for example, the doublets created by the use of loan-characters have little relationship with the original meanings of Chinese characters, only the shape of the characters is the same. The third question lacks empirical evidence. Shi [11] believes that “most of the doublets are the reduplication of monosyllabic state adjectives”, and proposes that “the semantic range of state adjectives is narrow, and there is no obvious ‘quantity’, and the reduplication is only to emphasize the descriptiveness of the A-form and enhance vividness and subjectivity”. It is a plausible proposition that by comparing the reduplication of modern property adjectives to obtain degree increments, it is renamed as reduplicating state adjectives to enhance descriptiveness. Because reduplication is a morphological process with grammatical value, what reduplication or affixation achieves is to determine the value change in the grammatical domain or grammatical category, such as the alternation of singular and plural nouns, the alternation of perfect and imperfect verbs. However, the “descriptiveness (vividity and subjectivity)” proposed by Shi seems to be a subjective evaluation concept, not the ontological and objective nature of state adjectives, and it is difficult to determine the value changes in it. Furthermore, in the Age of the Book of Poetry, reduplication as a morphological method has not yet been developed (Kennedy 1964) [10]. Therefore, the theory of motivation to constitute doublets in order to strengthen the descriptiveness of monosyllabic state adjectives cannot be established. Based on the above discussion, this paper attempts to re-propose the proposition. We assume that since some AA-forms and A-forms are morphologically and semantically related, we might as well assume that the monosyllabic state adjectives come from doublets, and check whether there is indeed an ontological connection between the two. We begin by observing the data. 3
Figure 1. The relationship between doublets and monosyllabic adjectives and related data. (A) 353 doublets, 669 occurrences; (B) 786 monosyllabic adjectives with 3334 occurrences 4; (C) 192 doublets corresponding to homonymous single-characters, appearing 422 times; (D) 192 single-characters corresponding to doublets, appearing 1810 times; (E) 21 doublets corresponding to single-characters of disyllabic words, appearing 44 times 5; (F) 594 monosyllabic adjectives without corresponding doublets, appearing 1525 times; (G) 140 single-characters only for doublets, appearing 208 times.
In Figure 1 The monosyllabic adjectives of C and D are associated with doublets, and the data are 192 items, but the perspectives are different. C indicates that 192 items of doublets have evolved into monosyllabic words, and 353 items still retain the form of doublets. D, on the other hand, means that 192 of the 786 monosyllabic words have been combined into doublets. Whichever view is taken, there may be a semantic or functional relationship between doublets and single characters, but it may also be a relationship of the same form. This is the question to be discussed below. E indicates that this part of doublets has not yet been monosyllablized. However, it remains to be seen whether the same characters between disyllabic words and doublets are the result of coincidence or evolution. F shows that at least about 75.57% of the characters have not been reduplicated from the point of view of doublets produced by character combinations. G means that from the perspective of the character usage in the Book of Poetry, each character in doublets neither appears independently nor is a constituent unit of any word other than doublets. What does this mean? 140 items of G (139 items of G. Kennedy’s statistics) are not a small number, and they appear 208 times, which cannot be accidental. So, we agree with Kennedy (1964) [10], these character items were created specifically to record doublets. The people of the West Zhou Period and the Spring and Autumn Period have mastered the method of inventing Chinese characters, and they are happy to use this skill to record the sounds spoken in their spoken language. Such sounds have never been recorded before, and they need the help of Chinese phonetic methods and techniques, so they are very natural and inartificial. The above discussion provides only prima facie evidence for conjectural results and has not yet been demonstrated. The correspondence between the single-characters of C and D and doublets only reflects the superficial relationship in the character forms, and does not discuss whether there is a real relationship in the meaning and function of the two. We need to really demonstrate the original nature of doublets in the Book of Poetry at a deep level.

2. Doublets Classification Based on Distribution of Character Usages

By index on the database of the Book of Poetry, we find that there are three kinds of character distribution relationships between doublets and single-characters. The terms herein are agreed as follows: A doublet refers to Chinese words consisting of two-homographic characters, called “重言 chóngyán” in Old Chinese. A couplet refers to Chinese words consisting of two characters, often alliterated or rhymed, called “联绵 liánmián”. A proclitic refers to disyllabic words with an initial-sound character, “词头词 cítóucí”, including “有字式 yǒuzìshì (characters with initial-sound yǒu)”; “其字式 qízìshì”, “斯字式 sīzìshì),” etc. An enclitic refers to structures with a post-auxiliary character, “(后附)衬字(结构) chènzì”, which are temporarily classified as couplets. A single-character refers to a monosyllabic state adjectives, which is a morpheme or a word. A disyllabic-word (“复音词”) includes couplets, proclitics, enclitics, and compound words [11,12,13]. 6 2.1. Type I: Characters among Doublets Never Appear as Isolated Single-Characters In this type there are 140 doublets (39.38%), which are what G. Kennedy called “Chinese characters created to record doublets” (1964) [10] 7. They do not appear in any other word-structures, and cannot be used alone, and do not form any disyllabic words in the Book of Poetry. The list is as follows (doublets are listed in single-character. The same to other lists of the paper below): 蔼ǎi, 嗷áo, 傍bēng, 菶běng, 怭bì, 熚yè, 儦biāo, 瀌biāo, 怲bǐng, 啴tān, 幝chǎn, 蚩chī, 冲chōng, 惙chuò, 佌cǐ, 踧dí, 忉dāo, 蹲cún, 峨é, 茷pèi, 幡fān, 騑fēi, 雰fēn, 唪běng, 杲gǎo, 仡yì, 阁gé, 耿gěng, 痯guǎn, 浩hào, 熇hè, 翯hè, 薨hōng, 喤huáng, 煌huáng, 哕huì, 翙huì, 踖jí, 濈jí, 伎qí, 畟cè, 矫jiǎo, 湝jiē, 斤jīn, 菁jīng, 兢jīng, 赳jiū, 踽jǔ, 鞙xuàn, 悁yuān, kuài, 骙kuí, 连lián, 粼lín, 栾luán, 律lǜ, 芒máng, 浼měi, 幪méng, 沵nǐ, 绵mián, 藐miǎo, 薿nǐ, 浓nóng, 庞lóng, 旁bēng, 伾pī, 僛qī, 锵qiāng, 蹻jiǎo, 谯qiáo, 翘qiáo, 钦qīn, 骎qīn, 睘qióng, 俅qiú, 祛qū, 渠qú, 瞿qú, 瀼ráng, 穰ráng, 陾réng, 懆cǎo, 汕shàn, 诜shēn, 駪shēn, 湜shí, 叟sōu, 穟suì, 傞cuō, 惕tì, 籊tì, 趯tì, 阗tián, 佻tiáo, 僮tóng, 慱tuán, 焞tūn, 脱tuì, 丸wán, 韡wěi, 濊huì, 习xí, 潝xi, 僊xiān, 哓xiāo, 翛xiāo, 潇xiāo, 嚣xiāo, 欣xin, 嬛qióng, 岩yán, 咽yuàn, 泱yāng, 洋yáng, 喓yāo, 驿yì, 弈yì, 殷yin, 呦yōu, 滺yōu, 俣yǔ, 噳yǔ, 麌yǔ, 彧yù, 蜎yuān, 畇yún, 战zhàn, 罩zhào, 晢zhé, 晣zhé, 蛰zhé, 蓁zhēn, 殖zhí, 挃zhì, 惴zhuì, 谆zhun, 啍tun, 灼zhuó, 訿zǐ The Book of Poetry is one of the earliest documents in China with the largest number of Chinese characters. But even so, the statement by G. Kennedy makes us know that the era of the Book of Poetry is still an era of large-scale creation of words and characters, and the “Chinese characters created to record doublets” in the Book of Poetry is the result of the creation of characters in this era. However, with the help of “Lexicon of Pre-Qin Oracle, Bronze Inscriptions and Bamboo Scripts” (Shang and Zhou Dynasties to the Spring and Autumn Period and the Warring States Period), we have a new understanding of this part of Chinese characters. 8 The Chinese characters used by doublets mentioned by G. Kennedy can also be divided into two groups, one is the characters that do not exist in the oracle and bronze inscriptions, and it is judged that the doublet recorder created new Chinese characters from the shape by using the rules of associative and pictophonetic characters. For example, there are 110 Chinese characters such as ǎi蔼,biāo儦,fēn雰,huáng煌, etc. Second, most of the Chinese characters represent the names of people, places, clans, family emblems and other specific meanings in Shang Dynasty, Western Zhou Dynasty and Spring and Autumn Period oracle bone inscriptions and bronze inscriptions. For example: chī蚩, jīn斤, páng庞, páng旁, xiāo嚣, gǎo杲, luán栾, ráng瀼, there are 12. The other 18 characters are from the Warring States Period, such as fēi騑, kuài哙, hào浩, máng芒, qīn钦, and zhàn战. Strictly speaking, these characters appearing in the inscriptions on bronze and tortoise shells have nothing to do with the meaning of doublets. We may think that doublets in the Book of Poetry only use such characters to create words, and some of them may have homophones with the borrowed characters, especially the onomatopoeia or mimetic words. Of course, this kind of creation may not be unique to the recorder of the Book of Poetry, but is gradually created and shared by the whole society. According to other documents, for example, the “The Book of Documents《尚书》” of the same period has “jīngjīng兢兢, hàohào浩浩, yìyì仡仡”, and the Chinese characters “lǜ律, tì惕, xiāo嚣” used separately, and later, the “Songs of the South《楚辞》” of the Warring States Period has “fēnfēn雰雰, hàohào浩浩, gǎogǎo杲杲, rángráng穰穰, tiántián阗阗”. These common characters prove that the creation of Chinese characters is an inevitable and indispensable way to record new words. It can also be inferred that the views of G. Kennedy are wise and desirable. For example, “biāo儦”, “biāo瀌” are actually different characters created for onomatopoeia or mimetic words in different contexts. Since the Chinese characters are created especially for doublets, they must not be a combination of single-characters. 2.2. Type II: Characters in Doublets Are Homographic to Single-Characters From the database, it is found that the number of homographs between doublets and single characters is 192 pairs. 9 A symbol homograph is the cornerstone of the idea of using single-character to form doublets. Do the facts support it? We further observe the relationship between the meaning of the two characters, which can be divided into two sub-groups. Type IIa: the meaning of doublets is unrelated to the meaning of single-characters, with 148 cases (41.92%) and 271 occurrences. The details are listed as follows: 卬áng, 敖ǎo, 板bǎn, 薄bó, 奔bēn, 麃biāo, 镳biāo, 采cǎi, 憯cǎn, 苍cāng, 草cǎo, 掺xiān, 虫chóng, 仇chóu, 楚chǔ, 旦dàn, 登dēng, 涤dí, 棣dì, 丁ding, 耳ěr, 发fā, 番fān, 反fǎn, 肺pèi, 冯píng, 逢péng, 弗fú, 茀fú, 浮fú, 皋gāo, 关guān, 管guǎn, 灌guàn, 膴wǔ, 虺huǐ, 嘒huì, 活huó, 缉qi, 几jǐ, 济jì, 渐chán, 将qiāng, 交jiāo, 骄jiāo, 胶jiāo, 揭jiē, 嗟jiē, 孑jié, 桀jié, 捷jié, 矜jīn, 京jīng, 究jiu, 居jū, 槛jiàn, 溃kuì, 离lí, 栗lì, 邻lín, 令lìng, 迈mài, 枚méi, 梦mèng, 靡mǐ, 勉miǎn, 莫mù, 泥ní, 孽niè, 旆pèi, 彭péng, 蓬péng, 平píng, 栖xi, 戚qi, 祁qí, 祈qí, 俟sì, 契qiè, 强qiāng, 跄qiāng, 溱zhēn, 青jing, 惸qióng, 裳cháng, 蛇yí, 牲shēn, 绳shéng, 施yì, 湿shē, 实shí, 蔌sù, 绥suí, 琐suǒ, 汤shāng, 滔tāo, 慆tāo, 陶yáo, 提chí, 佗tuó, 橐tuó, 唯wéi, 亹wěi, 温yùn, 沃wò, 闲xián, 宪xiàn, 萧xiāo, 偕xié, 泄yì, 骍xing, 休xiu, 许hǔ, 湑xǔ, 谑xuè, 熏xun, 言yán, 厌yàn, 晏yàn, 央ying, 阳yáng, 养yǎng, 业yè, 依yi, 猗yi, 揖jí, 抑yì, 绎yì, 逸yì, 殷yin, 英ying, 营yíng, 雝yong, 优you, 与yú, 愈yù, 渊yuān, 爰yuán, 跃tì, 牂zāng, 凿záo, 泽shì, 增zēng, 湛dān, 招zhāo, 烝zhēng, 秩zhì, 濯zhuó Many scholars have done a lot of researches on this phenomenon. Xiang [14] concluded that “chóngchóng蟲蟲” is the appearance of hot air transpiration, which has nothing to do with the insects; “shīshī湿湿” is the appearance of cow’s ears shaking, which has nothing to do with the “wet with water”; “dēngdēng登登” is the sound of smashing a wall with great force, which is not related to the “climbing”. In fact, this clearly points out that the correspondence between single-characters and characters in doublets is possibly that when doublets were created, people borrowed existing Chinese character symbols to record doublets. For example, “fúfú弗弗” (蓼莪: 飘风弗弗) should be a loan word from “fāfā发发”, “jiāojiāo胶胶” (郑风: 鸡鸣胶胶), in Mao H, Zheng X (2018) [15], “胶胶,犹喈喈也”. It is also a loan word. “cāngcāng鸧鸧” (商颂: 八鸾鸧鸧), the original word may be “qiāngqiāng锵锵”. Phonetic loan is a more convenient and concise way than creating new characters, so there are a large number of them. Here are a few examples (Table 1):
Table 1. The meaning of doublets is unrelated to that of single-characters.
Comparing the ancient interpretation of MaoZhuan with the modern annotation of Xiang (2014) [16], the meanings of single-characters listed above are completely different from those of doublets. For example, “我有旨酒,嘉宾式燕以敖I have delicious wine, you may enjoy my brew.” (Book of Odes: to guests), whose word is interpreted as “áo敖, wandering” in MaoZhuan. In contrast, “shuòrén áoáo, shuìyú nóngjiāo硕人敖敖,说于农郊, the buxom lady goes along, she passes outskirts to be wed.” (The Duke’s Bride)”, MaoZhuan explains “áoáo敖敖, long appearance”, that is tall appearance in the Dictionary of Book of Poetry by Xiang (2014) [16]. Obviously, the two are not the same thing at all. The English verses used in the paper are translated by Xu YC [17]. The use of loan characters to create new symbols and new words is one of the ways in which doublets are created, and the main point is to borrow sounds of existing characters. This is stated by Qiu [18]: “A character or a picture of something is borrowed as a phonetic symbol to record words that have the same or similar sound as the name of the word or the name of the thing. In this way, words for which it is difficult to create ideographs can also be recorded in writing”. We are convinced that Qiu’s insightful view could not be more aptly applied to the doublets in the Book of Poetry. The use of single characters to create doublets is only a phenomenon of recording phonetic words in writing, and has nothing to do with the reduplication of single characters (morphology). Type IIb: the meanings of doublets are related to the meanings of single-character words, including direct and extended meanings. There were 44 cases (12.46%, 149 times). This group of words is listed below. 哀āi, 安ān, 苾bì, 惨cǎn, 粲càn, 迟chí, 绰chuò, 蹙cù, 怛dá, 泛fàn, 芬fēn, 甫fǔ, 高gāo, 好hào, 皓hào, 赫hè, 桓huán, 皇huáng, 黄huáng, 简jiàn, 皎jiǎo, 駉jiong, 纠jiu, 睠juàn, 蹶guì, 蓼liǎo, 烈liè, 明míng, 冥míng, 穆mù, 凄qi, 萋qi, 悄qiǎo, 肃sù, 显xiǎn, 吁xǔ, 炎yán, 夭yāo, 曀yì, 翼yì, 幽you, 悠you, 昭zhāo, 振zhèn This paper judges the semantic relationship between a doublet and a single-character word mainly based on MaoZhuan, and uses other documents when a word is not explained in MaoZhuan. It is interesting that the meaning of a doublet is not the same as the original meaning of a single-character word, but Xǔ’s Origin of Chinese Characters sometimes directly uses the meaning of a doublet in the Book of Poetry as the original meaning of a single-character word. (Xiang, 1987) [14]. This also shows that the meaning of doublets is one of the sources of the meaning of single-characters (Xǔ, 1991) [19]. Please see the Table 2.
Table 2. Examples with related meanings between doublets and single-character words.
Observing these examples, we can see that the meaning of most doublets is consistent with the meaning of single-characters, mainly in the following groups. First of all, some content annotations directly show the relationship between the two meanings. For example, in the Book of Poetry, “是烝是享, 苾苾芬芬. shìzhēngshìxiǎng, bìbìfēnfēn. During the sacrifice the fat burned gives smell nice”. Zheng’s Notes said, “Since there are animals to offer, it is fragrant in bìbìfēnfēn”, compared with the single-character word “苾芬孝祀, 神嗜饮食. bìfēnxiàosì, shénshìyǐnshí”, Zheng’s Notes interpreted it as “bìbì fēnfēn, there is fragrance”. The line “bìbì fēnfēn” in the doublet clause is used as “bìfēn” in the clause with a single-character word, which obviously follows the meaning of the former. For another example, “于粲洒埽,陈馈八簋. yúcànsǎsǎo, chénkuìbāguǐ. Sprinkle and sweep the yards And arrange eight round dishes” (Book of Odes, Felling Wood), the exegesis of the single-character word in MaoZhuan is: 粲 (càn), bright appearance. Contrast the doublet: “西人之子, 粲粲衣服. xīrén zhīzǐ, càncàn yīfú. But as slaves the poor ones Serve all the officers. The son of a Westerner, can clothes” (Book of Odes, East and West, 小雅·大东), MaoZhuan explaining “粲粲càncàn, fresh appearance”, the meaning of the two words is similar, very much like in the mutual explanation. Second, some have the same meaning with doublets through the equivalence of proclitics. For example, “世之不显,厥犹翼翼. shìzhī búxiǎn, juéyóu yìyì. From age to age they’re blest; They work with care and zest.” (Book of epics-heaven’s decree), MaoZhuan “yìyì翼翼, respect”. However, we have noticed that the former sages have put forward the view that “doublets are equivalent to youzishi (words with initial-sounds “yǒu有”)” (Wu Ge, 1987:616 [20]). Take the verse as an example: “yǒuyán yǒuyì, gòngwǔ zhīfú有严有翼, 共武之服. Careful and strict we’d be; In battle dress we stand. (Book of Odes-General Ji Fu)”, in MaoZhuan “yì, respect.” Here yì = yǒuyì有翼, and yǒuyì is yìyì翼翼. This phenomenon will be discussed below again. There is also a kind of direct use of doublet meaning to explain single graphic words, such as: “振振鹭, 鹭于下. zhènzhèn lù, lù yúxià. Like egrets white, Dancers alight”. (鲁颂 Lu Song Hymns of Lu: The Ducal feast), The doublet exegesis “zhènzhèn, the appearance of flying”; And the sentence with single graphic word “振鹭于飞, 于彼西雝. zhènlù yú fēi, yúbǐ xīyōng. Rows of egrets in flight, Over the marsh in the west”. (周颂-振鹭 ZhouSong· Hymns of Zhou: The Guest Assisting at Sacrifice), the interpretation of MaoZhuan, “振振, 群飞貌, zhènzhèn, qúnfēi mào” directly applies the meaning of the doublet to the single-graphic word. From the whole doublet system, Type I and Type IIa have occupied 81.58% (288 items), the former is to create characters for doublets, and the latter is to borrow other characters for doublets, which have nothing to do with the meaning of single-character words. Therefore, it can be inferred that the doublets in the era of Book of Poetry have their own meanings. According to the above analysis, the number of type Ⅱb is relatively few, and it seems that the meaning of doublets is the same or similar to that of single-characters, but the real reason is that the meaning of a single-character is copied from the meaning of doublets. Statistically, type Ⅰ and type II (IIa and IIb) accounted for 94.05% of all doublets. That is, almost all doublets cannot come from the combination of single-characters. However, there are still a few doublets of Type Ⅱb that are somewhat special in nature, and from the current perspective, they have a certain “modernity” characteristic, that is, they show the characteristics of property adjectives of the later generations of the Book of Poetry, which will be named as IIbx. Table 3 lists some of these example words.
Table 3. Contrastive meaning of doublets and single-character words.
The modernity of words and expressions contains their historical processes, so there are differences in degree. For example, the monosyllabic bases of gāogāo (tall), cǎncǎn (miserable) and chíchí (late) have acquired the nature of modern property adjectives, for example, they can be modified by degree adverbs: hěngāo (very high), zuìgāo (the highest); hěncǎn (very miserable), jícǎn (extremely miserable); lüèchí (slightly late), tàichí (too late). They can be reduplicated: gāogāo gèzi (tall man), gāogāo jǔzhe (hold highly); cǎnyǎn yīnfēng (miserable gloomy wind), chíchí wèidòng (delayed in action), as well as other morphological and syntactic features. However, this is a later story, and about 3000 years ago they did not show the characteristics of property adjectives in the Book of Poetry. There are still a few similar words: hàohào (a happy look), jiǎnjiǎn (large), ānān (a calm look), míngmíng (wise), xiǎnxiǎn (remarkable), yōuyōu (far-reaching). Type IIbx is likely to be the beginning of the idea of doublet reduplication. For this reason, we have to make it clear that the correlation between the meaning of doublets and single-characters comes from the interpretation of scholars in later times, which is subjective. For example, in MaoZhuan: “xiéxié偕偕, strong appearance” (偕偕士子, 朝夕从事. being an officer strong, I’m busy day and night for the royal affairs); while the single-character “偕” in MaoZhuan is “together, all together” (执子之手, 与子偕老. give me your hand I’ll hold and live with me till old). There is no relationship between the two word meanings, and “偕偕” may be a loan character. However, in the Origin of Chinese Characters, “xié偕” is also explained as “strong”, and directly indicates that the meaning of the character comes from the doublet of the verse: 偕偕士子. A more enlightening study is that of Shen (2011) [8], who proposed a classification perspective on modern Chinese parts of speech that goes beyond convention, arguing that Chinese nouns, verbs, and adjectives are a large category that makes up a super-noun, opposite to ideophones. Compared to adjectives, whether single-characters or doublets, whether in the era of the Book of Poetry or in modern times, a doublet is a category of words that has not yet been deeply studied. Not only in the Book of Poetry, but throughout the pre-Qin period and even in modern times, it has always been on the side of ideophones with opposite to nouns, verbs, and adjectives, causing many confusion. People mistakenly view a doublet as the predecessor of reduplication, or believe that a reduplicating word is an inheritance of doublet, all of which stem from this. 2.3. Type III: Homographs of Doublets and the Characters Used in Disyllabic-Words The note in Section 1 specifically stresses that disyllabic-words in this article refer only to proclitics, enclitics and couplets. Some characters used by doublets appear in these words, that is, these single-characters do not appear independently, a total of 21, accounting for 5.94%. The Chinese characters corresponding to doublets and disyllabic-words are listed below. Proclitics: 鸧鸧qiāngqiāng--有鸧yǒuqiāng; 忡忡chōngchōng--有忡yǒuchōng; 荡荡dàngdàng--有荡yǒudàng; 霏霏fēifēi--其霏qífēi; 洸洸guāngguāng--有洸yǒuguāng; 喈喈jiējiē--其喈qíjiē; 瀰瀰mǐmǐ--有瀰yǒumǐ; 芃芃péngpéng--有芃yǒupéng; 翩翩piānpiān--有翩yǒupiān; 玱玱qiāngqiāng--有玱yǒuqiāng; 颙颙yúyú--有颙yǒuyú. Couplets and enclitics: 崔崔cuīcuī--崔嵬cuīwéi; 涣涣huànhuàn--判涣pànhuàn; 坎坎kǎnkǎn--坎其kǎnqí; 涟涟liánlián--涟猗liányī; 绵绵miánmián--绵蛮miánmán; 淠淠pìpì--淠彼pìbǐ; 翩翩piānpiān--有翩yǒupiān; 翩翩piānpiān--翩彼piānbǐ; 委委wēiwēi--委蛇wēiyí; 摇摇yáoyáo--漂摇piāoyáo; 嘤嘤yīngyīng--嘤其yīngqí. Firstly, these words all have the characteristics of onomatopoeias or mimetics, which is the most important way to create doublets (Jiang 2021) [21]. Secondly, the characters of doublets corresponding to disyllabic-words can be divided into two groups: proclitics and couplets. Only the single-characters in proclitics are assigned a meaning, The single-characters in couplets (and enclitics) take no meaning. This reason must be related to the statement by Wang (1987) [20] that “doublets are equivalent to yǒuzishi (有字式)”, and must also be the basis of MaoZhuan’s exegesis. Try to observe the table below. From the cases in Table 4, it is obvious that the exegesis method of MaoZhuan is to explain the root of proclitics without considering the initial characters. For example, “玱qiāng is voice of jade (héng)”, which explains the “玱qiāng” in the proclitic. This example comes from “有玱葱珩yǒuqiāng cōnghéng, And gem of pendant ring”. Again, “喈jiē is fast state”, which explains “喈jiē” in the clause “北风其喈, 雨雪其霏, běifēng qíjiē, yǔxuě qífēi, The sharp north wind does blow, And heavy falls the snow”. MaoZhuan says for the same doublet that “喈喈jiējiē is harmony heard far away”, which means sound. Later, Zhu (1958) [22] in Collection of Anotations for the Poetry said that “喈jiē is the sound of fast-speed”. To expand, “喈喈jiējiē” in the Book of Poetry describes a variety of sounds, such as the sound of yellow birds, cocks, drums and bells, and the sound of eight-bell cars.
Table 4. Examples of the homographic characters both in doublets and proclitics.
It should also be mentioned that “有荡yǒudàng”, MaoZhuan says that “dàng is easy” (鲁道有荡, 齐子由归. Lǔdào yǒudàng, qízǐ yóuguī. The way to Lu is plain; Your sister with her train), but does not explain “荡荡dàngdàng”. From the above examples, we can see that when explaining the single-characters in proclitics in MaoZhuan, the meanings of doublets follow, only consider the specific context to make the expression more concrete. For example, “有玱yǒuqiāng”, which is equivalent to “玱玱qiāngqiāng”, is embodied in a specific sound, that is, the collision of jade. If the expression of the proclitic is not much different from that of the doublet, the proclitic is directly explained by the doublet. For example, 忡忡chōngchōng, which in MaoZhuan is explained as worry (忡忡, 犹冲冲也), and comes from “未见君子,忧心忡忡. Wèijiàn junzǐ, yōuxīn chōngchōng. When my lord is not seen, I feel a sorrow keen”. (The grasshoppers) 有忡yǒuchōng, which in MaoZhuan is explained worried (忧心忡忡然), and is in “不我以归, 忧心有忡. Búwǒ yǐguī, yōuxīn yǒuchōng. I cannot homeward go; My heart is full of woe”. (complaint of a soldier) It is interesting to note that MaoZhuan does not give explanations for most of the characters used in the couplets corresponding to the doublets, or there are explanations but they are not related. For example, “陟彼崔嵬, zhìbǐ cuīwéi. The hill I climbing up”, MaoZhuan explains “cuīwéi describes mud mountains with stone”. There is also “习习谷风, 维山崔嵬. xíxí gǔfēng, wéishān cuīwéi. Strong winds blow high, But mountain stand”, MaoZhuan explains “cuīwéi refers to mountain top”. For another example, “继犹判涣jìyóu pànhuàn. To learn on what I should decide”, the character is interpreted by MaoZhuan as “判pàn means seperating; 涣huàn means scattering”, which is not a couplet in MaoZhuan and has nothing to do with the doublet “huànhuàn涣涣 (overflowing)”. About “绵蛮黄鸟miánmán huángniǎo. O hear the oriole’s song”, MaoZhuan explains “绵蛮miánmán is appearance of a little bird”, which has nothing to do with “绵绵miánmián (continuous)”. There is only one example in which “坎坎kǎnkǎn” is directly used to explain: “坎其击鼓kǎnqí jīgǔ. She beats the drum”, and in MaoZhuan “坎坎kǎnkǎn is the sound of beating a drum”, which seems to be related to the meaning of “坎其kǎnqí”. 10 In general, this means that it is difficult for a single-character in couplets to represent the holistic meaning of couplets independently. There are two exceptions, “淠彼pìbǐ” for example, in MaoZhuan it is explained as going by boat only for the single-character “淠pì”. At present we can not find it in the early or contemporaneous literature. For “涟猗liányī”, MaoZhuan only explains “涟lián”, saying that “Ripples are formed by the wind blowing on the surface of the water”. Xiang (2014) [16] clearly pointed out that “涟lián” and “猗yī” are not a couplet, nor are they at the same grammatical level. Since the period of the Six Dynasties in China, the vocabulary of “涟猗liányī” has become a disyllabic-word. One more example, “翩翩piānpiān” should be mentioned, which not only corresponds to the proclitic “有翩yǒupian” in form, but also corresponds to the couplet or enclitic “翩彼piānbǐ, and 翩其piānqí”, but it should only be related to the proclitic in semantics. Please read the words in Table 5.
Table 5. Examples of the homographic characters both in doublets and couplets.
It is not a new idea that a doublet is related to the meaning of a proclitic word. In the late Qing Dynasty, Wang XQ (1987) [20] put forward the view that “yǒuzishi (a type of proclitics) is equivalent to a doublet”. Wang X (1959) [23] pointed out that some “其霏qífēi” in the three schools of poetry were written as “霏霏fēifēi”. He said “If qífēi is not equivalent to fēifēi, this phenomenon is difficult to explain”. Therefore, he very much agreed with Wang XQ’s (1987) [20] viewpoint, and even put forward the viewpoint that qízishi (another type of proclitics) is also equivalent to a doublet. In brief, using a doublet to explain homographs in a proclitic seems to be related to each other in meaning, but the real reason remains to be investigated. However, there seems to be no connection between the doublets and the homographs in the couplets or the enclitics. This phenomenon does not support the view that the couplets originated from the doublets [2]. The reasons need to be further investigated. 11 From a statistical point of view, in all 353 doublets, the number is: Type I: 140 words, accounting for 39.66%; Type II: 192, accounting for 54.39% Type IIa: 148 words, accounting for 41.92%; Type IIb: 44 words, accounting for 12.46%; Type III: 21 words, accounting for 5.94% What is the value of the character classifications used by above doublets? Type I makes us know that the existence of the character-making procedures completely denies the possibility that the doublets come from the reduplication of single-characters; Type IIa makes us understand that even if the Chinese characters used in the doublets are the same as the single-characters, the objective data that the meaning of a doublet has nothing to do with the meaning of a single-character completely denies the assumption that the doublet comes from the reduplication of single-characters. The same forms of the two are only caused by the borrowing of characters. As for type Ⅱb, it is obviously caused by the fact that exegetes subjectively assign meanings to single-characters with doublets. Type Ⅲ can determine that the meaning of proclitics is derived from corresponding doublets. So far, we have basically proved the relationship between doublets and the corresponding Chinese single-characters and other disyllabic word, and revealed the true source of doublets.

3. Ideophones and Expressive Morphology as Motivators

3.1. Depictive Classification of Doublets The analysis of the use of characters from doublets in the Book of Poetry totally negates the view that doublets come from the reduplication of single-characters, and there is no grammatical reduplicating methods or morphological forms. 12 The cause and mechanism of the emergence of doublets in the Book of Poetry are closely related to the occurrence and development of human language. From the perspective of language ontology, any language begins with the emergence and enrichment of vocabulary. There must be some more direct way to create words before the traditional morphological methods of word formation, such as reduplication, are mature. Observing the meaning of doublets in the Book of Poetry, it is easy to notice that the later annotations of doublets are usually depictive in a pattern like the example in MaoZhuan: “叟叟sǒusǒu, (It is a) sound”; “赳赳jiūjiū, (It is) martial appearance”. This is actually the concept of onomatopoeic words and mimetic words, which shows that the production of a doublet is related to the speaker’s expression of coinage, and it is a subjective speech behavior with the help of real world sounds and scenes. From the etymological point of view, onomatopoeia and mimetics are the methods of creating words on the spot when people lack the vocabulary of the object they express, which is different from extracting existing words from the mental lexicon. Onomatopoeia is to imitate the sound of the objective world and refer to the thing or state represented by the sound, while mimetics is to simulate the thing or state representing the sound image in psychology, so as to create phonosemantic words, commonly known as ideophones. An ideophone is a very special vocabulary, which is often named by specific codes under the paradigm of modern linguistics, such as Chinese AB, xA, ABB, A里AB, ABA’B, etc. A doublet can be called AA-form. It is clear that a Chinese doublet has an ancient origin. 13 The common ground between onomatopoeia and mimetic is the depiction of things, which draws a dynamic picture in psychology, which is an important feature of expressive morphological coinage. The following is an observation of the annotations and categories presented in MaoZhuan: (1) onomatopoeia: 邻邻línlín is the sound of carriages moving 令令lìnglìng is a ring sound 挃挃zhìzhì is the sound of cutting rice 哓哓xiāoxiāo means fear (2) mimetics: 蚩蚩chīchī shows an honest appearance 猗猗yīyī shows a beautiful appearance 皓皓hàohào is white colour 营营yíngyíng expresses flying back and forth (of insects) (3) onomatopoeia and mimetics: 㷸㷸yèyè expresses the state of thunder and lightning. This example depicts the appearance of thunder and lightning, two actions of onomatopoeia and mimetic coexist. 谑谑xuèxuè means joking and joy. In the Origin of Chinese Characters the note for the verse is 善戏谑兮shànxì xuèxī , it means “at jesting he is good, but he is never rude”, which outlines the picture of the play with words, accompanied by words, laughter. (4) onomatopoeia or mimetics: 雝雝yōngyōng takes two meanings, one is the sound of wild geese, and another is happy appearance. 冲冲chōngchōng is the voice of cutting ice, or the state of decorative pendant. Among the four types of doublet ideophones, “邻邻línlín” seems to make people hear the sound of wheels rolling on the road, the squeaking sound of wheels rotating and rubbing, and the crisp sound of the collision of pendants on the frame, bringing together a picture of driving and moving activities. “蚩蚩chíchí” outlines the honest and his honest face and a restrained look of the person described by the speaker, which makes people feel pity, and the silent scene at this time seems to be a kind of voice that can be described. “㷸㷸yèyè” evokes a frightening picture of thunder and lightning. “谑谑xuèxuè” outlines the scene of laughter. The doublets such as “雝雝yōngyōng” and “冲冲chōngchōng” appear more than twice in the Book of Poetry, sometimes used as onomatopoeia, sometimes used as mimetics, which further proves that mimetics presents the appearance of onomatopoeic sounds in psychology. According to our statistics, there are 292 cases of mimetic doublets, 50 cases of onomatopoeic doublets, and 11 cases of onomatopoeia and/or mimetics in all the doublets in the Book of Poetry. Onomatopoeia and mimetics are both subjective expressions of things or states, a kind of sound depiction that arouses imagination according to the characteristics or states of objects. The recognition of the function of doublets in depicting pictures did not come into being in modern times. About 1500 years ago, Liu [24], a scholar of the Southern Liang Dynasty, vividly described the phenomenon of doublets in the Book of Poetry: “灼灼zhuózhuó” emphasizes the freshness of peach blossoms. “依依yíyí” is the appearance of willows, “杲杲gǎogǎo” is the state of sunrise, “瀌瀌biāobiāo” simulates the shape of rain and snow, “喈喈jiējiē” imitates the sound of orioles, and “喓喓yāoyāo” learns the rhyme of grasshoppers. The use of the characters “鲜xiān (fresh), 貌mào (appearance/state), 容róng (looks), 状zhuàng (shape), 声shēng (sound), and 韵yùn (rhyme)”, all vividly depicts the vivid scenes of the doublets: the flame-like red light of “灼灼zhuózhuó”, the graceful scene of “依依yíyí”, and the weak sound of “喈喈jiējiē” imitating a cute bird begging for food. These are the effects of scenes that are difficult to describe in other ordinary words. Why is a doublet so uniquely valuable? This needs to be examined in terms of ideophones and their theoretical basis. First of all, it should be pointed out that the Book of Poetry is one of the earliest recorded oral texts in ancient China (Xiang, 1987 [14]), which inherits a large number of mature vocabularies since Shang and Zhou Dynasties, especially the depictive vocabulary which is difficult to record in the oracle bone inscriptions and bronze inscriptions. 14 Secondly, with the development of narrative and poetic literature, the expression of feelings and meanings (different from oracular inscriptions) can not be avoided, prompting people to create a large number of Chinese characters to record them. Therefore, the depictive words such as doublets and couplets, which are polysyllables in spoken language, emerge explosively in written documents. The Book of Poetry is in such an era of character-making, word-making, and culture-making, which intentionally or unintentionally reveals a very important phenomenon in the history of the language, that is, ideophones are recorded by using the theory of expressive morphology. 3.2. Ideophones and the Theory of Expressive Morphology It has been mentioned that doublets can be classified as ideophones, but what is an ideophone? According to the research of Kilian-Hatz (2006) [25], before the 1970s, the study on ideophones was basically limited to African languages, and few scholars paid attention to it. Until the 1980s, especially since the 21st century, people have gradually realized that ideophones may be a universal feature of human language (Dixon 2010 [26]). The term ideophone was coined by the linguist Clement Doke (1935) [27] when he studied the African Bantu languages and defined it as “A vivid representation of an idea in sound. A word, often onomatopoeic, which describes a predicate, qualificative or adverb in respect to manner, color, sound, smell, action, state or intensity”. Since then, people have also given various definitions when studying ideophones in languages around the world. For example, Dingemanse (2011) [28] put forward the definition, “Ideophones are marked words that depict sensory imagery”. K. Akita and P. Pardeshi (2019) [29] summarize the definition of multiple people by saying that “They imitate a broad range of sensory experiences that involve auditory, visual, tactile, or other types of perception”. Dingemanse has made many statements about the unique characteristics of ideophones from a cross-linguistic and typological perspective. The five general characteristics can not only reveal the function of ideophones, but also be used as the criteria for judging ideophones. (1) ideophones are marked, i.e., they have structural properties that make them stand out from other words (2) they are words, i.e., conventionalized lexical items that can be listed and defined (3) they depict, i.e., they represent scenes by means of structural resemblances between aspects of form and meaning (4) their meanings lie in the broad domain of sensory imagery, which covers perceptions of the external world as well as inner sensations and feelings (5) ideophones form an open lexical class, i.e., a set of lexical items open to new additions Dingemanse (2019) [30] amended the definition at 5, “a member of an open lexical class of marked words that depict sensory imagery”. The added element of “openness” in this new definition is very much in line with the reality of East Asian languages. For example, the Chinese doublet dictionary compiled by Wang (1999) [31] takes more than 6700 entries, and the Chinese couplet dictionary compiled by Xie (2011) [32] has more than 6400 entries of doublets and couplets. In China, most of the studies on ideophones are classified under various terms, including Zhuangci状词 [33,34], Maoci貌词 (Ye, 1988) [35], Zhuangmaoci状貌词 (He, 1989) [36], Siyinge四音格 (Tian, 1986) [37], Dieyin/Diezi叠音/叠字 (Zhu, 1936) [38], state adjectives状态形容词 (Zhu, 1956) [39], vivid adjectives形容词生动式 (Lü, 1980) [40], and various code names, such as AA, AB, ABB, AABB, A里AB, etc. In ancient documentary, since the Northern Song Dynasty, various proper names have gradually appeared, for example, Zhang Y. (1985) [41] put forward “couplets联绵字”, Fang (1990) [42] put forward “lianyu謰语” (a kind of couplets) in the late Ming Dynasty. Expressive morphology is a concept proposed by Zwicky and Pullum (1987) [43] when discussing game expressions (secret words, riddles, puns, etc.) and ideophones. According to this theory, when people express the objective world, they will naturally form a picture of things and make a sound matching the picture to describe them. At this time, the sound exhaled may be an ideophone: a phonetic word with strong subjective imagination but matching the imaginary picture. Therefore, Zwicky and Pullum (1987) [43] argue that the words created by expressive morphology are produced by extra-grammatical rules and should “separate from garden-variety morphology, since they involve special phonetics, phonology, syntax, semantics, and pragmatics as well as exhibiting discontinuities with the morphology of the rest of the vocabulary”. On the basis of this understanding, Zwicky and Pullum (1987) [43] saw that the morphology of ideophones is very different from that of ordinary words, and that the rules of expressive morphology are not subject to the same conditions as those of ordinary morphology. So they say: “Expressive morphology is associated with an expressive, playful, poetic, or simply ostentatious effect of some kind”. Dingemanse (2017) [44] added: “Expressive morphology refers to the additive, playful, orderly processes of expressive reduplication and lengthening that we often find applied to ideophonic words in actual instances of use”. Perhaps this is why ideophones are characterized by phonological, structural, and semantic salience. However, such a narrative obviously pushes expressive morphology into the situation of subjective coinage of individual psychology and irregular dependence. Why is this so? In the Book of Poetry, there is a poem describing the scene of a date between a man and a woman: 东门之杨,其叶牂牂。昏以为期,明星煌煌。 dōng mén zhī yáng, qí yè zāng zāng, hūn yǐ wéi qī, míng xīng huáng huáng. On poplars by east gate, the leaves are rustling light. At dusk we have a date; The evening star shines bright. This scene depicts the scene that comes to mind when a lover comes to the East Gate at dusk: beside the city wall, there are big poplars with huge and luxuriant leaves, and the wind rustles. In the poetic realm, the lover looks up and the leaves flutter and shout. The poet could not find a ready-made word to describe such a ringing picture, so he borrowed the character “zāng” to be reduplicated and express it with the rhyme of “yáng”, which was a typical on-the-spot word creation. As time goes into the night, the starry sky shines brightly, and the poet reconstructs a character and a doublet “huánghuáng” to depict the natural scenery and the feelings in his heart. From the point of view of the nature of ideophones, this description is true and reasonable. Language begins with making words. When people feel the world, they will have the need to express things and emotions. They need to refer to things, state actions, describe states and express emotions. The objects involved should have names or codes, and be created and remembered. All of these are carried out under the conditions of sound use and sound restriction. The sound or silent things, actions and states in the objective world are gradually fixed by imaginary pictures that they think are like sounds, which is a spontaneous and natural complex process and a social pragmatic process. This process may be repeated over and over again, so that vocabulary increases and language matures.

4. Conclusions

The expression of the Book of Poetry is so unique that it represents the true face of early Chinese. The doublets, the couplets, the proclitics, the enclitics, and the polysyllabic names of people, places, and clans or tribes may be the witnesses of the early language of the Book of Poetry. This paper explores the question of which is the source and which is the flow of a doublet and a single-character word, aiming to lead to a great proposition: language begins with making words, a method of coinage beyond the rules of contemporary word formation. Contemporary grammar, regardless of its schools, is based on the system established by European languages. However, when we deal with the origins of the archaic doublets in Chinese, none of the inflectional, derivational, clitic, reduplicative, and compound lexical or morphological approaches seem to give a satisfactory answer. On the one hand, this is related to the lack of ideophones in European languages and the lack of relevant theories, on the other hand, it is related to the ancient generation stage of doublets, when Chinese was probably in the early stage of language development and had not yet developed morphological word-formation methods. However, the growth of language depends on coinage or word formation, which is the coinage value of “expressive morphology theory” to doublets described in this paper. Half a century ago, Wang (1980) [45] wisely reminded the academic community that “Chinese is not a monosyllabic language from the beginning; there were already a large number of disyllabic words in the pre-Qin era. Chinese disyllabic words have a special way of word formation; most of them are composed of disyllabic rhymes. The ancients called pure disyllabic words (words that can no longer be analyzed into two morphemes) lianmian words (couplets).” Although Wang only mentioned couplets here, we have no doubt that they can also include doublets, proclitics and so on discussed in this paper. The doublets are not only old words or structures, but also a grand imprint for later generations to explore the long history of Chinese evolution. We hope that the explorers of the Book of Poetry will not stop at the text of the Book of Poetry, nor will they stop at the 3000-year time limit delineated by the oracle bone inscriptions. They are more likely to trace back to the 5000-year ancient history of Chinese language, or even to see through the course of human language in East Asia for more than 10,000 years (Jiang Di, etc. 2014) [46]. When we enter the Book of Poetry and look up at the doublets in awe and veneration, it is as if we are standing in the wilderness of the old world, where the scene of onomatopoeia and mimetics of the doublets and its primitive appearance make us puzzled about the ancient word expressions and leads to a strong desire to explore their origin. No wonder G. Kennedy exclaims that the doublet is the “original form” of archaic Chinese. In other words, it is the origin of Chinese vocabulary, one of the starting points of early Chinese vocabulary, and it records the beginning of Chinese civilization.


1 The terms herein are agreed as follows: A doublet refers to two-homographic characters, which is “重言chóngyán” in Chinese and transliteration. A couplet refers to two-alliterated characters, which is “联绵liánmián”. A proclitic refers to a character with an initial-sound, “词头词cítóucí”, including “有字式yǒuzìshì (characters with initial-sound ‘yǒu’)”; “其字式qízìshì (characters with initial-sound ‘qí’)”, “斯字式sīzìshì (characters with initial-sound ‘sī’)”, etc. An enclitic refers to the structure with a post-auxiliary character, “后附衬字结构chènzì” in Chinese and transliteration. A single character refers to a monosyllabic state adjectives, which is a morpheme or a word. A disyllabic-word (“复音词”) includes couplets, proclitics, enclitics, and compound words. 2 Shi (2009) [47] explained that a doublet is produced by lengthening monosyllabic state words. This view has its roots. Wang Xian [23] said: “this paper basically agrees with Wang Yinzhi’s view that ‘有you, 其qi, 斯si and 思si’ are function words. Their grammatical meaning is that after they are combined with adjectives, they play a role in strengthening adjectives and make the whole structure equivalent to a doublet.” Xiang Xi (1987) [14] also holds this view: a doublet has basically the same meaning as a monosyllabic word, However, a doublet has a stronger descriptive value. 3 This paper builds a database of the doublets and words related to doublets of the Book of Poetry, and calibrates the words with Xiang Xi’s Dictionary of the Poetry (2014) [16]. According to Xiang Xi’s, the homomorphic words of the Poetry include a small number of nouns and verbs, but they have been controversial and excluded in this paper, including “zizi (子子), sunsun (孙孙), chuchu (处处), yanyan (燕燕), yuyu (语语), susu (宿宿), xinxin (信信)”. According to Yang (1979) [4], the monosyllabic adjectives in the Poetry are divided into two categories: property and state ones. The property adjectives can form reduplicated adjectives, such as “gao高 (high)” and “gaogao高高”, and state adjectives can form the doublets. However, this view is controversial, for example, Zhu (1985) [13] believes that “gao高” is really another character”gao皋”. Considering that reduplicating property adjectives are extremely rare, this paper tentatively classifies monosyllabic adjectives as doublets. The total number of doublets is 353. 4 In the era of Book of Poetry, the distinction of lexical catelogies was still chaotic, and the monosyllabic adjective was a general term that actually included not only state adjectives, but also state adverbs, state verbs, as well as cross-category words of adjective-noun, adjective-verb, and adjective-adverb, etc. In this paper, the single words in the Poetry that could serve as modifiers or predicates are included and counted as objects. For example, “Chang昌” can be used as a verb, adjective and noun, but as a noun, it is not taken or counted, such as “zǐ zhī chāng xī, sì wǒ hū táng xī; 子之昌兮,俟我乎堂兮” (国风·郑风), You looked strong and tall, And waited for me in the hall. “chāng” here is not counted as a noun. 5 In this paper, disyllabic words (复音词) includes couplets, proclitics, enclitics, but not doublets. See section 4 for details. 6 The couplets in this paper only refer to the adjective couplets. 7 Kennedy (1964) [10] counted 139, but did not list these doublets in his paper. 8 The database (by Academia Sinica) address is: https://inscription.asdc.sinica.edu.tw/ (accessed on 7 December 2018). 9 When counting single characters, a characters that appears in a compound is considered a single-character. For example, “jī缉” does not appear separately, but only in the couplet “缉熙jīxī”, and the doublet “缉缉jījī”, and the compound “jīyù缉御”. 10 The “其” attached to Table 4 is a proclitic word, and the “其” attached to Table 5 is an enclitic word, which is temporarily classified as couplets in Section 2.3. See Zhou Fagao (1973) [34] (pp. 296–299). 11 A very interesting phenomenon is proposed in a study by Gao (2013, 2014) [48,49], who argues that the syllabic framework of primitive Chinese is C1VC2v, with a single-character when C2 is 0 and bisyllabic-characters when C2 is non-0. The lowercase letter “v” is a rhyming-ending that is extremely easy to fall off. This phenomenon of single-character words in 燕国yānguó and 齐国qíguó dialects of the pre-Qin era is presented as a bisyllabic form in the contemporary Shenyang dialect. For example, the word “磕kē, stammer” is pronounced as “磕巴kēba” in Shenyang dialect. Therefore, Gao believes that early Chinese single-characters may actually contain two syllables. This point seems to be slightly similar to the enclitics in this paper. 12 G. Kenndey points out that “忧yōu worry” appears 82 times but does not reduplicate, indicating that there is no reduplicating production process. Quoted from Zhou Fagao (1973) [34] (p. 103). 13 There must have been a large number of doublets in the era of the Book of Poetry and before the Book of Poetry, some of which even lasted for 3000 years. However, this does not exclude that later generations continue to creat ideophones, but because of the abundance of other word-formation methods, the number of ideophones is gradually decreasing. 14 It refers to the technical difficulties in recording language with characters before the emergence of writing materials and tools such as bamboo slips and silk, and may also include the limited writing space of oracle bone inscriptions and bronze inscriptions.


The paper was presented at the “The 5th Symposium on Unearthed Literature and Archaic Chinese Studies and Chinese History Research” (September 2019, Department of Chinese Language and Literature, Fudan University) and the “Boya Linguistics Forum” (November 2022, Department of Chinese, Peking University). We would like to thank Yinfeng Liang and Baoya Chen for inviting us to attend the meeting. We would like to thank Guofu Long, Fengbin Yang, Changcai Zhao, Zhiping Wang, Ping’an Zhao and Yongxian Luo for their valuable comments on the revision of this paper.


This paper is one of the results of a major project of the National Social Science Foundation of China (21&ZD304), also supported by the National Natural Science Foundation of China “On the Origin and Evolution of East Asian People and Languages: A Globe Linguistic Perspective” (31271337).

Declaration of Competing Interest

The author declares no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.


Cao X. Reduplicating Characters in the Book of Poetry; The Commercial Press: Shanghai, China, 1980.
Zhu G. A Study on Disyllabic Words of the Book of Poetry; Henan People’s Publishing House: Zhengzhou, China, 1985.
Ye, S. The Cultural Interpretation of the Book of Songs - A Study of the Occurrence of Chinese Poetry; Shanxi People’s Publishing House: Xi’an, China, 2005.
Yang J. The State Adjectives in Pre-Qin Chinese.  Stud. Chin. Lang. 1979, 6, 426–436. [Google Scholar]
Sun J. Studies on the Reduplication Morphology of Ancient Chinese; Shanghai Educational Press: Shanghai, China, 2008.
Guo X. The Development of Word-information of Pre-Qin Chinese. Collections of Studies on Chinese History (The Supplement Edition); The Commercial Press: Beijing, China, 2005.
Zhang G. The Study on the Function and Recognition of Modern Chinese Adjective; The Commercial Press: Beijing, China, 2006.
Shen J. Rhythmic Structures and Adjectives in Chinese.  Chin. Lang. Learn. 2011, 3, 3–10. [Google Scholar]
Shi Q. From repetition to reduplication: on the diachronic change of AA patterns to adjectives in Chinese.  Res. Hist. Ling. 2009, 2, 32–53. [Google Scholar]
Kennedy GA. A Note on Ode 220. In Selected Works of George A. Kennedy; Far Eastern Publications Yale University: New Haven, CT, USA, 1964; p. 473.
Shi Q. The Historical Development of the Reduplication forms of Chinese Adjectives; The Commercial Press: Beijing, China, 2010.
Zhou F. The Grammar of Old Chinese: Word Information; Tailianguofeng Press, 1973.
Zhu X. ShijiZhuan by Zhu Xi’ Explanation; Zhonghua Book Company: Beijing, China, 1958.
Xiang X. Studies on Languages of the Book of Poetry; Sichuan People’s Publishing House: Chengdu, China, 1987.
Mao H, Zheng X. Maoshi Zhuanjian. (Calibration by Kong Xiangjun); Zhonghua Book Company: Beijing, China, 2018.
Xiang X. The Dictionary of the Book of Poetry (Revised Edition); The Commercial Press: Beijing, China, 2014.
Xu Y. The Book of Poetry; China Intercontinental Press/Zhonghua Book Company: Beijing, China, 2012.
Qiu, X. Outline of Typography (revised), The Commercial Press: Beijing, China, 2013.
Xu S. The Origin of Chinese Characters; Tianjin Chinese Classics Press: Tianjin, China, 1991.
Wang XQ. Shisanjia Yijishu (Calibration on by Wu G); Zhonghua Book Company: Beijing, China, 1987.
Jiang D. Tibetan Ideophones of ABA’B Model and the Expressive Morphological Theory.  Stud. Chin. Lang. 2021, 6, 673–723. [Google Scholar]
Zhu X. ShijiZhuan by Zhu Xi’ Explanation; Zhonghua Book Company: Beijing, China, 1958.
Wang X. The constructions correspondence to doublets: with you, Qi and Si in the structures.  Lang. Stud. 1959, 4, 9–43. [Google Scholar]
Liu X. A Translation for the Literary Mind and the Carving of Dragons; Zhonghua Book Company: Beijing, China, 1986.
Kilian-Hatz C. Ideophones. In Encyclopedia of Language and Linguistics, 2nd ed.; Elsevier: Oxford, UK, 2006; pp. 508–512.
Dixon RW. Basic Linguistic Theory; Oxford University Press: London; UK, 2010.
Doke, C. Bantu Linguistic Terminology; Longmans, Green, and Co.: London, UK, 1935.
Dingemanse M. The Meaning and Use of Ideophones in Siwu. PhD Thesis, Radboud University: Nijmegen, The Netherlands, 24 October 2011.
Akita K, Pardeshi P. Ideophones, Mimetics, and Expressives; John Benjamins Publishing: Amsterdam, The Netherlands, 2019.
Dingemanse M. ‘Ideophone’ as a comparative concept. In Ideophones, Mimetics, and Expressives; John Benjamins Publishing: Amsterdam, The Netherlands, 2019; pp. 13–33.
Wang W. A Dictionary of Chinese Doublets; Military Yiwen Publishing House: Beijing, China, 1999.
Xie J. A Dictionary of Chinese Coublets; Foreign Language Teaching and Research Press: Beijing, China, 2011.
Wang F, Wang D. Ideophones in Weining Dialect of Hmong language.  Lang. Stud. 1983, 2, 192–211. [Google Scholar]
Zhou F. The Grammar of Old Chinese: Word Information; Tailianguofeng Press, 1973.
Ye M. The Descriptives in Old Chinese.  J. Northwest Normal Univ. 1988, 2, 34–51. [Google Scholar]
He Y. The ideophones in the explanations of Maoshi Zhengyi.  J. Huazhong Normal Univ. 1989, 1, 115–124. [Google Scholar]
Tian D. Four-syllable Words in Tujia Language.  Minority Lang. China 1986, 5, 30–34. [Google Scholar]
Zhu G. The Origin of Poems.  J. East. Countries 1936, 33, 7. [Google Scholar]
Zhu D. A study on Adjective of Modern Chinese.  Lang. Stud. 1956, 1, 1–25. [Google Scholar]
Lü S. (Ed.) Eight Hundred Special Words of Modern Chinese; The Commercial Press: Beijing, China, 1980.
Zhang Y. Fugubian; Shanghai Bookstore (Publishing House): Shanghai, China, 1985.
Fang, Y. Tongya; Zhonghua Book Company: Beijing, China, 1990.
Zwicky AM, Pullum GK. Plain morphology and expressive morphology. In Proceedings of the Thirteenth Annual Meeting of the Berkeley Lingustics Society, 14–16 February 1987; pp. 330–340.
Dingemanse M. Expressiveness and system integration: On the typology of ideophones, with special reference to Siwu.  STUF 2017, 70, 363–385. [Google Scholar]
Wang L. The History of Chinese Language; Zhonghua Book Company: Beijing, China, 1980.
Jiang D, Kang C, Yan H. Evolution of word-syllable structures and the diversity of world languages.  Chin. Sci. Bull. 2014, 59, 3362–3368. [Google Scholar]
Shi Q.  From repetition to reduplication: on the diachronic change of AA patterns to adjectives in Chinese.  Res. Hist. Ling. 2009, 2, 32–53. [Google Scholar]
Gao J. Plosive Finals Retained in Disyllabic Etyma in Shenyang Dialect.  Stud. Lang. Ling. 2013, 2, 50–52. [Google Scholar]
Gao J. Chinese Collections of Archaic Remains of Disyllabic-monomorphomic Words.  NanKai Ling. 2014, 1, 28–39. [Google Scholar]