Improving data handling and analysis in the study of rhyme

Cahiers de Linguistique Asie OrientaleEast Asian Languages and Linguistics

February 25, 2020

Improving data handling and analysis in thestudy of rhyme patterns

Johann-Mattis ListMax Planck Institute for the Science of Human [email protected]

Abstract

By reviewing a recent quantitative study of rhyme patterns in Mandarin Chi-nese, this study shows how data handling and data analysis in the studyof rhyme patterns can be improved. Suggestions for improvement include(a) a consistent annotation of rhyme data, which is exhaustive and facilitatesdata reuse, and (b) emphasizes the importance of automated approachesfor exploratory data analysis, which can help to analyze rhyme data in animproved way, prior to applying statistical frameworks for hypothesis testing.

Keywords

rhyme patterns, Mandarin Chinese, rhyme analysis, data handling

Résumé

En examinant une récente étude quantitative des modèles de rimes en chi-nois mandarin, cette étude montre comment le traitement et l’analyse desdonnées dans l’étude quantitative de rimes peuvent être améliorés. Lessuggestions pour l’amélioration comprend (a) une annotation cohérente desdonnées de rimes, qui est exhaustive et facilite la réutilisation des données,et (b) souligne l’importance des approches automatisées pour l’analyse ex-ploratoire des données, qui peut aider à analyser les données de rimes demanière améliorée, avant d’appliquer des cadres statistiques pour les testsd’hypothèses.

000111222333

Authors copy, accepted with Cahiers de Linguistique Asie Orientale.Please cite as: List, Johann-Mattis (2020): Improving data handling and analysis in the studyof rhyme patterns. Cahiers de Linguistiques Asie Orientale. 49 (1). 1-15.

2 List

Keywords

motifs de rime – chinois mandarin – analise de rimes – traitement des don-nées

1 Introduction

Wang Zhihao’s study on “A Linguistic Study on Rhyming in the BeijingDialect” (this volume) is a considerable contribution to the growingfield of quantitative studies devoted to rhyming in Chinese linguisticsand beyond (Baxter 1992; List 2016; Bu 2019; List et al. 2019). Theamount of data which was annotated by the author in this study is im-pressive and may help encourage future studies on different epochsand varieties of Chinese. There are, however, two points where thestudy could be further improved. First, the data and the rhyme judg-ments are presented in a form that makes it extremely difficult, if notimpossible, to actually check the validity of the proposed rhyme judg-ments, not to speak of allowing scholars to build on the data in theirown research. Second, while the method the author chose to ana-lyze the data with may be in concordance with analysis techniquesproposed in the past by other scholars, the author missed the chanceto apply techniques for exploratory data analysis which allow for anadditional view on the data that may be particularly elucidating.Given the increased interest in quantitative rhyme analysis and the

potential that more, specifically young scholars, will follow up on this,I think it is useful to discuss the two problems in more detail here, asthis may increase the quality of future data handling and encouragescholars to try to think out of the box when it comes to analyzingrhyme data quantitatively. In the following, I will briefly discuss eachof the two points and then conclude by pointing to some future chal-lenges in rhyme annotation.

2 Data handling

When sharing data and code underlying a study, we do this for mul-tiple reasons. First, we want to make sure that the results can bereplicated. When sharing data and code, our colleagues can applythem on their own computers and make sure they achieve the same

Cahiers de Linguistique Asie Orientale February 25, 2020

Data handling and analysis 3

results. Second, we want to allow our colleagues to learn from ourstudy. By investigating our data and code, our colleagues can learnhow we analyzed the data, and can use our analysis to carry out sim-ilar analyses on different datasets. Third, we want to make sure ourcolleagues can challenge our research by inspecting our data andour code and searching for potential errors. As a fourth point, whichis often overlooked, we may want to give our colleagues a chance tobuild on our research, for example, by using our data as a basis thatthey further expand.All four aspects, replication, education, falsification, and amplifica-

tion, are vital for scientific research. In fact, as scientists, we haveall profited in the past from the fact that scientific research – by andlarge – follows these four principles of sustainability, even in thosecases where studies are not based on data and code.While scholars voluntarily subscribe to rigorous principles when it

comes to the citation of previous work, which is fundamental to ac-count for education, replication, and falsification, they are often sur-prisingly lax when it comes to submitting data and code. In manycases, it is hard to find the data underlying studies, since scholarsoften put the information where they can be found in a tiny footnotewithout further comments. Often, it is also hard to access the actualdata, since links may be broken, or data that had been promised tobe provided never made its way online. Having accessed the data,it is often difficult to actually use them, to interoperate them, as in-structions on the basic structure are missing, or data is shared inproprietary formats that can only be used by those who bought theoriginal software. This all makes it extremely difficult to reproduce agiven piece of research. These four aspects of findability, accessibil-ity, interoperability, and replicability are nowadays subsumed underthe label FAIR and propagated as one key guiding principle for thecompilation of research data (Wilkinson et al. 2016).The requirement that data underlying research should be FAIR is

a commonplace in most branches of science now. Although linguis-tics is rather slow in following up, one can note a recent awarenessamong scholars, journal editors, and reviewers, that it is not enoughto submit only the results, if a study itself was based on data andcode. One can also see that in the study byWang, since data and de-tailed instructions on how to arrive at the results, have been archivedwith Zenodo (https://zenodo.org), and instructions on how to carryout the calculations are also provided in the article.


4 List

Strictly speaking, however, Wang’s data is not entirely FAIR, as itis presented in a form that severely hampers interoperability, and ob-structs falsification and amplification. The reasons for this problemlies in the format used for the annotation of the rhyme judgments.When dealing with annotation, there are two basic principles that canbe used, namely stand-off annotation, where the information for agiven resource is provided in another document, and inline annota-tion, where the resource is rendered completely, and information isadded with the help of a markup language inside the original text.Recommendations for rhyme annotation have recently been pro-

posed and outlined by List (List et al. 2019). The proposed solutionconsists of two basic formats for rhyme annotation. The first format isan extended standoff format that renders rhyme data in tabular formalong with the original rhyme texts and allows for a very detailed anno-tation of rhyme sequences, including phonetic alignments (List 2014)of rhyme words. The second format is based on a very straightfor-ward inline annotation schema that was designed for the purposeof allowing scholars to annotate rhyme resources efficiently. Whilethe inline annotation schema is less powerful than the table-basedannotation, it provides enough functionality to annotate rhyme datain the form used by Wang, and it was used to digitize rhyme judg-ments in the work of Baxter (Baxter 1992) and Wang (Wáng 1980),as presented in List et al. (List et al. 2019).When looking at the data prepare byWang, we can see that neither

of the two methods was used directly. Instead, rhyme sequencesare represented in a rather idiosyncratic format in which rhyme se-quences are rendered in a spreadsheet, with one poem (or stanza?)per line. The first three columns of the spreadsheet indicate the pagenumber in the source, from which the rhymes were taken (Jiǎ 2009),the district from which the poem stems, and the source, from whichthe rhyme sequence was taken. The rhyme judgments are then in-dicated in the following columns of the spreadsheet in such a waythat all words that rhyme with each other are placed in consecutivecolumns in the same row, while changes in the rhyme in a givenrhyme sequence (a short poem or similar) are indicated by addingan empty cell in the row.As an example for the data itself, Wang shows the short poem�� kuàizǐ “Chopsticks”, on page 716 in the source, with rhyme wordsmarked in bold font here (underlined in Wang’s text):



Chinese Pīnyīn

�� Cháoyáng qū��, Gēliǎ yībiān dà,�� Jiànmiàn bù shuōhuà.��, Shuōhuà jiù dǐngzuǐ,�� Dǐngzuǐ jiù dǎjià.

The concrete annotation of the data, however, is rendered as fol-lows in Wang’s data:

Page Number District Source Rhyme Sequence

716 �� d(a)4 � h(ua)4 � j(ia)4

Thus, instead of annotating directly which words rhyme, Wang sim-ply lists the rhyme words in this rhyme sequence. While this mayseem to be enough to retrieve the original rhyme judgment, providedthe original resource is available, the format has several shortcom-ings: First, given that one rhyme sequence may have more than onerhyme, Wang has to make use of empty cells in the spreadsheet toindicate that a new rhyme sequence starts within the same poem.This also means that the format is not capable of rendering rhymesthat cross, such as rhyme schemas of the form ABAB, as they canbe frequently met in the literature. Second, since only rhyme wordsare given, but their position is not indicated, it is possible that theannotation itself is ambiguous, since words may be repeated in dif-ferent parts of a poem. This essentially means that it may not bepossible to identify the concrete rhyme judgments from the originalresource by simply looking at the short-cut format we find in Wang’sdata. Third, because it is difficult to understand the concrete rhymejudgments that have been made by the author, it is also difficult to crit-icize and correct them efficiently. Fourth, since the format imposeshuge restrictions on the kind of rhymes that can be annotated in atransparent manner, it cannot be used as a basis for an expanded,derived dataset.While I understand that our recommendations were published only

after Wang already had collected the rhyme data, it is important tonote that any project involving data should be planned in such a waythat the major aspects of replication, education, falsification, and am-


6 List

plification are guaranteed. In order to guarantee them, however, anydata collection needs to be preceded by a thorough planning of thedata collection. Wang’s data has not been wrongly collected per se,but given the size and the beauty of the original collection whichWang used, it is a pity that all this work cannot be easily re-employedin future studies.But how should one code one’s rhyme data in such a way that it can

be reused in later studies? For all analyses, I generally recommendto start with the inline annotation schema proposed in List et al. (Listet al. 2019), since this annotation can be carried out most efficiently,and is therefore ideal to get an analysis started. In this form, thepoem rendered above, would look as shown in Figure 1.

@title: ��@source: ��@page: 716

�� [a] �,�� [a] ��,�� [a] ��

Figure 1: Format for inline rhyme annotation.

Once the data have been annotated in this form, one can use thePoePy Python library (List 2019) and convert them into the more flex-ible table format. PoePy allows for a convenient rendering of rhymeword relations that can be used in publications, as shown in Table 4,which was produced from the inline annotation (code examples areprovided in the supplementary material accompanying this study).With the expanded format, the data can be annotated in many ad-

ditional ways. They can also be conveniently inspected and editedwith the help of the EDICTOR tool (List 2017), which was originallydesigned for the handling of etymological data in historical linguistics,but is likewise apt to handle rhyme sequences, as shown in Figure2.



ID STANZA LINE R:1

1 1.1 ��2 1.1 ��, �3 1.1 �� 4 1.1 ��,5 1.1 ��

Table 4: Tabular display of rhyme data with the help of PoePy.

Figure 2: Inspecting rhyme data in EDICTOR.

3 Data analysis

The analysis reported in Wang’s study is based on the assumptionthat rhyming is based on strict categories. Although it is mentioned inthe study itself that the traditional distinction between strict and freerhyming is unwarranted, and that an additional category should beadded, the author sticks to the idea that rhyming can in some way behandled in a strict way.For this reason, the author also refuses to apply the network-based

approach for the exploratory analysis of rhyme groups in rhyme cor-pora, as proposed in List (2016), emphasizing that:

It is a better choice when the phonology of the object lan-guage is unknown, because any given groups of wordsare based on researchers’ suppositions. However, thismethod is unnecessary when the research object is amodern language. The nature of rhyming suggests thatwords with the same final and tone do rhyme with eachother freely. So, finals and tones are naturally reliable


8 List

groups. The task is just to study the relation between dif-ferent finals or tones. (Wang, this volume, pp. XXX)

Com. Memb. Rhy. Variants

1 341 an en / 1, van / 30, uan / 70, ian / 112, an / 1282 217 ang uang / 42, iang / 67, ang / 1083 220 ao iao / 92, ao / 1284 262 ing uen / 1, uang / 1, ueng / 2, en / 2, in / 4, iong / 11, ong / 63, eng / 84, ing

/ 945 113 ou v / 1, ao / 1, iou / 49, ou / 626 171 a iao / 1, ie / 1, ua / 31, ia / 32, a / 1067 159 i y / 1, uei / 2, w / 12, v / 26, i / 1188 143 uo ua / 1, v / 1, u / 4, e / 51, uo / 869 97 ai ia / 1, anr / 1, uai / 17, ai / 7810 97 en eng / 3, vn / 11, uen / 24, in / 27, en / 3211 104 u u / 10412 95 ianr uor / 1, ier / 1, uair / 1, uar / 2, er / 2, ar / 4, iar / 5, vanr / 6, air / 8, uanr /

13, ianr / 26, anr / 2613 52 uei ei / 23, uei / 2914 50 ueir vnr / 1, uar / 1, anr / 1, uanr / 1, er / 1, vr / 1, engr / 1, wr / 2, inr / 2, uenr

/ 2, r / 2, uor / 2, ver / 2, eir / 3, ur / 3, ier / 3, yr / 4, ir / 5, enr / 6, ueir / 715 47 enr eir / 1, ingr / 1, yr / 2, engr / 2, r / 2, inr / 3, vnr / 3, ueir / 3, vr / 4, wr / ,

uenr / 5, ir / 6, enr / 1116 38 uei v / 1, i / 1, ei / 13, uei / 2317 41 ie ian / 1, e / 2, ve / 6, ie / 3218 25 ianr ir / 1, vanr / 1, air / 1, er / 1, ier / 1, ingr / 2, uanr / 4, ar / 4, ianr / 5, anr / 520 21 ueir anr / 1, vr / 1, ur / 1, enr / 2, yr / 2, uenr / 2, ir / 5, ueir / 7

Table 6: Major communities identified by the Infomap clustering.

This assumption is problematic in many regards. First, the authorassumes that it is possible to know the true phonology of a given di-alect variably, although the data from which the corpus is drawn itselfmay show multiple forms of variation, including not only space (di-atopic variation), but also time (diachronic variation) and social class(diastractic variation), and the individual situation, in which speech isuttered (diaphasic variation). The only situation in which the phonol-ogy of a given language can be reliably measured is when all thesedimensions of linguistic variations are reliably controlled for. Second,the author assumes that the phonology of a given language can pre-dict the rhyming behavior, but does not elaborate on additional fac-tors conditioning rhyming, such as, notably, culture, as representedby the rhyming practice which poets learn from poetry when theygrow up, cognition, as represented by overall principles of phonetic



similarity, and communication, as represented by the linguistic sys-tem the poet speaks or wishes to elicit. Third, the whole process bywhich rhyme data are collected is poorly described by a process inwhich poets pick all meanings they wish to express and then searchfor rhyme words to fill the slots. This process may hold for the pro-cess of poem writing (although the whole picture is beyond doubtmore complex), but the data collected in a rhyme analysis are basedupon the interpretation of a linguist analyzing a given poem. Whatthe poet itself thought – or if poets think at all when forming wordsinto speech – is a question that is not accessible to us.Given that we deal with so many uncertainties when analyzing cor-

pus data of rhyme judgments, it seems problematic to exclusivelyuse techniques for data analysis which assume that rhyme groupsare fixed entities that can be objectively detected prior to the analy-sis. It is all the more surprising, since in the study by Wang itself,it is mentioned that “Semi-free rhyming and free rhyming constitutea continuum. It is difficult to decide where to draw the boundary be-tween them.” (Wang, this volume). While there is nothing to sayagainst the analysis presented in Wang’s study, it is a pity that thealternative approach, which allows us to inspect the data in a muchless biased way, has not been pursued in this study, specifically alsobecause it is by no means difficult to apply the approach describedin List (2016).The major idea proposed in List (2016) is to model rhyme data in

form of a network in which individual rhyme words are representedas nodes and weighted links between the words are used to repre-sent how many times the words rhyme in a given corpus. Once anetwork has been reconstructed in this way, it can be analyzed withthe help of network techniques that search for communities in a givennetwork, with a community being defined by a partition of a networkin which the number of links inside the community is greater thanthat with other nodes in the network (Newman 2004). List (2016)proposed to use the Infomap algorithm by Rosvall and Bergström forthis purpose (Rosvall & Bergstrom 2008). The output of this analy-sis is a division of the network into communities, which, in theory,correspond to distinct rhyme groups in the data. Since it is possiblethat not only Wang, but also other scholars find the network analysisproposed in List (2016) difficult to apply, I thought it may be useful topresent the results of this analysis here.The supplementary material accompanying this study shows ex-


10 List

plicitly how the data of Wang can be converted to a rhyme network,and how this network can be analyzed with the help of the Infomapalgorithm, as proposed in List (2016). The analysis yields a networkconsisting of 2890 rhyme words, which consists of 102 connectedcomponents, with one very large component of 2566 rhyme words.That means that 89% of the rhyme words in the rhyme network areconnected directly or indirectly. This does not only confirm the find-ings in List (2016), where a similarly large connected componentcould be identified, it also shows that rhyming is best understoodas a continuum, and that completely clear rhyme groups are hard toidentify.

腔q(iang)1+

丁d(ing)1

封f(eng)1

空k(ong)1

涨zh(ang)3攻g(ong)1

笼l(ong)2

绷b(eng)1

公g(ong)5

镫d(eng)4

星x(ing)1+呈ch(eng)2

冷l(eng)3窿l(ong)5

象x(iang)4 腕(uan)4

翁(ueng)1+淌t(ang)3凰h(uang)2

洞d(ong)1

恙(iang)4

堂t(ang)2

望(uang)4妨f(ang)2倾q(ing)1糠k(ang)1

翁(ueng)1

容r(ong)2仗zh(ang)4

睛j(ing)5

当d(ang)1

公g(ong)1

屏p(ing)2

粮l(iang)2

兴x(ing)1

檩l(in)3

瓮(ueng)4

楞l(eng)2

弄n(ong)5

控k(ong)4

腔q(iang)1

镜j(ing)4

层c(eng)2轻q(ing)1

秧(iang)1蜓t(ing)2

腚d(ing)4

浓n(ong)2

工g(ong)1 抗k(ang)2

成ch(eng)2

钟zh(ong)1 灯d(eng)1 星x(ing)1

童t(ong)2

囊n(ang)2

嚷r(ang)3磬q(ing)4

长zh(ang)3争zh(eng)1

同t(ong)2

黄h(uang)2

藏c(ang)2

尝ch(ang)2

强q(iang)2

方f(ang)1 阳(iang)2笼l(ong)5

宁n(ing)2 偿ch(ang)2刚g(ang)1

王(uang)2

兵b(ing)1

龙l(ong)2

明m(ing)2

蜂f(eng)1

城ch(eng)2

岭l(ing)3

岗g(ang)3凶x(iong)1

平p(ing)2

纂z(uan)3

丧s(ang)4

账zh(ang)4

让r(ang)4

居j(v)1常ch(ang)2

网(uang)3

娘n(iang)2

亮l(iang)4

帐zh(ang)4

双sh(uang)1

峰f(eng)1 等d(eng)3

晾l(iang)4

成ch(eng)2+

向x(iang)4

样(iang)4

领l(ing)3唐t(ang)2

量l(iang)2

横h(eng)4 恨h(en)4

党d(ang)3

肠ch(ang)2

狼l(ang)2

工g(ong)1+

汪(uang)1

声sh(eng)1

放f(ang)4

墒sh(ang)1

崩b(eng)1

松s(ong)1

长ch(ang)2

相x(iang)4

蒋j(iang)3

铛d(ang)1

响x(iang)3

尾(i)3

弟d(i)5

旁p(ang)2

央(iang)1

腾t(eng)2

第d(i)4

装zh(uang)1

忙m(ang)2

硬(ing)4

亭t(ing)2

踪z(ong)1

命m(ing)4

京j(ing)1

枪q(iang)1

行h(ang)2

雄x(iong)2

升sh(eng)1

床ch(uang)2浆j(iang)1

筝zh(eng)1

证zh(eng)4炕k(ang)4

曲q(v)3疆j(iang)1

腾t(eng)5

亲q(in)1+

洋(iang)2

丰f(eng)1

庞p(ang)2

篱l(i)2

叮d(ing)1

剪j(ian)3

逛g(uang)4

霖l(in)2

案(an)4

密m(i)4

颖(ing)3

敢g(an)3+

夯h(ang)1+

忙m(ang)2+

唱ch(ang)4+静j(ing)4

稳(uen)3

噔d(eng)1

鹰(ing)1

杠g(ang)4

翎l(ing)2

晃h(uang)3供g(ong)4+

防f(ang)2

迎(ing)2

淌t(ang)3+

凳d(eng)4

绳sh(eng)2+

藤t(eng)2顶d(ing)3 正zh(eng)1

坑k(eng)1

光g(uang)1+

想x(iang)3+梆b(ang)1+

凉l(iang)1从c(ong)2

尾(uei)3

虹h(ong)2

晶j(ing)1

凌l(ing)2 兢j(ing)1

囊n(ang)5

通t(ong)1

桩zh(uang)1

青q(ing)1

障zh(ang)4

广g(uang)3

清q(ing)1

棚p(eng)2

蒙m(eng)1

宫g(ong)1

宗z(ong)1

茔(ing)2

烫t(ang)4铃l(ing)2

经j(ing)1

明m(ing)5

正zh(eng)4

行x(ing)2

警j(ing)3

程ch(eng)2

精j(ing)1

壮zh(uang)4

冰b(ing)1慌h(uang)1

殃(iang)1

笙sh(eng)1

昌ch(ang)1

盛ch(eng)2

疼t(eng)2

牲sh(eng)1+

晌sh(ang)3

弹t(an)5屄b(i)1

扮b(an)4+

观g(uan)4

酱j(iang)4+蹭c(eng)4

狱(v)4

具j(v)4

险x(ian)3缘(van)2恋l(ian)4嘀d(i)1拌b(an)4

练l(ian)4

裆d(ang)1

晕(vn)1

碗(uan)3

羡x(ian)4

舱c(ang)1

近j(in)4

面m(ian)4

品p(in)3

愣l(eng)4

伤sh(ang)1

缝f(eng)4

燕(ian)4

奔b(en)1

名m(ing)2爽sh(uang)3

将j(iang)4

场ch(ang)3

辆l(iang)4

香x(iang)1

跟g(en)1

按(an)4

利l(i)4

庄zh(uang)1

民m(in)2

掌zh(ang)5

祥x(iang)2

军j(vn)1很h(en)3

捶ch(uei)2 鸯(iang)1躯q(v)1

容r(ong)2+

丧s(ang)1

雨(v)3

尚sh(ang)5

朵d(uor)5

梆b(ang)1

情q(ing)2

筐k(uang)1

量l(iang)5

题t(i)2

酱j(iang)4

乡x(iang)1

淡d(an)4

拎l(in)1伞s(an)3

沿(ian)2担d(an)4

政zh(eng)4

商sh(ang)1

往(uang)3

瞒m(an)2

芒m(ang)2 笛d(i)2 杆g(an)3 法f(a)3

绊b(an)4

杖zh(ang)4

赏sh(ang)3

章zh(ang)1

僵j(iang)1

降j(iang)4

杏x(ing)4

盛sh(eng)4

帮b(ang)1

粱l(iang)5

粪f(en)4

拧n(ing)5

强q(iang)2+

沉ch(en)2

脓n(ong)2

忍r(en)3

当d(ang)1+

棱l(eng)1

问(uen)4

蹬d(eng)5

嗡(ueng)1

状zh(uang)4

上sh(ang)5+

蚣g(ong)1

棍g(uen)4

惊j(ing)1 诓k(uang)1

棂l(ing)2 菱l(ing)2放f(ang)4+

慌h(uang)1+

生sh(eng)5

当d(ang)4虫ch(ong)2

境j(ing)4甥sh(eng)5

哼h(eng)1

韵(vn)4

绳sh(eng)2

洋(iang)2+

蝇(ing)2

趟t(ang)1

巷x(iang)4

趟t(ang)4

冈g(ang)1

浑h(uen)2

病b(ing)4

饼b(ing)3

瓶p(ing)2

听t(ing)1

营(ing)2

春ch(uen)1

因(in)1

僧s(eng)1

本b(en)3

胧l(ong)2

粱l(iang)2

襟j(in)1

红h(ong)2

动d(ong)4

身sh(en)1

降x(iang)2

筋j(in)1群q(vn)2

仓c(ang)1

狂k(uang)2

荣r(ong)2

赚zh(uan)4

凉l(iang)2唱ch(ang)4

养(iang)3

厂ch(ang)3 上sh(ang)5 光g(uang)1

秤ch(eng)4

鸯(iang)5同t(ong)5

疯f(eng)1

隆l(ong)1

康k(ang)1

填t(ian)2

亡(uang)2

钉d(ing)1登d(eng)1享x(iang)3

庭t(ing)2生sh(eng)1

哄h(ong)1 样(iang)4+

送s(ong)4

轰h(ong)1 整zh(eng)3

楞l(eng)1

挺t(ing)3

量l(iang)5+

双sh(uang)1+弓g(ong)1

厅t(ing)1 脏z(ang)1

盈(ing)2冬d(ong)1

筝zh(eng)5停t(ing)2

拧n(ing)3

铜t(ong)2

障zh(ang)5

滚g(uen)3

塘t(ang)2

锋f(eng)1零l(ing)2

更g(eng)1+鸣m(ing)2

经j(ing)3

兵b(ing)1+

枉(uang)5

雄x(iong)2+

躺t(ang)3

想x(iang)3

苍c(ang)1攮n(ang)3

眶k(uang)1荒h(uang)1

皇h(uang)2

中zh(ong)1+

棒b(ang)4房f(ang)2江j(iang)1扔r(eng)1

宁n(ing)2+ 详x(iang)2

量l(iang)4

箱x(iang)1痛t(ong)4

郎l(ang)2

霜sh(uang)1

扬(iang)2

扛k(ang)2 陵l(ing)2

娘n(iang)5

撞zh(uang)4

煌h(uang)2

嚷r(ang)1

呛q(iang)4档d(ang)3

怆ch(uang)4

挡d(ang)3 创ch(uang)4

两l(iang)3

葬z(ang)4

穷q(iong)2

姓x(ing)4

炀(iang)2

厢x(iang)1杨(iang)2

哼h(eng)5

东d(ong)1 停t(ing)5

衡h(eng)2

敬j(ing)4

邦b(ang)1

罾z(eng)1隆l(ong)2

梦m(eng)4

乡x(iang)2

蹦b(eng)4

晴q(ing)2

拧n(ing)4

碰p(eng)4亮l(iang)5

棠t(ang)2兄x(iong)1

影(ing)3

躬g(ong)1

更g(eng)1

嗓s(ang)3风f(eng)1

情q(ing)5

窗ch(uang)1

征zh(eng)1

能n(eng)2

英(ing)1

赢(ing)2

性x(ing)4

兄x(iong)5

婶sh(en)3

忘(uang)4

榜b(ang)3

张zh(ang)1 上sh(ang)4

场ch(ang)2

省sh(eng)3

丈zh(ang)4糖t(ang)2

勇(iong)3

净j(ing)4

妆zh(uang)1

珑l(ong)2

羊(iang)2

讲j(iang)3

良l(iang)2

胸x(iong)1

耕g(eng)1

膛t(ang)2

忠zh(ong)1

凤f(eng)2

缨(ing)1

精j(ing)5

刑x(ing)2

增z(eng)1

冲ch(ong)1

戎r(ong)2

睁zh(eng)1

蓬p(eng)2

蒸zh(eng)1

咚d(ong)1

灵l(ing)2

营(ing)1

种zh(ong)3终zh(ong)1

烟(ian)1= 间j(ian)1=

边b(ian)1=

桶t(ongr)3 明m(ing)2+

蝈g(uor)1

土t(u)3+

岗g(ang)3+

味(ueir)4=

翅ch(wr)4=翠c(ueir)4=

气q(ir)4=

对d(ueir)4=

棍g(uenr)4=

细x(ir)4=

坠zh(ueir)4=

底d(ir)3=

玉(vr)4=

西x(ir)1=

鸡j(i)1+

七q(i)1+

机j(i)1+

一(i)1+

篮l(anr)2+

簪z(anr)1+ 花h(uar)1+

点d(ianr)3+帘l(ianr)2+

妞n(iour)1+弦x(ianr)2+

玩(uanr)2+

烟(ianr)1+

天t(ianr)1+枣z(aor)3 乐l(e)4+丈zh(ang)4+蹦b(eng)5

河h(e)2+

爬p(a)5馒m(an)2=罐g(uan)4= 碌l(iou)4= 三s(an)1+ 割g(e)1+ 园(vanr)2+ 有(iou)3= 笼l(ong)3=

短d(uan)3=茄q(ie)2=

爷(ie)2= 快k(uai)4+

卤l(u)3=捡j(ian)3=

娶q(v)3= 得d(er)2

晚(uan)3=

右(iou)4数sh(u)3+赛s(ai)4+

卦g(ua)4= 狗g(our)3

栗l(i)4=

饼b(ing)3=虮j(i)3=笼l(ong)2=彩c(ai)3+

庆q(ing)4

斗d(ou)4薯sh(u)3+

方f(ang)1+

桶t(ong)3

毛m(aor)2

套t(aor)2

闯ch(uang)3

喜x(i)3=

党d(ang)3+

挣zh(eng)4

重ch(ong)2

谎h(uang)3

统t(ong)3

锭d(ing)4

澎p(eng)2

掌zh(ang)3

媳x(i)2=红h(ong)2=

粽z(ong)4=

溜l(iour)4

篦b(i)4=聋l(ong)2=虫ch(ong)2=

蜡l(a)4=

辣l(a)4=

坏h(uai)4=懒l(an)3核h(er)2准zh(uen)3+扶f(u)2 九j(iou)3+劈p(i)1抬t(ai)2=

婆p(uor)2

后h(ou)4

台t(ai)2+

壳k(er)2 五(u)3+

圣sh(eng)4

灵l(ing)1

阳(iang)5

矿k(uang)4重zh(ong)4 中zh(ong)1

兴x(ing)4

盅zh(ong)1

功g(ong)1

疮ch(uang)1

形x(ing)2

昂(ang)2

懵m(eng)1

靖j(ing)4

芳f(ang)1

性x(ing)5

葱c(ong)1称ch(eng)1 胖p(ang)4

旺(uang)4

疔d(ing)1

薪x(in)1

洞d(ong)4

哄h(ong)3井j(ing)3订d(ing)4

蓉r(ong)2

冻d(ong)4

凤f(eng)4

应(ing)4

仗zh(ang)4+

熊x(iong)2

笋s(uen)3

冲ch(ong)4像x(iang)4

令l(ing)4横h(eng)2

晃h(uang)4

饭f(an)4+筒t(ong)3

蒙m(eng)2

鬃z(ong)1

琼q(iong)2

缝f(eng)2 定d(ing)4

逢f(eng)2

眼(ian)3+ 钮n(iou)3 对d(uei)4= 孙s(uen)1+ 声sh(engr)1吕l(v)3+

还h(uan)2颜(ian)2潭t(an)2前q(ian)2观g(uan)1剪j(ian)3

坛t(an)2 蒜s(uan)4 三s(an)1 龛k(an)1 锨x(ian)1 毡zh(an)1 断d(uan)4欠q(ian)5 扇sh(an)1

案(an)4

泉q(van)2烟(ian)1碗(uan)3般b(an)1

沾zh(an)1甜t(ian)2

严(ian)2

簪z(an)1

片p(ian)1

癣x(van)3

扮b(an)4砖zh(uan)1

源(van)2 甘g(an)1

煎j(ian)1 剜(uan)1

暄x(van)1

行x(ing)2+灯d(engr)1天t(ian)1=宾b(in)1+妹m(ei)4=威(uei)1然r(an)2+鼓g(u)3+米m(i)3+

关g(uan)1欢h(uan)1圆(van)2 男n(an)2弹t(an)2 喊h(an)3骗p(ian)4 炭t(an)4 厌(ian)4

编b(ian)1 篇p(ian)1 遍b(ian)4 官g(uan)1 算s(uan)1 颠d(ian)1 监j(ian)4缠ch(an)5南n(an)2 算s(uan)4弯(uan)1 缎d(uan)4纂z(uan)3 弹d(an)4

杆g(an)1肝g(an)1团t(uan)2参c(an)1惨c(an)3罕h(an)5杆g(an)3歉q(ian)4 班b(an)1占zh(an)4 旦d(an)4

墙q(iang)2=桌zh(uo)1=谢x(ie)4+盘p(an)2=扣k(ou)4+ 凳d(engr)4 烂l(an)4+跑p(ao)3+

馋ch(an)2= 箩l(uo)2+ 汉h(an)4= 肠ch(angr)2 刀d(ao)1= 娇j(iaor)1 餐c(an)1+ 窝(uo)1+后h(ou)4= 妹m(ei)4+

扇sh(an)4 帆f(an)1

嬉x(i)1 筲sh(ao)1=

箕j(i)1 高g(ao)1=

字z(y)4+

锅g(uor)1 四s(y)4+

袁(van)2 弹t(an)5 馆g(uan)3 兰l(an)2 燕(ian)4

月(ve)4做z(ou)4=

着zh(aor)2觉j(ve)2臭ch(ou)4=

娟j(van)1

抬t(ai)2+ 肉r(ou)4=

瓢p(iaor)2来l(ai)2+

顶d(ing)3=勋x(vn)1崽z(ai)3蛀zh(u)4 走z(ou)3=哪n(ar)3+

銮l(uan)2 缠ch(an)2 栓sh(uan)1敢g(an)3瞒m(an)2 鲜x(ian)1 满m(an)3

老l(ao)3+ 布b(u)4+ 帘l(ian)2= 戒j(ie)4+ 镜j(ingr)4 线x(ian)4+ 亮l(iang)4=锅g(uo)1=

伞s(an)3院(van)4圈j(van)4换h(uan)4罐g(uan)4片p(ian)4看k(an)4

嗒d(a)5多d(uo)1+前q(ian)2=长zh(ang)3+虫ch(ong)5小x(iaor)3败b(ai)4=

接j(ie)1

观g(uan)4迭d(ie)2 甸d(ian)4

完(uan)2

迁q(ian)1

捐j(van)1

穿ch(uan)1

线x(ian)4闲x(ian)2

演(ian)3 盘p(an)2暖n(uan)3

咱z(an)2

胆d(an)3

拌b(an)4 蛋d(an)4 汗h(an)4 穿ch(uan)1+ 变b(ian)4 见j(ian)4 乱l(uan)4 轩x(van)1店d(ian)4匾b(ian)3

牵q(ian)1 拳q(van)2 竿g(an)1 淹(ian)1 担d(an)1软r(uan)3 伸sh(en)1宽k(uan)1散s(an)4

饭f(an)4验(ian)4蝉ch(an)2旱h(an)4间j(ian)1慢m(an)4 圈q(van)1先x(ian)1钐sh(an)4 全q(van)2

悬x(van)2搬b(an)1难n(an)4撵n(ian)3 衫sh(an)1湾(uan)1

焉(ian)1骟sh(an)4钻z(uan)1权q(van)2碱j(ian)3 盐(ian)2

蟾ch(an)2砍k(an)3毽j(ian)4暗(an)4言(ian)2颤ch(an)4馅x(ian)4单d(an)1

炉l(u)2 狠h(en)3+ 蜊l(ir)2 匹p(i)3 毯t(an)3 手sh(ou)3+来l(ai)2=

篮l(an)2

悬x(van)4

卷j(van)3扦q(ian)1员(van)2 冠g(uan)1 箭j(ian)4 办b(an)4棉m(ian)1 咸x(ian)2

蛮m(an)2蹿c(uan)4炼l(ian)4勉m(ian)3坦t(an)5转zh(uan)3涟l(ian)2签q(ian)1馆g(uan)4站zh(an)4 馋ch(an)2

板b(an)3剑j(ian)4仙x(ian)1闪sh(an)3完(uan)1凡f(an)2三s(an)3唤h(uan)4 烂l(an)4 掮q(ian)2

环h(uan)2

短d(uan)3 缘(van)2

鞭b(ian)1 瓣b(an)4 建j(ian)4 反f(an)3 散s(an)3 摊t(an)1 丸(uan)2 选x(van)3绊b(an)4 船ch(uan)2粘zh(an)1 栾l(uan)2 灿c(an)4 善sh(an)4 念n(ian)4

堪k(an)1 褊b(ian)3雁(ian)4扮b(an)4+ 蚕c(an)2产ch(an)3石d(an)4田t(ian)2川ch(uan)1弦x(ian)2贩f(an)4填t(ian)2 甸d(ian)1 翻f(an)1掀x(ian)1

桃t(aor)2

几j(i)3=赶g(an)3=

拣j(ian)3=镰l(ianr)5

毛m(aor)5毡zh(anr)5

草c(aor)5

钱q(ianr)5

羊(iangr)5

宝b(aor)5

蛋d(anr)5

黄h(uang)2+ 快k(uai)4=

菜c(ai)4=

面m(ian)4=

香x(iang)1+

张zh(ang)1+

肚d(u)4=豆d(our)4

铺p(u)4=

裤k(u)4=六l(iour)4

醒x(ing)3

混h(uen)4

埋m(ai)2=丙b(ing)3

阱j(ing)3逮d(ai)3和h(e)2+ 黄h(uangr)2 膘b(iao)1= 脆c(uei)4+ 物(u)4

跪g(uei)4+

腰(iaor)1

药(iao)4=

集j(i)2+

吊d(iao)4=

叉ch(ar)4

房f(ang)2+ 完(uan)2=

犊d(ur)2

猴h(our)2 鞋x(ie)2=肉r(our)4

包b(aor)1

足z(ur)2

头t(our)2

节j(ier)2

球q(iour)2

食sh(wr)5

褂g(uar)4

分f(enr)1

高g(aor)1

油(iour)2

远(van)3= 碟d(ie)2=

散s(an)3=牛n(iour)2

橛j(ve)2=

扁b(ian)3= 瘸q(ve)2=

川ch(uan)2

餐c(an)1

肩j(ian)1

年n(ian)2

咽(ian)4

劝q(van)4恋l(ian)4

棍g(uenr)4+

二(r)4+

气q(ir)4+

地d(ir)4+

棉m(ian)2安(an)1 展zh(an)3贤x(ian)2莲l(ian)2

怨(van)4

谈t(an)2

寒h(an)2

现x(ian)4

玩(uan)2天t(ian)1

窜c(uan)1

穿ch(uan)2

点d(ian)3

寨zh(ai)4+ 婆p(uor)5扯ch(e)3+

馍m(uor)5

娘n(iang)2=

脉m(ai)4+

阳(iang)2=

瞅ch(ou)5

揉r(ou)5

幡f(an)1 痰t(an)2 柑g(an)1

天t(ian)2缆l(an)3 晚(uan)3 搀ch(an)1

含h(an)2绵m(ian)2

殿d(ian)4瘫t(an)1

淡d(an)4

攒z(an)3

然r(an)2

番f(an)1

岩(ian)2看k(an)5

贱j(ian)4

连l(ian)2

拦l(an)2

棺g(uan)1

研(ian)2 钱q(ian)2显x(ian)3眼(ian)3山sh(an)1援(van)2

半b(an)4 燃r(an)2绵m(ian)3辨b(ian)4难n(an)2

球q(iou)2+ 烟(ianr)1=片p(ianr)1=

苗m(iao)2=

帘l(ian)2

镩c(uan)1

丹d(an)1

铣x(ian)1

拴sh(uan)1险x(ian)3喘ch(uan)3 酸s(uan)1

眼(ian)3= 活h(uo)2+驴l(vr)2+

十sh(w)2+

西x(ir)1

挪n(uo)2+

叹t(an)4 段d(uan)4

范f(an)4

按(an)4面m(ian)4转zh(uan)4赚zh(uan)4

凯k(ai)3

落l(uo)4+矮(ai)3

场ch(ang)2+脖b(uo)2+ 者zh(e)3+

帮b(ang)1+窑(iao)2=

糕g(ao)1=

牛n(iou)2+官g(uanr)1=

天t(ianr)1=

肉r(ou)4+

六l(iou)4+灵l(ing)2=

烟(ian)1+

翻f(an)1+

绳sh(eng)2=

瓢p(iao)2+

者zh(e)3

桃t(ao)2+

扯ch(e)3

里l(i)3+

种zh(ongr)3

饼b(ingr)3

姜j(iang)1=

香x(iang)1=

匣x(ia)2=

耙p(a)2=

铞d(iaor)4

火h(uor)3意(i)4+

区q(v)1+

膀b(ang)3=

汁zh(wr)1

来l(ai)3 场ch(ang)3=

绑b(ang)3=

岗g(ang)3=

肉r(our)5

往(uang)3=

口k(our)3

财c(ai)4

冻d(ong)4=样(iangr)4手sh(our)3去q(v)4+堆d(uei)1=

香x(iangr)1

炕k(angr)4

背b(ei)4=亮l(iangr)4

极j(i)2+

窗ch(uangr)1

离l(i)2+ 溶r(ong)2=

冤(van)1

传ch(uan)2 赶g(an)3

娘n(iangr)5

边b(ian)1

滩t(an)1钻z(uan)4

指zh(w)3

余(v)2

待d(ai)4

蹿c(uan)4

钻z(uan)1甸d(ian)1

筲sh(ao)1

理l(i)3

棉m(ian)2

暄x(van)1

罕h(an)3

赶g(an)3

联l(ian)2

掀x(ian)1

天t(ian)2 娟j(van)1

锨x(ian)1

铣x(ian)1

唤h(uan)5

咽(ian)4

鞭b(ian)1

传ch(uan)2

见j(ian)4

幡f(an)1

比b(i)3

三s(an)1

驴l(v)2

谈t(an)2

干g(an)4

咱z(an)2温(uen)1

急j(i)2

川ch(uan)1

蕖q(v)2

是sh(w)4

巨j(v)4

叽j(i)1

叹t(an)4

连l(ian)2酸s(uan)1

避b(i)4

环h(uan)2悬x(van)2

言(ian)2颠d(ian)1

粘zh(an)1然r(an)2

沾zh(an)1边b(ian)1

摊t(an)1

腮s(ai)1

荠q(i)5

搬b(an)1

钱q(ian)2

弦x(ian)2

饭f(an)4=

衣(i)1山sh(an)1

搀ch(an)1

关g(uan)1 绵m(ian)3

仙x(ian)1

糊h(u)5衫sh(an)1

观g(uan)1

鼠sh(u)3颜(ian)2

断d(uan)4

蝉ch(an)2

烦f(an)2

蛐q(v)1

娶q(v)3

换h(uan)4

你n(i)3残c(an)2

晨ch(en)2

许x(v)3

船ch(uan)2

汉h(an)4

砖zh(uan)1

闪sh(an)3

饭f(an)4锯j(v)4

须x(v)1妻q(i)1

里l(i)3

捐j(van)1

集j(i)2

敌d(i)2

地d(i)4

白b(ai)5

梨l(i)2

龛k(an)1

婿x(v)4

皮p(i)2

子z(y)3

欺q(i)1

南n(an)2

悬x(van)4

晚(uan)3

先x(ian)1

裙q(vn)2

细x(i)4

间j(ian)1

烂l(an)4

馅x(ian)4

蜜m(i)4

改g(ai)3蛮m(an)2 盐(ian)2

蚕c(an)2

济j(i)4鸡j(i)1

干g(an)4=

为(uei)2

灌g(uan)4

去q(v)4

内n(ei)4餐c(an)1

片p(ian)1

鱼(v)2瞪d(eng)4

团t(uan)2

算s(uan)1

贱j(ian)4

盔k(uei)1 西x(i)5

厌(ian)4

骑q(i)2

迷m(i)2 尘ch(en)2 离l(i)2

烂l(an)4=

趣q(v)4

席x(i)2

赛s(ai)4

转zh(uan)4=

建j(ian)4

变b(ian)4

记j(i)4

戏x(i)4

己j(i)3

邻l(in)2

体t(i)3七q(i)1

滴d(i)1战zh(an)4戚q(i)1

计j(i)4提t(i)2

淳ch(uen)2

伸sh(en)1俊j(vn)4

梯t(i)1

礅d(uenr)1

黑h(eir)1 椹sh(enr)4

气q(ir)4

乐(ver)4 贝b(eir)4四s(yr)4

的d(er)5屁p(i)4

盖g(ai)4

踢t(i)1 捻n(ian)3=

奶n(ai)3戴d(ai)4

快k(uai)4

医(i)1

带d(ai)4

意(i)4啼t(i)2

气q(i)4

担d(an)1

藜l(i)2

蹁p(ian)2=

泣q(i)4

崖(ia)2

气q(i)5

的d(i)4

掰b(ai)1

息x(i)5棋q(i)2

易(i)4

时sh(w)2

舔t(ian)3

拽zh(uai)4

微(uei)1

姨(i)2

地d(ir)4哉z(ai)1

违(uei)2

会h(ueir)4信x(inr)4

根g(enr)4二(r)4

吹ch(uei)1

池ch(w)2

轩x(van)1

蹄t(i)2

槌ch(uei)2

司s(y)5锛b(en)1

履l(v)3

子z(yr)5

心x(inr)1

群q(vnr)2

根g(enr)1

龟g(ueir)1

门m(en)2

昏h(uen)1

粮l(iang)2+

绑b(ang)3

级j(i)2

匠j(iang)5

门m(en)5

兰l(an)2

乱l(uan)4

提t(i)2+

窄zh(ai)3

膀b(ang)3

功g(ong)1+

林l(in)2

螂l(ang)2

的d(i)1

镶x(iang)1挥h(uei)1

踩c(ai)3

低q(i)1信x(in)4

神sh(en)2孙s(uen)1

用(iong)4

真zh(en)1酸s(uan)2

唇ch(uen)2

寻x(vn)2

翻f(an)1

肩j(ian)1

殿d(ian)4

看k(an)4

段d(uan)4

论l(uen)4

款k(uan)3

蛋d(an)4

乓p(ang)1

馋ch(an)2

嘻x(i)1

散s(an)4 聚j(v)4

选x(van)3

眠m(ian)2

尺ch(w)3

善sh(an)4

绿l(v)4

半b(an)4

算s(uan)4

枝zh(w)1

元(van)2慢m(an)4

洇(in)1

线x(ian)4

汗h(an)4

惯g(uan)4

篱l(i)5

簧h(uang)2喜x(i)3

宜(i)2鹂l(i)2

钢g(ang)1

荡d(ang)4+

齐q(i)2人r(en)2+

难n(an)2

溺n(i)4

单d(an)1

篇p(ian)1

辨b(ian)4

篮l(an)2

冤(van)1

碴ch(a)2

如r(u)2

礼l(i)3

软r(uan)3

归g(uei)1

旗q(i)2

氅ch(ang)3

笛d(i)5

心x(in)1+

阴(in)1

窜c(uan)4

眉m(ei)2

云(vn)2

煎j(ian)1

凡f(an)2

千q(ian)1

厘l(i)2

勉m(ian)3

远(van)3

惨c(an)3

难n(an)4=

端d(uan)1

鲜x(ian)1

匀(vn)2

一(i)1

神sh(en)2+ 岸(an)4眼(ian)3

弹t(an)2银(in)2

魂h(uen)2

吞t(uen)1

人r(en)2

日r(i)4

们m(en)5

姜j(iang)1啼t(i)1

分f(en)1存c(uen)2

吁x(v)1

女n(v)5

况k(uang)4

全q(van)2研(ian)2

寒h(an)2

班b(an)1

焉(ian)1

雁(ian)4饷x(iang)3

圈q(van)1

音(inr)1

粉f(enr)3 讯x(vnr)4

味(ueir)4

推t(uei)1睡sh(uei)4

背b(ei)1

腻n(i)4

岁s(uei)4

画h(uar)4

泪l(ei)3

树sh(ur)4 顺sh(uen)5

吠f(ei)4踹ch(uai)4

闷m(en)1

稀x(i)1

堆d(uei)1

妹m(ei)4

飞f(ei)1

知zh(w)1

煤m(ei)2刺c(yr)4

溃k(uei)4

鼻b(i)2

讯x(vn)4

奔b(en)4

根g(en)1

柜g(uei)4

本b(enr)3

子z(yr)3

字z(yr)3

卫(uei)4

语(v)3

柜g(uei)4=

问(uen)4+

鬼g(uei)3

及j(i)2

腿t(uei)3水sh(uei)3

深sh(en)1

墩d(uen)1

碑b(ei)1

贵g(uei)4

婶sh(enr)3

枕zh(en)3

意(ir)4位(uei)4

会h(uei)4

妇f(ur)5

黑h(ei)1=翅ch(wr)4

泪l(ei)4卜b(uor)5

劲j(inr)4份f(enr)4

人r(enr)2阵zh(enr)4

穿ch(uan)1+

坛t(an)2

演(ian)3

辫b(ian)4

官g(uan)1

严(ian)2

蝠f(u)2

川ch(uan)2

怜l(ian)2

添t(ian)1

罢b(a)4

编b(ian)1

瓦(ua)3

剜(uan)1

碱j(ian)3

转zh(uan)4

涟l(ian)2

串ch(uan)4

癣x(van)3

女n(v)3

安(an)1

哩l(i)5

石d(an)4

占zh(an)4栓sh(uan)1

盘p(an)2

显x(ian)3脸l(ian)3

干g(an)1

玩(uan)2

伴b(an)4甜t(ian)2园(van)2

机j(i)1

颤ch(an)4

田t(ian)2

办b(an)4

点d(ian)3

溪x(i)1

般b(an)1

剑j(ian)4

拦l(an)2劝q(van)4

管g(uan)3敏m(in)3

缸g(ang)1

杯b(ei)1

针zh(en)1缠ch(an)5

淋l(in)2

墙q(iang)2源(van)2

店d(ian)4纲g(ang)1 茫m(ang)2炼l(ian)4

进j(in)4拳q(van)2 耪p(ang)3

罄q(ing)4

瓣b(an)4 旱h(an)4

弯(uan)1

禅ch(an)2息x(i)1荡d(ang)4

斤j(in)1

婚h(uen)1

盆p(en)2

恩(en)1

金j(in)1

伦l(uen)2

种zh(ong)4

汤t(ang)1

裳sh(ang)5

强q(iang)3

匠j(iang)4

梁l(iang)2

文(uen)2

扇sh(an)4棺g(uan)1

坦t(an)5

含h(an)2圈j(van)4

暗(an)4

弹d(an)4

箭j(ian)4贫p(in)2

遍b(ian)4

县x(ian)4潭t(an)2

粉f(en)3

片p(ian)4

奇q(i)2

村c(uen)1

攒z(an)3俭j(ian)3

奸j(ian)1展zh(an)3

万(uan)4

莲l(ian)2

砍k(an)3员(van)2

蒜s(uan)4

转zh(uan)3现x(ian)4

炭t(an)4

怨(van)4

牵q(ian)1

唤h(uan)4

戬j(ian)3

洗x(i)3

的d(i)5

三s(an)3

盼p(an)4

西x(i)1

围(uei)2焚f(en)2

祭j(i)4

均j(vn)1彩c(ai)3

印(in)4

饥j(i)1

追zh(uei)1

代d(ai)4

起q(i)3

击j(i)1

至zh(w)4

逵k(uei)2

黑h(ei)1

屯t(uen)2

贼z(ei)2背b(ei)4灰h(uei)1

败b(ai)4钻z(uanr)1

器q(i)4

句j(v)4

尿s(uei)1

穗s(ueir)4

瑰g(uei)1

肺f(ei)4

醉z(uei)4

禊x(ir)4

论l(uenr)4

事sh(wr)4

雷l(ei)2

份f(enr)2

亏k(uei)1林l(inr)2

水sh(ueir)2座z(uor)4

对d(uei)4得d(ei)3

皮p(ir)2

谁sh(ei)2

霉m(ei)2

唇ch(uenr)2

梅m(ei)2

盆p(enr)2

泪l(eir)4

称ch(en)4

袋d(ai)5

鬓b(in)4

脆c(uei)4

备b(ei)4驹j(vr)1

秸j(ier)1

春ch(uenr)1 紧j(in)3

音(in)1敦d(uen)1

累l(ei)4

罪z(uei)4

危(uei)1

辈b(ei)4

队d(uei)4

巾j(in)1

呸p(ei)1棱l(eng)2

非f(ei)1

宾b(in)1

撑ch(eng)1

朵d(uor)3

苇(uei)3

粒l(ir)4

恒h(eng)2

嘴z(uei)3

吨d(uen)1

关g(uanr)1

退t(uei)4

啡f(ei)1

赔p(ei)2

心x(in)1喂(uei)4

轮l(uen)2

跪g(uei)4

民m(in)2+

北b(ei)3

税sh(uei)4

美m(ei)3

朋p(eng)2衷zh(ong)1

平p(ing)2+

臣ch(en)2

农n(ong)2

劲j(in)4

息x(ir)5

诊zh(en)3

津j(in)1

亲q(in)1

里l(i)5

君j(vn)1

新x(in)1顺sh(uen)4

坟f(en)2

穰r(ang)2姻(in)1

凭p(ing)2

抢q(iang)3

痕h(en)2

饮(in)3

斟zh(en)1

参sh(en)1

坤k(uen)1

塞s(air)1

扮b(anr)4

伞s(anr)3

影(ingr)3

半b(anr)4

幡f(anr)1

纂z(uanr)3

三s(anr)1

奶n(air)5

鸦(iar)1

签q(ianr)1

带d(air)4

瓣b(anr)4

话h(uar)4

管g(uanr)3

伴b(anr)4

灰h(ueir)1

簪z(anr)1

瓜g(uar)1

妈m(ar)1

针zh(enr)1

八b(ar)1

衩ch(ar)3

井j(ingr)3

芽(iar)5

海h(air)3

色sh(air)3

干g(anr)1

院(vanr)4

鞍(anr)1

担d(anr)4

毽j(ianr)4

车ch(er)1

辫b(ianr)4

线x(ianr)4

扇sh(anr)4

鞋x(ier)2

铲ch(anr)3

被b(ei)4

槌ch(uei)5

费f(ei)4

类l(ei)4

配p(ei)4

媒m(ei)2

被b(eir)4

贝b(ei)4

随s(uei)2

催c(uei)1

垂ch(uei)2

肥f(ei)2

对d(ueir)4

锤ch(uei)2

军j(vnr)1

腿t(ueir)3

字z(yr)4米m(ir)3

坠zh(ueir)4

月(ver)4

贴t(ier)4

致zh(wr)4

髻j(ir)4

棍g(uenr)4

儿(r)2

嘴z(ueir)3

数sh(ur)4

村c(uenr)1

闷m(enr)4

墩d(uenr)1

罪z(ueir)4

趣q(vr)4

魁k(uei)2

帅sh(uai)4

惧j(v)4

实sh(w)2低d(i)1

子z(y)5

悲b(ei)1

回h(uei)2

底d(i)3

奈n(ai)4晒sh(ai)4

爱(ai)4

债zh(ai)4

队d(ueir)4 敢g(an)3

在z(ai)4

替t(i)4

丽l(i)4辟p(i)4

挤j(i)3

外(uai)4

怪g(uai)4

坏h(uai)4卖m(ai)4 材c(ai)2

葵k(uei)2

龟g(uei)1

坠zh(uei)4拜b(ai)4

已(i)3

治zh(w)4

浅q(ian)3

赖l(ai)4

宅zh(ai)2

在z(ai)5

素s(u)4

次c(y)4

歇x(ie)1

哀(ai)1

刺c(y)4

色s(e)3

别b(ie)2

点d(ian)3=

猜c(ai)1

卖m(ai)5

择zh(ai)2

四s(y)4

涕t(i)4

槐h(uai)2

材c(ai)5

采c(ai)3

伯b(ai)1

乖g(uai)1 紫z(y)3载z(ai)3

拐g(uai)3

缺q(ve)1

邪x(ie)2

谢x(ie)4

裂l(ie)4

顾g(u)4

卤l(u)3

蝶d(ie)2

富f(u)4

茄q(ie)2

鳖b(ie)1

务(u)4

斜x(ie)2

铺p(u)4

蝈g(uo)1

醋c(u)4

梳sh(u)1

楚ch(u)3

碟d(ie)2

署sh(u)3

买m(ai)3

籽z(y)3柿sh(w)4

古g(u)3

股g(u)5

竹zh(u)2

咧l(ie)5

截j(ie)2

结j(ie)2

戒j(ie)4

股g(u)3

夜(ie)4

父f(u)4

叶(ie)4

库k(u)4

撇p(ie)3

睬c(ai)3

翅ch(w)4肝g(an)1该g(ai)1

义(i)4

百b(ai)3

栽z(ai)1

绕r(aor)4

帽m(aor)4

迟ch(w)2

字z(y)4

驰ch(w)2

反f(an)3

霓n(i)2

日r(w)4 完(uan)2

缎d(uan)4

规g(uei)1

去q(v)5

窜c(uan)1

吃ch(w)1 基j(i)1年n(ian)2

俐l(i)4

台t(ai)2

摘zh(ai)1

苔t(ai)2

来l(ai)2

开k(ai)1

延(ian)2死s(y)3

季j(i)4

菜c(ai)4

线x(ian)4=

裁c(ai)2

渠q(v)2

堤d(i)1

拍p(ai)1 缆l(an)3

碓z(ueir)1

袄(aor)3

胆d(anr)3

泥n(ir)2

身sh(enr)1

裤k(ur)4

思s(yr)5

鸡j(ir)1

枝zh(wr)1

儿(r)5

耳(r)3

绳sh(engr)2

轨g(ueir)3

曲q(vr)3

笛d(ir)2

滚g(uenr)3

姨(ir)2

横h(engr)2

仁r(enr)2

褶zh(er)3

姐j(ier)3

馅x(ianr)4

燕(ianr)4角j(iaor)3

堆z(ueir)2

女n(vr)5

杏x(ingr)4

下x(iar)4

会h(ueir)3

神sh(enr)2

迹j(ir)4

驴l(vr)2哏g(enr)2吁(vr)4

起q(ir)3

瑞r(uei)4

竿g(anr)1

喜x(ir)3

咧l(ier)1

籽z(yr)3

底d(ir)3

鬼g(ueir)3

尾(ir)3

食sh(wr)2

粪f(enr)4

顿d(uenr)4

夔k(ueir)2

水sh(ueir)3门m(enr)2

吃ch(wr)1

集j(ir)2

身sh(enr)2

拉l(ar)3

短d(uanr)3

串ch(uanr)4眼(ianr)3

裙q(vnr)2

板b(anr)3

盖g(air)4

腕(uanr)4

马m(ar)3

脸l(ianr)3

袋d(air)4

屎sh(w)3

活h(uor)2

待d(ai)5

识sh(w)5

食sh(w)5

套t(aor)4

十sh(w)2

芽(iar)2

寨zh(ai)4

道d(aor)4挨(ai)1

抬t(ai)2

怀h(uai)2

麦m(ai)4

寺s(y)4

庙m(iaor)4

袢p(anr)4

呆d(ai)1

石sh(w)2

匙ch(w)2

害h(ai)5

篮l(anr)2

挨(ai)2

色sh(ai)3 派p(ai)4

埋m(ai)2轿j(iaor)4

太t(ai)4

孩h(air)2

圾j(i)1

乱l(uan)4=

排p(ai)2闹n(aor)4

直zh(w)2

扦q(ianr)1

筷k(uai)4

袋d(ai)4

批p(i)1

海h(ai)3

衬ch(en)4吉j(i)2

抱b(aor)4

嘎g(ar)2快k(uai)5 财c(ai)2 孩h(ai)2

才c(ai)5

白b(ai)2

瓠h(u)4

柴ch(ai)2

袁(van)2歪(uai)1

哨sh(aor)4

害h(ai)4块k(uai)4

牌p(ai)2 纸zh(w)3

迈m(ai)4

船ch(uanr)2

闸zh(ar)2

弹t(anr)2

镰l(ianr)2

茶ch(ar)2

宰z(ai)3

玩(uanr)2

杆g(anr)1

砖zh(uanr)1

圈q(vanr)2

捻n(ianr)3

前q(ianr)2

班b(anr)1

哪n(ar)3

家j(iar)1

丫(iar)1

托t(uor)1

衫sh(anr)1

肩j(ianr)1

盘p(anr)2

绢j(vanr)4

片p(ianr)4

面m(ianr)4

杆g(anr)3

卷j(vanr)3

边b(ianr)1

筷k(uair)4

饭f(anr)4

花h(uar)1

练l(ianr)4

馆g(uanr)3

官g(uanr)1

炭t(anr)4

摊t(anr)1

欢h(uanr)1

干g(anr)4

鞭b(ianr)1

圈q(vanr)1

鲜x(ianr)1

弯(uanr)2

园(vanr)2

檐(ianr)2

宪x(ianr)4

歌g(er)1

尖j(ianr)1

宅zh(air)2

弯(uanr)1

蛮m(anr)2

年n(ianr)2

烟(ianr)1

台t(air)2

钱q(ianr)2

辉h(uei)1

天t(ianr)1

帘l(ianr)2

山sh(anr)1

颤ch(anr)4

价j(iar)4

沿(ianr)2

筝zh(engr)5

盐(ianr)2

蛋d(anr)4 蒜s(uanr)4

菜c(air)4

颠d(ianr)1

罐g(uanr)4

缘(vanr)2

善sh(anr)4

庵(anr)1

碗(uanr)3

点d(ianr)3 府f(u)3服f(u)5

血x(ie)3

灭m(ie)4

杰j(ie)2节j(ie)2

裤k(u)4

袱f(u)5

布b(u)4

兔t(u)4

咧l(ie)1

鞋x(ie)2

宣x(van)1 饹g(e)5

捕b(u)3

铁t(ie)3

爹d(ie)1

土t(u)3

驹j(v)1

筵(ian)2

奶n(ai)5

数sh(u)3

镯zh(uo)2掖(ie)1

街j(ie)1

舞(u)3

参c(an)1

玄x(van)2

适sh(w)4

竿g(an)1

福f(u)2

煮zh(u)3

薯sh(u)3

稻d(ao)4

毒d(u)2

满m(an)3

迹j(i)4

泥n(i)2

喊h(an)3

瘫t(an)1埠b(u)5

诉s(u)4

番f(an)1杆g(an)1

米m(i)3权q(van)2

眯m(i)1

据j(v)4

甩sh(uai)3

湾(uan)1 序x(v)4

钗ch(ai)1

才c(ai)2

赎sh(u)2

瘸q(ve)2

之zh(w)1

李l(i)3

灾z(ai)1

断d(uan)4=

笔b(i)3

坯p(i)1

筛sh(ai)1

摆b(ai)3

弟d(i)4事sh(w)4

局j(v)2

食sh(w)2埃(ai)1

拆ch(ai)1

蒲p(u)2

咸x(ian)2

欢h(uan)1

丹d(an)1

前q(ian)2

烟(ian)1

堪k(an)1

滩t(an)1

看k(an)5岩(ian)2

萨s(a)5

镐g(ao)3

銮l(uan)2

贤x(ian)2

产ch(an)3

黎l(i)2

埠b(u)4

贩f(an)4

嘎g(a)2糊h(u)2

究j(iou)1峡x(ia)2

嗖s(ou)1

饱b(ao)3

流l(iou)2

志zh(w)4 粗c(u)1

帘l(ian)2

爷(ie)5

不b(u)4

恶(u)4潲sh(ao)4

部b(u)4

伍(u)5

魔m(uo)2

术sh(u)4

愁ch(ou)2

赌d(u)3

锣l(uo)2

住zh(u)4

果g(uo)3

鼬(iou)4

犊d(u)2

步b(u)4

堵d(u)3

学x(ve)2

咕g(u)5

呜(u)1

爷(ie)2

骨g(u)3姐j(ie)3

砚(ian)4

罐g(uan)4

扇sh(an)1

扦q(ian)1

燃r(an)2

穿ch(uan)1

掮q(ian)2

绵m(ian)2

夹j(ia)2

华h(ua)2

缴j(iao)3

痰t(an)2圆(van)2

拖t(uo)1

柱zh(u)4

户h(u)4

箩l(uo)5

业(ie)4

屋(u)1+

铺p(u)4+

隔g(e)2夫f(u)5+

割g(e)1

窝(uo)5 婆p(uo)5

乐l(e)4

做z(uo)4

磕k(e)1

涂t(u)5+

鸽g(e)1

托t(uo)1

窝(uo)1

磨m(uo)2

垛d(uo)4

阁g(e)2

雪x(ve)3

路l(u)4

何h(e)2

沫m(uo)4

罗l(uo)2

个g(e)4

和h(e)2

芦l(u)2固g(u)4

坷k(e)1

咐f(u)4

哥g(e)1

屋(u)1

嗑k(e)1

婆p(uo)2

泼p(uo)1

格g(e)1

热r(e)4

出ch(u)1+

剁d(uo)4

波b(uo)1

受sh(ou)4+

哭k(u)1+

卜b(uo)5

捉zh(uo)1

涂t(u)5

柁t(uo)2

助zh(u)4

梭s(uo)1

河h(e)2

沫m(uo)5

处ch(u)4

箍g(u)1

国g(uo)4

输sh(u)1

个g(e)3

可k(e)3

虬q(iou)2

喝h(e)1

忧(iou)1

活h(uo)2

磨m(uo)4

卧(uo)4

阔k(uo)4

蛄g(u)1

虎h(u)3

丑ch(ou)3澡z(ao)3 吼h(ou)3

烙l(ao)4

饽b(uo)5

壳k(e)2

嗦s(uo)5

绺l(iou)3

砣t(uo)2

酪l(ao)4

德d(e)2

脑n(ao)3

脖b(uo)2

靠k(ao)4

首sh(ou)3

喵m(iao)1

扭n(iou)3

黝(iou)1

武(u)3

舅j(iou)5

秃t(u)1

旧j(iou)4

瓜g(ua)5

僚l(iao)2肚d(u)4

苦k(u)3 主zh(u)3

腐f(u)5

篓l(ou)3

馍m(uo)2坐z(uo)4

臭ch(ou)4

搓c(uo)1

够g(ou)4

傲(ao)4

着zh(e)5

髅l(ou)2

姣j(iao)1

胡h(u)2

秋q(iou)1

闹n(ao)5

友(iou)3

到d(ao)5

购g(ou)4

护h(u)4

破p(uo)4

溜l(iou)4

说sh(uo)1

落l(uo)4

活h(uo)5

舅j(iou)4

禾h(e)5

国g(uo)2

得d(e)2

簌s(u)4

螺l(uo)2

鼓g(u)3

咕g(u)1五(u)3

吐t(u)3

妇f(u)5

母m(u)3

宿x(iou)3

烛zh(u)2

跷q(iao)1

夫f(u)1

补b(u)3

揪j(iou)1

欧(ou)5

溜l(iou)5

说sh(uo)1+

噜l(u)1

莫m(uo)4

果g(uo)5

暑sh(u)3朵d(uo)3

哥g(e)1+

锅g(uo)1 呵h(e)1

摸m(uo)1

驼t(uo)2

郭g(uo)1

务(u)5

我(uo)3

搁g(e)1饿(e)4 戈g(e)1 鹅(e)2

错c(uo)4

掇d(uo)5

禾h(e)2

拙zh(uo)2

簸b(uo)4

棵k(e)1

抹m(uo)3

泊b(uo)2

播b(uo)1过g(uo)4

恶(e)4

柯k(e)1

伙h(uo)3

饽b(uo)1

摞l(uo)4

和h(e)4

涩s(e)4

咯g(e)1

核h(u)2

坨t(uo)2

客k(e)4

误(u)4窠k(e)1

合h(e)2

夺d(uo)2

蜇zh(e)2

伯b(uo)2

握(uo)4

喝g(e)1

撮c(uo)1

喝h(e)5

我(uo)3+

挪n(uo)2

膊b(uo)2

户h(u)5

火h(uo)3

呼h(u)1

锁s(uo)3

缩s(uo)1

渴k(e)3

景j(ing)3

鞍(an)1

难n(an)4

胆d(an)3

骗p(ian)4

暖n(uan)3

浪l(ang)4

穹q(iong)2

短d(uan)3

禧x(i)3

站zh(an)4

监j(ian)4

鸦(ia)1

画h(ua)4

冠g(uan)1

泉q(van)2

咱z(an)2+

男n(an)2

半b(an)4+

间j(ian)1+

瑕x(ia)2

差ch(a)1

馆g(uan)4

立l(i)4

簪z(an)1

镰l(ian)2

甸d(ian)4

看k(an)4+

取q(v)3

秸j(ie)1

栾l(uan)2

还h(uan)2

闲x(ian)2

脾p(i)3

院(van)4

傅f(u)5

榴l(iou)5

跃(iao)4

淹(ian)1

出ch(u)1初ch(u)1

午(u)3

窕t(iao)3

柑g(an)1

猪zh(u)1

吒zh(a)1

范f(an)4

念n(ian)4

艳(ian)4

馆g(uan)3

稼j(ia)4

甘g(an)1

漆q(i)1

鬟h(uan)5

尖j(ian)1

拄zh(u)3

援(van)2

伏f(u)2

蟾ch(an)2

实sh(w)5

拴sh(uan)1

宽k(uan)1

拟n(i)3

褊b(ian)3

天t(ian)1

骟sh(an)4

钐sh(an)4

签q(ian)1

哒d(a)1

毡zh(an)1

切q(ie)1

蓝l(an)2

淀d(ian)4

边b(ian)1+

无(u)2

旦d(an)4钻z(uan)4

喘ch(uan)3

除ch(u)2

缠ch(an)2

稼j(ia)1

撵n(ian)3

琶p(a)5

板b(an)3

原(van)2

嫌x(ian)2

揖(i)1

献x(ian)4

蛤h(a)2

捞l(ao)1

鸟n(iao)3

锄ch(u)2

摔sh(uai)1

箩l(uo)2

丘q(iou)1

挑t(iao)1

棉m(ian)1

逗d(ou)4

完(uan)1

镩c(uan)1

鸭(ia)1

手sh(ou)3

了l(iao)3

砸z(a)2

乍zh(a)4

留l(iou)2

焦j(iao)1

牢l(ao)2

道d(ao)4

岔ch(a)4

吕l(v)3

压(ia)1

卷j(van)3

电d(ian)4

歉q(ian)4

角j(iao)3

搞g(ao)3

爸b(a)5

靴x(ve)1

标b(iao)1

兽sh(ou)4

书sh(u)1

帆f(an)1

欠q(ian)5

萨s(a)4

灿c(an)4

脊j(i)3

扮b(an)4

天t(ian)1+

插ch(a)1

大d(a)4

檐(ian)2

裟sh(a)1

砂sh(a)1

麻m(a)2

担d(an)1+

话h(ua)4+

达d(a)2

刮g(ua)1

娃(ua)2

搭d(a)1

下x(ia)5

毽j(ian)4

架j(ia)4

加j(ia)1

八b(a)1

咱z(a)2

打d(a)3=

鸹g(ua)1

啦l(a)5

撒s(a)1

验(ian)4

园(van)2+

叭b(a)1

嫁j(ia)4

痧sh(a)1

穿ch(uan)2

哪n(a)3

牙(ia)2

罚f(a)2

哈h(a)1

田t(ian)2+

匾b(ian)3

发f(a)1

减j(ian)3

扎z(a)1

话h(ua)4

掐q(ia)1

法f(a)5

干g(an)4+

搽ch(a)2

拔b(a)2

便b(ian)4+

嗒d(a)1

轧(ia)4

腊l(a)4

霸b(a)4

胯k(ua)4

仨s(a)1+

全q(van)2+

爸b(a)4

麻m(a)2+

瓜g(ua)1+

呀(ia)5

他t(a)1+

麻m(a)2=

吗m(a)5

卦g(ua)4

麻m(a)5

善sh(an)4+

驾j(ia)4

家j(ia)1+

妈m(a)1+

八b(a)1+

花h(ua)1+

袜(ua)4

嘛m(a)5

伐f(a)2

沙sh(a)1

褂g(ua)4

哑(ia)3

叉ch(a)1

献x(ian)4+

滑h(ua)2

杀sh(a)1

抓zh(ua)1

丫(ia)1

乏f(a)2

哇(ua)1

喇l(a)3

把b(a)3

骂m(a)4

叭b(a)5

夸k(ua)1

差ch(a)4

发f(a)5

貌m(ao)4

邋l(a)1瓜g(ua)1

汉h(an)4+

巴b(a)5

吧b(a)5

嘎g(a)5

嫂s(ao)3=

小x(iao)3=

好h(ao)3=

教j(iao)4

妙m(iao)4=

薄b(ao)2

枣z(ao)3=

罩zh(ao)4

脚j(iao)3=

烤k(ao)3=饺j(iao)3=

笑x(iao)4岛d(ao)3

校x(iao)4

少sh(ao)3

妞n(iou)1

教j(iao)1

跑p(ao)3 盗d(ao)4

小x(iao)3

料l(iao)4

好h(ao)3

冒m(ao)4

少sh(ao)4

寿sh(ou)4

高g(ao)1

要(iao)4

遥(iao)2

扣k(ou)4 撅j(ve)1

阶j(ie)1

珠zh(u)1

足z(u)2

烤k(ao)3

淘t(ao)2

怒n(u)4

痨l(ao)2+

瞧q(iao)2

好h(ao)3+

硝x(iao)1

妙m(iao)4

袄(ao)3

道d(ao)5

轿j(iao)4

掉d(iao)5

捎sh(ao)1

灶z(ao)4谣(iao)2牛n(iou)2刀d(ao)1

夫f(u)5

求q(iou)2

晓x(iao)3

告g(ao)4 票p(iao)4

觉j(iao)4

浇j(iao)1

撒s(a)3

图t(u)2

窖j(iao)4

蒿h(ao)1

庙m(iao)4

叫j(iao)4

袍p(ao)2

嚎h(ao)2

宝b(ao)3

苗m(iao)2

绕r(ao)4

药(iao)4

交j(iao)1

㞎b(a)3

草c(ao)3

烧sh(ao)1

羞x(iou)1

妞n(iou)5

油(iou)2

谷g(u)3树sh(u)4

歌g(e)1

秀x(iou)4

科k(e)1泡p(ao)5

猴h(ou)2

萝l(uo)2

沟g(ou)1

狗g(ou)3

蕉j(iao)1

仇ch(ou)2狗h(ou)3

豆d(ou)4

呵h(e)5

周zh(ou)1

坡p(uo)1

孝x(iao)4

舍sh(e)3

勒l(e)4

瞅ch(ou)3

勾g(ou)1痨l(ao)2

羔g(ao)1劳l(ao)2

奴n(u)2

噜l(u)5

有(iou)3

着zh(uo)2

溜l(iou)1

多d(uo)1

车ch(e)1

丢d(iou)1 所s(uo)3

座z(uo)4

哭k(u)1

受sh(ou)4

辽l(iao)2

佛f(uo)2

烧sh(ao)5

瘦sh(ou)4

毫h(ao)2

熬(ao)2胞b(ao)1

熟sh(ou)2州zh(ou)1

拃zh(a)3

她t(a)1

年n(ian)2+

迁q(ian)1

话h(ua)5

扎zh(a)1

塔t(a)3

俩l(ia)3

探t(an)4=

散s(an)3

丸(uan)2 花h(ua)1

马m(a)3=

啦l(a)1

煞sh(a)4

爬p(a)2

啊(a)4

山sh(an)1+

拉l(a)1

么m(a)5

傻sh(a)3

他t(a)1

洒s(a)3

娃(ua)5

呱g(ua)1

蟆m(a)5

茶ch(a)2

擦c(a)1

马m(a)3

寡g(ua)3

发f(a)4

刷sh(ua)1

寇k(ou)4

巴b(a)1

趴p(a)1

稼j(ia)5妈m(a)1

芽(ia)2 家j(ia)5

挖(ua)1

遢t(a)5

把b(a)4

仨s(a)1

翘q(iao)4

倒d(ao)3

宝b(ao)2

扫s(ao)3

前q(ian)2+渣zh(a)1

拉l(a)5

化h(ua)4

铰j(iao)3

刨p(ao)2

瘩d(a)5

战zh(an)4+

纱sh(a)1

洼(ua)1

担d(an)4+

保b(ao)3

怕p(a)4

挂g(ua)4

铡zh(a)2杷p(a)5

豹b(ao)4

展zh(an)3+

吓x(ia)4

蜡l(a)4

掉d(iao)4=

打d(a)3

家j(ia)1

下x(ia)4

价j(ia)4

缥p(iao)1

搅j(iao)3

它t(a)1

爪zh(ua)3

呀(ia)1

衙(ia)2

兰l(an)2+厦sh(a)4

袖x(iou)4

糟z(ao)1

罗l(uo)5

久j(iou)3厚h(ou)4

兜d(ou)1

口k(ou)3

吵ch(ao)1

造z(ao)4+

学x(iao)2

高g(ao)1+

鹿l(u)4导d(ao)3 雀q(iao)3 柳l(iou)3

绣x(iou)4

娇j(iao)1

帽m(ao)4

桥q(iao)2

樵q(iao)2

捣d(ao)3

由(iou)2

毛m(ao)2

包b(ao)1

搜s(ou)1

着zh(ao)2

盒h(e)2

报b(ao)4诌zh(ou)1

姥l(ao)5摇(iao)2

敲q(iao)1

操c(ao)1飘p(iao)1

条t(iao)2

叟s(ou)3

优(iou)1

炮p(ao)4

闹n(ao)4

宵x(iao)1

遭z(ao)1

巧q(iao)3

套t(ao)4

脚j(iao)3

耗h(ao)4

消x(iao)1

猫m(ao)1

泡p(ao)4

俏q(iao)4

表b(iao)3

咬(iao)3

叔sh(u)1

邀(iao)1

尿n(iao)4

收sh(ou)1

约(iao)1刁d(iao)1

秒m(iao)3

条t(iao)5

讨t(ao)3

辣l(a)4

虾x(ia)1

糙c(ao)1

早z(ao)3妈m(a)5咂z(a)1

罕h(an)5拿n(a)2

豪h(ao)2+

赵zh(ao)4

钓d(iao)4

描m(iao)2

乎h(u)5

喳zh(a)1

哨sh(ao)4

跤j(iao)1

瞎x(ia)1

钩g(ou)1

萄t(ao)5

吊d(iao)4跳t(iao)4啕t(ao)2

彪b(iao)1幽(iou)1郊j(iao)1

楼l(ou)5

窑(iao)2

壶h(u)2姑g(u)1

吵ch(ao)3

挎k(ua)4

膘b(iao)1

疃t(uan)3

恼n(ao)3嫂s(ao)3

薅h(ao)1暴b(ao)4 游(iou)2

撂l(iao)4

堡b(ao)3

驼t(uo)5楼l(ou)2

绸ch(ou)2

找zh(ao)3

酒j(iou)3曹c(ao)2

陶t(ao)2

壕h(ao)2

姥l(ao)3

摸m(ao)1

揍z(ou)4休x(iou)1

鳅q(iou)5

走z(ou)3

鬏j(iou)1

挑t(iao)3

九j(iou)3

透t(ou)4

招zh(ao)1

绞j(iao)3

擞s(ou)3

跑p(ao)3=

揉r(ou)2

娥(e)2

雹b(ao)2

爆b(ao)4

折sh(e)2

粥zh(ou)1

偷t(ou)1

号h(ao)4

到d(ao)4

糕g(ao)1

球q(iou)2

梢sh(ao)1

头t(ou)5

枣z(ao)3

膏g(ao)1

老l(ao)3

豪h(ao)2

造z(ao)4

辙zh(e)2

囚q(iou)2

饶r(ao)2

肉r(ou)4

鹞(iao)4=

抛p(ao)1

漂p(iao)1

修x(iou)1

照zh(ao)4

腰(iao)1

豆d(ou)5

货h(uo)4

锹q(iao)1 芦l(u)5

头t(ou)2

着zh(ao)1

悠(iou)1

凑c(ou)4

萧x(iao)1

端d(uan)1

练l(ian)4烦f(an)2

万(uan)4

疃t(uan)3 原(van)2

玄x(van)2

眠m(ian)2

联l(ian)2

罕h(an)3

串ch(uan)4

淀d(ian)4岸(an)4

千q(ian)1

边b(ian)1+

筵(ian)2

灌g(uan)4

干g(an)4

脸l(ian)3

县x(ian)4

鬟h(uan)5

电d(ian)4管g(uan)3

奸j(ian)1

款k(uan)3

添t(ian)1檐(ian)2

坝b(a)4+

挠n(aor)2

懒l(an)3=

镰l(ian)2=气q(i)4+ 羔g(aor)1

幌h(uang)3=甜t(ian)2= 刀d(aor)1

战zh(an)4

羡x(ian)4

怜l(ian)2

残c(an)2盼p(an)4

园(van)2

俭j(ian)3

窜c(uan)4 唤h(uan)5鞍(an)1

禅ch(an)2

干g(an)1

惯g(uan)4

元(van)2

镰l(ian)2汉h(an)4戬j(ian)3伴b(an)4远(van)3

酸s(uan)2 辫b(ian)4 嫌x(ian)2 尖j(ian)1 献x(ian)4蓝l(an)2

椒j(iao)1

桃t(ao)2

斗d(ou)3

桃t(ao)5

抛p(ao)4

藕(ou)3耗h(ao)4=

甲j(ia)3

闸zh(a)2

褡d(a)1

杈ch(a)5

劁q(iao)1

酬ch(ou)2

帽m(ao)4=

嚼j(iao)2

叫j(iao)4=

叨d(ao)5

朝ch(ao)2

漏l(ou)4

袍p(ao)2=

抄ch(ao)1

抱b(ao)4

效x(iao)4

裘q(iou)2

倒d(ao)4

掉d(iao)4

笤t(iao)2

杈ch(a)4

涝l(ao)4

镖b(iao)1

到d(ao)4=

大d(a)4+

毛m(ao)5

抓zh(ua)3

瞎x(ia)1+

臜z(a)1

巴b(a)4

法f(a)2=

嚓c(a)1

帕p(a)4

划h(ua)4

凹(ua)1

喳zh(a)1+

拉l(a)3

叭b(a)1+

嘎g(a)1

着zh(ao)2=

牢l(ao)2+

觉j(iao)4=

饶r(ao)2+

勺sh(ao)2

叫j(iao)4+

悄q(iao)1

凿z(ao)2

流l(iou)4六l(iou)4

头t(ou)1

兜d(ou)5

佑(iou)4

浴(v)4

驮t(uo)2

刘l(iou)2

调d(iao)4

侯h(ou)2

赫h(e)4

走z(ou)4

做z(ou)4

馊s(ou)1

露l(ou)4

轴zh(ou)2

韭j(iou)3

逃t(ao)2

落l(ao)4雕d(iao)1

瓢p(iao)2

滔t(ao)1

挠n(ao)2

州zh(ou)2

槽c(ao)2

跎t(uo)2

涛t(ao)1

乔q(iao)2

掏t(ao)1

Figure 3: Full rhyme network derived from Wang’s data

When applying the Infomap community detection algorithm to iden-tify larger rhyme groups in the data, we find as many as 201 com-munities. In contrast to the connected components in the network,these communities are rather well-balanced. There are 19 communi-ties with 20 or more rhyme words. Together they make up for 2293



nodes, which corresponds to 80% of the rhyme words in the data.The major components of these communities, as reflected by thePīnyīn pronunciations of the rhyme words involved, are provided inTable 6.What we can see from this analysis is that the network has a rather

strong community structure, which can be seen as first evidence thatrather well-definable rhyme groups play an important role in rhyming.However, we can also see that the Infomap community detectionanalysis has separated certain rhymes from each other that we wouldrather group together, such as communities number 13 and 16, 14and 20, as well as 12 and 20. Apart from this, however, the com-munities reflect the rhyme groups as discussed by Wang rather well.That the automated community detection analysis subdivides somegroups that obviously belong together results from the sparsenessof the data and should not surprise us further. It seems reasonableto assume that the communities will become more pronounced withmore data being considered.Since it would require too much time to investigate all of the com-

munities in this sample, I will concentrate on the first and largest oneidentified by the Infomap algorithm. This comprises rhymes centeringaround rhyme words with finals pronounced as -an, according to theannotatin provided in Wang’s data, along with the finals -ian, -uan,and -van (phonetically yan). Wang’s analysis seems to indicate thatthe whole group as such does not reflect free rhyming per se, butrather involves a subtype of free rhyming, which is called semi-freerhyming. Given that our community analysis included all four finalsinto the same group, it seems worthwhile to investigate this groupfurther.As a first step, consider Figure 4, which shows two views of a sub-

graph derived by taking all rhyme words with one f the four finals -an,-ian, -uan, or -van, following the pronunciation provided in Wang’sdata. The first view shows the network in a force-directed layout,while the second view distinguishes the four pronunciation groups.What can be seen from this figure, without applying any further

technique alone, is that the four subgroups seem to rhyme freely witheach other. The big picture does not allow us to assume any form ofstrong intransitivity in rhyming, even if there may be some tendenciesamong the subtypes.This can be further confirmedwhen applying the quantitativemethod

proposed by List et al. (2017), where assortativity (Newman 2003) is


12 List

轩x(van)1

钻z(uan)1

摊t(an)1

建j(ian)4

千q(ian)1

筵(ian)2

蟾ch(an)2

监j(ian)4

联l(ian)2

疃t(uan)3

艳(ian)4

栓sh(uan)1

褊b(ian)3

年n(ian)2

蚕c(an)2

瘫t(an)1

险x(ian)3

窜c(uan)1

喊h(an)3

篮l(an)2

牵q(ian)1

丹d(an)1

悬x(van)2

船ch(uan)2

番f(an)1

剜(uan)1

搀ch(an)1

暄x(van)1

完(uan)1帆f(an)1

完(uan)2

拴sh(uan)1

焉(ian)1

胆d(an)3

煎j(ian)1

连l(ian)2

缆l(an)3

编b(ian)1

签q(ian)1

扇sh(an)1

搬b(an)1

满m(an)3

碱j(ian)3

箭j(ian)4

迁q(ian)1

喘ch(uan)3

颠d(ian)1

龛k(an)1

贩f(an)4

锨x(ian)1

单d(an)1

权q(van)2

棺g(uan)1

毽j(ian)4

担d(an)4+

兰l(an)2+

汉h(an)4+

前q(ian)2+

展zh(an)3+

盘p(an)2

钱q(ian)2

鞍(an)1

潭t(an)2

弯(uan)1

边b(ian)1

穿ch(uan)1

严(ian)2

软r(uan)3瞒m(an)2

淹(ian)1

鲜x(ian)1

产ch(an)3

咽(ian)4

谈t(an)2

衫sh(an)1

淡d(an)4

石d(an)4

鬟h(uan)5

先x(ian)1

暗(an)4

般b(an)1

原(van)2

班b(an)1

扮b(an)4+

添t(ian)1

铣x(ian)1

缘(van)2

嫌x(ian)2

欢h(uan)1

肩j(ian)1

伴b(an)4

寒h(an)2

欠q(ian)5

冠g(uan)1

含h(an)2

毡zh(an)1

餐c(an)1

卷j(van)3

缎d(uan)4

弦x(ian)2

缠ch(an)5

幡f(an)1

眠m(ian)2

片p(ian)4

算s(uan)1

干g(an)4

镰l(ian)2

攒z(an)3

丸(uan)2

面m(ian)4

炼l(ian)4

拌b(an)4

捡j(ian)3=

晚(uan)3=

短d(uan)3=

散s(an)3= 扁b(ian)3=

远(van)3=

烟(ian)1=

间j(ian)1=懒l(an)3=烂l(an)4+

眼(ian)3=线x(ian)4+

赶g(an)3=

天t(ian)1=

拣j(ian)3=

边b(ian)1=

帘l(ian)2= 毯t(an)3 馋ch(an)2= 汉h(an)4= 然r(an)2+烟(ian)1+ 前q(ian)2=

餐c(an)1+镰l(ian)2=

砚(ian)4全q(van)2+面m(ian)4=三s(an)1+甜t(ian)2=眼(ian)3+罐g(uan)4=完(uan)2=

纂z(uan)3

园(van)2+

田t(ian)2+

善sh(an)4+

献x(ian)4+

山sh(an)1+

岩(ian)2

棉m(ian)1

远(van)3

泉q(van)2

看k(an)5

戬j(ian)3

甸d(ian)1

旦d(an)4

罕h(an)3

闪sh(an)3

歉q(ian)4

袁(van)2

淀d(ian)4

三s(an)3 雁(ian)4

反f(an)3

棉m(ian)2

沾zh(an)1

伞s(an)3

担d(an)1+

栾l(uan)2粘zh(an)1

减j(ian)3

镩c(uan)1

残c(an)2

演(ian)3

宣x(van)1

环h(uan)2川ch(uan)2

间j(ian)1+

短d(uan)3

天t(ian)1+

绵m(ian)2

盘p(an)2= 翻f(an)1+懒l(an)3 馒m(an)2=

便b(ian)4+

钐sh(an)4板b(an)3

片p(ian)1

甘g(an)1骟sh(an)4

捻n(ian)3=

乱l(uan)4=

断d(uan)4=

线x(ian)4=

点d(ian)3=

蹁p(ian)2=善sh(an)4

咸x(ian)2

拳q(van)2 叹t(an)4

填t(ian)2

蓝l(an)2

甜t(ian)2

遍b(ian)4

鞭b(ian)1 员(van)2

蒜s(uan)4

篇p(ian)1研(ian)2

展zh(an)3

旱h(an)4

掮q(ian)2

案(an)4

圈j(van)4

团t(uan)2

坦t(an)5

唤h(uan)5

年n(ian)2+

散s(an)3

绵m(ian)3

凡f(an)2

延(ian)2

转zh(uan)4=饭f(an)4=

烂l(an)4=

灌g(uan)4

干g(an)4=

馆g(uan)4

探t(an)4=

男n(an)2

罐g(uan)4

担d(an)4

扦q(ian)1

难n(an)4=

燕(ian)4

汉h(an)4

瓣b(an)4

剪j(ian)3

半b(an)4+绊b(an)4

半b(an)4

钻z(uan)4元(van)2

店d(ian)4

敢g(an)3

站zh(an)4

扮b(an)4

蛋d(an)4

练l(ian)4

蛮m(an)2

田t(ian)2

算s(uan)4

散s(an)4

悬x(van)4

看k(an)4馅x(ian)4

赶g(an)3

剑j(ian)4

骗p(ian)4

献x(ian)4线x(ian)4

端d(uan)1尖j(ian)1

殿d(ian)4

晚(uan)3

岸(an)4

念n(ian)4

还h(uan)2

辫b(ian)4

奸j(ian)1冤(van)1 院(van)4

园(van)2

滩t(an)1

砖zh(uan)1见j(ian)4

参c(an)1仙x(ian)1

禅ch(an)2串ch(uan)4

羡x(ian)4 宽k(uan)1

观g(uan)1

言(ian)2烟(ian)1 兰l(an)2

玩(uan)2

全q(van)2

官g(uan)1

銮l(uan)2

山sh(an)1

显x(ian)3

南n(an)2

脸l(ian)3

蝉ch(an)2

关g(uan)1断d(uan)4

川ch(uan)1

眼(ian)3选x(van)3

汗h(an)4

前q(ian)2

惯g(uan)4

管g(uan)3

怜l(ian)2

颤ch(an)4

三s(an)1

然r(an)2

传ch(uan)2

涟l(ian)2

战zh(an)4+

馋ch(an)2

浅q(ian)3

转zh(uan)3

穿ch(uan)1+

现x(ian)4

燃r(an)2电d(ian)4

穿ch(uan)2

占zh(an)4

辨b(ian)4

癣x(van)3

拦l(an)2

贤x(ian)2

款k(uan)3

颜(ian)2

变b(ian)4

酸s(uan)1

扇sh(an)4

勉m(ian)3

酸s(uan)2

转zh(uan)4

怨(van)4

蹿c(uan)4

天t(ian)1俭j(ian)3

圆(van)2

干g(an)1

杆g(an)3

闲x(ian)2

翻f(an)1

帘l(ian)2

安(an)1

杆g(an)1

暖n(uan)3

点d(ian)3

烂l(an)4

舔t(ian)3

檐(ian)2

唤h(uan)4

圈q(van)1敢g(an)3+痰t(an)2

捐j(van)1

弹d(an)4

罕h(an)5

边b(ian)1+

掀x(ian)1

簪z(an)1

肝g(an)1乱l(uan)4

匾b(ian)3

弹t(an)5

贱j(ian)4

援(van)2

县x(ian)4

咱z(an)2

竿g(an)1

难n(an)4

柑g(an)1

缠ch(an)2

难n(an)2

烦f(an)2

天t(ian)2

炭t(an)4

办b(an)4

间j(ian)1

盐(ian)2段d(uan)4

甸d(ian)4

玄x(van)2

窜c(uan)4

厌(ian)4咱z(an)2+

湾(uan)1

莲l(ian)2

源(van)2

坛t(an)2

范f(an)4

战zh(an)4

娟j(van)1腕(uan)4

干g(an)4+

验(ian)4

赚zh(uan)4

看k(an)4+

饭f(an)4+

换h(uan)4

盼p(an)4

万(uan)4

碗(uan)3

饭f(an)4

馆g(uan)3

按(an)4

惨c(an)3

撵n(ian)3

灿c(an)4

劝q(van)4

弹t(an)2

沿(ian)2

砍k(an)3

担d(an)1观g(uan)4

堪k(an)1

慢m(an)4

恋l(ian)4

-an -van

-ian

-uan

拴sh(uan)1

罐g(uan)4=栓sh(uan)1

钻z(uan)4断d(uan)4=

转zh(uan)3

观g(uan)4

蹿c(uan)4

转zh(uan)4=

短d(uan)3=

穿ch(uan)2

款k(uan)3

湾(uan)1

镩c(uan)1

惯g(uan)4

疃t(uan)3

唤h(uan)5

完(uan)2=

腕(uan)4

纂z(uan)3唤h(uan)4窜c(uan)4缎d(uan)4 鬟h(uan)5丸(uan)2

宽k(uan)1

冠g(uan)1

短d(uan)3

断d(uan)4

棺g(uan)1

晚(uan)3

栾l(uan)2

穿ch(uan)1+

算s(uan)1

劝q(van)4

选x(van)3

冤(van)1

袁(van)2

员(van)2

元(van)2

园(van)2

原(van)2

援(van)2

泉q(van)2

园(van)2+

娟j(van)1

圆(van)2

拳q(van)2

玄x(van)2怨(van)4

权q(van)2

暄x(van)1卷j(van)3远(van)3

源(van)2

缘(van)2悬x(van)2

癣x(van)3

悬x(van)4

院(van)4

捐j(van)1

宣x(van)1

远(van)3=

全q(van)2+

全q(van)2

圈q(van)1

圈j(van)4

轩x(van)1

盘p(an)2=

砍k(an)3饭f(an)4=

干g(an)4=

按(an)4

散s(an)3=蝉ch(an)2

参c(an)1

案(an)4

禅ch(an)2懒l(an)3=

难n(an)4=

探t(an)4=

甘g(an)1

凡f(an)2

兰l(an)2

缆l(an)3

贩f(an)4

扮b(an)4+

坦t(an)5蛮m(an)2赶g(an)3=散s(an)3痰t(an)2 扮b(an)4柑g(an)1

烂l(an)4=

灿c(an)4

半b(an)4+

馋ch(an)2=

懒l(an)3

毯t(an)3

烂l(an)4+

三s(an)3

帆f(an)1

三s(an)1+

看k(an)5

罕h(an)3

肝g(an)1

旦d(an)4

餐c(an)1+

馒m(an)2=

寒h(an)2

慢m(an)4

满m(an)3竿g(an)1汉h(an)4=

瘫t(an)1簪z(an)1伞s(an)3攒z(an)3担d(an)4弹t(an)2

残c(an)2堪k(an)1

潭t(an)2

半b(an)4汉h(an)4

难n(an)2范f(an)4

丹d(an)1

盼p(an)4

淡d(an)4

蚕c(an)2

闪sh(an)3

办b(an)4

烂l(an)4翻f(an)1

男n(an)2炭t(an)4

喊h(an)3石d(an)4

产ch(an)3

拦l(an)2战zh(an)4缠ch(an)5衫sh(an)1

胆d(an)3

看k(an)4+

翻f(an)1+颤ch(an)4

伴b(an)4

饭f(an)4+

扇sh(an)4罕h(an)5占zh(an)4惨c(an)3杆g(an)1散s(an)4南n(an)2弹d(an)4般b(an)1

缠ch(an)2

汗h(an)4

蛋d(an)4

拌b(an)4

看k(an)4

难n(an)4

搬b(an)1

三s(an)1

善sh(an)4

毡zh(an)1

站zh(an)4

暗(an)4

单d(an)1

反f(an)3

盘p(an)2

担d(an)4+

兰l(an)2+

干g(an)1

展zh(an)3+

汉h(an)4+

战zh(an)4+

干g(an)4+

坛t(an)2

山sh(an)1+

善sh(an)4+

然r(an)2

山sh(an)1

咱z(an)2

饭f(an)4

干g(an)4

燃r(an)2

然r(an)2+

谈t(an)2

展zh(an)3

安(an)1

叹t(an)4

敢g(an)3

杆g(an)3

钐sh(an)4

滩t(an)1

赶g(an)3

片p(ian)1甜t(ian)2颠d(ian)1

变b(ian)4怜l(ian)2

店d(ian)4鲜x(ian)1严(ian)2烟(ian)1 前q(ian)2莲l(ian)2

迁q(ian)1

献x(ian)4

建j(ian)4

演(ian)3

险x(ian)3

铣x(ian)1

奸j(ian)1

厌(ian)4

甸d(ian)1

骗p(ian)4点d(ian)3

田t(ian)2

点d(ian)3=

鞭b(ian)1

千q(ian)1

现x(ian)4

念n(ian)4

船ch(uan)2

环h(uan)2

串ch(uan)4

乱l(uan)4=

蒜s(uan)4

砖zh(uan)1

段d(uan)4

完(uan)2

酸s(uan)2

转zh(uan)4

万(uan)4

川ch(uan)2

灌g(uan)4

晚(uan)3=

玩(uan)2

官g(uan)1

完(uan)1

窜c(uan)1

乱l(uan)4

馆g(uan)3

川ch(uan)1

管g(uan)3

端d(uan)1暖n(uan)3

酸s(uan)1赚zh(uan)4

穿ch(uan)1

观g(uan)1銮l(uan)2钻z(uan)1

团t(uan)2

剜(uan)1

关g(uan)1

换h(uan)4

罐g(uan)4

欢h(uan)1

算s(uan)4

弯(uan)1

馆g(uan)4

碗(uan)3

还h(uan)2

传ch(uan)2

喘ch(uan)3

软r(uan)3

编b(ian)1

天t(ian)1

年n(ian)2

钱q(ian)2

篇p(ian)1

连l(ian)2

眼(ian)3

殿d(ian)4

边b(ian)1

面m(ian)4

闲x(ian)2

间j(ian)1

先x(ian)1

绵m(ian)2

前q(ian)2=

田t(ian)2+

前q(ian)2+

煎j(ian)1

片p(ian)4

见j(ian)4

便b(ian)4+

县x(ian)4

言(ian)2脸l(ian)3

辨b(ian)4帘l(ian)2

线x(ian)4辫b(ian)4尖j(ian)1

烟(ian)1+

献x(ian)4+

年n(ian)2+

间j(ian)1+

天t(ian)1+

肩j(ian)1

涟l(ian)2

镰l(ian)2

镰l(ian)2=

掮q(ian)2

掀x(ian)1

岩(ian)2

甜t(ian)2=

褊b(ian)3

砚(ian)4

俭j(ian)3

扁b(ian)3=帘l(ian)2=

淹(ian)1焉(ian)1

羡x(ian)4

电d(ian)4

天t(ian)1=间j(ian)1=烟(ian)1=

绵m(ian)3

淀d(ian)4

檐(ian)2

眼(ian)3=

显x(ian)3

燕(ian)4

匾b(ian)3

监j(ian)4

线x(ian)4+

拣j(ian)3=

剪j(ian)3

验(ian)4

戬j(ian)3

签q(ian)1

甸d(ian)4

馅x(ian)4

毽j(ian)4

嫌x(ian)2

浅q(ian)3

筵(ian)2

联l(ian)2

恋l(ian)4

锨x(ian)1

舔t(ian)3

研(ian)2

贤x(ian)2

歉q(ian)4

碱j(ian)3

眠m(ian)2

扦q(ian)1

箭j(ian)4

剑j(ian)4

仙x(ian)1

延(ian)2

咽(ian)4

天t(ian)2

练l(ian)4

撵n(ian)3沿(ian)2填t(ian)2弦x(ian)2

雁(ian)4

面m(ian)4= 蹁p(ian)2=炼l(ian)4捻n(ian)3=线x(ian)4=捡j(ian)3= 边b(ian)1=勉m(ian)3

牵q(ian)1

棉m(ian)2

颜(ian)2

眼(ian)3+

棉m(ian)1咸x(ian)2

贱j(ian)4欠q(ian)5

艳(ian)4

添t(ian)1

减j(ian)3

遍b(ian)4

盐(ian)2

边b(ian)1+

弹t(an)5

沾zh(an)1

餐c(an)1

担d(an)1+

骟sh(an)4

担d(an)1

班b(an)1

敢g(an)3+

搀ch(an)1

烦f(an)2

咱z(an)2+

岸(an)4

瞒m(an)2

蟾ch(an)2

旱h(an)4蓝l(an)2

馋ch(an)2

龛k(an)1

扇sh(an)1

粘zh(an)1

瓣b(an)4

绊b(an)4

摊t(an)1

板b(an)3

含h(an)2

鞍(an)1

篮l(an)2

幡f(an)1

番f(an)1

Figure 4: Two views of the subgraph of all rhyme words with finals in -(iuv)an.

used to measure to which degree rhyming is influenced by vowel pu-rity in Old Chinese rhyme data. Assortativity tests whether a certainpartition of the nodes in a network into groups is also reflected by thenetwork structure of the network itself. While assortativity was usedto test if rhyming favors identical vowels or not, we can use the sameidea to test if distinctions in the medials in rhyme words with final -anfind a reflection in the graph structure.In order to do this, we first compute the assortativity coefficient of

the subgraph shown in Figure X above, with respect to the four sub-groups -an, -ian, -uan, and -van. To check if the result is significant,we then carry out 1000 random trials in which we randomly assignthe nodes in the network to one of the four groups, and re-computethe assortativity coefficient for each of these trials. We then calculatethe standard deviation of the 1000 trials and compute, by how manystandard deviations the attested assortativity differs from the one weachieve in our random trials.The results of this analysis (which is included as part of the sup-

plementary material) yield an assortativity coefficient of 0.01 for thesubgraph, and a mean assortativity coefficient of 0.00 for the ran-dom trials, with a sigma score of 2.21. This means on the one handthat there is a significant difference between the randomly assignedgroupings and the groupings given in the data themselves. On theother hand, however, it also shows that the amount of assortativemixing in the data is extremely low. An assortativity coefficient of 0indicates that there is no evidence for assortative mixing in the data,or – to apply this to our rhyme network – that there is no evidence



for different rhyme groups in a given set of nodes. A coefficient of0.01, as we find it in this experiment, indicates a very small tendencyfor words from the same subgroup to rhyme with each other. Butthis tendency is so small that it can almost be neglected. To addsome context to the obtained score of 0.01: in List et al. (2017), allreconstruction systems that were tested for vowel purity showed as-sortativity coefficients of 0.5 and higher.In the light of data-driven network approaches, there is no evidence

for semi-free rhyming in the group of -an finals, contrary to what waspostulated by Wang. Given the continuous character of rhyming, asadmitted by Wang as well, it may not even be needed to postulatesuch a group. Instead, it seems more useful to report results basedon the data themselves, and in this context, measures like assortativ-ity have the advantage that they are not binary, but can take rhymegroups involving multiple subgroups into account.

4 Conclusion

In this small review of Wang’s quantitative study on rhyming in Bei-jing Chinese, I have tried to show how data handling and analysisin the study of rhyme patterns can be improved. With respect todata handling, I have concentrated on recently proposed annotationframeworks which have the advantage of reflecting rhyme analysesin a replicable way that can be easily expanded, criticized, and modi-fied. With respect to the analyses, I have tried to illustrate the impor-tance of network approaches when trying to identify consistent rhymegroups in a given dataset. By showing how tests for assortative mix-ing can be used to assess whether a rhyme group should be furthersubdivided into smaller groups, I have tried to propose an alterna-tive for the binary test procedure used in Wang’s study. While this isnot intended to reject Wang’s analysis, which is valuable in its ownright, I think it is desirable for future analyses to include network tech-niques as well, since these techniques offer an important additionalperspective on rhyme data.


14 List

Supplementary Material

The supplementary material accompanying this study contains thesource code and the data required to replicate the studies reportedhere, along with a description explaining how to apply the code. Ithas been uploaded to the Open Science Framework, where it can bedownloaded from https://doi.org/10.17605/OSF.IO/C9VDP.

Acknowledgments

This researchwas funded by the ERCStartingGrant 715618 “Computer-Assisted Language Comparison” (http://calc.digling.org). Theauthor thanks one anonymous reviewer for help with the manuscript,and expresses his gratitude to Wang Zhihao for sharing the data un-derlying the study transparently.

References

Baxter, William H. 1992. A handbook of Old Chinese phonology . Berlin: deGruyter.

Bu, Aisong. 2019. An optimality-theoretical analysis of Chinese Great VowelShift . Beihang University. (Master thesis).

Jiǎ, Zhī�� (ed.). 2009. Zhōngguó gēyáo jíchéng, Běijīng juàn�� [China Folk Rhymes Collection, Vol. Beijing] . Beijing: ChinaISBN Center.

List, Johann-Mattis. 2014. Sequence comparison in historical linguistics.Düsseldorf: Düsseldorf University Press.

List, Johann-Mattis. 2016. Using network models to analyze Old Chineserhyme data. Bulletin of Chinese Linguistics 9(2). 218–241.

List, Johann-Mattis. 2017. A web-based interactive tool for creating, inspect-ing, editing, and publishing etymological datasets. In Proceedings of the15th Conference of the EuropeanChapter of the Association for ComputationalLinguistics. SystemDemonstrations, 9–12. Valencia: Association for Com-putational Linguistics.

List, Johann-Mattis. 2019. PoePy. A Python library for handling annotatedrhymes. Jena: Max Planck Institute for the Science of Human History.https://github.com/lingpy/poepy/releases.

List, Johann-Mattis &Hill, NathanW. & Foster, Christopher J. 2019. Towardsa standardized annotation of rhyme judgments in Chinese historical phonol-ogy (and beyond). Journal of Language Relationship 17(1). 26–43.



List, Johann-Mattis & Pathmanathan, Jananan Sylvestre & Hill, NathanW. &Bapteste, Eric & Lopez, Philippe. 2017. Vowel purity and rhyme evidencein Old Chinese reconstruction. Lingua Sinica 3(1). 1–17.

Newman, M. E. J. 2003. Mixing patterns in networks. Physical Review E67. 026126.

Newman, M. E. J. 2004. Analysis of weighted networks. Physical Review E70(5). 056131.

Rosvall, M. & Bergstrom, C. T. 2008. Maps of random walks on complex net-works reveal community structure. Proc. Natl. Acad. Sci. U.S.A. 105(4).1118–1123.

Wáng, Lì��. 1980. Shījīng Yùndú�� [Rhyme readings in the Bookof Odes] . Shànghǎi��: Shànghǎi Gǔjī��.

Wilkinson, Mark D. & Dumontier, Michel & Aalbersberg, Ilsbrand J. & Apple-ton, Gabrielle & Axton, Myles & Baak, Arie & Blomberg, Niklas & Boiten,Jan-Willem & da Silva Santos, Luiz B. & Bourne, Philip E., et al. 2016. TheFAIR Guiding Principles for scientific data management and stewardship.Scientific Data 3.


Documents

Improving data handling and analysis in the study of rhyme