世界語言中分類詞、性別詞與複數標記的分與合： GIS的類型學研究 - 政大學術集成

全文

(1)國立政治大學語言學研究所碩士學位論文 National Chengchi University Graduate Institute of Linguistics Master Thesis. 指導教授: 何萬順博士 Advisor: Dr. One-Soon Her. 立. 政治大. ‧ 國. 學. 世界語言中分類詞、性別詞與複數標記的分與合：. ‧. GIS 的類型學研究. y. Nat. sit. n. al. er. io. A GIS Typological Analysis of the Convergence and Divergence among Numeral Classifiers, Genders and Plural Markers in the World’s Languages. Ch. engchi. i n U. 研究生：唐威洋撰 Student: Marc Tang. 中華民國一○四年七月 July, 2015. v.

(2) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(3) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. Copyright © 2015 Marc Tang All Rights Reserved. i n U. v.

(4) 摘要本論文的主要目的在對於分類詞、性別詞以及複數標記在語言當中的地域分佈提出解釋．在前人的研究當中，這三項元素被認為是名詞句中平衡資訊的重要工具 (Greenberg, 1990; Aikhenvald, 2000)：分類詞語言主要位於東南亞和南美洲部分地區，而具有性別詞或複數標記的語言大多出現在歐洲、非洲和美洲部分地區．我們提出的論證如下：即便這三樣元素外表上具有歧異，它們會呈現當今所見的地域分佈原因在於它們共有的兩項標記功能：可數性質及語意分類．分類詞同時滿足兩者而性別詞及複數標記分別滿足其一；依照此邏輯，我們預測有分類詞的語言不會同時具有性別詞及複數標記而反之亦然．本文中我們透過句法形式和語意功能的比較提出論證並透過類型學、地理及歷史的角度分析來自世界上最大的二十個語系（印歐，漢藏，亞非，尼日爾-剛果，南島，達羅毗荼，阿爾泰，南亞，壯侗，尼羅－撒哈拉，烏拉，高加索，等語系）的 155 個語言．架構上，第一章簡單對研究題目進行介紹，第二章呈現前人研究的匯整，第三章包含我們的理論論證以及我們對於分類詞、性別詞及複數標記分與合的解釋．隨後的第四章中，我們提出類型學和地理資訊系統(GIS)的證據；最後在第五張和第六章我們分別點出本研究的限制以及結論．. 立. 政治大. ‧. ‧ 國. 學. n. al. er. io. sit. y. Nat. 關鍵詞：分類詞、性別詞、複數標記、類型學、地理資訊系統. Ch. engchi. i n U. v.

(5) Abstract This thesis aims at providing an explanation for the typological and areal distribution between numeral classifiers, genders (noun classes) and grammatical plural markers. Within previous studies, these three components are considered as different devices to balance information in noun phrases (Greenberg, 1990; Aikhenvald, 2000). Numeral classifier languages are mainly present in South-East Asia and parts of South-America, while languages with genders and grammatical plural markers are generally attested in Europe, Africa and parts of the Americas. We propose that despite their apparent divergence, the three elements display this particular geographical distribution due to their convergent features of count/mass distinction and semantic classification: Numeral classifiers carry both functions, while genders and plural markers separately fulfill one of them. Following this logic, we expect that a language with numeral classifier do not have simultaneously the systems of genders plus plural markers and vice-versa. Theoretical evidence via formal syntactic form and semantic function comparison is proposed and further supported by typological, geographical and historical analysis of 155 languages that are mainly part of the 20 biggest language groups in the world, e.g. Indo-European, Sino-Tibetan, Afro-Asiatic, Niger-Congo, Austronesian, Dravidian, Japonic, Altaic, Austro-Asiatic, Tai-Kadai, Creole, Nilo-Saharan, Uralic, Quechuan, Hmong-Mien, Mayan, North Caucasian, Language isolates among others. Chapter 1 presents a brief introduction of the subject while chapter 2 displays the literature review. Chapter 3 includes our theoretical discussion. 立. 政治大. ‧. ‧ 國. 學. sit. y. Nat. n. al. er. io. proposing explaining the convergence and divergence among numeral classifiers, genders and plural markers, followed by typological and geographical evidence via GIS (Geographic Information System) in Chapter 4. Finally Chapter 5 and 6 contain the limitations of our study and its conclusion.. Ch. engchi. i n U. v. Key words: Numeral classifier, Gender, Plural marker, Typology, Geographic information system.

(6) Table of contents List of maps..................................................................................................................... i List of tables ...................................................................................................................ii List of figures ............................................................................................................... iii List of abbreviations ..................................................................................................... iv 1. Introduction ................................................................................................................ 1 2. Literature review ........................................................................................................ 4 2.1 Numeral classifiers..................................................................................... 4 2.2 2.3 2.4. Genders ...................................................................................................... 8 Plural markers .......................................................................................... 14 Summary .................................................................................................. 16. 政治大 3. Theoretical discussions ............................................................................................ 17 立 3.1 Numeral classifiers and plural markers.................................................... 17 ‧. ‧ 國. 學. 3.2 Numeral classifiers and genders .............................................................. 18 3.3 Genders and plural markers ..................................................................... 22 3.4 Summary: Triangle relationship .............................................................. 22 4. Supporting evidence................................................................................................. 27 4.1 Methodology ............................................................................................ 27 4.2 Typological evidence ............................................................................... 30. sit. y. Nat. n. al. er. io. 4.2.1 Optimal marking .......................................................................... 32 4.2.2 Under-marking ............................................................................. 35 4.2.3 Over-marking ............................................................................... 42 4.2.4 Summary ...................................................................................... 47 4.3 GIS evidence ............................................................................................ 48 4.3.1 Numeral classifiers and plural markers........................................ 50 4.3.2 Numeral classifiers and genders .................................................. 51 4.3.3 Genders and plural markers ......................................................... 52 4.3.4 Summary ...................................................................................... 54 5. Limitations ............................................................................................................... 55. Ch. engchi. i n U. v. 6. Conclusion ............................................................................................................... 56 REFERENCES ................................................................................................................. 57 Appendix 1: List of languages and related references ................................................. 63.

(7) List of maps Page 2 2 2. 7. Content Numeral classifiers in the world languages (Aikhenvald, 2000:122) Genders in the world languages (Aikhenvald, 2000:78) Numeral classifiers and genders, adapted from Aikhenvald (2000:78,122) Numeral classifiers in the world languages (Aikhenvald, 2000:122) Genders in the world languages (Aikhenvald, 2000:78) Plural marking in the world, adapted from Haspelmath (2013) and Dryer (2013) Display of 155 surveyed languages (by dots). 8 9 10. Display of 155 surveyed languages (by speaker population) Categories distribution of 155 languages in database (by dots) Categories distribution of 155 languages in database (by speaker. 29 48 49. y. Distribution of genders and plural markers, adapted from Aikhenvald (2000:78), Haspelmath (2013) and Dryer (2013) Distribution of genders and plural markers in 155 languages. io. 15. Nat. 14. ‧. 13. 學. 12. 立. coverage) Distribution of numeral classifiers and plural markers, adapted from Aikhenvald (2000:122), Haspelmath (2013) and Dryer (2013) Distribution of numeral classifiers and plural markers in 155 languages Distribution of numeral classifiers and genders, adapted from Aikhenvald (2000:78,122) Distribution of numeral classifiers and genders in 155 languages. sit. 11. 政治大. n. al. er. 4 5 6. ‧ 國. Map 1 2 3. 16. Ch. engchi. i. i n U. v. 4 8 14 28. 50 51 51 52 53 53.

(8) List of tables. 立. 政治大. 學 ‧. io. sit. y. Nat. n. al. er. 7. Content Proto-Bantu noun class meanings (Richardson, 1967; Welmers, 1973) Main differences between genders and classifiers (Dixon, 1986) Usage of general classifier and specific classifier (Chinese) Genders and plurals co-occurrence in French Semantic functions of numeral classifiers, genders and plural markers Different combinations of numeral classifiers, genders and plural markers Distribution of nominal case marking among category 1, 2 and 3. ‧ 國. Table 1 2 3 4 5 6. Ch. engchi. ii. i n U. v. Page 9 20 21 22 23 31 37.

(9) List of figures Figure Content 1 Interaction of numeral classifiers, genders and plural markers 2 Preliminary database of 80 languages. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. iii. i n U. v. Page 23 27.

(10) List of abbreviations ABS (C) CL DEF EX (F) GL IMP IN LOC. absolutive common gender classifier definite exclusive feminine gender goal imperative inclusive locative. M (M) (N). measure word masculine gender neutral gender. PFV PL PRES REC SG 1SG 2SG 3SG. perfective plural present recipient singular first person singular second person singular third person singular. ‧. ‧ 國. 學. n. al. Ch. sit er. io. first person plural second person plural third person plural. y. Nat. 1P 2P 3P. 立. 政治大. engchi. iv. i n U. v.

(11) Chapter 1 Introduction The main purpose of this thesis is to explain the typological and areal distribution between numeral classifiers, grammatical genders (also known as noun class) and grammatical plural markers. Within previous studies, these three components have been considered as devices to balance information in noun phrases (Greenberg, 1990; Aikhenvald, 2000; Kemps & Regier, 2012). However, their nature diverges: First, numeral classifiers carry semanticity related to the noun and appear when the noun is accompanied by a numeral (Aikhenvald, 2007). They are generally present in East and South-East Asian languages (Gil, 2013), an example from Chinese being: yi tiao yu ‘one CL-long shape fish’, the numeral classifier highlights the long shape of the main noun: fish. Second, genders are viewed as pure syntactic agreement elements. They divide all (or nearly all) nouns into rigid classes and are realized on other constituents in the form of agreement (Dixon, 1986; Corbett, 1991; Grinevald, 2000).. 立. 政治大. ‧. ‧ 國. 學. This system is very common in Indo-European and African languages (Corbett, 2013), e.g. the masculine and feminine distinction in French: livre ‘book, masculine’ and table ‘table, feminine’. Third, plural markers are commonly defined as syntactic marking of plurality (mass/count feature) for nouns, also realized by agreement (Corbett, 2000). Their presence is noted within most languages of the world, except for South-East Asia and Australia where they may be optional (Haspelmath, 2013). A. sit. y. Nat. n. al. er. io. typical example would be the -s in English, e.g. ‘one book’ and ‘two books’.. Ch. i n U. v. Our framework is as follows: based on the literature review, we provide a clear definition for the three elements in discussion: numeral classifiers, genders and plural markers. They were compared by pair in the past but no previous research ever combined them together to explain their typological and geographical distribution and this is the main contribution of this study. As an example, Greenberg (1990) hypothesized a long time ago that numeral classifiers and plural markers tend not to co-occur. Then, Borer (2005) provided theoretical syntactic evidence and said even more directly that they are the same thing. But until now no precise statistics and. engchi. cross-language typological evidence were provided. Moreover, the geographical distribution of the three elements have been studied separately (Aikhenvald,2000; Gil, 2013; Corbett, 2013; Haspelmath, 2013) but their results have not been combined to analyze their convergence and divergence in languages of the worlds. As an example in Map 1 and 2, Aikhenvald (2000) proposed two separate maps for the geographical distribution of numeral classifiers and genders. 1.

(12) Map 1. Numeral classifiers in the world languages (Aikhenvald, 2000:122). Map 2. Genders in the world languages (Aikhenvald, 2000:78). 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. i n U. v. We innovatively combined them, showing that their language area does not overlap: classifiers are mainly in Asia while genders stand in Europe and Africa.. engchi. Map 3. Numeral classifiers and genders, adapted from Aikhenvald (2000:78,122). 2.

(13) As a summary, the relationship between numeral classifiers, genders and plural markers have been discussed by pairs in previous studies, but never combined all three together to explain their typological and geographical distribution. This is precisely the gap we plan to fill in our study. Following this chapter (Chapter 1) which presents a brief introduction of the subject, Chapter 2 displays the literature review. Chapter 3 includes our theoretical discussion explaining the convergence and divergence among numeral classifiers, genders and plural markers, followed by typological and geographical evidence via GIS (Geographic Information System) in Chapter 4. Finally Chapter 5 and 6 explain the limitations of our study and offers a conclusion.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 3. i n U. v.

(14) Chapter 2 Literature review In the linguistic field, researchers are interested in systems of classifiers since they can reflect how speakers view the world (Haas, 1942; Burling, 1965; Craig, 1986), taking as an example the analysis of Lakoff (1987) of the language Dyirbal where women, fire and dangerous things are categorized under the same classifier. In this section we will briefly introduce the related literature and clarify the definition concerning numeral classifiers, genders and plural markers.. 2.1 Numeral classifiers. 政治大. Geographically speaking, the main concentration of numeral classifiers can be found at a single zone in East and South-east Asia, but reaching out both westwards and. 立. ‧ 國. 學. eastwards through the Indonesian archipelago and then into the pacific, coming to an end in some western part of the Americas (Gil, 2013), as displayed in Map 4.. ‧. Map 4. Numeral classifiers in the world languages (Aikhenvald, 2000:122). n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Numeral classifiers mainly carry two semantic functions: noun classification and countability marking. First, regarding semantic classification, as stated by Tai & Wang (1990) and Her & Hsieh (2010): A numeral classifier categorizes a class of nouns by picking out some salient perceptual properties, which are permanently associated with entities named by the class of nouns. Taking as an example in (1a) the 4.

(15) numeral classifier highlights the long shape of the noun yu ‘fish’. While in (1b) with the phrase yi CL-zhi gou ‘one dog’, the classifier zhi carries the feature of animacy, which is a feature shared by the following noun gou ‘dog’, thus making the combination of the two words possible. On the other hand, phrases such as *一隻書 yi CL-zhi shu ‘one book (animate)’ would be semantically ill-formed. (1) Sample of numeral classifiers (Chinese) a. Highlights the long shape of the fish 一條魚. b. Highlights the animacy of the dog 一隻狗. yi 1. yi 1. tiao CL-long shape. yu fish. ‘one fish’. zhi CL-animacy. gou dog. ‘one dog’. 政治大. Major agreement is reached on the fact that numeral classifiers are required in various languages of the world (Greenberg, 1990; Aikhenvald, 2000), especially in Asia and. 立. ‧. ‧ 國. 學. the Americas. The form may vary in terms of word formation and word ordering, but it does not alter the status of the numeral classifiers. As an example in Baniwa, an Arawakan language, the classifier counts as a morpheme directly combined to the numeral, e.g. apa-api mawipi ‘one-CL:hollow blow.gun+CL:long.thin) ‘one blowgun’ (Aikhenvald, 2007:480). While in Chinese the classifier is an individual word located between the numeral and the noun. A general sample of ordering is displayed in (2), where we can realize that numeral classifiers concatenate with a quantifier, locative,. sit. y. Nat. n. al. er. io. demonstrative or predicate to form a nexus that cannot be interrupted by the noun which it classifies (Allan, 1977), as underlined between the numeral and the classifier. Few apparent exceptions are proposed in the literature (Watters, 1981; Adams, 1989), but since they are still subjects of discussion in formal syntactic terms (Kihm, 2005; Her, 2012), we will not talk about them in our study.. Ch. engchi. i n U. v. (2) Different types of word order with numeral classifiers (Her, 2015) Ordering Language sample Numeral-Classifier-Noun Chinese, Vietnamese, Bengali Noun-Numeral-Classifier Thai (Tai-Kadai group) Classifier-Numeral-Noun Noun-Classifier-Numeral Classifier-Noun-Numeral Numeral-Noun-Classifier. Garo (Tibeto-Burman group) Jingpho, Bodo (no language) (no language). 5.

(16) Following the idea of connection with numerals, we come to the second main semantic function of numeral classifiers: countability marking. It is commonly accepted that one of the primary purposes of numeral classifiers is to facilitate quantification of the noun and mark its mass/count feature (Thompson, 1965; Quine, 1969; Stein, 1981; Link, 1991). As demonstrated in (3). (3) Sample of numeral classifiers and measure words (Chinese) a. Numeral classifier with count noun b. Numeral classifier with mass noun 三本書 *三個水 san three. ben CL-volume. shu book. san ge three CL-general. ‘three books’ c. Measure word with count noun 三箱書. 立. ‘*three water’ d. Measure word with mass noun 三瓶水. 政治大 san ping. shui three M-bottle water ‘three bottles of water’. ‧ 國. 學. san xiang shu three M-box book ‘three boxes of books’. shui water. ‧. Taking an example from Chinese in (3a) with the noun shu ‘book’: when enumerating. sit. y. Nat. the quantity a classifier is required to denote that the following noun is countable, as in the phrase san ben shu ‘three CL-volume book’. On the other hand, if we observe the mass noun shui ‘water’, the phrase *san ge shui ‘three CL-general water’ in (3b). n. al. er. io. is semantically ill-formed, since water is not countable and cannot apply numeral classifiers. The alternative for mass nouns is a measure word (Tai & Wang, 1990) which denotes the quantity of the noun, as demonstrated by the phrase san ping shui ‘three M-bottle water’ in (3d). Measure words also apply to countable nouns to describe quantity in specific measures, as in (3c). This differentiation has different terms attested in the literature, e.g. sortal and mensural classifiers, nevertheless the definition is still the same therefore we do not list out all the possible naming here.. Ch. engchi. i n U. v. As a summary for numeral classifiers and measure words: numeral classifiers classify nouns by highlighting one of their essential features, meaning that they do not bring new information but do facilitate referent tracking in discourse. On the other hand, measure words carry new information of quantity but only highlight an accidental feature of the following noun (Her and Hsieh, 2010), e.g. in (3c-d) the fact that books may be put into boxes and water into bottles is not an essential or necessary specificity contained in their subject concept, while the ‘volume’ feature of the books highlighted by numeral classifier in (3a) is. Another example would the that the noun 6.

(17) ‘bachelor’ already encodes the feature of being ‘unmarried’ but does not mention whether the person is happy or not, ‘unmarried’ being an essential feature of a bachelor while ‘happy’ is an accidental feature. In these terms, the numeral classifiers are semantically redundant while measure words are not. An example is demonstrated in (4a), where the omission of numeral classifiers due to specific written register (e.g. in the bible) is acceptable and does not create change in the meaning of the phrase itself. However, in (4b) if the measure words are omitted the meaning is entirely different, e.g. changing from ‘5 boxes of loaves’ to ‘5 loaves’. (4) Semantic redundancy of numeral classifiers (Her and Hsieh, 2010:544) a. Numeral classifiers omission does not change the meaning 五個餅二條魚五餅二 = wu five. ge CL. bing loaf. er two. 立. tiao CL. yu fish. wu. bing. er. 魚 yu. 政治大five loaf two fish ‘five loaves and two fish’. ‧ 國. 學. ‘five loaves and two fish’ b. Measure words bring new information of quantity and cannot be omitted 五箱餅二箱魚 ≠ 五餅二. wu bing er yu five loaf two fish ‘five loaves and two fish’. ‧. wu xiang bing er xiang yu five box loaf two box fish ‘five boxes of loaves and two boxes of fish’. 魚. y. Nat. sit. One last point necessary to highlight is that the presence of numeral classifiers in a. n. al. er. io. language does not imply that they are obligatory. As demonstrated in (5a) from Minangkabau, an Austronesian language of Indonesia, and from Malay in (5b) where the usage of numeral classifiers is optional. The reason of this optionality is due to specific situations such as context, but since it is not our main subject of discussion, we follow Gil (2013)’s methodology and count optional numeral classifier languages as numeral classifier languages. Further details are displayed in the typological evidence section.. Ch. engchi. i n U. v. (5) Sample of optional numeral classifiers a. Minangkabau (Gil, 2013). b. Malay (Nomoto, 2013:12). duo (ikue) two (CL) ‘two dogs’. dua (buah) two (CL) ‘two books’. anjiang dog. buku book. As a summary, we realize that languages with numeral classifiers differ from others primarily with respect to the following characteristic feature (Senft, 2015): In 7.

(18) counting inanimate as well as animate referents the numerals (obligatorily) concatenate with a certain morpheme, the “numeral classifier”. This morpheme classifies or quantifies the respective nominal referent according to semantic criteria. Therefore, linguists generally differentiate between classifiers and measure words (quantifiers). These two categories are usually defined as follows: numeral classifiers classify a noun inherently, they designate and specify semantic features inherent to the noun and divide the set of nouns of a certain language into disjunct classes. On the other hand, measure words classify a noun temporarily, it can be combined with different nouns and designate a specific feature of a certain noun that is not inherent to it.. 2.2 Genders. 政治大. Within previous studies, the biggest database is found in Aikhenvald (2000) and WALS (Corbett, 2013) among others. It is generally agreed that genders are the most common systems of nominal classification. They are mainly attested in Africa, Europe, Australia, Oceania, while their distribution is existent but not continuous in the Pacific, Asia and Americas, as demonstrated in Map 5.. 立. ‧. ‧ 國. 學. Map 5. Genders in the world languages (Aikhenvald, 2000:78). n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. First of all, it is necessary to point out that the gender discussed here does not only involve the feminine/masculine distinction of nouns, but is related to the noun class system of the world languages. In gender system (also known as noun class system) languages, all nouns of the lexicon are assigned to a number of classes. Saying that a language has two genders implies that there are two classes of nouns which can be 8.

(19) distinguished syntactically by the agreement they take (Senft, 2015). The most basic distinction would indeed be masculine/feminine as in French where livre ‘book’ is masculine and télécommande ‘remote control’ is feminine. Generally, the assignment logic is considered as opaque. In French for example, little clue can be found to explain why a book is masculine while a remote control is feminine. However, other more transparent noun classes are existent, such as animate/inanimate, edible among others. This profusion is especially spotted in languages from the Niger-Congo group, as an example near 20 noun classes are recognized for Proto-Bantu (Corbett, 1991), the exact number being slightly different when applying different definitions and methodologies. An example may still be demonstrated in Table 1 where we see different meanings related to each noun class, including humans (1/2), trees (3/4), fruits (5/6), liquid masses (6), animals (9/10), abstract nouns (14) among others.. 政治大 Meanings. Table 1. Proto-Bantu noun class meanings (Richardson, 1967; Welmers, 1973). 立. Noun Class. Humans, other animates. 學. ‧ 國. 1/2. Kinship terms, proper names. 3/4. Trees, plants, non-paired body parts, other inanimates. 5/6. Fruits, paired body parts, natural phenomena. 6. Liquid masses. 14 15. al. n. 12/13. y. sit. io. 11. Animals, inanimates. Long thin objects, abstract nouns. er. 9/10. Manner. Nat. 7/8. ‧. 12/22. v i n C hAbstract nouns, mass nouns e n g cInfinitive hi U Diminutives. 16.17,18. Locatives (near, remote, inside). 19. Diminutive. 20/22. Augmentive (diminutive). 21. Augmentive pejorative. It is interesting to note that within Proto-Bantu each noun class related to countable nouns can be divided into two branches: one denoting singular and the other plural, as in Table 1 for Humans, the marker of noun class 1 is for pointing out that the noun belongs to the category of humans and is singular. On the other hand, noun class 2 also highlights the human feature of the noun, but with plurality. Nevertheless, for sake of simplicity, the two respectively related variants such as noun class 1 and 2, 3 and 4, 5 and 6 among others are combined in the same raw since they refer to the 9.

(20) same semantic category. Following this logic, categories with uncountable nouns will only have one class, since they do not need to differentiate between singular and plural form, e.g. noun class including liquid masses (6), abstract nouns (11), abstract nouns and mass nouns (14). It is commonly viewed that the main function of genders is to facilitate referent tracking in discourse through semantic classification of nouns (Dixon, 1986; Nichols, 1989; Corbett, 1991). As demonstrated in (6), thanks to gender differentiation, the speaker can avoid repeating the entire noun while keeping the same level of precision: since in French ‘book’ is masculine while ‘remote control is feminine’, the two nouns may be directly replaced by the pronouns ‘he’ and ‘she’. (6) Example of referent tracking with genders (French) A : Le livre et la télécommande. 立. 政治大 the(F) remote control (F). sont où ? where? chaise chair. ‧. ‧ 國. 學. the(M) book(M) and are ‘Where are the book and the remote control?’ B : Il est sur la table et elle est sur la he is on the(F) table(F) and she is on the ‘He is on the table and she is on the chair.’. sit. y. Nat. Such a situation would not be possible in Chinese: the neutral pronoun ta ‘it’ would create ambiguity such as ta zai zhuozi shang ta zai yizi shang ‘it is on the table and it. n. al. er. io. is on the chair’. Without the gender helping referent tracking, the speaker would be forced to point out the nouns, as in shu zai zhuozi shang yaokongqi zai yizi shang ‘the book is on the table and the remote control is on the chair’. Otherwise the hearer would not be able to know which referent is on the table and which one is on the chair. Similar situation is found with English which only has a pronominal gender system, it is explained in the later part of this section. Nevertheless, non-gender languages will rely on different ways to fill the gap of ease in discourse. This will be explained in the theoretical discussion sections.. Ch. engchi. i n U. v. To avoid confusion in terminology, it is important to highlight that nouns such as ‘brother’ and ‘sister’ denoting male and female are not enough to constitute a gender system. As stated by Corbett (2013), the defining characteristic of gender is agreement: a language has a gender system only if we find different agreements ultimately dependent on nouns of different types. In other words, there must be evidence for gender outside the nouns themselves, as demonstrated by the example with Russian in (7). The three phrases display similar number, case and syntactic 10.

(21) structure yet the different genders of the nouns are reflected in agreement to the verb: ‘magazine’ is masculine, ‘book’ is feminine and ‘letter’ is neutral. (7) Gender agreement with verb in Russian (Corbett, 2013) a. Žurnal. ležal. na stole.. magazine(M) lay(M) on table ‘The magazine lay on the table.’ b. Kniga book(F). ležal-a na stole. lay(F) on table. ‘The book lay on the table.’ c. Pis´mo letter(N). ležal-o na stole. lay(N) on table. 立. ‘The letter lay on the table.’. 政治大. ‧ 國. 學. ‧. Following this logic, Chinese does have lexical contrasts such as nanhai ‘boy’ versus nühai ‘girl’, but does not have a gender system. This is a matter of lexical semantics, and not a gender system. As demonstrated in (8), we can see that the verb does not show agreement with the different subjects, as opposed to the Russian sample in (7).. y. Nat. al. n. magazine not.see. le.. er. bujian. io. a. zazhi. PFV. Ch. ‘The magazine disappeared.’ b. shu book. bujian. le.. not.see. PFV. sit. (8) No gender agreement with verb in Chinese. engchi. i n U. v. ‘The book disappeared.’ c. xin letter. bujian. le.. not.see. PFV. ‘The letter disappeared.’ Another confusing example would be the changes obtained from derivational morphology (as in English: actor/actress), once again this type does not form a gender system. The reason is that there can be numerous similar oppositions, concrete versus abstract for example, none of which would be counted as grounds for postulating a 11.

(22) grammatical category in the language in question (Corbett, 2013). Similarly, inflectional markers are not adequate to propose a gender system: as in French, the word final inflection –e is often viewed as an indicator of feminine gender as in (9). (9) Masculine and Feminine nouns in French a. pont bridge, masculine rideau curtain, masculine b. table table, feminine chaise chair, feminine However, numerous exceptions can be found, as an example livre ‘book’, téléphone ‘telephone’, exemple ‘example’ are all masculine nouns even if they do have a word final -e, as demonstrated in (10) by the agreement they take on the determiner and adjective. This shows once more that agreement evidence should be our primary basis for definition of gender systems.. 政治大. 立. un. grand. one(M) big(M). livre. book(M). ‧. ‧ 國. a.. 學. (10) Agreement of exceptions in French. ‘a big book’. y. nouveau téléphone.. sit. un. Nat. b.. al. n. ‘a new telephone’. er. io. one(M) new(M) telephone(M). Ch. engchi. i n U. v. In general, we may observe that agreement is the principal criteria for the definition of gender system, whether involving verbs, determiners, adjectives, among others. Nevertheless, one question may arise when analyzing languages such as English, which display gender differences on pronouns but not on verbs, as mentioned previously in this section. An example is demonstrated in (11), the pronouns do change according to masculine/feminine subjects but the verb keeps the same form. (11) Pronominal gender in English a. He is tall. b. She is tall. Languages like English, which mark gender only on personal pronouns, are referred to as pronominal gender languages. They are difficult to categorize for two reasons, as 12.

(23) stated by Audring (2008:95): First, they do not show the syntactic repetitiveness or redundancy of gender marking, which is considered to be the main function of gender: to facilitate reference tracking and discourse coherence. In other words, genders are normally marked across word classes such as determiners, adjectives, verbs among others, as shown in (12) with French, but pronominal gender systems do not display this phenomenon, as demonstrated in (11). Second, pronouns are the least canonical of the possible target for agreement: they are phrase-external and even clause-external with regard to their antecedent while agreements targets are expected to share a local domain with their antecedent, e.g. the phrase. (12) Gender agreement across word classes in French La grande voiture est arrivée. the(F) big(F) car(F) is arrived(F) ‘The big car has arrived.’. 立. 政治大. ‧. ‧ 國. 學. Nevertheless, we still count pronominal gender systems as grammatical gender systems in our study due to two reasons. First, most scholars do include the connection between the anaphoric pronoun and its antecedent as agreement (Barlow 1991, 1992:134–152; Siewierska 2004:221–227; Corbett 1991, 2001, 2006, 2013).. sit. y. Nat. The main reason for such a choice is as stated and demonstrated by Audring (2008:96): co-reference alone cannot explain why pronouns mirror the features of their antecedents in languages that do not have a semantics-based gender system. As. n. al. er. io. an example in (13), the semantic of the two sentences is the same but the pronouns still vary according to the noun’s gender, which would be difficult to explain without referring to agreement: in (13a) fototoestel ‘camera’ is of neutral gender while camera ‘camera’ in (13b) is of common gender, resulting in the use of the different pronouns dit and deze. Second, even though the behavior of pronominal gender and common gender is different, e.g. the gender system of English is definitely less pervasive than in French or Russian, including them, however, makes little difference to the overall picture, since they are rare (Corbett, 2013).. Ch. engchi. i n U. v. (13) Agreement evidence for pronominal gender systems in Dutch a.. b.. Dit fototoestel is niet van mij, het is van mijn broer. DEF.SG.N camera(N) is not of me 3SG.N is of my brother ‘This camera is not mine, it’s my brother’s.’ Deze camera is niet van mij, die is van mijn broer. DEF.SG.C camera(C) is not of me, 3SG.C is of my brother ‘This camera is not mine, it’s my brother’s.’ 13.

(24) As a summary, the main purpose of the gender system (or noun class system) is to facilitate discourse by providing a device of reference tracking. This system classifies all the nouns of a language into specific classes, which can be recognized through agreement with different constituents of the clause, e.g. determiner, adjective, verb among others. Following this criterion, it is then necessary to differentiate between grammaticalized gender systems and sporadic features contrasts. The second situation is only a difference in the referents of the nouns, as the masculine/feminine distinction between ‘brother’ & ‘sister’ or ‘actor’ & ‘actress’, which can be obtained via different nouns in the lexicon or inflectional/derivational morphology.. 2.3 Plural markers Besides noun classification, another major distinction marked on the noun is number, meaning the singular/plural differentiation (Corbett, 1991), as an example the prefix -s in English, which is generally combined to the noun to denote the notion of plural, as demonstrated in (14). This system of morphological change is also the most common in the world languages (Dryer, 2013).. 立. 政治大. ‧ 國. 學. ‧. (14) Example of plural marking (English) one table three tables one book three books one cat three cats. er. io. sit. y. Nat. al. n. The geographical distribution of languages with plural marking system is displayed in Map 6, the main area of concentration in black color being Europe and Africa.. Ch. engchi. i n U. v. Map 6. Plural marking in the world (Adapted from Haspelmath, 2013 & Dryer, 2013). 14.

(25) When looking in to details, we may realize that obligatory plural marking of all nouns is found throughout western and northern Eurasia and in most parts of Africa (Haspelmath, 2013). While optional plural marking is common in Southeast and East Asia, and complete lack of plural marking is particularly found in New Guinea and Australia. As mentioned previously, the singular/plural distinction is indeed present in most languages of the world (Haspelmath, 2013), however the form may vary. For languages such as French, English, Spanish, among others, number marking occurs through morphological process, as demonstrated in (14). For some other languages, it involves more complicated procedure, as in Standard Arabic, speaker can denote plural by changing the noun stem or by suffixing, depending on the form of the noun, as demonstrated in (15). If the noun ends with a feminine marker -a, as in (15a), the plural form will involve a change of the ending of the noun to -aat. On the other hand, if the noun does not have the feminine marker, as in (15b) the change will be applied. 立. 政治大. ‧ 國. 學. on the noun stem itself. This phenomenon which is actually linked to gender is further explained in the theoretical sections.. n. al. sit er. io. men(M). y. Nat. rijaal. ‧. (15) Different plural marking systems (Arabic) a. kalima word(F) kalimaat words(F) b. rajul man(M). Ch. i n U. v. Nevertheless, one point must be highlighted regarding the definition of plural: in our study we apply the criteria of agreement, as proposed by Corbett (2000). The main reason is that the presence or absence of plural marker itself is not sufficient, as demonstrated by the example: ‘three sheep’ in English. If the existence of marker was our sole key parameter, we would have to count sheep here as singular, since we don’t see the suffix -s. However, we know that it is plural from verb agreement, as in ‘three sheep are sleeping’ and not ‘*three sheep is sleeping’. This allows us to filter out some elements such as collective markers and associative markers which do not. engchi. denote real plurality (Comrie and Vogel, 2000; Daniel and Moravcsik, 2013). As an example for collectivity, there is the marker men in Chinese, as in lao shi men ‘the group of teachers’, the marker men here highlights the homogeneous group feature of the teachers, rather than pointing out their additive plurality as common plural markers do. A similar situation is observed for associative plurals carrying the meaning ‘X and other people associated with X’, as an example in Japanese: 15.

(26) Tanaka-tachi ‘Tanaka and his associates’, the difference with collective plurals being that the group members are heterogeneous. In both cases, the markers do not carry grammatical agreement. Following this logic, we can realize that plural markers also denote the mass/count distinction between nouns (Her, 2012). As an example in (16), only count nouns can apply plural while mass nouns must rely on a measure word (also named quantifier) to specify the quantify of the noun, as in three cups of water, the measure word-like cup provides detailed information on how much water the speaker is talking about. Further details supporting this evidence are provided in the theoretical sections. (16) Mass/count distinction in nouns (English) three books three tables *three water *three salt. 立. 政治大. al. er. io. sit. y. Nat. 2.4 Summary. ‧. ‧ 國. 學. As a summary, plural markers can occur in various forms therefore it is preferred to rely on agreement to distinguish between grammatical plural markers and other elements denoting similar but different notions of plural. Moreover, one of the main functions of plural marker is to denote the mass/count feature of the noun.. n. Previous studies generally agree on the fact that plural markers denote the mass/count distinction between nouns, while genders (or noun classes) provide noun classification, both of these functions being testable through agreement. Numeral classifiers have a different behavior but still carry the two functions simultaneously. Finally, in geographical terms, it appears that the presence of numeral classifiers in a language coincide with the absence of grammatical genders and grammatical plural markers. As an example, South-East Asia is a hot spot for numeral classifier languages, which usually do not have genders or grammatical plural markers in their system.. Ch. engchi. 16. i n U. v.

(27) Chapter 3 Theoretical discussions In this section, we will combine the separated information of numeral classifiers, grammatical genders and grammatical plural markers together. This is done by groups of two, such as numeral classifiers & plural markers, numeral classifiers & genders, genders & plural markers. Finally, we propose an overview of the triangle relationship between these three elements.. 3.1 Numeral classifiers and plural markers It is claimed that the main function of grammatical plural markers is similar to numeral classifiers: count/mass distinction (Tsuo, 1976; Borer, 2005; Her, 2012; Doetjies, 2012). Taking an example from Her (2012): the -s suffix in English, which is applicable to all count nouns, is seen as a general classifier, similar to the Chinese general classifier ge in (17a) and (17b); the two thus share the same constituent. 立. 政治大. ‧ 國. 學. ‧. structure, as in (18). Following this logic, English relies on the plural marking system therefore does not have numeral classifiers, vice-versa in Chinese, which uses numeral classifiers instead.. y. Nat. sit. al. er. beizi cup. n. san ge 3 CL b. three cups. io. (17) Comparison of numeral classifiers (Chinese) and plural markers (English) a. 三個杯子. Ch. engchi. i n U. v. (18) Syntactic structure of numeral classifiers (Chinese) and plural markers (English). The evidence of this approach is a combination of syntax and mathematics. Starting from Greenberg (1990), researchers such as Au Yeung (2005), and Yi (2011), Her (2012a) among others proposed to compare the structure of [Num C] as [n × 1]. In a multiplicative operation, the multiplicand is null if its value is 1, meaning that it can 17.

(28) be omitted, which is exactly the same rule applying for numeral classifiers and plural markers. A sample from Her (2012) is demonstrated in (19), where we can realize that the mathematical operation is the same, with the only difference that Chinese employs a numeral classifier e.g. ge and English uses a plural marker -s, but they can both be considered as a multiplicand with the value of 1. (19) Unification of numeral classifiers and plural markers Chinese: [[3 × 1] cup] = [3 ge beizi] English: [[3 × 1] cup] = [3 –s cup] A further proof of their redundancy and resemblance to a null multiplicand is the fact that in both languages numeral classifiers or plural markers are generally required however they can be omitted without affecting the meaning of the phrase. In Chinese *san beizi ‘three cup’ or in English *three cup, even though it is ungrammatical, there can be no misunderstanding of its meaning. Examples can even be found in English itself where the noun does not carry a plural marker but has the plural meaning, e.g. three fish, three deer, three sheep.. 立. 政治大. ‧ 國. 學. ‧. Another evidence is that in English, for example, the plural marker -s is still required even when the value of the numeral is smaller than 1, e.g. 0.5 apples and 0 apples and not *0.5 apple and *0 apple, indicating that -s here has no relation with plurality. The plural marker -s thus serves the same function as a general numeral classifier. sit. y. Nat. n. al. er. io. highlighting the countability of the noun. It may be argued that the plural marker does not occur with the number 1, e.g. *one cup(*s), while it does in Chinese, e.g. yi ge beizi ‘one CL cup’. However, Her (2012) explains from a mathematical approach that in a clause where the numeral is one, we would obtain the equation of [1× 1]. In this case, either the multiplier (the numeral) or the multiplicand (numeral classifier or plural marker) can be omitted without affecting the result. This optionality is typologically demonstrated by the study of Her (2012), as an example in Khasi, an Austro-Asiatic language in India, when the numeral is one, the numeral classifier is obligatorily omitted (Temsen 2007), while in Persian it is the numeral (Gebhardt. Ch. engchi. i n U. v. 2009).. 3.2 Numeral classifiers and genders Numeral classifiers are proposed to converge with grammatical plural markers in terms of mass/count distinction. A similar situation can be observed with genders in the domain of nominal classification. As explained in the literature review, the main 18.

(29) purpose of genders is to facilitate referent tracking in discourse. Numeral classifiers can have the same function, as demonstrated in Chinese in (20a), with a phrase such as wo mai le yi ben ‘I bought one CL-volume’. Even though the noun is not present, the hearer is still able to narrow down the possibilities to a book since the numeral classifier is highlighting the feature of a volume. In (20b), since the numeral classifier is narrowing down the noun to something with a 2D flat surface, we may deduce that the speaker is talking about papers, cards, posters, among others. (20) Referent tracking via numeral classifiers (Chinese) 了一本 a. 我買 wo. mai. le. yi. Ben. I buy PFV one CL-volume ‘I bought one (book shape thing).’ 了一張 b. 我拿. 政治大 yi zhang 立 one CL-2D surface. ‧ 國. 學. wo na le I take PFV ‘I took one (2D flat thing).’. ‧. It is not possible to obtain the exact referent since each numeral classifier represents a category of nouns sharing the same feature. The effect is similar with the gender example of book and remote control in French in (6), the noun classification can help the speaker and hearer in conversation however it cannot replace each noun with. sit. y. Nat. n. al. er. io. precise accuracy. To reach such precision would imply that each noun has a distinct category, which would be against the original principle of economy. To summarize, numeral classifiers and genders converge in terms of semantic function, meaning that they both fulfill the purpose of noun classification.. Ch. engchi. i n U. v. Nevertheless, the two systems do have divergences: in general, gender systems are more rigid than classifiers, as listed in Table 2. As an example, genders obligatorily mark every noun in a language regardless of register or context, while classifiers may be optional in specific situations, as demonstrated in (4) via Chinese. Moreover, genders have a closed system compared to classifiers, meaning that it is very rare to see a new gender appear in a language, while it is quite common to observe the emergence of new numeral classifiers. As in French, the usage of two genders has been stable since a long period of time, while the usage of numeral classifiers shows the opposite in Chinese, in which new numeral classifiers developed along the ages, sometimes replacing the previous ones. This also leads to the fact that genders are generally outnumbered by classifiers in terms of average quantity, as an example, 19.

(30) most languages have only two or three genders, with Bantu at most may have near 30 genders attested in previous studies, however Chinese on itself have more than 100 numeral classifiers in the lexicon. Moreover, the rigidness of genders can also be observed in grammatical agreement, which is not a requirement on the classifiers side. Table 2. Main differences between genders and classifiers (Dixon, 1986) Gender system. Classifier system. 1. Classify all nouns. Do not classify all nouns. 2. Into a smaller number of classes (from 2-20). Into a larger number of classes. 3. Of a closed system. Of an open system. 4. May fuse with other grammatical categories. Independent constituent. (Def, Num, Case) 6 7. Can be marked on noun. Not affixed to noun. variation. classes at speaker’s will. No variation in register. Formal/ informal uses. 治 Realized in agreement政 patterns Marked once 大 N uniquely assigned 立 to a class with no speaker N possibly assigned to various 學. 8. ‧ 國. 5. ‧. Finally, genders rarely generate speaker individual variation or change of usage according to register, while classifiers do. As an example from French with genders, a table is labeled as feminine and it is rare to observe a speaker assign a different class to it. Furthermore, the table will remain feminine regardless of the speaker and hearer. sit. y. Nat. n. al. er. io. involved in conversation. On the other hand, numeral classifiers are the total opposite: one noun can be assigned different numeral classifiers depending on the feature the speaker wishes to highlight, resulting in speaker variation. Taking the action of pointing out a fish as an example, in (20a) the speaker may use a numeral classifier related to the tail feature of the fish, but it is also possible to highlight its long shape in (20b) or its animacy in (20c).. Ch. engchi. i n U. v. (21) Different numeral classifiers on the same noun (Chinese) a. Highlights the tail feature of the fish 一尾魚 yi wei 1 CL-tail ‘one fish’. yu fish N-fish as frame and CL-tail as profile. 20.

(31) b. Highlights the long shape feature of the fish 一條魚 yi tiao 1 CL-long shape ‘one fish’. yu fish N-fish as frame and CL-long shape as profile. c. Highlights the animacy feature of the fish 一隻魚 yi 1. zhi CL-animacy. yu fish. ‘one fish’ N-fish as frame and CL-animacy as profile. 政治大 Furthermore, numeral classifiers usage may vary according to different registers. As 立 an example in Chinese casual speech, the speaker may heavily rely on the general ‧. ‧ 國. 學. classifier ge which can be associated to almost every noun in the lexicon. However, in formal situation, it is usually preferred to use specific classifiers for demonstrating education attainment and respect to the hearer, as demonstrated in Table 3.. y. Nat. Table 3. Usage of general classifier and specific classifier (Chinese) Official (Specific classifier). sit. Casual (General classifier) b. yi ge zhuozi ‘one CL-general table’. c. yi ge fangzi ‘one CL-general house’. yi dong fangzi ‘one CL-building house’. d. yi ge wenzhang ‘one CL-general article’. article’. n. al. Ch. yi wei lao shi ‘one CL-person teacher’. er. yi ge laoshi ‘one CL-general teacher’. io. a. yi zhang zhuozi ‘one CL-2D surface table’. i n U. v. e n gyicpian h iwenzhang ‘one CL-written thing. As a summary, numeral classifiers and genders converge in terms of semantic function but diverge in their represented form. Their common semantic function would let us expect a tendency not to co-occur in languages (Dixon, 1982; Corbett, 1991; Aikhenvald, 2000; Blench, 2012), as we deduced for numeral classifiers and plural markers. Previous studies involving big data of languages such as Aikhenvald (2000) do propose results supporting this hypothesis. However, since the overlap would be semantic rather than syntactic, it is not expected to be forbidden, rather only dispreferred. A similar case can be found in our explanation of numeral classifiers with the number one, since they both represent the mathematical value of 1, they are displaying a semantic overlap yet still can occur, just in a low frequency. Following 21.

(32) this logic, it is indeed unusual for a language to have both genders and numeral classifiers but examples can still be found: Tariana, a North Awarakan language spoken in Brazil is attested to have a gender system and three sub-types of classifiers (Aikhenvald, 1994). Other examples of genders and classifiers coexistence can also be observed in various languages such as Tidore (van Staden, 2000:77-81) or Ngan’gityemerri which shows the development from generic classifiers into genders (Reid, 1997). Further statistical evidence is provided in the GIS section.. 3.3 Genders and plural markers The convergence of genders and plural markers is motivated by the fact that they both carry different semantic functions. Genders fulfill semantic classification while plural markers represent countability marking. The two systems separately fulfill the two main classifications present in noun phrases, leading us to the expectation that they should co-occur in languages. This is actually what is observed in most languages of the world, taking French as an example, gender marking is existent as demonstrated in the previous sections. Furthermore, plural marking is also realized through suffixes, as. 立. 政治大. ‧ 國. 學. demonstrated in Table 4.. ‧. Table 4. Genders and plurals co-occurrence in French. y. Plural. trois grandes tables. ‘one(F,sg) big(F,sg) table(F,sg)’. ‘three big (F,pl) tables (F,pl)’. n. er. al. un petit livre ‘one(M,sg) small(M,sg) book(M,sg)’. sit. une grande table. io. Masculine. Nat. Feminine. Singular. trois petits livres. v i ‘three small(M,pl) books(M,pl)’ n Ch U engchi. Syntactically this is also possible, since genders and count/mass distinction are supposed to be assigned at the lexical level rather than by syntactic processes (Her, 2012). Moreover, the semantic functions and syntactic forms of the two elements are both different therefore no conflict occurs among them. As a resume, we realize that genders and plural markers separately fulfill the two main classification functions of noun phrases therefore they are expected to complement each other, as a contrast to numeral classifiers which merge these two functions into one entity. 3.4 Summary: Triangle relationship We propose that despite their apparent divergence, the three elements display a particular distribution in languages due to their common features. As displayed in 22.

(33) Table 5, the noun phrase normally requires: countability (mass/count distinction) and noun classification. First, numeral classifiers carry both simultaneously, as in the example where the numeral classifier assigns the noun to the category of ‘long-shape objects’ and highlights the fact that it is countable. If it was a mass noun, the noun phrase would use a measure word (quantifier) instead of a numeral classifier, e.g. shi bang rou ‘ten M-pound meat’ (Li & Thompson, 1981). Second, genders provide semantic classification to facilitate referent tracking, as demonstrated in the introduction (Luraghi, 2011; Contini-Morava & Kilarski, 2013). Finally, grammatical plural markers point out the countability of the noun, as in English where only count nouns can take plural marking, e.g. ‘some tables’ but ‘some water’ (Sanches & Slobin, 1973; Greenberg, 1990; Ghomeshi & Massam, 2012). Table 5. Semantic functions of numeral classifiers, genders and plural markers. 政治大 Yes Chinese ‘one CL-long fish’. Countability Classification Example Yes. Gender. No Yes. Yes. French ‘table, fem’/ ‘book, mas’. No. English –s ‘three tables’. 學. Plural marker. 立. ‧ 國. Num classifier. ‧. From features in Table 5 we can obtain the relationship graph in Fig 1. We hypothesize that in a language, numeral classifiers and plural markers are mutually exclusive (line 1), numeral classifiers and genders have the tendency to not co-occur (line 2) while genders and plural markers are expected to co-occur (line 3).. er. io. sit. y. Nat. al. n. Figure 1. Interaction of numeral classifiers, genders and plural markers. Ch. engchi. i n U. v. In line 1, numeral classifiers and plural markers both mark countability in semantic function and syntactic form, therefore should not occur together (Borer, 2005; Hsieh, 2009; Her & Chen, 2013). In line 2, numeral classifiers and genders share the semantic function of classification so are expected not to co-occur (Dixon, 1986:111, Blench, 2012). However, since the overlap is in semantic function rather than syntactic form, the redundancy is acceptable but would occur in low frequency 23.

(34) (Greenberg, 1974; Li, 2000). In other words, the agreement of gender is syntactic, therefore it has no connection with countability and it may co-occur with numeral classifier. However, they have the same semantic function of classification, making the overlap redundant, thus dispreferred. In line 3, genders and plural markers bear separately the two types of information: countability and classification. Therefore, they should appear in the same environment, unless another carrier fulfilling the same purpose is found, e.g. noun classifiers, morphological case marking, among others. This exceptional situation is demonstrated in the typological section. A typological example would be that the lack of gender and number markers in Sino-Tibetan, Miao-Yao and Tai nouns is somewhat compensated for by numeral classifiers, which serve to individuate nouns and may be said to ‘agree’ with particular classes of nouns e.g. flat, round, elongated objects among others (Matisoff, 1991:496).. 政治大. It is necessary to explain that in line 1 the two elements are mutually exclusive since both numeral classifiers and grammatical plural markers share the same semantic function and syntactic form, while syntactic stacking is not possible. Moreover, they. 立. ‧. ‧ 國. 學. distinguish between two separate meanings, e.g. in English ‘book’ and ‘books’ are definitively different, but marking the mass/count distinction twice as in ‘*bookss’ would not add anything new to the word, making the redundancy useless. However, for line 2 the shared feature of noun classification is not in syntactic form but in semantic classification and does not distinguish a meaning, rather it is an optional system of noun classification for referent tracking. In this sense, one could add as. sit. y. Nat. n. al. er. io. many categories as one wants in order to obtain a maximum of precision, as it would be when adding adjectives to a noun, e.g. the big yellow French book. A similar process with noun classification would result in something like ‘the book, masculine/ inanimate/ square-shape’. This makes the overlap acceptable, but due to economy tradeoff in language, it is still expected to appear in low frequency. In other words, since the purpose of noun classification is to highlight a specific feature of the noun, adding more than one would cancel the stand-out effect of a feature, therefore it is dispreferred.. Ch. engchi. i n U. v. Combining the deduction mentioned previously, we may display the two functions via the concept of continuum: each language is searching for a balance between economy and expressiveness. For the speaker, the more economical the better, since less energy is invested in speaking. On the opposite side for the listener, the more expressive the better, since less energy is needed to interpret the message. In other words, countability and classification are two sources of information available for noun phrases, the most equilibrated situation occurs when both functions are fulfilled once 24.

(35) each, while marking only one of it or neither of them leads to more economy but less expressiveness and vice-versa, as demonstrated in (22). (22) Equilibrium continuum between economy and expressiveness within languages Economy Expressiveness << --------------------------------------------------------------------------- >> Countability Classification Language. ✓ ✓ Piraha Finnish. Usan. ✓ ✓ Chinese/ French. ✓ ✓✓ ✓✓ ✓✓ ✓ ✓✓ Tidore Hungarian Marathi. The most equilibrated point is expected to occur in the middle, when both countability and classification are fulfilled once, e.g. Chinese has numeral classifiers which carry countability and classification marking once, French relies on genders for classification and plural markers for countability. Other distributions may occur but the frequency is predicted as low. As an example, when the language prefers economy. 立. 政治大. ‧ 國. 學. ‧. over expressiveness, it would only include one function between countability and classification, e.g. the Piraha language from the Amazonas only shows countability marking, while Finnish only marks countability via plural markers. A most extreme example in this direction would be Usan, a language from Papua New Guinea, which does not mark either of the two functions. On the other hand, a language may also. sit. y. Nat. n. al. er. io. tend to enhance expressiveness. In this case it would mark one function twice, e.g. in Tidore noun classification is fulfilled once by numeral classifiers and a second time by genders. Most extreme cases also exist, as in Marathi which is attested to have numeral classifiers, genders and plural markers, apparently resulting in double marking for countability and classification. Our speculation regarding the high ratio of balanced languages in the middle of the continuum is further supported by our data in the GIS section, but it is also necessary to point out that beside countability marking and noun classification, it is possible that other functions may be required in the noun phrase. Moreover, the hierarchy between the two main functions involved in (21) also needs further discussion. Due to resource limitation and avoidance of a too large scale. Ch. engchi. i n U. v. for this paper, these issues are left for further research to develop. As a summary, we expect that the two functions of countability (mass/count) and classification may explain the distribution of numeral classifiers, grammatical genders and grammatical plural markers. First, countability marking is syntactic therefore it is not possible to show stacking, following this logic we propose that numeral classifiers 25.

(36) and grammatical plural markers should be mutually exclusive. Second, noun classification is semantic therefore it is possible to be represented in more than one form. Thus, to avoid losing the function of highlighting a noun’s feature it would still be dispreferred. Under this hypothesis, numeral classifiers and grammatical genders are expected to show the tendency of not co-occuring. Finally, since grammatical genders and grammatical plurals separately fulfill countability and noun classification, they are not affecting each other and are expected to co-occur. It is also necessary to point out that numeral classifiers, grammatical genders and grammatical plural markers may not be the only carriers fulfilling the function of countability and classification marking, e.g. noun classifiers, grammatical case marking are also attested to be plausible candidates. However, in this study we first focus on the most common three categories of numeral classifiers, genders and plural markers to obtain a general overview. A brief analysis of noun classifiers and case marking is still presented in the typological evidence section, nevertheless more detailed research in this subject is required for further studies.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 26. i n U. v.

(37) Chapter 4 Supporting evidence This section is divided in three parts: first we briefly explain the methodology of data gathering. Second, we run statistical typological analysis to verify our hypothesis in the languages of the world. Third, based on GIS system, we display and explain the areal distribution and historical development of numeral classifiers, grammatical genders and grammatical plural markers in the world’s languages.. 4.1 Methodology To gather sufficient data and test our hypothesis, we choose to expand our coverage as much as possible to get a big picture of the languages in the world, and then we go into details for each language. Two main criteria are set in terms of language coverage:. 立. 政治大. ‧. ‧ 國. 學. speaker population and language genealogy. For speaker population, we aim at including in our sample a quantity of speakers reaching 60% of the world population, as to follow the actual tendency of the actual world. The second condition of language genealogy is motivated by the fact that language diversity must also be taken into consideration. As an example, English is one of the most spoken languages in the world however it may be resulting from the influence of domains such as economy and politics rather than being solely dependent on a well-structured language system.. sit. y. Nat. n. al. er. io. Therefore, we need to cover different language groups to test if our hypothesis does predict a correct tendency. Our preliminary data contained 80 languages with their information on numeral classifiers, genders and plural markers. It was obtained via the merge of three biggest open-source database: Gil (2013) for numeral classifiers, Corbett (2013) for genders and Haspelmath (2013) for plural markers, as displayed in Figure 2.. Ch. engchi. i n U. v. Figure 2. Preliminary database of 80 languages. 27.

(38) Then, we checked which main language groups or geographical regions were not sufficiently covered in these 80 languages to enlarge our inventory obtaining a database of 155 languages, as displayed in Map 7 with each dot representing a language. This result is obtained through the combination of the biggest available open source database: Gil’s (2013) survey on numeral classifier languages, Corbett’s (2013) discussion on genders, Haspelmath’s (2013) research on plural markers, SIL (Ethnologue) language records and information from professor Her research team in the syntax lab of National Chengchi University, e.g. data on 454 numeral classifier languages. The detailed list is provided in Appendix 1. Map 7. Display of 155 surveyed languages (by dots). 立. 政治大. ‧. ‧ 國. 學 er. io. sit. y. Nat. al. n. In terms of world population the speakers of the 155 languages we gathered reached 65% of the world’s population according to SIL data, while in terms of language genealogy we covered 30% (39/129) of the main language groups in the world, e.g. Indo-European, Austronesian, Sino-Tibetan among others. Our genealogical coverage may seem insufficient, however we already included in our study 90% (18/20) of the top 20 biggest main language groups in the world (Indo-European, Sino-Tibetan, Afro-Asiatic, Niger-Congo, Austronesian, Dravidian, Japonic, Altaic, Austro-Asiatic, Tai-Kadai, Creole, Nilo-Saharan, Uralic, Quechuan, Hmong-Mien, Mayan, North Caucasian, Language isolates), which account for 99% of the world speakers. Ch. engchi. i n U. v. population according to SIL, therefore we estimate it sufficient. Following the distribution on the map in Map 5, we may realize that Africa, Europe and Asia are covered however Australia and the Americas would need more representative languages. The main reason for this point is that the literature is quite controversial on whether the languages in these areas rely on numeral classifiers, noun classifiers, genders (noun classes) or other types of systems, therefore to avoid incorrect 28.

(39) judgment of categorization, we do not include languages with different structures attested. Finally, to assure that our coverage in terms of language speakers is sufficient, in Map 8 we also display the language speaker distribution via polygons of the QGIS software. The black area represents the presence of speakers from the 155 languages, showing that in demographic terms we do cover the main parts of the globe, since every continent is colored. Map 8. Display of 155 surveyed languages (by speaker population). 立. 政治大. ‧ 國. 學 ‧. Before going into the details of the typological evidence it is also necessary to highlight that some records of previous studies database have been modified within our coding. One example would be Chinese. In Gil (2013), Chinese is attested to have numeral classifiers, and noted by Corbett (2013) as genderless which is indeed. sit. y. Nat. n. al. er. io. supported by our data. However, Haspelmath (2013) database on nominal plural categorize Chinese as having optional plural marking on human nouns. As a reminder, we defined that our study only targets grammatical plural markers, this criteria automatically rules out Chinese since it is a language without number agreement. Therefore in our database we count Chinese as without plural markers. A possible source of confusion could be the Chinese collective marker men, but since it is more analyzed as a clitic without grammatical agreement rather than a grammatical plural marker attached to the noun, we do not count it as a plural marker (Rijkhoff, 2000:240). Further evidence on agreement and definiteness are shown in (23).. Ch. engchi. (23) Collective marker men in Chinese a. Phrase with one teacher 一位老師出去了 yi wei laoshi chuqu le one CL-person teacher go out PFV ‘One teacher went out.’ 29. i n U. v.

(40) b. Phrase with men added 老師們出去了 men chuqu le laoshi teacher Collective go out PFV ‘The teachers went out.’ c. Conflict of indefinite marker and men *那裡有一些老師們 men nali you yixie laoshi there is some teachers Collective ‘There is some the teachers there.’ As demonstrated in (23a-b), when men is attached to the noun lao shi ‘teacher’, it seems to reflect plural however it does not have grammatical agreement with the verb. Moreover, men shows definiteness, as shown in (23c) with the impossible co-occurrence with indefinite marker such as yi xie ‘some’, which is different from the conventional plural markers, e.g. with -s in English, teachers would be indefinite and some teachers entirely grammatical. Therefore, the men is interpreted here as a. 立. 政治大. ‧ 國. 學. ‧. collective marker which highlights the homogenous feature of the group members, rather than being a grammatical plural marker. As a reminder, the detailed coding and references for each language are provided in Appendix 1.. sit. y. Nat. n. al. er. io. 4.2 Typological evidence. Ch. i n U. v. Following our assumptions in Table 5, we know that mathematically eight combinations are possible for the presence (＋) or absence (－) of numeral classifiers,. engchi. genders and plural markers. According to our concept of continuum in (22), only two of them should be optimal in the world languages, with countability and classification fulfilled once either by numeral classifiers alone (Chinese) or the combination of genders and plural markers (French). Three are under-marked: if the language favors economy, one or both of the two functions may not be marked, resulting in under-marking of noun information, e.g. languages only marking countability with plural markers, just marking classification via genders or not marking either of the two. Finally, three are over-marked: if the language prefers expressiveness, one or both of the two functions may be marked twice, e.g. languages having numeral classifiers and genders would experience an overlap of noun classification, while languages with numeral classifiers and plural markers display double marking of countability. The last extreme situation would be languages attested with numeral 30.

(41) classifiers, genders and plural markers, resulting in a double marking of both countability and classification. The result of categorizing the 155 languages of our database by the eight combinations mentioned is displayed in Table 6. Table 6. Different combinations of numeral classifiers, genders and plural markers Plural Frequency 1 Optimal marking. Num CL. Gender. Countability Classification (Syntactic) (Semantic) ＋－－＋. Example. Qty. %. (155). French, Swahili. 88. 57%. 2 Optimal marking. －. ＋. ＋. －. Chinese, Japanese. 28. 18%. 3. Under-marking. ＋. －. －. －. Afrikaans, Basque. 13. 8%. 4. Under-marking. －. －. －. ＋. Piraha. 1. 1%. 5. Under-marking. －. －. 5. 3%. 6. Over-marking. －. 2. 1%. 7. Over-marking. ＋. －－ Usan, Yidiny 治政＋＋＋ Mai Brat, Tidore 大＋＋－ Hungarian, Tuvaluan. 9. 6%. 8. Over-marking. ＋. ＋. 9. 6%. 立. ＋. ＋. German, Marathi. ‧ 國. 學. ‧. Combination 1 and 2 both have the two main marking of noun phrases: countability and noun classification represented once therefore are optimal. For combination 3, it only has the marking of countability (plural markers), however since semantic classification (genders) is rather optional, its absence can be tolerated. The similar logic applies to combination 4, which only has the noun classification represented by. sit. y. Nat. n. al. er. io. genders. Combination 5 includes languages with the most economic but least expressive system without either marking. On the other hand, combination 6 has a semantic redundancy of noun classification through the co-occurrence of numeral classifiers and genders, thus is also possible but marked. Combinations 7 and 8 have redundancy of countability marking (numeral classifiers and plural markers) which is also marked, as explained in the theoretical discussions.. Ch. engchi. i n U. v. In terms of typological distribution, the results show that 87% of the data is in accordance with our predictions. Furthermore, the anomalies can be explained when analyzing the details. We provide a quick overview first in the following while further details are demonstrated in the incoming sections. For category 7 and 8, two situations can generally be found: first, the languages are attested to have numeral classifiers but their inventory is very limited, as in German which only has one optional numeral classifier attested (Sussex and Cubberley, 2006) or Tuvaluan with just some classifier-like elements (Besnier, 2000; Gil, 2013). Second, even though the language does have numeral classifiers and plural markers, they do not co-occur on the noun, as 31.