The Morphology of Loanwords in Urdu: the Persian, Arabic and English Strands Submitted by: Riaz Ahmed Islam In fulfilment of the requirements for the degree of Doctor of Philosophy Supervised by: Dr. S. J. Hannahs & Prof. Maggie Tallerman School of English Literature, Language & Linguistics Newcastle University October, 2011 © 2011 Riaz Ahmed Islam All rights reserved i Abstract The Morphology of Loanwords in Urdu: the Persian, Arabic and English Strands Language contact and the influence of one language on another are very common phenomena. Persian, Arabic and English have influenced various languages globally. Urdu is one of the recipient languages from these three sources, and shows linguistic features borrowed from them. This study focuses on the Persian, Arabic and English loanword morphology in Urdu. Loanwords from Persian/Arabic are far older than English loans, and function like native Urdu words. Therefore, native Urdu morphological structures and those from Persian and Arabic are treated as the patterns for English loans. The discussion describes the patterns and then the processes involving English loans in the light of these patterns. The hypothesis is that the affixation, whether inflectional or derivational, may be based on native Urdu patterns but that the compounding of English loans is more frequent with Persian and Arabic loanwords. This is a major factor, which needs to be established. It is equally vital to know whether Urdu also borrows any derivations of an English loan, as it did with Persian/Arabic loans with or without any morphological changes. Almost nothing is written on the morphology of loanwords, from the three languages, into Urdu. Furthermore, there is no theory on loanwords specifically dealing with the morphological adaptation of loans. So, the present work is descriptive and deals with the characteristics of the morphological structures from native Urdu, Persian, Arabic and English. Due to space restrictions, the primary focus is on gender/number and case morphology, and derivation of by affixation and by compounding. The study is divided into six chapters. The discussion begins in the first chapter with an introduction to the study and an overview of the sociolinguistic background of Urdu. It also discusses the influence of English loanwords on South Asian languages in general and Urdu in particular. The chapter exemplifies pluralisation of English loans and compound verb forms with the words of recipient languages. The next three chapters focus on inflectional morphology, derivation by affixation and derivation by compounding. In chapter 2, only the gender, number and case morphology of the Urdu noun with relevance to the three sources comes under discussion. Chapter 3 focuses mainly on derivational affixes in Urdu from the three sources. Although a discussion of morphological issues is the main concern, some phonological and semantic issues with relevance to morphology are also included. However, phonological issues are only discussed in connection with Arabic loanword phonology, which shares in lexical creations e.g. ɣʊnɖa ‘scoundrel’. Derivation of new words is also very frequent by means of compounding. Constituents from two different sources very often interact and are rather more frequent than normal, i.e. native + native, combinations in Urdu. Therefore, rather than looking at their source languages, Chapter 4 focuses on various types of compounds i.e. endocentric, exocentric and copulative etc. Chapter 5 discusses the features of English loanwords adaptation in the light of Persian and Arabic loanwords adaptations. Morphological changes occur both on the inflectional and derivational level. On the inflectional level, the changes are more frequent and based on the native Urdu patterns. Derivational changes are seen in various loans, but the adaptability is limited in the derivation of other categories irrespective of native Urdu or Persian and Arabic patterns. It is far less frequent than the adaptation seen in Persian and Arabic loans. It is more frequent with native Urdu affixes, but the formation of compounds is more frequent with Persian and Arabic loan constituents. Thus, the hypothesis made in the beginning of the study is supported. English loan affixes have not found a place in formal Urdu, although they are used informally. The chapter draws some conclusions. Chapter 6 then presents a summary of the discussion made in the thesis, and presents the implications of the study. Chapter 1 Morphology and Loanword Morphology 1.1. Introduction Language contact and the influence of one language on the other are very common phenomena. Persian, Arabic and English are some of the great languages of the world. They have influenced various other languages globally. Although a great language in itself, Urdu is also one of the recipient languages. Lexical influence from various sources is a common occurrence, but it is significant to note that a recipient language shows the loan morphological and phonological features. The same can be seen in Urdu as well, but almost nothing has been studied on any of the linguistic aspects. A great many features are unexplored. This study attempts to bring some of these aspects to the surface. It focuses on the loanword morphology in Urdu and looks at loanwords from Persian, Arabic and English. This study focuses equally on the morphology of loanwords from these languages. However, the loanwords from the former two are far older than English loans, and function like native Urdu words. They are far more morphologically integrated, which means that they have adopted various morphological changes. Therefore, in this perspective, the investigation into morphological adaptation of English loans is done by looking at the patterns of native Urdu morphological structures and the morphological structures of the older loans. Thus, the morphological structures of native Urdu, Persian and Arabic are treated as the model for the morphological adaptation of English loans, which is discussed in the final part of this study. As English loanwords are more recent, it is necessary to look briefly at some of the background of English loanwords in Urdu. 1 This is because the Urdu lexicon is basically an amalgamation of the native Urdu words and Persian and Arabic loans. Therefore, the morphological structures which are apparent as a whole are also an amalgamation of the morphological structures from these three sources. ‘Native Urdu’ is understood to mean a word belonging to Urdu, Hindi or even Sanskrit, provided that it is commonly used in Urdu. The reason for this is that linguists generally treat Urdu and Hindi as the same language, known by different names, i.e. Urdu, Hindi, Hindustani and Hindi-Urdu. According to Rai (2000: 11), “one man’s Hindi is another man’s Urdu”. Linguists normally use the term Urdu-Hindi or Hindi-Urdu to study the common features despite the fact that some lexical, morphological and phonetic differences do exist. Sanskrit is considered to be the mother of both, but due to deep and distinct influences, Urdu is highly Persianised/Arabicised, while Hindi is highly Sanskritised. Studying loanwords and loanword integration, the distinction and combination of two criteria of conformity, i.e. conformity of loanwords to the source language form and conformity to the target language system, help us better understand the diachronic processes of progressive integration of loanwords (Winter 2008: 156). By invoking conformity in both, one of two alternative possibilities must be addressed. (1) The properties of a loanword are unchanged in a borrowing language and they match those of the source language. (2) The properties of a loanword have changed, and they now match those of a target language system in that the loanword seems to be part of borrowing language vocabulary. Analysing conformity to the (SLF) Source Language Form and conformity to the (TLS) Target Language System, the main goal of this study is to establish first the relation between the Source Language (Persian and Arabic) and the Target Language (Native Urdu), with relevance to the morphology of loanwords. Then, treating the morphological structures of native Urdu, Persian and Arabic as the three faces of Urdu morphology as a whole, it is interesting to look at English loanword 2 morphology. The hypothesis is that the affixation, whether inflectional or derivational, may be on native Urdu patterns but that the compounding of English loans is more with the older loans (Persian and Arabic), which form a bigger part of the Urdu Lexicon. This is one of the factors which need to be established. Moreover, it is equally important to know whether Urdu also borrows any derivations, of an English loan, as it borrowed Persian and Arabic derivations with or without any morphological changes. The study generally observes the extent of the English loanword adaptation, in the light of Persian and Arabic loanword adaptation, and establishes if it is as complete as that of the two older loan influences. There are other related questions, e.g. what are the structural patterns after the loanword integration; how are the loanwords modified, and which of the elements are affected? It is also of interest to note whether Urdu borrows English affixes in the same way as it has borrowed Persian and Arabic affixes. Two points in particular need to be considered. Regarding Persian and Arabic influence, there is extremely little written on Urdu linguistics in general and its morphology in particular. There is nothing written on English loanword influence on Urdu either. Therefore, the lack of any related literature is an obstacle. Moreover, there is no theory of loanwords dealing specifically with morphological adaptation of loans. In this context, the present work does not fit clearly into any specific theoretical framework. Therefore, there is need for good, careful descriptive work before serious theoretical analysis can be carried out. So, much of the current work is descriptive and deals with the characteristics of the morphological structures from the three prominent sources i.e. native Urdu, Persian and Arabic. Following this, the discussion of the morphological adaptation of English loans describes the processes involving English in the light of the three sources of morphological structures already shown. 3 Morphological adaptation includes all aspects of derivational and inflectional integration. However, due to space restrictions, the primary focus is on some major points, including gender/number and case morphology, the derivation of nouns, adjectives and verbs by affixation and by compounding. The proposed study is divided into six chapters. The examination starts in the first chapter with an introduction to English loanwords in Urdu, and more generally morphology and loanword morphology of the South Asian languages. A general picture is sketched regarding the status of English loanwords and how nouns and verbs in particular are adapted in some South Asian/ Indo Aryan languages. In the subsequent chapters, the focus is on the morphology of the Urdu noun itself (Chapter 2), derivation by means of affixation (Chapter 3) and by compounding (chapter 4). Chapter 5 takes a close look at the adaptation of English loans. The final chapter (Chapter 6) concludes the discussion summarising Urdu morphological features and those of English loans. The key point in this chapter is to have a look at the general implications of the study. In order to discuss these points, the first factor required is to understand morphology or word structure, in particular inflection and derivation. Beyond that, it is equally important to understand the issues involved in loanword morphology. 1.2. Morphology Morphology is the study of words and their internal structures mainly through the analysis of morphemes, affixation, reduplication (partial or full) and various sorts of compounding. Morphemes are various types of word parts, e.g. prefix, stem, infix and suffix. Thus, a single Urdu word, e.g. laɪlmi ‘unawareness’ can be broken into three morphemes, i.e. the Arabic prefix la ‘un’, Arabic stem ɪlm ‘knowledge/awareness’ and native Urdu suffix -i. The prefix la converts the noun ɪlm into an adjective laɪlm which is then converted into another noun laɪlmi 4 by the suffix -i. Sometimes, independently occurring words in a compound can also participate in morphological processes. For example, the Urdu word kəmra-e-dʒəmaət̪ ‘classroom’ is a compound word and both the constituents kəmra ‘room’ and dʒəmaət̪ ‘class’ are two independently occurring words. The function of the infix -e- is to link both the constituents and give possessive meaning. It replaces the semantic genitive ka/ ki/ke, which cannot occur in compounds. Many morphologists make a distinction between inflection and derivation, which are often associated with affixation (the attachment of a morpheme). Inflection does not change the category but rather the form of the same word, e.g. plæt ‘plate’, plætæ(cid:619) ‘plates’, plæto͂ (plural oblique) plæto (plural vocative). In the given Urdu examples, plæt is the singular noun, -æ(cid:619) is a plural marker in the second, while -o͂ and -o show the oblique and vocative forms of the noun, respectively. Inflected forms are variants of one and the same word. Derivation, on the other hand, means new word formations from a base word. This may change the category of the base, but a derivation may also be in the same category as the base word. A noun, for instance ʊsul ‘principle’, can be converted into the adjective ʊsuli ‘in principle’, the noun and adjective beʊsul ‘(someone) unprincipled’ and an adverb ʊsulən ‘in principle’. A loanword typically follows the morphological rules of a borrowing language. It may be helpful to know how languages influence each other and what morphological results occur. 1.2.1. Loanword Morphology When speakers of a language want to identify with each other, they may find themselves adjusting their speech to eliminate the more obvious differences in pronunciation or vocabulary. But when multilingual speakers interact with each other, it often involves 5 language contact. This can be seen in Kupwar, India, where multilingualism has led to the convergence between the local dialects of two Indic languages, Marathi and Urdu, and two Dravidian languages, Kannada and Telugu (Thomason 2001: 45). Kupwar is a small town of 3000 inhabitants who speak the four languages. Due to centuries of contact, the varieties of the four languages are similar to each other syntactically and are distinct from the varieties spoken in other parts of India. Gumperz and Wilson (1971) note that historically, grammar is thought to be the most persistent, while lexicon is the most changeable component of a linguistic system, but in Kupwar the grammar is adaptable and the lexicon is the most persistent. The words of each language have been retained and coupled with a common syntax. They observe that the adaptations are far-reaching and multilateral, and each language has adopted some features from the others. Bilingualism and education are two important factors. Presently, electronic media is also an important factor. Note the following two examples. The first is Greece, where most people are educated and monolingual (Tatsioka 2008: 129), and the other is the Indian Subcontinent where most people are bilingual and uneducated. In both these regions, only a small number of people know English, but most people frequently use English words and phrases. Expressions like oh God! are very common. This is mainly due to media influence. The consistent use of such borrowings makes them part of the borrowing language. However, loanwords and simple borrowings are two different terms which distinguish them from each other. Haugen (1950: 212) uses the term “borrowing” for the attempted reproduction of some patterns of one language into the other, and the term “loanwords” for one type of borrowing. Loanwords are single words or compounds, but borrowings may be stems or full phrases. Loanwords are also different from code-switching, as Poplack et al (1989: 390) state that the 6 former follows the rules of the recipient language, while code-switching is a linguistic term denoting the concurrent use of more than one language, or language variety, in conversation. The study of how a language reacts to the presence of foreign words – whether it rejects, translates or freely accepts them – may give us insight into its formal tendencies. The adaptation of a loanword depends both on its word class (verb, noun, adjective) and on its similarity to native words and to loanwords from other languages already existing in the borrowing language. The native Urdu affixes are productive and form new words in combination with the words from a different source. But some of the loan affixes e.g. Persian -ana → zalɪmana ‘tyrant’ also derive a good number of words with roots from other sources e.g. Arabic in this case. The productivity of these affixes in general, although systematic, shows some irregularity as well. Seidenberg and Gonnerman (2000: 354) state that, “Morphological systems are quasi regular. They are productive and systematic but also show many seemingly irregular forms.” They note that the inconsistencies in word structures reflect the uncertainty about the nature of morphological units and greatly complicate the task of interpreting empirical results; for example, bakery is related to bake and cannery to can but what is groce in grocery? While looking at similar inconsistencies, Ahmed (2008: 2) states that Urdu “lexical items, prefixes, and suffixes are difficult to study on the spoken level since their occurrence is quite unpredictable.” Although Urdu morphology follows some patterns and rules, the rules may be complex and may cut across linguistic components e.g. morphology and phonology. It cannot be termed ‘unpredictable’, as Ahmed claims. Nevertheless, there are certain deviations from the rules of each pattern. 7