Introduction

From descriptive linguistic point of view, structural analysis of word refers to anatomical analysis of morphemic structure of words. Keeping this in mind, in this chapter an attempt is made to analyse forms and functions of morphemes used in formation of Bengali words. In course of this study, an effort is also made to analyse various morphological and semantic properties of the morphemes so as to understand intricate patterns of inflection and derivation used in generation of final forms. Information obtained from structural analysis of Bengali words is a valuable input in descriptive analysis of words, as well as in language teaching, grammar book writing, and language processing. It is useful for developing a morphological processor that will identify morphemes and other formative elements that are used to generate words in Bengali. Moreover, it can also be utilized to develop a part of speech (POS) tagger that will annotate words based on their syntactic roles in a sentence within a piece of text.

There are several appropriate and approved ways of using morphemes in formation of words in Bengali. Since morphemes are quite recurrently used in different Ways and manners for word formation in Bengali, it is interesting to know which morphemes are used and in what manner for this purpose. To understand this phenomenon, it is better to take a turn from the traditional path and treat words as separate linguistic entities with unique surface structures. Based on this assumption, analysis of surface structure of words shows that morphemes are nothing but building blocks which are used to construct words that are considered as the basic units of linguistic expressions in speech and writing of a speech community.

This chapter is developed in the following ways. In the following sections some general discussion on the orthographic representation of Bengali words is presented and the general and most frequent word formation methods in Bengali are discussed.

Introduction

Postposition is one of the most important linguistic components of the grammar of the Bengali language. ‘Use of postposition to denote case relations is found in IA, Kol and Dravidian’ (Chatterji, [1926] 1993). Proper understanding of form and function of postpositions used in modern Bengali texts carries tremendous linguistic advantages in understanding the language in its present form and usage. In a general sense, what is preposition in English is postposition in Bengali (and in many other Indian languages) as these two unique linguistic elements exhibit close similarities in syntactic–semantic functions although they occur in two different positions in the sentence in the respective languages. It is ‘useful to treat postpositions as a word class in Bangla, not only to help foreign learners, but also because many of the locative noun forms or perfective participles concerned have changed or expanded their meaning in their use as postpositions’ (Thompson, 2010).

Although postpositions used in modern Bengali texts are functionally similar to prepositions used in English, the prepositions are used before nouns and pronouns in English (e.g., of father, in house, by hands, in morning, at station, etc.) while postpositions in Bengali are usually used immediately after nouns and pronouns in sentence.

Introduction

Reduplication is one of the highly productive word formation processes in Bengali. It is quite frequently used in the language for various linguistic reasons and purposes. Also, it is used in different ways and manners to serve various means of information-sharing and communication. Although it is mostly used to express a sense of multiplicity of various countable items, it is also used as a process to refer to the act of continuation of an action or an event.

The process of reduplication is indeed very rich and robust in Bengali. Due to this fact, each and every word used in the language is virtually capable of being reduplicated. However, due to several orthographic inconsistencies, reduplicated words are often written with or without a space between two members. It is observed that the lexical item, which is reduplicated, can either be a word, a morpheme, a syllable, a sequence of syllables, or a string of characters, which does not form a prosodic constituent such as morpheme, syllable, or root, etc. It is also observed that reduplication can occur to all word-classes including the pronouns and indeclinable. In most cases, the process signifies plurality of an item or continuity of an action. Moreover, all kinds of echo and onomatopoeic words, which are quite prevalent in the language, also fall within the list of reduplication.

In this chapter I have tried to analyse the Bengali reduplicated forms with reference to their structure to explore how various parts of words tend to concatenate to generate the final forms and what kind of morphophonemic alternation takes place in the whole process of generation of reduplicated form.

This study may help readers to identify the processes these words undergo without considering their semantic properties.

Introduction

In linguistics, the term ‘word’ is a conceptual enigma. Although linguistically it is treated as one of the fundamental compositional units of a language, in the history of linguistics, the independent identity of a word has often been questioned, challenged or ignored. Scholars of linguistics have been often reluctant to acknowledge the separate linguistic identity of words because of their varied surface forms and functions within a piece of text.

This leads Matthews (1974) to argue that it is not a word, but a morpheme, which is important in establishing relation of a word with phonology, syntax and semantics. Aronoff (1981) also ignores the separate identity of word as he considers word as nothing but a phonetic string, which is connected to other linguistic entities lying outside the string.

Selkirk (1983), on the other hand, treats word from a purely syntactic point of view as she identifies a set of word formation rules, which are applied to generate words. Bybee (1985) interlinks words with morphology to argue that a word should never be studied free from its meaning, as the meaning of morphemes and contexts determine several linguistic properties used in formal expression of words.

Jensen (1990) argues that it is not words but morphemes, which are primary structural units, and which are typically but not necessarily meaningful. Therefore, it is sensible to focus on morphemes rather than on words in word formation. Spencer (1991) tries to build up an interface that underlies between morphology and phonology to understand linguistic entity of words. His basic idea of word includes inflectional morphology with an underlying interface between syntax and morphology by which one can explore the processes involved in word formation.

In essence, word is a complex linguistic concept, which is difficult to define in straight terms. Citing different examples from various languages as well as observations of various experts of the field, it can be argued that the concept of word is actually interlinked with several linguistic issues such as, pronunciation, lexicology, orthography, morphology, grammar, meaning, derivation, inflection, convention, usage, etc. which directly or indirectly play crucial roles in identification of words in a natural language.

The book is all about the descriptive study of surface form and usage of Bengali Words based on data and information obtained from a large corpus of modern Bengali written prose texts. The corpus contains more than five million words of written text samples collected from more than eighty five different subject domains published between 1981 and 2000. This book has exhaustively used all kinds of data, information and examples from this corpus only. This book is, therefore, the first exhaustive Work on modern Bengali words with close reference to actual data and information of real-life language use.

In this book, I have made an empirical attempt to analyse Bengali Words and other lexical items from the perspective of their surface orthographic forms to understand the internal structure of their composition with a clear focus on their functional roles in various contexts of their usage in texts. To achieve this goal, I have to bridge up an intelligible link between the internal composition and external representation of the words in the interface of usage and function of the Words in texts. Although several attempts have been made by earlier scholars to analyse Bengali words from etymological or morphological point of view, serious attempt has hardly been made to explore into the structure or surface forms of words. The primary motivation is to look into the texture of Bengali words as they are found to be used in texts and analyse the information embedded Within their structure from the lexico-semantic as well as functional perspective. The corpus data, used for this study, supply large amount of information about the form and usage of words – based on which I have been able to draw deductive inferences with regard to the patterns of use of words and Word-formative elements in the language. The issues that are adequately addressed in different chapters of this book include decomposition of words, interpretation of function of Word-formative elements and analysis of lexico-semantic identities of these components in relation to their function in words.

Introduction

The term affix that I have considered in this chapter includes all inflectional and derivational elements as well as other formative elements, which play important roles in the formation of words in Bengali. These are the most productive components in the act of generating new words in the language. If the total number of words available in modern Bengali is divided into two types, namely, the non-affixed words and the affixed words, it will be clearly noted that the number of affixed words far exceeds the number of non-affixed words. Due to this fact, these formative elements ask for attention in structural analysis of words in the language.

In a general sense, the list of affix includes the deviational and inflectional elements, which are tagged with root, stems or bases to generate new derived and inflected words, respectively. While the derivational affixes generally take part in the formation of new verbs, nouns, and adjectives, inflectional affixes are mostly used in formation of inflected nouns, pronouns, and conjugated verb forms. That means derivational affixes are mostly operative within the parts of speech of verbs, nouns, and adjectives, while inflectional affixes are mostly operative within the part of speech of nouns, pronouns, and verbs. Words belonging to other parts of speech such as, postposition, demonstrative, adverb, and indeclinable do not normally use affixes to form new words in the language.

Although affixes are considered dependent morphemes, these are productive in the sense that these are regularly tagged with stems, bases, and words to form new words in a language (Spencer, 1991).

Introduction

In this chapter I have made a systematic corpus-based study to show how the vocabulary of modern Bengali language is full of a large number of new English words and terms that have recently crept into the Bengali vocabulary to make room for new concepts, items, and ideas. This study can also show that these newly borrowed words and terms have made exiting equivalent words and terms of Bengali functionally redundant. I have made an empirical attempt to explore how the vocabulary of the present Bengali language is undergoing linguistic metamorphosis at the lexical level by drawing heavily from the English vocabulary as a result of several linguistic and extralinguistic factors, and how new English general words, as well as scientific and technical terms, are being incorporated into the Bengali language through various methods of naturalization.

I have presented here some of the findings, which I have obtained from his analysis of a Bengali corpus of prose texts (1981–1995) as well as from a lexical database manually compiled from various types of modern Bengali prose texts. In a systematic way I have shown how naturalization is taking place in Bengali vocabulary with complex operation of linguistic processes such as adoption and adaptation at lexical, phonological, and morphological levels. This pilot study plants the seeds for further investigation into this area with some large databases compiled from heterogeneous texts of the modern Bengali language.

In theory, naturalization is an ongoing process by which various linguistic items of a language are gradually naturalized or nativized in another language by means of applying various morpho-phonological processes, which are quite active in the borrower language. Naturalization becomes active in a language when a need arises for increasing the stock of vocabulary for addressing various linguistic and non-linguistic needs of the language users.

The conditions and the motivations of naturalization are almost similar to those observed in the case of lexical borrowing (Hockett, 1958).

Problems with words in Bengali

In the context of processing Bengali words through a computer, there may arise several issues that are directly linked with surface structure of words. These issues may create problems in manual and computer-based counting of number of words in a corpus. They can also create problems in morphological processing of words. These issues come up because there is hardly any consistency in orthographic representation of words in written Bengali texts. The high irregularities in writing of inflected words, proper names, adjectival forms, adverbial forms, compound words, reduplicated words, onomatopoeic words, hyphenated words, etc. present a daunting task before an investigator in normalizing the surface forms of words for generating a lexical database as well as developing a word processing system for the works of language technology.

Perhaps, this is not the problem of the Bengali language alone. Most of the Indian languages also face similar problems due to inconsistencies in orthographic representation of words in writing. Therefore, before the work of frequency count of words is carried out on the Bengali corpus, it was necessary to adopt the following strategies without which it might have created some wrong observations and false deductions.

1. • Each unbroken string of characters is separated with a white space before and after it is treated as a single word unit. The white space before and after the string has worked as a marker of word boundary.

2. • All compound words, adjectival forms, and adverbial forms written with a space in between two members are treated as single word units to maintain consistency in counting their frequency of occurrence in the corpus. That means a white space existing between the two formative members is ignored and the entire form is treated as a single word unit.

3. • Words with a hyphen mark in between two formative members (e.g., compound words, prefixed words, reduplicated words, etc.) are treated as single word units.

4. • Consecutive occurrence of two identical forms of words, separated by space or hyphen, is treated as an instance of a reduplicated word.