The Problem of Defining Word - Coursework Example

November 26, 2007 The problem of defining ‘word’ The Oxford English Dictionary defines the term ‘word’ as: “a sound or combination of sounds forming a meaningful element of speech, usually with a space on either side of it when written or printed, used as part (or occasionally as the whole) of a sentence.” This definition, however, attempts to incorporate different approaches in linguistics that are not necessarily in agreement with each other. ‘a sound or combination of sounds’ would refer to a phonological definition of ‘word’; ‘a meaningful element of speech’ is a lexical approach to the definition of the term; and ‘with a space on either side of it when written or printed’, would be a definition in terms of an orthographic approach. Linguists have always struggled with the definition of the term ‘word’. “No one has ever been able to define the word word despite gargantuan efforts to do so. The linguistic concept of word: an analytic bibliography by Alphonse Juilland and Alexandra Roceric is a 118-page bibliography of books and articles (unsuccessfully) attempting to define word over the past 3 millennia. Why can no one define word? Maybe because words simply do not exist; rather, the sentences we speak are composed of lexemes and morphemes and these two linguistic objects differ too much to be subsumed under one concept.” (Dr Goodword’s Office) The smallest meaning-bearing elements of language are termed as morphemes. In any language, morphemes mark the crucial categories of grammar such as progressive aspect, plural number, past tense, negation, definite, possession, etc. Morphemes can be defined in a language-independent manner. However, it is to be borne in find that no adequate language-independent definition of the word as a unit has been agreed upon (Karlsson, 1998, p. 83). Prepositions, prefixes such as ‘re’, ‘ex’ and ‘un’, suffixes such as ‘ing’, ‘s’ and ‘ed’, and particles like ‘the’, ‘not’ and ‘so’ would fall in the category of morphemes. Lexemes, on the other hand, are anything that is used in speech to name or refer to things in the real world. Nouns, verbs and adjectives would therefore fall in the category of lexemes. The objective of this paper is to examine why it is so difficult to define the term ‘word’ defies linguistic definition. This will be done by adopting a two-pronged approach: i. Analysis of the different perspectives which has attempted to define the term ‘word’ to identify why a common definition or meaning is not possible. ii. Examination of the differences between morphemes and lexemes to identify why these two basic linguistic objects cannot be covered under a common definition of the term ‘word’. The difference concepts of ‘word’ Words are defined as orthographic units in the printed or visual form. An orthographic word is a sequence of signs, symbols or letters which has a white space at either end but no white space in the middle. This is a rather visual definition of ‘word’ and has no existence in speech or audio form of the language. Orthographic words run into problems with compounds. ‘Roof top’ can also be written as rooftop, and roof-top. In the first case it is two orthographic words, while in the second and third case it is a single orthographic word. All the three forms nevertheless have the same meaning. Other examples of similar compounds are ice-cream, pocketknife, doorkeeper, commonsense, toyshop, etc. Since the rules of orthography in English do not specify which compounds should be written with a white space in between, individual preferences vary considerably. Though the orthographic concept of word is helpful in learning to read and write English, and is important in the linguistic study of writing systems, orthographic words are irrelevant for most linguistic purposes and therefore hold very little linguistic interest. There are also many languages which do not use space to separate words, and could hence not be brought under an orthographic definition. However, English orthography tells us something important about the language (Bauer, 2004). Phonological words is more related to audio perceptions where a word is defined as a unit of pronunciation based on some criteria that vary from language to language. The phonological word is defined as a piece of speech. In English the most important criteria is that the phonological word will contain only one main stress. A sentence in any language will have certain stress points. For example, when the sentence ‘The young boys’ll have to go there’ is pronounced in a relaxed colloquial manner, five main stress points fall on the words ‘young’, ‘boys’, ‘have’, ‘go’ and ‘there’. The sentence can therefore be broken up into the following phonological words: [The young] [boys’ll] [have to] [go] [there] What is evident here that all the phonological words do not correspondent to other acceptable units such as grammatical units. The phonological word ‘boys’ll’ does not lend itself to any grammatical interpretation. Phonological words are important in the study of phonetics and pronunciation, but are irrelevant to the study of grammar. A lexeme or a lexical item or object, is defined as an abstract unit of the lexicon or vocabulary of a language. A lexical item has an identifiable meaning or function. It has to be represented in writing or in speech by one of the many grammatical forms that it can assume. The canine animal ‘lion’ can be represented either using the singular form ‘lion’ or the plural ‘lions’. ‘Lion’ and ‘lions’ are the grammatical forms of the same lexical abstract item ‘lion’. In the same manner, the abstract lexical item ‘give’ has five grammatical forms – ‘give’, ‘gave’, ‘given’, and ‘giving’. Only the lexical items are listed by dictionaries. ‘Lion’ and ‘lions’ are both listed under ‘lion’; all the grammatical forms of the lexical item ‘give’ will also be listed under give and not separately. The number of grammatical forms for each lexical item can vary greatly. In English, the lexical item ‘be’ has the maximum of eight grammatical forms while most have fewer than this, and some have only one. One of the grammatical forms of each lexical item is chosen to represent the lexical item. This grammatical form is known as the citation form or dictionary form and is used to name the lexical item, refer to it, and list it in the dictionary. In English, the citation form of a lexical item carries no grammatical markings. This base form is the natural choice for the citation form. For a noun like ‘lion’, the singular ‘lion’ is chosen as the citation form, and not the plural ‘lions’; for a verb like ‘give’, the infinitive ‘give’ is taken as the citation form, and not an inflected form such as ‘given’ or ‘giving’; for the adjective ‘small’, the positive form ‘small’ is taken, and not an inflected form such as ‘smaller’ or ‘smallest’. However, the availability of such grammatically unmarked base forms for a majority of lexical items are almost unique to the English language and are not available in other languages. Trying to define words in terms of lexical items, runs into problem in the case of defective lexical items. A defective lexical item lacks some grammatical forms that are inherent to lexical items of its class. ‘Furniture’ and ‘sheep’ have no plurals while ‘oats’ and ‘people’ have no singular form. In the case in which one of the missing forms is the form that is normally chosen as the citation form, then one of the other forms has to be chosen. Since there is no ‘oat’ the plural ‘oats’ with grammatical marking has to be chosen as the citation form of the lexical term oats. A lexical item therefore usually has one grammatical form or base form with no grammatical markings and other grammatical forms with grammatical markings, which are assumed by the lexical item for grammatical purposes. The grammatical forms with grammatical markings are known are as the inflected items of the lexical item. In exceptional cases such as ‘police’, ‘sheep’, ‘with’ and ‘beware’, there are no inflected forms of the lexical items. This approach attempts to define ‘words’ from the perspective of grammatical word-forms. However, besides the overt grammatical forms that have been discussed, many lexical items exhibit what can be termed as implicit grammatical form. The verb ‘give’ take different forms in the past and past participle tense. ‘He gave the book’ and ‘He has given the book’. However, many other verbs do not. ‘He completed the paper,’ and ‘He has completed the paper’. This would imply that the past tense ‘completed’, and the past participle tense ‘completed’ are in fact two different implicit grammatical forms. Another example can be cited in the case of present tense form and infinite form of English verbs. The form of the verb ‘give’ remains the same in ‘You should give donations’, which is the infinite, as also in ‘I always give donations,’ which is in the present. Even though there is no change in form, many linguists would still hold the view that there are in fact two grammatical forms depending on the usage. Not only do lexical items have different inflected forms, they also give rise to entirely new lexical items through what is known as derivation. Derivation is the construction of a new lexical item from an existing lexical item by the addition of a prefix or a suffix. ‘Fishy’ is a derivative of ‘fish’ which has only one grammatical form and by itself is a noun. ‘Fishy’ as in ‘There is something fishy going on,’ is an adjective. ‘Fishy’ is therefore derived from ‘fish’ by the addition of a suffix. It is a distinct lexical term, and finds a separate entry in dictionaries. Similarly, ‘catty’ comes from ‘cat’, and ‘doglike’ comes from ‘dog’. Other examples of derivation by the addition of suffixes are ‘nudism’ from ‘nude’, ‘happiness’ from ‘happy’ and ‘kingdom’ from ‘king’. Examples of derivation by the use of prefixes are ‘unhappy’ from ‘happy’, ‘rewrite’ from ‘write’ and ‘undo’ from ‘do’. Some examples of derivations however do not follow any distinct pattern and are irregular such as in ‘destruction’ from ‘destroy’, and ‘comprehensible’ from ‘comprehend’. Again, English has a derivation that is called zero-derivation or conversion in the case of which the there is no affixation to the lexical item. The lexical item is simply shifted from one class of words to another depending on the usage. The adjective ‘slow’ as in ‘the slow worker’ is converted to a verb in ‘slow it down’; then again, the noun ‘shoe’ in ‘new shoes’ is converted to a verb in ‘shoe the horses’. In these examples the, the zero-derived words has to be considered as belonging to a different lexical item from the source, i.e. the ‘slow’ which is a noun and ‘slow’ which is a verb must be considered to be two different lexical items. It becomes all the more confusing when the direction of derivation in the case of zero-derived items is not obvious. The word ‘punch’ is used as a verb in ‘he punched me’ and as a noun in ‘he gave me a punch’. One is a derivative of the other, but it is not possible to distinguish between the derivative and the root. The definition of words is further complicated in English by lexical items that appear to be words by a certain set of rules, but which are found in two or three parts sometimes separated by other material. Phrasal verbs which consist of a verb and one or two particles fall in this category. Make up, take on, put off, put up with, etc. are examples of phrasal verbs. Though the phrasal verbs may comprise two or three orthographic or phonological words, their meanings, based on their application, could be very different from their constituents. For example the phrasal verb ‘made up’ can be used as in ‘he made up a story,’ or ‘he made up with his girlfriend,’ or ‘she made up her face’. In many cases, the constituents of phrasal verbs can also be separated by other material so that they are discontinuous as in ‘turn the lights on’ instead of ‘turn on the lights’. Then there are the prepositional verbs that consist of a verb, or sometimes an adverb, and a preposition as a single unit as in call on, take to, look down on, etc. Even though syntactically the prepositions would seem to combine with the receiving noun to form prepositional phrases as in ‘we called on Susie’, the meanings that the prepositional verbs imply put them in the category of single lexical items. Where to fit in the Clitics and the Short Forms The definition of the term ‘word’ faces a peculiar barrier in the form of clitics. A clitic represents lexical item and a grammatical word form, but do not make a phonological word by itself. A clitic depends on its host, a more substantial item, to form a phonological word. The English articles a, an and the along with the auxiliary ‘ll are examples of clitics. The articles bind with a following host and the auxiliary binds a preceding host to form phonological words. Simple examples are ‘he read the book’, ‘eat an apple a day’, and ‘she’ll come today’. A clitic which precedes a host is known as a proclitic, and a clitic that follows a host is known as an enclitic. Though clitics are deemed to be marginal words, they could be said to fall in status somewhere between an autonomous word and an affix. Then we have the many short forms – the abbreviations, the logograms, the contractions, the acronyms and initialisms, and the clipped forms. Where and how do these short forms fit in on the definition of the term ‘word’? An abbreviation is a short way of writing a word or a phrase, using alphabets and full stops. Abbreviations can be written only, and have no pronunciation of their own. They are usually read by their full forms or spelled out letter by letter. Examples are ‘Prof.’ for ‘Professor’, ‘Dr’ for ‘Doctor’, ‘km’ for ‘kilometer’, ‘C’ for ‘degrees Celsius’, and ‘e.g.’ for ‘for example’. An abbreviation may be considered as an orthographic word but is not a word in any other sense. It is not a part of spoken language. A written character which is not an alphabet but which conventionally represents a word or a collection or sequence of words is known as a logogram. Monetary symbols such as ‘$’ for ‘dollars’, arithmetic symbols such as ‘+’ for ‘plus’ and digits such as ‘6’ for ‘six’ are examples of logograms. A logogram represents a lexical item or a grammatical word-form, and could fit the definition of orthographic words unless rejected on the basis of non-alphabetic representations. Contractions such as she’ll, won’t, and I’m constitute of single orthographic and phonological words but comprise multiple lexical items and grammatical word forms. Any generic definition of word would have to be accorded a considerable degree of flexibility to accommodate contractions. Acronyms are formed by extracting the first letters of a phrase or a name consisting of several words and putting them together to form a separate entity. BBC for the British Broadcasting Corporation, AIDS for Acquired Immuno-Defficiency Syndrome, UK for the United Kingdom and NATO for North Atlantic Treaty Organization (NATO). According to British convention, when a formation of this kind is pronounced by reading it out letter by letter, it is initialism; but it is read out as an ordinary word as in NATO and AIDS, then it is an acronym. Certain acronyms such as ‘scuba’ for ‘self-contained underwater breathing apparatus’, ‘laser’ for ‘light amplification by the stimulated emission of radiation’ and ‘radar’ for ‘radio detection and ranging’, have been transformed to perfectly ordinary lexical items better known by the acronyms then their root phrases. An initialism or an acronym can indeed be considered to be a lexical item. An acronym does not differ from any other lexical item except in its unusual written form. An initialism is a lexical item with an unusual written form and an unusual pronunciation. A clipped form is a part of a longer phrase or word. The process of extracting the part from the whole is known as clipping. ‘Gym’ for ‘gymnasium’, ‘bike’ from ‘motorbike’, ‘flu’ from ‘influenza’, and ‘phone’ from ‘telephone’ are examples of clipped forms. Clipped forms are genuine lexical items. They can take on grammatical inflections typical of their word class such as plurals in phones, gyms, bikes, etc. Verb clipped forms can accept ordinary verbal inflections. Like any other lexical items, clipped forms can form compounds as in ‘phone book’, ‘porn star’, etc. The disparate morphemes and lexemes Lexemes refer to the real world and semantic content i.e. it has a readily identifiable meaning. The morphemes, on the other hand, have little or no identifiable meaning, but have one or more grammatical functions. Lexemes undergo derivation, morphemes do not. While ‘guitarist’ is derived from the lexeme ‘guitar’, ‘runner’ from ‘run’ and ‘bigness’ from ‘big’; morphemes such as ‘the’, ‘that’, ‘of’, ‘for’ and ‘it’ have no derivatives. Lexemes always come with a sound and a meaning, but morphemes can have either sound or meaning, but not both. The pronunciation of lexemes remain more or less fixed wherever they may occur in a phrase, the pronunciation of morphemes, especially that prefixes and suffixes, can vary to a great extent. Finally, lexemes belong to an open unlimited class. There is not limit to the number of nouns, verbs and adjectives in a language. New lexemes may be derived from existing lexemes in the same language or borrowing from other languages. Morphemes are however limited in number to around 200 in the English language. Lexemes and morphemes can therefore never be clubbed together. And as long as this distinction remains, the definition of ‘word’ will always remain tentative. The result is the many ways in which the meaning of words is interpreted. The answer to the question: ‘what is a word?’ will always remain an enigma. The utilization of morphemes as basic representational units in a statistical language model instead of words seems a promising course (Creutz & Lagus, 2002). References 1. Korytskyy, O. (2003), Morphemes and Lexemes as Language Determinants, CADSM’2003, February 18-22,2003, Lviv-Slasko, Ukraine, pp. 552 - 554 2. Karlsson, F. (1998). Yleinen kielitiede. Yliopistopaino, Helsinki University Press, pp. 83 3. Bauer, L. (2004). Adjectives, Compounds and words, Nordic Journal of English Studies 3/1 (=Worlds of Words: A tribute to Arne Zettersten) pp. 7-22. 4. Breutz, M. & Lagus, K. Unsupervised Discovery of Morphemes, Association for Computational Linguist. Read More