Words and morphemes

How can we determine what counts as a lexeme (word) in a language? Without a secure criterion of wordhood, it will be hard to decide – especially in unfamiliar languages – what units we should be trying to attribute meanings to. For European languages with a well-established tradition of literacy, this question usually does not arise: words are the units surrounded by spaces in standard orthography. This definition of ‘word’ will not take us very far, however, for two reasons. The first is that languages which have only recently been written down often have a very fluid practice of word-division. A meaning-bearing unit considered by one speaker as only part of a word will not infrequently be written as a separate word by another speaker (see Dixon and Aikhenvald 2002: 7–9 for details). Speakers of Northern Sotho (Niger-Congo; South Africa), for instance, show two ways of writing the sentence meaning ‘we shall skin it with his knife’ (Dixon & Aikhenvald 2002: 8, quoting Van Wyk 1967). The first is to put spaces between each morpheme:

These differing practices may sometimes become conventionalized in such a way that closely related, typologically similar languages may adopt differing orthographic conventions. For example, Northern Sotho’s relations, Southern Sotho (Niger-Congo, South Africa) and Tswana (Niger Congo, Botswana) are usually written according to the convention in (11), while Zulu (Niger-Congo, South Africa) and Xhosa (Niger-Congo, South Africa), whose morphological structure is entirely equivalent to that of the other group, typically follow the convention in (12).

The second reason to be suspicious of writing as an indicator of word hood is that orthographic practice itself is not even stable within long standing traditions of literacy. An unbroken tradition of literacy links Modern and Ancient Greek. Yet Ancient Greek was written without any word-division, whereas modern Greek observes the norms familiar from languages like English. We would obviously not want to say, however, that Ancient Greek did not have words. Similarly, the reform of German spelling rules made standard (for a trial period) in German schools since 1998 resulted in strikingly different word divisions, as can be seen from the following list:

Linguists have advanced many criteria for the demarcation of the word as an isolable linguistic unit. One common criterion is that of ‘potential pause’: words are units before and/or after which pauses can be found in spoken language. For languages like Chinese, which lack complex morphology, this criterion may be workable. But for languages which show even a small degree of morphological complexity, like English, it is clearly unsatisfactory. Thus, Dixon and Aikhenvald (2002: 11) point out that one may well pause at morpheme boundaries within a single word, for example ‘it’s very un- suitable.’ (Similarly, expletives in English can be inserted within what we normally consider a single word: abso-bloody-lutely.) Bloomfield’s famous definition of ‘word’ (1933: 178), as ‘a minimum free form’, i.e. the minimal unit which may appear on its own without any additional grammatical material, is clearly insufficient: many canonical words like the, of or my do not usually appear alone, but must presumably be considered as fully fl edged words.

In order to introduce some clarity into the confusion over wordhood, it seems necessary to distinguish two different levels on which words may be defined. The first is the phonological level. Here, divisions between words are determined according to the domain of application of phono logical rules and processes. In Dagbani (Gur, Northern Ghana), for instance, the clearest description of stress can be given by assuming the existence of a unit – the phonological word – each example of which bears only one main stress, normally on the penultimate syllable (Olawsky 2002: 206). In order, therefore, to determine whether a given phonetic string is a phonological word in Dagbani, one need only count the number of main stresses: if the unit in question has more than one main stress, then it is more than one phonological word. Furthermore, the fact that the penultimate syllable is typically the tonic (accent-bearing syllable) allows us to determine where the word boundaries lie. Many languages are like Dagbani in calculating stress on the basis of phonological words: as a result, stress is typically a useful indicator of the phonological word. Other indicators are also found, however. Like Dagbani, Bare (Northern Arawak, Brazil) shows penultimate stress (Aikhenvald 1996: 494). But this language possesses an additional marker of phonological wordhood: aspirated consonants can only be found in word-initial position (Aikhenvald 1996: 494): as a result, given a string with n aspirated consonants, one is guaranteed of the presence of at least n phonological words.

 The phonological level alone will often not be enough to demarcate word boundaries. Thus, stress in Dagbani and Bare is only mostly on the penultimate syllable: exceptions are possible, and this can lead to ambiguity in word division. As a result, the grammatical level of wordhood must also be considered. Dixon and Aikhenvald (2002: 19) propose three criteria for the recognition of those linguistic units which are independent grammatical words: cohesiveness, fixed order and conventionalized coherence and meaning. The last criterion ‘indicates that the speakers of a language think of a word as having its own coherence and meaning. That is, they may talk about a word (but are unlikely to talk about a morpheme)’ (Dixon and Aikhenvald 2002: 20).

QUESTION How reliable a criterion of grammatical wordhood is this? Do speakers ever talk about the meaning of morphemes? What about the first two criteria, cohesiveness and fixed order? Ancient Greek (Indo-European, Eastern Mediterranean) provides a clear illustration of both. Ancient Greek verbs were obligatorily multi-morphemic, consisting of at least the elements root + inflection, as in the verb meaning ‘cure’, therapeu-o-:

These elements must co-occur: the verb root therapeu- cannot occur with out an inflectional suffix, and the suffix cannot occur without a verb root. The combination of verb root and inflectional affix thus constitutes a word on the criterion of cohesiveness. These forms also illustrate fixed order, in that one cannot invert the order root-suffix: the inflectional markers are suffixes, not prefixes. As a result, the combination verb root + inflection constitutes an unambiguous grammatical word in Ancient Greek.

If words are the clearest type of meaning-bearing unit in a language, they are certainly not the only ones: the domain of meaningfulness extends both above and below the threshold of the individual word. Below word level, morphemes, by definition, have meanings. Given the definition of a morpheme as the ‘minimal meaning-bearing unit’ of language, it is clearly impossible to conceive of a morpheme without a meaning – even if it is often hard to specify exactly what this meaning is. Quite often in linguistic analysis, it proves surprisingly difficult to come up with a settled analysis of the meaning of a given morpheme. This is the case, for instance, with the meanings of the possessive suffix -s and of many morphemes involved in the verbal tense/aspect system in English (see 9.2): semanticists agree that these morphemes have meanings, but disagree about exactly what they are.

Above the level of the individual word, phrasal verbs and compounds are two clear cases where a single meaning is associated with a combination of lexemes. Phrasal verbs consist of one or sometimes two ‘full’ verbs followed by one or more particles, as in (16):

QUESTION Noun compounding is an extremely frequent means of word formation in English, and shows many different types of meaning relation between the compounded elements: a tree house is a type of house in a tree, but a lighthouse is a type of ‘house’ which contains a light, and a poorhouse was an institution for the accommodation of the poor. A computer problem is a problem with a computer, and a zebra crossing is a crossing that is striped like a zebra. Find twenty examples of noun compounds from a newspaper, and describe the semantic relationships between the constituent parts. Can you discern any regularities?

Idioms, discussed in 1.4.3 in relation to throw in the towel, also demonstrate the existence of units of meaning associated with several words simultaneously, and we will consider the question of the meaning of grammatical constructions in a later chapter (10.3). Thus, although we most often think of meaning as something belonging to individual words, we must actually recognize that words are only the most obvious of a number of meaning bearing units.

اخر الاخبار

اخبار العتبة العباسية المقدسة

بالتعاون مع قسم الشعائر.. المواكب العزائية تحيي ذكرى شهادة الإمام جعفر الصادق (عليه السلام) في كربلاء

المجمَع العلمي يقيم الاختبار المركزي لطلبة الحفظ في بغداد

قسم الشؤون الطبية ينظّم زيارة لدعم الأطفال المصابين بالسرطان في مستشفى الإمام الحسن (عليه السلام)

قسم الشؤون الفكرية يطلق مسابقة (حامل اللواء) بنسختها الثانية في بغداد

المزيد

الآخبار الصحية

بديل طبيعي للفلورايد يحمي أسنانك من التسوس

تحذير للرجال.. أطعمة شائعة الاستهلاك قد تضر بصحة البروستات

اكتشاف بديل طبيعي لـ"أوزمبيك" يخفض الوزن دون آثار مزعجة

خبراء يكشفون لماذا تنام النساء أكثر من الرجال ؟

المزيد

الآخبار التكنلوجية

الأرض "تومض" ليلا!.. كوكبنا ازداد سطوعا بنسبة 16% في غضون 8 سنوات

أحد أخطر البراكين العملاقة على وجه الأرض يمتلئ بالصهارة!

علماء يطورون تقنية جديدة تُسرّع التئام الجروح عبر تنشيط خلايا الجلد

بعد مهمة "أرتميس 2".. أوروبا تخطط لإرسال رواد فضاء إلى القمر

المزيد

مواضيع ذات صلة

Mismatches between grammatical and phonological words

The units of meaning

Word-based and meaning-based approaches to definition

The history of the dictionary

Meaning and use

Meanings as brain states

Meanings as concepts/mental representations

Breaking the circle

Object language and metalanguage

Levels of meaning

Compositionality

Sense/reference/denotation/connotation