Units of grammar
When grammarians break complex strings of language down into parts, they do so only as far as the smallest unit of meaning: the morpheme. Of course, individual speech sounds are smaller than most morphemes, but most individual speech sounds do not function as morphemes and therefore do not carry meaning. While ‘grammar’ in the broader sense might encompass a model of phonology, this area has its own complex set of terms that we do not explore here. The diagram in Figure 14.6 illustrates the grammatical units of varying sizes for which linguists have developed a set of terms. Some of these grammatical units should already be familiar from earlier chapters in the book. The sentence is represented as the largest grammatical unit because larger pieces of discourse consist of sentences joined together in a variety of ways.
As we have seen, the morpheme is the smallest unit of language that can carry meaning. Some words, like house, consist of a single morpheme, while others, like house-s or employ-ment consist of more than one morpheme. The study of morphology, then, is the study of word structure. Morphemes that can stand alone, like house, are free morphemes, whereas those that need to attach to something, like plural -s, are bound. The simplest possible form of a content morpheme is called a root; this may be free, like house, or bound, like pseudo-. Bound morphemes like -ment or -s which do not have content meaning are called affixes. There are two types of affix: the derivational affix and the inflectional affix.
The derivational affix creates new words, often belonging to a different word class (we return to word classes below). In English, affixes that change word class are suffixes, which means that they attach to the end of words. For example, the verb employ plus the suffix -ment becomes a noun employment. The noun nation plus the suffix -al becomes the adjective national. Suffixes can be stacked; consider the noun nation-al-is-ation, for example. English also has some prefixes that do not affect word class, but do affect the meaning of the word (for example, de-nationalise, or un-do). These also fall within the category of derivational affixes.
The inflectional affix, which is also a suffix in English, does not change the category of the word, nor does it affect the content meaning. Instead, it marks a subclass of that word. Another way of saying this is that it marks a different grammatical form of the same lexical item. Some English inflectional morphemes are illustrated in Table 14.1. Some of the grammatical terms in the left-hand column will make more sense by the end of this section.
Of course, this brief discussion of morphology rests upon the assumption that we have a clear notion of what it means to describe something as a word. However, there are a number of different ways of defining this term (Trask 2004). We are used to thinking of a word in terms of an orthographic word: something that is written as a single unit. However, this does not necessarily tell us anything about spoken language, which is of primary interest to linguists. Orthographic systems are man-made and vary enormously, sometimes revealing little about the structure of the language they represent. A phono logical word is a unit of pronunciation, defined according to the phonologic al rules of that language. In English, a phonological word usually contains one main stress. In rapid speech, some parts of an utterance are ‘glued together’ into single phonological words, which do not correspond to our idea of where the word boundaries lie from a meaning or grammar perspective. Trask (2004) provides the following example, where the bracketed units in (4b) correspond to phonological words:
As this example shows, the boundaries laid down by the system of pronunciation do not always correspond with the boundaries laid down by meaning or grammar. While the phonological word reveals much about the phonological structure of a language, it is less useful in the study of grammar.
A third definition of ‘word’ is lexical item, a term that we have relied upon throughout earlier parts of the book. This term means a unit of our mental ‘dictionary’ (or encyclopaedia), and this is the sense in which linguists use the term. A lexical item has a more or less identifiable meaning (like cat) or function (like this). However, recalling the discussion of inflectional morphology above, each lexical item may have a number of grammatical word forms. Nouns like cat, for example, have both a singular and a plural form (cat–cats), and verbs like go have a whole list of forms (go, goes, went, going, gone). The list for be is even longer (be, am, are, is, was, were, being, been). Adjectives like big also have a number of forms (big, bigger, biggest). We can think of each lexical item, then, as a bundle of forms, although some lexical items, like my, have only one form in English.
