Let’s have a look at the morphemes in the following words: extended, discussion, suggests, characterisation, antithesis. We should have no problems separating out the morphemes:
(a) extended consists of extend and the past tense morpheme -ed;
(b) discussion consists of discuss and the morpheme -ion that results in a noun being formed;
(c) suggests consists of suggest and the morpheme -s, used when the subject is singular (but not I) and the verb is in the present tense;
(d) characterisation consists of characterise and the morpheme -ation which works a bit like -ion mentioned above; we can split of characterise further into character and -ise (also spelt -ize) that results in a verb being formed;
(e) antithesis consists of the prefix anti- and thesis.
We can therefore talk about morphemes that are bound (ie they cannot stand alone, like -ion) and those that a free (ie they can stand alone, like character). The bound morphemes are often those can be attached to the beginnings and endings of other morphemes, and these are affixes – those attached to the beginnings of other morphemes are prefixes and those attached to the endings are suffixes. (Occasionally an affix might be attached to the middle of words, like abso-bloody-lutely: this would be an infix.)
Affixes are either derivational or inflectional (also spelt: inflexional). A derivational affixes changes either the meaning or the word class of the original morpheme/word (or both). Thus, -ion, -ation and -ise are derivational morphemes because they change the word class of the words to which they were suffixed (discuss, a verb, becomes discussion, a noun; characterise, a verb, becomes characterisation, a noun; character, a noun, becomes characterise, a verb); anti- is also a derivational morpheme because of the change in meaning (antithesis is different in meaning from thesis). Character, discuss and thesis in the above examples are roots. (The root is the base form of a word which cannot be further analysed without loss of the word’s identity.)
If the word class or the essential meaning is not changed, then the affix is inflectional. They can also be called inflections (inflexions) on its own. In English today, we have inflexions for tense (-ed) and aspect (-ing), for person (-(e)s), for plural (-(e)s), for the possessive (’s) and for comparisons (-er, -est). The element of word structure without any inflections is known as the stem. Therefore, the word characterising is made up of characterise (the stem) + ing (inflectional affix). And from the earlier paragraph, we know characterise can be broken up into character (the root) + ise (derivational affix).
Word classes (‘Parts of speech’)
I have already used labels like nouns and verbs. I will give some quick working definitions here for the various word classes in English. I will start with the traditional definitions because they tend to be most accessible, but because these are usually based on meaning, there will be problems if we use these as the sole criterion for classifying words. This is because words function within structures (and this is what grammar is about!) and we should be able to describe them in structural terms too.
A verb is traditionally a ‘doing’ or ‘action’ word, like sit, throw, accuse or discuss. (Formally, we can identify verbs as words that display tense contrast (eg sit v. sat – you can’t do the same thing with seat, for example) and other things like aspect, voice, mood, person and number. Functionally, the verb can act as a minimal predicate.)
A noun is traditionally defined as the ‘name of a person, place or thing’, like chair, house, apple or sky. A reasonable good test to see if a word is a noun is to put a determiner (the, a(n), my, your, etc.) before it (my seat, the constituency, a calamity).
An adjective is traditionally a ‘describing’ word, like naughty, silly, regretful or antagonistic. Note that these words describe persons, places or things, not actions. Functionally, we can say that adjectives are words whose main function is to modify a noun.
An adverb is a word that describes an action of a verb, like quickly or carelessly. The derivational morpheme -ly is often used to convert adjectives to adverbs (naughty à naughtily; regretful à regretfully).
A preposition is an item that precedes a noun phrase, and often indicates meanings to do with location in physical or abstract time and space – like by, to, at, in and onto.
A conjunction is a type of word whose chief function is to connect morphemes (inter- and intranational) words (a short and fat boy), phrases (a short boy and a fat boy) or clauses (I shouted at the fat boy and he stared at me).
One element that now stands out if we consider English orthography is that the system is not fully phonological. We have already noted non-English influences in the etymological principle of spelling. Here is another principle of English spelling.
Consider the past-tense inflection. (We will leave aside the question of strong verbs, ie verbs that can change the vowel sound like ring v. rang, for the moment.) In most cases we can add the -ed inflection despite the fact that the inflection will be pronounced differently depending on the verb to which it is attached, eg drenched, dented and dared where the inflection is pronounced /t/, /Id/ and /d/ respectively. (And of course many speakers also use /@d/ instead of /Id/.) The spelling remains more or less constant while the pronunciations change.
Consider also the common morphemes in the following words:
(a) magic, magician
(b) photograph, photography, photographic
(c) nation, national
The spellings in (a) make the relationship between magic and magician clear although this would be easily lost in the pronunciations /"madZIk/ and /m@"dZISn/. In the same way, the morpheme photo in photograph and photographic are pronounced the same way /"foUtoU/, but rather differently in photography /f@"tQ/. It is the spelling that preserves the link. We see the same link preserved in the spellings in (c), which are not reflected in the pronunciations /"neISn/ and /"naSnal/.
(It doesn’t matter if you are not totally familiar with the phonetic symbols at this stage. For them to display properly, you need to have installed the Times Roman Phonetics font. If you haven’t done so already, click here.)
We can therefore argue, from these examples, that English spelling reflects English morphology, over against English phonology. (Although we must note that there are exceptions such as pronounce v. pronunciation; maintain v. maintenance; leaf v. leaves.)
Clearly morphology plays an important role in word building, and we can be fairly creative in combining some morphemes and yet never risk being misunderstood. The suffix -able or -ible can be attached to many verbs, like recordable, generalisable or contactable. Other affixes that have a lot of potential for being used more widely are pre-, non- and ex-.
We will have more to say about word building when we consider the nature of English vocabulary.
There is a branch of linguistics that attempts to classify all human languages, regardless of their history or their genetics, based on their structural properties, in particular whether they use inflections or not. This branch of linguistics is known as the typology of language or typological linguistics. (We will have opportunity to consider the genetic classification in Week 6.) Languages can be, broadly speaking:
(a) analytic or isolating: meaning that the words in the language are invariable (eg Chinese and Vietnamese);
(b) synthetic or fusional: meaning that words typically contain more than one morpheme and there is a wide range of inflections (eg Latin, Greek and Arabic).
Other terms that might be used are agglutinative and polysynthetic, which we will not focus on. We can see analytic and synthetic as two general tendencies that are gradable.
Any Chinese speaker learning, say, German today will be struck by the range of forms of a particular word. This difficulty can be expressed in terms of the need to shift from an analytic language to a more synthetic language.
Let us now think about the inflections that were available in an earlier form of English, known as Old English (OE, also called Anglo-Saxon) and present-day English (PDE).
If we consider the number of forms of the king there are in English today, we’d have to say that there are two, possibly four if we include the possessive, of each:
the king the
kings the king’s the kings’
In OE, here are the forms of the king:
(a) se cyning – singular, nominative
(b) þone cyning – singular, accusative
(c) þæs cyninges – singular, genitive
(d) þæm cyninge – singular, dative
(e) þy cyninge – singular, instrumental
(f) þa cyningas – plural, nominative and accusative
(g) þara cyninga – plural, genitive
(h) þæm cyningum – plural, dative
We can see at once that the range of inflections is much higher in OE, with Ø (this means zero inflection), -es, -e, -as, -a, -um. These are to do not only with number (singular v. plural) but also with case. The labels nominative, accusative, etc. are case labels. Case indicates the relationship between the noun and the verb and the other nouns. Here’s roughly what they indicate and italicised examples in a clause from today’s English.
nominative: subject of a clause, typically the ‘doer’ (He chased me.)
accusative: direct object, typically the ‘patient’ or ‘affected’ (He chased me.)
genitive: possessor or source (He chased my dog.)
dative: indirect object, typically the ‘recipient’ (He gave me a dirty look.)
instrumental: means, instrument (He tripped me with a stick.)
We also see a wide range of forms of the: se, þone, þæs, þæm, þy, þa, þara. But these are not the only ones: others include þæt, seo, þon and þære. What were these extra forms of the for? The answer is that it depends not only on number and case but also on gender. Nouns in OE were assigned to three classes of gender: masculine, feminine and neuter. This was largely arbitrary but it had consequences, as seen earlier, in the form of words used. Adjectives were also inflected for number, case, gender and additionally where a form of the was used. Here are some examples involving dola (= foolish), cyning (= king, masculine), bearn (= child, cf. Scottish bairn, neuter) and ides (= woman, feminine). We will only look at the nominative case.
(i) se dola cyning
(j) þæt dole bearn
(k) seo dole ides
(l) þa dolan cyningas
(m) þa dolan bearn
(n) þa dolan idesa
(o) dol cyning
(p) dol bearn
(q) dolu ides
(r) dole cyningas
(s) dolu bearn
(t) dolu idesa
So, apart from the base form dola, OE speakers also needed to contend with dole, dolan, dol and dolu. Let us now summarise how the OE system is different from PDE based on the system of inflections.
(1) OE nouns were inflected for number (hund-hundas, deor-deor, oxa-oxan, fot-fet, cyning-cyningas, ides-idese, scip-scipu). Nouns are still inflected for number today (dog-dogs, deer-deer, ox-oxen, foot-feet, king-kings, ship-ships). OE seems to have a much wider range of ways of forming plurals, including -as, Ø, -en, -e, -u and vowel mutation. The PDE system is more regular, with the majority taking on -es or -s. We still have some irregular plurals which we inherited from OE. Some other irregular plurals are due to loan words retaining the foreign plurals (eg criterion-criteria from Latin, and analysis-analyses from Greek).
(2) OE nouns had grammatical gender (which determined the forms of the adjectives and determiners used with them). Where gender exists in PDE, this is natural gender.
(3) OE nouns were inflected for case. This is no longer the true of PDE. We are, however, left with some vestiges of this system in our pronominal system today:
|
|
nominative, singular |
accusative, singular |
genitive, singular |
nominative, plural |
accusative, plural |
genitive, plural |
|
|
1st person |
I |
me |
my, mine |
we |
us |
our, ours |
|
|
2nd pers. |
you |
you |
your, yours |
you |
you |
your, yours |
|
|
3rd pers. |
masc. |
he |
him |
his |
they |
them |
their, theirs |
|
fem. |
she |
her |
her, hers |
||||
|
neuter |
it |
it |
its |
||||
(4) Adjectives were also inflected in OE. This is no longer the case in PDE.
If we therefore contrast OE to PDE, it seems clear that although PDE is an inflected language, the range of inflexions has been reduced, and the grammatical categories of case and gender do not exist for most ordinary nouns any more (if we discount the case of the genitive or possessive).
In fact, if we examine some English-based pidgins and creoles, we can see how the evolution is taken a step further. (For our purposes, a pidgin is a simplified language arising out of the need of different communities who do not share a language coming together for a restrictive purpose such as trade. Creoles are pidgins that have become the mother tongue of a community.) In pidgins and creoles, the plural inflection is dropped (two book, dem creature). When plural meaning has to be indicated, particles may be used (The rabbit dem eat it all). The possessive is also not employed (dat man house) or a particle might be added (De coat a fi me = ‘that coat is mine). Case distinctions for pronouns are also not employed (She see he come, take he coat, and go; Carry dat book to she teacher).
The evolution seems to have been from a synthetic language (OE) to a more analytic language (PDE, especially English-based pidgins and creoles). We shall explore the whys and wherefores elsewhere in the module.