Unit 3: Basic Morphology

(Comments on Activities)

Activity I

Since this activity does not specify any method for dividing words up, a number of possible methods could easily be devised. Having read on into the Unit though, you should be able to come back to check your response to this, and divide the words into morphemes. The suggested morphemic division is as follows:

anti+social 2 morphemes

eat+s 2 morphemes

fly+ing 2 morphemes

ostrich+es 2 morphemes

pre+dict+able 3 morphemes (but raises some issues about the nature of a morpheme)

pre+cook+ed 3 morphemes (compare this to predictable - they are not the same)

uncouth 1 or 2? (no morpheme "couth" in English).

Remarks

If the basic definition of a morpheme is something along the lines of "minimal unit of meaning", this gives us a reasonably clear morphemic division for the first four items. Ostriches provides an example of a spelling rule devised to assist with the representation of the pronunciation of the plural morpheme. We don't write *ostrichs, but we could. Don't confuse spelling with morphology.

With predictable we have a slight problem. Pre- is reasonably classifiable as a productive prefix meaning something like "before", although you might need to know Latin to be aware of this. (Similar remarks would apply to anti- in antisocial.) Unfortunately, there is no word "dict" in English. However, we could still argue that it is a morpheme, as it is an element occurring in several other contexts, e.g. contradict, diction, dictation, edict etc., in which we know that the meaning derives from the Latin word dicere - to say. The perception of morphemic status would then depend on our awareness of the etymology of English, and that could be argued to be questionable from a descriptive point of view. If people have no knowledge of Latin, and no conscious or subconscious awareness of the common thread of meaning carried by an element such as "dict", can we really say that "dict" acts as a morpheme for them? A further point is that words like this often came fully-formed into English from Latin, or via mediaeval French. In other words, the actual formation process had already taken place in a different language. A more recent example would be a word like cigarette. We could argue that this consists of "cigar" + "ette", in which the suffix is a morpheme meaning "small". But this is not really an English word-formation process (except in a few nonce examples and sometimes facetiously), and cigarette came whole into English. For these reasons, it seems defensible to treat words like predict and cigarette as one morpheme. This is a knotty issue in morphology. In the end we probably need to distinguish between morphemes based on etymology and morphemes based on speaker awareness of meaning, even if that "awareness" may in fact be subconscious for much of the time. (see also discussion of disruption in Activity 3).

With precooked, pre- is a prefix with a much more recognisable meaning than it has in predict. This is further emphasised by the different stress patterns of the two words. The "beforeness" of pre- in precooked is much nearer the surface of most speakers' awareness.

Uncouth presents a more serious problem. Here, although un- can dealt with in the same way as anti-and pre-, there is no meaningful item "couth". What this tends to show is that the processes of word formation sometimes ride roughshod over what the general system of word formation suggests is possible. This item is unusual. The fact that we cannot really say whether it has two morphemes or one does not invalidate the principle of morphological division of words.

Activity 2

All these words obviously end in -ic, which is still a productive method of forming adjectives in English. If we add-ic to a noun, the resulting adjective means something like "having the characteristics or properties of the noun". An example would be something like sulphuric (in passing, note the difference in meaning between sulphuric and sulphurous). The problem is that there are many words in English with this -ic ending which have come down to us in that form from Latin or Greek. In ancient Greek, -___s_was one of the commonest adjectival endings. This was then carried to Latin in words such as classicus. Do we want to go that far back in order to justify the treatment of -ic in modern English as a morpheme? This seems unnecessary and unwise, since we would never know where to stop. What we could do instead is to argue that the use of the -ic ending (or rather its -___s predecessor) ceased to be a productive item, or "formative", even by the time of classical Latin. So words like economic came into English fully formed, either from Latin, or perhaps from mediaeval French, which in turn came from Latin. This would make economic one morpheme and uneconomic two. Our judgement about this is being made on the basis of our knowledge of the history of the word. In this case it probably fits quite well with most people's perceptions as well.

Words in this list which derive from classical words, and came fully-formed into English from Greek or Latin or French, are cynic (dog-like), frantic (cognate with frenetic), panic (relating to the god Pan) and topic (relating to a place). These we can say are all one morpheme for the same reason that economic is one morpheme.

Basic is different, since there is no recorded use of it until the nineteenth century, and it therefore belongs to a different group of -ic words, i.e. those which are formed by suffixation to an existing English word, base. This happened as a result of the need for a scientific vocabulary. The trend probably started in the eighteenth century, and was well under way by the mid-nineteenth century. So basic is two morphemes, base + -ic. Periodic, nationalistic and gnomic can be included in this group, periodic for the same reason as basic, and nationalistic and gnomic because the trend has now spread beyond the realms of science into more general abstract vocabulary. In other words, an ancient Greek word-formation process which more or less died out has been revived in English (and doubtless other European languages) in the last couple of hundred years, initially in the domain of science, but now more widely. Notice that in nationalistic, and perhaps also gnomic, the -ic ending does not simply mean "having the characteristics of". There seems to be an attitudinal element as well - probably disapproving. This is a fairly recent development.

In preparing these notes, reference was made to the OED entries for the words in question, and also to the entry for the suffix -ic. This is something anybody can do. The point is not to pursue the etymology of words exhaustively, but to understand the relationship between etymology and judgements about the morphological status of words in modern English.

Activity 3

The instruction to this exercise was a little loose. The Unit should have distinguished more clearly between free and bound morphemes first. Free morphemes are lexical morphemes, e.g. crack,, which are neither inflexional nor derivational. Bound morphemes are generally inflexions and derivational affixes (but see discussion of disruption below). The decision about whether a morpheme is inflexional or derivational applies then to bound morphemes. The table below has a column for free morphemes as well. Some words are dealt with separately below the table.

free morpheme	inflexional morpheme	derivational morpheme
crack	-ed
crack		-er
linger
sing		-er
harbinger
immense
modest		im-
harness
beat	?0
soluble		in- + dis- + (soluble) + -ility

beat

See Unit for discussion of this type of word. It may be first person singular, or it may be past tense, or it may be infinitive. Out of context it is hard to say how many morphemes it is.

disruption

This word does not seem to have a free morpheme. The "rupt" element comes from Latin, and occurs in words such as corrupt, interrupt, rupture etc. The case is similar to prediction (Activity 1), where we said that "dict" could perhaps be regarded as a morpheme on historical grounds. The same goes for "rupt", but there is also an argument for saying that neither "dict" nor "rupt" are meaningful to most people. Certainly it seems unlikely that any new words would be formed by combining these elements with others. Furthermore, as already pointed out, these words came whole into English. We could therefore argue strictly that a word must contain at least one free morpheme. If that is so, it quickly deals with problem cases like uncouth, predict and disruption. Since none of the potential component morphemes of these words are free, it follows that each must in fact be one morpheme in its own right. This line of reasoning offends some morphologists who like to focus on the Latin or Greek or other origins of words, but it has a simplicity about it which works quite well for modern English.

were

This must be the past tense of be. So there are at least two morphemes here, the lexical morpheme be, and the grammatical morpheme "past". There would also often be a plural element involved here, which would represent a further grammatical morpheme. The whole lot is rolled up into one word. This aspect of morphology is really hard.

Activity 4

The apostrophe is just a spelling convention, borrowed from French as recently as the 17th century. The -s represents the Germanic genitive case of the noun, the only case left in English apart from the base form of the noun, sometimes known as the common case. The genitive case signifies possession (John's car), but also more general attribution (Pilgrim's Progress), duration etc. {a year's absence). Possession is not so easily attributed to inanimate entities, so that examples such as the house's walls are rare. Case endings are traditionally thought of as inflexions of the noun. But in English, the genitive is actually somewhat more like an agglutinative particle, since it does not necessarily appear on the head noun in a noun phrase. In The King of Spain's beard, the king possesses the beard, but the marker of possession appears elsewhere, on Spain.

car's

Here the apostrophe represents a spoken contraction of has, which is then added to car. So we are not dealing with the formation of a new word, but a variant representation of two words. The 's is a morpheme signifying perfectivity,

broken

Inflexion -the past participle variant of break.

did

Inflexion - the past tense of do.

outside

This appears to be two morphemes, out and side, joined together. Neither out nor side is inflected here. We could see this as a case of compounding, similar to the word formation processes of an isolating language like Chinese, and relatively common in English as well (see carpark below). On the other hand, the meaning of outside seems not entirely derivable from our usual understanding of out and side, which might lead us to treat it as one morpheme. We can compare it with beside, which we are much more likely to see as one morpheme, since the meaning of be- is unavailable to us. The meaning of a compound word is more or less opaque with respect to its apparent constituent morphemes, and our sense of its relative opacity or transparency will depend greatly on the history of the word, and our awareness of its history and etymology. Words such as outside, inside and beside could just as easily be seen as having been formed by a process of agglutination, with side as the head and the prefixes added.

carpark

A more straightforward case of compounding, with transparent meaning of the constituent morphemes. Typical of isolating languages.

anything

Similar to outside and carpark. But the words any+thing could of course be a noun phrase having a determiner +head structure. There is a more structural basis for this compounding in other words.

immobiliser

Two morphemes here, "immobilise" (free) + "-er" (bound derivational).

couldn't

There seem to be three here:

a. Lexical element, which we can label CAN.

b. Past tense.

c. Negativity.

Activity 5

	FINITE FORMS	NON-FINITE FORMS
move	move, moves, moved	move, moving, moved
leave	leave, leaves, left	leave, leaving, left
drink	drink, drinks, drank	drink, drinking, drunk (drunken - see below)
shrink	shrink, shrinks, shrank, ?shrunk (see below)	shrink, shrinking, shrunk (shrunken - see below)
lie (to fib) lie (to recline)	lie, lies, lied lie, lies, lay (past tense!)	lie, lying, lied lie, lying, lain
lay (to place, deposit etc.)	lay, lays, laid	lay, laying, laid
beat	beat (present and past tenses), beats	beat, beating, beaten
cleave	cleave, cleaves, cleaved, ?clove (see below)	cleave, cleaving, cloven, cleft (see below)
go	go, goes, went	go, going, gone
have	have, has, had	have, having, had
be	am, are, is, was, were	be, being, been

The base form in the finite column is the present tense, all persons except third. The base+s form is the present tense, third person. Then there is the past tense. Be is obviously exceptional (see below). In the non-finite column, the base form is the infinitive. Then there are the present and past participles.

This table does not give any archaic familiar forms such as (thou) goest, hast or art.

Some further comments about individual items:

drunken and shrunken

These are not really parts of the verb in modem English, but adjectives derived from an archaic past participle (see Unit 10, Basic Lexis).

shrunk

Many verbs in English, whether they came originally from French, Latin or German, have had a chequered history as regards their past and past participle forms. Shrunk has certainly been used as the past tense in earlier centuries, even though many grammars will list shrank as the "correct" past tense nowadays. There may be a difference depending on whether the word is being used literally ("My jumper shrunk in the wash") or figuratively ("She shrank from his evil intentions"). In the film Honey, I Shrunk The Kids, it is a moot point whether the word is a "real" past tense, i.e. chosen as a past tense by someone who would not count himself as a dialect speaker, or a dialectal past tense, chosen by someone who uses it as part of a regional variant, and who might, if pressed, say that actually shrunk is the past participle. This latter phenomenon would be very common with the word done, for example. If a football manager says: "The lad done great today", this isn't because he doesn't know the word did. It is simply part of his variant speech in that context.

lie and lay

These words are increasingly hard to distinguish. It is common to hear people in all walks of life using lay in the sense of lie. John Major has been heard to mock Tony Blair in Parliament by saying: "I don't suppose the honourable gentleman lays awake in bed worrying about it." Here are some sample sentences:

lie = fib

Governments always lie.

She lied to me about her past.

He has lied to every teacher in the school.

lie = recline

Edinburgh lies to the west of London.

They lay on the grass and watched the cricket.

It had lain there since Roman times.

lay (standard)

Then we lay the bricks, one by one, like this.

They laid him in a shallow trench.

It hasn't laid any eggs for weeks.

lay (non-standard)

I think I'll just lay down for a while.

They laid in until about ten.

beat

The only verb in English in which the present and past tenses have the same form while the past participle has a different form.

cleave

This is generally an archaic word, and the non-finite forms tend to appear only in fixed phrases such as "cleft in two". Cloven is probably much like drunken, and almost entirely adjectival - cloven hoofs etc.

be

This is the only verb which more or less retains different forms for each person, as inflecting languages tend to.

Activity 6

This is further practice along the same lines as Activity 5. Use the verb chart and comments above to check your answers.