Unit 10 - basic lexis

The purpose of this unit is to introduce some basic concepts to do with word meaning.

Definitions

When you don't know the meaning of a word, your instinct is probably to look it up in a dictionary. This works well enough for most practical purposes, but we should not treat dictionary definitions with too much reverence. They are only approximations, and can never hope to capture everything about the meaning of a word. Sometimes they are very contentious. For example, some dictionaries define Yorkshire as a "former county" of England, apparently on the grounds that it is not an administrative unit. The citizens of that region are said to be upset by this. Yet the same fate of being defined as a "former county" also befell Middlesex, and it seems that few citizens of that region are upset. The reasons for this difference must be very complex, but we can assume that the meaning of Yorkshire has something about it which former appears to deny, whereas Middlesex is more solidly former.

Activity 1

Think about what elements of the meaning of the word Yorkshire contribute to rejection of the label former? How does Middlesex differ? Consider how you would explain this difference for a few moments then click on the link below for our thoughts.
click for our comments

The difference, of course, relates to the social values which attaches to a word such as Yorkshire but not to words such as Middlesex. For many people, the term Yorkshire is still very much functional in specifying both a region and a regional identity. To suggest that there is something 'former' about Yorkshire, is thus to somehow suggest that there is something 'former' about the Yorkshire identity. Few such associations attach to the term Middlesex - it is not nearly so functional in specifying a region and an identity.

Whatever core meaning we think a word has, the actual meaning of a word in use is likely to be more complex than, or even quite different from, the core meaning we thought it has. This is particularly true of common words. Indeed, the more common a word is, the more likely it is that it will have more than one meaning.

Activity 2

Consider these dictionary entries for girl and boy but do not look the words up in any other dictionaries for the time being.

boy n. 1. a male child; youth. 2. a man regarded as immature or inexperienced: he's just a boy when it comes to dealing with women. 3. See old boy. 4. the boys. Informal. a group of men, esp. a group of friends. 5. (esp. in former colonial territories) a native servant. 6. short for boyfriend. 7. boys will be boys. youthful indiscretion or exuberance must be expected and tolerated. 8. jobs for the boys. Informal. appointment of one's supporters to posts, without reference to their qualifications or ability. - interj. 9. an exclamation of surprise, pleasure, contempt etc. [C13 (in the sense: male servant; C14: young male): of uncertain origin; perhaps related to Old English Boia, man's name; from Old High German Buobo, man's name.

girl n. 1. a female child from birth to young womanhood. 2. a young unmarried woman; lass; maid. 3. Informal. a sweetheart or girlfriend. 4. Informal. a woman of any age. 5. an informal word for daughter. 6. a female employee, esp. a female servant. 7. (usually pl. and preceded by the) Informal. a group of women, esp. acquaintances. [C13: of uncertain origin; perhaps related to Low German Gšre boy, girl]

Consider the following questions, perhaps jotting down some thoughts in the spaces provided. Then click for our thoughts on the matter.

1. Are there any aspects of either definition which you disagree with or object to?

Our thoughts

Disagreements and objections.
It's important to be aware that in linguistically well-informed discussions of such topics, we should focus on the question of whether the definition is linguistically accurate, whether it accurately reflects the way the word is used in real life, rather than being too much influenced by personal beliefs about how a particular word SHOULD be used, or how you would PREFER it to be used. If you disagree or object to some aspect of a definition, you must be careful to distinguish between objecting to the truth of what is being asserted about the use of a word, and objecting to the fact that the word is used in that way.

2. Are there differences between the use of boy and girl which these definitions do not bring out?

Our thoughts

This is an open-ended question. We might, for example, argue that the definition could have made more out of the interesting fact that the term girl and girls is somewhat more readily applied to women of all ages than is the case with boy and men of all ages. This fact seems to reflect certain gender biases in society. The definition might also have mentioned the pejorative and prejudicial use of boy to refer to African American males in some circles. No doubt there are other points which may have occurred to you.

and finally:

3. Write definitions for the words lad and lass which take account of the likely context of use of each word.
After doing this, compare your version with several dictionaries.

Our comments

The definition should take some note that the two terms have (a) an association with informal styles - in many dialects the words would only used be in casual, friendly or even intimate contexts, and (b) certain regional dialectical associations - the term lass, for example, has association with Scottish English.

In using a word we are doing more than simply referring to an entity or concept in the real world. We are also exercising a choice, determined by a range of social and linguistic factors. Boy and lad, girl and lass: each pair seems to be doing much the same thing, i.e. pointing to a young male, or female, human being. Yet the meanings conveyed by each term in these two pairs go well beyond this simple act of reference.

It is conventional to distinguish reference from sense. Reference is the relationship between the word and the object or concept in the world. Sense is the relationship between the word and other words. That is, sense has more to do with our choice of one word rather than another and the way we combine that word with others we have chosen. In this way of looking at meaning, boy means not girl (also not man, not dog etc.), and girl means not boy, not woman etc. Notice that lad also means not boy, not youth, not juvenile, not young male human, not knave, not page etc., i.e. not any of these lexical choices, as opposed to not that object in the world. With Yorkshire, it is clear that people are not necessarily referring simply to an administrative unit (i.e. the unit which is Yorkshire not Lancashire), but to a whole complex of meanings which are 'Yorkshire' not 'Lancashire'.

It is commonly said that words mean what they mean because of how they are used. Reference is only one aspect of this use. Repeated use of a word to refer to an entity will indeed confirm its referential meaning, although there are many words to which it is hard or impossible to give any referential meaning, e.g. the, however. Repeated use of a word in certain social or linguistic contexts will help to confirm its sense. The meaning of a word, therefore, arises out of its use as a referring item and its context of use. 'Context' here is a broad term, which needs dividing into several categories.

Social Context

This refers to variables such as region, class, age, gender etc. Many words are neutral with regard to social context: that is, they are used by everyone, and mean the same thing no matter who uses them, where or when. Words like telephone, paper-clip, dictionary seem to be of this type. A word such as tummy is clearly different. You would not find it used in medical text-books, for example. If it did appear in one, it would surely be as part of a cited dialogue between, say, a doctor and a child. If a doctor used the word tummy in a discussion with you about your own health, you might find it patronising. An element of meaning would be attached to it by using it outside the social boundaries you expect for that word.

Generic context

Activity 3

The following words are clearly marked or restricted by social or generic context in one way or another. That is, they are much more likely to be used by certain speakers than others, at certain times and in certain places etc. In the spaces provided, briefly note down the likely social or generic context of use for each, indicating, for example, whether the word is associated with a particular subject area or specialisation, with a particular region or dialect, with a certain level of informality/formality, and so on. Then check the feedback column for comments.

wee (as in small)
feedback	Marked for region - Scots, but more widely known.
pikelet
feedback	Marked for region, not widely known outside the Midlands.
platelet
feedback	Generically restricted to the field of medicine. Most people go through life without knowing about platelets, but of course non-specialists who happen to suffer from blood disorders might well be admitted to the 'platelet-using' speech community, if only temporarily.
wee (as in urine)
feedback	Marked for age, and informality. Usually used by and to children (under about 5?). Sometimes used informally by adults to each other, but probably not by children aged 5 to 15? More euphemistic than pee, a semi-taboo item.
bathroom
feedback	As a euphemism, marked for region - US English - now invading UK?
cool
feedback	Marked for youth. Originally US? Musical connection suggests some generic restriction as well.
wotcha
feedback	Marked for informality. Also class and region? British
gules
feedback	Generically restricted to the language of heraldry.

Linguistic context

This refers to a range of restrictions arising from the immediate linguistic environment. Grammatical restrictions are one kind, but of more immediate interest are semantic and other less easily defined restrictions. Some linguistics books attempt to distinguish grammatical from semantic restrictions by means of different deviant sentences, usually invented for the purpose. For example, in the sentence Sincerity admired the girl, it has been suggested that we have a violation of a semantic restriction, because sincerity is a non-human entity which is not semantically capable of being the subject of a verb like admire. From a strictly grammatical point of view, it has also been argued, there is nothing wrong with the sentence; it does not break any grammatical rules. By contrast, the sentence The girl admired sincerity, does break a grammatical rule and is thus a violation of a grammatical restriction. This distinction is well-embedded in the linguistic literature, but its usefulness is also a matter of dispute.

Now consider the unidiomatic sentence The snow was absolutely bucketing down. This does not appear to break any grammatical rules. So is the violation a semantic one? There seems to be no real reason why snow should not fall by the bucketful. Is it perhaps only liquids that can bucket down? This seems a little doubtful, since even to say 'the stream was bucketing down' or 'water was bucketing down from the overflow pipe' sound odd. In the end, it seems that only rain 'buckets down', and it can probably only do this directly from the sky. (In passing, notice how unusual the use of buckets down is in the last sentence - have you ever seen this verb used in the present tense like this before?) What we appear to be saying is that it is part of the meaning of bucket down that the subject must be rain. There is quite a jump from saying that the subject of admire must be human to saying that the subject of bucket down must be the actual word rain. Moreover, while we certainly find it difficult to conceive of sincerity admiring anything, we do not find it difficult to conceive of snow in buckets. So to call our snow sentence a semantic violation seems, at the very least, to be introducing a different slant to the term semantic. Recent computer-aided studies of lexis have made several interesting discoveries about the more precise restrictions affecting words.

Activity 4

What linguistic restrictions apply to the use of each of the following adjectives? Consider, for example, restrictions on positions the word can occupy in the sentence, restrictions on the range of words the term in question can modify or describe, and so on. Write your comments in the spaces provided, then click on 'feedback' for ours.

NB:Speakers who are asked to state restrictions on the use of words do not usually have immediate access to large numbers of examples. They obviously have to rely on their intuition about the language and may not find it easy to articulate their response. Some grammatical restrictions are pretty obvious. Other restrictions are much less obvious, but can be revealed by looking at a large number of examples. Note that the existence of a very dominant pattern of use for a word does not necessarily exclude other possibilities. The comments in this Unit, and throughout the course, frequently make reference to a large computer-held corpus of examples. The importance of this will be commented on in detail later.

drunk
feedback	a predicative adjective, i.e. it is used after the verb to be (and a few similar verbs), but not in front of a noun. We say: He is often drunk, or I got drunk last night, but not: *He is a drunk lout. Some very rare exceptions do occur, e.g. drunk drivers.
drunken
feedback	an attributive adjective, i.e. it is used in front of a noun, e.g. He is a drunken lout, but not on its own after the verb to be: *He is often drunken. Virtually no exceptions. Most adjectives can be used either way. (These are both clearly grammatical restrictions.)
blithering
feedback	attributive (The idiot was blithering) but also highly restricted to idiot, and one or two other nouns such as fool*..
rancid
feedback	normally restricted to a set of nouns which refer to fatty substances, e.g. oil and butter. Out of 152 corpus examples, more than 140 show the word used in this way. Amongst these there are a few cases of rancid liquids, e.g. milk, sweat, and a few examples of the meaning being transferred to another item, but where a fairly concrete connection with the usual use is still traceable, e.g.: a rancid odour, a rancid old bed. Metaphorical extension is also possible, e.g. the rancid rhetoric of class warfare, rancid right-wing views. It is odd that both these examples are from the discourse of politics. The restriction here appears to be semantic. Part of the meaning of rancid is its restriction to fatty substances. If a metaphorical extension is involved, that element of meaning is transferred in some way.
torrential
feedback	almost exclusively restricted to rain, which is also the immediately following word in the great majority of cases. There are a few related words such as rainfall, downpour, storm, stream etc. Out of 404 examples there are three metaphorical uses: torrential speeches, torrential inventiveness, torrential emotion. Is this a similar case to rancid? Do speakers feel that the meaning of torrential is dependent upon rain in the same way that the meaning of rancid is dependent on fatty substances? Is part of the meaning of torrential the fact that it is only used of rain? This seems not very likely. The related noun torrent has wider reference, more usually referring to flowing water, e.g. streams etc., and it also has well-used metaphorical meaning, e.g. torrents of abuse. So, although torrential is in fact almost exclusively used with rain, speakers' knowledge of the related word torrent probably makes the meaning of torrential less dependent on rain. Compare this with rancid, which has no related noun.
taciturn
feedback	seems to be applied almost exclusively to men. A wide range of words referring to men can occur, e.g. chap, individual. Some more general words such as family, people, etc. are also possible. Out of 130 examples there is one clear example of a female referent: Lester Piggot's wife is described as taciturn by the journalist Jean Rook. Why are only men taciturn? Are speakers consciously aware of this restriction?
beige
feedback	colour term with a wide range of possible nouns, but falling into some clearly defined groups, e.g. vehicles, items of clothing, cosmetics, buildings and parts of buildings. Out of 591 examples, beige car (3), plus various examples such as beige Toyota, beige carpet (5) beige walls (5) beige sweater (7) and so on. Not a single instance of beige hair. Why not? Is beige hair impossible, or is it just an accident of the corpus? If it is impossible, why is it?
lovely
feedback	keep reading....

Words and Social Reality

The distinction between social context, generic context and linguistic context is not absolute. Taciturn is hardly ever used of women. Is this merely a curious linguistic fact, similar to the fact that beige is never used of hair? Or is there an underlying element of meaning in taciturn which relates to the world, or the way we see the world? It is clearly not the case that only men are habitually silent. It may be the case that habitual silence in a woman is perceived as less remarkable, and is thus less in need of vocabulary to refer to it. If true, this is a social fact, a fact about people's attitudes to men and women, which is then subtly reflected in the language.

Similar remarks apply to lovely, except that the semantics of lovely are far more complicated than those of taciturn. Lovely is used of both human and non-human subjects. Let us put the non-human use on one side, and consider how the word applies to the two nouns man and woman. Your instinct might be that lovely is more likely to apply to women. If so, you would be wrong, at least according to the corpus of examples we are using, which has 80 instances of lovely man and only 35 of lovely woman. Lovely has two broad meanings when applied to people: physical appearance, and personality or character. Which is more common? Where men are concerned, it is clearly the personality which is lovely, although appearance examples do also occur. With women, appearance is the more common meaning, but not by a huge margin. Personality meanings do also occur. More intriguingly, when lovely is applied to men, the men are nearly always old, retired or dead. Lovely men are most often talked of in the past. Who it is actually calling a man lovely (is it a term used more often by women of men?) might also be significant. More could be said but space forbids. The important point is that speakers must have some quite refined semantic criteria which allow or discourage the use of lovely + man or woman. These seldom appear in dictionaries and are not necessarily amenable to conscious explanation by speakers.
Notice that a social explanation for the strange distribution of taciturn is not a comment on who uses the word. There is no evidence to hand at present that taciturn is more likely to be used by men than women or vice versa. This is different from a case like purl.

Activity 5

What restrictions, if any, can you identify for the following words? Write your comments in the spaces provided, then click on 'feedback' for comments.

pot-bellied
feedback	72 examples, of which about 24 refer to Vietnamese pigs and most of the remainder to people, all male. Notice how an individual who is described as both taciturn and pot-bellied seems almost inconceivable as female - well, almost. There are a few examples of things like pot-bellied stoves (4). Pot-bellied pig is not only frequent but unitary. Similar examples are greater spotted woodpecker or hump-hacked whale. A pot-bellied stove is a type of stove. Notice the relationship between the frequency of a combination, the frequency of its constituent parts, and its unitary status. The combination pot-bellied man is just as common as pot-bellied stove, yet we are not tempted to assign it unitary status - it does not refer to a type of man. This is partly because the noun man is much more frequent than the noun stove, and partly because the use of pot-bellied with stove is less literal.
arrant
feedback	32 examples, of which 18 are arrant nonsense. The remainder are mainly negative qualities such as hypocrisy, greed, stupidity etc. There are one or two nouns referring to people, e.g. arrant cowards, communists. These restrictions are linguistic. Semantically, the word is obviously negative; there is no *arrant wisdom, *arrant beauty etc. However, the huge preference for nonsense is less easily explicable in ordinary semantic terms. This kind of patterning is known as collocation, similar to torrential rain. Notice that arrant is also attributive. We do not say: *This nonsense is arrant. - a grammatical restriction on the word. Presumably arrant is also marked for social class, level of education etc.?
sautéed
feedback	Exclusive to food items, as one would expect, and most of the examples are from recipe books etc. Within the category of food, the matter is trickier. The restrictions on the word seem to be partly explicable in semantic (i.e. linguistic) terms. The cooking technique involved is performed on solid, edible, usually previously uncooked, items, usually (but not necessarily) cut up into small pieces, usually in butter or wine, using moderate heat and stirring, in a particular kind of pan, and so on. The corpus contains the following interesting example: He'd eat metal bolts if they were sautéed in butter. The author has conferred edibility on the inedible by using a word which is known to be exclusively used for food, and chosen a technique which suits bolts from the point of view of solidity, stirability etc. But there are items which would probably be ruled out completely. *Sautéed yoghurt (too liquid?), * sautéed icing-sugar (too powdery?), *sautéed whole haunch of venison (too large, but possible in a giant's kitchen?). In social terms, the word may be somewhat restricted to certain types of cooking (its French origin suggests that). What does one get for breakfast in a transport cafe - sautéed or fried mushrooms?
vacuum cleaner
feedback	No obvious restrictions.
almighty
feedback	The most common use of the word is as a noun referring to God, i.e. The Almighty. (about 200 examples out of 617). Easily the most common noun to go with almighty used as an adjective is also God, 177 out of 617 examples, but many of these examples are profane exclamations. Almighty is one of a small group of adjectives which can follow the noun they modify. Only a few nouns behave like this with almighty. We have Almighty God and God Almighty (but not *row almighty). Both occur in ordinary religious contexts and as exclamations. The latter is more common than the former in the corpus (Almighty God: 75. God Almighty: 102), doubtless a reflection of the relative amounts of fiction and religious text in the corpus. The placing of almighty after God is more frequent in exclamations than in religious texts. The much rarer Christ Almighty (29 examples) occurs exclusively as a profane exclamation, not in religious texts, and *Almighty Christ does not occur at all. Other regular combinations include: almighty row (26), almighty dollar (12), almighty explosion (4). There are no *almighty mountains, trees, cakes, armies, ministers etc. There is a clear predominance of a religious (i.e. social) context for almighty, as would be expected, but there is also a curious relationship between the postpositioning of the word (i.e. a linguistic marking) and its exclamatory function. Used in non-religious contexts to mean very great or powerful, the word selects from a relatively narrow range of collocating nouns. It is hard to pin down any obvious semantic explanation for this.

In a later unit we shall consider meaning again. In particular, we shall explore questions such as: Are there words which mean what they mean irrespective of context? What do words such as the and however mean, and how do they mean that? Are there any two words which mean exactly the same thing?