Free download english morphology

Carousel Next. What is Scribd? English Morphology. Uploaded by Alvaro Alonso. Document Information click to expand document information Description: Description of English Morphology. Original Title English Morphology pdf. Did you find this document useful? Is this content inappropriate? Report this Document. Description: Description of English Morphology. Flag for inappropriate content. Download now. Original Title: English Morphology pdf. Related titles. Carousel Previous Carousel Next. Unit 1 the Study of Morphological Structure of English.

Jump to Page. Search inside document. Syntax: studies the combination of words. Compound word: 2 free roots: handbook. Inflection is more productive -s; -ing; -ed Inflection has more regular meanings: - s plural , -ed past Dictionaries include derived lexemes but not inflected words. Adds lexical meanings to the root: important un-important mother mother-hood Creates new lexemes.

Documents Similar To English Morphology pdf. Munna Hossain Khan. Pot Fairuz. Cristina Rocas-Bisquera. Abdullah Shaghi. Akshay Hinduja. Anika Reza. Dana Si David. Vichitra Kumarian. Khairuddin Zurina. Arul Dayanand. The Handbook of Morphology - Spencer and Zwicky. Billie Gee. Cristina-Andreea N. Do you think it? Yeah, confirm it. This is the site that will offer all guides that probably you need. Are guide's collections that will make you feel interested to check out?

Rochelle Lieber, Ingo Plag to refer currently, you need to follow this page constantly. Keep in mind. By seeing this internet site, you have started to make new deal to constantly be up-to-date. This book presents a comprehensive, data-rich, theory-neutral description of English word formation, including inflection and derivation, compounding, conversion, and such minor processes as subtractive morphology.

It also offers analyses of the theoretical challenges these phenomena present. It is the first to make systematic use of large linguistic corpora, including the Corpus of Contemporary American English, the British National Corpus, and the American National Corpus by which, for example, the authors are able to measure the productivity of different patterns and to trace semantic developments as they happen. After setting out their methodology and theoretical assumptions, the authors describe word formation and inflection in contemporary English.

They give equal weight to form and meaning and cover nominalizations, agentive forms, comparatives, root and synthetic compounds, as well as more recondite topics such as the abstract noun-forming suffixes -hood, -dom, and -ship, neoclassical compounds, and the morphology of numbers. They examine the relations between orthography and phonological form. While their focus is on contemporary morphology, they trace the history of phenomena wherever doing so helps to understand and explain current form and function.

The final part of the book shows how the data assembled within it bear on current theoretical issues and reveal new lines of research. This outstanding book will interest all scholars and students of English and of linguistic morphology more generally. There is no doubt that the authors have achieved this main goal This book fills a gap in the literature and will be an essential reference for anyone interested in the morphology of English.

It can be of benefit to instructors, linguistic researchers, and students, both beginners and advanced, looking for a quick overview of the many issues, problems, and intricacies of English morphology He is co-editor of the journal Morphology. Bauer, Rochelle Lieber, Ingo Plag as referrals, going to search the title and also theme in this website is. You could likewise as soon as possible to review guide that is currently downloaded.

It will certainly relieve you any place you require guide soft file to read. Novel crystal form Form P of Doxofylline with new rod shape morphology has been identified and fully characterized by a variety of analytical techniques such as. Both parsing and generation are even simpler if OM is interpreted as a finite state transducer under local constraint evaluation.. It remains to be seen, if the specific.

The application of two level morphology to non concatenative German morphology The application of t w o l e v e l morphology to non concatenative German m o r p h o l o g y Harald Trost. Past studies of the variable ING have demonstrated regular and stable social and stylistic conditions across English speech communities around the world, factors which shape the. Show more Page. Download now 9 Page. Review "This comprehensive book, which covers all aspects of English morphology, is a needed reference work.

But does it follow that all the word forms of a lexeme must always share the same root morpheme? The answer is yes, but seldom at least in English. Because it is a verb, we expect it to have a past tense form, and this expectation is not disappointed. Should we say, then, that go and went are allomorphs of one morpheme? Most linguists would say no; rather, they would treat this as showing that one lexeme may be represented by two or more quite distinct root morphemes not allomorphs.

The term given to this phenomenon is suppletion; go and went are said to be distinct roots and hence distinct morphemes , standing in a suppletive relationship as representatives, in different grammatical contexts, of one lexeme. This is because suppletion is generally seen as a relationship between forms of the same lexeme, whereas allomorphy need not be. The discussion so far in this chapter has been rather general.

However, these sections will provide opportunities to illustrate a few further general issues and notions as well. These borrowings from Latin and elsewhere are discussed further in Chapter 9. I have already mentioned two teeth, men where there is a change in the vowel of the root — or, more precisely, an allomorph of the root with a different vowel from the singular. An obvious question, therefore, is: if the plural and singu- lar forms of these nouns are the same, how can we tell whether they are singular or plural?

The answer is: according to the syntactic context. Consider the following examples: 12 A deer was visible through the trees. It is true that the relationship is not hard- and-fast: there are plenty of domesticated and game animals which have regular -s plurals e. In Section 4. But does that mean that all nouns referring to countable entities have both singular and plural forms? Not quite. Those scissors belong in the top drawer. Your pants have a hole in the seat.

However, for these lexemes, there is a conventional circumlocution or periphrastic form: pair of pants and pair of scissors as in That pair of scissors belongs in the top drawer. But what about the noun phrases themselves?

The choice between singular and plural there is determined not by grammar but by meaning, one may think — by what the speaker wants to say. If so, does this contrast really deserve to be called grammatically conditioned? Despite the freedom to choose between, say, this pianist and these pianists as subjects of 9 , there is still a sense in which English grammar affects the choice between singular and plural. It does so in the sense that it imposes the choice.

In talking about a series of weekly piano concerts, we are free to be vague about the number of pianists who perform — except that we are forced by English grammar to be precise about whether there is one that pianist or more than one these pianists.

Like- wise, if I see a cat or some cats in the garden, I cannot report what I have seen without making it clear whether there was just one cat, as in 16 or more than one cat, as in A formulation that is deliberately vague on that issue, such as 18 , is unacceptable: 16 I saw a cat in the garden. The best we can do to express the intended content of 18 is use a circumlocution like one or more cats or at least one cat. That does not mean that one cannot distinguish between one object and several when talking Chinese; it is just that the distinction is not imposed by Chinese grammar, which permits ambivalence about plurality.

Compare the meaning of 14a with that of 19 and 20 : 19 That pair of scissors belongs in the top drawer. On the other hand, 14a is vague in just the way that 17 was meant to be; it can be interpreted as synonymous with either 19 or The singular—plural distinction is the only grammatical distinction that is expressed morphologically in English nouns. These classes are so called because their membership can be added to, and indeed is added to constantly as new words come into use.

By contrast, one does not expect in English to encounter a new pronoun a word such as I or she or us or a new preposition a word such as in or at or without. However, determiners deserve a mention here because some of them, like nouns, display a singular—plural contrast, and pro- nouns combine a singular—plural contrast with contrast unique to them, between subject and non-subject forms.

We have already encountered the distinction between this and these, as in this pianist and these pianists.

In English, the same technique is used for one small closed class of lexemes, namely personal pronouns. If one replaces John and Mary with the appropriate pronouns in these two examples, the outcome is as in 26 and 27 : 26 He loves her. He and him are sometimes said to contrast in case, he belonging to the nominative case and him belonging to the accusative case. One possibility is to say that these are pronoun forms belonging to a third case, the genitive or possess- ive, which stand in for apostrophe-s forms in noun phrases that consist only of a personal pronoun.

But these are issues of syntax rather than morphology. Mary gives a lecture every year. Mary gave a lecture last week. Mary is giving a lecture today. Mary has given a lecture today. The lecture is always given by Mary. Mary may give a lecture. Mary wants to give a lecture. Mary and John give a lecture every year. The contrast between present at 28a and past at 28b is a contrast of tense. The other dimensions of contrast manifested in 28a are person third person versus the rest and number singular versus plural, just as for nouns and pronouns.

In fact, most verbs have only four forms, because the past tense and the perfect or passive participle forms are the same. I will not list them all here, however, because the study of these irregularities belongs to grammar rather than to word- formation. But they deserve mention here because their various forms distinguish an unusually small or large range of grammatical words. On the basis of our experience with plurals of countable nouns and past tense forms of verbs, then, you will probably expect that every adjective lexeme should possess a comparative and a superlative form or, at any rate, every adjective denoting a property that can be present to a greater or lesser degree.

In some languages, a lexeme may have hundreds or even thousands of distinct forms. Why languages should differ so enormously in this respect is a fascinating question, but one that we cannot delve into here. What word form represents each of the following grammatical words?

Which of the forms in question 2 are irregular? Are any of them suppletive? Identify at least one adjective, not mentioned in the chapter, that has a suppletive comparative form. Which ones? Consult a native speaker, if necessary. Do not be surprised if different speakers disagree!

Aronoff discusses the fact that the same word form serves as both perfect participle and passive participle in English, despite the fact that syntactically the two are quite distinct. Citing similar examples, he points out the wider implications of this phenomenon for morpho- logical theory.

What lexeme could this be? This question is easy to answer when we notice that, alongside performance, there is a plural form performances. This tells us something about the relationship between perform and performance: it is a relationship not between word forms but rather between lexemes.

Thus derivational morphology is concerned with one kind of relationship between lexemes. Some bases are roots, whether bound e. What I have just called word classes are the same as what in traditional terminology are called parts of speech and what many contemporary linguists call lexical categories. That belief is incorrect. If you feel tempted by it, please do not skip this section!

That seems scarcely appropriate. Resembling, one may think, hardly counts as an activity. To say that for example my great- uncle William resembles a giraffe is not to report some action of his, but rather to describe him. Does that mean, then, that a lexeme cannot have both noun forms singular and plural and verb forms past, third person singular present, and so on?

The more interesting question, then, is: do such roots exist? The answer is certainly yes. The answer is no, for two reasons. Although numerous, it is still heavily outnumbered by the proportion that is either purely noun-like in its grammatical behaviour e. Even there, however, it seems generally necessary to distinguish nominal i. A second kind of reason has to do with English in particular. She stated that it would rain. She knew that it would rain. She denied that it would rain.

She admitted that it would rain. She acknowledged that it would rain. For all of these sentences we can identify a nominal counterpart, that is a counterpart of the form her … that it would rain: 2 a.

In morphological terms, therefore, it makes sense to say that the verbal construction in 1 is basic, the nominal construction in 2 being derived from it. If we look only at 3 and 4 , we have no basis for deciding whether these lexemes are basically nominal or basically verbal: 3 a. She hoped that it would rain. She feared that it would rain. Either way, these ambivalent words present the problem of determining which word class the basic form belongs to.

Some introductory treatments of English grammar talk as if not just many but all adverbs end in -ly. Also, the very exist- ence of some of these words seems arbitrary. I use quotation marks here to identify non-existent but plausible lexemes.

It is merely an accident that some of these words have come into general use while others have not, so those that do exist must be lexically listed. However, this stress difference has many parallels compare Canada and Canadian, mathematics and mathematician , and many linguists would argue that it is due to a phonological process.

If so, then the base to which -ian is attached in historian for example can be regarded as the same as the free allomorph history. Some of these nouns are formed from bases other than the free form of the corresponding adjective, e. This gappiness is a reason for counting all nouns in -ity as lexical items, and its implications will be discussed further in Chapter 8. Here are just a few: 13 -ance, -ence, e. This semantic waywardness will be discussed further in Chapter 8, along with a phonological restriction on the use of noun-forming -al.

By contrast with some languages, however, the derivational use that English makes of vowel change is minimal. Languages that exploit it much more consistently are members of the Semitic family, such as Arabic and Hebrew. Because it is so common, most dictionaries do not attempt to list all un- adjectives. It is more restricted than un-, largely for historical reasons such as will be discussed in Chapter 9.

However, such forms in italics in 23 can also be adjectives: 23 a. The party-goers sounded very drunk. The car seemed more damaged than the lamp-post. We have already encountered -able in 22 , where the vari- ant, or allomorph, -ible is also illustrated. Here are some: 27 -ful, e. Semantically, the examples in 31 — 34 are mostly straightforward, although those with de- are less so: to decompose is not to undo the creative work of a musical composer!

Intransitive verbs, such as lay in 36b , lack such an object. Jill laid the book on the table. The book lay on the table. Jill boiled the water. The water boiled. More will be said about such historical developments in Chapter 9. Hence, when the roots to which they are attached are bound e.

However, because most of the bases to which -ate is attached are bound roots, it does not clearly favour either adjectival or nominal bases. It turns out that the adjectives that can be bases for deriving -en verbs are all monosyllabic and all end in plosives the sounds usually spelled p, b, t, d, c k and g in English or fricatives including the sounds usually spelled s, th, f and v.

That is a question about productivity, so we will defer it to Chapter 8. However, the starting-point for an answer is to look for adjectives which end in plosives or fricatives but for which there is no corresponding verb in -en. There is no need to wait until Chapter 8 before embarking on this search! Some of the reasons for this are historical, and will be discussed in Chapter 9.

Why so? Underlying this puzzle are big questions about the status of the word as a linguistic unit — questions too big and controversial to be tackled here. However, more will be said about unpredictability in derivation when we discuss productivity in Chapter 8. Here are ten adjectives. Give examples to show that it can also be used to derive nouns from other nouns. Give examples to show that it can also be used to derive adjectives from nouns and from other adjectives.

Can you identify any phonological characteristic that the -ar adjectives share? Two pioneering works on derivational morphology within modern linguistic theory are Aronoff and Jackendoff They deserve high priority for any reader who wants to go beyond introductory texts. In this chapter we will look at compounds, that is words formed by combining roots, and the much smaller category of phrasal words, that is items that have the internal structure of phrases but function syntactically as words.

As we will see, some types of compound are much commoner than others. There are also some styles of writing for example, newspaper headlines in which compounds are especially frequent. Roots in English are mostly free rather than bound. How can we tell, then, whether a pair of such roots constitutes a compound word or a phrase, that is a unit of sentence structure rather than a complex word? Consider the expressions a green house, with its literal meaning, and a greenhouse, meaning a glass structure not usually green in colour!

They live in a white house, not a yellow one! Apart from stress, a second criterion traditionally used for dis- tinguishing compounds from phrases is semantic: a compound tends to have a meaning that is more or less idiosyncratic or unpredictable. This is true of most of the compounds in 1 — 5.

This criterion must be treated with caution, however, because, as we noted in Chapter 2, being semantically unpredictable does not correlate exactly with being a word.

All the same, it is true that words are more likely to be lexical items than phrases are, so treating semantic idiosyncrasy as an indicator of compound status will not often be misleading. All the compounds in 1 — 5 are nouns, and compound nouns are indeed the commonest type of compound in English.

We will examine them in detail in later sections. Meanwhile, Sections 6. Nevertheless, a variety of types exist which may be dis- tinguished according to their structure: 6 verb—verb VV : stir-fry, freeze-dry 7 noun—verb NV : hand-wash, air-condition, steam-clean 8 adjective—verb AV : dry-clean, whitewash 9 preposition—verb PV : underestimate, outrun, overcook Only the PV type is really common, however, and some compounds with under-, over- and out- do not need to be classed as lexical items.

You will notice that all these compounds have a verb as the rightmost element, and also that, with most of them, the activity denoted by the compound as whole is a variety of the activity denoted by that right- most element.

Let us call these compounds right-headed, the rightmost element being the head. Most English compounds are right-headed, but not all, as we shall see in Section 6. In overactive at 12 , the head of the compound is the adjective active derived from the verb act in the fashion described in Section 5. More will be said about the implications of this kind of structuring in Chapter 7. All the compounds in 10 — 12 are right-headed. There are also a few compound adjectives that are not right-headed, but we will discuss them along with all headless compounds in Section 6.

That is not surprising. Most of these are also right-headed, although we will defer further discussion of headed- ness to Section 6. In fact, almost any pair of nouns can be juxtaposed in English so as to form a compound or a phrase — provided that there is something that this compound or phrase could plausibly mean.

The issue of meaning turns out to play an important part in distinguishing two kinds of NN com- pound. Consider the four examples at Does each one have a precise interpretation that is clearly the most natural, on the basis of the mean- ings of their two components? For hair restorer, the answer is surely yes: it most naturally denotes a substance for restoring hair growth.

The difference in precision with which we can interpret hair restorer on the one hand and hairnet etc. These expected or required nominal concomitants to a verb are called its arguments. For example, an X-restorer, whatever X is, something or some- one that restores X. It is time to introduce some terminology, for convenience. Let us call a NN compound like hairnet or mosquito net, in which the right-hand noun is not derived from a verb and whose interpretation is therefore not precisely predictable on a purely linguistic basis, a primary or root compound.

Yet another term sometime used is synthetic compound. In this respect they are like most NN compounds and most compounds generally — but not all, as we shall see in the next section.

However, whereas a greenstone is a kind of stone and a blackboard is a kind of board, a faintheart is not a kind of heart but a kind of person — someone who has a faint heart, metaphorically. So, although heart is a noun, it is not appropriate to call heart the head of the com- pound.

Rather, faintheart is headless, in the sense that its status as a noun is not determined by either of its two components.

A few VN-type compound nouns resemble secondary compounds in that the noun at the right is interpreted as the object of the verb: 18 pickpocket, killjoy, cutpurse These too are headless, in that a pickpocket is not a kind of pocket, for example. An implication of these analyses is as follows: if the fact that heart and pocket are nouns is really irrelevant to the fact that faintheart and pickpocket are nouns too, we should expect there to be some headless nouns in which the second element is not a noun at all — and likewise, perhaps, headless adjectives in which the second element is not an adjec- tive.

Both expectations turn out to be correct. The nouns at 19 can be seen as a special case of this, where the base is a verb plus another word some- times constituting a lexical item , as illustrated in 20 : 20 a. The plane took off at noon. The chairman wrapped the meeting up. The students sat in during the discussion. They live in a very downmarket neighbourhood. Sporadically, however, we encounter a kind of compound where at least one com- ponent is reproduced only partially.

These are known as blends. Examples of partial blends, where only one component is truncated, are talkathon from talk plus marathon and cheeseburger from cheese plus hamburger.

Intermediate between an acronym and a blend is sonar from sound navigation and ranging. It does not follow that any string of capital letters represents an acronym. This semantic predictability is crucial to the coining of new technical terms using these elements.

Apart from containing bound roots, anthropology differs in two other ways from most compound nouns. Firstly, it has a central linking vowel -o- that cannot conclusively be assigned to either root. In this respect it resembles many combining-form compounds. In this respect it resembles e. Not surprisingly, some combining forms can function in this way too in other words, the dividing line between combining forms and other bound roots is not sharp : for example, soci- and electr o - from 23 also occur, indeed much more commonly, in social and electric.

Given that combining forms, and the compounds that contain them, are so untypical of compounds in general, it is natural to ask how English has come to acquire them. We will have more to say about these circumstances in Chapter 9. There is a clear difference between compound word structure and sentence structure here. But there are also complex items that function as words, yet whose internal structure is that of a clause or phrase rather than of a compound.

There is no standard term for these items, so I will introduce the term phrasal words. An example of a phrasal word is the noun jack-in-the-box.

Though structurally a phrase, then, it behaves as a word. Despite its hyphens, therefore, brother-in-law is not a word at all but a phrase although also a lexical item — a combination discussed in Chapter 2. Can phrases other than noun phrases constitute phrasal words?

How do they form their plural: like attorney generals, or like attorneys general? It seems better, therefore, to treat them as examples of something that we have not so far encountered: endocentric words which, untypically, have their head on the left rather than on the right. On the other hand, if you prefer the latter sort of plural attorneys general , they seem more akin to brother s - in-law: not words but lexicalised phrases. In Chapter 7 I will have more to say about a fact that I have not emphasised so far: one or both of the component words in a compound may itself be a compound, so there is in principle no upper limit to the size of compounds.

We have also seen that at least one syntactic relationship can be expressed within compounds just as well as within sentences, namely the verb—object relationship or perhaps one should say the action—goal relationship , as in hair restorer. One might ask, then, why English, or any language, needs both com- pound word-structure and clause-structure side by side: could not just one do the work performed in actual English by both?

That is an import- ant question, but unfortunately one for which there is no generally agreed answer. Further discussion of it is therefore a task for research papers, rather than for an introductory textbook such as this. Which of the following are compound words, which are phrases, and which are phrasal words? Of the compounds not the phrases or phrasal words in Exercise 1, which are endocentric and which are exocentric? Of the compound nouns in Exercise 1, which are primary or root compounds and which are secondary or verbal compounds?

Identify with the help of a dictionary, if necessary the sources of the following blends or acronyms: brunch, motel, radar, modem, laser. Each of these words is a compound containing at least one bound Graeco-Latin combining form. Be warned, however, that these writers treat as compounds some noun—noun collocations that I analyse as phrases. However, this is not the dominant view among con- temporary morphologists.

For an opposed view, see Anderson , reviewed by Carstairs-McCarthy A classic discussion of secondary compounds is Lieber See also section 4. Some words are so predictable, indeed, that they do not have to be listed as lexical items. This predictability of meaning depends on how the structure of complex word forms guides their interpretation.

Even with words that are lexically listed, unless their meaning is entirely different from what one might expect, such guidance is relevant. This chapter is about how it operates, and also in Section 7. In some words, structure is straightforward. Because there are only two elements in this word form, it may seem there is not much to say about its structure.

Sections 7. Finally, in Section 7. In Chapter 6 we saw that most compounds are headed, with the head on the right. Consider, however, the role played by the head house of a compound such as greenhouse.

Many though not all linguists therefore treat -er as the head of teacher in just the same way as house is the head of greenhouse. In the derived words teacher and helpful, therefore, the two components are not equal contributors, so to speak; rather, the righthand element as in most compounds has a special status. Is our expectation disappointed, then?

Consider the relationship between helpful and unhelpful. In helpful, -ful has a clearly wordclass-determining role because it changes a noun, help, into an adjective. In unhelpful, however, un- has no such role; rather, it leaves the wordclass of helpful unchanged see Section 5. This characteristic of un- is not restricted to adjectives, moreover.

This strongly suggests that the head of of all these words is not un- but the base to which un- is attached helpful, tie, ease etc. Similar arguments apply to re-: rearrange, repaint and re-educate are verbs, just as arrange, paint and educate are. Examples are unhelp- fulness and helplessness. Likewise, helplessness contains -less by virtue of the fact that it contains helpless.

Once that is recognised, the apparent need to make special provision for -ful and -less when they appear inside complex words, rather than as their rightmost element, disappears. However, this topsy-turvy usage has become well established in linguistic discussions. The points in a tree diagram from which branches sprout are called nodes. The nodes in 3 and 4 are all labelled, to indicate the wordclass of the string that is, of the part of the whole word that is dominated by the node in question.

The information about structure contained in tree diagrams such as 3 and 4 can also be conveyed in a labelled bracketing, where one pair of brackets corresponds to each node in the tree: [[un-[[help V]N-ful]A]A-ness]N, [[[help V]N -less]A-ness]N.

One thing stands out about all the nodes in 3 and 4 : each has no more than two branches sprouting downwards from it. This kind of uncertainty was discussed in Chapter 2. Another salient point in all of 3 — 6 is that more than one node in a tree diagram may carry the same wordclass label N, V, A.

However, it has considerable implications for the size of the class of all possible words in English. One cannot so easily demonstrate that there is no such thing as the longest word in English; but it is not necessary to do so in order to demonstrate the versatility and vigour of English word-formation processes. The issue of how new words can be formed will be taken up again in Chapter 8. The same applies to compounds: any compound has just two immediate constituents. In Chapter 6, all the compounds that were discussed contained just two parts.

This was not an accident or an arbitrary restriction. To see this, consider for example the noun that one might use to denote a new cleaning product equally suitable for ovens and windows. Parallel to the secondary compound hair restorer are the two two-part compounds oven cleaner and window cleaner.

Can we then refer to the new product with a three-part compound such as window oven cleaner? The answer is surely no. Window oven cleaner is not naturally interpreted to mean something that cleans both windows and ovens; rather, it means something that cleans window ovens that is, ovens that have a see-through panel in the door.

As simple compounds, verb—noun and Regan—Gorbachev certainly sound odd. Never- theless verb—noun contrasts denotes crucially contrasts between verbs and nouns, not contrasts some of which involve verbs and others of which involve nouns; therefore verb—noun deserves to be treated as a subunit within the whole compound verb—noun contrast.

Likewise, a Reagan— Gorbachev encounter necessarily involves both Reagan and Gorbachev, not just one of the two, so Reagan—Gorbachev deserves to be treated as a subunit within Reagan—Gorbachev encounters. In Chapter 6 we concentrated on compounds with only two members. But, given that a compound is a word and that compounds contain words, it makes sense that, in some compounds, one or both of the components should itself be a compound — and 8 , with its most natural interpretation, shows that this is indeed possible, at least with compound nouns.

Window oven, if it is a com- pound, should have its main stress on the lefthand element, namely window — and that seems correct. Again, that seems correct.

Nevertheless, where no contrast is implied or stated such as between marketing and manufacture , the most natural way of pronouncing the example at 9 renders window the most prominent element. Can we then conclude that all complex compound nouns follow the left-stressed pattern of simple compound nouns? Before saying yes, we need to make sure that we have examined all relevant varieties. It may have struck you that, in 8 and 9 , the compounds-within-compounds are uniformly on the left.

We have not yet looked at compounds or potential compounds in which it is the righthand element in fact, the head that is a compound. Again, we are assuming that no contrast is implied — between a holiday trip and a business trip, say. Yet it would be strange if a compound noun cannot itself be the head of a compound noun, given that any other kind of noun can be. The best solution seems to be to qualify what was said in Chapter 6 about stress in compound nouns.

The usual pattern, with stress on the left, is overridden if the head is a compound. In that case, stress is on the right — that is, on the compound which constitutes the head. Another way of expressing this is to say that the righthand component in a compound noun gets stressed if and only if it is itself a compound; otherwise, the lefthand component gets stressed.

It is also consistent with a more complex example such as 12 , involving internal compounds on both left and right branches. Discussing these instances leads us to the question of whether a unit larger than a word that is, a phrase can ever be a constituent of a compound word.

Its structure seems clear: it is a phrase consisting of two words, an adjective nuclear and a noun physicist. So, if the interpretation of linguistic expressions is always guided by their structure, it ought to mean a physicist who is nuclear. Instead, this expression means someone who is an expert in nuclear physics.