EN 302 Finals Notes PDF

The sound patterns of language Ever on the search for legal jokes not necessarily connected with the death penalty, I consulted a friend who is still practising. She said a member of her chambers was in court one Monday morning when the judge said, "I'm afraid we'll have to adjourn this case, I have...

The sound patterns of language Ever on the search for legal jokes not necessarily connected with the death penalty, I consulted a friend who is still practising. She said a member of her chambers was in court one Monday morning when the judge said, "I'm afraid we'll have to adjourn this case, I have written my judgment out, but I left it in my cottage in Devon and I can't get it sent here until tomorrow." "Fax it up, my Lord," the helpful barrister suggested, to which his Lordship replied, "Yes, it does rather." Quoted in Crystal (1995) In the preceding chapter, we investigated the physical production of speech sounds in terms of the articulatory mechanisms of the human vocal tract. That investigation was possible because of some rather amazing facts about the nature of language. When we considered the human vocal tract, we did not have to specify whether we were talking about a fairly large male, over six feet tall, weighing over 200 pounds, or about a rather small female, about five feet tall, weighing 100 pounds. Yet those two physically different individuals would inevitably have physically different vocal tracts, in terms of size and shape. In a sense, every individual has a physically different vocal tract. Consequently, in purely physical terms, every individual will pronounce sounds differently. There are, then, potentially thousands of physically different ways of saying the simple word me. In addition, each individual will not pronounce the word me in a physically identical manner on every occasion. Obvious differences occur when the individual is shouting, is asking for a sixth martini, or is suffering from a cold. Given this vast range of potential differences in the actual physical production of a speech sound, how do we manage consistently to recognize all those versions of me as the form [mi], and not [ni], or [si], or [ma], or [mo], large extent by the study of phonology. Phonology Phonology is essentially the description of the systems and patterns of speech sounds in a language. It is, in effect, based on a theory of what every speaker of a language unconsciously knows about the sound patterns of that language. Because of this theoretical status, phonology is concerned with the abstract or mental aspect of the sounds in language rather than with the actual physical articulation of speech sounds. Phonology is about the under- lying design, the blueprint of the sound type, that serves as the constant basis of all the variations in different physical articulations of that sound type in different contexts. Thus, when we think of the [t] sound in the words tar, star, writer and eighth as being 'the same', we actually mean that, in the phonology of English, they would be represented in the same way. In actual speech, these [t] sounds are all very different. However, all those articulation differences in [t] sounds are less important than the distinction between the [t] sounds in general and the [k] sounds, or the [f] sounds, or the [b] sounds, because there are meaningful consequences related to the use of one rather than the others. These sounds must be distinct meaningful sounds, regardless of which individual vocal tract is being used to pronounce them, because they are what make the words tar, car, far and bar meaningfully distinct. Considered from this point of view, we can see that phonology is concerned with the abstract set of sounds in a language which allows us to distinguish meaning in the actual physical sounds we say and hear. Phonemes Each one of these meaning-distinguishing sounds in a language is described as a phoneme. When we considered the basis of alphabetic writing in Chapter 2, we were actually working with the concept of the phoneme as the single sound type which came to be represented by a single symbol. It is in this sense that the phoneme /t/ is described as a sound type, of which all the different spoken versions of [t] are tokens. Note that slash marks are conventionally used to indicate a phoneme, /t/, an abstract segment, as opposed to the square brackets, as in [t], used for each phonetic, or physically produced, segment. An essential property of a phoneme is that it functions contrastively. The basis of the contrast in meaning between the form’s fat and vat, or fine and vine. This contrastive property is the basic operational test for deter- mining the phonemes which exist in a language. If we substitute one sound for another in a word and there is a change of meaning, then the two sounds represent different phonemes. The consonant and vowel charts presented in Chapter 5 can now be seen as essentially a mapping out of the phonemes of English. The terms which were used in creating that chart can be considered 'features' which distinguish each phoneme from the next. If the feature is present, we mark it with a plus (+) sign; if it's not present, we use a minus (-) sign. Thus, /p/ can be characterized as [-voice, +bilabial, +stop] and /k/ as [-voice, +velar, +stop]. Because these two sounds share some features, they are sometimes described as members of a natural class of sounds. The prediction would be that sounds which have features in common would behave phonologically in some similar ways. A sound which does not share those features would be expected to behave differently. For example, /v/ has the features [+voice, +labiodental, +fricative] and so cannot be in the same 'natural class' as /p/ and /k/. Although other factors will be involved, this feature- analysis could lead us to suspect that there may be a good phonological reason why words beginning with /pl-/ and /kl-/are common in English, but words beginning /vl-/are not. Could it be that there are some definite sets of features required in a sound in order for it to occur word-initially before /l/? If so, then we will be on our way to producing a phonological account of permissible sound sequences in the language. Phones and allophones While the phoneme is the abstract unit or sound-type ('in the mind'), there are many different versions of that sound-type regularly produced in actual speech ("in the mouth'). We can describe those different versions as phones. Phones are phonetic units and will appear in square brackets. When we have a set of phones, all of which are versions of one phoneme, we refer to them as the allophones of that phoneme. For example, the [t] sound in the word far is normally pronounced with a stronger puff of air than is present in the [t] sound of the word star. If you put the back of your hand in front of your mouth as you say tar, then star, you should have some physical evidence of the aspiration (the puff of air) accompanying the [t] sound in the initial position of tar (but not in star). This aspirated version is represented more precisely as [th]. That's one phone. In the last chapter, we noted that the [t] sound between vowels in a word like writer often becomes a 'flap', which we represented as [D]. That's another phone. In the pronunciation of a word like eighth, the influence of the final dental [e] sound causes a dental articulation of the [t] sound. This would be represented more precisely as [t]. That's yet another phone. There are other variations of this sound which, like [t], [D] and , can be represented differently in a detailed, or narrow, phonetic transcription. Because these variations form a set of phones, they would typically be referred to as allophones of the phoneme /t/. The crucial distinction between phonemes and allophones is that substituting one phoneme for another will result in a word with a different meaning (as well as a different pronunciation), but substituting allophones only results in a different (and perhaps odd) pronunciation of the same word. Let's take another brief example. In English, there is a difference in pronunciation of the /i/ sound in words like seed and seen. In the second word, the effect of the nasal consonant [n] makes the [i] sound nasalized. This nasalization can be represented by a diacritic ["], called 'tilde', over the symbol [I] in narrow phonetic transcription. So, there are at least two phones, [i] and , used in English to realize a single phoneme. They are allophones of/i/in English. It is possible, of course, for two languages to have the same pair of phonetic segments, but to treat them differently. In English, the effect of nasalization on a vowel is treated as allophonic variation because the nasalized version is not meaningfully contrastive. In French, however, the pronunciation [me] is used for one word mets, meaning 'dish', and [me] for a different word main, meaning "hand". Also, [so] for seau, meaning 'pail', contrasts with [sö] for son, meaning 'sound'. Clearly, in these cases, the distinction is phonemic. Minimal pairs and sets Phonemic distinctions in a language can be tested via pairs and sets of words. When two words such as par and bar are identical in form except for a contrast in one phoneme, occurring in the same position, the two words are described as a minimal pair. More accurately, they would be classified as a minimal pair in the phonology of English. (Arabic does not have this contrast between the two sounds.) Other examples of English minimal pairs are fan-van,bet-bat,site-side. Such pairs have been used frequently in tests of English as a second language to determine non-native speakers' ability to understand the contrast in meaning resulting from the minimal sound. When a group of words can be differentiated, each one from the others, by changing one phoneme (always in the same position), then we have a minimal set. Thus, a minimal set based on the vowel phonemes of English would include feat, fit, fat, fate, fought, foot, and one based on consonants could have big, pig, rig, fig, dig, wig. Phonotactics This type of exercise involving minimal sets also allows us to see that there are indeed definite patterns to the types of sound combinations permitted in a language. In English, the minimal set we have just listed does not include forms such as lig or vig. As far as I know, these are not English words, but they can be viewed as possible English words. That is, your phonological knowledge of the pattern of sounds in English words would allow you to treat these forms as acceptable if, at some future time, they came into use. They represent 'accidental' gaps in the vocabulary of English. It is, however, no accident that forms such as [fsIg] or [rnIg] do not exist or are unlikely ever to exist. They have been formed without obeying some constraints on the sequence or position of English phonemes. Such constraints are called the phonotactics of a language and are obviously part of every speaker's phonological knowledge. Because these constraints operate on units larger than the single segment, or phoneme, we have to consider the basic structure of that larger phonological unit called the syllable. Syllables and clusters A syllable must contain a vowel (or vowel-like) sound. The most common type of syllable in language also has a consonant before the vowel, often represented as CV. Technically, the basic elements of the syllable are the onset (one or more consonants) and the rime. The rime (also written as 'rhyme") consists of the vowel, which is treated as the nucleus, plus any following consonant(s), treated as the coda. Thus, syllables like me, to or no have an onset and a nucleus, but no coda. They are known as 'open' syllables. When a coda is present, as in the syllables up, cup, at or hat, they are called 'closed' syllables. The basic structure of the kind of syllable found in English words like green (CCVC), eggs (VCC), and (VCC), ham (CVC), I (V), do (CV), not (CVC), like (CVC), them (CVC), Sam (CVC), I (V), am (VC), is shown in the accompanying diagram. Both the onset and the coda can consist of more than one consonant, also known as a consonant cluster. The combination st is a consonant cluster (CC) as onset in the word stop, and as coda in the word post. There are many CC onset combinations permitted in English phonotactics, as in black, bread, trick, twin, flat and throw, with approximants (/w/, /r/, /l/) frequently appearing in second position. (Note that throw begins with only two consonants, /er/, once again showing that spelling is not a good guide in phonology.) English actually can have larger onset clusters, as in stress and splat, consisting of three consonants (CCC). The phonotactics of these larger onset consonant clusters in English is not difficult to describe. The first consonant must always be /s/, followed by one of the voiceless stops (/p/, /t/, /k/) and then one of the approximants (/r/,/,/w/). You can check if this description is adequate for the combinations in splash, spring, strong, scream and square. Does the description also cover the second syllable in the pronunciation of exclaim? How about /ek-skleym/? Remember that it is the onset of the syllable that is being described, not the beginning of the word. It is quite unusual for languages to have consonant clusters of this type. Indeed, the syllable structure of many languages (e.g. Hawaiian or Japanese) is predominantly CV. It is also noticeable in English that large consonant clusters are frequently reduced in casual conversational speech, particularly if they occur in the middle of a word. This is just one example of what is often discussed in terms of ‘co-articulation effects'. Co-articulation effects In much of the preceding discussion, we have been describing the speech sounds as if they are always pronounced carefully and deliberately, almost in slow motion. Speech isn't like that very often. Mostly our talk is fast and spontaneous, and it requires our articulators to move from one sound to the next without stopping. The process of making one sound almost at the same articulation effects, called 'assimilation' and 'elision'. Assimilation When two phonemes occur in sequence and some aspect of one phoneme is taken or copied' by the other, the process is known as assimilation. In terms of the physical production of speech, one might assume that this regular process is occasioned by ease of articulation in everyday talk. In isolation, you would probably pronounce /I/ and /æ/ without any nasal quality at all. However, in saying words like pin and pan, the anticipation of forming the final nasal consonant will make it 'easier to go into the nasalized articulation in advance and consequently the vowel sounds in these words will be, in precise transcription, [I] and []. This is a very regular feature of English speakers' pronunciation. So regular, in fact, that a phonological rule can be stated in the following way: 'Any vowel becomes nasal whenever it immediately precedes a nasal.' This type of assimilation process occurs in a variety of different contexts. It is particularly noticeable in ordinary conversational speech. By itself, you may pronounce the word can as [kæn], but, if you tell someone I can go, the influence of the following velar [g] will almost certainly make the preceding nasal sound come out as [D] (a velar) rather than [n] (an alveolar). The most commonly observed 'conversational' version of the phrase is [aykǝngo]. Notice that the vowel in can has also changed to schwa [ə] from the isolated- word version []. In many words spoken carefully, the vowel receives stress, but in the course of ordinary talk, that vowel may no longer receive any stress and reduce to schwa. For example, you may pronounce and as [and] in isolation, but in the casual use of the phrase you and me, you almost certainly say [en], as in [yuənmi]. Elision Note that in the last example, in the environment of preceding and follow- ing nasals, the [d] sound of and has simply disappeared. The [d] sound is also commonly 'omitted' in the pronunciation of a word like friendship, [frenšIp]. This 'omission' of a sound segment which would be present in the deliberate pronunciation of a word in isolation is technically described as elision. In consonant clusters, especially in coda position, /t/ is a common casualty in this process, as in the typical pronunciation [æspeks] for aspects, or in [himǝsbi] for he must be. You can, of course, slowly and deliberately pronounce the phrase we asked him, but the process of elision in casual speech is likely to produce [wiæstIm]. Vowels also disappear, as in [V] for every, [IntrIst] for interest, [kæbnIt] for cabinet, and [spowz] for suppose. These two processes of assimilation and elision occur in everyone's speech and should not be treated as a form of sloppiness or laziness in speaking. In fact, consistently avoiding the regular patterns of assimilation and elision used in a language would result in extremely artificial-sounding talk. The point of investigating phonological processes (only a very small number of which have been explored here) is not to arrive at a set of rules about how a language should be pronounced, but to try to come to an under- standing of the regularities and patterns which underlie the actual use of sounds in language. Words and word-formation processes When, in about 1820, a congressman named Felix Walker was accused of speaking drivel - which, evidently, he was-he replied that he was speaking to the people of Buncombe County, North Carolina, his district. Almost immediately his congressional colleagues began referring to any political claptrap or bombast as speaking to Buncombe. Soon the phrase had spread beyond Washington and was being abbreviated to buncombe, often respelled bunkum, and eventually further contracted to bunk. Debunk did not come until 1927. Bunkum in turn begat hokum-a blend of hocus and bunkum. Thus with a single fatuous utterance, the forgotten Felix Walker managed to inspire half a page of dictionary entries. Bill Bryson (1994) Around 1900, in New Berlin, Ohio, a department-store worker named J. Murray Spangler invented a device which he called an electric suction sweeper. This device eventually became very popular and could have been known as a spangler. People could have been spanglering their floors or they might even have spanglered their rugs and curtains. The use could have extended to a type of person who droned on and on (and really sucked), described as spanglerish, or to a whole style of behavior called spanglerism. However, none of that happened. Instead, Mr Spangler sold his new invention to a local businessman called William H. Hoover whose Hoover Suction Sweeper Company produced the first 'Hoover'. Not only did the word hoover (without a capital letter) become familiar all over the world, but in Britain, people still talk about hoovering (and not spanglering) their carpets. The point of this small tale is that, although you had never heard of Mr. Spangler before, you really had no difficulty coping with the new words: Spangler, spanglering, spanglered, spanglerish or spanglerism. That is, you can very quickly understand a new word in your language (a neologism) and accept the use of different forms of that new word. This ability must derive in part from the fact that there is a lot of regularity in the word-formation processes in your language. In this chapter, we shall explore some of those basic processes by which new terms are created. Word-formation processes In some respects, the study of the processes whereby new words come into being in a language like English seems relatively straightforward. This apparent simplicity, however, masks a number of controversial issues, some of which we shall consider in the following chapter. Despite the disagreements among scholars in this area, there do seem to be some regular processes involved, and in the following sections we shall cover the technical terms used to describe those processes and identify examples currently in use which are the result of those processes. It should be remembered that these processes have been at work in the language for some time and many words in daily use today were, at one time, considered barbaric misuses of the language. It is difficult now to under- stand the views expressed in the early nineteenth century over the "tasteless innovation" of a word like handbook, or the horror expressed by a London newspaper in 1909 over the use of the newly coined word aviation. Yet many terms of recent currency cause similar outcries. Rather than act as if the language is being debased, we might prefer to view the constant evolution of new terms and new uses of old terms as a reassuring sign of vitality and creativeness in the way a language is shaped by the needs of its users. Let us consider the ways. Coinage One of the least common processes of word-formation in English is coinage, that is, the invention of totally new terms. The most typical sources are invented trade names for one company's product which become general terms (without initial capital letters) for any version of that product. Older examples are aspirin, nylon and zipper, more recent examples are kleenex, teflon and xerox. It may be that there is an obscure technical origin (e.g. te(tra)- fluor)-on) for such invented terms, but after their first coinage, they tend to become everyday words in the language. Borrowing One of the most common sources of new words in English is the process simply labeled borrowing, that is, the taking over of words from other languages. Throughout its history, the English language has adopted a vast number of loan-words from other languages, including alcohol (Arabic), boss (Dutch), croissant (French), lilac (Persian), piano (Italian), pretzel (German), robot (Czech), tycoon (Japanese), yogurt (Turkish) and zebra (Bantu). Other languages, of course, borrow terms from English, as can be observed in the Japanese use of suupaamaaketto ('supermarket') and rajio ('radio'), or Hungarians talking about sport, klub and futbal, or the French discussing problems of le stress, over a glass of le whisky, during le weekend. A special type of borrowing is described as loan-translation, or calque. In this process, there is a direct translation of the elements of a word into the borrowing language. An interesting example is the French term un gratte- ciel, which literally translates as 'a scrape-sky', or the German Wolkenkratzer ('cloud scraper'), both of which were used for what, in English, is normally referred to as a skyscraper. The English word superman is thought to be a loan-translation of the German Übermensch, and the term loan-word itself is believed to have come from the German Lehnwort. Nowadays, some Spanish speakers eat perros calientes (literally 'dogs hot"), or hot dogs. The American concept of 'boyfriend' was a borrowing, with sound modification, into Japanese as boyifurendo, but as a calque into Chinese as 'male friend' or nan pengyu. Compounding In some of those examples we have just considered, there is a joining of two separate words to produce a single form. Thus, Lehn and Wort are combined to produce Lehnwort in German. This combining process, technically known as compounding, is very common in languages like German and English, but much less common in languages like French and Spanish. Obvious English examples would be bookcase, fingerprint, sunburn, wall- paper, doorknob, textbook, wastebasket and waterbed. This very productive source of new terms has been well documented in English and German, but can also be found in totally unrelated languages, such as Hmong, in South East Asia, which combines hwj ("pot') and kais ("spout") to produce hwjkais ('kettle'). The forms pajkws ('flower' + 'corn' = 'popcorn') and hnabloojtes ("bag' + 'cover' + 'hand' = 'glove") are recent creations. Inis combining of two separate forms to produce a single new term is also present in the process called blending. However, blending is typically accomplished by taking only the beginning of one word and joining it to the end of the other word. In some parts of the United States, there's a product which is used like gasoline, but is made from alcohol, so the 'blended' term for referring to this product is gasohol. If you wish to refer to the combined effects of smoke and fog, there's the term smog. In places where they have a lot of this stuff, they can jokingly make a distinction between smog, smaze (smoke + haze) and smurk (smoke + murk). Some other commonly used examples of blending are bit (binary/digit), brunch (breakfast/lunch), motel (motor/hotel), telecast (television/broadcast) and the Chunnel (Channel/ tunnel), connecting England and France. The recent phenomenon of fund-raising on television that feels like a marathon is typically called a telethon and if you are excessively crazy about video, you may be called a videot. Infotainment (information/entertain ment) and simulcast (simultaneous/broadcast) are also new blends from life with television. To describe the mixing of languages, people refer to Franglais (French/English) and Spanglish (Spanish/English). In order to send information fast, you may use a telex (teleprinter/exchange) or, via computer, a modem (modulator/demodulator), or you may decide to send a fax. But that's not a blend. It's an example of our next category. Clipping The element of reduction which is noticeable in blending is even more apparent in the process described as clipping. This occurs when a word of more than one syllable (facsimile) is reduced to a shorter form (fax), often in casual speech. The term gasoline is still in use, but occurs much less frequently than gas, the clipped form. Common examples are ad ('advertisement'), bra ('brassiere'), cab ('cabriolet'), condo ('condominium"), fan ('fanatic'), flu, perm, phone, plane, pram, pub and sitcom ('situation comedy"). English speakers also like to clip each other's names, as in Al, Ed, Liz, Mike, Ron, Sam, Sue and Tom. There must be something about educational environments that encourages clipping because just about every word gets reduced, as in chem, exam, gym, lab, math, phys-ed, poly-sci, prof and typo. A very specialized type of reduction process is known as backformation. Typically, a word of one type (usually a noun) is reduced to form another word of a different type (usually a verb). A good example of backformation is the process whereby the noun television first came into use and then the verb televise was created from it. Other examples of words created by this process are: donate (from 'donation'), opt (from 'option'), emote (from 'emotion'), enthuse (from 'enthusiasm'), liaise (from 'liaison') and babysit (from 'babysitter"). Indeed, if you backform anything, you have used a back- formation. One very regular source of backformed verbs in English is based on the pattern: worker- work. The assumption seems to have been that if there is a noun ending in -er (or something close in sound), then we can create a verb for what that noun-er does. Hence, an editor must edit, a sculptor must sculpt and burglars, peddlers and swindlers must burgle, peddle and swindle. A particular type of backformation, favored in Australian and British English, produces forms technically known as hypocorisms. First, a longer word is reduced to a single syllable, then-y or-ie is added to the end. Perhaps the most familiar versions of this process are the words movie ('moving pictures') and telly ("television'). It has also produced Aussie ('Australian'), barbie ("barbecue'), bookie ("bookmaker'), brekky ("breakfast') and hankie ("handkerchief"). You can probably guess what Chrissy pressies are. Conversion A change in the function of a word, as, for example, when a noun comes to be used as a verb (without any reduction), is generally known as conversion. Other labels for this very common process are 'category change' and 'functional shift'. A number of nouns, such as paper, butter, bottle, vacation, can, via the process of conversion, come to be used as verbs, as in the following sentences: He's papering the bedroom walls; Have you buttered the toast?; We bottled the home-brew last night, They're vacationing in France. These conversions are readily accepted, but some examples, such as the noun impact being used as a verb, seem to impact some people's sensibilities rather negatively. The conversion process is particularly productive in modern English, with new uses occurring frequently. The conversion can involve verbs becoming nouns, with guess, must and spy as the sources of a guess, a must and a spy. Phrasal verbs (to print out, to take over) also become nouns (a printout, a takeover) One complex verb combination (want to be) has become a very useful noun as in He isn't in the group, he's just a wannabe. Verbs (see through, stand up) also become adjectives, as in see-through material or a stand-up comedian. Or adjectives, such as dirty, empty, total, crazy and nasty, can become the verbs to dirty, to empty, to total, or the nouns a crazy and a nasty. You may even hear of people doing the nasty. Some compound nouns have assumed adjectival or verbal functions, exemplified by the ball park appearing in a ball-park figure or asking some- one to ball-park an estimate of the cost. Other nouns of this type are carpool, mastermind, microwave and quarterback, which are all regularly used as verbs. Other forms, such as up and down, can also become verbs, as in They up the prices or We down a few brews. It is worth noting that some converted forms shift substantially in meaning when they change category. The verb to doctor often has a negative sense, not normally associated with the source noun a doctor. A similar kind of reanalysis of meaning is taking place with respect to the noun total and the verb run around, which do not have negative meanings. However, after conversion, if you total your car (= verb), and your insurance company gives you the runaround (= noun), then you will have a double sense of the negative. Acronyms Some new words, known as acronyms, are formed from the initial letters of a set of other words. These can remain essentially 'alphabetisms' such as CD ('compact disk') or VCR ('video cassette recorder") where the pronunciation consists of the set of letters. More typically, acronyms are pronounced as single words, as in NATO, NASA or UNESCO. These examples have kept their capital letters, but many acronyms lose their capitals to become every- day terms such as laser (light amplification by stimulated emission of radiation"), radar ("radio detecting and ranging'), scuba ('self contained underwater breathing apparatus") and zip ('zone improvement plan") code. You might even hear talk of a snafu which is reputed to have its origins in *situation normal, all fouled up', though there is some dispute about the f- word in there. Names for organizations are often designed to have their acronym represent an appropriate term, as in 'mothers against drunk driving' (MADD) and 'women against rape' (WAR). Some new acronyms come into general use so quickly that many speakers do not think of their component meanings. Recent innovations in banking such as the ATM ("automatic teller machine') and the required PIN ('personal identification number') are regularly heard with one of their elements repeated, as in I sometimes forget my PIN number when I go to the ATM machine. Derivation In our list so far, we have not dealt with what is by far the most common word-formation process to be found in the production of new English words. This process is called derivation, and it is accomplished by means of a large number of small 'bits' of the English language which are not usually given separate listings in dictionaries. These small 'bits' are called affixes and a few examples are the elements un-, mis-, pre-, -ful, -less, -ish, -ism, -ness which appear in words like unhappy, misrepresent, prejudge, joyful, careless, boyish, terrorism and sadness. Prefixes and suffixes In the preceding group of words, it should be obvious that some affixes have to be added to the beginning of a word (e.g. un-). These are called prefixes. The other affix forms are added to the end of the word (e.g. -ish) and are called suffixes. All English words formed by this derivational process use either prefixes or suffixes, or both. Thus, mislead has a prefix, disrespectful has both a prefix and a suffix, and foolishness has two suffixes. Infixes There is a third type of affix, not normally to be found in English, but fairly common in some other languages. This is called an infix and, as the term sug- gests, it is an affix which is incorporated inside another word. It is possible to see the general principle at work in certain expressions, occasionally used in fortuitous or aggravating circumstances by emotionally aroused English speakers: Hallebloodylujah!, Absogoddamlutely! and Unfuckingbelievable!. In the movie Wish You Were Here, the main character expresses her aggravation (at another character's trying to contact her) by screaming Tell him I've gone to Singabloodypore! The expletive may even have an infixed element, as in godtripledammit!. We could view these 'inserted' forms as a special version of infixing. However, a much better set of examples can be provided from Kamhmu, a language spoken in South East Asia. These examples are taken from Merrifield et al. (1962): ("to drill") see-srnee ('a drill') ('to chisel') toh-trnoh ('a chisel') ("to eat with a spoon") hiip-hrniip ('a spoon") ('to tie') hoom-hroom ('a thing with which to tie" to verbs to form corresponding nouns. If this pattern is generally found in the language and you know that the form krnap is the Kamhmu word for 'tongs', then you should be able to work out what the corresponding verb 'to grasp with tongs' would be. It is kap. Multiple processes Although we have concentrated on each of these word-formation processes in isolation, it is possible to trace the operation of more than one process at work in the creation of a particular word. For example, the term deli seems to have become a common American English expression via a process of first 'borrowing' delicatessen (from German) and then 'clipping' that bor- rowed form. If you hear someone complain that problems with the project have snowballed, the final term can be noted as an example of 'compounding', whereby snow and ball have been combined to form the noun snowball, which has then undergone 'conversion' to be used as a verb. Forms which begin as 'acronyms' can also undergo other processes, as in the use of lase as a verb, the result of 'backformation' from laser. In the expression, waspish attitudes, the form WASP ("white Anglo-Saxon Protestant") has lost its capital letters and gained a suffix (-ish) in the 'derivation' process. An acronym that never seems to have had capital letters comes from 'young urban professional', plus the -ie suffix, as in hypocorism, to produce the word yuppie (first recorded in 1984). The formation of this new word, however, was helped by a quite different process, known simply as analogy, whereby words are formed to be similar in some way to existing words. Yuppie was made possible as a new word by the earlier existence of hippie and the other short-lived analogy yippie. The term yippie also had an acronym basis ("youth international party'), but was generally used for stu- dents protesting the Vietnam war in the United States. One joke has it that yippies just grew up to be yuppies. And the process continues. Another analogy, with the word yap ('making shrill noises'), has recently helped label some of those noisy young professionals as yappies. Many such forms can, of course, have a very brief life-span. Perhaps the generally accepted test of the 'arrival' of recently formed words in a language is their published appearance in a dictionary. However, even this may not occur without protests from some, as Noah Webster found when his first dictionary, published in 1806, was criticized for citing words like advocate and test as verbs, and for including such 'vulgar' words as advisory and presidential. It would seem that Noah had a keener sense than his critics of which new word-forms in the language were going to last. Morphology BAMBIFICATION: The mental conversion of flesh and blood living creatures into cartoon characters possessing bourgeois Judeo-Christian attitudes and morals. Douglas Coupland (1991) Throughout the preceding chapter, we approached the description of processes involved in word-formation as if the unit called the 'word' was a regular and easily identifiable form. This doesn't seem unreasonable when we look at a text of written English, since the 'words' in the text are, quite obviously, those sets of things marked in black with the bigger spaces separating them. Unfortunately, there are a number of problems with using this observation as the basis of an attempt to describe language in general, and individual linguistic forms in particular. Morphology In many languages, what appear to be single forms actually turn out to contain a large number of 'word-like' elements. For example, in Swahili (spoken throughout East Africa), the form nitakupenda conveys what, in English, would have to be represented as something like I will love you. Now, is the Swahili form a single word? If it is a 'word', then it seems to consist of a number of elements which, in English, turn up as separate 'words'. A very rough correspondence can be presented in the following way: It seems as if this Swahili 'word' is rather different from what we think of as an English 'word'. Yet, there clearly is some similarity between the both. Perhaps a better way of looking at linguistic forms in different languages would be to use this notion of 'elements' in the message, rather than to depend on identifying 'words'. The type of exercise we have just per- formed is an example of investigating forms in language, generally known as morphology. This term, which literally means 'the study of forms', was originally used in biology, but, since the middle of the nineteenth century, has also been used to describe that type of investigation which analyzes all those basic 'elements' which are used in a language. What we have been describing as 'elements' in the form of a linguistic message are more technically known as morphemes. Morphemes We do not actually have to go to other languages such as Swahili to discover that 'word- forms' may consist of a number of elements. We can recognize that English word-forms such as talks, talker, talked and talking must consist of one element talk, and a number of other elements such as -s, -er, -ed, -ing. All these elements are described as morphemes. The definition of a morpheme is "a minimal unit of meaning or grammatical function". Let's clarify this definition with some examples. We would say that the word reopened in the sentence The police reopened the investigation consists of three morphemes. One minimal unit of meaning is open, another minimal unit of meaning is re- (meaning "again"), and a minimal unit of grammatical function is-ed (indicating past tense). The word tourists also contains three morphemes. There is one minimal unit of meaning, four, another minimal unit of meaning -ist (meaning 'person who does something"), and a minimal unit of grammatical function -s (indicating plural). Free and bound morphemes From these two examples, we can make a broad distinction between two types of morphemes. There are free morphemes, that is, morphemes which can stand by themselves as single words, e.g. open and tour. There are also bound morphemes, that is, those which cannot normally stand alone, but which are typically attached to another form, e.g. re-,-ist,-ed, -s. You will recognize this last set as a group of what we have already described in Chapter 7 as affixes. So, all affixes in English are bound morphemes. The free morphemes can be generally considered as the set of separate English word- forms. When they are used with bound morphemes, the basic word-form involved is technically known as the stem. For example: It should be noted that this type of description is a partial simplification of the morphological facts of English. There are a number of English words in which the element which seems to be the 'stem" is not, in fact, a free morpheme. In words like receive, reduce, repeat we can recognize the bound morpheme re-, but the elements- ceive, -duce and -peat are clearly not free morphemes. There is still some disagreement over the proper characterization of these elements and you may encounter a variety of technical terms used to describe them. It may help to work with a simple distinction between those forms like -ceive and -duce as "bound stems' and other forms like dress and care as 'free stems". Free morphemes What we have described as free morphemes fall into two categories. The first category is that set of ordinary nouns, adjectives and verbs which we think of as the words which carry the 'content' of messages we convey. These free morphemes are called lexical morphemes and some examples are: boy, man, house, tiger, sad, long, yellow, sincere, open, look, follow, break. We can add new lexical morphemes to the language rather easily, so they are treated as an 'open' class of words. The other group of free morphemes are called functional morphemes. Examples are: and, but, when, because, on, near, above, in, the, that, it. This set consists largely of the functional words in the language such as conjunctions, prepositions, articles and pronouns. Because we almost never add new functional morphemes to the language, they are described as a "closed" class of words. Bound morphemes The set of affixes which fall into the 'bound' category can also be divided into two types. One type we have already considered in Chapter 7, the derivational morphemes. These are used to make new words in the language and are often used to make words of a different grammatical category from the stem. Thus, the addition of the derivational morpheme-ness changes the adjective good to the noun goodness. The noun care can become the adjectives careful or careless via the derivational morphemes -ful or -less. A list of derivational morphemes will include suffixes such as the -ish in foolish, the -ly in badly and the -ment in payment. It will also include prefixes such as re-, pre-, ex-, dis-, co-, un- and many more. The second set of bound morphemes contains what are called inflectional morphemes. These are not used to produce new words in the English language, but rather to indicate aspects of the grammatical function of a word. Inflectional morphemes are used to show if a word is plural or singular, if it is past tense or not, and if it is a comparative or possessive form. English has only eight inflectional morphemes, illustrated in the following: Let me tell you about Jim's two sisters. One likes to have fun and is always laughing. The other liked to study and has always taken things seriously. One is the loudest person in the house and the other is quieter than a mouse. From these examples, we can see that two of the inflections, -'s (possessive) and -s (plural) are attached to nouns. There are four attached to verbs, -s (3rd person present singular), -ing (present participle), -ed (past tense) and -en (past participle). There are two inflections, -est (superlative) and -er (comparative) attached to adjectives. Note that, in English, all inflectional morphemes listed here are suffixes. There is some variation in the form of these inflectional morphemes, with, for example, the possessive sometimes occurring as -s' (those boys' bags) and the past participle as -ed (they have finished). Derivational versus inflectional The difference between derivational and inflectional morphemes is worth emphasizing. An inflectional morpheme never changes the grammatical category of a word. For example, both old and older are adjectives. The -er inflection (from Old English -ra) simply creates a different version of the adjective. However, a derivational morpheme can change the grammatical category of a word. The verb teach becomes the noun teacher if we add the derivational morpheme-er (from Old English -ere). So, the suffix form -er can be an inflectional morpheme as part of an adjective and also a distinct derivational morpheme as part of a noun. Just because they (-er) look the same doesn't mean they do the same kind of work. In both cases, they are bound morphemes. different languages, have not been fully resolved by linguists. The solutions to these problems are clearer in some cases than in others. The relationship between law and legal is a reflection of the historical influence of other languages on English word-forms. The modern form law is a result of a borrowing into Old English from Old Norse, over 1,000 years ago. The modern form legal is a borrowing from the Latin form legalis ("of the law"). Consequently, there is no derivational relationship between the two forms in English, nor between the noun mouth (an Old English form) and the adjective oral (a Latin borrowing). It has been pointed out that an extremely large number of English forms owe their morphological patterning to languages like Latin and Greek. Consequently, a full description of English morphology will have to take account of both historical influences and the effect of borrowed elements. Morphs and allomorphs The solution to other problems remains controversial. One way to treat differences in inflectional morphemes is by proposing variation in morphological realization rules. In order to do this, we draw an analogy with some processes already noted in phonology (Chapter 6). If we consider 'phones' as the actual phonetic realization of 'phonemes', then we can propose morphs as the actual forms used to realize morphemes. Thus, the form cat is a single morph realizing a lexical morpheme. The form cars consists of two morphs, realizing a lexical morpheme and an inflectional morpheme ('plural'). Just as we noted that there were 'allophones' of a particular phoneme, then we can recognize allomorphs of a particular morpheme. Take the morpheme 'plural'. Note that it can be attached to a number of lexical morphemes to produce structures like 'cat + plural', 'sheep + plural", and 'man + plural'. Now, the actual forms of the morphs which result from the single morpheme 'plural' turn out to be different. Yet they are all allomorphs of the one morpheme. It has been suggested, for example, that one allomorph of 'plural' is a zero-morph, and the plural form of sheep is actually 'sheep + ". Otherwise, those so-called 'irregular' forms of plurals and past tenses in English are described as having individual morphological realization rules. Thus, 'man + plural' or 'go + past', as analyses at the morpheme- level, are realized as men and went at the morph-level. Syntax After a lecture on cosmology and the structure of the solar system, William James was accosted by a little old lady who told him that his view of the earth rotating round the sun was wrong. "I've got a better theory," said the little old lady. "And what is that, madam?" inquired James politely. "That we live on a crust of earth which is on the back of a giant turtle." "If your theory is correct, madam,' he asked, "what does this turtle stand on?" "You're a very clever man, Mr. James, and that's a very good question," replied the little old lady, "but I have an answer to it. And it's this: the first tur- tie stands on the back of a second, far larger, turtle, who stands directly under him." "But what does this second turtle stand on?" persisted James patiently. To this, the little old lady crowed triumphantly, "It's no use, Mr. James, it's turtles all the way down." Adapted from J. R. Ross (1967) In the course of the preceding chapter, we moved from a consideration of general grammatical categories and relations to specific methods of describing the structure of phrases and sentences. If we concentrate on the structure and ordering of components within a sentence, we are studying what is technically known as the syntax of a language. The word 'syntax' came originally from Greek and literally meant 'a setting out together' or 'arrangement'. In earlier approaches to the description of syntax, as we saw in Chapter 9, there was an attempt to produce an accurate analysis of the sequence or the ordering 'arrangement' of elements in the linear structure of the sentence. While this remains a major goal of syntactic description, more recent work in syntax has taken a rather different approach in accounting for the ‘arrangements’ we observe in the structure of sentences. Generative grammar Since the 1950s, particularly developing from the work of the American linguist Noam Chomsky, there have been attempts to produce a particular type of grammar which would have a very explicit system of rules specifying what combinations of basic elements would result in well-formed sentences. (Let us emphasize the word "attempts" here, since no fully worked-out grammar of this or any other type yet exists.) This explicit system of rules, it was pro- posed, would have much in common with the types of rules found in mathematics. Indeed, a definitive early statement in Chomsky's first major work betrays this essentially mathematical view of language: "I will consider a language to be a set (finite or infinite) of sentences" (Chomsky, 1957:13). This mathematical point of view helps to explain the meaning of the term generative, which is used to describe this type of grammar. If you have an algebraic expression like 3x+2y, and you can give x and y the value of any whole number, then that simple algebraic expression can generate an end- less set of values, by following the simple rules of arithmetic. When x = 5 and y = 10, the result is 35. When x = 2 and y = 1, the result is 8. These results will follow directly from applying the explicit rules. The endless set of such results is 'generated' by the operation of the explicitly formalized rules. If the sentences of a language can be seen as a comparable set, then there must be a set of explicit rules which yield those sentences. Such a set of explicit rules is a generative grammar. Some properties of the grammar A grammar of this type must have a number of properties, which can be described in the following terms. The grammar will generate all the well- formed syntactic structures (e.g. sentences) of the language and fail to generate any ill-formed structures. This is the 'all and only' criterion (i.e. all the grammatical sentences and only the grammatical sentences). The grammar will have a finite (i.e. limited) number of rules, but will be capable of generating an infinite number of well-formed structures. In this way, the productivity of language (i.e. the creation of totally novel, yet grammatical, sentences) would be captured within the grammar. The rules of this grammar will also need the crucial property of recursion, that is, the capacity to be applied more than once in generating a structure. For example, whatever rule yields the component that chased the cat in the sentence This is the dog that chased the cat, will have to be applied again to get that killed the rat and any other similar structure which could continue the sentence This is the dog that chased the cat that killed the rat.... You can do the same recursive thing with phrases specifying a location, beginning with, The book was on the table. This sentence tells us where the book was. Where was the table? Near the window? Okay, where was the window? In the hallway? Okay. Putting this type of recursive effect into a single sentence will lead us to: The book was on the table near the window in the hallway beside the.... There is, in principle, no end to the recursion which would yield ever-longer versions of this sentence, and the grammar must provide for this fact. Basically, the grammar will have to capture the fact that a sentence can have another sentence inside it, or a phrase can have another phrase of the same type inside it. (Recursion is not only to be found in descriptions of sentence structure. It is an essential part of the little old lady's view of the role of turtles in cosmic structure, as quoted at the beginning of this chapter.) This grammar should also be capable of revealing the basis of two other phenomena: first, how some superficially distinct sentences are closely related, and second, how some superficially similar sentences are in fact dis tinct. We need some exemplification for these points. Deep and surface structure Two superficially distinct sentence structures would be, for example, Charlie broke the window and The window was broken by Charlie. In traditional terminology, the first is an active sentence and the second is passive. The dis- tinction between them, it can be claimed, is a difference in their surface structure, that is, the syntactic form they take as actual English sentences. However, this difference in superficial form disguises the fact that the two sentences are very closely related, even identical, at some less 'superficial' level. This other 'underlying' level, where the basic components shared by the two sentences would be represented, has been called their deep structure. The deep structure is an abstract level of structural organization in which all the elements determining structural interpretation are represented. So, the grammar must be capable of showing how a single underlying abstract representation can become different surface structures. Structural ambiguity On the second point noted above, let us say that we had two distinct deep structures expressing, on the one hand, the fact that 'Annie had an umbrella and she whacked a man with it'; and, on the other hand, that 'Annie whacked a man and the man happened to be carrying an umbrella.' Now, these two different concepts can, in fact, be expressed in the same surface structure form: Annie whacked a man with an umbrella. This sentence is structurally ambiguous. It has two different underlying interpretations which would be represented differently in the deep structure. Groucho Marx knew how to have fun with structural ambiguity. In the film 'Animal Crackers', he first says One morning I shot an elephant in my pyjamas, then follows it with How he got into my pyjamas I'll never know. In the non-funny interpretation, the structural unit in my pyjamas is an addition, attached to the end of the structural unit I once shot an elephant. In the alternative (ho, ho) interpretation, the structural unit an elephant in my pyjamas is a necessary internal part of a structure that would otherwise be incomplete, I once shot.... Phrases can also be structurally ambiguous, as when you come across an expression like old men and women. The underlying interpretation can be either old men plus old women or old men plus women (no age specified). The grammar will have to be capable of showing the structural distinction between these underlying representations. Different approaches We have considered some of the requirements which would have to be met by a complete syntactic description of a language. However, this area of linguistic investigation is notorious for giving rise to very different approaches to producing that description. For some, the only relevant issues are syntactic ones, that is, how to describe structure, independently of 'meaning' considerations. For others, the 'meaning component' is primary. In some later versions of generative grammar, the level of deep structure is essentially taken over by a 'meaning' or semantic interpretation which is assigned a structural or syntactic form in its surface realization. (We shall explore 'meaning' issues in Chapters II and 12.) Unfortunately, almost everything involved in the analysis of generative grammar remains controversial. There continue to be many different approaches among those who claim to analyze language in terms of generative grammar, and many more among those who are critical of the whole system. Rather than explore controversies, let us look at some of the really basic features of the original analytic approach and see how it is all supposed to work. First, we need to get the symbols straightened out. Symbols used in syntactic description We have already introduced some symbols (in Chapter 9) which are quite easily understood as abbreviations for the grammatical categories involved. Examples are 'S' (= sentence), 'N' (= noun), 'Art' (= article) and so on. We need to introduce three more symbols which are commonly used. The first of these is in the form of an arrow, and it can be interpreted as 'consists of". It will typically occur in the following format: NP-ArtN This is simply a shorthand way of saying that a noun phrase (e.g. the book) consists of an article (the) and a noun (book). The second symbol used is in the form of parentheses, or round brackets- (). Whatever occurs inside these brackets will be treated as an optional constituent. Perhaps an example will make this clear. You can describe an object as the book, or as the green book. We can say that both the book and the green book are examples of the category, noun phrase. In order for a noun phrase to occur in English, you may require an article (the) and a noun (book), but the inclusion of an adjective (green) is optional. You can include an adjective, but it isn't obligatory. We can capture this aspect of English syntax in the following way: NP→ Art (Adj) N This shorthand notation expresses the idea that a noun phrase consists of an obligatory article and an obligatory noun, but may also include an adjective in a specific position. The adjective is optional. The third symbol used is in the form of braces, or curly brackets - {} These indicate that only one of the elements enclosed within the brackets must be selected. They are used when there is a choice from two or more constituents. For example, we have already noted, in Chapter 9, that a noun phrase can consist of an expression like the woman (Art N), or she (pro- noun), or Cathy (proper noun). We can, of course, write three single rules, as shown on the left below, but it is more succinct to write one rule, as shown on the right below, which incorporates exactly the same information: NP→ ArtN NP→ pronoun NP→ Art N pronoun NP proper noun proper noun It is important to remember that, although there are three constituents in these curly brackets, only one of them can be selected on any occasion. We can now present a list of symbols and abbreviations commonly found in syntactic descriptions: S sentence N noun Pro pronoun PN proper noun V verb Adj adjective Art article Adv adverb Prep preposition NP noun phrase VP verb phrase PP prepositional phrase = 'consists of = 'ungrammatical sequence' () 'optional constituent' {)= 'one and only one of these constituents must be selected' Labeled tree diagrams In Chapter 9, we considered ways of describing the structure of sentences that (basically) concentrated on the linear sequence of constituents. It is, of course, possible to show the same sequence as, in a more explicit way, 'hierarchically' organized. So, to bring out the hierarchical organization of the labeled and bracketed constituents shown on the left below, we can show the same information in the form of a tree diagram, as on the right below: NP NP Art N Art N [The] [monkey] The monkey This type of tree-diagram representation contains all the grammatical information found in the other analyses, but also shows more explicitly the fact that there are different levels in the analysis. That is, there is a level of analysis at which a constituent such as NP is represented and a different, lower level at which a constituent such as N is represented. Here's how a whole sentence would look in a tree diagram: S NP Art VP NP Art N The monkey ate a banana If you start at the top of this tree diagram, you are starting with a sentence (S) and then dividing the sentence into two constituents (NP and VP). In turn, the NP constituent is divided into two constituents (Art and N). Finally, one word is selected which fits the label Art (the), and another which fits N (monkey). Phrase structure rules We can view this tree-diagram format in two different ways. In one way, we can simply treat it as a static representation of the structure of the sentence at the bottom of the diagram. We could propose that, for every single sentence in English, a tree diagram of this type could be drawn. The alternative view is to treat the diagram as a 'dynamic' format, in the sense that it represents a way of 'generating' not only that one sentence, but a very large num- ber of sentences with similar structures. This alternative view is very appealing since it should enable us to generate a large number of sentences with only a small number of rules. These 'rules' are usually called phrase structure rules, and they present the information of the tree diagram in an alternative format. So, instead of the diagram form on the left below, we can use the notation shown on the right below: NP VP S→ NP VP The rule is then read as: "a sentence consists of a noun phrase followed by a verb phrase". In addition to rules of this type which generate structures, we can also have lexical rules which indicate the words to be used for constituents such as N. For example: N→ (boy. girl, dog.... This means that N is rewritten as boy, or girl, or dog. We can create a set of extremely simple (and necessarily incomplete) phrase structure rules which can be used to generate a large number of English sentences: S→ NP VP Art (Adj) N NP- PN "} VP→ V NP (PP) (Adv) PP → Prep NP N{boy, girl, dog) PN (George, Mary) (a,the) Art Adj → {small, crazy} V→ (saw, followed, helped} Prep -> {with, near} Adv→ (yesterday, recently} These rules will generate the grammatical sentences shown below as (1) to (7), but will not yield the ungrammatical sentences shown as (8) to (10): 1. The girl followed the boy. 2. A boy helped the dog. 3. The dog saw a girl. 4. Mary helped George recently. 5. George saw a dog yesterday. 6. A small dog followed Mary. 7. The small boy saw George with a crazy dog recently. 8. "Boy the Mary saw. 9. *Helped a girl. 10. "Small dog with girl. This small set of rules is a good start on creating a phrase structure grammar of English, but we still have not incorporated recursion. Back to recursion The phrase structure rules, as presented, have no recursive elements. Each time we rewrote a symbol from the left, we did not include that symbol on the right side of any arrow. We have to be able to repeat some symbols on the right side of the arrow. That is the essence of recursion. We need, for example, to have sentences included within other sentences. We know that Mary helped George is a sentence. We also know that Cathy thought Mary helped George is a sentence. And, being tediously recursive, we know that John said Cathy thought Mary helped George is a sentence. In order to capture these structures in our rules, we need to add V→ (said, thought) and PN (Cathy, John) to our lexical rules. We also need to add a crucial recursive rule that says: VP→V S. With these minor additions, we can now represent the structure of a more complex sentence. NP VP V NP VP V S NP VP NP John said Cathy thought George helped Mary In principle, there is no end to the recursion of sentence structures of this type in the English language and our rule (VP-V S) represents that fact. Transformational rules One other feature of our phrase structure rules is that they will generate all sentences with fairly fixed word order to the constituents. For example, adverbs will always come at the end of their sentences if we follow the rules we have just illustrated. That is fine for generating the first sentence below, but how would we get the second sentence? (1) George helped Mary yesterday. (ii) Yesterday George helped Mary. We can think of the yesterday element as having been 'moved to the beginning of the sentence in (ii). In order to do this, we need a set of rules which will change or move constituents in the structures derived from the phrase structure rules. These are called transformational rules. Essentially what they do is take a "branch' of the 'tree' away from one part of the tree diagram, and attach it to a different part. Here is an example of a movement transformation: NP VP Adv NP VP NP Adv V NP (George helped Mary yesterday) (Yesterday George helped Mary) We would, of course, specify which constituents can be moved, from where and to where. One of the best arguments for having transformational rules involves what seems to be the movement of a very small element in English sentence structure. We recognize that the following two sentences have a great deal in common: (i) Doobie picked up the magazine. (i) Doobie picked the magazine up. These sentences contain a verb-particle construction (verb= pick; particle= up) which can be symbolized as: V→Vb part. It is clear that the particle can be separated from the verb and 'moved to the end of the sentence. A constituent structure analysis, as described in Chapter 9, would have some difficulty accommodating this type of structure. A phrase structure analysis would have to create two distinct tree diagrams. Yet, we intuitively recognize that these two sentences must come from a single underlying source. Let us propose a single tree diagram source which produces a string of elements like: NP Verb Particle NP. Under circumstances like these, let us then propose the optional transformation called 'Particle Movement', Phrase Structure Tree NP VP NP Vb part Art N Particle Doobie picked up the magazine Movement: Doobie picked the magazine up which takes that structural description and yields the structural change to: NP Verb NP Particle. By using this simple transformational rule, we have provided the means for explicitly relating the two structures in sentences (i) and (ii) above as 'surface' variations of a single underlying structure. It may not seem much, but this type of transformational analysis solved a number of tricky problems for previous syntactic descriptions. There is, of course, much more involved in generative syntax and other methods of syntactic description. (We have barely scratched the surface structures.) However, having explored some of the basic issues in the syntactic description of language, we must move on, as historically the generative linguists had to do, to come to terms with the place of 'meaning' in linguistic description. This leads us, in the following chapter, to a consideration of the role of semantics.