матковська

118 Cognitive Exploration of Language and Linguistics

(r)and Africa, the idea (r)of it. This is often called the “intrusive r”. You sometimes get an “intrusive r” within a word, as in withdraw(r)ing, saw(r)ing.

5.6.2 Stress, tone, intonation

Stress is a property of syllables. A stressed syllable is produced with more energy than an unstressed syllable. Stressed syllables, therefore, are more “prominent” than unstressed syllables. They are typically longer and louder than unstressed syllables, and are produced with greater clarity. Unstressed syllables tend to be short, and are often pronounced rather indistinctly.

Each word in English has a distinctive word stress pattern. Some words are, amongst others, distinguished by stress location: SUBject vs. subJECT. The phonetic symbol for stress is [Á] placed before the stressed syllable: [Ás%bdŠekt], [s6bÁdŠekt].

Within an utterance, stress can highlight the important words, often by suggesting a contrast. Compare: HE didn’t do that, He DIDN’T do that, He didn’t DO that, He didn’t do THAT. It is not di cult to construct contexts in which each of these variants would be appropriate.

Tone is also a property of syllables. In a tone language like Chinese, most syllables are associated with a characteristic pitch melody. This means that the same syllable spoken with a di erent tone each time has a di erent meaning. The pitch melody assigned to the syllable is just as much a part of the word as is the phonemic structure of the syllable.

Intonation is the melody superimposed on an utterance. Intonation is of importance in English for signalling the function of an utterance (e.g. as a statement or question), and for expressing speaker attitudes (see Chapter 4.4.1 and 4.4.2). It may be very interesting to compare the many intonations that can be associated with really.

5.7 Sounds in context

It is not su cient to study sounds in isolation since sounds may change under the influence of other sounds when words are combined with other words. In the longer units of word groups or sentences, the sounds of single words undergo massive changes such as linking, elision, assimilation etc.

Chapter 5. The sounds of language 119

5.7.1 “Linking”

One kind of change involves adding linking elements at the boundaries of words. Examples from English are the linking and “intrusive r”, discussed in the preceding section.

5.7.2 Elision

Sounds are often omitted in the stream of speech, especially in informal speaking styles. This is known as elision. Elisions should not be thought of as “careless”, or “lazy”. On the contrary, not to use elided pronunciations in relaxed, informal speech could be perceived as pedantic.

In careful speech, library would have three syllables. In informal speech, it could have only two [laibri]. Likewise, ordinary has four syllables only in very careful speech.

Consonant clusters are often the target of elision, i.e. one or more consonants in a cluster are elided. The elided consonants are nearly always alveolar or dental.

Clothes in careful speech is [kl6uðz]. In informal speech it is [kl6uz]. This pronunciation is especially likely if the next word begins with a consonant, as in clothes cupboard.

And typically loses the final stop, especially if the following word begins with a consonant: you and me [ju 6n mi].

Next and last generally lose their final /t/ before a word beginning with a consonant: last night [las nait].

/h/ is generally elided in unstressed syllables. In I saw HIM (with stress on him), the /h/ is pronounced. If him is unstressed, the /h/ is elided: I SAW him [ai Ásf Im] or [ai Ásf 6m]. It is even possible for an “intrusive r” to appear in the phrase: [ai sf rIm].

5.7.3 Assimilation

Assimilation is a process whereby one sound causes an adjacent sound to be “more similar” to itself. Assimilation can be progressive (a sound influences the following sound), or retrogressive (a sound influences a preceding sound).

Progressive assimilation is illustrated by the alternative pronunciations of the plural morpheme. After a voiceless consonant, plural “s” is voiceless: cats [kæts]. After a voiced sound (either consonant or vowel), plural “s” is voiced:

120 Cognitive Exploration of Language and Linguistics

dogs [df:z], bees [biz]. Further examples of progressive assimilation are the devoicing of the normally voiced [l, r] when these sounds occur after a voiceless consonant in a syllable-initial cluster: please, pray [pliz,® prei]® .

Retrogressive assimilations are frequent in English. News has a final voiced [z]. Yet in newspaper voiceless [p] causes preceding [z] to become [s]: [njuspeip6].

The above examples illustrate voicing assimilation, i.e. the voice/ voicelessness of a segment “spreads” into a neighbouring segment.

Place assimilation is when the place of articulation of a consonant spreads into a neighbouring consonant. Retrogressive place assimilation is frequent in English. Thus, good boy may be spoken as [:~b bfi], good girl as [:~: :8l]. The nasal in the negative prefixes un- and in- often assimilates to the place of articulation of a following consonant. In unbelievable the negative prefix is followed by a bilabial, and may be pronounced [6m], while in unconscious it is followed by a velar, and may be pronounced [6]]. In unfavourable one might get the labiodental nasal [X].

Nasal assimilation occurs when one segment takes on the nasality of a neighbouring segment. Vowels often nasalize before a nasal consonant: can’t [kãnt].

Assimilation can be total, i.e. a sound can become identical to its neighbour. By retrogressive voicing assimilation, is Sam [Iz sæm] becomes [Is sæm].

Some assimilations (e.g. the devoicing of /z/ in newspaper) are obligatory within word boundaries, often, however, they are optional, and tend to be more frequent the more informal and relaxed the speaking style.

Assimilation can sometimes appear to change the phonemic structure of a word. In the example good boy, the final [d] of good is changed to [b], i.e. an allophone of the /d/ phoneme has been replaced by an allophone of a di erent phoneme, i.e. /b/. In other cases, assimilation replaces one allophone of a phoneme by another allophone of the same phoneme, as when the [l] of play becomes voiceless. Although we cannot pursue this matter here, facts of this nature have led some linguists to question the theoretical status of the phoneme, as traditionally defined. For example, given that comfort is pronounced [k%Xf6t], and that the use of [X] represents place assimilation to the following [f], to which phoneme should [X] be assigned, to /m/ or to /n/?

5.7.4 Palatalization

Palatalization is a rather common process in which the palatal glide [j] causes a preceding obstruent to be articulated in the palatal region.

Chapter 5. The sounds of language 123

of accent, i.e. the regional or social di erences in pronunciation. Diphthongs are combinations of two vowels in one syllable. Since di erent sounds may be variants of one and the same phoneme, this category is of a psychological rather than physical nature. It is what in a given language is considered to be meaningdiscriminating. Two di erent sounds are two di erent phonemes if they cause a di erence in meaning as in a minimal pair like pear and bear. Di erent sounds that do not create a di erence in meaning like the [th] in top and the [t] in stop are allophones, which in this case occur in complementary distribution. This means that they are bound to a given position: [th] can only occur in initial position, [t] in non-initial position. If the context does not play a role, allophones are in free variation. We must also distinguish between a phonetic transcription, describing all the allophones of a phoneme, and a phonemic transcription, only taking care of the phonemes.

In addition to speech sounds, also larger entities such as the syllable, stress, tone and intonation are important. A phonological syllable consists of a vowel(- like) sonorant core, i.e. a vowel or diphthong, represented as V and optionally a consonant (C). Languages di er very strongly in their patterns of syllable structure. The position which a phoneme can have in a syllable is known as its distribution. In English many consonants can be combined into consonant clusters. If a phoneme is not pronounced in a given position, e.g. /r/ source in British English, we may have homophones as with source and sauce. Also stress and tone are properties of syllables, whereas intonation is the melody superimposed on an utterance. Syllables are grouped into words and therefore word stress is needed to mark the main syllable. The flow of sentences causes the individual words to be adapted in various ways. Linking elements may have to be added between words ending and beginning with a vowel, elision may be needed, and especially consonants may have to be adapted to each other, which is known as assimilation. We distinguish between two types of voice assimilation: progressive assimilation as in dogs /d#gz/ and retrogressive assimilation as in hotdog /h#d#g/. Place assimilation occurs in good boy /g~bfi/. Other processes of adaptation to the speech stream are palatalization, vowel reduction and the use of weak forms in unaccented syllables and strong forms in accented ones.

124 Cognitive Exploration of Language and Linguistics

5.9 Further reading

Good introductions on phonetics are Ladefoged (1993) and Catford (1990) and on phonology Katamba (1982) and Lass (1991). Specific treatments of British English are Gimson (1989) and Giegerich (1992). Cognitive approaches are Nathan (1994, 1996, 1999) and Taylor (2002).

Assignments

1.The underlined segments in the following words represent di¬erent pronunciations. Group the segments accordingly and find the appropriate terms to characterize the di¬erences.

a.thin – then – mother – cloth – clothes

b.sees – seize – cease – seizes – ceases – house – houses

2.Compare the written forms and the pronunciations of the following words and (i) say whether they rhyme or not, (ii) write the words in phonemic transcription

a.horse – worse

b.heart – heard – beard

c.lumber – plumber

d.tough – bough – dough – hiccough

e.broom – brook – brooch

f.tomb – bomb – womb

g.roll – doll

h.golf – wolf

i.seize – sieve

j.kind – kindle

3.a. Do you think it would be good idea if English spelling more closely represented

pronunciation?

b.Can you see any disadvantages if English spelling were 100% phonemic?

c.Comment on Mark Twain’s plans for the improvement of English spelling: For example, in Year 1 that useless letter “c” would be dropped to be replased either by “k” or “s”, and likewise, “x” would no longer be part of the alphabet. The only kase in which “c” would be retained would be the “ch” formation,

which will be dealt with later. Year 2 might reform “w” spelling, so that “which” and “one” would take the same konsonant, wile Year 3 might well abolish “y” replasing it with “i” and Iear 4 might fiks the “g/j” anomali wonse and for all.

Chapter 5. The sounds of language 125

Jenerally, then, the improvement would kontinue iear bai iear with Iear 5 doing awai with useless double konsonants, and Iears 6–12 or so modifaiing vowlz and rimeining voist and unvoist konsonants. Bai Iear 15 or sou, it wud fainali bi posibl tu meik ius ov thi ridandant letez “c”, “y” and “x” — bai now jast a memori in the maindz ov ould doderez — tu riplais “ch”, “sh”, and “th” rispektivli.

Fainali, xen, aafte sam 20 iers ov orxogrefkl riform, wi wud hev a lojikl, kohirnt speling in ius xrewawt xe Ingliy-spiking werld.

4.Voicing

a.It is not possible to produce voiced sounds while whispering. Why not? Consequently, a whispered utterance of the word hand ought to be virtually indistinguishable from a whispered utterance of the word and. Why?

Try it, and see!

b.Is it possible to distinguish the words Sue and zoo, cease and seize, do and too, in whisper? If you find that it is possible (which you should!), what explanation can you o¬er?

(Hint: [d] and [t], [z] and [s], are not only distinguished by presence vs. absence of voice, but by other features as well. What are these?)

5.Consonants

a.The first sound of yes is very similar, phonetically, to the final sound of say. Yet you would probably want to say that the first sound of yes is a consonant, and the final sound of say is a vowel. Why?

b.Try to isolate the “k” sound in keen and the “k” sound in cool. How do they di¬er? Say the sounds independently of the words in which they occur.

6.Phonemes and allophones

If you consider the environments in which they occur, you will discover that “h”- sounds and []] are in complementary distribution in English. State the environments in which these sounds occur as precisely as possible. Would you want to say that “h”- sounds and the velar nasal are allophones of one and the same phoneme? Why not? What additional criteria, over and above the fact of complementary distribution, need to be invoked in identifying the phonemes of a language?

7.Vowels

a.Make a pure (i.e. unvarying) “i”-type vowel, as in see. Make the vowel as front as possible, and as high as possible. Now make a pure “u”-type vowel, as in too. The [u] vowel should be as back as possible, and as high as possible, and with prominent lip rounding.

6Language, culture and meaning

Cross-cultural semantics

6.0Overview

The previous chapters have shown repeatedly that linguistic conceptualization may be radically di erent in various, even closely related languages. This applies to concepts expressed through all aspects of linguistic structure, i.e. the lexicon, morphology, syntax, and even in phonology, at the level of tone and intonation.

This chapter will look into cross-linguistic semantic di erences in a systematic way. We will present a method for pinpointing semantic distinctions and for exploring their cultural relevance. A key question is whether di erences in linguistic conceptualization play a central role in language and thought or whether they are rather marginal. Both positions have been advocated. The first is known as linguistic relativity, in its extreme form as linguistic determinism. The second is known as universalism and holds that all people all over the world basically think in the same way. This chapter proposes a compromise between the extremes: Most linguistic concepts are indeed language-specific, but there is also a small number of universal linguistic concepts which occur in all languages. These universal concepts can be used as a “neutral” basis for paraphrasing the huge variety of language-specific and culture-specific concepts in the languages of the world. This is illustrated firstly for lexical concepts, then for grammatical concepts, and finally for the cultural norms of behaviour which underlie people’s behaviour in di erent cultures.

6.1 Introduction: Linguistic relativity and universalism

A key question is whether language influences thought or thought influences language. Both positions have been advocated. The first is known as linguistic