Cambridge Introducing Phonology 1-What is phonology

Cambridge Books Online http://ebooks.cambridge.org/ Introducing Phonology David Odden Book DOI: http://dx.doi.org/10.1017/CBO9780511808869 Online ISBN: 9780511808869 Hardback ISBN: 9780521826693 Paperback ISBN: 9780521534048 Chapter 1 - What is phonology? pp. 1-18 Chapter DOI: http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge University Press CHAPTER What is phonology? 1 This chapter introduces phonology, the study of the sound systems of language. Its key objective is to: ◆ introduce the notion of phonological rule ◆ explain the nature of sound as a physical phenomenon ◆ highlight the tradeoff between accuracy and usefulness in representing sound ◆ distinguish between phonetics and phonology ◆ contrast the continuous and discrete aspects of linguistic sounds ◆ introduce the notion of “sound as cognitive symbol” PREVIEW sound symbol transcription grammar continuous nature of speech accuracy KEY TERMS Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 2 INTRODUCING PHONOLOGY Phonology is one of the core fields that composes the discipline of lin- guistics, which is defined as the scientific study of language structure. One way to understand what the subject matter of phonology is, is to con- trast it with other fields within linguistics. A very brief explanation is that phonology is the study of sound structure in language, which is dif- ferent from the study of sentence structure (syntax) or word structure (morphology), or how languages change over time (historical linguistics). This definition is very simple, and also inadequate. An important feature of the structure of a sentence is how it is pronounced – its sound struc- ture. The pronunciation of a given word is also a fundamental part of the structure of the word. And certainly the principles of pronunciation in a language are subject to change over time. So the study of phonology even- tually touches on other domains of linguistics. An important question is how phonology differs from the closely related discipline of phonetics. Making a principled separation between phonetics and phonology is difficult – just as it is difficult to make a principled separation between physics and chemistry, or sociology and anthropology. A common characterization of the difference between pho- netics and phonology is that phonetics deals with “actual” physical sounds as they are manifested in human speech, and concentrates on acoustic waveforms, formant values, measurements of duration meas- ured in milliseconds, of amplitude and frequency, or in the physical prin- ciples underlying the production of sounds, which involves the study of resonances and the study of the muscles and other articulatory struc- tures used to produce physical sounds. On the other hand, phonology, it is said, is an abstract cognitive system dealing with rules in a mental grammar: principles of subconscious “thought” as they relate to lan- guage sound. Yet once we look into the central questions of phonology in greater depth, we will find that the boundaries between the disciplines of phonetics and phonology are not entirely clear-cut. As research in both of these fields has progressed, it has become apparent that a better understanding of many issues in phonology requires that you bring pho- netics into consideration, just as a phonological analysis is a prerequisite for any phonetic study of language. As a step towards understanding what phonology is, and especially how it differs from phonetics, we will consider some specific aspects of sound structure that would be part of a phonological analysis. The point which is most important to appreciate at this moment is that the “sounds” which phonology is concerned with are symbolic sounds – they are cog- nitive abstractions, which represent but are not the same as physical sounds. The sounds of a language. One aspect of phonology considers what the “sounds” of a language are. We would want to take note in a description 1.1 Concerns of phonology Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 of the phonology of English that we lack a particular vowel that exists in German in words like schön ‘beautiful,’ a vowel which is also found in French (spelled eu, as in jeune ‘young’), or Norwegian (øl ‘beer’). Similarly, the consonant spelled th in English thing, path does exist in English (as well as in Icelandic where it is spelled with the letter þ, or Modern Greek where it is spelled with �, or Saami where it is spelled t�), but this sound does not occur in German or French, and it is not used in Latin American Spanish, although it does occur in Continental Spanish in words such as cerveza ‘beer,’ where by the spelling conventions of Spanish, the letters c and z represent the same sound as the one spelled � (in Greek) or th (in English). Rules for combining sounds. Another aspect of language sound which a phonological analysis would take account of is that in any given lan- guage, certain combinations of sounds are allowed, but other combina- tions are systematically impossible. The fact that English has the words brick, break, bridge, bread is a clear indication that there is no restriction against having words begin with the consonant sequence br; besides these words, one can think of many more words beginning with br such as bribe, brow and so on. Similarly, there are many words which begin with bl, such as blue, blatant, blast, blend, blink, showing that there is no rule against words beginning with bl. It is also a fact that there is no word �blick1 in English, even though the similar words blink, brick do exist. The question is, why is there no word �blick in English? The best explanation for the nonexistence of this word is simply that it is an accidental gap – not every logically possible combination of sounds which follows the rules of English phonology is found as an actual word of the language. Native speakers of English have the intuition that while blick is not actu- ally a word of English, it is a theoretically possible word of English, and such a word might easily enter the language, for example via the intro- duction of a new brand of detergent. Fifty years ago the English language did not have any word pronounced bick, but based on the existence of words like big and pick, that word would certainly have been included in the set of nonexistent but theoretically allowed words of English. Contemporary English, of course, actually does contain that word – spelled Bic – which is a type of pen. While the nonexistence of blick in English is accidental, the exclusion from English of many other imaginable but nonexistent words is based on a principled restriction of the language. While there are words that begin with sn like snake, snip and snort, there are no words beginning with bn, and thus �bnick, �bnark, �bniddle are not words of English. There simply are no words in English which begin with bn. Moreover, native speakers of English have a clear intuition that hypothetical �bnick, �bnark, �bniddle could not be words of English. Similarly, there are no words in English which are pronounced with pn at the beginning, a fact which is not only demonstrated by the systematic lack of words such as �pnark, �pnig, �pnilge, What is phonology? 3 1 The asterisk is used to indicate that a given word is non-existent or wrong. Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 but also by the fact that the word spelled pneumonia which derives from Ancient Greek (a language which does allow such consonant combina- tions) is pronounced without p. A description of the phonology of English would then provide a basis for characterizing such restrictions on sequences of sounds. Variations in pronunciation. In addition to providing an account of pos- sible versus impossible words in a language, a phonological analysis will explain other general patterns in the pronunciation of words. For exam- ple, there is a very general rule of English phonology which dictates that the plural suffix on nouns will be pronounced as [�z], represented in spelling as es, when the preceding consonant is one of a certain set of consonants including [�] (spelled sh) as in bushes, [�] (spelled as ch) as in churches, and [ j�] (spelled j, ge, dge) as in cages, bridges. This pattern of pro- nunciation is not limited to the plural, so despite the difference in spelling, the possessive suffix s2 is also subject to the same rules of pro- nunciation: thus, plural bushes is pronounced the same as the possessive bush’s, and plural churches is pronounced the same as possessive church’s. This is the sense in which phonology is about the sounds of language. From the phonological perspective, a “sound” is a specific unit which com- bines with other such specific units, and which represent physical sounds. Phonetics, on the other hand, is about the concrete, instrumentally meas- urable physical properties and production of these cognitive speech sounds. That being the case, we must ask a very basic question about pho- netics (one which we also raise about phonology). Given that phonetics and phonology both study “sound” in language, what are sounds, and how does one represent the sounds of languages? The question of the physical reality of an object, and how to represent the object, is central in any science. If we have no understanding of the physical reality, we have no way of talking meaningfully about it. Before deciding how to represent a sound, we need to first consider what a sound is. To answer this question, we will look at two basic aspects of speech sounds as they are studied in phonetics, namely acoustics which is the study of the properties of the physical sound wave that we hear, and articulation, which is the study of how to modify the shape of the vocal tract, thereby producing a certain acoustic output (sound). 1.2.1 Acoustics A “sound” is a complex pattern of rapid variations in air pressure, travel- ing from a sound source and striking the ear, which causes a series of neural signals to be received in the brain: this is true of speech, music and random noises. 1.2 Phonetics – what is physical sound? 4 INTRODUCING PHONOLOGY 2 This is the “apostrophe s” suffix found in The child’s shoe, meaning ‘the shoe owned by the child.’ Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 The horizontal axis represents time, with the beginning of each word at the left and the end of the word at the right. The vertical axis represents displacement of air particles and correlates with the variations in atmos- pheric pressure that affect the ear. Positions with little variation from the vertical center of the graph represent smaller displacements of air particles, such as the portion that almost seems to be a straight horizontal line at the right side of each graph. Such minimal displacements from the center cor- respond to lower amplitude sounds. The portion in the middle where there is much greater vertical movement in the graph indicates that the sound at that point in time has higher amplitude. While such a direct representation of sounds is extremely accurate, it is also fairly uninformative. The difference between these words lies in their vowels (ee versus i), which is the part in the middle where the fluctuations in the graph are greatest. It is difficult to see a consistent difference just looking at these pictures – though since these two vowels are systematically distinguished in English, it cannot be impossible. It is also very difficult to see similari- ties looking at actual waveforms. Consider figure 2 which gives different repetitions of these same words by the same speaker. Waveforms. A concrete way to visually represent a sound is with an acoustic waveform. A number of computer programs allow one to record sound into a file and display the result on the screen. This means one can visually inspect a representation of the physical pattern of the variation in air pressure. Figure 1 gives the waveforms of a particular instance of the English words seed and Sid. What is phonology? 5 FIGURE 1 Waveforms of speech‘seed’ ‘Sid’ Time a m p l i t u d e Absolute accuracy is impossible, both in recording and meas- urement. Scientific instruments discard information: micro- phones have limits on what they can capture, as do recording or digitiz- ing devices. Any representation of a sound is a measure- ment, which is an idealization about an actual physical event. Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 6 INTRODUCING PHONOLOGY Visual inspection gives you no reason to think that these sets of graphs are the same words said on different occasions. The problem is that while a physical waveform is a very accurate representation of a word, it provides so much information that we cannot tell what is important and what is not. Since we are interested in the part which makes these two words sound different, we might get a clearer picture of the physical differ- ence by expanding the scale and looking just at a part of the vowel. Vowels are periodic, which means that the pattern of their wave- form repeats over time. The display in figure 3 gives a portion of the FIGURE 2 Different repetitions of words FIGURE 3 Closeup waveform of vowels of seed, Sid a m p l i t u d e ‘seed’ ‘Sid’ Time a m p l i t u d e Time ‘seed’ ‘Sid’ Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 vowels from the middle of the words seed and Sid, involving around 30 milliseconds (ms) of each of the words (the entire word in each of these two examples actually lasts approximately 600 ms, so this is a small part of the entire word). We can indeed see that there is a pat- tern which is repeated (although successive repetitions are not perfect reproductions). Though there are visible differences between the waveforms, the basis for distinguishing these vowels remains unclear. Sound spectra. We need a better analytical technique than just looking at raw sound, to be able to talk precisely about properties of these sounds. We therefore need to understand some basic properties of physical sounds. All sound waves are definable in terms of three properties that characterize a sine wave familiar from trigonometry, namely frequency measured in cycles per second also known as Hertz (Hz), amplitude meas- ured in decibels (dB), and phase measured in the angular measure radi- ans. These characteristics suffice to define any sine wave, which is the ana- lytic basis of sounds. The property phase, which describes how far into the infinite cycle of repetition a particular sine wave is, turns out to be unim- portant for the study of speech sounds, so it can be ignored. Simple sine waves (termed “pure tones” when speaking of sounds) made up of a single frequency are not commonly encountered in the real world, but can be created by a tuning fork or by electronic equipment. Speech sounds (indeed all sounds) are complex waveforms which are vir- tually impossible to describe with intuitive descriptions of what they “look like.” Fortunately, a complex waveform can be mathematically related to a series of simple waves which have different amplitudes at different fre- quencies, so that we can say that a complex waveform is “built from” a set of simple waves. Figure 4 shows a complex wave on the left which is con- structed mathematically by just adding together the three simple wave- forms of different frequencies and amplitudes that you see on the right. What is phonology? 7 FIGURE 4 Complex wave and the component simple waves defining it The complex wave on the left is made from simple sine waves at 100, 200, and 300Hz, and the individual components defining the complex wave are graphed on the right. The most prominent component (the one Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 Changing the amplitude of one such component changes the overall character of the waveform. A complex wave is mathematically equivalent to a corresponding series of sine wave components, so describing a com- plex wave directly is equivalent to describing the individual components. If we see two differently shaped complex waves and we can’t describe their differences directly in terms of the complex waves, we can instead focus on the equivalent series of sine wave components, and describe the dif- ferences in terms of very simple information about component frequency and amplitude. Just as a single complex waveform can be constructed from a series of sim- ple waves at different frequencies and amplitudes, a single complex wave- form can also be mathematically broken down into a series of components which have different frequencies and amplitudes. Rather than graph the full shape of each specific sine wave component – which becomes very hard to understand if there are more than a handful of components – we can simply graph the two important values for each of the component sine waves, the amplitude and frequency. This is known as a spectrum: it is the defining fre- quency and amplitude components of a complex waveform, over a fixed peri- od of time. The spectrum of the waveform in figure 4 is plotted in figure 6, where the horizontal axis corresponds to frequency from 0 to 7,000 Hz and the vertical axis corresponds to amplitude from 0 to 60 dB. Note that in this display, time is not represented: the spectrum simply describes amplitude and frequency, and information about how long a particular complex wave- form lasts would have to be represented somewhere else. 8 INTRODUCING PHONOLOGY with the highest amplitude) is the one at 100 Hz, the thinnest line which makes one cycle in the chart: it has an amplitude of 60 dB. By comparison, the component at 200 Hz (graphed with a medium-weight line, which makes two cycles in the chart on the right) has the lowest amplitude, 40 dB. The 300 Hz component, graphed with the thickest line, has an inter- mediate amplitude of 50 dB. It is the amplitudes of the individual compo- nents which determine the overall shape of the resulting complex wave. Now we will see what happens when we change this artificial sound to make the 200 Hz component be the most prominent component and the 100 Hz one be less prominent – if we simply switch the amplitudes of the 100 Hz and 200 Hz components, we get the wave shown in figure 5. FIGURE 5 Effect of changing component amplitude Downloaded from Cambridge Books Online by IP 202.45.54.138 on Mon Feb 11 09:53:54 WET 2013. http://dx.doi.org/10.1017/CBO9780511808869.003 Cambridge Books Online © Cambridge University Press, 2013 What is phonology? 9 This is a very simple spectrum, representing an artificially constructed sound containing only three components. Naturally occurring sounds have many more components than this. Since complex sounds can be mathematically broken down into a series of simple components, we can use this very useful tool to look at the vowel sounds of seed