An Introduction to the Science of Phonetics

Nigel Hewlett, Janet Mackenzie Beck

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781136500039
• Edition: 1
• Topic: Lingue e linguistica
• Subtopic: Linguistica

The book is designed as an introduction to the scientific study of speech. No prior knowledge of phonetics is assumed. As far as mathematical knowlege is concerned, all that is assumed is a knowledge of simple arithmetic and as far as possible concepts are dealt with on an intuitive rather than mathematical level. The anatomical material is all fully explained and illustrated. The book is arranged in four parts. Part 1, Basic Principles, provides an introduction to established phonetic theory and to the principles of phonetic analysis and description, including phonetic transcription. Part 2, Acoustic Phonetics, considers the physical nature of speech sounds as they pass through the air between speaker and hearer. It includes sections on temporal measurement, fundamental frequency, spectra and spectrograms. Part 3, Auditory Phonetics, covers the anatomy of the ear and the perception of loudness, pitch and quality. The final part, Part 4, covers the articulatory production of speech, and shows how experimental techniques and tools can enhance our understanding of the complexities of speech production.Though the audience for this bookis mainly students and professors in the Speech Sciences, it will also be valuable to any students studying hearing science and acoustics. The book is well supported with figures, tables, and practice boxes with experiments.

1	Introduction

In attempting to choose a title that would accurately cover the broad range of topics in the book, we eventually decided on An Introduction to the Science of Phonetics. Much of the material in the book could go by the name of speech and hearing science. However, phonetics is a more inclusive name. It allows us to include a chapter on speech perception, the stage beyond hearing. It also implies a link to other aspects of language (phonology, syntax, semantics, and pragmatics). Most important, it includes the tradition of systematic, trained auditory observation, which is the concern of the first section of this book. Even where this technique is not used alone (as it still is in many applications—in many applications in speech pathology, for example), it remains an important ingredient of the most successful work in phonetics/speech and hearing science.

Phonetics is the scientific study of the sounds of speech, including their production and their perception. Like eating and walking, speech comes naturally and so for the most part we don’t consciously consider what the process involves. Studying phonetics requires a systematic, conscious consideration of how speech sounds are made, what they sound like, and how they compare with each other. Beyond the present chapter, the book is divided into four sections. The system of description and analysis that has become established is described in Part I. It includes phonetic transcription, which is a very useful tool in all branches of phonetics. Part II, on acoustics, describes the physical nature of speech sounds during their fleeting existence in the air between speaker and hearer. In one sense, this acoustic signal is speech. It is what the speaker produces and the hearer hears. A large and well-established body of knowledge is now available on the acoustic structure of speech. Part III deals with hearing and perceiving speech. Of course, there are other things to be heard besides speech (music, for example, and environmental noises like thunder and traffic noise, the hearing of which can be very useful to us) and hearing science is wider than a study of the hearing of speech, although for humans speech is surely the most important noise around. Likewise, on the production side, the subject of Part IV, humans can produce other sounds besides speech. But in this case there can be even less doubt that the other sounds that can be produced by the human vocal tract, like crying and laughing and moaning and whistling, take second place to speech.

This book attempts a fairly comprehensive introduction to the subject and some readers may wish to read selectively. Much of the material in Part I will ease the way to an understanding of later sections, particularly Parts II and IV Certainly, some knowledge of phonetic transcription would be essential for these later sections. Parts III and IV assume an understanding of material in Part II. However, Parts III and IV can be read independently of each other. We have tried to make the text as simple as it can be, while (hopefully) avoiding misleading oversimplifications. Some chapters will be found more difficult than others, depending on the prior knowledge of the reader, although no previous knowledge of phonetics is assumed nor is any specialist knowledge of anatomy, physiology, or mathematics. For the most part, applying the rules of arithmetic is all that will be demanded of the reader, so far as mathematics is concerned (although, admittedly, there is a square root or two in chap. 17). The only real exception is logarithms, which are needed for understanding the semitone scale (chap. 9), the decibel scale (chap. 10), and some graphs related to hearing (in Part III). Logarithms are explained in the text and the reader is not expected to have prior knowledge of them. Many readers will find chapters 10 and 17 the most difficult in the book. A detailed understanding of chapter 10 is not necessary for understanding chapters 11 and 12, but it is important for understanding some of the content of Part III, on hearing. Chapter 17 could be omitted, without causing difficulties in reading chapters 18 and 19, the final chapters of the book.

Most chapters contain practice boxes at intervals in the text. They describe articulatory activities to be carried out by the reader or straightforward test questions to be answered. Many chapters also have a set of exercises at the end. All these activities and exercises are focused on the acquisition or improvement of phonetic skills (transcribing speech, interpreting spectrograms, specifying hearing thresholds, and so forth) and we urge readers not to skip them.

Phonetics, (the Rest of) Linguistics, and Psychology

Linguistics is the scientific study of language, including, notably, the areas of semantics, syntax, and phonology. Whether phonetics should be counted as part of linguistics or as a separate discipline is an argument too obscure to be of much interest here (or anywhere else, possibly). However, there are obviously connections and overlap in the study of different areas of language. Modern linguistics is much concerned with the nature of mental representations, including mental representations of the phonetic forms of words, and ultimately with answering the question: What is the nature of human language? Many linguists assert that linguistics itself is situated within the field of psychology.

In the production of speech, in real time, a sentence is formulated in the brain and is transformed into a sequence of movements of the speech organs, so speech is the result of both mental and physical activity. In the comprehension of speech, a listener hears an acoustic signal, which is transformed into a sentence in their brain. Explaining the processes of production and comprehension is the province of phonetics in combination with (the rest of) linguistics and psychology. The relationship between the mental activity and the physical activity that are involved in speech is a complex and unusual one. The physical movements of the tongue, lips, and so on are goal-oriented actions, like walking and eating. The physical forces to be overcome may vary according to the context: for example, producing a “b” sound, while lying down, involves overcoming different physical forces from producing a “b” sound while standing up. Similar things can be said of a physical activity like swallowing. However, the goal of swallowing is relatively easy to define: to get a bolus of food or liquid traveling down toward the stomach, without choking. The goals of the movements made in speaking are not so easy to define. They communicate meanings, by means of words, so the need to define relationships between physical events and mental phenomena soon arises. Also, the word goal implies the possibility of success or failure. How do we measure the success or failure of a speech utterance? Speech communication requires, in addition to the speaker, at least one listener, and the success or otherwise of an utterance lies with whether or to what extent it has successfully transmitted intended meanings to a listener, or listeners. A little thought around this issue will soon produce some interesting complications with judging whether or not a particular piece of speech has achieved its goal. Although it is not directly addressed to such questions, which trespass into other areas of linguistics and psychology, this book aims to give the reader the phonetic knowledge that is required in order to pursue them.

Speech Sounds, Phonemes and Pronunciational Variants

We referred to “speech sounds,” which gives an impression that speech consists of sequences of separate sounds, just like the writing on this page consists of sequences of separate letters. However, unlike the letters on this page, speech sounds do not come separated, because a speech utterance is a continuous stream of sound. Nevertheless, it is common to record and analyze such a stream as if it arises from a number of discrete sounds put together. This is the approach taken, for the most part, in this book. A pronunciation of the word “bun,” for example, is treated as a sequence of three individual speech sounds: a “b” followed by a “u” (“Λ” in phonetic transcription) followed by a “n.” This approach probably seems convincing, even obvious. It may indeed be the correct one but it does become less, rather than more, obvious with increasing knowledge of phonetics.

When a speech sound (or a sequence of speech sounds) is referred to in writing, by a letter, it is either written between slashes or between square brackets: for example, either /p/ or [p]. Roughly, /p/ stands for all instances of “p.” It stands for a sound category that functions as a contrastive unit in the language in question, the name for which is a phoneme.¹ When, on the other hand, the letter is used to denote an actual utterance of the phoneme, on one particular occasion, it is enclosed in square brackets. Square brackets are also used to denote systematic pronunciational variants: For example, the phoneme /p/ in English has (at least) two systematic variations in its pronunciation, according to whether it occurs at the beginning of a syllable or elsewhere. The reasons behind the distinction should become clearer after a reading of the first few chapters of this book.

A proper exploration of the issues raised in the preceding paragraph is the domain of phonology, which is the study of speech sound systems in different languages, how they are represented mentally, and how they function in distinguishing one word from another. A good introduction to phonology is Carr (1999). Understanding phonology requires some prior knowledge of phonetics. A knowledge of the material in Part I of this book would suffice for much of the phonological literature. However, more and more in recent years, research in phonology has used, as evidence, findings from experimental phonetics, in which case it requires a knowledge of material in Parts II to IV.

Accents of English

Where possible, illustrations of speech sounds in this book are taken from English. However, there are many varieties of English. Although differences in syntax and vocabulary are not an issue in a book on phonetics, differences in pronunciation certainly are. A group of people who pronounce their language in approximately the same way are said to speak with the same accent. Accents may be defined more or less narrowly. Accent differences are associated with geographical and sociological factors. The normal orthography does not change according to accent. The word “where,” for example, is pronounced in quite different ways, according to accent, but it is always spelled the same.

We make most frequent reference to three (broadly defined) standard accents: General American, RP and Scottish English. General American is spoken widely across the whole of the United States. RP (sometimes also called BBC English or Southern British Standard English) is the accent of most upper-class English people (and some Welsh, Irish, and Scottish, too). RP stands for Received Pronunciation, the meaning of which is obscure. Scottish English (more precisely, Standard Scottish English) is the accent of most middle-class speakers in Scotland. There are some radical differences among these accents. Scottish English has a rather different vowel system from most other accents; for example, the words “look” and “Luke” share the same pronunciation, but the words “heard” and “herd” are pronounced differently. Both General American and Scottish English pronounce the “r” sound in words like “floor” and “horse.” They share this feature with, for example, Irish English and Canadian English. In RP, there is no “r” sound after a vowel, so the words “flaw” and “floor” are pronounced the same. It shares this feature with, for example, Australian English and New Zealand English.

No attempt is made here to give a systematic description of accent variation. The standard work on accents of English, from a phonetic point of view, is Wells (1982). Wells could be read comfortably by anyone who has worked through Part I of this book. The study of associations between phonetic variables and sociological variables goes by the name of sociophonetics.

Applications of Phonetics

This book is focused on what might be called the core of phonetic science but phonetics finds applications in many related fields. This section gives an idea of the range of applications of phonetics.

Psychology of Communication. From one point of view, phonetics is situated within the broad field of communication. Most of the time, even in the modern technological age, speech occurs in the context of people taking turns to be speaker and listener in a face-to-face conversation. In such a situation, many other factors (visual signals of one sort or another, for example) interact with the phonetic signal in complex ways. For example, it may be possible to convey sarcasm entirely by tone of voice. Alternatively, it may be conveyed by facial or other gestures or by some combination of the phonetic and the visual. In any case, whether or not sarcasm is conveyed depends on the listener as well as the speaker (cf. the remarks made in the introductory section of this chapter, about speech communication). What one person perceives as sarcasm, another may not. Phonetics forms an important part of the complex study of face-to-face communication.

Speech Technology. Speech technology is a large and growing field with, by now, quite a long history. For many years we have been able to make a permanent recording of speech and other sounds, either by mechanical or electrical means, and play it back. Nowadays, most recordings are in digital form. Digital recordings tend to be better quality and they are more convenient for editing and storage. Speech can now be produced by machine (speech synthesis) and recognized by machine (speech recognition), although, at the time of writing, there remains substantial room for improvement, especially in the case of speech recognition. Phonetic science contributes in all sorts of ways to such developments. For example, economies can be made with playback systems if it is known how the ear responds to sound. Perceptual encoding, as it is called, is a technique in which certain features are deleted from the recorded material in the knowledge that the human ear will not detect any difference.

Speech synthesis must contend with contextual variation. The “b” in “bun” is phonetically different from the “b” in “nub.” So it is no good just telling the machine to “produce ab”; speech synthesis systems have to find ways of taking account of context and taking account of the fact that a speech utterance is a continuous stream of sound. A problem for speech recognition (greatly underestimated in the early days) is the sheer amount of variability in speech, which human listeners cope with but that easily confounds machines. Sources of variability include the identity of the speaker (factors like sex, age, and health, as well as accent), the topic under discussion, and the degree of informality of the occasion.

Forensic Phonetics. Suppose, for example, that in a criminal trial the prosecution presents evidence in the form of a tape recording of a threatening phone call. The prosecution claims that the voice on the recording belongs to the defendant. The defendant claims it is not his voice. Phoneticians have given evidence as expert witnesses in a number of cases of this sort. The techniques of analysis that are involved in such work include both trained listening skills (described in Part I of this book) and acoustic analysis (described in Part II).

Clinical Phonetics. Last, but not least (considering it is both authors’ main interest!), we should mention clinical phonetics. Disorders of speech and of hearing, although not usually life threatening, or even obviously disabling at first sight, nevertheless have catastrophic effects on people’s ability to participate normally in social and economic life. The disorders cover a wide spectrum. They differ according to the nature of the syndrome and the age of the affected person. For example, a difficulty with producing speech requires a different form of intervention from a difficulty with hearing it; abnormal structure of the speech apparatus (a cleft palate, for example) creates different problems from a reduced precision of control over the movements of the speech organs; being deaf from birth produces quite a different situation from that of acquiring deafness late in life; and so on. In wealthier countries, at least, a specialist profession, or professions, supplies a service of diagnosis and treatment for people suffering from speech, language, or hearing disorders. In the United States, audiologists and speech pathologists undergo a common basic training. In Britain, training for audiologists and speech and language therapists is separate. In some parts of Europe, there is a distinction between logopaedists (community-based practitioners...