Introduction to English Linguistics

1 The sounds: phonetics

1.1 Introduction

Speaking is such a normal and everyday process for us that most of the time we do not consciously think about what we are doing. Fortunately, you might say. Imagine you had to think carefully about every sound in every word in every sentence you want to produce. It could take hours to finish a single sentence. Luckily, there is no need for this: we have developed such efficient routines for speaking that most of the necessary actions do not require conscious thought. You could compare it to walking: once you have learned what to do, some sort of automatism takes over.

This works fine as long as we stick to our respective native language. The situation changes, however, when we start learning a new, foreign language. Not only are the words different, but in many cases the foreign language also has some sounds which are unfamiliar. German learners of English, for instance, very often have problems with the "lisping" sound in words such as bath, therapy, or mathematics (we use italics whenever we cite words as examples). There are no German words which include this type of sound. That does not mean, of course, that native speakers of German cannot achieve a correct pronunciation of bath or therapy, but before they can do so they have to learn how to produce the new sound. English learners of German, on the other hand, encounter the same problem with the vowel that appears in German Müsli 'muesli' and Hüte 'hats' (we use single inverted commas to indicate the meanings of examples cited). This vowel is not part of the pool of sounds which English speakers use to construct the words of their language, the English sound inventory (we use bold print whenever we introduce an important new term).

There are some general conclusions we can draw from this. Firstly, languages may use only a subset of all possible speech sounds. In fact, there is no language which makes use of all of them. Secondly, languages differ in which sounds they include in their inventory: German uses a different selection than English does. Foreign language learners are thus bound to encounter sounds which do not occur in their native language and which they do not have routines for. They have to learn the gestures necessary to produce these unfamiliar sounds. In other words, learners have to find out which muscle movements in which combination and sequence are required for the production of the respective new sound.

There is an entire subdiscipline of linguistics, phonetics, which deals with these and other characteristics of speech sounds. It focuses on questions such as the following: What types of speech sounds do we find in the languages of the world and in individual languages? How can we describe these sounds? Which criteria can we use to distinguish different sounds?

Several approaches have been taken to the investigation of speech sounds and different branches of phonetics have developed, each focussing on a different aspect of speech. Articulatory phonetics aims at describing the process of articulation. How do we create speech sounds? In what way does the production of one sound differ from that of another? Which articulation-related criteria can we use to distinguish and classify different speech sounds? Acoustic phonetics, on the other hand, concentrates on the physical properties of the speech sounds themselves. What is the physical reality of a speech sound and how can we measure acoustic differences between speech sounds? Which physical properties are characteristic of particular sounds? Finally, auditory phonetics investigates how speech sounds are perceived and processed by the listener. In the following sections we will deal mostly with articulatory phonetics.

The remainder of this chapter is structured as follows: section 1.2. introduces some notational conventions used in phonetics which are essential for the discussion of speech sounds. In 1.3. we give a general overview of the nature of speech sounds and their production before discussing in more detail how we can describe and classify sounds in section 1.4., focussing on the sounds we find in English. A conclusion in section 1.5. summarises the findings of this chapter.

1.2 Spelling vs. pronunciation: the representation of speech sounds

We said above that the sound that occurs in English therapy and mathematics does not exist in German. What about the German words Therapie or Mathematik, you might want to object. The letters are the same, alright. However, if you pronounce the words, you find different sounds in English and in German: only English has the lisping sound that we referred to earlier, whereas the corresponding sound in the two German words is the same as that which occurs initially in words such as Tee 'tea'.

What does this tell us? For one thing, the examples show that we have to strictly distinguish between letters and sounds, since we do not always get the same sound for the same letter. Even within a single language, there is no one-to-one correspondence of sound and orthographic symbol. Some more examples of this phenomenon are given in (1):

(1)  knight would who doubt honest though

If we assumed that every word in (1) had the same number of letters and sounds, we would expect the word knight to consist of six distinct sounds, would should have five sounds, and so on. If you listen to someone pronouncing these words, you will notice that, clearly, this is not what we find. Some of the letters do not seem to have a corresponding sound at all: all of the above words have more letters than sounds. The conclusion we can draw from this is that spelling does not necessarily reflect the sound structure of words, or, to put it differently, spelling and pronunciation are two pairs of shoes and ought to be kept apart. To illustrate this point further, let us investigate some more words. What can you find out about the relation of spelling and pronunciation when looking at the words in (2)?

(2) a. rune b. beat who head shoe great moon heard you heart true

If you pronounce the words in (2a), you will realise that they all share the same vowel sound. Yet, they differ in spelling. For one and the same vowel sound, we find an amazing total of six different spellings. Turning to the words in (2b), we observe a related phenomenon, again focussing on the vowels and their representation in writing: This time, the spelling is the same in all five words, <ea>, but the pronunciation is a different one each time.

It is obvious from the examples above that spelling does not help us if we want to talk about and refer to sounds, since the correspondence between letters and sounds is not one-to-one. This, however, is exactly what we would like to have when we talk or write about sounds, so that, whenever we use a certain symbol, people know exactly which sound we are talking about. Therefore phoneticians have established transcription systems which have a separate symbol for each sound. The most well-known and most standardised transcription system is the International Phonetic Alphabet (IPA). You may already be familiar with phonetic transcription from dictionary entries or the vocabulary parts of your foreign language textbooks in school, where the IPA or a similar system of phonetic symbols is used to indicate the pronunciation of words. Many (though not all) of the symbols in fact look like ordinary letters, but it is important to keep in mind that, unlike letters, phonetic symbols directly represent sounds in a one-to-one fashion. In order to indicate which kind of unit we are talking about, we use two different sets of brackets: angled brackets "< >" for letters, and brackets "[ ]" for sounds. Thus, for instance, the orthographic representation of the word referring to a piece of garment you wear on your feet is <sock>, but the representation of its sounds, its 'phonetic representation', is [sk]. In the latter case each symbol represents uniquely one sound.

Of course, if you want to use the IPA symbols, you need to be familiar with the correspondences between symbols and sounds. Which sound does a given symbol refer to? There is no way of telling just from the look of it. You can think about phonetic symbols as labels of some kind, not unlike labels you use for musical notes, for instance. The names we use to distinguish different notes, such as 'a', 'g', or 'd', enable us to talk about these notes without giving complicated examples and explanations. Phonetic symbols function in much the same way.

The list in (3) gives the IPA symbols and sample words for all sounds that occur in the standard accent of British English, which is known under the name of 'Received Pronunciation' (RP). Note that the complete IPA...