English Phonetics: Consonants PDF

2 English Phonetics: Consonants (ii) 2.1 Central vs Lateral In discussing the alveolar approximant [ɹ], we said that the air escapes along a central groove (of the tongue, in this case; the same kind of groove can be formed by the lips). This is true for all of the fricatives and ap...

2 English Phonetics: Consonants (ii) 2.1 Central vs Lateral In discussing the alveolar approximant [ɹ], we said that the air escapes along a central groove (of the tongue, in this case; the same kind of groove can be formed by the lips). This is true for all of the fricatives and approximants described in chapter 1: they are all central fricatives and approximants. However, it is possible to produce fricatives and approximants in which this is not the case. For instance, in the first sound in lift, the centre of the blade of the tongue forms a stricture of complete closure with part of the alveolar ridge, but the articulation which ‘counts’ is that between the sides of the tongue and the alveolar ridge. Since the sides of the tongue form a constriction of open articulation with the alveolar ridge, and no friction is created, we refer to this sound (transcribed as [l]) as a voiced alveolar lateral approximant. Since English fricatives and approximants are typically central, we will use the term ‘lateral’ for laterals, and omit the term ‘central’ in describing central fricatives and approximants in English speech. The sounds [l] and [ɹ] are, clearly, quite similar: both are approximants, both 49 are voiced, both are alveolar. The principal difference is that the former is lateral and the latter central.1 2.2 Taps and Trills We have said that, for a great many speakers of English, the sound at the beginning of words such as rat, rope, reap, etc. is a post-alveolar approximant: [ɹ]. The same is true of the sound which occurs after stops in words such as prude, true, creep, etc. However, some speakers utter, not an approximant, but a sound which is very like a voiced alveolar stop of very short duration. Many Scots utter this sound, rather than [ɹ], after stops, as in the words just cited. During the articulation of this sound, the blade of the tongue comes into a momentary constriction of complete closure with the alveolar ridge. This sound, transcribed as [ɾ], is referred to as a voiced alveolar tap (or flap). This is also the sound that many American speakers have instead of [t] or [d] in words such as Betty, witty, rider, heady, etc. Speakers of certain accents of English may utter neither an [ɾ] nor an [ɹ] in words such as rat, rope, reap and prude, true, creep, but a sound referred to as a voiced alveolar trill. Trills are produced by holding one articulator (e.g. the blade of the tongue) next to the other (e.g. the alveolar ridge) in a constriction of complete closure, but without the same muscular pressure as one finds in stops. The result is that air pressure builds up behind the closure and forces it open; the air pressure then reduces, and the muscular pressure again creates a constriction of complete closure. This sequence may be 50 repeated in quick succession, producing, in the case of an alveolar trill, a series of taps of the tongue against the alveolar ridge. The alveolar trill is transcribed as [r], but is relatively rare. Scots are often said to produce this sound; however, most speakers of Scottish varieties of English typically produce, not an alveolar trill, but an alveolar tap. 2.3 Secondary Articulation We have said that the lateral approximant [l] is alveolar. However, laterals may also be produced with an additional articulation, such as one formed between the back of the tongue and the velum, i.e. a velar articulation. When this happens, we may distinguish between the alveolar articulation as the primary articulation and the velar one as the secondary articulation. Where a secondary articulation is velar, this process is referred to as velarization: we say that the lateral is velarized. A velarized lateral approximant is transcribed using the velarization diacritic, thus: [ɫ]. This sound is often referred to as ‘dark l’.2 Where a secondary articulation is palatal (formed between the front of the tongue and the hard palate), this process is referred to as palatalization; we say that the lateral is palatalized. A palatalized lateral is transcribed using the palatalization diacritic, thus: [lj]. The term ‘clear l’ is often used to refer to [lj], or to [l] (neither palatalized nor ‘dark’). In subsequent chapters, we will consider the status of ‘dark l’ and ‘clear l’ in different accents of English. 51 2.4 Affricates We have, thus far, distinguished three classes of consonant according to degree of constriction: stops, fricatives and approximants. Consider the first sound in chip: it is like a stop in that there is complete closure between the blade of the tongue and the palato-alveolar region. However, it is like a fricative in that it clearly involves friction. That friction occurs during the release phase of the closure, which we referred to in 1.3.1. Sounds produced with a constriction of complete closure followed by a release phase in which friction occurs are called affricates. We might say that one of the main differences (place of articulation apart) between the first sound in tip and the first sound in chip is that, during the release phase of the [t] in tip, there is no friction of the sort one finds during the release phase of the first sound in chip. We might therefore think of affricates as stops with a slow, fricative, release phase. The affricate in chip is a voiceless palato-alveolar affricate, transcribed as [tʃ]. Its voiced counterpart is [dʒ], the first sound in jury, joy, etc.3 These two affricates occur in the speech of most speakers of English. In later chapters, we will examine some other affricates which occur in the speech of speakers of certain accents of English. 2.5 Aspiration The first stop in pit, we said, is a voiceless bilabial stop. So too 52 is the first stop in spit. But the bilabial stop in pit differs phonetically from the bilabial stop in spit: if you hold the palm of your hand up close to your mouth when uttering pit, you will feel a stronger puff of air on releasing the bilabial stop than you will when you utter spit. That ‘stronger puff of air’ phenomenon is called aspiration: we say that the bilabial stop in pit is an aspirated voiceless stop, whereas the stop in spit is unaspirated. Aspirated voiceless stops are transcribed with the aspiration diacritic ([ʰ]), so that the bilabial stop in pit is transcribed as [pʰ]. Unaspirated stops are transcribed without that diacritic, so that the bilabial stop in spit is transcribed as [p]. 2.6 Nasal Stops We have been making an assumption in our discussion thus far, concerning the position of the velum in the production of the speech sounds we have described. We have assumed that, in all of these sounds, the air from the lungs is escaping only through the mouth (the oral cavity). This is true if the velum is in the raised position, such that it prevents the flow of air out through the nasal cavity (see figure 1). In all of the sounds discussed thus far, the velum is indeed raised: we describe all such sounds as oral sounds. But the velum may be lowered, to allow escape of air through the nasal cavity (see figure 1). Sounds produced with the velum lowered, and with air escaping through the nasal cavity alone, are referred to as nasal stops.4 These may occur at most places of articulation; 53 let us consider those which are relevant for the study of English. While nasal stops may be either voiced or voiceless, they are typically voiced in most human languages; we will therefore ignore voiceless nasal stops and use the term ‘nasal stop’ to imply ‘voiced nasal stop’. Bilabial nasal stops (transcribed [m]) entail, as one would expect, complete closure between the lips, voicing, and escape of the air through the nasal cavity. An example is the first consonant in map. Labio-dental nasal stops (transcribed [ɱ]) entail complete closure between the lower lip and the upper teeth, voicing, and escape of the air through the nasal cavity. An example is the second consonant in pamphlet. In English, they occur before labio-dental sounds, as in this case. The nasal stop articulation in cases such as these reflects a process of assimilation. Assimilation processes are processes in which one sound becomes similar to an adjacent sound. In this case, the nasal is assimilated to the following fricative, in the sense that it ‘takes on’ the place of articulation of the fricative. Such processes involve a principle of ease of articulation. In this case, if the nasal in pamphlet is articulated at the same place as the following fricative, this saves the speaker the articulatory effort of moving from a bilabial to a labio-dental articulation. We will return to such processes in chapter 6. Dental nasal stops (transcribed as [n̪]) entail complete closure between the tip of the tongue and the upper teeth, voicing, and escape of the air through the nasal cavity. An example is the second consonant in tenth. As in this case, they occur before other dental sounds, and this too is a matter of 54 assimilation involving place of articulation. Alveolar nasal stops (transcribed as [n]) entail complete closure between the blade of the tongue and the alveolar ridge, voicing, and escape of the air through the nasal cavity. An example is the first sound in not. Velar nasal stops (transcribed as [ŋ]) entail complete closure between the back of the tongue and the velum, voicing, and escape of the air through the nasal cavity. An example is the last sound in sing or the nasal stop as it is often articulated (especially in faster or more casual speech styles) in the word incredible. Once again, the latter case involves assimilation. Notes 1 The central approximant [ɹ] also differs from [l] in having tongue body retraction and lip rounding. We will see shortly that alveolar laterals may be produced with retraction too. 2 The term ‘dark l’ can also be used to refer to lateral approximants in which the body/back of the tongue is retracted and/or lowered. Accents of English vary with respect to the exact articulatory nature of their ‘dark l’s: some are velarized, while others have no velar articulation, but have, instead, retraction and/or lowering of the back/body of the tongue. Such retraction can lead to loss of alveolar contact, and thus to [l]-vocalization, in which the articulation becomes vowel-like. 3 Some authors transcribe [tʃ] as [č] and [dʒ] as [ǰ]. We should, if we were to stick strictly to the conventions of the International Phonetics Association, transcribe both affricates 55 with a ‘tie bar’ above the two symbols; we depart here from the conventions of the IPA chart, which does not contain an ‘affricate’ category. 4 The term ‘nasalized’, as opposed to ‘nasal’, is used to describe sounds in which air escapes through both cavities, the oral and the nasal. The term ‘nasal’ is used to describe sounds in which the air escapes through the nasal cavity alone. Exercises 1 Listen to Track 2.1 at www.wiley.com/go/carrphonetics. For each of the words on the recording, identify (a) any oral stops, (b) any fricatives, (c) any approximants, (d) any affricates and (e) any nasals. For each sound that you identify, say whether it is voiced or voiceless and what its place of articulation is (e.g. the word stop: voiceless alveolar stop [t] and voiceless bilabial stop [p]; voiceless alveolar fricative [s]; no approximants, affricates or nasals). The words are: bring licking fever thinking assure measure heated year 2 Listen to Track 2.2. Which of the words on the recording begins with an affricate, and which (if any) with a stop? tune chip dune June Many speakers of English typically utter words like tune 56 and dune with an affricate at the beginning of the word. This means that dune and June are typically indistinguishable. None the less, when asked in a phonetics class whether they utter words such as dune with an affricate, such speakers often deny that they do. These speakers typically have a more careful pronunciation of words such as tune and dune, in which there is a [j]. Notice, however, that there is no such more careful pronunciation of words like chip and June: one never hears these pronounced with [tj] and [dj]. In order to explain the difference between dune and June, we need to say that the speaker in some sense intends to utter [dj] in dune, but that ease of articulation results in a palato- alveolar affricated release of the stop closure, rather than a transition from an alveolar closure to a stricture of open approximation between the front of the tongue and the hard palate. In the case of June and chip, the intended articulation is a palato-alveolar affricate. If you are a speaker of General American, you may well never utter a [j] in words like tune and dune, in which case you will utter a stop followed by a vowel. However, you may well also have been told at school that the ‘correct’ pronunciation of such words has a [j] after the stop. Your speech may well vary with respect to the presence or absence of the [j]. If your speech does vary in this way, how do you pronounce noon? 57 3 Listen to Track 2.3. Give a phonetic transcription of each of the words on the recording, using a ‘V’ for the vowels. The words are: lull pear reap throws think misjudges churches incorrec You may well have noticed that the nasal stop in think is velar, rather than alveolar. It requires considerable conscious effort to utter that nasal stop as alveolar, and when one does so, the resulting pronunciation sounds quite unnatural. This appears to be the result of a process of anticipatory assimilation: the tongue adopts the articulatory position for the velar stop [k] during the pronunciation of the nasal. But what about the nasal-plus-velar stop sequence in incorrect? Many speakers of English find it easier to utter an alveolar, rather than a velar, nasal there, despite the fact that cases like incorrect also contain a sequence of a nasal stop followed by a velar stop. Do you have any hunches as to why the two cases should be different? 58