Session 5: Applications of Phonetics and Phonology
Vowels
Before moving on to larger units of speech, it makes sense to have a closer look at the second major building block in our phoneme inventory besides consonants: vowels.
Vowels are produced by letting air flow through the articulatory system without any significant obstruction. The vocal cords always vibrate when a vowel is produced and the continuous stream of air makes it possible to lengthen or shorten vowel sounds, a distinction that can differentiate meaning in some languages.
The central qualities that allow us to describe vowels are height, backness and roundedness. They allow us to describe where a vowel sound originates (height and backness) and the shape of the lips during articulation (roundedness).
The above graphic should give you a good idea of height and backness and in what way they affect the quality of vowels.
The vowel chart (originally developed by phonetician Daniel Jones) combines all three features (height, backness, roundedness) into a single model to describe the realization of the so-called cardinal vowels in the oral cavity.
Because the decisive articulator determining the quality of vowel sounds is the tongue, it is possible for vowel quality to change over the duration of articulation if the position of the tongue changes. If the articulatory configuration shifts from one vowel into another the resulting composite sound is a so-called diphthong (or gliding vowel).
Examples for pure vowels:
sit /ɪ/ => high front vowel (unrounded)
foot /ʊ/ => high back vowel (rounded)
man /æ/ => middle central vowel
Examples for diphthongs:
time /aɪ/
face /eɪ/
choice /ɔɪ/
Differences between dialects and sociolects of English are often marked by contrasts in vowel quality.
Suprasegmental phonology / prosodics
Suprasegmental phonology studies intonation and other aspects of speech that extend over more than one segment:
Intonation
Intonation refers to the contrastive use of pitch or melody in speech (ger. Tonhöhenverlauf, Sprechmelodie). Different levels of pitch (tones) are used in particular sequences (contours) to express a wide range of meanings. For example, we often make use of the difference between a falling and a rising pitch pattern in statements and questions.
- They’re waiting. (information)
- They’re waiting? (question)
- They’re waiting??! (surprise)
The part of a sentence over which a particular intonation pattern extends is called an intonation phrase. The intonation phrase is a unit of information rather than a syntactically defined unit, but it often overlaps with syntactic units like phrases, clauses, or sentences.
Most languages exhibit a general downward trend of pitch (declination) over the course of an intonation phrase. The completion of a full grammatical unit such as a declarative sentence is often signaled by a distinctive fall in pitch. Incomplete utterances, such as mid-sentence clause breaks where the speaker intends to show there is more coming, often exhibit a slight rise in pitch.
Connected Speech
Speech is a continuous stream of sounds without a definite borderline between each word. When we communicate with each others, we adapt our pronunciation to our audience and tend to speak at a pace which is convenient for us, rather than speaking clearly. This causes changes to the ‘shape’ of words. As a result, certain words are lost, and some phonemes are linked together while speaking. These changes are described as features of connected speech.
Among the phonological processes that affect connected speech are:
Features of Connected Speech
Weak Forms: Some English words can occur in a full and a weak form, because English exhibits qualities of a stress-timed language. That means that, while we try to keep an equal interval between stressed syllables and give the phrase rhythm, we tend to leave out non-essential words. Consequently, conjunctions, pronouns and articles (i.e. function words) are often reduced or even lost.
Examples of words which have weak forms are:
- and: fish and chips. (fish´n chips)
- can: She can dance better than I can. (1st “can”= weak, 2nd “can” = full)
- of: A cup of tea.
- have: Have you eaten? (weak)/ Yes, I have. (full)
- should: Well, you should have told me. (“should” and “have” are weak)
Assimilation: This process alters sounds so that they becomes similar (partial assimilation) or identical (total assimilation) to a neighboring or nearby sound.
There are different types of assimilation: regressive/ anticipatory, progressive and reciprocal.
Examples:
Common sound deletions
- int(e)rest, sim(i)lar, lib(a)ry, diff(e)rent, t(o)night.
/ t / and / d / = consonants often elided
- chris(t)mas, san(d)wich
/ h /= this sound is often left out
- you shouldn´t (h)ave
Phrasal verbs can show how we link closing consonants and beginning vowels across word boundaries, e.g. Get out ( getout ), Come out ( cumout )
Intrusion and Linking: We often put an extra sound (/j/, /w/, /r/) between two vowel sounds, because it marks the transition sound between the two vowels. This is regarded as intrusion.
Examples:
/ j /
- I / j / agree, They / j /are here!
/ w /
- I want to/ w/eat, Do/ w/it!
/ r /
- The media / r /are to blame, Law(r)and order.
A lot of times we drag final consonants to initial vowels or vice versa, therefore consonants and vowels can be linked also.
Examples:
- Get on. (geton ), Not at all. (notatall ), Come on. (comon)
How does connected speech affect our communication?
Native speakers normally do not have a problem with unclear utterances caused by connected speech, as they can assume what the missing part could be within that context. Non-native speakers, on the other hand, sometimes have difficulty predicting which lexical item may or may not appear in a particular context. This already is a significant problem for learners. However, the non-native speaker not only has to recognise the use of reduced forms but also use them himself, unless he wants to risk sounding fairly unnatural. Furthermore, the listener will have trouble to identify the points of focus if the speaker uses too many stressed forms. In conclusion, aspects of connected speech are of significant importance for people who learn a new language.
Key Terms
Before moving on to larger units of speech, it makes sense to have a closer look at the second major building block in our phoneme inventory besides consonants: vowels.
Vowels are produced by letting air flow through the articulatory system without any significant obstruction. The vocal cords always vibrate when a vowel is produced and the continuous stream of air makes it possible to lengthen or shorten vowel sounds, a distinction that can differentiate meaning in some languages.
The central qualities that allow us to describe vowels are height, backness and roundedness. They allow us to describe where a vowel sound originates (height and backness) and the shape of the lips during articulation (roundedness).
The above graphic should give you a good idea of height and backness and in what way they affect the quality of vowels.
The vowel chart (originally developed by phonetician Daniel Jones) combines all three features (height, backness, roundedness) into a single model to describe the realization of the so-called cardinal vowels in the oral cavity.
Because the decisive articulator determining the quality of vowel sounds is the tongue, it is possible for vowel quality to change over the duration of articulation if the position of the tongue changes. If the articulatory configuration shifts from one vowel into another the resulting composite sound is a so-called diphthong (or gliding vowel).
Examples for pure vowels:
sit /ɪ/ => high front vowel (unrounded)
foot /ʊ/ => high back vowel (rounded)
man /æ/ => middle central vowel
Examples for diphthongs:
time /aɪ/
face /eɪ/
choice /ɔɪ/
Differences between dialects and sociolects of English are often marked by contrasts in vowel quality.
Suprasegmental phonology / prosodics
Suprasegmental phonology studies intonation and other aspects of speech that extend over more than one segment:
- stress is associated with syllables
- rhythm, tempo and intonation are associated with phrases and sentences
Intonation
Intonation refers to the contrastive use of pitch or melody in speech (ger. Tonhöhenverlauf, Sprechmelodie). Different levels of pitch (tones) are used in particular sequences (contours) to express a wide range of meanings. For example, we often make use of the difference between a falling and a rising pitch pattern in statements and questions.
- They’re waiting. (information)
- They’re waiting? (question)
- They’re waiting??! (surprise)
The part of a sentence over which a particular intonation pattern extends is called an intonation phrase. The intonation phrase is a unit of information rather than a syntactically defined unit, but it often overlaps with syntactic units like phrases, clauses, or sentences.
Most languages exhibit a general downward trend of pitch (declination) over the course of an intonation phrase. The completion of a full grammatical unit such as a declarative sentence is often signaled by a distinctive fall in pitch. Incomplete utterances, such as mid-sentence clause breaks where the speaker intends to show there is more coming, often exhibit a slight rise in pitch.
Connected Speech
Speech is a continuous stream of sounds without a definite borderline between each word. When we communicate with each others, we adapt our pronunciation to our audience and tend to speak at a pace which is convenient for us, rather than speaking clearly. This causes changes to the ‘shape’ of words. As a result, certain words are lost, and some phonemes are linked together while speaking. These changes are described as features of connected speech.
Among the phonological processes that affect connected speech are:
- assimilation (changing sounds)
- elision (losing sounds)
- intrusion and linking (adding or joining sounds between words)
Features of Connected Speech
Weak Forms: Some English words can occur in a full and a weak form, because English exhibits qualities of a stress-timed language. That means that, while we try to keep an equal interval between stressed syllables and give the phrase rhythm, we tend to leave out non-essential words. Consequently, conjunctions, pronouns and articles (i.e. function words) are often reduced or even lost.
Examples of words which have weak forms are:
- and: fish and chips. (fish´n chips)
- can: She can dance better than I can. (1st “can”= weak, 2nd “can” = full)
- of: A cup of tea.
- have: Have you eaten? (weak)/ Yes, I have. (full)
- should: Well, you should have told me. (“should” and “have” are weak)
Assimilation: This process alters sounds so that they becomes similar (partial assimilation) or identical (total assimilation) to a neighboring or nearby sound.
There are different types of assimilation: regressive/ anticipatory, progressive and reciprocal.
- regressive/ anticipatory: articulation of the following sound will be anticipated. In most cases assimilation is regressive
- progressive: articulation of a sound continues in the next sound, which means it will be maintained. Progressive assimilation is rare.
- reciprocal: two sounds that produces a third one. (Example: don’t you)
Examples:
Common sound deletions
- int(e)rest, sim(i)lar, lib(a)ry, diff(e)rent, t(o)night.
/ t / and / d / = consonants often elided
- chris(t)mas, san(d)wich
/ h /= this sound is often left out
- you shouldn´t (h)ave
Phrasal verbs can show how we link closing consonants and beginning vowels across word boundaries, e.g. Get out ( getout ), Come out ( cumout )
Intrusion and Linking: We often put an extra sound (/j/, /w/, /r/) between two vowel sounds, because it marks the transition sound between the two vowels. This is regarded as intrusion.
Examples:
/ j /
- I / j / agree, They / j /are here!
/ w /
- I want to/ w/eat, Do/ w/it!
/ r /
- The media / r /are to blame, Law(r)and order.
A lot of times we drag final consonants to initial vowels or vice versa, therefore consonants and vowels can be linked also.
Examples:
- Get on. (geton ), Not at all. (notatall ), Come on. (comon)
How does connected speech affect our communication?
Native speakers normally do not have a problem with unclear utterances caused by connected speech, as they can assume what the missing part could be within that context. Non-native speakers, on the other hand, sometimes have difficulty predicting which lexical item may or may not appear in a particular context. This already is a significant problem for learners. However, the non-native speaker not only has to recognise the use of reduced forms but also use them himself, unless he wants to risk sounding fairly unnatural. Furthermore, the listener will have trouble to identify the points of focus if the speaker uses too many stressed forms. In conclusion, aspects of connected speech are of significant importance for people who learn a new language.
Key Terms
- vowels
- height
- backness
- roundedness
- monothongs, diphthongs
- suprasegmental phonology /prosodics
- pitch
- intonation
- connected speech
- weak forms
- assimilation
- elision
- intrusion / linking
Dec
04
2007
04
2007
--
Presentation for Session 5
Below is the link to the presentation for last week’s session.
Session 5 – Applications of Phonetics and Phonology
The following issues will be briefly discussed next session, before we move into new territory (morphology):
Session 5 – Applications of Phonetics and Phonology
The following issues will be briefly discussed next session, before we move into new territory (morphology):
- prosodics
- connected speech
Nov
28
2007
28
2007
--
Session 4: Sound and Meaning
Distribution of allophones
Before looking at sequences of speech sounds and how they are arranged in words and utterances, it pays off to have another look at how the allophones of a phoneme are distributed, specifically what variants can occur in what kinds of surroundings (have a look at the previous summary if you need a refresher on those two terms).
Analyzing slightly larger units of speech allows us explain some of the variation in how an individual phoneme is realized. If one allophone of a phoneme can always be found in a certain place and never in another, this is described as complementary distribution:
An example to illustrate this is the phoneme /p/ with its realization as [pʰ] – the little h indicated that the sound is aspirated – and [p], which is unaspirated. Both sounds are allophones of the phoneme /p/ and they occur in complementary distribution, meaning one always occurs where the other doesn’t and vice versa. The aspirated version occurs when /p/ is the syllable onset and followed by a stressed vowel (as in the word pin), while the unaspirated version occurs in all other situations (e.g. in spin or top). It is important to note that while the two sounds are actually phonetically slightly different, exchanging one for the other (e.g. aspirating the /p/ in top) would not have any effect on the meaning of the word. If that seems strange to you, consider again that other languages are different in that respect and that in them these two variants of /p/ may not be allophones but distinct phonemes. Finally, another source of phonetic difference in how a phoneme is realized is free variation. Free variants are the result of individual or dialectal differences, such as vowel quality in different varieties of British and American English.
Syllables
If phones are the smallest units of measurement in speech production, syllables are what follows them. Essentially a syllable is a vowel with optional consonants clustered around it. The vowel forms the so-called nucleus of the syllable, while any consonants coming after the vowel are referred to as the coda. Depending on whether the nucleus is followed by a coda or not, we describe the syllable as either closed or open.
The example below shows how syllable structure can be accurately described:
word: map (phonetically [mæp])
syllable structure: CVC
The word map has one syllable. It consists of the consonant [m], followed by the vowel [æ] and ends with the consonant [p] – therefore it has the syllable structure CVC. This is an example for a closed syllable (the vowel nucleus is followed by a consonant coda).
By contrast, the following example is an open syllable and does not have a coda:
you (phonetically [yu])
CV
Remember that the letter y may represent a consonant (as in this example) or a vowel, as in happy. Don’t be fooled by the strange spelling conventions of English!
Consonant Clusters
Several consonants can stack at the beginning or end of a syllable, forming a so-called consonant cluster. The following example demonstrates this:
nests
CVCCC
In English, a maximum of three consonants can stack at the beginning of a syllable, while a maximum of four consonants can succeed the nucleus, leading to this phonotactic description of syllable structure:
(CCC)V(CCCC)
Here are a few more examples for better illustration:
splice
CCCVC
monosyllabic (one syllable), closed
easy
VCV
disyllabic (two syllables), open
axe
VCC
monosyllabic (one syllable), closed
ex-pla-na-tion
VCC CCV CV CVC
polysyllabic (four syllables), syllables are closed-open-open-closed
Stress
Stress is a means of emphasizing syllables (or, in some cases, words) in spoken language. In contrast to French (as one example) where words are generally stressed on the last syllable, English does not have a fixed word stress, as these examples show:
‘asymmetrical
ab‘duction
emplo‘yee
Stress also serves an important grammatical function in English, as it is capable of indicating word class. For example, the word survey can be either a verb or a noun:
(1) We want to surVEY all viewers of Channel 5 in order to learn more about their tastes.
(2) This SURvey indicates that the students are extremely bored.
In the first sentence survey is a verb and stressed on the second syllable, whereas in the second sentence it is a noun and stressed on the first syllable. Generally function words such as and, to and of (which are often monosyllabic) are unstressed in English.
Have a look at the following sentence and think about how shifting word stress affects the meaning.
(3) JOHN doesn’t like pie.
= John doesn’t like pie, Mary does
(4) John DOESN’T like pie.
= Someone assumed that he likes pie, but he actually doesn’t
(5) John doesn’t LIKE pie.
= He doesn’t just like it, he loves it!
(6) John doesn’t like PIE.
= He doesn’t like pie, but he’s crazy about donuts.
Key Terms
Before looking at sequences of speech sounds and how they are arranged in words and utterances, it pays off to have another look at how the allophones of a phoneme are distributed, specifically what variants can occur in what kinds of surroundings (have a look at the previous summary if you need a refresher on those two terms).
Analyzing slightly larger units of speech allows us explain some of the variation in how an individual phoneme is realized. If one allophone of a phoneme can always be found in a certain place and never in another, this is described as complementary distribution:
Complementary distribution is the mutually exclusive relationship between two phonetically similar segments. It exists when one segment occurs in an environment where the other segment never occurs.from the SIL Glossary of Linguistic Terms
An example to illustrate this is the phoneme /p/ with its realization as [pʰ] – the little h indicated that the sound is aspirated – and [p], which is unaspirated. Both sounds are allophones of the phoneme /p/ and they occur in complementary distribution, meaning one always occurs where the other doesn’t and vice versa. The aspirated version occurs when /p/ is the syllable onset and followed by a stressed vowel (as in the word pin), while the unaspirated version occurs in all other situations (e.g. in spin or top). It is important to note that while the two sounds are actually phonetically slightly different, exchanging one for the other (e.g. aspirating the /p/ in top) would not have any effect on the meaning of the word. If that seems strange to you, consider again that other languages are different in that respect and that in them these two variants of /p/ may not be allophones but distinct phonemes. Finally, another source of phonetic difference in how a phoneme is realized is free variation. Free variants are the result of individual or dialectal differences, such as vowel quality in different varieties of British and American English.
Syllables
If phones are the smallest units of measurement in speech production, syllables are what follows them. Essentially a syllable is a vowel with optional consonants clustered around it. The vowel forms the so-called nucleus of the syllable, while any consonants coming after the vowel are referred to as the coda. Depending on whether the nucleus is followed by a coda or not, we describe the syllable as either closed or open.
The example below shows how syllable structure can be accurately described:
word: map (phonetically [mæp])
syllable structure: CVC
The word map has one syllable. It consists of the consonant [m], followed by the vowel [æ] and ends with the consonant [p] – therefore it has the syllable structure CVC. This is an example for a closed syllable (the vowel nucleus is followed by a consonant coda).
By contrast, the following example is an open syllable and does not have a coda:
you (phonetically [yu])
CV
Remember that the letter y may represent a consonant (as in this example) or a vowel, as in happy. Don’t be fooled by the strange spelling conventions of English!
Consonant Clusters
Several consonants can stack at the beginning or end of a syllable, forming a so-called consonant cluster. The following example demonstrates this:
nests
CVCCC
In English, a maximum of three consonants can stack at the beginning of a syllable, while a maximum of four consonants can succeed the nucleus, leading to this phonotactic description of syllable structure:
(CCC)V(CCCC)
Here are a few more examples for better illustration:
splice
CCCVC
monosyllabic (one syllable), closed
easy
VCV
disyllabic (two syllables), open
axe
VCC
monosyllabic (one syllable), closed
ex-pla-na-tion
VCC CCV CV CVC
polysyllabic (four syllables), syllables are closed-open-open-closed
Stress
Stress is a means of emphasizing syllables (or, in some cases, words) in spoken language. In contrast to French (as one example) where words are generally stressed on the last syllable, English does not have a fixed word stress, as these examples show:
‘asymmetrical
ab‘duction
emplo‘yee
Stress also serves an important grammatical function in English, as it is capable of indicating word class. For example, the word survey can be either a verb or a noun:
(1) We want to surVEY all viewers of Channel 5 in order to learn more about their tastes.
(2) This SURvey indicates that the students are extremely bored.
In the first sentence survey is a verb and stressed on the second syllable, whereas in the second sentence it is a noun and stressed on the first syllable. Generally function words such as and, to and of (which are often monosyllabic) are unstressed in English.
Have a look at the following sentence and think about how shifting word stress affects the meaning.
(3) JOHN doesn’t like pie.
= John doesn’t like pie, Mary does
(4) John DOESN’T like pie.
= Someone assumed that he likes pie, but he actually doesn’t
(5) John doesn’t LIKE pie.
= He doesn’t just like it, he loves it!
(6) John doesn’t like PIE.
= He doesn’t like pie, but he’s crazy about donuts.
Key Terms
- complementary distribution – free variation
- aspiration
- syllables
- nucleus – coda
- closed- open
- consonant clusters
- syllable stress, word stress
Nov
25
2007
25
2007
--
YouTube Linguistics – Xhosa, New York English
Clicks in Xhosa
The clip above features a speaker of Xhosa, an agglutinative tonal language of the Bantu family, spoken in South Africa. The distinctive clicks of Xhosa are just one example for a speech sound with phoneme status that does not exist in English. Note that tonal languages (in which intonation distinguishes meaning) make up a significant percentage of the world’s languages.
New York English 1
New York English 2
New York English 3
These three clips give an interesting introduction into the phonology of New York English – and into the culture as well. Pay attention not just to the examples that Hipstomp gives, but also to how his speech is affected when he is talking quickly. We will discuss some aspects of connected speech in the next session.
The clip above features a speaker of Xhosa, an agglutinative tonal language of the Bantu family, spoken in South Africa. The distinctive clicks of Xhosa are just one example for a speech sound with phoneme status that does not exist in English. Note that tonal languages (in which intonation distinguishes meaning) make up a significant percentage of the world’s languages.
New York English 1
New York English 2
New York English 3
These three clips give an interesting introduction into the phonology of New York English – and into the culture as well. Pay attention not just to the examples that Hipstomp gives, but also to how his speech is affected when he is talking quickly. We will discuss some aspects of connected speech in the next session.
Nov
22
2007
22
2007
--
Presentation for Session 4
Below is the link to the presentation for today’s session.
Session 4 – Sound and Meaning
The following issues will be repeated next session in particular:
Session 4 – Sound and Meaning
The following issues will be repeated next session in particular:
- free variation and complementary distribution of allophones
- syllable structure
- stress
Nov
21
2007
21
2007
Session 3: The Sounds of English
The two primary fields of linguistics concerned with speech sounds – those sounds that are used by humans to communicate – are phonetics and phonology. Both areas are mutually dependent. Phonetics describes the concrete, physical dimension of sounds, such as whether they are voiced or voiceless and their place and manner of articulation. The aspect of sound production is particularly what articulatory phonetics is concerned with, while acoustic and auditory phonetics deal with the characteristics of sound waves and how they are perceived by the human ear.
While phonetics deals with the form of sounds (how they are produced, heard and how they can be described), phonology is concerned with the function of sounds, that is, with their meaning in a given language. By systematically studying phonological differences between languages, it is for example possible to predict what sounds the learner of a second language will have difficulties with or why certain languages are judged as more difficult to learn in terms of pronunciation than others.
Phones
The basic unit of phonetics is called a phone, which is basically any human speech sound. Remember, phonetics is only concerned with “sounds as such”, so any sound that comes out of a person’s mouth can be called a phone. In contrast to this, the basic unit in phonology is the phoneme, which is any sound in a language that differentiates meaning. In linguistic contexts, phones are often expressed by placing brackets around a transcription (e.g. [dæns] for American dance).
Phonemes
The relationship of sound and meaning can be explained by looking into whether a difference in sound structure causes a shift in meaning or not. Try this by saying the following words out loud:
look – book – cook – took
You will notice right away that their sound patterns are similar except for the initial sound (l, b, c, t). The fact that replacing one sound with another (for example, l with b) yields a different meaningful word in English demonstrates that the speech sounds l and b are phonemes in the English language. Linguists normally write phonemes with slashes around the transcriptions, e.g. /l/ and /b/. An case like look – book that demonstrates that /l/ and /b/ are phonemes is called a minimal pair.
Now compare this with another example:
tea – he
At first this might be confusing. While the spelling of look, book etc happens to be similar, except at the beginning of the word where the distinct phoneme occurs, the spelling of tea and he is not similar. But they still form a minimal pair for the phonemes /t/ and /h/, because the rest of the sound pattern is identical.
The key here is to recognize that we are dealing with sounds, not spelling. Two sounds may be distinct phonemes while being represented by the same letters, or be completely identical in terms of sound structure but look different in writing.
Examples:
see – sea
Identical sound, different spelling – not a minimal pair, because we’re looking at a difference that exists only in writing.
the – me
The final sound looks similar in writing, but is there is an obvious sound difference between short and long e. However, these two words are not suitable candidates for a minimal pair test, as the rest of the sound pattern is not identical.
Allophones
What, then, about sounds which are different but do not differentiate meaning? Take this example:
lip – pill
While the difference is slight, you might notice that /l/ does not sound exactly the same in lip and pill (try to keep track of where you place the tip of your tongue). Such a difference depends on many factors – in this case whether the sound is at the beginning or end of the word. Other examples for such factors include dialectal differences (think about how British vs. American speakers say dance or France) and there is even a certain degree of difference among individuals. The decisive contrast between this and the examples above is that such variants don’t differentiate meaning. The ls in lip and pill are both allophones of the phoneme /l/.
What is it good for?
Why is it important whether we are dealing with allophones of the same phoneme or with entirely different phonemes? Have a look at this table describing the phoneme inventory of Standard Mandarin, the official language of China. One difference that you are likely to notice is that Standard Mandarin lacks the voiced bilabial, alveolar and velar plosives /b/, /d/ and /g/. However, it has aspirated (ger. behaucht) versions of these consonants, which are distinct phonemes: pʰ, tʰ, kʰ. In other words, a difference that does not distinguish meaning in English (aspiration) is a salient difference in Mandarin Chinese, while another one (voicing of plosives) distinguishes meaning in English but not in Chinese.
Side by side
The following table gives an summarizing overview of the differences between phonetics and phonology.
The human vocal tract
What is for linguistic purposes identified as the vocal tract fills several functions, among them breathing and ingesting food. The production of speech sounds is essentially realized by directing the flow of air through the articulatory system in specific ways – for example by letting air escape gradually in a sort of hiss, by letting air pressure build and then suddenly releasing it, by letting the vocal cords vibrate etc.
Describing speech sounds
Speech sounds are usually described via their articulatory qualities, i.e. their
One basic phonetic differentiation that can be made when classifying speech sounds is that they fall into two relatively distinct categories:
Place of articulation
The following list describes the main places of articulation for English consonants. Note that the use of certain places and manners of articulation is common in some languages but not in others. For example, Arabic has two pharyngeal consonants that English lacks (/ħ/ and /ʕ/). An example that you are familiar with are the Umlaut vowels that occur in German but not in English and the dental fricative (the “th”) that is common in English but not in German.
Lips
A sound that is produced by pressing the lips together is called bilabial.
Sounds: /p/, /b/, /m/ (and, to some extent, the labial-velar approximant /w/)
A sound that involves using the lips and teeth together is described as labio-dental.
Sounds: /f/, /v/
Teeth
A sound that is created by placing the tip of the tongue behind the upper teeth is dental.
Sounds: /θ/, /ð/
Alveolar ridge
A sound that is produced by tapping the tongue against the area a bit behind the teeth (called the alveolar ridge) is referred to as alveolar.
Sounds: /t/, /d/, /n/, /s/, /z/, /ɹ/, /l/
Alveolar ridge and hard palate
A sound that originates between the alveolar ridge and the hard palate is called palato-alveolar or post-alveolar.
Sounds: /ʃ/, /ʒ/, /tʃ/, /dʒ/
Hard palate
A sound that comes from the middle section of the roof of the mouth (the hard palate) is called palatal.
Sound: /j/
Soft palate (velum)
A sound the is produced in the upper back area of the mouth (the soft palate or velum) is described as velar.
Sounds: /k/, /g/, /ŋ/
Throat
A sound that originates in the throat (or, more specifically, the glottis) is referred to as glottal.
Sound: /h/
Manner of articulation
The term manner of articulation is generally used to explain how a sound is produced. Place, manner and voicing are usually named together, allowing us to describe /z/ as a voiced alveolar fricative.
Plosives
These sounds (also referred to as stops) occur when there is an initial blockage of both the oral and nasal cavities of the vocal tract (and therefore no air flow), which is then suddenly released.
Sounds: /p/, /b/, /t/, /d/, /k/, /g/
Nasals
When articulating a nasal the air flow completely bypasses the oral cavity, instead flowing through the nose. The precise position of the tongue during articulation determines the resulting sound.
Sounds: /m/, /n/
Fricatives
Fricatives are produced by forcing air through a narrow channel made by placing two articulatory organs closely together (for example, upper lip and lower teeth in /f/).
Sounds: /f/, /v/, /θ/, /ð/, /s/, /z/, /ʃ/, /ʒ/, /h/
Africatives
Africatives combine plosive with fricative qualities, first blocking the air stream and then slowly releasing it (in contrast to normal plosives, which release pressure suddenly).
Sounds: /tʃ/, /dʒ/
Approximants
The articulation of these speech sounds involves only very slight obstruction of the air flow, which is why some approximants are considered to be relatively close to vowels (so-called half-vowels). The exact realization of the approximant /r/ (as either /ɹ/ or /ɻ/) is one of the characteristic differences between British and American English dialects*.
Sounds: /l/ (lateral-alveolar approximant), /j/ (palatal approximant), /w/ (labial-velar approximant), /ɹ/ (lateral approximant in British RP), /ɻ/ (retroflex approximant in American English)
* Note that there is really no single British or American English in the precise linguistic sense. British could theoretically include Scots (which is widely regarded as a seperate language), Received Pronunciation and Cockney, while American would geographically conflate Canadian English with Southern US-American and countless other regional and social varieties. Always keep in mind that tags like British English and American English are idealized blanket labels which are generally not specific enough for linguistic purposes.
Key Terms
While phonetics deals with the form of sounds (how they are produced, heard and how they can be described), phonology is concerned with the function of sounds, that is, with their meaning in a given language. By systematically studying phonological differences between languages, it is for example possible to predict what sounds the learner of a second language will have difficulties with or why certain languages are judged as more difficult to learn in terms of pronunciation than others.
Phones
The basic unit of phonetics is called a phone, which is basically any human speech sound. Remember, phonetics is only concerned with “sounds as such”, so any sound that comes out of a person’s mouth can be called a phone. In contrast to this, the basic unit in phonology is the phoneme, which is any sound in a language that differentiates meaning. In linguistic contexts, phones are often expressed by placing brackets around a transcription (e.g. [dæns] for American dance).
Phonemes
The relationship of sound and meaning can be explained by looking into whether a difference in sound structure causes a shift in meaning or not. Try this by saying the following words out loud:
look – book – cook – took
You will notice right away that their sound patterns are similar except for the initial sound (l, b, c, t). The fact that replacing one sound with another (for example, l with b) yields a different meaningful word in English demonstrates that the speech sounds l and b are phonemes in the English language. Linguists normally write phonemes with slashes around the transcriptions, e.g. /l/ and /b/. An case like look – book that demonstrates that /l/ and /b/ are phonemes is called a minimal pair.
Now compare this with another example:
tea – he
At first this might be confusing. While the spelling of look, book etc happens to be similar, except at the beginning of the word where the distinct phoneme occurs, the spelling of tea and he is not similar. But they still form a minimal pair for the phonemes /t/ and /h/, because the rest of the sound pattern is identical.
The key here is to recognize that we are dealing with sounds, not spelling. Two sounds may be distinct phonemes while being represented by the same letters, or be completely identical in terms of sound structure but look different in writing.
Examples:
see – sea
Identical sound, different spelling – not a minimal pair, because we’re looking at a difference that exists only in writing.
the – me
The final sound looks similar in writing, but is there is an obvious sound difference between short and long e. However, these two words are not suitable candidates for a minimal pair test, as the rest of the sound pattern is not identical.
Allophones
What, then, about sounds which are different but do not differentiate meaning? Take this example:
lip – pill
While the difference is slight, you might notice that /l/ does not sound exactly the same in lip and pill (try to keep track of where you place the tip of your tongue). Such a difference depends on many factors – in this case whether the sound is at the beginning or end of the word. Other examples for such factors include dialectal differences (think about how British vs. American speakers say dance or France) and there is even a certain degree of difference among individuals. The decisive contrast between this and the examples above is that such variants don’t differentiate meaning. The ls in lip and pill are both allophones of the phoneme /l/.
What is it good for?
Why is it important whether we are dealing with allophones of the same phoneme or with entirely different phonemes? Have a look at this table describing the phoneme inventory of Standard Mandarin, the official language of China. One difference that you are likely to notice is that Standard Mandarin lacks the voiced bilabial, alveolar and velar plosives /b/, /d/ and /g/. However, it has aspirated (ger. behaucht) versions of these consonants, which are distinct phonemes: pʰ, tʰ, kʰ. In other words, a difference that does not distinguish meaning in English (aspiration) is a salient difference in Mandarin Chinese, while another one (voicing of plosives) distinguishes meaning in English but not in Chinese.
Side by side
The following table gives an summarizing overview of the differences between phonetics and phonology.
phonetics | phonology |
sounds as such | sounds as parts of a sound system |
language use (parole) | language system (langue) |
language-independent | language-dependent |
substance | function |
concrete | abstract |
phone [ ] | phoneme / / |
What is for linguistic purposes identified as the vocal tract fills several functions, among them breathing and ingesting food. The production of speech sounds is essentially realized by directing the flow of air through the articulatory system in specific ways – for example by letting air escape gradually in a sort of hiss, by letting air pressure build and then suddenly releasing it, by letting the vocal cords vibrate etc.
Describing speech sounds
Speech sounds are usually described via their articulatory qualities, i.e. their
- place of articulation (where in the vocal tract they are generated)
- manner of articulation (how they are generated)
- voiced or voiceless (whether they make the vocal cords vibrate or not )
One basic phonetic differentiation that can be made when classifying speech sounds is that they fall into two relatively distinct categories:
- vowels, which are produced by letting air flow through the articulatory system without any constraints and
- consonants, which feature some sort of obstruction of the air flow in the vocal tract
Place of articulation
The following list describes the main places of articulation for English consonants. Note that the use of certain places and manners of articulation is common in some languages but not in others. For example, Arabic has two pharyngeal consonants that English lacks (/ħ/ and /ʕ/). An example that you are familiar with are the Umlaut vowels that occur in German but not in English and the dental fricative (the “th”) that is common in English but not in German.
Lips
A sound that is produced by pressing the lips together is called bilabial.
Sounds: /p/, /b/, /m/ (and, to some extent, the labial-velar approximant /w/)
A sound that involves using the lips and teeth together is described as labio-dental.
Sounds: /f/, /v/
Teeth
A sound that is created by placing the tip of the tongue behind the upper teeth is dental.
Sounds: /θ/, /ð/
Alveolar ridge
A sound that is produced by tapping the tongue against the area a bit behind the teeth (called the alveolar ridge) is referred to as alveolar.
Sounds: /t/, /d/, /n/, /s/, /z/, /ɹ/, /l/
Alveolar ridge and hard palate
A sound that originates between the alveolar ridge and the hard palate is called palato-alveolar or post-alveolar.
Sounds: /ʃ/, /ʒ/, /tʃ/, /dʒ/
Hard palate
A sound that comes from the middle section of the roof of the mouth (the hard palate) is called palatal.
Sound: /j/
Soft palate (velum)
A sound the is produced in the upper back area of the mouth (the soft palate or velum) is described as velar.
Sounds: /k/, /g/, /ŋ/
Throat
A sound that originates in the throat (or, more specifically, the glottis) is referred to as glottal.
Sound: /h/
Manner of articulation
The term manner of articulation is generally used to explain how a sound is produced. Place, manner and voicing are usually named together, allowing us to describe /z/ as a voiced alveolar fricative.
Plosives
These sounds (also referred to as stops) occur when there is an initial blockage of both the oral and nasal cavities of the vocal tract (and therefore no air flow), which is then suddenly released.
Sounds: /p/, /b/, /t/, /d/, /k/, /g/
Nasals
When articulating a nasal the air flow completely bypasses the oral cavity, instead flowing through the nose. The precise position of the tongue during articulation determines the resulting sound.
Sounds: /m/, /n/
Fricatives
Fricatives are produced by forcing air through a narrow channel made by placing two articulatory organs closely together (for example, upper lip and lower teeth in /f/).
Sounds: /f/, /v/, /θ/, /ð/, /s/, /z/, /ʃ/, /ʒ/, /h/
Africatives
Africatives combine plosive with fricative qualities, first blocking the air stream and then slowly releasing it (in contrast to normal plosives, which release pressure suddenly).
Sounds: /tʃ/, /dʒ/
Approximants
The articulation of these speech sounds involves only very slight obstruction of the air flow, which is why some approximants are considered to be relatively close to vowels (so-called half-vowels). The exact realization of the approximant /r/ (as either /ɹ/ or /ɻ/) is one of the characteristic differences between British and American English dialects*.
Sounds: /l/ (lateral-alveolar approximant), /j/ (palatal approximant), /w/ (labial-velar approximant), /ɹ/ (lateral approximant in British RP), /ɻ/ (retroflex approximant in American English)
* Note that there is really no single British or American English in the precise linguistic sense. British could theoretically include Scots (which is widely regarded as a seperate language), Received Pronunciation and Cockney, while American would geographically conflate Canadian English with Southern US-American and countless other regional and social varieties. Always keep in mind that tags like British English and American English are idealized blanket labels which are generally not specific enough for linguistic purposes.
Key Terms
- phonetics – phonology
- phone – phoneme – allophone
- vocal tract
- vowel – consonant
- voicing – place of articulation – manner of articulation
:.