The second “extra” allophone of /t/ is a glottal stop, represented by this symbol: [Ɂ]. To produce this sound, the vocal cords close tightly and then open quickly. It’s like the beginning of a small cough, or the middle sound when we say “huh-uh” to mean “no.”
The phoneme /t/ can be pronounced as a glottal stop when two things happen:
When the syllable before it is stressed, and the syllable after it is unstressed.
When the syllable after it is /ən/ or syllabic /n/. (That is, /ə/ disappears and /n/ is lengthened and becomes a whole syllable. The symbol for syllabic /n/ is [n̩].)
It’s not absolutely necessary for learners to pronounce the flap [ɾ] or glottal stop [Ɂ] allophones of /t/, but they need to understand them when they hear them. And in normal American English speech, they will hear them often.
A third allophone of /t/ occurs when /t/ comes before /r/. In this case, /t/ sounds very much like /ʧ/. The tongue is moving farther back in the mouth toward the palate, getting ready to say /r/, and so the /t/ is also pulled a little farther back. This happens naturally through the process of assimilation, a sound change in which one sound becomes more similar to a sound that comes before or after it. This happens naturally and doesn’t need to be specifically taught.
Light and dark /l/: The consonant /l/ is often said to have two allophones: “light” or alveolar /l/ (with the symbol [l] ) and “dark,” or velarized /l/ (with the symbol [ɫ]), occurring in different positions:
[l]: at the beginning of a syllable, especially before front vowels, in words like light, left, and believe.
[ɫ]: at the end of a syllable and before back vowels, in words like low, ball, and pool.
However, in the speech of most Americans, the difference between these two allophones is small, and some Americans don’t make this distinction at all. Instead, they pronounce a sound like a dark /l/ in all positions. Because of this, if your pronunciation model is American English, the distinction between dark and light /l/ does not need to be a high priority. You and your students have more important things to think about.
Consonant clusters are groups of two or more consonant sounds in a row, as in stop, strong, desk, desks, or sister. (It’s important to remember that we’re talking about groups of consonant sounds, not necessarily consonant letters. These are not always the same thing. For example, ship and sing each have groups of two consonant letters, but each group only spells one sound. On the other hand, the letter “x” as in six represents a consonant cluster of two sounds: /ks/.)
Consonant clusters in English can occur at the beginning, middle, or end of words. There are restrictions on how many consonants can occur in a particular position, and which consonants can occur together. For example, in English, /sk/ as in sky, /pl/ as in play, and /spr/ as in spring are possible combinations at the beginning of a word, but /sd/, /fp/, and /zpr/ are not.
The Keynote “Syllable Structure and Consonant Clusters” gives more information and examples of what kinds of consonant clusters are possible in English.
Simplification of consonant clusters: We generally encourage students to pronounce every sound in a consonant cluster. However, there is one situation when it’s acceptable to simplify a consonant cluster, that is, to omit one of the consonants. When there are three or more consonants in a row, the middle one is sometimes dropped. (The first or last consonant is not dropped.) This happens most often when the middle consonant is a stop, /θ/, or /ð/. For example:
tests might sound like /tɛsts/ or /tɛs/
asked might sound like /æskt/ or /æst/
months might sound like /mʌnθs/ or /mʌns/
sixths might sound like /sɪksθs/ or /sɪks/
Another way native speakers make consonant clusters easier to pronounce is by resyllabification. That is, they split up a consonant cluster so that the last consonant goes with the syllable after it. For example when we say: The cats are sleeping, the final /s/ in cats sounds like it joins the following word: The cat sare sleeping.
Learner problems with consonants
The problems learners have in pronouncing new sounds will vary, depending on the learners’ native language and its sound system. To predict and explain what problems learners might have, teachers need at least a basic knowledge of the phonological system of the learner’s language. Some sources of problems can be:
New sounds: When learners try to pronounce a sound that doesn’t exist in their own language, it’s naturally difficult at first, and they may substitute a similar (but not identical) sound from their own language. For example, many languages don’t contain the phonemes /θ/ or /ð/, so speakers of those languages often have a hard time hearing and distinguishing these new sounds. When they try to say the new sounds, they often substitute more familiar sounds, like /s/, /f/, or /t/ for /θ/ and /z/, /v/, or /d/ for /ð/. It’s important to help students hear and understand that there actually is a difference between the new sounds and the familiar first-language sounds so they can begin to pronounce the new sounds more accurately.
Familiar sounds in new environments: There can also be sounds that are easy for learners to pronounce in some phonetic environments, but difficult in others. For example, the glide /w/ is not a problem for Japanese or Korean speakers when followed by most vowels. Saying we, way, or wine is not hard. However, when /w/ is followed by /uw/ or /ᴜ/, it’s more of a problem. Words like woman, wood, and woo are a pronunciation challenge. This is because those languages have sound combinations similar to /wi/ /we/, and /wa/, but not combinations like /wu/. The fact that the sounds of /w/ and /uw/ are very similar can also make it hard for learners to pronounce them in sequence, and “wood” can end up sounding like /uwd/.
The same situation happens with /y/ before the similar vowel sounds /iy/ and /ɪ/. It’s hard for many learners to distinguish year and ear or yeast and east.
Final consonants: Consonants at the ends of words are often more troublesome than the same consonants at the beginnings of words. This is especially true for students whose native language does not allow any consonants at the ends of words, or perhaps only a limited set of consonants.
When learners have trouble pronouncing final consonants, they cope in different ways, depending partly on their language background. Speakers of some languages tend to omit final consonants. For example, they might pronounce meet as /miy/ or back as /bæ/. Speakers of other languages might add an extra vowel after the final consonant, pronouncing meet as /miytə/ or back as /bæku/.
Because these changes fit the pattern of what the learners are used to doing in their own language, they don’t realize that they’re changing anything. They unconsciously reshape new words to fit the familiar pattern of their own language. Still, both of these types of changes can make it hard for listeners to recognize the words that the speakers are trying to say.
Consonant clusters: Different languages also have different restrictions on what kinds of syllables and consonant combinations are possible. Some languages don’t have consonant clusters at all. Others have fewer clusters than English, or they allow different combinations of consonants. (On the other hand, some languages have even more consonant clusters than English!) Learners whose languages have different syllable structure rules than English may have trouble pronouncing some words with consonant clusters.
Learners cope with unfamiliar consonant clusters in different ways. They might omit one or more of the consonants. For example, they might pronounce section as /sɛʃən/ or spring as /spɪŋ/ or /pɪŋ/. Other learners add an extra vowel before or between the consonants. For example, school might become /ɛskuwl/ or spring might become /sᴜpᴜrɪŋ/.
(These changes follow a similar pattern to the ways learners deal with unfamiliar final consonants.)
All of these are changes that learners unconsciously produce to make words more comfortable and easier to pronounce. However, they also make it much harder for listeners to understand what the speaker is trying to say. Teachers need to help students understand and practice the patterns of English syllable structure to make their speech more understandable.
What about table titles and borders?
Boxes: Consonants (phonemic symbols and examples)
Demonstrating Manner of Articulation
Here are tips for illustrating a few of the manners of articulation. (Sorry, I don’t have tips for all of them.) Click on the tiny pictures to enlarge them.
Stops: A pop gun works like a stop. The tube is like the vocal tract. The cork plugs the tube, then when you push the handle, air builds up and pushes the cork out. This is how a stop works.
Fricatives: When you push air out of a balloon pump, it passes out through a small opening, making a hissing sound like a fricative.
Nasals: To show that air comes out the nose rather than the mouth, have students hold a mirror under their noses and say /m/. The mirror should fog up. Or have them hold their noses closed. If no sound comes out, it’s a nasal.
Liquids: Use the image of water (liquid) flowing smoothly in a calm river to illustrate these sounds. Contrast them with the “rougher” types of sounds, which are more like a roughly flowing river with lots of rocks and rapids.
Classification of American English Consonant Phonemes
Keynote: Some changes in consonant sounds
Examples of flaps and glottal stops
Allophones of voiceless stops
Keynote: syllable structure and consonant clusters
Voicing of English consonants
Place of articulation
Manner of articulation
The Vowels of American English
How do we describe vowels?
Vowels are sounds in which the air stream moves through the vocal tract very smoothly; there’s nothing blocking or constricting it. The first sounds in the words extra, only, and apple are vowels. In general, every word and every syllable must have a vowel sound (although, as we saw in the last chapter, the consonants /n/, /l/, and /r/ can sometimes be stretched out to be a syllable in themselves). Vowels are the “heart” of syllables.
The vowel sounds are not exactly the same in all varieties of English. In fact, when we compare different dialects of English, both the major national varieties and the many dialects within each variety, we can easily see that vowel sounds vary much more than consonants.
For most speakers of American English, there are 14 vowel sounds, or 15 if we include the vowel-like sound in words like bird, her and turn. Sometimes people assume that there are only five vowels in English: A, E, I, O and U, but this is incorrect. These are vowel letters, not vowel sounds. English has many more vowel sounds than vowel letters. Each vowel letter can represent more than one sound. For example, the letter “a” can be used in spelling /æ/ as in hat, /ey/ as in hate, /ɑ/ as in car, or /ɛ/ as in care.) There’s not a one-to-one correspondence between letters and sounds.
Categories for describing vowels
When we describe vowels, we need a new set of categories. When we describe the vowels of English, we talk about:
Tense and lax vowels
Simple vowels, glided vowels, and diphthongs
Why do we need different categories than for consonants? The consonant categories just don’t work very well with vowels. Since all vowels in English are voiced, there’s no point talking about voiced and voiceless vowels. Method of articulation doesn’t work well because vowels are all pronounced very similarly—with no blockage or friction in the mouth. (In fact, we could think of “vowel” as one more method of articulation.) Place of articulation almost works as a category, but since the no parts of the articulatory system actually touch when we say a vowel, we talk about “tongue position” instead of telling which parts of the articulatory system are touching.
When you say a vowel, even a small change in the position of your tongue can make a big change in how the vowel sounds.
Where is the highest, tensest, or most active part of the tongue? It’s difficult to use “place of articulation” to describe where this part is, since the tongue isn’t actually touching anything. Instead, the way we describe this position is something like graphing a point in math or finding a location on a map. We give two “coordinates” to describe where this point is, like the x and y coordinates of a point on a graph in math or the latitude and longitude of a point on a map.
Vertical position: high / mid / low
Horizontal position: front / central / back
We can describe the tongue position for vowels by telling both of these positions. For example,
a low front vowel = /Q/ as in “cat.”
a mid central vowel = /√/ as in “but.”
a high back vowel = /uw/ as in “boot.”
We can show the tongue positions for different vowels by using a diagram called a “vowel quadrant.”
The vowel quadrant is a way of representing the tongue position for vowels. Each section of the diagram shows a different tongue placement, vertically and horizontally.
Are the lips very rounded, somewhat rounded, relaxed, stretched wide?
The shape of the lips can change the sound of vowels.
Are the lips very rounded, just a little rounded, relaxed, or stretched wide?
In English, the vowels /uw/, /U/, /ow/, and /ç/ are pronounced with the lips rounded.
The vowels /iy/, /I/, /ey/, /E/, /Q/, /√/, /´/, and /a/ are unrounded.
Tense and Lax Vowels
How tense or tight are the muscles of the tongue during the sound? This isn’t totally real. Look at Ladefoged again.
Lax vowels don’t come at end of words
Simple vowels, glided vowels, diphthongs
Combine simple and glided vowels? A two-way contrast wouldn’t confuse people so much. Make it a continuum? More like reality.
Is the sound constant throughout the vowel, or does it change a bit at the end?
If the tongue stays in the same position throughout the vowel, it’s a simple vowel.
If the tongue position changes just a little, it’s a glided vowel.
If the tongue position changes a lot, so it sounds like two separate vowel sounds blended together, it’s a diphthong.
When people (especially British phonologists or people talking about phonics) call vowels “long” or “short,” they don’t mean that they’re identical except for length. We need to disabuse students of the idea that the only difference between vowel pairs like /iy/ and /I/ or /uw/ and /U/ is that one is long and the other is short. Many of them have learned this, but it just doesn’t work.
/ɚ/ as a vowel
American vs. British vowels
Partly the difference is in the sounds themselves and the vowel system.
Partly it’s in the way phonologists describe them.
Some people want to emphasize the fact that some of the vowels are actually two sounds combined, so they prefer the symbols like /ey/ and /ow/. Some want to do that long/short vowel thing, which REALLY doesn’t work for American English, so they use i: and u: and confuse the heck out of people.
a comparison between a typical American phonemic alphabet and a typical British phonemic alphabet. You can see that the consonant symbols are very similar, but the vowel symbols are very different. You might wonder why we need different sets of phonemic symbols for two varieties of the same language. There are two main reasons:
The vowels and vowel systems of NAE and RP are actually different.
There are differences in how phonologists in the two countries analyze vowels, especially diphthongs and vowel length.
Changes in vowel sounds
Vowel sounds are sometimes changed by the sounds around them.
Length: Vowels are usually shorter in duration before voiceless sounds and longer before voiced sounds. They’re longest of all when they come at the end of a word. Compare:
bed / bet bead / beat man / mast hill / hit
/r/ coloring: Some vowel contrasts are neutralized before /r/. Look at these words:
bead / bid / beer “Bead” /biyd/ and “bid” /bId/ are separate words with different vowel sounds. But we could pronounce “beer” either /bIr/ or /biyr/ without changing its meaning.
load / laud / lord In the same way, “load” /lowd/ and “laud” /lçd/ have contrasting vowels, but with “lord,” we could say /lçrd/ or /lowrd/ without changing the meaning.
/l/ coloring: To a lesser extent, vowel contrasts before /l/ are also sometimes weaker. Examples:
heal / hill / he’ll she’ll we’ll I’ll you’ll they’ll
Nasal coloring: Vowels followed by a nasal sound also tend to be nasalized. Examples:
seem seen sing can can’t
Citation form: The way we pronounce a word when we’re saying it very carefully is its citation form—the unreduced form. For example, the citation form of “to” is /tuw/.
Reduced form: The way we pronounce a word in normal speech, when it isn’t being stressed, is tis reduced form. For example, the reduced form of “to” is /t´/.
Vowel sounds are often changed (reduced) in unstressed syllables. They become less clear, and often turn to /ə/ (schwa). Only a small number of vowel sounds commonly occur in unstressed syllables.
Redo this table:
sofa aqua quota Sarah about around subtract offend
(/´/ is by far the most common vowel in unstressed syllables
water mother curtail entertainment
(This sound can be thought of as a combination of /´/ and /r/)
music reddish sleeping
city pretty candy
meadow narrow shadow hotel rosette
into venue menu
Vowels are slippery. They vary tremendously. But this doesn’t mean they don’t matter or that you can just do what you want.
Add vowel length Keynote
Boxes: Keynote-The vowels of American English
Vowels of American English—symbols and examples
Vowel quadrant (vowels)
Vowel quadrant (consonants)
Pronouncing Can and Can’t
The contrast between the words “can” and “can’t” is often especially troublesome.
“Can’t” is usually stressed. It sounds like /kQnt/ or /kQn//, with a clear /Q/ sound.
I can’t go with you. Who can’t afford a new car?
“Can” is usually unstressed. It often sounds like /k´n/ or /kn/, with a reduced vowel.
I can go with you. Who can afford a new car?
But when “can” is alone, with no verb after it, it’s stressed.
Yes, I can. I can tomorrow, but not right now.
Review (currently empty)
Teaching the Sounds
This chapter contains suggestions for introducing and practicing individual consonant or vowel sounds. Ways of practicing a broader array of pronunciation skills, emphasizing topics such as intonation, stress, and rhythm, will be in Chapter 10, “Beyond Repeat After Me.”
Before you begin
Think about: Age of students, proficiency level, goals, how much time you have
Listening and noticing need to come before successful production.
Suggested steps (from Celce-Murcia)
Introducing/re-introducing new sounds
Ways to introduce new sounds: Different age groups. How much explanation? Should I use technical terms in explanations to students? Should I use IPA/phonemic symbols? Representing sounds with colors
It’s not enough to just tell people to raise or lower their tongues or to just “point out” the difference between sounds. Noticing is good, and helping people to notice, but if they could get it right by just hearing about it, they would have done it already.
Teaching with toys and gadgets, using multisensory methods
“Scripts” for describing each sound using simple, concrete language.
/f/ and /v/: “bite your lip.” Not really a good idea. Implies that your teeth should be much lower down on your lip than they really should be. /θ/ and /ð/ stick out your tongue
Using a dental model--details about how to demonstrate sounds. Make demo video. Make them feel and understand it, not just memorize
How can you tell it’s retroflex/lateral? Put your tongue in position, then suck in air and see where it’s old.
Straw for practicing /r/ and /l/
Lollipop for th
When explaining/demoing sound: Demo --> try it. Explain --> try it. Don’t forget the “try it” part. (Input-output-input-output)
How much technical language should we use in teaching?
Is it necessary for students to learn the official names of the parts of the articulatory system? Should we use terms like “alveolar ridge” and “hard palate” in our explanations?
If the students are children, then certainly not. Children learn better through simpler, more concrete means—through models, demonstrations, imitation, and actions. Technical language is inappropriate and ineffective for teaching them.
As children reach junior high or high school age, they are better able to understand verbal explanations and abstract concepts. However, it still works better to keep explanations simple, concrete, and practical. “Show, don’t just tell” works best for this age group too.
If the students are adults, the situation is a little different. Some adults, especially those who have a scientific or medical background, appreciate knowing the real names of things. Others find them confusing or burdensome—just one more obstacle to reaching their language-learning goals. Adults also benefit from learning through pictures, models, and demonstrations, not just lectures.
It’s important to get to know your students and their attitudes and preferences, and adjust your methods and explanations you use as you learn what works best for them.
For students of any age, it is more important to feel and understand what’s happening inside their mouths than to remember the technical terms for its parts. Learning about the articulatory system is a tool to help us reach our pronunciation goals, not a goal in itself.