Phones
Using his nose, mouth, and throat, a person can make a great many distinct sounds: snorting, grunting, whistling, creaking, clicking, and so on. Most of the sounds that most languages use involve vibrating the vocal cords while passing air between the tongue and some part of the mouth, but there are many exceptions – notably the nasal sounds, which are made in the nose rather than the mouth, but also the creaks and clicks that many languages besides English do make practical use of.
Once we understand how speech sounds are made, we can classify them in a scientific way. Why would we want to? Imagine you’ve just begun to study a language. Your friend, who speaks that language, tells you that your pronunciation sounds odd, but when you ask him why, he can’t tell you – “It just sounds odd.” Not very helpful, right? Without an explanation of the difference between the sound you’re making and the sound you should be making, how can you improve your pronunciation? Phonetics gives you the tools to give and understand such explanations. If you and your friend both have knowledge of phonetics, he can tell you: “You’re aspirating your voiceless stops. They aren’t supposed to be aspirated.” Combining your phonetics with a little phonology, you can guess why you’re making this particular mistake – voiceless stops in your native language (L1) are always aspirated in certain environments, whereas in your target language (L2) they’re never aspirated. Knowing how your L1 is going to try to sneak into your L2 (L1 interference), you can police your own speech and make sure this doesn’t happen.
If all that sounded like gibberish to you, it’s time to learn what words like aspirated, voiceless, and stop mean! On this page you can learn about the various speech sounds and how they’re classified. Although the terms you’ll see here may seem unnecessarily arcane for such a basic act as speaking, remember that knowledge is power, and since speech is the ultimate human superpower, imagine how powerful a deeper knowledge of it could make you!
Kinds of phones
There are two basic kinds of phones,
These are useful descriptions. Before we delve deeper into the nature of vowels and consonants, let’s see where those words come from. (Whenever I want to understand a thing better, I look at where its name came from, since the name often gives clues about its nature.)
Vowel is an English respelling of the Old French vouel, from Latin vocālis, source of our word vocal. Vowels are vocal sounds. But what does vocal mean? Vocālis is derived from vox ‘voice’. Now we have the true meaning of a vowel: a voiced sound, one spoken with the voice.
On first read, this may seem redundant. Aren’t all speech sounds spoken with the voice? After all, a person who’s lost her voice has lost her ability to speak. But if we think of “voice” as a charge carried by a stream of air from the lungs, we come to see that blocking this voice-charged stream is as useful to speech as is allowing it to pass. The sounds that block or at least slow down the stream are consonants, from Latin consonāre ‘to sound together’, from com- ‘together’ + sonāre ‘make a sound’. When the ancient Romans were first giving names to speech sounds, they supposed that these air-blocking sounds could only be made together with vowel sounds; using the same logic today, we might call them co-sounds. (Of course, today we know this logic to be not quite true. Some consonants can in fact be pronounced as whole syllables without vowels. But this is the exception and not the rule.)
The main division between vowels and consonants, then, is this: if a sound’s main feature is the escape of voice-charged air, it’s a vowel; if the main feature is the blockage of this air, it’s a consonant. (Another way to divide vowels and consonants is that a vowel is used as the core sound, or nucleus, of a syllable, while a consonant is used as the boundary between two syllables. But there’s an exception to this: glides, also called semivowels, which are vowels that are used “consonantally” to form syllable boundaries. You can learn more about glides on the “Phonotactics” section of the Phonology page.)
Vowels
Now that we’ve established what a vowel is, there is much more we can say about it. Vowels seem uncomplicated at first – just open your mouth and go! – but think of the difference between the vowel in moon and the French equivalent, lune. How would you explain that difference? Or how about the last vowels in English fishhook and Korean
Height
The higher the tongue is in the mouth – that is, the closer to the roof of the mouth – when pronouncing a vowel, the higher the vowel is said to be. There are three main height settings: close (sometimes called high), mid, and open (sometimes called low). In between are the intermediate settings near-close, close-mid, open-mid, and near-open. If you think of the vowel space in the mouth as a coordinate plane, height is measured on the y-axis.
Backness
The word may sound silly, and this feature may as accurately be called “frontness,” but the concept is a useful one; backness forms the x-axis, just as height forms the y-axis, of the vowel space. The closer to the back of the mouth the blade of the tongue is when pronouncing a vowel, the more “back” the vowel is said to be. The three main backness settings are front, central, and back, with intermediate settings of near-front and near-back.
Rounding
Unlike height and backness, which are sliding scales on which a vowel can have any of several values, rounding is a two-way switch: either a vowel is rounded or it isn’t. We call a vowel “rounded” if we round our lips, or pucker them into an “O” shape, to pronounce it. On a vowel chart, rounded vowels are written to the right of unrounded vowels that occupy the same x, y space.
Using just those three features, we can describe any vowel fully! For example, the vowel in E. moon is a high back rounded vowel, whereas the vowel in F. lune is a high front rounded vowel; they differ in backness only. Likewise, the last vowel in E. fishhook is a near-high near-back rounded vowel, whereas the one in K.
Consonants
Just as vowels have three defining features that allow them to be fully described, consonants have five defining features that can describe them (more or less) fully. These are place of articulation, manner of articulation, voicing, aspiration, and tension.
Place of articulation
First, let’s look at which parts of the mouth are useful for making consonant sounds. The obvious most important shaper of sound is the tongue, whose name in Latin (lingua) gives us our words language and linguistics, but other features of the mouth’s geography are just as important. In fact, a few oral sounds don’t even involve the tongue. Here are the places of articulation, those areas of the mouth that we use to speak.
Lips
The lips can be brought together to block or redirect the airstream from the lungs; the resulting sounds are called bilabial (Latin bi- ‘two’ + labia ‘lip’). The bottom lip can also be raised to touch the top teeth; the resulting sounds are called labiodental (L. labia ‘lip’ + dēns ‘tooth’).
Teeth
In addition to the labiodental sounds, the tongue can also be stuck between the top and bottom teeth to make the interdental (L. inter ‘between’ + dēns ‘tooth’) sounds, or touched to the back of the front teeth to make dental sounds.
Alveolar ridge
This is the hard gum ridge that dips down just behind your top teeth. When the blade of the tongue contacts this ridge, the resulting sounds are called alveolar.
Palate
Also called the hard palate, this is the ridged roof of the mouth that slopes upward as your tongue travels along it toward the back of your mouth. Sounds made between the top of the tongue and the palate are called palatal sounds.
Velum
Also called the soft palate, this is the spongy area at the very back of your mouth. It’s as far back as you can reach with the tip of your tongue. Sounds made between the top of the tongue and the velum are called velar sounds.
Uvula
The dangly thing at the back of your mouth. English doesn’t use it to make speech sounds, but some languages, like Arabic, do. Consonants made against the uvula are called uvular.
Pharynx
The place where your nasal (nose) and oral (mouth) cavities meet and your throat begins. As with the uvula, English doesn’t have a practical use for it, but Arabic does. Sounds that the root of the tongue makes against the pharynx are called pharyngeal.
Glottis
Part of the larynx (voice box), the glottis includes the vocal cords and the opening between them. This opening can be closed, completely or partially, to create glottal sounds. Since they’re made in the throat, glottal consonants don’t involve the tongue.
Manner of articulation
Now that you know where the various consonant sounds are made, we can begin to examine how they’re made.
Stops
Also called plosives (for the burst of air that often accompanies them), stops are exactly what they sound like: they stop the air stream for a moment before allowing it be released through the mouth. This moment usually doesn’t last very long, but it can be exaggerated for emphasis. Depending on where the air is blocked, stops can be bilabial, interdental, dental, alveolar, palatal, velar, or glottal.
Fricatives
From L. fricāre ‘rub’ (source of E. friction), these sounds don’t block the air stream completely, but partially, squeezing it between two parts of the mouth (usually the tongue and an area of the top of the mouth). Depending on where the squeezing happens, fricatives can be bilabial, interdental, alveolar, alveo-palatal (between the alveolar ridge and the palate), palatal, velar, or glottal.
Affricates
These supersounds are a combination of a stop and a fricative. Every affricate begins as a stop and opens up into a fricative; in other words, the air stream is completely blocked, but the blockage then releases partially so that the air can escape, in a limited fashion, between two parts of the mouth. (If the blockage released fully, no fricative sound would be made, so the sound would be a simple stop, not an affricate.) The most common affricates begin as alveolar stops and end as either alveolar or alveo-palatal fricatives.
We get the word affricate by prefixing fricative with L. ad ‘to’ – an affricate is a transition from a stop to a fricative.
Liquids
Like fricatives in that they obstruct the air stream partially to move it between the tongue and the mouth, but unlike fricatives in that they obstruct it less, so no rubbing occurs. Liquids can be subdivided into lateral (L-like) consonants, which force the air to flow around the sides of the tongue, and rhotic (R-like) consonants, which guide the air stream over the top of the tongue. Liquids may be alveolar, alveo-palatal, palatal, or velar.
Sometimes a liquid can, without a vowel, take up an entire syllable, as the second syllable of riddle. These liquids are called syllabic. Usually, though, liquids are non-syllabic, like the first sound in riddle.
Taps
Like stops but shorter – instantaneous, in fact. Unlike stops, which can be held out and “charged up,” taps can last for only the moment that the blade of the tongue contacts the roof of the mouth.
Taps are a distinctive part of the consonant inventories of some languages, like Spanish, Korean, and Japanese; in others, like English, they aren’t an official part of the inventory but occur regularly in certain environments (e.g., between two vowels, as -t- in water and -tt- in butter).
Nasals
Made by closing the velar flap, which redirects the air stream from the mouth to the nose. Nasals are commonly bilabial, alveolar, or velar.
You may be wondering whether “nasal” isn’t a place of articulation rather than a manner of articulation, since the nose is certainly a place in the human head! To understand why “nasal” is a manner rather than a place of articulation, consider that the labels for different PoAs are mutually exclusive (the same consonant can’t be both alveolar and velar, for example), as are the labels for different MoAs (the same consonant can’t be both a stop and a liquid). Meanwhile, any PoA can be paired with any MoA (a consonant can be both alveolar and fricative, or velar and a stop). To decide whether “nasal” is a PoA or MoA, check which labels it can pair with. A consonant can be nasal and bilabial or nasal and velar, but not nasal and fricative or nasal and liquid. The labels “nasal” and “fricative” describe two different things, and the label “liquid” describes another thing altogether. Therefore, nasals may safely be classified as a manner, not a place, of articulation.
Voicing
Like rounding of vowels, voicing of consonants is easy to understand because it’s a switch with only two settings, on (voiced) and off (voiceless). To voice a consonant, vibrate your vocal cords as you form it.
While all vowels are voiced by definition, any consonant (considered as a combination of place and manner of articulation) can be either voiced or voiceless, since the vocal cords can be either buzzing or not buzzing while the consonant is pronounced. Because voicing is distinctive in some languages (like modern English, Spanish, and Japanese) but not in others (like Chinese, Korean, Old English, and Icelandic), it can be one of the trickiest aspects of pronunciation to master in a foreign language.
Voicing is the difference between the initial consonant sounds in pat and bat, fat and vat, thin and then, sue and zoo.
Aspiration
A strong puff of air that accompanies a stop. This feature is also a two-way switch: a consonant can be aspirated or unaspirated. Because aspiration is not distinctive in English, native English speakers aren’t trained to notice it, but you can feel its effects if you hold your hand in front of your mouth and say the words pot and spot. Though you’ll feel a little air blown against your hand from both words, you’ll feel a lot more from pot. That’s because English has a rule – not a rule you’ve ever learned, but one you follow instinctively! – that says initial stops are aspirated while medial stops in consonant clusters are not.
Languages that use aspiration as a meaningful element of pronunciation include Chinese, Korean, and Hindi; languages that do not include English (which uses aspiration regularly but unmeaningfully) and Japanese (which doesn’t use it at all). While aspiration is most commonly applied to voiceless stops, a few languages such as Hindi apply it to voiced stops.
Tension
This is an unusual feature of consonants in that it applies also, and even more commonly, to vowels. Another switch with two settings (tense and lax), tension is the property of whether the vocal muscles are stretched tight or not. If they are tight, as when you make the long vowel sounds in kale, keel, coal, and cool, the vowel is tense; if they are left loose, as when you make the short vowel sounds in Cal, kill, and cull, the vowel is lax.
Apart from vowels, stop consonants can also be made tense if they are “charged up” – that is, pronounced through stretched vocal muscles that deprive the sound of any voicing or aspiration it would otherwise have. That said, a tense consonant is not simply a voiceless unaspirated consonant; it has more intensity and a faster release.
Tensing of consonants is not distinctive, and indeed will sound odd if done out of place, in English, but is a core part of the Korean consonant system.
The IPA
Now you know how to describe almost any speech sound, vowel or consonant, precisely. But mentioning every feature of the phone you want to discuss can be cumbersome. If I want you to think of a particular sound, I can tell you it’s “the a in cat,” but if you don’t happen to speak my variety of English, you won’t know which sound I mean. More objectively I could call it a “near-open front unrounded vowel,” but who wants to chew on that every time we talk about vowels?
Thankfully, the International Phonetic Association, which exists to deal with these issues, has devised the International Phonetic Alphabet (IPA), a series of symbols that can represent a whole bundle of vowel or consonant features with a single typed character. For a chart of these symbols, click here.
If you’re serious about improving your ability in any language, native or non-, you owe it to yourself to learn the IPA. Pronunciation isn’t quite an exact science, but these symbols are as close as it gets!
A phone can be represented in writing by enclosing its IPA symbol in brackets, like this:
Sometimes, when it’s not important to distinguish between a phone and a phoneme, I’ll leave out the brackets or slashes, like this:
