Types of phonemes
Broadly speaking, we can divide mouth sounds up into vowels and consonants.
This isn't really accurate, so let's characterise them more sanely.
Vowels
Vowels are pronounced with continuous smooth voicing, so are part of a larger class called sonorants.
Vowels are any sound made by the mouth in which the channel of air through the mouth is clear.
No part of the mouth is touching any other part.
Vowel axes
The vowel itself is determined by three axes:
Tongue height — distance from the palate, the roof of the mouth
Tongue depth — the tongue's location forward, central or backward in the mouth
Lip "rounddness" — the horizontal width of the lips
Rhotic vowels
Vowels can be classed as rhotic or non-rhotic. Rhotic means that the tongue is curled upwards, similar to the R sound, which adds an "R colouring" to the vowel.
This is the fourth axis of vowel quality.
If you really explore the vowel chart, with mindful attention to the feeling and mindful control of your mouth, you will discover that the standardised IPA system for each language is vastly incorrect.
I am not speaking simply about accents, but rather that the teams who agree on IPA systems for each language seem to lean towards simple-looking symbols instead of accurate symbols.
This is actually a terrible thing, because for people who use the IPA to improve their pronunciation of a language they're learning, the incorrect IPA of their native & target languages causes them to have a misunderstanding of what they're learning.
Consonants
Actually, "consonants" is not a real class. Let's describe these as:
Non-Vowels
Ok, that's not much more helpful, but at least we're clear now that there are vowels and other sounds that are not vowels.
Let's go deeper:
Plosive–Stops
Plosives are named after the fact that they cause an explosion. Actually, some languages also have implosions, and this is why the sound class is called plosives instead of explosives.
Every sound exists in the flow of time, and so every sound has a start, a middle and an end.
aaapaaa
For there to be an explosion, there must first be a blockage of air flow.
Prior to the ex/im-plosion, the sound can be called a stop.
aaaap
Some languages use these positions at the end of syllables, solely to stop the airflow, and do not release the blockage with any kind of sound. They simply stop before opening the blockage. This actually changes the shape of the end of the vowel before it.
In other languages, stops at the end of syllables are still plosives, and the mouth position is released, so a tiny sound is heard. If this sound is missing, it is difficult or even impossible for native speakers of that language to hear the word correctly, because their ears are not trained to hear such subtle changes in vowel closure. If you are a learner of that language, you should release the final plosive position correctly, because you do not know which other sounds the native speakers may interpret your unreleased stop position to be.
If the sound is used before a vowel, it results in a typical plosion release.
paaa
Fricatives
Fricatives are named after the friction involved.
To produce such friction, two parts of the mouth must be touching, with loose force between them, so that airflow or voice causes the two parts to vibrate against each other. If the force between them is too hard, the vibration cannot occur. If the force is too weak, the vibration also cannot occur, and the sound will be a strange kind of closed-mouth vowel.
A fricative is a continuous type of sound, and as such it must have no discernible beginning or end. It must have the same quality throughout. It is best to practise fricatives with a very long stretch, as this trains your mouth into producing the sound correctly:
Practise fricatives with length
Do not practise fricatives as a singular moment of sound: "fa".
This method makes it likely that you will pronounce a plosive instead, even if you use the right position.
Instead, practise fricatives as a long moment of sounds: "ffffffff".
This method ensures that you are maintaining one position, producing a steady flow of air or voice, and allows you to experiment with applying enough pressure through the parts of your mouth.
Glides
Glides, also usually known as approximants, are a class of sounds in which two parts of the mouth come extremely close to each other but don't touch.
As with everything else inside this universe, glide sounds exist within the movement of Time, so they have a beginning, a middle and an end. Let's demonstrate with "w":
aaaawaaaa
Note how the mouth moves through this position:
the mouth is open, in "a"
the mouth closes into the "w" position
the mouth opens up again into "a".
This is why we can call it a glide.
A glide can exist at the start of a vowel:
waaaaa
and at the end of a vowel:
aaaaaw
and between two vowels:
aaawaaa
In most languages, it is vital that you do not touch the two parts of the mouth together. If you do, you will create a plosive or a fricative, which might be interpreted as a different phoneme, and thus as a different word.
You cannot base your pronunciation variations on your native language pronunciation system. The way that your brain filters phones & phonemes is not the same as the people you are speaking to!
The primary characteristic of a glide is the motion.
The secondary characteristic is its maximum point of movement.
All glides can theoretically exist as vowels. The difference is that a vowel takes up a moment of time, typically considered as a "syllable"; whereas a glide maxima is very short, existing just as a point for mouth movement to pass through.
Laterals
Laterals are pronounced with continuous smooth voicing, so are part of a larger class called sonorants.
Laterals are a strange class. There are actually quite a few, but English only features two — and if you were to ask natives, most would only notice one.
The lateral in English is L, but L features two distinct allophones: Light L and Dark L.
The manifestation of Light or Dark L is entirely natural, dependent on the L's position within a syllable and vowels and consonants, so the English alphabet doesn't need to distinguish them. However, many languages do not have Dark L, so English learners struggle to pronounce it, and usually pronounce Light L very carefully in positions that native English speakers would really struggle to do!
Other languages feature more laterals than just L.
Nasals
Nasals are pronounced with continuous smooth voicing, so are part of a larger class called sonorants.
Nasals are produced by passing voice vibration through the nose. This is achieved by pressing two parts of the mouth together with full closure, forcing the vibration to transfer through the mouth into the nose.
Nasals can also be, very rarely, unvoiced.
Nasals can be pronounced without a vowel, so are a continuous type of sound.
In English, the word "not" is very commonly contracted into the verb before it, such as in isn't, wasn't, doesn't, didn't, can't, couldn't, won't, wouldn't, shan't, shouldn't, hasn't, haven't, hadn't.
This contraction can get more complex with the addition of the perfective "have": can't've, couldn't've, won't've, wouldn't've, shan't've, shouldn't've, hadn't've.
T clips the N
In these cases, the nasal /n/ is followed by a ⟨t⟩. This T may often not be pronounced as an audible /tʰ/ — however, the T absolutely still has an effect! The T causes the N to be clipped short, the voicing stops suddenly. This is extremely important, because without the T, many of these words sound like other words. In some cases, they sound like the opposite meaning!
can't → can
won't → won, one
shan't → Sian, Sean, shawn, Shaun, /ʃ-n/
hadn't → had an, hidden
"isn't context enough to understand me?"
You may be of the opinion that "context is enough for the listener to work out what I mean".
a)
in English speech culture, we make the effort to be as understandable for the listener as possible. We are not aware of this, but as a widely-travelled language teacher & learner, I observe that some cultures speak with laziness, placing the onus of interpretation fully upon the listener.
While it would be lovely for English native speakers to be more flexible in their listening patterns, in reality this is not going to happen, because there are hundreds of millions (if not more than a billion) native English speakers already, we all have different accents, and there are a further billion or so second-language English speakers, also with their own accents and pronunciation patterns and localised "mistakes".
b)
English has "connected speech"; there is little to no vocal separate between words. As such, mispronouncing one word can have severe knock-on effects through the rest of your sentence, and thus also through the rest of your conversation! The misinterpretation may even cause the listener to second-guess something that they had already understood.
P clips the M
There are several minimal pair words in English ending in MB and MP.
In these cases, the B extends the M sound, while the P shortens the M sound. Just as the NT set above, the MP must be short and sharp! If it is not, it sounds like MB or M.
lamp → lamb
dump → dumb
ok I can't recall any more right now
K clips the Ŋ
There are several minimal pair words in English ending in NG and NK.
In these cases, just as with MP and NT, the K makes the Ŋ (ng) into a short, clipped sound.
think → thing
thank → thang
wink → wing
pink → ping
bank → bang
ect
N is not ん or M or NG!
Many languages have nasals at the end of syllables.
Many languages do not care what kind of nasal is at the end!
In many cases, this results in a free-rotation blending between N, M, Ŋ (ng) and any other nasal consonants. It can even include nasalised vowels, such as in French and Portuguese.
Japanese has a nasal syllable ん, which can be any one of approximately six different nasals, dependent upon the next syllable's sound. ん before a T is an N sound; ん before a P is an M sound.
Learners of English from these languages consistently mispronounce syllable-final nasal letters, directly related to whatever the common patterns are from their native language.
In English, the three nasals affect the meaning of the word, and affect the transition into the next word via "connected speech". It is vital that you pronounce the right nasal sound!
M uses two lips.
N uses the tongue, pressed behind the top teeth — in the same position as T.
NG uses the root of the tongue, pressed up at the back of the mouth — in the same position as K.
To Do
vector venn diagram of phoneme classes — sonorants, liquids, nasals, semivowels, etc.
Last updated