good vibes: a conversation about voiced and unvoiced consonants and other speech noises

“When I use a word,’ Humpty Dumpty said in rather a scornful tone, ‘it means just what I choose it to mean — neither more nor less.’

’The question is,’ said Alice, ‘whether you can make words mean so many different things.’

’The question is,’ said Humpty Dumpty, ‘which is to be master — that’s all.”
― Lewis Carroll, Through the Looking Glass

Canto: Okay, so now we’re getting into phonetics, is it? I’ve heard recently that some consonants are voiced, some unvoiced. Can you tell me what that means?

Jacinta: I think phonemics is the word. Or maybe phonology. Or maybe it is phonetics. Anyway don’t worry about the terminology, let’s look at your question. If I tell you that these five consonants are unvoiced: t, s, f, p, k, and that these five consonants are voiced: d, z, v, b, g, play around with those consonants in your mouth, that magnificent musical instrument, and see if you can work out the difference.

Canto: Okay, wow, I’ve noticed something. When I put my hand in front of my mouth and utter the first five, the unvoiced, I feel a blast of air hitting my hand. It doesn’t happen with the voiced consonants, or not nearly so much. Well, actually, no, ‘s’ isn’t like that, but the other four are. So that’s not it, though it’s an interesting thing to observe. But thinking voiced and unvoiced, that gets me somewhere. The voiced consonants all seem to be louder. Compare ‘z’ to ‘s’ for example. I seem to be forcing a sound out of my mouth, a kind of vibration, whereas ‘s’ is just a ‘ssss’. A vibration too, of course, but softer. Unvoiced, I get that. ‘t’ seems to be just a mere touching of mouth parts and pushing air past them to make this very soft sound, ‘whereas ‘g’, ‘d’ and ‘b’ are more forceful, louder. And ‘v’, like ‘z’, makes a loud vibration. It’s funny, though – even as I make the sounds, and focus on how they’re made in my mouth, I’m damned if I can work out clearly the mechanics of those sounds. But of course researchers have got them thoroughly sorted out, right?

Jacinta: Well you’ve got the distinction between voiced and unvoiced pretty right. The key is that in a voiced consonant the vocal chords vibrate (actually, they’re vocal folds – they were mis-described way back in the day, actually as vocal cords, and the n
ame has stuck, with a musical embellishment). Here’s a trick: take the pair of consonants you mentioned, ‘s’ and ‘z’, and sound them out, while putting your hand to your throat, where the voice-box is…

Canto: But it’s not really a box?

Jacinta: The larynx, responsible for sound production among other things. A housing for the vocal folds. So what do you fee?

Canto: Yes I feel a strong vibration with ‘z’, and nothing, or the faintest shadow of a vibration with ‘s’.

Jacinta: So now try ‘f’ and ‘v’. Then t/d, p/b and k/g.

Canto: Got it, and never to be forgotten. So that’s all we need to know about voiced and unvoiced consonants?

Jacinta: It’s something that could be done with learners – without overdoing it. I’d only point it out to learners who are having trouble with those consonants. And it’s intrinsically interesting, of course.

Canto: So this raises questions about speech generally, and that great musical instrument you mentioned. Is the regular patterning of sound by our lips, our tongues and so on to make speech, is that very different for different languages, and is this a barrier for some people from different language backgrounds to learning English?

Jacinta: Well you know that there different ways of speaking in English, what we call accents and dialects, so there are different ways of saying English. For L2 learners, especially if they take up their L2 – in this case English – later rather than sooner, it’s unlikely that they’ll lose their L1 accent, but this is unlikely to affect comprehension if they can get the syntax right.

Canto: I’ve noticed that Vietnamese speakers in particular have trouble producing some English word endings. What’s that about?

Jacinta: The Viet language, like a lot of Asian languages, doesn’t have consonantal word endings. So that’s why they ‘miss’ plurals in speech (s,z), as well as saying ‘I lie’ for ‘I like’, and the like. They have the same problem with t, v, j and other consonants. They also have trouble with consonantal combos in the middle of words. And according to the ESLAN website, they ‘struggle greatly with the concept of combining purely alveolar sounds with post palatal ones’.

Canto: Eh?

Jacinta: Okay let’s learn this together. An alveolar consonant is one that employs the tongue against or close to the superior or upper alveolar ridge. That’s on the roof of the mouth getting close to the upper teeth, and it’s called alveolar because this is where the sockets of the teeth – the alveoli – are. You can feel a ridge there. English generally uses the tongue tip to produce apical consonants while French and Spanish, for example, uses the flat or blade of the tongue to produce laminal consonants.

Canto: So can you give me an example of an apical alveolar consonant?

Jacinta: Yes, the letter n is called an alveolar nasal consonant. Try it, and note that the tongue tip rests on the alveolar ridge and sound is produced largely through the nasal cavity. The letter t is a voiceless alveolar stop consonant. It’s called a stop because it stops the airflow in the oral cavity, and it’s voiceless as there’s no vibration of the vocal folds. On the other hand the letter d is a voiced alveolar stop, differing from t in that it involves a vibration of the vocal folds, a ‘voicing’.

Canto: Mmm, but I notice that with d the tongue is a little less forward in where it hits the upper palate – behind the alveolar ridge, whereas with t you’re almost at the base of the upper teeth.

Jacinta: Well, there are four specific variants of d. Your specific variant is postalveolar, whereas the other three are more forward – dental, denti-alveolar and alveolar.

Canto: So there’s this complex combinations of stops – stopping the airflow – voiced and unvoiced, where the vocal folds come into play (or not), nasalisation and other soundings, all of it pretty well unconscious, and delivered with various levels of stress (in both senses of the term). It’s all pretty amazing, and it’s no wonder that those interested in AI and robotics have realised that embodied consciousness is where it’s at, because we’re surely a long long way from developing a robot that can manage anything equivalent to human speech. And that’s just in terms of phonology, never mind syntax and morphology. But I’ve got a few other ‘sound’ terms knocking around in my head that I’d like explained. Tell me, what are fricatives and plosives?

Jacinta: Okay, well this is all about consonants. The letters p,t,k (unvoiced) and b,d,g (voiced) are all plosives in that ‘air flow from the lungs is interrupted by a complete closure being made in the mouth’. With fricatives – unvoiced f and s, voiced v and z – ‘the air passes through a narrow constriction that causes the air to flow turbulently and thus create a noisy sound’. I’m quoting from the World Atlas of Language Structures (WALS). So for example, the difference between rice and rise is that the former uses an unvoiced fricative and rise uses a voiced one – very peculiar because rise uses ‘s’ which sounds like ‘z’ and ‘rice’ uses ‘c’ which can lead learners astray with the ‘k’ sound. If you’re interested in learning more…

Canto: We both are.

Jacinta: WALS online is a great database with 151 chapters describing the structural features of the world’s languages – phonological, grammatical and lexical. It’s published by the Max Planck Institute for Evolutionary Anthropology and should be a great starting place for an all-round knowledge of human language.

Canto: Just another of those must-reads…