Skip to main content
Loading Explore guides…
English S Sound Pronunciation · Acedra
Home
Explore
Languages
English S Sound Pronunciation
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
47 segments
Copy
Hide video
0:00
Mayday, mayday, can you hear us? Can you– We are sinking, we’re sinking. –What are you sinking about? The idea that English has a most important consonant isn't something I'm aware of hearing from many other people, but long years spent or misspent in English Phonetics have convinced me it's true. The sound I have in mind is extremely important for communication, but it's really tricky to pronounce. Speakers have various ways of bringing it off,
0:31
and some might say they don't all succeed. For learners of English it's even harder, because this English sound is one that many other languages don't quite have, despite what both spelling and phonetic symbols may tell you. And the situation’s made even harder by the repeated failure of books and teachers to describe it correctly. Okay, enough of the mystery. The sound, or sound family, that I'm talking about is the one most typically associated with the letter S.
1:02
Now if you're about to click away because you think S is easy, or boring, I beg to differ. It's a sound children often find really challenging, I certainly did, and many adults continue to have non-standard pronunciations that can be criticised or made fun of. Or even become part of your image. We must strive for a sane and just peace. Sufferin' succotash We'll drill down into the physicality of S type sounds in a bit, but first I want to expand on
1:35
why they’re so important in English. There are several reasons. One is that /s/, along with its variants /z/ and /ɪz/, means something. In fact it means a number of things, all of them important. One of these meanings is plurality. One cat, several cats, one consonant, several consonants, etc. Also one dog, several dogs, where the S can take on the voicing of the preceding sound. Of course there are languages like Chinese that don’t make such a big deal out of plurality,
2:11
but in English we always want to know whether we’re talking about one, or more than one. Another meaning of ‘s’ is possession, ownership. The cat’s whiskers, Geoff’s video etc. ‘This government works’ and ‘this government's works’ do not mean the same thing.
2:31
Yet another meaning is third person in the present tense. That video sucks, she goes to school, etc. It’s also the contracted form of ‘is’, for example it’s here, that’s right. And the contracted form of ‘has’, for example what’s happened or it’s just finished. In the British National Corpus of Spoken English, contracted S meaning ‘is’ or ‘has’ was the seventh most frequent word or morpheme in the language, more common than ‘to’, or ‘of’, or ‘that’.
3:05
And note that these are all endings. Huge numbers of English learners find any final consonant a problem. Across the world’s languages, there’s a preference for words to end in vowels, and in many languages practically all words end in vowels. But English has to be awkward. Most English words end in consonants, and thousands of English words end in several consonants. And then the -s suffix goes at the very end.
3:34
Now when we’re learning foreign languages we tend to work through the pronunciation of difficult words sound by sound, and many learners of English will prioritize sounds that come earlier over those that come later. L2 learners and users of English often operate on a last-in, first-out basis, so the S on the end can easily get dropped or under-articulated, even when it’s carrying all that meaning. But native or L1 speakers of English don't do this. I can’t call to mind
4:04
a single process of English which deletes a final /s/ or /z/. They’re just too important. And here's another reason English S needs special attention. It's not just always retained, no matter how many consonants precede it, it’s a lot louder than it is in many other languages. Which raises two questions. Why does it have to be so loud, and what is it that speakers do to make it so loud? I think one of the reasons S has to be so loud in English is to distinguish it from TH.
4:38
English is quite unusual in contrasting three pairs of fricatives involving the tip of the tongue, θ s ʃ and ð z ʒ. What really makes TH different is how quiet it is compared with loud fricatives like S and SH which are called sibilants. And that's an opportune moment to say Thank you to Surfshark
5:02
for sponsoring this video. I decided to get really accurate measures of the loudness differences between these sounds at University College London. And on a busy train to London where I get warnings about unsecured connections I don't want to take any chances. There are too many people out there who want access to my details. So it's one place I really do always turn on my VPN. Just one click and I'm protected. Surfshark's super easy to use on my laptop and my phone,
5:32
in fact the number of devices supported is unlimited. And because Surfshark relocates me anywhere I like across a hundred countries, I can do my research and find all the illustrative clips that my viewers enjoy without fear of being geoblocked. It's the same principle that’ll let you watch all those Netflix libraries or avoid being prevented from accessing your own services when you're abroad.
5:55
So if you're considering a new VPN and you'd like to support this channel and get an extra 3 months for free, just use my special code. There's a 30-day money-back guarantee and the link’s in the description below. So thanks to Surfshark, the VPN with added sibilants. To get really accurate measures of differences in loudness, or more objectively sound pressure level, I've come here to UCL to make use of something I remember quite clearly from
6:22
my days here as a student long long ago, the anechoic chamber. I remember coming in here for the first time as a student and it felt as though somebody had put their fingers in my ears. The ceiling, the walls and the floor of the anechoic chamber are covered in these sound-absorbent wedges. That's what makes the room acoustically dead, there's no reverberation at all. Nothing. Just checking that you can hear me ok? I can hear you fine.
7:05
So there we have a calibrated recording. Gordon Mills of Psychology and Language Sciences measured my S and SH as respectively 7 and 9 dbSPL higher than my TH, that's more than twice the sound pressure level. Even though I gave TH my best shot, close to a high quality mic in a super quiet environment. And here we can see spectrograms of the American thin, sin and shin, in the Oxford Advanced online dictionary.
7:32
thin sin shin These of course are the initial consonants, All three are relatively high up compared with the rest of each word, they have acoustic energy in the higher frequencies, the more tinny part of the sound spectrum. The lower frequencies are the more boomy part, which is why nasal sounds like N are right down at the bottom. SH goes down here, to about 2kHz. S is higher, it only comes down to about 4 kHz, and the TH energy is way up here.
8:04
However, what really makes TH different from the other two is how quiet it is. I can demonstrate the distinctive quietness of the TH by actually deleting it completely. And guess what? thin sin shin
8:20
You can still hear it as thin, even if the TH noise is gone. You might be wondering, why doesn’t it just sound like ‘in’? The answer is that you can still hear the effect on the following vowel of the tongue position for the TH. If I say these three words to camera, I can deliberately make the TH so weak that practically no sound comes out. Thin, sin, shin. You can still identify the first word because of the way my tongue transitions into
8:49
the following vowel, even if I hide my mouth so you aren’t influenced by the visual cues.
8:58
And if you’ve ever edited YouTube videos, you may well have seen on the audio track that S and SH get their own bump of loudness, often like a separate syllable, but TH can be barely visible. The greater power of S compared with TH isn’t just a matter of decibels. These two sounds often come into direct contact in English, as in months and clothes, and especially ‘is the’, ‘was that’, ‘does this’ etc. Getting from one to the other can
9:30
be tricky. s‿θ‿s‿θ or z‿ð‿z‿ð. And English speakers like to simplify this sound sequence. But remember I said that English doesn't drop S? So it’s the TH that gets thrown under the bus. Take the word ‘months’ for example. Native speakers love this word. We frequently refer to periods of six months and eighteen months, whereas many L2 users of English often say ‘half a year’ and, even less English, ‘one year and a half’. Now in ‘months’,
10:06
the S is always clearly articulated, but the TH is passed over very quickly and very often dropped completely, most of all in American English, making ‘months’ into a rhyme with ‘dunce’. over the past few months for many many months eleven months in the 3 months since that swim ended events over the last several months Non-native L2 users of English, on the other hand, tend to operate on a last-in-first-out basis, and
10:39
will often take care to pronounce the English TH while dropping the all-important and meaningful S. You are here for how many month? Neun Monate? The combination of a word-final S and a word-initial TH is extremely common in English, and again, L2 speakers will often put more effort into the TH. That’s totally understandable, given the universal tendency for beginnings to be stronger than endings. But again English has to be awkward. That meaningful final S is so
11:13
much more important in English, that it can totally assimilate the feeble initial TH. Resistance is futile. –Woah, this is so cool. So ‘is that right’ often becomes ‘is zat right’, and in rapid speech the two z’s can merge into one, izat right? A famous instance of this is in the world of cricket, where the appeal to the umpire ‘how’s that’ has even developed its own conventionalized spelling, howzat? But how's that gonna help us end extreme poverty in the world?
11:44
L2 users will quite often assimilate the other way round, saying not ‘is zat right’ but ‘ith that right?’ This vulnerability of the English TH sounds is also demonstrated by the various dialects that don't use them at all, substituting other sounds instead, typically /f/ and /v/. This replacement of TH is generally known as TH-fronting. It used to be very localized, but it’s become a lot more common in Britain in recent years.
12:13
So when your thumbnail is put up against another thumbnail Note that the subtitute sounds which these native dialects use are, like the TH sounds themselves, non-sibilant. L2 speakers, however, will often replace English TH with their sibilants, for example pronouncing think as sink. This is the German coastguard. –We are sinking, we're sinking. What are you sinking about? Actually the English TH sounds are pretty easy to make,
12:53
as long as you’ve got a tongue and some teeth, and can blow air between them. You know how to whistle don't you Steve? You just put your lips together and blow S is a much trickier and more complicated sound to produce, especially English S. Remember, it has to be much louder than TH, and considerably higher-pitched than SH. This isn’t the case in languages with fewer contrasts. European Spanish has /θ/,
13:20
but doesn’t have a contrast between /s/ and /ʃ/. And this allows the sound that it writes with the letter ‘s’ to be somewhere between English /s/ and /ʃ/. [Spanish] Here again we can see American English spectrograms. math mass mash
13:56
and this is a Spanish ‘más’. The Spanish sound is visibly intermediate between English /s/ and /ʃ/. Japanese, on the other hand, has no contrast between /θ/ and /s/. And this allows the sound that it romanizes with the letter ‘s’ to be somewhere between English /s/ and /θ/. To my ear, these words could be ‘death’ and ‘math’.
14:30
But in fact they’re these extremely common verb forms. [Japanese] And this is the Japanese form ‘mas(u)’. It’s visibly intermediate between English /θ/ and /s/, just as the European Spanish sound was intermediate between English /s/ and /ʃ/. mas(u) math mass mash más
15:13
But teachers and learners of English pronunciation can be hypnotized by letters and symbols, so they tend to ignore ‘s’, thinking, oh we've got that letter ourselves, that’s no problem, and they focus their attention on the scary TH thing that has not just one but two special symbols. More than once I’ve failed to convince Japanese teachers of English pronunciation that they should consider devoting more time to the so-often meaningful English
15:43
‘s’ than to English ‘th’. The ‘s’ sounds of English are much stronger and sharper, and that applies to the lenis /z/ as well as the even stronger fortis /s/. So how do sibilants, especially English sibilants, get to be so loud? It’s a good question, and the answer exposes how misleadingly described they sometimes are. Can you see anything not quite right about these text book diagrams? Try putting your vocal organs
16:14
in exactly these positions and see if you can produce the sounds they're intended to describe. If you'd like a hint It's not the index finger I'm taking about, though we'll see later that a finger can help. What I have in mind is those teeth. In their book the Sounds of the World's Languages, Ladefoged and Maddieson describe sibilants as produced by 'the high velocity jet of air formed at a narrow constriction going on to strike the edge of some obstruction such as the teeth.' (p. 138)
16:46
And there are text books that make this clear with photos of the mouth shape. This is why people with no front teeth don't make good sibilants. I've told you a million times, my teeth are coming in, it's called cleidocranial dysplasia and why, when you watch people speaking, their jaws generally move together for sibilants. You can make say an N with your mouth open, /n/, but not S or SH. And if you have nicer, bigger teeth
17:14
than mine, you often see a clear gateway made by the upper and lower teeth for sibilants. Well this is the central issue as you heard there. There is no transition to net zero carbon emissions. Soon, over a quarter of the Australian population will be over the age of 65. It's actually such a shame that so many are sharing their emotional opinions on the subject without even understanding. Animators can be more aware of this than some phoneticians. You are a sad, strange little man.
18:02
Inside the mouth, the tip of the tongue is generally grooved and held against the palate, channeling that concentrated jet of air towards the back of the teeth, which is what makes the noise.
18:19
chops Now the closer the tip of the tongue gets to the teeth, the less possible it is to fire a powerful jet of air at the teeth. Japanese speakers typically have their tongue quite close to the teeth, which is where the English TH sounds are typically made.
18:39
However, I watch a lot of YouTube videos, which means watching a lot of close-ups of people talking directly to camera. And it's clear to me that a very large percentage of English speakers also make S with their tongue tip at or very near their teeth. Now I'm not just taking about speakers with what’s often described as a lisp. This refers to the placement of the tongue forward in the mouth, so it can be visible a lot of the time,
19:07
in such a way as to weaken or prevent that intense sibilant noise we saw on the spectrograms. Speaks with a noticeable lisp There is only one way to celebrate this I think there's a continuum between what people think of as a lisp and an S that would be thought of completely standard but is made nonetheless with the tongue tip at the teeth. But how can you make a loud, clear S if the air is squeezing out between the tongue tip
19:40
and the teeth? What is there beyond the teeth to act as the all important acoustic obstacle? The answer, of course, is the lips. Very many speakers, perhaps more nowadays than used to be the case, produce S by shooting air between the tongue tip and the teeth so it strikes the lips, usually the bottom lip, /s/. And when you watch speakers like this, you see their bottom lip being raised every time they make an S or a Z. Everyone at some point in their life has experienced
20:16
the sensation of having plugged up ears. And having plugged up ears really sucks. I say thank you for the trust you have placed in us.
20:30
YouTube processing is very complex with newer codes and newer methods
20:38
and that processing spits out YouTubers can't see and for videos like this, the final file size
20:48
about a fresh hour
20:54
processing is a necessary process
21:00
If I try to make these bottom lip S’s, they aren’t particularly loud, because we're all built differently. But watch how pulling my bottom lip away just kills it. My impression is it's much rarer to use the top lip as the acoustic obstacle, but it's certainly possible. And notice that if you do use your top lip as the obstacle, the lower jaw becomes irrelevant and, contrary to what I said earlier, you can actually make an S with your mouth open, at least a little bit.
21:39
Now are there any other possible obstacles you could use to make an S beside the lips? Well frankly you could use practically anything. Like a finger. So thank you again to Surfshark for sponsoring this video.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat