Hi, I'm Taylor and welcome to Crash Course Linguistics!
Spelling can be a mess.
Take the letters -ough.
They're pronounced differently in tough, cough, though, through, bough, and brought.
The letters don’t always match the sounds they represent!
What's more, people can pronounce the same words in different ways.
There’s a difference between "I. Am. Going. To. Get. You."
and "I'm gonna getcha".
If we want to talk about the different ways people can pronounce the same spellings, or which sounds exist in different languages, it's useful to have a more precise understanding of how language sounds work.
The study of speech sounds is called phonetics, and the writing system that linguists use to write pronunciation very precisely is called the International Phonetic Alphabet.
[THEME MUSIC]
Speech sounds don’t include coughing, sneezing, or belching.
They’re the sounds that spoken languages combine with each other to make words.
While there are a finite number of speech sounds, we can use them to make an infinite number of words.
To understand speech sounds, we first need to understand how the human body produces them.
We start with the lungs, which push air up through the vocal folds.
Also known as vocal cords, your vocal folds are thin membranes that sit across your throat and vibrate to make sound.
This path, which moves from lungs to lips, is known as the vocal tract.
We can change the shape of any part of the vocal tract to make, or articulate, different sounds.
It’s like a wind instrument--when air goes through it, it makes a sound.
You’re like a squishy, wet, meat clarinet that can articulate sounds.
I’m sure that’s an image you’ll never forget.
For spoken languages, the tongue, mouth and rest of the vocal tract are the articulators, the parts of the body that move to make language.
For signed languages, the articulators are the hands, face and other visible parts of the body.
There are five major features of signed language articulation:
Handshape, such as the difference between the ASL sign for ‘red’ and ‘sweet’.
Palm orientation, such as the difference between ‘balance’ with the palms down and ‘maybe’ with the palms up.
Movement, such as moving in a circle for ‘always’ or a line for ‘scold’.
Location, like "mother" at the chin or "father" at the forehead.
And expression, also called non-manual signs, like furrowed eyebrows to make something a yes/no question.
There are some conventions for writing signed languages down, but there is less consensus about how to write them than with spoken languages.
Describing all these systems in detail would be a whole other video!
Spoken language linguists also describe sounds in relation to how they’re made with the articulators.
Consonants are speech sounds made by closing the vocal tract in some way, while vowels are made by making it a different shape but leaving it open.
We’ll look more closely at vowels in the next episode.
To dive into how linguists write spoken languages down, let’s start with some consonant sounds you know,
which happen to be some of the most common sounds in languages across the world: /p/, /t/ and /k/.
Try making these yourself, and pay attention to what your mouth is doing: /p/, /t/, /k/.
These sounds all require you to close off part of the vocal tract to make the air stop before you let it explode out, which is why they’re known as stops or plosives.
These three sounds are made at three different spots in the vocal tract, which linguists call places of articulation.
We often name these places according to the parts of the vocal tract used to make the sound.
So, since [p] is made by pressing both of your lips together, it’s a bilabial sound.
[t] is made by pressing the front part of the tongue against the hard ridge just behind your teeth, known as your alveolum, so it's an alveolar sound.
Some people also call the alveolum the "pizza ridge" because that's where you burn your mouth when you take a nice big bite of hot slice!
And [k] is made by pressing the back of the tongue against the soft part of the roof of your mouth, known as your velum, so it's a velar sound.
We can put these sounds on a diagram of a mouth to show these different places of articulation.
The square brackets here indicate that they’re phonetic sounds, not English letters.
Let’s look at two more sounds we recognize: [f] and [s].
Unlike our stops, these involve the air continuously flowing out of, and causing friction in, the mouth.
So this type of sound is a fricative.
[f] is made using the bottom lip and the top teeth.
That makes [f] a labiodental sound.
[s] is pronounced at the alveolar ridge, just like [t].
This means [s] and [t] have the same place of articulation, but are different in their manner of articulation, because [t] is plosive and [s] is fricative.
/k/ has a corresponding fricative too, /x/.
It’s not in most varieties of English, but it’s the sound in
Scottish ‘loch’ or Spanish "jalapeño".
Now try this with me.
Put your hand on your neck and make the sound [s] and then turn it into a [z].
[sssssssszzzzz]
Your lips and tongue stay in the same position.
You're not changing the place or manner of articulation.
But something does happen in your throat.
For [ssssss], your vocal cords aren't vibrating, while for [zzzzz] they are.
Seriously, try it!
The buzzing of your vocal cords is known as voicing.
[s] is a voiceless sound, and [z] is a voiced sound.
Each of these sounds we’ve discussed so far has a near-identical voiced twin, like [f] and [v], [p] and [b], or [t] and [d].
Even the velar fricative, /x/, has one.
It’s found in many languages, including Arabic, Basque and Navajo, and is written with this curly Y symbol, based on the Greek letter gamma.
To continue our tour of the vocal tract, another manner of articulation is nasal sounds.
We make them by closing the mouth but opening the nasal passage using the uvula.
In English we have [m], which goes with our bilabials [p] and [b].
We have [n], too, which goes with our alveolars [t] and [d].
We also have the nasal sound at the end of ‘sing’ and in the middle of ‘hunger’, which goes with our velars [k] and [ɣ ].
We normally write this [ŋ] sound with an N and a G.
But that’s not that useful for phonetics, because some “NG” words, like “hanger,” don’t have the [ŋ] sound.
/ŋ/ needs its own symbol, so we use this little n with a tail stolen from a g, called an engma.
In fact, the word engma has a velar nasal in it!
How convenient!
Our mouth diagram is beginning to look a little cluttered.
We have our places of articulation, which move from left to right as we move further back into the vocal tract. We can treat these like columns in a table.
Now, we can make the different manners of articulation into rows.
Great, this chart is such an improvement!
We know it’s based on the mouth, but it’s a lot easier to read!
Because it's hard to make a chart in three dimensions, we'll
represent voicing by placing the symbols in pairs inside the boxes, where the first symbol is always voiceless.
And when there's only one symbol by itself, it's always voiced.
We could keep going by paying attention to what the mouth is doing as we pronounce different sounds in English and other languages until we have a comprehensive table of possible speech sounds.
For example, there are a ton of other manners of articulation.
We can add Approximants, like [r], Trills, like the R in Spanish ‘perro’, and Taps or Flaps like the T in ‘butter.’
There are also Lateral Fricatives, like the Welsh pronunciation of the name Llewellyn, and Lateral Approximants, like the English pronunciation of Llewellyn.
There are more sounds we haven't gotten to, especially in other places of the vocal tract which get constricted in various manners.
But let's take a step back from the examples for now.
It turns out, this table we've been building is a chart of the International Phonetic Alphabet, also known as the IPA.
The IPA is a finite set of symbols used by linguists to represent all of the sounds that are relevant for spoken language.
Let’s head to the Thought Bubble to find out more about the IPA.
The International Phonetic Alphabet was created in the late 1800s by a group of linguists who were tired of putting a pronunciation key at the beginning of every document.
If we could just make one system, where each sound had exactly one symbol, they thought, then people could just learn it once.
And then we'd be able to write about all the sounds relevant to spoken languages, without any confusion.
Unlike attempts to write down all the words in the world, which end up going on forever since we keep inventing new ones, this project actually succeeded!
There are only a finite number of ways that we can configure our mouths and throats to make sounds, and only some of those are used in languages.
The phoneticians who started the IPA were mostly from England and France, so they created symbols using Latin and Greek alphabets, which they were familiar with.
And it comes from the days before computers, so many symbols are rotated versions of existing letters.
That way, they wouldn’t need to go forge a whole new set of letterforms or create a new kind of typewriter just to explain how to pronounce “Gavagai” or something.
Occasionally, a new symbol gets added, such as the voiced labiodental flap in 2005.
But for the most part, we've figured it out!
Thanks Thought Bubble.
The IPA chart is a bit like the periodic table, in that its structure tells us a lot about the properties of each sound.
This means that even if you don't know a symbol immediately, you can figure out what it sounds like based on its row and column.
Or, you can look up a clickable IPA chart where you can click on each symbol and hear it pronounced.
I’ve tried it.
It’s pretty fun!
The empty white boxes are sounds that could exist, but haven't been found in any known language yet, while the shaded gray boxes are sounds that we just don't think the human body is capable of making.
There are also two categories of other consonants that don’t fit in the consonant table the way it's commonly structured.
The first are sounds where the articulators touch in two places, called co-articulated consonants.
They don’t easily fit into any one column in the table.
For example [w] not only involves the lips coming together, but also a constriction at the velum towards the back of your mouth.
The second are consonants that aren’t made with air flowing out of the mouth, or pulmonic air flow, so they’re known as non-pulmonic consonants.
This category includes sounds where the air is sucked into the vocal tract, or implosive consonants, and sounds where the glottis is sealed up and air is forced out of the upper part of the vocal tract, known as ejectives.
The final group of non-pulmonic consonants are clicks.
You make a dental click sound when you make a disapproving Tsk!
Some languages, like Zulu and Xhosa and the Khoesan languages, include click sounds in the standard set of consonants used to make words.
And that’s how linguists represent all the consonants!
Whew! The International Phonetic Alphabet was crucial before it was easy to make and share recordings of language.
It’s for this reason that opera singers, speech pathologists, and accent coaches all find it useful to work with the IPA.
Now that recording and digital sound is everywhere, the IPA helps us teach computers to recognize and copy human speech.
The IPA is also great for helping you understand new speech sounds you encounter when you start learning a new language.
Unlike reading something in standardized English spelling, reading something in IPA is an intimate experience.
You're reading it back in exactly the accent of a specific person.
But so far, we've actually only talked about half the IPA — the consonant half.
In the next video, we’ll look at vowels.
Thanks for watching this episode of Crash Course Linguistics, which is produced by Complexly and PBS.
Are you craving more humanities in your YouTube diet?
Then head over to Storied, PBS’ home for the liberal arts geek in all of us.
Storied features Monstrum, a show about monsters and myths, and what these stories say about us.
As we gear up for Halloween, Monstrum’s host and writer Dr. Z is doing a deep dive on zombies, tracing their evolution from Haitian folklore to the brain-eating creatures we know and love today.
Subscribe to Storied for Monstrum and other great shows, and tell them Crash Course sent you!