Some things are easy to tell apart, like chocolate and broccoli.

Others can be hard to differentiate, like violet and indigo. But when it comes

to the sounds of language, what you notice and what you don’t is a matter of what you

already know. I’m Moti Lieberman, and this is the Ling Space.

Every sound that gets used in language is made by moving air through some combination

of your mouth, nose and throat. This is what’s known as the vocal tract.

When you move your tongue and lips to make a particular sound, that changes the

shape of the vocal tract, and in turn, that changes the properties of whatever sound you're going to make.

So for example, the difference between a [p] and a [t], which are both sounds where no

air can escape the mouth, is that in one, you close your lips, like [p], and in the other, you

move your tongue to the alveolar ridge, that raised area right behind your top front teeth, like [t].

So we make different language sounds by having variation in the way we shape our vocal tracts.

But how much variation is enough to make a real difference in what we hear? How far do

we have to move our tongues around before we get to a new sound of our language, that’s

totally different from any other sound? It turns out there’s no universal answer to

that question. There’s no target inside your mouth that says “This far! No farther!

Or you’ll be a different sound!” The inside of your mouth is a continuous space – there’s

no actual dividing line between the different areas, no special gap or marker saying

“NOW YOU ARE IN THE ALVEOLAR ZONE”.

Every language in the world has to come to its own conclusions about how it can divide up

the different ways we can shape the mouth. Every learner, every speaker, needs to know

what those meaningful differences are. Those basic sounds a language has, those ones where

if you swap one for another, you really notice something has changed? Those sounds are

known as phonemes, and knowing what your language’s phonemes are is an essential part of

knowing that language. Phonemes are what you use to build words; if you change a phoneme, you

get a different word.

And I really mean that literally - the simplest way that linguists test whether a pair

of sounds are phonemes in a language or not is to see if switching them – and only them

– changes the meaning of the entire word. Any pair of words that differ in only one

sound, but have their meaning change, is known as a minimal pair. If you can find a minimal

pair, then both of the sounds that got swapped have to be phonemes.

Let’s look at some examples. Take the two words [blu] and [glu]. Blue glue, right? The

only thing that’s different about them is that first sound; one starts with

a [b] and the other starts with a [g], but the rest of the sounds are the same. That one change

causes the meaning of the whole word to change! Blue and glue are a minimal pair; moving [b]

to [g] makes a meaningful difference. And that means that /b/ and /g/ must be different

phonemes in English.

And we can do this with any pair of sounds in English – those sounds can be at the beginning,

like blue and glue; at the end, like cheek and cheat; or in the middle like beat and

bit. Those minimal pairs show us that /b/ and /g/, /k/ and /t/, and /i/ and /ɪ/ are

all different phonemes in English. And you could go through and find pairs like this for all

the basic sounds of English.

But like I said before, every language comes up with its own set of phonemes. So that means

that some things that are different in one language may be part of the same phoneme

in another language. So, take [i] and [ɪ]. As we just saw, those are different phonemes

in English; we know this because we have the minimal pair [bit] and [bɪt]. But if you’re a Turkish

or a Spanish speaker, those two sounds are part of the same vowel phoneme – the space in your mouth

where you can make that [i]-ish sound is just bigger, rather than being cut into two.

So to them, [bit] and [bɪt] sound pretty much the same. It’s similar for a pair of

consonants like [l] and [ɹ]; English might have a minimal pair like [lej] and [ɹej],

but Japanese only has one sound like that, like [ɾej]. This is part of

why people speaking a language other than their native tongue often sound like they

have some kind of foreign accent. They’re using their phonemic inventory and applying it to

a language with a different one!

And, lest you think English is special, it’s not. Other languages carve up the sound spectrum

in a way English doesn’t. So if you’re a French speaker, there’s a difference between

[u] and [y], like in the minimal pair [vu] and [vy], which mean you and sight. If

you’re an Arabic speaker, there’s a difference between [q] and [k], like in [qalbi] and [kalbi],

which are my heart and my dog. English speakers probably had a lot of difficulty telling the

difference between those.

Now this makes a lot of sense - we don’t make sounds the same way every time. Our tongues hit slightly

different points, our mouths close slightly differently… any number of factors could be different.

If we had to pay attention to every single detail of every phoneme, we’d never have any time to

pay attention to what anyone was saying. It’d be all, “hey, you just said ‘NERV,’

but the last time, you said ‘NERF’ (with a slightly less heavy v), and was that important?

Oh man, was that important?” No. This way, you know what’s important, and it’s

only the sounds that change meanings.

There’s a lot of research out there that shows that we can’t really tell the difference

between two sounds that aren’t phonemes in our language. For example, a 2000 study

examined responses in the brain to two different syllables, [ta] and [da]. Now, these sounds ranged

between [t] and [d] at set intervals. There’s a boundary in how we perceive [t] and [d] – if

the vocal folds vibrate later, that’s a [t], and if they vibrate earlier, that’s

a [d]. English learners notice changes in how long it takes for that vibration to start

only if they cross over that boundary – if it moved the sound from a [t] to a

[d], or vice versa. But if it varies a bunch without crossing the boundary, if it stays

in [t]-land or [d]-opolis? English learners don’t care. Even their brains don’t care.

Or how about this: French doesn’t have an [h] sound, so it's really hard for a French speaker

to hear when there’s an [h] in English. A 2011 neurolinguistic study found that English speakers

really boggle when they hear the last word in a sentence like “The mechanic filled

the tire with hair.” It’s just not right. But French speakers don’t care – to them,

it still sounds like the tire got filled with air. Nothing more normal than that! So their

brains just process it normally.

One last case about what variation we can hear: let’s think about the sentence “Let’s

scoop this goop.” Or in particular, let’s focus on “scoop” and “goop.” If you’re

an English speaker, that [k] and [g] sound different, right? Except… they’re not.

That last pair was exactly the same, except for the [s] at the beginning. Listen again:

“goop” and “goop.” But /k/ and /g/ are phonemes in English, right? So we should

be able to hear the difference between them there. That’s what phonemes are, right?

So how can that work?

That’s the work of allophones and variation that’s driven by rules, which will be our

topic for next week's video. Cliffhanger! That’s right, we’ve reached the end of the Ling

Space for this week.

For now, though, if you could hear the difference between my phonemes, you learned that

speech sounds are made by changing the shape of our vocal tracts; that there are no universal

differences between one sound and its neighbour; that phonemes are what we build words from,

because they make meaningful differences; that every language comes up with its own

set of phonemes, and that we can use minimal pairs to figure them out; and that we can only hear

the differences between sounds that are phonemic in our language.

The Ling Space is written and produced by me, Moti Lieberman. It’s directed by Adèle-Elise

Prévost, our production assistant is Georges Coulombe, and music and sound design is by

Shane Turner. Our educational consultants are Level-Up Learning Solutions, and our graphics

team is AtelierMuse. We’re down in the comments below, or you can bring the discussion over

to our website, where we have some extra material on this topic. Check us out on Twitter and

Facebook, and if you want to keep expanding your own personal Ling Space, please subscribe.

And we’ll see you next Wednesday. Go dté tú slán!