Describing Language: Sounds

This is the second post in my series on describing language: linguistics for the non-linguists in my life. In this post I am going to talk about the sounds in language and the how and why of describing sounds.

 The first step in describing a language is describing its sounds.  Sounds are really, very important because each language has a set a sounds that contrast with each other and let speakers make up different words by using them.  But, the contrast set isn’t made up of single sounds so much as it is made up of sound categories.  Think for a minute about the letter t. We use this letter to represent the category of sounds we call “t”. ts can be at the beginning of a word, a middle of a word, at the end of the word; ts can appear in clusters. Say a few words outloud: table, butter, bat, stop. Do the ts sound the same? When I make them, they do not. The t in table sounds like it has a puff of air after it but the t in stop does not. The t (well, ts) in butter is really fast, just barely a tap on the roof of my mouth and my vocal cords are vibrating in my throat (something that isn’t true for stop and table!) and the t in bat feels like it’s happening in my throat. But, they’re all recognizable as t!

So, figuring out the sounds in a language is a little complicated.

Claire Bowern in her book called Linguistic Fieldwork suggests that you start by eliciting greetings and then some words. Last summer at CoLang 2014, we started with numbers and animals. The goal is to listen as carefully as possible and write down exactly what you hear. In order to help with doing that, linguists use a tool called the international phonetic alphabet or IPA. This alphabet is ideal for trying to describe contrasts because each symbol only ever represents one sound. So, all of my different ts above would be written down differently. As you can see below, this way of writing down sounds makes it really easy to see the differences because each difference has its own way of being represented. In addition to seeing the different ts, you can also see that vowels in the IPA don’t match with the spelling. This takes some getting used to but it  is so handy for pronunciation! 

table: [thebɫ]
 stop: [stap]

butter: [bʊɾɚ]

Once you have a good set of words, you try to find the environments where each sound can be found. This is how you determine what sounds are contrasts, they belong in the set of sounds in the language, and what sounds are complements, they belong in a group that makes up a particular sound in a language. Which, can be a little confusing. Each language has a set of sounds but each sounds is also its own set? Yes, possibly. Like with t above, it is possible that you have a sound that every native speaker thinks of as t but, in particular circumstances is pronounced one way but in other circumstances is pronounced another way. When you have a sound that is distinctive, linguists call it a phoneme. When you two sounds that could be the same phoneme but one sound is said in one circumstance and the other sound is said in another circumstance, these sounds are allophones of the same phoneme. Why would that happen? It might happen because the sound is affected by the sounds around it. Ideally, you’d have way more than four words so that you’d be able to see that there is a pattern and that there are rules that can describe when you hear one sound to represent t and when you hear the other. Below is an example of the environments.

[t]: s_a

[th]: #_e
[ɾ]: ʊ_ɚ

[ʔ]: æ_#

Sometimes you get lucky and you what are called minimal pairs. These are pairs of words that mean different things but only differ in a single sound. And English example is “Jess” and “Chess”. They only differ in their first sound. (And, to be technical, they only differ in one property of their first sound. J is voiced and Ch is not.) When you have a good set of environments you can make a guess about what all of the sounds in the language are. Of course, every new word you collect is more evidence for those sounds. Hopefully, you can get to the point where every new word is more evidence for your hypotheses and not a reason to rewrite the rules.

This is been a very brief overview of how linguists start to investigate the sounds of a language. Please come back for our next in the series: Morphology (or word parts!) 

 For further reading on Phonology:

Introductory Phonology by Bruce Hayes

Introducing Phonology by David Odden

twittergoogle_pluspinteresttumblrinstagramby feather
This entry was posted in Language, Life is like that and tagged , , , , , . Bookmark the permalink.