Language learning is busy. One advantage to being out of the tribe is having more time to work uninterrupted on work that needs to be done, and one of those things is an alphabet for the Kovol language. Before we left we were working on gathering and triple checking the phonetic spelling of a word list of about 400 words. We can analyse this data and suggest an suitable alphabet.

This will be the first of 4 blog posts (unless of course I enjoy myself so much I keep talking) detailing the process of choosing an alphabet for a language that doesn’t have one yet. The four steps are:

  1. Eliciting data
  2. Phonemic interpretation
  3. Phonemic analysis
  4. Choosing symbols for the phonemes.
Gerdine writes stuff down, the Dutch way!

This first post is introductory to introduce the vocabulary and ideas and will also cover how we built our word lists.
The first thing to cover is phonetics and phonemics: that is, how we record some language data we can work with and why that wouldn’t make a good alphabet to start a literacy program with.

Phonetics: 1 symbol for 1 sound

Let’s start by recording some language data we can analyse. One question is how do we write it down in our notebooks? When a Kovol person says good morning how should we write it down?

  • samog bili?
  • samohg bilee?
  • samog bilie?

Having 5 different nationalities on our team will inevitably result in disagreements if we’re all going on what we were taught at school!

We need a symbol set that we can use to accurately write the sounds we hear consistently. We also need a symbol set that can handle sounds we don’t have in our own languages.

A trick we sometimes use is to use words we know how to pronounce to spell new ones: phonetics “phone – e – ticks”, but that’s a bit unwieldy and probably not consistent over the 5 nationalities in our team! Also our English letters represent more than one sound (think about the o in one and only).

Classic missionary photo, a book rather than a knife or cigarette in hand

The international phonetic alphabet is the answer. Phonetics [foᵘnɛtɪks] (we write phonetic script in square brackets) enables us to record sounds we hear consistently and it can handle all the strange (to us) sounds we might hear.
There’s a symbol for the English th sound [θ], the labial click used in Africa [ʘ], the o in one [ɔ], the o in only [oᵘ] and the Kovol bilabial fricative [β].

So we end up writing good morning as [ sɑmɔg bili]

Using phonetics we can even write the tone and rhythm of the words we hear. (Thankfully in Kovol those things aren’t used to convey differences in words!)

In theory we’re able to accurately and consistently write words as we hear them and since we’re all using the same symbol set our phonetics should match. In practice it’s a little tricky! Our mother tongues continue to influence what we hear and how we write things and disagreements over phonetics inside the team is the norm in about 75% of words, it seems.

We should be able to write consistently, but as human beings we’re not very consistent! This applies to us language learners writing the language down, but it also applies to Kovol speakers! Sometimes we hear abbreviated speech or maybe spoken in a different dialect. Sometimes people make mistakes when they speak, sometimes words sound different when carefully spoken in isolation as opposed to when in the middle of natural flowing speech, and some people just speak funny 🙂

Hence the need for the four of us learning the Kovol language to triple check the words in our word list together. Our time for doing this was cut short, but we had several weeks of Friday morning meetings discussing (arguing) over our word list and what we found.

Phonetics vs Phonemics

A phonetic alphabet is unlikely to be a good alphabet for literacy though. Phonetics is how an outsider writes a language, and it’s not the same as how an insider perceives it. In English we have a th sound. If you listen carefully to yourself saying “thistle” and “this”, though, you’ll see that the th isn’t actually the same sound, there’s a slight difference. In “thistle” the th sound is pronounced voicelessly (vocal chords off) [θ], and in “this” it’s voiced [ð].

An outsider would write [θɪsəl] and [ðɪs], but as an English man I know that it’s just th. It’s no big deal, and it doesn’t need a unique letter of the alphabet to represent it. This goes even more so for vowels seeing as we manage to cram 27 different vowel sounds into just 5 alphabet characters

Phonemics is the process of finding clues in the language data that tip you off as to what the internal understanding of the language is and identifying phonemes. Phonemes are the phonetic sounds that matter for conveying meaning. [ð] and [θ] can be united into the same phoneme, but to do so with [v] and [w] would just be confusing (was the veil in the temple torn in two, or the whale?).

By trying to understand a language from an insider’s perspective we can reduce the number of symbols needed in our alphabet to just the ones that carry essential meaning, and hopefully end up with an alphabet that is intuitive for Kovol people to learn to read and write with because it matches how they think their language works.
Jumping ahead it looks like [l] and [r] will be the same phoneme in the Kovol language (in a medial environment – that means in the middle of a word). Most people don’t seem to be able to hear the difference, and those that do know that it’s just pronunciation and it doesn’t mean anything. Our literacy program would suffer if we start teaching, and absolutely insist that [l] and [r] are very different (because they are in English). Our poor students would become confused and frustrated because we’re insisting they look at their language as an English speaker. People would drop out, people wouldn’t learn to read, and people would lose enthusiasm for learning God’s word because ‘they’re too stupid to read’. A good orthography (alphabet) hopefully sidesteps these problems.

Analyse the phonetics

Once our word list is built, the first step is to go through it and identify all of our phonetic sounds and build some charts. Here’s the Kovol consonants and vowels. Have a look at Wikipedia if you want to hear what sounds those symbols represent.

Nerd with me and appreciate this graph


Shirley Payne · 06/01/2021 at 3:52 am

Oh my. What a challenge. I’ve been praying for this part of the language acquisition.

walkerwife · 06/01/2021 at 7:35 am

How complicated! I am definitely not a linguist!!! Praying on that the Lord will help you to get this sorted, so that the Kovol people will be able to read and write, and learn from the Word of God.

Lois S. · 07/01/2021 at 11:58 pm

Does the graph indicate incidence per 100 words, 1000 sounds, or what? Do you have a similar graph for English, so we can compare? Thanks for filling us in on the challenges involved. I know that the translation of the Low German Bible was done with an orthography that did not feel right to my dad as a native speaker, and so he didn’t feel like it was helpful to use it. Blessings as you try to intuit how to do this.

Colette Harding · 09/01/2021 at 6:40 am

Thanks Steve. Very helpful to begin to understand some of the challenges ahead of you. What wonderful progress you’ll be able to make with all the free time you’ll have when the twins are born …?! :-). Seriously though, We’re praying all will go smoothly with your Covid tests and flights nest week.

Leave a Reply