In the olden days of Bible Translation, development of a working orthography (aka alphabet) for an unwritten language was a long and drawn out process. Having elicited several hundred words, the translator/linguist would do a phonological analysis, looking for how the different sounds were used and distributed in the language. On the basis of this analysis a list of phonemes (the building blocks of the sound system) was developed and then symbols were assigned to each phoneme. The resulting alphabet was then tested and refined and tested again. The entire process could and often did take months to complete.
When we started work among the Lower Ramu language groups back in 2002 we were introduced to a new method for developing alphabets. This new method relied more on input from the speakers of the language and less on the linguistic expertise of the expat translator. Using this method, alphabets could be developed for many languages during the course of a two-week long Alphabet Design Workshop. Here is how it worked in the Marangis language group where we conducted an alphabet design workshop in 2003.
Writing a story in their language for the first time.
The first day of the workshop saw about a dozen men and women ready to work on an alphabet for their previously unwritten language. That morning we gave them the assignment that each of them would write a short story, using their knowledge of the english alphabet, in the Marangis language. If there were sounds that they did not know how to write they were just supposed to make up a temporary symbol, or leave a blank. We would work on finding a way to write it later. At first they were a bit intimidated by the idea but most of the participants soon got into the flow of the assignment.
Once the stories were completed, each writer read their story out loud to the group. When they had all read their stories the group selected one to put on the blackboard. The author carefully wrote his story on the board and then the fun began.
At the top of the blackboard we listed each letter that had been used in the story, in roughly the same order as the english alphabet. Then starting with the first letter on the list, we asked the participants to give us three words which start with that sound, three words with the sound in the middle, and three words ending in that sound. These words were listed on a second blackboard. This is actually harder than it sounds and it was often a bit of a struggle to find nine words for each sound. And some sounds simply do not occur in all positions in a word.
Working on the Alphabet Book.
As we wrote out the words, occasionally a new sound would be found that had not occured in the original story. Those sounds would be added to the alphabet on the top of the blackboard. As the workshop went on the list of words and sounds grew until by the end of the third day we had covered all the sounds so the participants’ satisfaction. Now it was time to decide how to write those sounds.
Much of this was fairly straight forward. Since english is the national language of Papua New Guinea, we use the english alphabet as the basis for the alphabets we develop for the vernacular languages. And many of the sounds in the vernacular correspond nicely with the sounds in english. So, for example, we write the “s” sound as s and the “t” sound as t. But what about sounds that do not have a corresponding english sound?
In this case we try to find an letter in the english aphabet that is somewhat close to the sound in the vernacular. Or if there is simply no good correspondence we might use what we call a special character. A common special character in PNG languages is the barred i. This is an i with a cross through it and it represents a central vowel that does not exist in english.
There is often a lot of discussion as the participants work to decide how to write their language. Our role is to offer alternatives, show what other language groups have done with similar sounds, and help guide the discussion. Since it is their languages it is ultimately their decision how to write it.
When they had made decisions for how to write all the sounds, we asked them to go back and re-write their original stories using the new alphabet. These stories were published in an Alphabet Book for the language. This book included a list of the sounds of the language and the corresponding letters and also all the stories that were written in the course. This was the first book ever published in the Marangis language.
Charles reads a story in his own language for the first time.
Five years later, the Marangis people have published a large number of books in their language and have used this alphabet extensively. And so far we have found no additional sounds and there have been no problems with the alphabet that they developed. While this approach does not work with every language, in the case of the Marangis people it allowed them to quickly develop an alphabet and start writing in their own language.