Phonology and Tone

1. Deliverables for the community:

   1. Alphabet chart with pictures and a word for each phoneme and similar posters

   2. Orthography guide - appropriate for all intended audiences

2. Deliverables for linguistic purposes:

   1. Orthography proposal (more technical, bridge between phonology and orthography, with reasons for non-obvious decisions)

   2. Phonology write-up (including morphophonology, later a published phonology within a grammar)

   3. Write-up of tone system (could be part of the phonology write-up)

Phonology provides a clear map of how a given language arranges the sounds of its words. Such a map is indispensable for creating a successful way of writing the language. It is also a basis for all subsequent language research.

Good orthographies need to be decided early:

Consistency checks in Paratext at end stages of a translation process highlight the need for making good decisions with the community for orthography and word division. These decisions are best made early in the process of translation and reviewed regularly on the basis of consistency checks. We recognise that the publishing of scripture is often the benchmark for establishing conventionalisation of the language and sets the writing of the language in concrete. The scriptures are often seen as prestigious for the language or add prestige to the language.  For this reason many early missionary orthographies have remained in use around the world despite inconsistencies in the analysis of the language.  Once an orthography is set in place it is hard to change. (Story told by Mary Salisbury)

No more, no less:

Fu Gud su luved the wuld that he gave his unly Sun.

All the individual sounds of a language need to be represented in the orthography (alphabet), but not the sounds that don’t make a meaningful difference. In the sentence above, the letter ‘o’ in English spelling has been replaced by ‘u’ and ‘r’ has been left out. Isn’t it hard to read?  But we know that the letter ‘o’ represents several different sounds. A good alphabet should distinguish each different sound consistently.  It is an accident of history that English has a confusing spelling system. We can still learn to read English, but it takes years at school to do so.

Hose who can each, hose who can’, each eachers.

In this sentence the letter ‘t’ has been missed out entirely.  In Cook Islands Māori, the glottal stop represents a sound in the language, but it hasn’t been written consistently in their alphabet for 200 years! This example shows how hard it would be for them to read without the glottal stop. (Story told by Mike Cahill)

ANS: Those who can teach, those who can’t, teach teachers.

We don't need to write all phonetic differences in the orthography

Ideally an alphabet should represent only the sounds that make a meaningful difference to words, not all the phonetic differences that native speakers don’t even notice they are making.

I like a lituw amond miuwk in latte.

In NZE the sound l is sometimes missed out or pronounced more like a vowel before consonants and at the ends of words, but native speakers don’t notice they are doing so.  They don’t really need those two types of l to be distinguished in their writing, do they!

In many Polynesian languages vowels can be lengthened in the formation of reduplicated words (e.g. Pkp: ngalepe ‘torn’; ngālelepe ‘torn (plural subject)’; ngālepelepe  ‘torn to pieces’), and grammatical words (e.g. ko and ia) can have a lengthened vowel before a short word containing two vowels, but not before longer words (e.g. Pkp: kō Mele (Mary), but ko Petelo (Peter), iā Mele, but ia Petelo).  This type of vowel lengthening is predictable.  Would you write this type of vowel lengthening in an orthography for this language? 

Phonological research can be accomplished in several ways. The well-known traditional approach starts with phonetic transcriptions, i.e. with what linguists think they hear (at times underpinned by acoustic analyses of recordings) and basing the search for phonemic contrasts on (quasi-) minimal pairs.

A more recent alternative approach, well-tested but not well-known yet, starts instead with what speakers perceive they are saying. It also establishes the set of phonemes by their place in the system of sounds, according to the structural linguistics tradition. We believe this innovative approach to linguistic fieldwork has the potential to become a game changer for linguistics in BT projects:

• quick access to the phonemic level of the sound inventory (consonants, vowels), adding phonetic detail later as needed

• achieve several goals simultaneously: linguistic analysis, awareness raising for all, training of all, participants and facilitators

• early written documentation of some results (by linguists, facilitators) for backflow of initial literacy material into the community

• involvement of the local community as much as possible, thus offering higher chances of sustainability (ownership, localisation)

Some Literature on Participatory Research in Linguistics:

Constance Kutsch Lojenga 1996: Participatory Research in Linguistics. Notes on Linguistics 73:13–27.

Tim Stirtz: Rapid Grammar Collection.

Stirtz, Timothy, Michael Cahill and Philip Davison (Eds). Participatory linguistics: Methods and case studies from around the world. Language Documentation & Conservation Special Publication no. 29. Honolulu: University of Hawai'i Press.