AFII Online Museum · Room of Living Tongues

Giving the Tongue Its Letters

How a linguist sits down beside a language that has never been written, and how, together with its people, a page is finally made to hold it.

Room I

The Listening

There are, at this hour, more than seven thousand languages spoken on earth, and a great many of them have never once been set down on a page. A linguist does not walk into such a community with an alphabet already in his pocket. The first work is not writing at all — it is listening, for months or years, until the shape of the language begins to surface: its consonants and vowels, its tones, its rhythms of word and clause.

SIL International, which has now helped shape orthographies for more than 1,300 languages since its founding in 1934, describes orthography itself as something far larger than an alphabet chart: it is "the way in which a language is expressed in written form, including symbols, punctuation, decisions on where to break words and where to join them together"[1] — a whole architecture, not a coat of paint.

Field Note No two languages wear the same orthography well. A system built for one tongue's sounds, tones, and word-shapes will not simply transfer to the next. Each writing system must be cut, like a garment, to the body of the one language it is meant to clothe.[1]
Room II

The Sieve of Sound

Long before any letter is chosen, the linguist must answer a quieter question: which sounds in this language actually change meaning? Every language produces far more sounds, physically, than it uses meaningfully — a puff of breath after a "p," a slightly different vowel color at the end of a word. The linguist's task is to sort the phonetically real from the meaningfully real: to find the phonemes.

The classic tool for this is the minimal pair — two words that differ by exactly one sound and mean two different things. English speakers do this instinctively every time they hear the difference between pat and bat, or let and lit.[2] A field linguist builds hundreds of such pairs from a language with no dictionary yet to consult, listening for the sound-swaps that split meaning from meaning:

  1. Gather the raw phonetic data — record words in narrow transcription, in brackets, exactly as spoken.[3]
  2. Search for minimal or near-minimal pairs among the sounds in question. If found, the sounds are separate phonemes, full stop.[3][4]
  3. Where no minimal pair turns up, list every environment in which each sound occurs — what comes before it, what comes after.[3][4]
  4. Look for complementary distribution: if one sound appears only where the other never does, they are likely a single phoneme wearing two faces — allophones.[4]
  5. Name the basic form and the derived form, and write the rule that predicts when one becomes the other.[4]

This method has a lineage. Kenneth Pike, the great practitioner of it in the mission-linguistics tradition, called his own manual simply Phonemics: A Technique for Reducing Languages to Writing[5] — a title that says plainly what the whole discipline is for.

Room III

The Workshop Table

Here the exhibit turns a corner that the field itself turned only in the last half-century. "The days when an outside linguist would come into a language community and unilaterally establish an orthography are almost totally gone," SIL's own guidance now states plainly.[6] The linguist's analysis supplies the raw phonemic map; but the alphabet itself — which letters, which marks, which shapes feel like ours — belongs to the community that will read it.

In Papua New Guinea this principle became a literal workshop: the Alphabet Design Workshop, in which native speakers themselves handle the phoneme inventory a linguist has surfaced and choose, symbol by symbol, how their language will look on a page — weighing familiarity with a national language's script, ease of typing, and simple visual beauty.[7]

On the Wall "Community involvement is critical at every stage of development, from initial planning, to wading through the language discovery and initial proposal, to testing and approval. If an orthography is to be actually put into use, then the potential users must accept and approve it."[6]

A growing body of guidance now exists for exactly this handoff — from Wedekind and Wedekind's practical workshop manual on symbolization and sociolinguistic testing, to Easton and Wroge's community-interaction training for Papua New Guinean teacher trainers, to Hosken's field guide for writing up an orthography description once the community has chosen it.[8]

Room IV

Every Orthography Is Political

An alphabet is never only a linguistic object. Cahill and Rice's standard reference on the subject devotes early chapters not to phonology but to government policy, education systems, and what they title, without softening it, "Sociolinguistic factors: All orthographies are political."[9] Which script a community adopts, how closely it echoes a national or colonial language's spelling, whose dialect becomes "standard" — every choice carries weight beyond the page.

SIL's own working notes on orthography reform are candid about the constraints under which most of this work happens: "most of the languages still requiring new orthographies are spoken in communities without electricity, let alone computers with internet connections and the capacity for complex fonts."[10] Tone marking is a recurring case in point — many languages depend on pitch to distinguish words, yet even where diacritics exist to mark it, communities often set them aside for the plainer look of unmarked text.[10][11]

The overriding rule, repeated across the literature, is simple to state and hard to satisfy: the community must be happy with, and in control of, its own orthography.[10]

Room V

Testing It on Living Readers

No orthography is finished at the workshop table. It must be tried on real readers — children learning to sound out their own mother tongue for the first time, adults deciding whether a page in their language reads faster or slower than one in the language of school and government. Literacy specialists such as Susan Malone have built whole training manuals around exactly this stage of mother-tongue education planning.[12]

Dictionary and materials development often rides alongside this testing, and here too the center of gravity has shifted toward the community itself. Where nineteenth- and twentieth-century lexicography for under-resourced languages was typically "spearheaded by a single linguist, most often someone not from the community,"[13] newer workflows — rapid word collection, community-run dictionary projects, native documenters trained directly in the tools — now let speakers themselves decide which words belong, and how.[13]

Room VI

Two Case Stones

Konni — Ghana Native American Communities — North America
Linguist Michael Cahill developed the Konni orthography working directly in the language community, later serving as SIL's International Linguistics Coordinator for eleven years — a case often cited as a model of sustained, technically grounded fieldwork married to community partnership.[14] A number of Native American communities have instead chosen to build their own writing systems on English spelling patterns — "English-based Practical Orthographies" — sometimes setting aside more linguistically precise systems designed by outside linguists. Scholarship on this pattern argues the reason is not ignorance of the linguistics, but a reasoned preference: a system without ongoing local infrastructure to teach it "will ultimately and repeatedly be abandoned" in favor of the spelling readers already half-know.[15]

Placed side by side, the two stones teach the same lesson from opposite directions: an orthography survives only when it fits the actual life of the people who must use it — their schools, their neighboring languages, their daily habits of literacy — and not merely the elegance of its phonemic logic.

Room VII

Why AFII Keeps This Room

Artists for Israel International exists inside this same long labor. The Great Commission does not wait for a language to already have letters; Scripture translation has, across the centuries, been one of the great drivers of orthography development itself — the very reason thousands of unwritten languages received their first alphabet at all. The goal of reaching all seven thousand of the world's languages by 2033 runs through rooms exactly like this one: a linguist, a notebook of minimal pairs, and a community deciding, together, what its own words should look like on a page.

Artists for Israel International

"How then shall they call on him in whom they have not believed? and how shall they believe in him of whom they have not heard? and how shall they hear without a preacher?" — Romans 10:14