Writing Yorùbá: Ajami, Crowther, and the Orthography Committees
How Yorùbá came to be written in Arabic and then Roman script, who decided the spelling rules, and why the subdots and tone marks are not optional.
intermediatehigh·
Loading…
How Yorùbá came to be written in Arabic and then Roman script, who decided the spelling rules, and why the subdots and tone marks are not optional.
The Yorùbá wey dem quote, the proverbs, oríkì, ẹsẹ Ifá, word list headwords, Odù names and citations dey exactly as the corpus record dem, for every language.
Dem don use two writing systems write Yorùbá. The older one na Ajami, Yorùbá wey Muslim scholars write with modified Arabic letters, wey dem record from late nineteenth century . The standard one wey dem dey use now na Roman script, wey Church Missionary Society missionaries and Yorùbá Christians develop from the 1840s, standardise for one Lagos conference for 1875, and government committees come revise for the 1960s and 1970s . The current system dey use dot under letters take distinguish letters and accents take mark tone, and the two of dem dey structurally necessary, no be for decoration.
To understand this history dey important because particular people design the writing system for particular purposes, and some of those purposes na for gospel evangelism. The writing system no be neutral thing, and to know who take which decision dey help reader see where the language own logic stop and where translator theology start.
Ajami na the general name for African languages wey dem write with modified Arabic script, practice wey dey across Muslim West Africa for Hausa, Fulfulde, Wolof, Kanuri, Nupe, and Yorùbá among others . E follow the spread of Islam and Arabic literacy, and for the Nigerian region e expand well after dem establish Sokoto Caliphate for early nineteenth century .
For Yorùbá specifically, the surviving written records come later pass wetin the Roman-script records fit make person expect. The oldest documented Yorùbá Ajami material date back to the late nineteenth century, with early examples wey dem attribute to scholars like Badamasi b. Musa Agbaji around 1891 . Ilọrin, with im Islamic scholarly establishment, na principal centre, and dem don catalogue and publish manuscript collections from the Ilọrin Emirate .
Two points need care here. First, no doubt say Arabic-script literacy don dey among Yorùbá Muslims well before 1891, but the surviving Yorùbá-language Ajami texts come later pass that literacy, so claims say very old Yorùbá Ajami literary tradition dey don pass wetin documented evidence show. Second, Ajami no be historical dead end: recent decades don see new Yorùbá Ajami texts for religious writing, short stories, advert, and political pamphlets, and e better to describe the tradition as something wey still dey continue pass something wey don die .
The technical wahala with Ajami na say Arabic script dey mark consonants and long vowels well, and short vowels only through optional diacritics, while Yorùbá need seven different vowels plus three tones. Yorùbá Ajami writers expand the script to manage am, but dem never agree on any single standardised Yorùbá Ajami orthography, and na one reason why e no turn the general written medium .
The Roman orthography come out from the Church Missionary Society work for the 1840s and 1850s. The central figure na Samuel Àjàyí Crowther (c. 1806-1891), one Yorùbá man wey dem capture for slave trade when e be boy, British naval squadron free am, bring am land for Freetown, e go school for Sierra Leone, dem ordain am, and eventually consecrate am as the first African Anglican bishop .
Crowther publish one Yorùbá vocabulary and grammatical account for 1843, followed by A Grammar of the Yoruba Language for 1852 and one expanded A Vocabulary of the Yoruba Language the same year . Dem dey generally describe the 1843 work as the first grammar and vocabulary of an African language wey native speaker of that language produce, wey be solid achievement and dem suppose recognise am as that alongside the criticisms wey follow for translation-and-bias.
Crowther no work alone or from nothing. C. A. Gollmer propose one orthography for 1847, wey enter into CMS 1848 Rules for Reducing Unwritten Languages to Alphabetic Writing in Roman Characters, the general missionary template wey dem apply across plenty languages . The early orthography no dey consistent, and Crowther own 1852 grammar no mark tone systematically .
Church Missionary Society call one conference on top Yorùbá orthography on 28 and 29 January 1875 for CMS Faji Mission House, Lagos, with Crowther among the prominent African participants . The conference resolve plenty inconsistencies for existing spelling practice and agree on the conventions wey stand for most of the next century, including the subdotted letters ẹ, ọ, ṣ and the use of acute and grave accents for High and Low tone while dem leave Mid without mark .
The 1875 system remain in use for almost one hundred years . Na im make Yorùbá Bible, Yorùbá hymnal, and Yorùbá newspaper from the colonial period still dey readable for reader today with only small adjustment.
Pressure for revision begin build up for mid twentieth century. Àjàyí (1960) call for new conference make dem reform the orthography for school use . Ayọ̀ Bamgbóṣé, wey later become Nigeria first professor of linguistics, publish Yoruba Orthography through Ibadan University Press for 1965, one linguistic appraisal of the existing system with concrete reform proposals, as e argue from the difficulties wey native speakers dey experience when dem dey use am .
Bamgbóṣé come chair the Western Nigeria Ministry of Education Committee on Yoruba Orthography from 1966, and edit the report, wey confirm the underdots on the open vowels ẹ and ọ and on the fricative ṣ, and settle the marking of High, Low, and Mid tone . Dem establish the current standards after the 1974 Joint Consultative Committee on Education report .
Contention no end for there. Disagreement still dey on top spelling conventions, word division, and pass everything on top how consistently dem suppose require tone marks for ordinary printed Yorùbá . Dis na live debate on top language planning, no be matter wey dem don settle.
Two distinct types of mark dey, and to mix dem together dey cause confusion.
Dot wey dey under letter na part of di letter. Di dot wey dey under ẹ, ọ, ṣ dey show say na different sound: ẹ [ɛ] different from e [e], ọ [ɔ] different from o [o], ṣ [ʃ] different from s [s]. To write ese instead of ẹsẹ̀ dey like to write bit instead of beat, na different word, no be say na careless spelling of di same word.
Tone mark dey carry meaning for grammar and word. As e dey explained for yoruba-phonology-and-tone, acute mark dey show High tone, grave mark dey show Low tone, and Mid tone no get mark. Because Yorùbá tone dey separate words from each other, if person remove tone mark, e dey remove information wey di reader fit no fit recover.
Dem don measure how big di confusion reach. As Iroro Orife dey work with corpus wey get one million words for Yorùbá, e find out say 85 percent of all words dey carry tone or dot mark, say 32 percent of words wey no get mark match two or more different words wey get mark, and say 64 percent of one-syllable words wey no get mark get plenty different forms wey get mark . Di average number of choices for each word wey no get mark, wey dem dey call lexical diffusion, na 1.47 . Because majority of Yorùbá verbs na one syllable, di confusion dey affect pass di exact words wey dey carry di action for sentence .
Clear example from dat work show di wahala :
Di common argument wey people dey use defend writing without tone, say person wey sabi read go fit understand from context, na true but e get limit. E dey work for normal text wey person familiar with, but e no dey work for names, words wey person no know before, poetry, church text, dictionary entries, and anything wey machine suppose process. E also no dey work for di exact kind of reader wey dis corpus dey target: person wey dey try learn how to read language wey dem dey speak but nobody ever teach dem how to read am.
Every character wey Yorùbá need dey inside Unicode, but di way dem encode am get some wahala for di structure wey make sense make person understand, because e dey explain why Yorùbá text online dey often scatter.
Unicode get single precomposed character for ẹ (U+1EB9) and ọ (U+1ECD) and ṣ (U+1E63). Wetin e no provide na single precomposed character for vowel wey get dot under wey still carry tone mark, like ẹ́ or ọ̀ . Dem reject proposals for dat kind precomposed characters, based on di general Unicode principle say dem no go add new precomposed forms if person fit combine marks build am . So ẹ́ must be encoded as base character plus at least one combining mark.
Three practical consequences follow from dis.
First, to use dead key type am dey hard. Standard dead key for keyboard dey point to one single code point, and since ẹ́ no get any, standard dead-key system no fit produce am directly . Special keyboard layout for Yorùbá dey solve dis problem, but di user must install am.
Second, di same word wey person dey see fit be different byte sequences inside computer. Di dot under fit dey encoded as combining dot below (U+0323) or, for text wey old or non-standard tools produce, as one of different combining marks wey resemble am, and di combining marks fit appear for order wey no be standard . Two words wey look di same on top screen go come show say dem no match inside software, and dis one dey scatter search, sorting, and deduplication. Di solution na Unicode normalisation, to apply NFC or NFD consistently across di whole corpus, wey be standard practice for Yorùbá language technology work.
Third, how e dey display depend on di font. Fonts wey no properly arrange combining acute on top letter wey already get dot under go make di marks clash or shift go wrong place. Fonts wey get good support for combining marks dey handle am well.
Di result na say most Yorùbá text for internet na plain ASCII wey no get any mark at all, no be say writers reject di spelling rules, but because devices and applications make am difficult . Dis one na technology-adoption problem instead of linguistics problem, and people dey work on am: automatic diacritic restoration, wey dey treat text without mark as source language and text with mark as target, now dey reach error rate wey dey below 5 percent with neural sequence-to-sequence models, and dem don release pre-trained models and datasets open for public .
Di Yorùbá writing system don reach about 180 years old for di Roman alphabet form, na Yorùbá people including Crowther help shape am well well, dem make am formal for one mission conference for 1875, and Nigerian linguists wey work through government committees revise am for 1966 and 1974. Dem design am well for di language: e dey mark di difference between di seven vowels and di three tones without wasting marks.
Di main practical problem wey e get na say people dey drop di marks plenty, and dis one don produce one generation of Yorùbá speakers wey no fit read dia own language comfortably and dem come dey think say na dia own fault. No be dia fault. Person fit learn di spelling system within one afternoon, and learning-yoruba-resources explain how to do am.
What Yorùbá is as a language, which languages it is related to, how its dialects vary across Yorùbáland, and how many people speak it today.
How missionary and colonial translation produced the English words now used for Yorùbá religion, what Crowther's Bible did to Yorùbá religious vocabulary, how Èṣù came to mean Satan, and how to read older sources critically.
The Yorùbá press, D. O. Fágúnwà and the Yorùbá novel, Tutuola's contested place, the travelling theatre, Yorùbá film, and where the language stands in education and online.
How an adult who speaks some Yorùbá but cannot read it can start, covering the alphabet, tone marks, greetings, core vocabulary, keyboards, and named quality resources with links.
What oríkì are, the different kinds, who performs them and why, and how they carry history that written records do not.
The Yorùbá sound system and its three tones, with worked minimal pairs showing how pitch alone changes a word's meaning.