How This Corpus Handles Evidence

Almost everytin wey dey spoil when pipo dey write about Yorùbá tradition dey spoil di same way: dem dey write three different kind of statement for di same voice, so di reader no go fit know which one be which. "Odùduwà come down from di sky for Ilé-Ifẹ̀," "radiocarbon date from Ifẹ̀ show say heavy settlement start from around di eighth century," and "di native pipo dey worship idol and na superstition dey control dem" na all sentence about Ifẹ̀, and dem no get anytin at all in common as evidence. Dis file dey explain how dis corpus take separate dem.

Di three register

Register one: wetin di tradition tok about imself

Dis one cover ìtàn (story, history, di account of how sometin take happen), oríkì (praise poetry wey dey attached to person, lineage, town or òrìṣà), ẹsẹ Ifá (di verse dem of di Ifá corpus), proverb dem, song dem, and di account wey priest dem, elder dem and title-holder dem give about dia own institution dem.

Dis ones na primary source wey dey very important and dem no dey neutral. Na pipo wey get position to defend dey perform Yorùbá oral genre dem. oríkì wey dey praise one lineage na dat lineage produce am and na for dia own self dem produce am. Dynastic ìtàn wey dey establish say one ruling house come from Odùduwà na claim to legitimacy wey dem make for competitive ground where oda house dem dey make oda claim dem. Na babaláwo dey pass di Ifá corpus down, and na di corpus itself dey give dem authority. None of dis make di material unreliable; e make am evidence of one particular kind, wey historian dem dey handle by asking say who dey perform, for who, and against which rival account.

When dis corpus report material for dis register, e dey tok am clearly: "di tradition tok say," "di ìtàn of Ilé-Ifẹ̀ give," "inside di Odù Ọ̀sá Méjì dem tok say." Dat kind phrase dey do real work and no be say person dey try dodge word.

Register two: wetin historian and archaeologist dem document

Dis one cover excavation and how dem date am, linguistic reconstruction, art-historical and material analysis, di reconstruction of trade from goods and residue dem, documentary record where dem dey, and di whole work of academic history.

E get im own limit dem, and e good make dem state dem. Radiocarbon dey give date range for sample, no be date for an event, and di connection between di two need argument. Linguistic reconstruction dey give relative sequence and rough time-depth, no be calendar year dem. Excavation for Yorubaland small well well compared to di size of di region and dem face only few site pass, especially Ilé-Ifẹ̀, so when evidence no dey, na very often because digging no dey. Where claim for dis register depend on single excavation or single scholar, dis corpus dey tok am clearly.

Di most important recent work for dis register for Yorùbá deep history na Akínwùmí Ògúndìran im The Yorùbá: A New History, wey combine archaeology, historical linguistics, environmental science and oral tradition into one single account and trace Yorùbá origin go di region of di Niger-Benue confluence wit expansion later wey climate change partly drive 1. Dem cite am throughout di history section.

Register three: wetin colonial and missionary source dem claim

Dis one cover Church Missionary Society journal dem, wetin traveller dem write, colonial administrative report dem, and di early ethnography wey missionary dem and official dem write.

Dis source dem dey indispensable and dem get systematic distortion. Dem dey indispensable because for big part of di nineteenth century na dem be di only continuous written record of Yorùbá life, and because na mostly Yorùbá agent dem keep di CMS journal dem especially, pipo wey be di first literate generation of Yorùbá people, wey make dem be Yorùbá source as well as mission source 2. Dem get distortion because di author dem dey inside argument. Missionary wey dey describe òrìṣà shrine dey describe sometin wey im dey dia to abolish, wit vocabulary wey im theology supply, for reader dem for England wey dia donation dey fund di mission. Di way J. D. Y. Peel read dis archive dem establish di two fact together at once: di journal dem rich well well, and di author dem "saw heathenism as an absence or vacuum, rather than as something with durability, style, and an ethos of its own" 2

Dis corpus dey cite colonial-era source dem freely and always dey label dem. Where dat kind source na di origin of claim wey still dey circulate, dem dey trace di claim go back to am and dem dey give di context of di claim. Di file on di demonization na mostly exercise for exactly dis tin.

Why di separation matter

To mix register one wit register two dey produce celebratory error: tradition wey dem restate as established history, so dat "Odùduwà come from di east" come turn to evidence of migration from Arabia or Egypt. To mix register three wit register two dey produce hostile error: missionary characterisation wey dem restate as ethnographic fact, so dat "di Yorùbá dey worship idol and dem no get any idea of God" go enter reference literature and stay dia for one century. Di two error get di same shape. To keep di register dem apart na di single most useful discipline wey dey.

Di confidence marker dem

Every file dey carry confidence: high | medium | contested for im frontmatter, and di marker dey point to di file im central claim dem, no be to every sentence wey dey inside am.

High mean say multiple independent source dem of good quality support di claim and working scholar dem no dey seriously dispute am. Yorùbá na tone language wit three phonemic tone dem. Dem organize di Ifá corpus into 256 Odù. Dem write Ifá enter UNESCO Representative List for 2008 3. Dis ones na high-confidence claim dem.

Medium mean say support dey for the claim but the foundation no too strong, or na interpretation wey reasonable scholars fit disagree on, or e involve to reconstruct from evidence wey no direct. Most of the history before nineteenth century dey fall under here. Na so most translation of philosophical vocabulary be too, because to translate àṣẹ or ìwà enter English dey involve interpretive choices wey the English word come cover up.

Contested mean say scholars dey actively disagree, and the file mention the different positions instead of to pick one. Whether e correct to describe Yorùbá religion as monotheistic na contested matter: E. Bọ́lájí Ìdòwú argue for 1962 for wetin e call diffused monotheism, wey be say the òrìṣà na different ways to conceptualise the attributes of Ọlọ́dùmarè, and people don criticize that interpretation say e carry Christian theological frame put inside system wey no arrange itself like that 4. The date for the Ifẹ̀ sculptural corpus na contested matter. Whether specific dynastic sequences really happen for history na contested matter. Anywhere wey you see this marker, make you expect say the file go lay out the disagreement instead of to settle am.

Small note on wetin confidence markers no be. Dem no be way to measure how important something be or how strongly anybody believe am. Say millions of people believe one religious claim with full conviction no mean say e be high-confidence historical claim, and to talk so no be comment on the religious standing of the claim. The marker dey describe evidence wey dey support am, and nothing else.

Knowledge wey initiates hold, and wetin make am remain as gap

Better part of Yorùbá religious knowledge no dey open to the public. Na inside initiation dem dey pass am down, under conditions wey include restriction on who fit receive am and wetin dem fit talk for outside. This one apply to some parts of the Ifá corpus, to the interior content of particular òrìṣà cults, to Ògbóni and Òṣùgbó procedure, to some parts of masquerade practice, and to plenty pharmacological and ritual knowledge wey herbalists and priests hold.

This corpus no dey try to provide that material, because of three reasons.

The first reason dey straight to evidence matter. Anywhere wey dem restrict knowledge, wetin dey circulate for public no dey reliable by default. "Secret" material wey dem publish na usually half-story from informant wey no get right to give am out, or wetin outsider reconstruct, or wetin dem fabricate. Any corpus wey repeat am go dey present low-grade material for the exact points wey readers want certainty pass, wey be the worst place to dey inaccurate.

The second reason be say the restriction itself na fact about the system and e suppose dey described instead of to dodge am. Say Ifá training dey take years, say na order-by-order dem dey learn Odù, say na certain stages dem dey give certain material, say some things na only inside sacred grove dem fit talk dem: all these na structural features of how dem arrange and pass the knowledge down, and dem dey tell you real truth about the tradition. To report the structure dey give more information pass to smuggle small pieces of the content come out.

The third reason be say the alternative na speculation, and speculation for this area get habit wey don establish to dey turn citation. Guess wey dem publish for one book go enter footnote for the next one, and by the third book e don turn fact.

So anywhere this corpus reach boundary, e dey state am clear like this: this material na initiate-held, dem no publish am, wetin public evidence confirm na the following, and the corpus no dey cross beyond that point. Files wey this one be their main limitation dey carry the marker confidence: contested.

The translation rule

Any Yorùbá text wey dem quote anywhere for this corpus, whether proverb, oríkì, ẹsẹ Ifá, song or prayer, dem dey present am in three layers: the original with complete diacritics, literal gloss wey show wetin the Yorùbá really talk including the metaphor before dem polish am, and idiomatic English rendering wey mention the translator name. After that, note on wetin the translation lose.

This rule dey because translation for this field na where most of the damage happen and na where plenty of am still remain. The clearest example na the mission-era translation of Èṣù as Satan inside the Yorùbá Bible, wey be translator decision wey happen for mid-nineteenth century wey get consequences wey still dey shape how Yorùbá people dey understand their own heritage 5. That decision no dey visible to person wey dey read the Yorùbá Bible; e dey look like fact about Yorùbá religion instead of choice about English. To show the layers na the way to make such choices visible.

Anywhere published translations differ well well, this corpus dey show the two and mention the translators name. Anywhere wey mission-era translation clearly take shape from the translator theology, dem dey state am and show wetin the Yorùbá really talk. Dem no dey silently correct sources at all.

Wetin this corpus dey do when e no know

E dey talk say e no know, and e dey talk wetin go fit settle the question. That formulation dey appear everywhere. E dey more useful pass confident sentence, and na the only honest thing to write when evidence no dey.