How Yoruba History Is Actually Known
The kinds of evidence historians use to reconstruct the Yoruba past, what each kind can and cannot prove, and how one 1921 book shaped everything written since.
Yoruba history before the nineteenth century na from four separate sources of evidence dem take dey reconstruct am, and dem hardly dey fully agree: oral tradition wey dem dey perform and pass down inside Yoruba institutions, archaeology and the radiocarbon dates wey follow am come, historical linguistics, and written sources wey people from outside wey sabi Arabic and Europeans write. Each one dey answer different kind question. Oral tradition dey tell you wetin community dey talk about who dem be and where their authority come from, wey be historical fact about the community even if person no fit confirm the things wey happen as dem describe am. Archaeology dey give you material remains wey get date, but e almost no dey talk anything about names, royal dynasties, or why people do wetin dem do. The honest truth na say scholars dey combine these evidence through strong argument, no be just to add dem together, and competent scholars wey read the same evidence dey reach different conclusion.
This file dey explain how the work dey go. E dey here because reader wey no understand how dem take bring out claim about Yoruba past no go fit know the difference between archaeological finding wey get proper date, reconstruction wey make sense, interpretation of tradition wey people dey drag, and assertion from colonial time wey people don repeat so tey e come dey sound like fact now. Those four things dey appear side by side inside most popular accounts without any difference.
Itan: oral tradition as a Yoruba genre, not as raw data
The Yoruba word for historical account na ìtàn, from the verb tàn, wey mean to spread out or unfold, so ìtàn literally mean something wey dem spread out make people look am . Ìtàn no be just one single genre. E cover dynastic narrative, story of how dem take found towns, the narrative parts of ẹsẹ Ifá, and the origin stories wey join with òrìṣà. Beside am, oríkì dey, wey be praise poetry for people, lineages, towns, and deities, wey Karin Barber don show say na the deepest single storehouse of historical reference inside Yoruba verbal art, precisely because e no be narrative . Oríkì dey gather in pieces, each one attach to a name, and e dey keep praise-names and allusions long after the story wey explain dem don loss. Barber argument na say oríkì dey work by placing things side by side instead of sequence, so e dey keep material wey story for don wipe comot just to make the story flow well .
Three layers:
Ọ̀yọ́ ni ilé, ìlú mìíràn ni ọjà.
Ọ̀yọ́ / na / house, town / other / na / market.
Ọ̀yọ́ na home; every other town na marketplace.
Translator: rendered here from one Ọ̀yọ́ oríkì formula wey spread well well; na my own translation from the transcribed Yoruba and make person take am as medium confidence, no be like published scholarly translation.
That line no be description, na claim. E dey place ilé, the permanent house and by extension ancestral home, against ọjà, the market, wey in Yoruba idiom get the meaning of place wey person dey visit and leave. The proverb Ayé lọjà, ọ̀run nilé, the world na market and heaven na home, dey use the same pairing, so the Ọ̀yọ́ formula dey borrow existing cosmological contrast take make political claim about who dey first. English fit carry the words but no fit carry the fact say the audience already know the other proverb. Na this exact kind material get historical value as evidence of Ọ̀yọ́ self-assertion and no get any value as evidence about timeline of settlement.
The critical point for method: ìtàn and oríkì na inside institutions wey get interest dem dey perform dem. The arọ̀kin, the group of royal bards for Ọ̀yọ́ palace, na palace officials wey their recitation dey give legitimacy to the king line wey dey rule . Origin traditions na, as Robin Law put am, political charters, and wetin dey inside dey change according to the political needs of that time . Law study on Odùduwà traditions argue say claims say dem come from Odùduwà spread go outside over time as style wey royal families take claim rank wey dem no dey claim before, wey mean say the way the tradition spread dey trace how political idea spread instead of how population move .
None of this thing mean say oral tradition useless. E just make am evidence of particular type. E be strong evidence for how a kingdom or community understand and defend itself at the time dem record am, moderate evidence for order of rulers within few centuries before the recording, and weak evidence for exact dates or for things wey happen many centuries earlier.
Samuel Johnson and the shape of everything downstream
Dem born Samuel Johnson (1846 to 1901) for Freetown, Sierra Leone, to Yoruba parents wey dem free from slave ship, and na great-grandson of Aláàfin Abíọ́dún of Ọ̀yọ́ . E return come Yorubaland when e be small pikin, train as Church Missionary Society agent, dem ordain am as deacon for 1880 and priest for 1888, and dem station am for Ọ̀yọ́ from 1881 . E actively participate for the peace negotiations wey end the nineteenth-century wars, e no just be person wey dey look from far. E complete im manuscript for 1897 and send am go CMS for London, where e take loss. Im brother, the doctor Obadiah Johnson, rebuild the work from Samuel notes; the reconstructed manuscript dey inside the ship Appam when German raider capture am during the First World War and e end up for America, and paper shortage during the war come delay the printing further. The History of the Yorubas from the Earliest Times to the Beginning of the British Protectorate finally come out for 1921, twenty years after the author die .
The reason wey Johnson give na say "educated natives of Yoruba are well acquainted with the history of England and with that of Rome and Greece, but of the history of their own country they know nothing whatever" The book come turn, and still dey till today, the single book on Yoruba wey people quote pass . Na that one be the problem just as e be the big achievement.
Na Robin Law 1984 article for History in Africa be di standard critical work on di mata . Law attack di assumption say wetin Johnson write down exactly match wetin e hear without any problem . Three clear issues dey inside am: Johnson use written source and oral source, but e no always separate dem; di text wey survive na another person reconstruct am from notes, so we no fit always tell Samuel words apart from Obadiah own; and di oral traditions wey Johnson collect, Islamic and Christian influence plus nineteenth-century political condition don already change dem before e reach Johnson hand . For example, Law show say Yoruba origin traditions of dat period carry things wey connect to Islamic legend, including di story of Nimrod and Abraham, wey mean say di tradition wey Johnson record as ancient thing na partly something wey dem mix together for nineteenth century .
Di main distortion wey matter pass na perspective issue. Johnson na Ọ̀yọ́ man, Anglican clergyman, and descendant of Aláàfin, wey dey write inside and for Ọ̀yọ́, and im book treat Ọ̀yọ́ as di center, while e treat other Yoruba polities mostly based on how dem relate to Ọ̀yọ́. Di story frame, wey make Ilé-Ifẹ̀ di spiritual foundation and Ọ̀yọ́ di political top while other kingdoms na branches, come be di default structure for Yoruba history for most of di twentieth century. Historians don correct am in three main ways.
First, historians of di eastern and coastal kingdoms, including Ijẹ̀bú, Ìjẹ̀ṣà, Èkìtì, Ondó and Ọ̀wọ̀, don show say these polities get institutional histories wey no come from Ọ̀yọ́, and for many cases, dem even old pass di Ọ̀yọ́ imperial period. Second, archaeology don set dated chronology for Ilé-Ifẹ̀ wey no depend on dynastic tradition, and wey place Ifẹ̀'s period of glory centuries before Ọ̀yọ́'s imperial expansion. Third, J. D. Y. Peel and others don show say "Yoruba" as single ethnic label for all these polities na nineteenth-century product itself, so any book wey dey describe one united Yoruba people from di earliest times dey carry modern category go put for past .
Read Johnson, but read am as primary source: document wey one particular man produce from im own particular position for one particular moment; e dey very valuable for late eighteenth and nineteenth century Ọ̀yọ́ where e get access to court informants and live through di events, but make people treat am with care for every other place.
Archaeology: wetin dated material fit talk and wetin e no fit talk
Archaeology for Yorubaland dey give absolute dates through radiocarbon and thermoluminescence, physical evidence of production technologies, and evidence of how settlements big reach and how dem arrange dem. E no dey give names, and na chronology of sites e dey establish, no be of dynasties.
Di main body of work for Ilé-Ifẹ̀ start with Frank Willett, wey do rescue excavations for Ita Yemòó for December 1957 and January 1958 afta labourers wey dey clear land for construction discover copper-alloy castings, and Willett continue other excavation campaigns reach 1963 . Di dates wey Willett get for Ita Yemòó fall inside eleventh to thirteenth century CE; im work for Òrun Ọba Adó bring out dates from eighth to twelfth century . One 2021 study wey reconstruct Willett excavation records note say di site popular well well, but surprisingly, dem publish very small detail about di archaeology, and say na unpublished records from Willett dem need take recover di stratigraphy . Dat one na general condition for Ifẹ̀ archaeology: dem recover several of di most famous objects during rescue conditions or by chance, without strong stratigraphic control, and dat one dey limit how tightly dem fit date dem.
Di recent work wey sure pass na on top glass. James Lankton, Akin Ige and Thilo Rehren propose for 2006 say di high-lime, high-alumina glass wey dem see for Ilé-Ifẹ̀ na locally dem make am, no be say dem import and remelt am, as people bin think before . Abidemi Babalola excavations for Ìgbó Olókun between 2010 and 2012 bring out pass 12,000 glass beads plus hundreds of crucible fragments wey get glass for body and other production debris . Di composition argument clear well well in a way wey stylistic argument no fit clear: this glass get roughly 10 to 20 percent alumina and 12 to 20 percent lime, one pattern wey no dey inside European, Middle Eastern or Asian glass, and e match local high-alumina pegmatite sands wey dem mix with snail shell as lime source . Therefore, dem don establish say sub-Saharan Africa bin dey produce primary glass for early second millennium CE through chemistry, no be by inference.
Wetin archaeology no fit do na to attach these findings to people wey tradition mention their names. No excavated object for Ifẹ̀ carry writing wey name any ruler. To talk say particular casting belong to particular Ọ̀ọ̀ni na art-historical argument from iconography and style, no be archaeological findings, and na so dem suppose label am.
Linguistics
Yoruba dey inside di Yoruboid cluster, wey get other members like Ìgálà and Itsẹkiri, inside di bigger Volta-Niger grouping of Niger-Congo . Early classifications put am inside Kwa and later inside Benue-Congo; di main position wey scholars dey hold today na Volta-Niger alongside Igbo, Edo and di Gbe languages . Di reconstructed ancestor language, Proto-Yoruboid, na near di Niger-Benue confluence scholars generally place am .
Linguistic evidence dey support two things well: how languages relate, and rough sequence of how dem take separate from each other. But e no dey support exact dates well. Glottochronological dates wey dey circulate for the Yoruboid split, wey people dey usually put around three thousand years before now, depend on assumptions about how fast words dey change, wey historical linguists themselves dey treat as something wey no reliable for exact chronology . Make una take any specific century number for proto-language as rough estimate. Wetin linguistic evidence establish well na say the Yoruboid languages separate from one common ancestor wey dey northeast of where Yoruba heartland dey today, and dis one agree with, even though e no prove, how Ogundiran take reconstruct Yoruba origins for the Niger-Benue confluence region .
Dialect geography na the second linguistic tool wey dey even more reliable. The internal diversity among Yoruba dialects, and how deep the difference between the northwestern, central, and southeastern groups reach, dey set limit for how dem fit reconstruct settlement sequence. Areas wey get plenty dialect diversity generally dey show say people don settle there reach long time pass areas wey get small diversity.
Arabic and European written sources
Arabic sources for the West African interior, especially the North African and Andalusian geographical tradition, get plenty information for Sahelian states and very small for the forest zone. Dem no give any early reference wey person fit use about Ifẹ̀ or Ọ̀yọ́. Claim say some particular medieval Arabic geographers describe Yoruba states na wetin person suppose check well against the real text before dem repeat am; plenty of dis kind claims for popular writing no dey survive checking.
European written sources start with Portuguese people for the coast from late fifteenth century. Dem write dem during that same time, wey be their strong point, and dem focus on coastal areas, rely on second-hand information about the interior, and business interest influence wetin dem write, wey be their limit. Dutch, English, and French accounts gather come plenty through the seventeenth and eighteenth centuries. Hugh Clapperton and Richard Lander reach Ọ̀yọ́-Ile for 1826, and na dem write the first European eyewitness description of the Ọ̀yọ́ capital, and wetin dem write na solid reference point wey person fit take check oral traditions about the empire late period.
The nineteenth century change the written record totally. Church Missionary Society agents, wey majority of dem na Yoruba instead of European, keep journals wey dem submit give CMS, and dis ones form one massive collection of dated records wey dem write almost for that same time. Peel book wey be Religious Encounter and the Making of the Yoruba stand on top dis archive, and e argue say dis African agents no just be ordinary informants, but dem be authors wey get substantial autonomy, and say as the first generation of Yoruba wey go school to read and write, dem play central role for how modern Yoruba consciousness take form . The CMS journals na the single most important documentary source for nineteenth-century Yoruba social history. Dem still be mission documents, wey converts write, for missionary society, about people wey the writers dey try convert, and person must read dem with that frame visible.
Wetin follow from all this
Four practical rules dey guide how dem write this section.
Dem keep different registers separate. When claim come from tradition, dem label am as tradition. When e come from dated excavation, dem give the date and the method. When e come from colonial or missionary source, dem identify am like that and state the position of the source.
Dates carry confidence level. Dem give archaeological dates with their ranges. Dynastic chronologies before roughly 1700 na reconstructions from how long dem estimate say rulers reign and king lists, and dem mark dem like that.
Dem name disagreement clearly, dem no dey smooth am over. Where Law and Ogundiran, or Willett and Blier, interpret the same evidence differently, dem give both positions and the evidence for each one.
Dem no carry tradition promote to fact, and dem no demote am to legend. Say dem no fit verify one account as real event no mean say e no true as statement of how people understand their own foundation, and that understanding on its own na historical matter wey deserve precise description.