What an Òrìṣà Actually Is

An òrìṣà is a focus of devotion in Yorùbá religion: a named power with a history, a personality, a body of poetry, a priesthood, a shrine, and a set of people who belong to it. Beyond that basic description, scholars disagree substantially about what an òrìṣà fundamentally is, and the disagreement is not a failure of research. It reflects a genuine variety inside the tradition itself, where some òrìṣà are remembered as people who once lived and others are described as forces that existed before people did, and where a great many are both at once depending on which town you ask.

The word does not translate cleanly into English. This page sets out the range of definitions, explains why the standard English renderings distort, and describes how the òrìṣà stand in relation to Olódùmarè, the supreme being, and to the ancestral dead.

The word itself

Òrìṣà is written òrìṣà in standard Yorùbá orthography and pronounced with a low tone on the first syllable, low on the second, mid on the third. The Anglicised spellings orisha and orixá (Brazil) and oricha (Cuba) all descend from it.

The etymology is contested. Bọlaji Idowu proposed orí (head) plus ṣà (to select, to choose), producing something like "selected head" or "chosen source" 1 Other proposals derive it from orí plus ṣe or connect it to ṣà in the sense of to sift or to strain. None of these is secure, and the honest position is that the etymology of òrìṣà is not settled. Where a source presents one derivation confidently, treat that as a proposal rather than a finding.

The related term ìmọ̀lẹ̀ or imalè is used in some texts and regions as a near-synonym or as a term for a particular class of divinity, and ebọra is used for a class of powerful, often dangerous entities. These terms are not uniformly distinguished across Yorùbá country, and different scholars sort them differently.

The three-way scholarly split

The standard tripartite classification comes from J. Ọmọṣade Awolalu, whose Yoruba Beliefs and Sacrificial Rites (1979) divides the òrìṣà into primordial divinities, deified ancestors, and personified natural forces and phenomena 2.

Primordial divinities are those said to have existed before the world in its present form, emanating from Olódùmarè without human involvement. Ọbàtálá, Ọ̀rúnmìlà, and Èṣù are typically placed here.

Deified ancestors are people held to have lived, acted consequentially, and then become objects of devotion. Ṣàngó is the standard example: Yorùbá tradition itself identifies him as a former Aláàfin of Ọ̀yọ́, a historical king. Ògún and Oduduwa are also often placed here.

Personified natural forces cover the rivers, hills, the sea, and phenomena. Ọ̀ṣun, Ọya, Yemọja, Ọba, and Erinlẹ̀ are all rivers as well as òrìṣà, and the identification is not metaphorical: Ọ̀ṣun is the river at Òṣogbo in a sense that has no clean English parallel.

The classification is useful and it is also leaky, and Awolalu's own presentation acknowledges that the categories overlap. Ògún appears in the sources as an iron-working culture hero, as a hunter-king of Irè, and as the principle of iron and of the cutting edge itself. Ṣàngó appears as the fourth Aláàfin of Ọ̀yọ́ and as thunder. Deciding that one of these is the "real" identity and the other a later accretion is a scholarly move that the tradition does not itself require.

Bọlaji Idowu and the argument about monotheism

Idowu's Olódùmarè: God in Yoruba Belief (1962) is the single most influential book on this subject and also the most contested 1. Idowu, a Methodist theologian who became Patriarch of the Methodist Church Nigeria, argued that Yorùbá religion is fundamentally a belief in one supreme God, Olódùmarè, and that the òrìṣà are best understood as ministers, functionaries, or manifestations of that one God rather than as independent deities. He coined the term diffused monotheism for this position.

Idowu's motivation is legible and should be stated plainly: he was answering the missionary and colonial charge that African religion was crude polytheism or fetishism, and he answered it by demonstrating that a supreme God stood at the centre of Yorùbá thought. This was a real corrective to a real slander, and the book remains indispensable for its documentation.

The criticism of Idowu, which is now the majority scholarly position, is that "diffused monotheism" imports a Christian theological frame and flattens what the sources actually show. J.D.Y. Peel argued that Idowu and his contemporaries were engaged in a project of making Yorùbá religion respectable in Christian terms, and that this project systematically overstated the centrality of Olódùmarè relative to the òrìṣà in actual practice 3. In documented ritual life Olódùmarè receives almost no direct cult: there are no temples to Olódùmarè, no priesthood, no festivals, no sacrifices addressed to him. Devotion goes to the òrìṣà. A theology built on Olódùmarè's centrality describes something other than what people do.

The counter-position is not that Yorùbá religion has no supreme being. Olódùmarè is unambiguously present in Ifá verse, in proverb, and in ordinary speech, and Idowu did not invent him. The argument is about weight and about whether "monotheism" is the right container. Readers should hold both: Olódùmarè is real in the tradition, and the tradition's active religious life is òrìṣà-centred. 04-cosmology treats Olódùmarè at length.

Karin Barber and the reciprocal relationship

The most important single argument about what an òrìṣà is comes from Karin Barber's 1981 article "How Man Makes God in West Africa: Yoruba Attitudes towards the Orisa," published in Africa 4.

Barber's finding, drawn from fieldwork in Òkukù, is that the òrìṣà are maintained and kept in existence by the attention of human beings. Her abstract states the claim directly: Yorùbá traditional religion contains the notion that gods are made by men, and far from indicating scepticism or a decline of belief, this seems to be a central impulse to devotion; without the collaboration of their devotees, the òrìṣà would be betrayed, exposed, and reduced to nothing 4.

This needs care, because it is easy to misread as debunking. Barber is not saying that Yorùbá people privately think the òrìṣà are fictions. She is describing a genuinely different structure of relationship from the one a reader trained in Christianity expects. In the Christian model the deity is prior, self-sufficient, and unaffected by whether anyone worships. In the model Barber documents, an òrìṣà and its devotees are mutually constituting. The devotee gives attention, praise, sacrifice, and public reputation. The òrìṣà gives protection, children, prosperity, and cure. If the devotees stop, the òrìṣà's ọlá, its greatness and visibility, diminishes, and an unattended òrìṣà can effectively lapse. Conversely a small òrìṣà with energetic devotees can grow.

Barber connects this to the structure of Yorùbá social life more broadly, where a "big man" is big precisely because he has followers, and followers are attracted by evidence that following pays. The relationship between a devotee and an òrìṣà runs on the same logic as the relationship between a client and a patron: reputation is built by visible acts of support flowing both ways, and both parties are diminished if the exchange stops.

Barber developed a related argument in "Oriki, women and the proliferation and merging of orisa," on how òrìṣà multiply and fuse over time through the praise-poetry that women perform, which is the mechanism by which a local river power can accumulate the attributes of a widely known òrìṣà, or a widely known òrìṣà can absorb a local one 5. This is why òrìṣà counts are unstable and why the same name attaches to noticeably different figures in different towns.

Why "god" is a poor translation

"God" carries three implications that do not hold for òrìṣà.

Self-sufficiency. The Christian God does not need worshippers. On Barber's account the òrìṣà do 4.

Ontological separation. "God" in English implies a categorical difference in kind from human beings. Many òrìṣà were, on the tradition's own account, human beings. The line between an important ancestor, a founding king, and an òrìṣà is a gradient rather than a boundary, and the same person can be described in all three registers.

Universality. The Christian God is God everywhere. An òrìṣà is characteristically tied to a place, a lineage, a river, a town. Ọ̀ṣun is emphatically the Ọ̀ṣun river and Òṣogbo. To be a devotee is usually to belong to a particular local congregation of a particular local òrìṣà, not to hold a general belief.

"Deity" and "divinity" carry the first and third of these problems in slightly weaker form and are the least bad common English options, which is why most careful scholarship uses "divinity" or leaves òrìṣà untranslated. This corpus leaves it untranslated.

Why "idol" is worse than poor

"Idol" is not a translation. It is a polemic.

The word enters the description of Yorùbá religion through nineteenth-century missionary writing, where it does specific work: it asserts that the object of devotion is the physical carving, and that worshippers are confused about the difference between wood and power. Both halves of that assertion are false about Yorùbá practice.

The carved figure, the stone, the iron, the pot, and the beads at a shrine are not the òrìṣà. They are where the òrìṣà is addressed and where its àṣẹ is concentrated and made available. Rowland Abiodun's work on Yorùbá art and aesthetics develops the vocabulary the tradition itself uses for this, in which an image has ìwà (essential character) and àṣẹ (the power to make things happen) and is understood in terms of its capacity to mediate rather than its capacity to resemble 6. A Yorùbá devotee who has an image consecrated understands perfectly well what wood is. The relationship is closer to what a Catholic understands about a consecrated host or a relic than to what "idol" describes, and the missionaries who used the word about Yorùbá shrines were largely from traditions that also condemned Catholic practice in the same terms.

The word also does not translate anything. There is no Yorùbá term whose meaning is "idol." Ère means carved figure or image, and is used of any statue including secular ones. Òrìṣà means òrìṣà. The English word arrives with the missionaries and stays because it was written into dictionaries and school books. 01-foundations and 04-cosmology treat this demonization history directly.

A further note for readers raised Christian: the King James translation of the Hebrew Bible and the Greek New Testament uses "idol" and "graven image" for a set of ancient Near Eastern practices that Yorùbá religion is not historically or structurally continuous with. Applying the term to òrìṣà shrines is a nineteenth-century polemical transfer, not a finding.

The numbers: 201, 401, 601, 801

Yorùbá texts give the number of òrìṣà or of irúnmọlẹ̀ variously as 201, 401, 601, and 801, and occasionally as 1,440 or 1,700. These are not competing censuses. They are a rhetorical form.

The Yorùbá numeral system is vigesimal, built on twenties, with distinct lexemes at ogún (20), ọgọ́rùn-ún (100), igba (200), irínwó (400), and ọ̀kẹ́ (20,000) 7. Igba, two hundred, functions in ordinary speech as a general large quantity, and irínwó, four hundred, likewise.

The formulations then work by adding one to a round large number. Igba plus one gives 201. Irínwó plus one gives 401. The "plus one" is the point. It signals that the set is not closed, that whatever number you name there is one more, and that the count is a way of saying "innumerable" rather than a total. The same move appears in the Ifá corpus and in oríkì, where a large round number plus one is a standard idiom of completeness that exceeds counting.

A common formulation in Ifá texts places two hundred irúnmọlẹ̀ on Olódùmarè's right hand and two hundred on the left, giving four hundred, with the additional one making 401. Another formulation gives 801 by a similar construction. Jacob Olupona titled his study of Ilé-Ifẹ̀ City of 201 Gods, using the Ifẹ̀ formulation, and the book documents the actual working religious life of the city rather than attempting to enumerate 201 discrete divinities 8.

The practical reality on the ground is different from all of these. Any given Yorùbá town has an active religious life involving perhaps a dozen to a few dozen òrìṣà with real shrines, priesthoods, and festivals, and the specific roster differs town to town. William Bascom's fieldwork and the ethnographic record generally support a count in the low tens per locality, with a much longer tail of òrìṣà known by name in one or two places only 9. There is no canonical list, no central authority that maintains one, and the merging and splitting that Barber describes means the boundaries of any list are unstable in principle 5.

Anyone who presents a definitive numbered list of the òrìṣà is presenting a construction, usually a diaspora one where the ritual system did require a fixed working set.

Òrìṣà and Olódùmarè

The relationship as the tradition presents it: Olódùmarè is the source, and the òrìṣà are the powers through which the world is administered and through which human beings deal with the divine. Olódùmarè is not approached directly in cult. Ifá verses and proverbs describe Olódùmarè as the one who cannot be seen and who gives the final word, and the standard formulation is that a matter settled by Olódùmarè cannot be appealed.

Missionary and early scholarly writing sometimes described this as Olódùmarè being "withdrawn" or as a deus otiosus, a god who has gone idle. That description imports an assumption that direct cult is the measure of a deity's importance. Idowu argued against it strongly 1, and the current scholarly consensus is that "withdrawn" mischaracterises a structure in which Olódùmarè's non-approachability is itself a theological statement about scale rather than evidence of neglect.

The practical division is that Olódùmarè set destiny and the terms of existence, and that everything a person actually negotiates about their life, health, children, and work is negotiated with an òrìṣà, with one's own orí, and through Ifá. 04-cosmology develops this, including orí and àyànmọ́.

Òrìṣà and ancestors

These are distinct categories that touch at the edges.

An ancestor, baba-nlá or in ritual context an egúngún, is a specific dead person of a specific lineage. Their claim on the living runs through descent. You honour your own ancestors because they are yours, and someone from another lineage has no relationship to them. 04-egungun in this section treats this at length.

An òrìṣà is not lineage-bounded in the same way. Ṣàngó is available to anyone Ṣàngó calls, in any town where there is a Ṣàngó congregation. A person can be initiated to an òrìṣà with no ancestral connection to it.

The categories touch because some òrìṣà are, on the tradition's own account, ancestors who became so widely honoured that they exceeded their lineage. Ṣàngó is an Ọ̀yọ́ king, and Ògún at Irè, and Ọduduwa at Ifẹ̀. The process by which a very great dead person becomes an òrìṣà is precisely the process Barber describes, where accumulated devotion and reputation produce the transformation 45. Ọduduwa in particular is honoured at Ilé-Ifẹ̀ as an òrìṣà and simultaneously claimed as the literal ancestor of the Yorùbá royal lineages, and the two claims are not held apart.

There is a third category that is neither: òrìṣà ilé, household or lineage òrìṣà, where a particular descent group holds a particular òrìṣà because an ancestor of that lineage established the relationship. Here the two structures overlap directly. 02-orisa-worship-organization handles this.

What the tradition says of itself, versus what historians document

Three registers should be kept separate throughout this section.

What the tradition says. Òrìṣà are real powers, they act in the world, they are approached at shrines, they respond to sacrifice and praise, and the itan (narratives) about them recount things that happened. Within the tradition these are not symbols.

What historians document. The historically attestable material is thinner and different in kind. Ṣàngó's identification with a fourth Aláàfin of Ọ̀yọ́ appears in Samuel Johnson's The History of the Yorubas, compiled in the 1890s from Ọ̀yọ́ court tradition and published posthumously in 1921, and Johnson was a Christian convert and Anglican clergyman writing with an explicit interest in rendering Yorùbá history in a form respectable to a Christian readership 10. Archaeology at Ilé-Ifẹ̀ documents a substantial urban centre with sophisticated metal and terracotta sculpture from roughly the eleventh to fifteenth centuries, which constrains but does not confirm the Ọduduwa narratives. 03-history handles the evidence problem properly.

What colonial and missionary sources claimed. That the òrìṣà were demons, that the practice was fetishism, that the shrines were idols, and that the whole system was a degradation from an earlier purer monotheism. These claims are documented artifacts of the nineteenth century and are not evidence about the tradition. They are evidence about the people who wrote them.

Blending these three is the single most common failure in popular writing about the òrìṣà, in both directions: sceptical writing that treats devotees as confused, and enthusiast writing that presents itan as documented history.