A babaláwo casts the palm nuts, counts what's in his hand, and makes a mark in the powder on the tray. Single stroke or double. He does it eight times, and the pattern of strokes is the odù — the sign that names the reading.
Everybody who has looked at Ifá has seen this. Marks scratched into powder on a wooden tray.
And the standard question — where did that come from — has an awkward history. The marks are made in dust, they're binary, they come in sets of eight, and there's a North African sand-divination tradition, ʿilm al-raml, that also makes binary marks in dust. The resemblance has bothered scholars for a century.
Odwirafo's answer is a single word: the word fa means to scrape. And what does the scraping, on his account, is a bird's beak.1
Follow the image, because it's the whole argument.
Tahuti's bird is the crested ibis — the habui. An ibis feeds by driving its long curved bill into mud and soft ground and dragging it, working under the surface for what's buried there.
That is, physically, a scraping action. And it leaves marks.
So the claim assembles: the divinity of divine wisdom takes the form of a bird whose characteristic action is scratching the surface to reach what is beneath it — and the divination system named after him consists of making scratch-marks in powder to reach what is beneath a situation.1
Odwirafo's framing: "When you're doing divination, you're scraping the surface so you can go under the surface to find out what's going on in the spirit realm."1
The practitioner is doing what the bird does. The marks are a reenactment.
It's worth being explicit, because this corpus produces few arguments of this quality.
It explains a form, not just a name. Most etymological claims in this material connect two words. This one connects a word to a practice — and specifically to the otherwise-unexplained fact that the marks are made by scratching rather than by, say, arranging objects or casting lots onto a surface. The physical gesture is the thing being accounted for.
It has independent iconographic support. Kemetic material shows Tahuti's bird with its beak to the ground. That's not evidence manufactured for the argument; it's how ibises are depicted because it's how ibises behave.
And it predicts the right details. If divination marks are ibis-scratches, you'd expect them made in a loose medium that takes an impression — powder, sand, dust — rather than incised in wood or drawn in ink. Which is what the tray actually holds.
That's three kinds of fit from one claim, and none of them requires trusting a dictionary.
The sharpest thing in the segment is that Odwirafo turns the argument against a Yoruba authority on Yoruba practice.
Wándé Abímbọ́lá — the most significant scholar of Ifá in the twentieth century, an awíṣe of Ifá himself — allowed in print that the resemblance between Ifá's marks and Arabian sand divination raised the possibility of derivation. Odwirafo quotes the hedge and rejects it:
"It doesn't come from Arabian sand divination. The reason why fa means to scrape or scratch [is] because that's the title of Tahuti as the great divine habui bird who's using his beak to scrape or scratch the surface of the ground."1
And he adds the Kemetic term for making lines in the ground as further support.1
What makes this worth a case study is the structure of the disagreement. The insider scholar hedged toward external derivation; the diaspora practitioner supplies an internal explanation the insider didn't reach for.
That's an unusual configuration and it isn't obviously wrong. Abímbọ́lá's caution was appropriate — the resemblance to raml is real, the contact is historically documented, and a responsible scholar names the possibility. But naming a possibility isn't establishing it, and an internal account that explains the gesture is doing something the external account doesn't: raml explains the binary structure and the medium; it doesn't explain why the tradition's own word for the practice means to scratch.
The honest position is that both may be partly right — an indigenous scratching-divination that later absorbed structural features from contact. The corpus doesn't consider that, and it's the most likely answer.
The load-bearing assumption, stated plainly.
The argument needs fa in Yoruba — the name of the system — to be the same morpheme as a Kemetic term for scraping, and needs both to connect to Fa the Vodun palm-divinity and Tahuti's epithets.
That is a chain across three language families, and it is exactly the kind of chain Odwirafo's Comparative-Linguistic Method documents as failing the standards of comparative-historical linguistics.
But notice how much of the argument survives if the etymology fails. The gesture-explanation doesn't need it. If Yoruba fá independently means to scrape — and it does; the sense is ordinary Yoruba — then Ifá's marks being scratch-marks named after scratching is an internal Yoruba fact requiring no Kemetic connection at all.
The Tahuti material adds a cosmological reason for the gesture. It isn't needed to defeat the raml hypothesis. Yoruba can do that on its own.
Which means the corpus has, again, attached a strong local argument to a weak comparative one and made them stand or fall together.
There's a second piece of Tahuti iconography in the same session, and it belongs here because it's the same reading applied to a different image.
Odwirafo points at depictions of Tahuti and Seshat together, both bending backward, reaching to grasp an egg — and reads the enclosing structure they stand inside as the fallopian architecture of the great mother, with the pair positioned where the ovum is received.1
The reading is of a piece with the physiological mirror method documented elsewhere in this build: take a sacred image, ask what anatomical structure it maps onto, and read the deities as the forces operating there.
Its strength is that it explains a posture. Bending backward to grasp is an odd way to draw a divinity, and the standard readings treat it as compositional. Its weakness is that the method has no failure condition — any enclosing curve can be read as an anatomical structure by a reader who has decided to.
What holds the two Tahuti images together is the same verb. The bird scratches under the surface to reach what is buried. The pair reach backward to grasp what is about to come forth. Both are retrievals from a place you cannot see into — which is, again, the tradition's model of how knowing works.
You've met a claim that a ritual practice is explained by an etymology.
Ask what the etymology explains that nothing else does. Here: the gesture. Plenty of accounts explain the binary structure and the medium; only this one explains why the practitioner scratches.
Then check whether the explanation works inside one language before it reaches across several. If Yoruba fá means scrape, the local argument is complete and the comparative chain is optional. Arguments frequently import machinery they don't need, and the imported machinery is where they get attacked.
Look at the iconography independently. A depicted bird with its beak in the ground is evidence available without any linguistic claim, and it either supports the reading or doesn't.
And consider the composite answer, which contested-origin disputes usually reach eventually. Indigenous practice plus later structural borrowing accounts for more of the evidence than either pure position, and it is almost never the position either party argues for.
One extension worth recording, because it ties the cluster together.
If the divinity of wisdom is characterised by going under the surface, that is not only a description of divination. It's a description of what wisdom is in this system.
The behaviour cycle documented on Per Ankh Nut turns entirely on depth of retrieval — whether you go to a shallow layer and pull up your conditioning, or go deep and reach the actual value of the thing. That is the same image: the surface, and what's under it, and the work of getting there.
And the aku, the faculty Tahuti governs, is described as the capacity to attune to what is in harmony with divine order — which is to say, to reach the level where that becomes apparent.
So the bird's action is the tradition's model of cognition. Divination is the ritual instance; the behaviour cycle is the everyday instance. Both are scratching.
That coherence is the strongest thing about the claim, and the corpus doesn't assemble it — the pieces sit in three different sessions.
Strongest evidence. The gesture-explanation, which accounts for a feature no competing account addresses, plus iconography that exists independently of the argument.
Central tension: the comparative chain. Fa across Yoruba, Vodun and Kemetic is asserted, not demonstrated — and the argument doesn't need it.
A second tension. The raml question is treated as settled by the internal explanation. It isn't. An indigenous practice can absorb external structure, and the binary-in-eight arrangement is the feature the resemblance actually concerns.
A third. Abímbọ́lá is cited to be refuted. He is the most authoritative figure in the field and an initiate; a diaspora teacher overturning him on Yoruba practice requires more than an etymology, and the corpus doesn't acknowledge the asymmetry.
Open question. Does the ikin method have a documented history predating Islamic contact in West Africa? That's an archaeological and historical question with a real literature, and it bears directly on the raml dispute. Nothing in the range read consults it.
This is the best argument in the six sessions and it takes about ninety seconds.
He isn't comparing dictionaries here. He's looking at a bird, looking at a practice, and noticing they perform the same action — which is a different order of observation from most of what surrounds it.
And the boldness is real. Taking on Abímbọ́lá, on Ifá, from outside Yorubaland, is not a small move, and he makes it without hedging.
Set against Odwirafo Source Tensions, the pattern this build keeps finding recurs in its most costly form. The local argument is strong and self-sufficient; the comparative extension is weak and unnecessary; and they are welded together. A reader who rejects the trans-family etymology — which most trained readers will — loses the gesture argument along with it, and the gesture argument is the one worth keeping.
Terminology Drift Across Recensions. That page tracks how a practice-name's referent migrates across recensions while the word stays fixed — the term surviving intact and the thing it names quietly changing underneath it.
Set beside the raml dispute, that's the mechanism the corpus's binary framing cannot accommodate. Odwirafo's question is did Ifá come from Arabian sand divination, yes or no — and the drift literature suggests that's rarely how practices behave. A system with an indigenous gesture and an indigenous name can absorb an external structural feature, retain both names, and leave a record that looks like independent origin to one party and derivation to the other.
The insight the pairing produces is a test rather than a verdict. Where a practice is contested between indigenous and borrowed origin, the components usually have different histories, and the way to find out is to date them separately. The gesture, the medium, the binary structure, the count of eight, the corpus of verses — these need not have arrived together. Odwirafo's etymology is good evidence about the gesture. Abímbọ́lá's caution is about the structure. Both can be right, and the dispute persists because each is defending a different component while appearing to argue about the same thing.
One handshake. The concept is a contested-origin practice claim; the mechanics of how practices actually travel is its honest purchase.
Sharpest implication. The gesture argument defeats the derivation hypothesis on the one point the derivation hypothesis never addressed — why the practitioner scratches — and it does so without leaving Yoruba. Fá means to scrape in Yoruba; Ifá's marks are scratch-marks; that is an internal fact. The Kemetic ibis supplies a cosmological reason for the gesture and is not required to establish it, and by welding the two the corpus puts its best local argument inside its weakest comparative one.
Generative questions.
If gesture, medium, binary structure and verse corpus can be dated separately, which components of Ifá actually predate contact — and does anyone in either camp argue at that resolution?
The bird scratches under the surface; the behaviour cycle turns on depth of retrieval; the aku attunes below appearances. Is going beneath the surface this tradition's general model of knowing, and has anyone stated it as such?