Banebdjedet is usually shown as a ram, or a ram-headed man. One head.
But there's another form. In some depictions he has four heads — four ram heads on one body, with the solar disc and the serpents.1
Egyptology reads the four heads as the four bau of major divinities united in one form, which is well attested and reasonable.
Now go and look at a diagram of the thyroid.
There is one gland at the front of the neck. And behind it, in variable positions, there are four parathyroid glands.2
One in front. Four behind.
Because it is a number matching a number.
Almost every siting in this corpus works by functional analogy — this force regulates, that organ regulates, therefore. Analogies are cheap and hard to break.
This one makes a numerical claim. Four heads, four glands. It could have been three. It could have been two. Human parathyroid count is usually four but genuinely varies — some people have three, some have five or six.
And the corpus states the standard figure and the variability together: humans usually have four, located on the back of the thyroid, in variable positions.3
A guess would have to land on four specifically, about small glands most people have never heard of, that are easy to miss in dissection and were not distinguished as separate organs until the nineteenth century.
Whatever else is true, that's a fact worth stating precisely.
The corpus reads the function off the reference and it matters for the reading.
The parathyroids produce parathyroid hormone in response to low blood calcium, and it plays a key role in regulating calcium in the blood and within the bones.4
So: the thyroid regulates the body's metabolic fire. The parathyroids regulate calcium — which is to say, the mineral content of the skeleton.
And Banebdjedet is master of the djed, and the djed is the spine.5
The four glands behind the one gland are regulating the bone the whole name is about.
That's a second-order fit. Not just four-and-four, but four glands whose job is the maintenance of the very structure the divinity is named for mastering.
And now the honest problem, which is that the same teacher gives a different explanation of the four heads, minutes later, in the same session.
The four heads are the ba of Ra, the ba of Shu, the ba of Geb, and the ba of Asar — four spirits contained in one vessel.6
His image for it: you're driving a car, four people get in, they contribute direction and input, but you're the one holding the wheel.7
That's the standard Egyptological reading, and he gives it fully.
He does not reconcile the two. He does not say the four bau are the four glands, or that one reading is figurative and the other literal, or that both hold at different levels.
He gives the parathyroid reading, moves through other material, and then gives the four-bau reading, and both stand.
The vault's rule is that contradictions are preserved, not silently resolved, and this is a clean case for it.
Because the two readings are not compatible by simple addition.
The parathyroid reading says the four heads are a count of a physical structure. The four-bau reading says they are four named divine spirits riding in one vessel. One is anatomical enumeration; the other is theological composition.
You could force a synthesis — the four spirits are seated in the four glands — but the corpus never says that, and inventing it would be building vault content under the teacher's name.
What's more interesting than a synthesis is that this is the corpus's own evidence that its readings are underdetermined. Given one image, its method produced two incompatible outputs within minutes, and it didn't notice.
[n=1 OBSERVED] as an internal contradiction. And it's the single clearest instance in the build of the method generating more than one answer from the same input.
The parathyroid reading makes a claim that is, in principle, datable.
If the four-headed form predates any plausible route to knowing about the parathyroids, the correspondence is either coincidence or something harder to explain.
If it's late — Ptolemaic, say — the reading weakens considerably, because late-period iconography absorbed a great deal.
And there's a second check: do other Kemetic divinities appear in four-headed form? If four-headedness is a general convention for composite divinity, the number is doing generic work and the parathyroid match is coincidence.
Neither check is run. Both are cheap.
He shows the one-headed form. Then the four-headed form, with the disc and the serpent.
And he asks the question directly: if he's the ba, the master of the djed pillar, and he governs the thyroid — why would he have four heads also? What is that expression?8
Then he pulls up the parathyroid diagram.
He also says, plainly, that this is something Egyptologists don't know and that people studying Kemetic material are unaware of.9 Which is a claim about novelty and it's probably accurate — the identification is his.
You look up the parathyroids and find they were not identified as distinct organs until the 1800s, and that they are small, pale and easy to mistake for fat.
Which makes four an odd number to have arrived at by any route other than counting them.
You watch the same session produce two explanations for one image and notice that both felt convincing while he was saying them.
Not because either was dishonest. Because the method produces outputs, and a method that always produces an output will produce two when asked twice.
That's the most useful thing this page contains, and it isn't about parathyroids.
This is the corpus's strongest claim, so it deserves the most careful weighing in the build.
What is certainly true. The four-headed form exists. Humans usually have four parathyroid glands. The glands sit behind the thyroid. The thyroid sits over the vertebrae. The name means master of the djed.
None of that is disputed by anyone.
What is the claim. That the four heads depict the four glands.
What supports it. The numerical match, which is specific. The positional relationship — four behind one — which matches the iconography's structure of four heads on one body. And the functional fit, since the glands regulate the bone the god is named for.
Three independent alignments, and that's more than any other siting in the course gets.
What weakens it.
The corpus supplies a rival reading itself and doesn't choose. That is unusually damaging, because it isn't an outside objection — it's the same method, on the same image, giving something else.
The dating is unaddressed. Everything hinges on it and nobody looked.
Four-headedness may be conventional. If composite divinities are generally shown four-headed, the number carries no information.
And the direction of reasoning is unclear. The heads were known first; the glands were found to match. That is confirmation-shaped rather than prediction-shaped, and confirmation-shaped reasoning is what produces the corpus's weaker claims elsewhere.
The verdict. This is the best-constrained claim in twelve hours of material, and it is still [n=1 OBSERVED] — one image, one match, with a live alternative from the same speaker and two cheap checks unrun.
Which is the honest shape of the whole corpus in miniature. Its strongest case is genuinely interesting, genuinely unproven, and one afternoon's work away from being much better or much worse understood.
Strongest support. A specific number matching a specific number, in a structure that is small, obscure, and was not distinguished in Western anatomy until the nineteenth century — combined with the glands' function being maintenance of the exact bone the god's name is about.
The tension, and it is the corpus's own. The four-bau reading is given by the same teacher in the same session and is not reconciled. Preserved, not resolved.
A second. Dating is unaddressed and the whole claim turns on it.
Open. Two cheap checks. Does the four-headed form predate any plausible anatomical knowledge? And is four-headedness a general convention for composite divinities, in which case the number is generic?
Against The Thing Sitting on Top of Your Spine, this page is the second half of one siting and the stronger half.
That page argues the thyroid from function and position. This one adds a count, which is a different and better kind of evidence — functional analogies can be constructed for almost anything, and numbers cannot.
Against What She Wears Is Where She Is, this is that page's flagship case and its clearest problem, together.
Flagship, because the image supplies information the anatomy didn't already contain — the number came from the picture.
Problem, because that page's central worry is that the method has no stopping rule, and here the same image yields two readings and the method offers nothing to choose between them.
Authority Construction and Belief
Both pages are about how a claim acquires the feel of being established, and this one is a specimen worth keeping because the claim may well be true.
The authority-construction analysis shows that confidence, specificity and delivery do most of the work that evidence is assumed to do. A precise claim delivered evenly reads as settled; a hedged claim reads as weak, regardless of which is better supported.
Here both readings of the four heads are delivered identically — same pace, same confidence, same absence of hedging. And they cannot both be right.
The insight the pairing produces is that register carries no information about support, and that this is a property of the speaker's practice rather than of any individual claim.
Which yields a usable habit for reading any corpus: the confidence of delivery is not a signal, and where a source is uniformly confident, you have to grade claim by claim from the outside because the source will not grade them for you.
And the sharpest version of that here: the parathyroid claim may be the best thing in twelve hours of material. It arrives sounding exactly like the Christmas-tree derivation.10 A listener calibrating on tone would weight them equally, and would be wrong about both.
Sharpest implication. The corpus's single best claim and its own refutation of that claim were delivered minutes apart, in the same voice, by the same person, about the same image — which is the strongest available evidence that this method produces readings rather than finding them.
Generative questions.