In the middle of a Latin conjuration for water elementals, the sentence stops making sense. It runs to UT ELEMENTUM SPIRITUS AQUAE NN — and then two capital letters that are not Latin, not Hebrew, and not a name.1 A footnote explains them: NN, you will replace here the name of the desired elemental.
That is the whole documentation. One clause, in one footnote, attached to one formula — and it is the only place in this entire body of material where the practice of substituting a variable into a printed formula is stated outright.
NN is real and old. It stands for nomen nescio — I do not know the name — and it is the standard Latin placeholder in legal, liturgical and ritual texts for a name to be supplied when the form is used. A baptismal rite prints N.; a grimoire prints N.N.; a Roman legal formula does the same.
So the convention is inherited rather than invented, and using it is the correct thing to do. What is unusual is how little is made of it.
Count the appearances and the imbalance is the finding. NN appears five times in the converted file — once in the water conjuration where the footnote explains it, and four more times afterwards, in formulas where nothing explains anything. The reader is expected to have caught it the first time and to carry it forward.
And the same device, without the letters, runs through the whole practical material. Eleven exorcisms share one frame with only the substance noun swapped. The four Guardians of the Towers share one sentence with a name and a direction changed. The Phoenician Star runs eight epithets through one line. The base-oil consecration runs four. The seven-pointed star runs four more.
In none of those is the substitution mentioned. The formulas are printed complete, one after another, as though each were its own text — and a reader who has not noticed the pattern will learn thirty formulas where there are five. ⚠ So the material documents variable substitution exactly once, in a footnote, for one specific case, while relying on it everywhere. Reported as counted.
The distinction is worth drawing because it changes what a practitioner has to know. A frame slot is filled from a closed idea. Exorciso igitur te creatura takes a substance, and there are only so many substances in the curriculum — metal, stone, oil, herb, earth, fire. A reader can work out what belongs there by looking at the other ten instances.
NN is filled from a list. The footnote says to insert the name of the desired elemental, and the elementals are named in the tables that follow: Primordial Kings of Water — MENAS, HAMAN, KOLOS, KARSOS, ONADOS, HASON, who deal with the Obedience of Waters. So NN is not a category slot. It is an address slot, and what goes in it comes off a roster. Which means the conjuration formula is not a spell for water elementals in general. It is a spell for one named being at a time, and the choice of which is the operator's.
That is exactly the position the conjuration school describes — you address a being whose portfolio you know — and here it is implemented as a blank in a sentence.
The NN footnote is a key to the rest of the material, and it is easy to miss because of where it sits. Once a reader knows that a printed formula can contain a slot, the whole spellbook reads differently. Nothing else is marked. The exorcisms do not print creatura NN. The Guardians do not print NN EGO RECIPERO. The Phoenician Star does not print DEUS NN. Every one of them prints a filled example instead — and a filled example is indistinguishable from a fixed text unless you have several to compare.
Which is why the super is calix error is worth so much. A gemstone formula that still says upon this chalice is a slot caught mid-substitution, and it demonstrates from the inside what the NN footnote states from the outside: these are forms with variables, being filled in.
Two pieces of evidence, one explicit and one accidental, and together they settle what kind of thing the formulas are.
What fills NN is not free. The tables that follow the footnote supply the names, arranged by rank and by element, and the water entry gives the shape of all of them: Primordial Kings of Water — MENAS, HAMAN, KOLOS, KARSOS, ONADOS, HASON. Deal with the Obedience of Waters.
Six names, one rank, one stated business. So the practitioner choosing what to put in the slot is choosing among named officers with a portfolio, which is exactly the information the conjuration school says matters most about any being you address.
⚠ The full roster across all four elements was not censused in this pass, and the ranks above and below Primordial Kings are not recorded here. Re-check the plate.
⚠ Three things a practitioner would need are missing. Nothing says which formulas have slots. NN is marked in one; the rest are not marked at all.
Nothing says how to pronounce a filled slot. In a system where a name is a number resolving to a code, and where a name spoken carelessly is a different name, inserting a foreign name into a Latin sentence raises a real question about what the resulting sentence sums to. It is never asked.
And nothing says what happens to an unfilled slot. A practitioner who recites the water conjuration exactly as printed will say the letters N, N out loud in the middle of it. No instruction addresses that.
Short, and mostly a matter of noticing. When you meet NN, substitute. Put in the name of the elemental you are actually addressing, taken from the tables — MENAS, HAMAN, KOLOS, KARSOS, ONADOS or HASON for the Primordial Kings of Water, and the corresponding names for the other elements.
Do not say the letters. They are a placeholder, not a word. And read every other formula for slots that are not marked. Compare an exorcism with the next exorcism; compare the four guardian invocations; compare the eight lines of the Phoenician Star. What is identical is the frame and what differs is the variable — and knowing which is which is the difference between learning five sentences and memorising thirty.
The footnote is quoted exactly; the count of NN occurrences is this build's own, run against the converted file.
**NN as nomen nescio is a genuine and well-attested convention in Latin liturgical, legal and ritual texts. Its use here is correct and conventional.
⚠ NN appears five times and is explained once. Reported as counted. ⚠ The elemental name-lists supply what goes in the slot**, and the water rulers are quoted above. The full roster across all four elements was not censused in this pass.
⚠ This build's working notes carried a further claim about where four of the weapon exorcisms come from — that the Chalice's derives from the Emerald Tablet, the Sword's from Lévi, and the Pentacle's and Wand's from psalm texts. That claim was not verified in this pass and is not carried onto this page. It is recorded here as outstanding rather than repeated as fact. Unresolved: nothing states whether other formulas contain unmarked slots, and the material never raises the question.
Follow the footnote and the formulas are working forms. A slot is marked, a substitution is instructed, and the practitioner is expected to complete the sentence before saying it — which is how a functioning ritual text behaves. Follow the presentation and they are finished artefacts. Every other formula is printed complete, with no blanks, no brackets and no marks, as texts kept secret and hidden away by the Wizards of the Eldest Tradition and handed down unchanged.
Both are true of the same pages, and the difference is invisible to a reader working from any single formula. Only somebody comparing several will see the seam.
Akan-Only Language in Ritual — Both traditions are exact about what may be said in a ritual formula, and in both that exactness is treated as doing real work rather than as pedantry. Neither allows a practitioner to improvise the wording.
The Akan rule fixes the language entirely: do not speak English at the shrine, even to ancestors who never knew Akan, because the medium itself corrupts the space. Nothing is variable. A formula is what it is, in the language it is in, and a name spoken in the wrong tongue is a fault.
Here a name from one language is dropped into a sentence in another, by design, and neither corpus notices the question that raises. UT ELEMENTUM SPIRITUS AQUAE is Latin and MENAS is not; the slot is built to take a foreign name. In a system where a name is a number resolving to a code, a mixed-language sentence has a mixed-language sum — which the Akan practice would treat as the whole problem and this one never mentions. One tradition protects the medium and the other protects the address, and only the first has a doctrine about what happens when they conflict.
Sharpest implication. The substitution device is stated once, in a footnote, for one formula — NN, you will replace here the name of the desired elemental — and used silently across five frames and thirty-odd printed formulas. A reader who misses the footnote will learn each one as a fixed text.
Generative questions.