The 72 Spirits of Goetia

The catalogue has a shape before it has a meaning, and learning the shape first is the difference between a working practice and a wall of names. Three ranks hold seventy-two entries: nine Presidents, thirty-six Dukes, thirty Kings, and beneath the ranks, families of two, three or four. Each entry gives a rank, a visible form, a stated office, and a set of powers, which makes this the most structured entity material in Western grimoire literature and the easiest to use badly. The failure mode is not evocation. It is shopping, treating a catalogue of jurisdictions like a list of products and reaching for the impressive one instead of the apt one. What follows builds the beginner path: how to read an office, how to choose one entry, how to run a short working that stays inside the source text, and how to notice when a name has been advertised to you as something it is not.

Read the Structure Before the Names

The arrangement is not decorative. Three ranks, nine, thirty-six, thirty, total seventy-two, and the numbers themselves have been read as much as the entries have. A structure that regular is easier to hold in the mind, which is the main practical argument for it and a fairly good one.

The rank tells you something about the office rather than the power. Presidents are given the broad and headline domains in the text, Dukes handle substantial earthly administration, Kings handle smaller and more technical matters. That gradient is a claim about how the original compiler organised the material, and it is worth respecting because it tells you which entries a beginner will be able to read as coherent.

Families group entries that share a domain or a method. Working inside one family for a first series of workings is the single most useful habit in this material, because the differences between adjacent entries become visible in a way they never do when you jump randomly between ranks.

tracing the source material

The genealogy runs through Jewish and Christian demonological writing rather than through any single folk tradition. The most significant ancestor is the Pseudepigrapha Book of Enoch, where a set of fallen watchers is instructed to bind an order of spirits, and the instruction produces a numbered list. The seventeenth-century Conjuratio Ben-Maimon builds on that structure, and the Lemuria add the domestic framing of a protective rite performed at a specific season. The compiled catalogue that survives is late relative to its sources and was shaped by the anxieties of the period that produced it.

The entities were administrative or disciplinary officials in that scheme. They issue commands, they supply information, they control bounded domains. Nothing in the source material describes them as torturers of the innocent, and that characterisation is a modern addition with a long commercial history. The original entries read as a bureaucratic hierarchy, which is both stranger and more usable than the popular version.

What Modern Coverage Adds and What It Misses

Goetia as a visible brand, with the video essays, the app, the commentary, the tier lists, is a twenty-first-century phenomenon sitting on top of a much older text. That layer is not worthless. It has made a difficult primary material accessible to people who would otherwise never open it, and it has produced some genuinely careful summaries.

It has also flattened the catalogue. A domain that took a careful source reading to understand gets compressed into a single dramatic epithet, and the epithet then circulates further than the reading ever did. The practical result is that a beginner arrives with a hierarchy in their head that the source does not support, and the hierarchy makes the correct entry look like the wrong one.

The second distortion is the flattening of rank. Once every entry is presented as equally available and equally powerful, the distinctions that make the catalogue usable disappear. The first nine are not simply the nine most impressive, and the last thirty are not filler. Treat the flattening as a marketing artefact rather than a description.

How to Choose One Entry

The final step is the whole practice. The catalogue rewards the entry that fits, and it punishes the entry that flatters. Every seasoned reader in this material has made the same mistake of choosing on the description rather than the office, and the repair is always the same: go back and read the office.

1. Write the practical situation in one sentence, with no magical language in it at all.
2. List the words in that sentence that describe a domain rather than an outcome.
3. Search the catalogue by domain word, and read every entry in that domain rather than only the famous one.
4. Check the rank and note it, because the rank tells you how the entry expects to be addressed.
5. Read the powers in the source text twice, and mark anything that is a capability rather than a gift.
6. Pick the least dramatic entry that fully covers the domain.

A Short Working That Stays Inside the Source

The seventh step is what makes the practice cumulative. A working with no record is an event, and events teach you almost nothing. A working with a one-line record becomes a series, and a series is where you start being able to tell which framing produced which result.

1. State the situation in one sentence, using the domain word rather than a wish.
2. Name the entity and state its office back to yourself in one sentence, so you are working the text rather than an impression.
3. Draw the seal or write the name in a form you can reproduce quickly and consistently.
4. Set a short time limit, ten minutes being plenty, and hold to it even if nothing seems to be happening.
5. Speak the request in the first person, present tense, as a specification rather than as a plea.
6. Close by releasing the working explicitly rather than just stopping.
7. Write the time, the framing, and one line about your state during the working.

Why Goetic Workings Stall

Three causes account for most stalled series. The first is choosing by appearance, which produces a mismatch between the request and the office and then produces a flat series that gets read as evidence about the material rather than about the choice. The second is escalation, where a bounded request grows until the entity is being asked to override conditions no symbolic operation could touch. The third is role confusion, where a jurisdiction gets asked to deliver something belonging to a different jurisdiction entirely.

The repair in all three cases is to go back to the source entry and read the office again, slowly, including the parts that do not match what you wanted. The parts that do not match are the specification. An entry that can deliver everything has no specification, and an entity without a specification gives you no useful handle on what happened.

Testing a Working Without Fooling Yourself

Pre-register the outcome before the working, in writing, with a measure and a date. A Goetic request will usually be about a bounded thing, so a bounded measure is available. Record misses with the same care as hits, because the miss rate in this material is high and a series that reports only the hits is not a series at all.

Watch for the specific distortion that comes with this material, which is attribution to a person. A favourable outcome after a working will be credited to the entity, and the same outcome would probably have arrived anyway. The cleanest test is to run a working you expect nothing from, with the same structure, and compare. A working that produces nothing is the control your practice needs and almost nobody runs.

Tools, Materials and a Series Log

A spirit selector that filters by domain is worth having, because domain filtering is the selection method described above and doing it by memory is unreliable. A short session preparation checklist is useful for the one thing beginners most often skip, which is deciding the request before starting. Beyond those, the requirements are modest.

• A domain-filtering selector, used to shortlist entries before reading rather than to choose directly.
• A session preparation checklist that forces the request into a single written sentence.
• A series log with a column for rank, one for the office, one for the outcome and one for the date.
• Paper rather than a screen, since the entries are long and reading them on a phone is where the office gets skipped.
• One hour a week, held across eight to ten workings, which is enough to see a pattern if there is one.

Building a Series Instead of Collecting Names

The mistake that ends most Goetic practice is collection. Reading entries without working them produces the pleasant sensation of having done the work, and it never produces a result you can inspect. The alternative is a series: one entry, one domain, several workings, one log.

Ten workings against the same entry in the same domain will teach you more than ten workings against ten different entries, because the constant is what makes the variable legible. If the outcomes drift, the drift is information. If they do not, that is also information, and it is much harder to dismiss than a single flat result.

When a series finishes, write a paragraph about what changed in your behaviour across it. Most people find that the behavioural change is the durable part, and that the outcome they originally requested was the least interesting thing that happened.

What the Source Does Not Claim

The catalogue is a demonological document with no evidence of external agency behind it, and the entries are better read as a structured symbolic vocabulary than as a set of beings with wills. That reading is not a demotion. A vocabulary of seventy-two bounded domains is more useful for practice than a claim about entities, because it can be checked, compared, and applied without anyone having to believe anything.

The corollary is that the material is safest read as description. Ask what an entry claims it can do, and treat the answer as a specification of a technique rather than a report of a person. The distance from that reading to the popular one is where most of the difficulty in this field comes from, and closing it is the beginner's real work.

Where This Fits in a Broader Practice

This material pairs well with a sigil system, because both work by specification. The sigil carries an intention compactly, and the catalogue supplies a domain vocabulary for stating the intention precisely. Where the sigil alone tends toward vagueness, the office statement fixes it, and the pair is more reliable than either.

It also pairs badly with escalation and with collection, and those two habits are the same habit wearing different clothes. A practice that accumulates tools, entries and materials feels like deepening right up until the point where nothing is actually being applied. One entry, one domain, one log, repeated, will outperform an impressive shelf every time.

Frequently Asked Questions

Which spirit in this catalogue is associated with wealth?

Mammon is the entry given the domain of riches, and a duke later in the sequence, Clauneck, is described as granting riches directly. Paimon is associated with dignities and confirmations of favour, which is a different request and gets conflated with money constantly. Read the stated office in your own source text rather than trusting a summary, because the summaries vary.

Who are the nine Presidents?

Read them down the page and you find Bael, Agares, Vassago, Samigina, Marbas, Valefor, Aamon, Belphegor, then Saleos. Each of them carries a documented rank, appearance and area of power in the source. The nine are given the broadest domains in the catalogue, which is why they appear most often in modern coverage and why beginners frequently start there, though a narrower entry is usually easier to work with.

Are these spirits demons in any real sense?

They are, historically, rows in a demonological catalogue assembled from earlier Jewish and Christian material, and nothing suggests any of them acts on anybody. Practically the catalogue is a rich symbolic vocabulary with unusually consistent structure, which is a perfectly good reason to use it and a poor reason to describe it as communication.

Do I need a circle of protection or a full traditional setup?

No. A written specification, a mark, a stated time, and a place to record the result covers the substance of the work. The elaborate apparatus in the older sources is largely a matter of form, and adding it to a beginner practice tends to add anxiety rather than precision.

How many workings should I run before deciding it does not work?

Enough to make the comparison meaningful, which in practice means at least six or eight against the same entry in the same domain with a written prediction each time. A single flat result tells you very little, because flat results are also what an unresponsive day produces.