The 72 Spirits of Goetia

The seventy-two spirits of the Ars Goetia are usually encountered in the wrong order, which is why a large part of the twenty-first century treatment of this material is a content genre rather than a reference work. The catalogue presents seventy-two entries in three ranks, nine Presidents, thirty-six Dukes and thirty Kings, grouped into families of two to four. The grouping is not a hierarchy. A Duke is not above a King, and a family does not indicate an alliance. The material descends from Jewish and Christian demonological sources, principally the pseudepigraphal Book of Enoch, arriving in Europe through the seventeenth-century Conjuratio Ben-Maimon and the Lemuria material. In the original scheme these were officials, administrative or disciplinary in function, not torturers. The focus that the media has given them pushed the entries out of that context entirely, and the entry that survives closest to its source is also the one least used in the popular material. What follows is an attempt to treat the catalogue as what it structurally is: a filing system with four fields per entry, useful for organising a working, and misleading in specific, nameable ways.

The catalogue that became a content genre

A catalogue of seventy-two entries with four fields each is, structurally, a database. It has been reorganised as entertainment, which means the fields have been selected for effect rather than for utility. The rank has become a difficulty rating, the office has become a job title in a fantasy franchise, and the list of powers has become a shopping list of abilities a practitioner might acquire.

None of that is fraud, but all of it is drift, and drift is worth naming because it is the normal condition of this material. The seventeenth-century material was already a compilation of compilation, assembled for a purpose that had little to do with the sources it drew on. The twenty-first century version is a second compilation on top of the first, with the same property.

Where the numbers stop matching the text

The three-rank structure is one of the most internally consistent features of the catalogue and one of the most frequently misquoted in circulation. Countdowns of the most powerful demons routinely place a President above thirty Dukes, which is a category error, since the ranks are a division of function and not a ladder of strength. Family groupings are similarly misread, being treated as alliances or as hierarchies when the grouping is simply the arrangement of the source text.

There is also a version of the count that varies between editions, which is a real editorial feature rather than a modern error. The relevant point for a practitioner is narrower: if a secondary source's numbers do not match the primary text, the secondary source is not a reliable guide to structure, however reliable it is on atmosphere.

The economy built on which-demon-for-what

The dominant contemporary use of the catalogue is a matching exercise. A problem is described, a demon is selected, a working is built. This format has an obvious appeal and one clear problem: it selects on the field that matters least. The office describes what an entity was for within a system of administration. Selecting on it is selecting on a label, in the same way that choosing a tool by the name on its handle is choosing a tool by its label.

The alternative is to select on the form, which is the visual description attached to the entry, and then to write the office yourself in terms narrow enough to be checkable. That reverses the usual order and produces workings that are auditable, at the cost of requiring more from the practitioner up front. It is a better use of the material and it is considerably less fun.

Officials, and the century that forgot

The most consequential distortion is the substitution of torturer for official. In the source material the spirits are given positions in a hierarchy with defined functions, comparable in form to a bureaucracy with an uncomfortable remit. The administrative reading is not a sanitising interpretation; it is closer to the structure of the text than the alternative.

This matters practically, not just historically. An official is bounded by its office. Its powers are described in the language of its function, and the description is therefore an instruction about scope. A torturer has no scope, which makes the entry unfalsifiable and therefore useless for building anything. The entries whose offices are narrow are the entries that can still be worked with.

The families that do not mean what you were told

Families of two to four entries appear throughout the catalogue and are frequently described as houses, courts or hierarchies with internal politics. The grouping is present in the source, and its function there appears to be organisational convenience in the arrangement of the text. Reading political structure into it is a projection that most readers will not notice themselves making.

The practical advice is to treat a family as a search convenience and nothing more. If you are working with a family, the reason should be that you found two adjacent entries with related functions, not that you have discovered a faction. The second reason produces workings whose authority comes from a structure nobody can check.

Reading a power list without believing it

The powers field is the most quoted and the least operational part of any entry. It is written in the language of wonder rather than instruction, and reading it as a capability list produces an aim that cannot be tested. A spirit described as able to grant invisibility yields an aim that cannot fail, which is the same defect that makes a creative aim like write more useless.

Reading the powers as descriptive of a function rather than as a menu is more work and produces something usable. Whatever the entry claims about moving, hiding or persuading, ask what the entry's office would need in order to do that job. The answer is usually narrower than the text and always more specific, and specificity is what a working requires.

The one question the genre never asks

The matching format never asks what happens when the selection is wrong. Nothing in the popular treatment has a failure mode, because a failure would require a criterion, and a criterion would require the entry to have specified a result. The whole structure is arranged to make the outcome unfalsifiable.

Adding one question repairs most of the damage. Before building a working, write the condition that would show you had selected the wrong entry, and check it after a month. Most selections will be ambiguous rather than wrong, which is itself useful, since an ambiguous selection is a signal to narrow the aim rather than to increase the intensity.

What a working with a Goetic row can honestly be

A defensible working looks like this: take one row, use the office as the raw material, write your own office in one sentence with a checkable condition, design a sigil from that sentence, and keep a log for a fixed period. What you have built is a commitment device with a specific office, dressed in a tradition's vocabulary. The dressing is optional and the device is real.

What you have not built is a consultation with a described intelligence. The catalogue's own structure does not support that claim, since the fields it provides are a name, a form, an office and a list of abilities, and none of those constitute a personality that could be addressed. Anyone who tells you otherwise is reading a personality into four fields.

The material as a filing system instead

Used properly, the catalogue is a very good indexing device for a certain kind of practice. It offers seventy-two distinct offices, which is seventy-two prompts for narrowing an intention, and a form field, which supplies a visual vocabulary you would otherwise have to invent. Neither use requires believing anything about the entries, and both survive contact with people who hold none of the metaphysics.

The discipline is to treat it as a catalogue and not as a bestiary. A catalogue is consulted, indexed, cross-referenced and put down again. A bestiary is read, absorbed and gradually assimilated into a picture of the world. The first use leaves you with seventy-two tools. The second leaves you with a cosmology you did not choose.

Two centuries of invention layered on one text

It is worth being precise about how much is layered. The Enoch material is ancient and its transmission to the demonological compilations is a documented path. The Conjuratio Ben-Maimon is a specific seventeenth-century document. Everything describing these spirits as torturers, as the servants of a single arch-fiend, or as instruments of deliberate cruelty is later, and much of it is nineteenth century in origin.

The practical consequence is that the version most people have absorbed is the youngest layer. When you build a working from the material, you are choosing which layer to work with, and the choice is invisible unless you make it deliberately. The layer nearest the source is administrative and usable. The youngest layer is dramatic and not.

What survives the critique

Almost everything practical. The entries are still seventy-two distinct functional descriptions, they are still arranged in three ranks that mean something, they are still grouped, and they still supply seventy-two prompts for narrowing an intention. The office field remains the most actionable part of the material and it remains the least used part of it.

What does not survive is the framing. The torture reading, the single-arch-fiend reading and the difficulty-ranking reading are all incompatible with the structure of the source text. None of that matters if you are using the catalogue as an indexing device, and all of it matters if you are trying to build a coherent picture of what the spirits are, because the picture will not cohere and the incoherence will be attributed to the source rather than to the additions.

The entry you would actually use

If you take one row from this catalogue, take one with a narrow administrative office, a form that is a diagram rather than a body, and a powers field short enough to read in one line. There are several such entries. Choose on the office, write your own narrower office beside it, and ignore the rest of the row.

Then keep a log for thirty days with three columns: the office as you wrote it, the number of occasions on which you did the thing it names, and what you would have done anyway. The third column is the one that keeps the exercise honest, and it is the one that no secondary source will suggest. It is also the reason a narrow entry beats a dramatic one. You can audit a narrow office. You can only admire a dramatic one.

Frequently Asked Questions

What are the three ranks in the Ars Goetia?

Nine Presidents, thirty-six Dukes and thirty Kings, according to the catalogue as it stands. The ranks divide function and do not rank strength, so a President is not more powerful than a King. Treating the ranks as a difficulty ladder is one of the most common misreadings of this material and it inverts the structure of the source text.

Were the seventy-two spirits originally torturers?

The entries describe administrative or disciplinary officials within a hierarchy, and in that framing the powers are described in the language of function and scope. The torturer reading is a much later layer, largely nineteenth century in origin. The distinction is practical rather than merely historical, because a bounded office can be written as a checkable condition while an unbounded one cannot.

Do the family groupings in the catalogue matter?

They are useful for finding related functions and nothing more. A family of two to four entries is an arrangement of the source text, not an alliance, a court or a faction. Reading political structure into the grouping produces workings whose authority rests on a hierarchy that nobody can check, which is the same problem as citing a power list instead of writing an office.

Can I use the catalogue without believing in the spirits?

That is the most productive use of it, and it is what most practitioners in the tradition actually do. A catalogue of seventy-two offices is a set of prompts for narrowing an intention, plus a visual vocabulary. Neither use requires a commitment about what the entries are, and both produce workings that can be audited against a log.

Why do most sources rank the demons by power?

Because the matching format needs a selector and power makes a memorable one. A list ordered by power is easy to consult and easy to remember, and it produces a decision in seconds. Ordering by office is slower and produces a decision you can check, which is why the popular format and the usable format diverge so completely.