The 72 Spirits of Goetia

The seventy-two spirits of the Ars Goetia are the most reproduced piece of western grimoire material in circulation and the least well understood. Between the seventeenth-century text, the twentieth-century compilations and the present-day media, they have acquired a reputation their original context does not support. This page works through the structure of the list, traces where it came from, explains what an entry actually instructs, and sets out a way to work with the material that requires no belief in any of the claims attached to it. It is written for practitioners who want the source rather than the brand, and for readers who have been handed a recommendation without a domain and would like to know how the selection is supposed to work.

What the list is, structurally

The Ars Goetia presents seventy-two spirits arranged in three ranks. Nine Presidents, thirty-six Dukes and thirty Kings, grouped into families of two, three or four and numbered within their classes.

The rank is a statement about the size of the jurisdiction rather than about power in any comparable sense. Presidents have kingdoms or seats, Dukes have duchies, Kings have cities and noble estates. Reading the ranks as a hierarchy of supernatural authority is a modern invention and it is the first misunderstanding worth removing.

Tracing the entries back to their sources

The printed text descends from Jewish and Christian demonological sources, particularly the Pseudepigrapha Book of Enoch, arriving in Europe through the seventeenth-century Conjuratio Ben-Maimon and the Lemuria material, and reaching print in the Lesser Key of Solomon.

The Enochic passages are a list of fallen angels classified by the sins they were accused of. That is the ancestor: a taxonomy of transgression, roughly contemporary with other early Jewish and Christian lists, that acquired names, offices and genealogies as it travelled.

What the classical office actually was

In their own setting these figures were administrative or disciplinary officials. An angel who confined ribald spirits was a prison officer. The list is closer to a bureaucracy of constraint than a collection of torturers.

The transformation to modern horror arrived largely in the twentieth century, and it is worth noticing that it required very little textual change. The names, ranks and descriptions were already there. What was added was an emotional register, supplied by illustration, film and genre fiction rather than by the source.

How to read a single entry

Each entry gives a name, a rank, a family, a seat or domain, a stated power, and the shape of the seal. Read in that order the text is remarkably consistent and almost entirely concerned with administrative capacity.

The powers are domains of knowledge and command. Wisdom, discovery, transformation, invisibility, memory, arithmetic. Very few entries describe anything that would function as an attack on a person, which is a useful corrective to the general impression created by the popular versions of the same list.

1. Name and rank, which indicate a jurisdiction rather than a threat level
2. Family grouping, which usually corresponds to shared accusers or shared functions
3. Seat and domain, the most practically informative fields in any entry
4. Stated power, nearly always a capacity to know, command or transform
5. Seal description, needed for the visual working and nothing else

Step one, choose by domain rather than reputation

The single most common beginner error is choosing by fame. Names that appear constantly in online material are chosen for reasons of visibility, not suitability, and several of the most requested are among the least clearly specified in the text.

Choose instead by reading three or four entries and picking the one whose stated domain matches your situation. This is slower, and it produces a materially different working from the one the recommendation algorithm would have given you. It is also the only selection method with any textual basis at all.

Step two, set the terms of the working

A goetic working in its original framing is a request for assistance within a stated domain, made with a duration, a form of address and a stated limit on what is being asked for. Write those four things down before the working.

Two limits are worth including by default. A time limit, so the working is not open-ended, and a refusal clause, stating that nothing is to be done that would harm a third party or override someone's will. Neither is in the source text. Both make the working easier to justify and much easier to record honestly afterwards.

Step three, construct the seal

The seal is a device in a defined layout, and the traditional method is exacting. Each spirit has a specific shape with specific elements in specific positions, and an improvised one is not the same object as the specified one.

Draw it slowly from a text rather than from an image, and check it against the description before use. This is a two-hour task that people routinely skip by tracing a picture, and the skip is where most disappointing accounts of goetic work originate. If you are going to do this at all, the seal is the part to do properly.

Step four, the sequence of the working

Draw in the circle, place the seal, speak the office of the spirit with the name and the rank, state the request, and state the limits. The order matters more than the wording, and it is roughly the same order in every text that gives one.

Keep the whole thing short. The goetic material has a reputation for drama and the source is dry: a list of instructions for a formal approach to a named official. Readings that take twenty minutes to deliver are a modern addition and they tend to be the part people report forgetting afterwards.

Step five, closing and recording

Close the circle, dispose of or store the seal according to your own discipline, and write the entry the same day. State what was requested, what was the time limit, and what in fact happened, including nothing.

The record is what makes a second working comparable to the first, and goetic practice is unusually well suited to record keeping because the text is specific enough to say whether you did it correctly. Most disappointing results turn out, on inspection, to have been a correctly specified working aimed at a domain that would never have produced the hoped-for outcome.

What actually tends to happen

The most reliable report is a change in the practitioner's own attention to a domain. People who work with a spirit of arithmetic or discovery find themselves noticing arithmetic problems and unsolved questions they had walked past. That is a real effect, it is attributable to the focus, and it is not the same as the entity acting on the world.

The second most common report is nothing, and the third is an event in the practitioner's life that gets attributed to the working afterwards. None of these is a reason to stop, and the first of them is a reason to keep good records, since attention effects are the most reliably produced outcome in the whole practice.

Why the modern brand is a separate thing

What circulates today as Goetia is a layered construction. The seventeenth-century text provides names, ranks and offices. Twentieth-century occultists provide the working methods. Present-day media provides the imagery, the threat register and the commercial category.

Separating those layers is the most useful thing a reader can do with the material, and it changes the practice. You end up working with a formal apparatus that was designed to be used, rather than with a horror aesthetic that was designed to be sold.

• Seventeenth-century layer: names, ranks, offices, stated powers and seal descriptions
• Twentieth-century layer: the working methods, which are largely modern ceremonial practice
• Media layer: the imagery and the threat register, added for genre rather than theology
• Commercial layer: the modern category, where the material is sold as an experience

A defensible first working

Pick a domain you can check, write the four terms, take an hour over the seal, do a short working, and record it. Set the review at three weeks and do not move it.

This will be less dramatic than the versions of the practice circulating online, and it will produce something you can compare against a second working. That comparison is the entire point, and it is unavailable to anyone working from an account that never got written down.

Frequently Asked Questions

Which spirit is associated with wealth, and is it safe to work with?

Clauneck, a Duke, and Mammon, a King, are the two most commonly named in the wealth material, and both are associated in the source with riches, discovery and the removal of obstacles rather than with creating money. The safety question is better answered by attention than by the spirit chosen: a working aimed at your own financial position carries the ordinary risks of financial desperation if it fails, and those risks are larger than anything in the text.

Who are the nine Presidents, and what are they actually for?

Nine Presidents rule kingdoms or hold seats, and the set is dominated by figures associated with command, transformation and the arbitration of disputes: Balam, Bifrons, Alloc, Vine and others. Their practical use in a historical framing is arbitration and authority, which is a different business from what the modern material implies. Choosing among them is a question of stated domain, and the entries say plainly what each one is competent in.

Are these demons in any real sense?

The honest answer is that the question cannot be settled from the sources, because the sources are texts of a religious tradition rather than records of an observation. What can be said is that the modern category of a demon as an adversary is largely a twentieth-century development, and that the original material reads as a list of confined figures with administrative roles. Most people practising with this material are doing so as a formal exercise, and that is a coherent position regardless of how the ontology lands.

What do I get from a paid goetic app that a book does not?

Convenience in three specific places: finding the right entry by domain rather than by memory, holding the seal description at the size you need while drawing, and keeping a dated record of what you asked for. Those are real and a printed grimoire makes all three tedious. What a book gives you that an app does not is the surrounding context, which is where most of the interpretation lives. If you already own one good edition, the app is a convenience purchase rather than a source.

Is it safe to do this alone, and what should I have in place first?

The physical risks are the same as for any candle-based working, and the psychological ones are the ones worth planning for. A written record, a stated end date, and an ordinary commitment somewhere else in your week are the practical preparations, and they are more important than any protective preparation. Most accounts of distress in this area involve a person who had stopped sleeping, had stopped telling anybody what they were doing, and had no review date at all.