The 72 Spirits of Goetia

Most material on the Goetia is arranged as a list to browse: seventy-two names, a rank, a face, a handful of powers. A list is useful for finding a name and useless for conducting anything. What a working session needs is a procedure, and the procedure is easier to build than most of the literature implies. This piece takes the catalogue as it stands in the Ars Goetia and turns it into a repeatable protocol: read the stated office before the reputation, choose by domain rather than by fame, run six stages with a fixed shape, and write down what happened while it is still ambiguous. The claim underneath is that structure, not intensity, is what makes a session repeatable.

Read the Office Before the Reputation

Rank, visible form, office and a set of powers are what every Ars Goetia entry supplies, and the office is the field that carries operational information. Reputation is what a century of video essays added on top, and reputation rarely tracks the office. Some of the most sought-after names have narrow, technical jurisdictions. Several much-discussed entries are administrative figures whose stated business is entirely procedural.

The practical consequence is that you start from the office column and work outward. If the task is about a legal or institutional obstacle, find the spirits whose office concerns courts, contracts, offices and obstructions. If the task concerns loss, find those whose office concerns discovery and recovery. The match will occasionally be a name nobody has ever mentioned, which is exactly the value of doing it this way.

Skipping that step is the commonest error in the whole body of material, and it produces sessions that are theatrically dramatic and pointlessly aimed at a figure whose stated business has nothing to do with the situation in front of you.

Ranks Are Instructions, Not Hierarchies of Power

Three ranks structure the catalogue: nine Presidents, thirty-six Dukes and thirty Kings. Those labels are not a statement about how dangerous each spirit is. They are a statement about what kind of authority the entry is asking you to borrow and for how long.

The Presidents are granted wide authority over large domains, and working with one is conventionally framed as requiring a more demanding preparation and a fuller receipt. The Dukes and Kings are more numerous, more specific and considerably easier to work with. Most of the seventy-two sit in those lower two ranks, and the majority of serious practitioners in the modern literature work almost entirely with them.

Treat rank as a parameter rather than a ladder. Working with a King does not mean you have taken a step down from a President. It means you have borrowed a narrower instrument, and for most practical purposes the narrower instrument is the one that fits the job.

The Working Sheet

Before any session, one page, five fields. The office you are working with, copied from the source rather than from a summary. The situation, described in one paragraph in your own words with the actual stakes named. The window, meaning the date and hour you have chosen and why. The method, which is the sigil, the name, the frame. And the outcome line, left blank until the date has passed.

The blank outcome line is the part that does the work. Writing it before the session, where you can see it empty, produces a modest and specific intention. Filling it in afterwards, from memory, produces whatever the session seemed to be about, which is usually what you wanted it to be about.

A sheet per session, kept in one file, is also the only mechanism that will tell you which methods actually produce trackable results for you. Three sessions in and the file already shows more than any amount of further reading would.

Choosing by Domain and Not by Popularity

Write the situation down first, then read domains until one of them maps onto the actual shape of the problem. A question about money is not a single domain. It might be a question about a contract, a question about reputation that unlocks money, a question about a specific person, or a question about finding something lost that happens to be money. Each of those lands on different entries.

This is why the sessions that produce anything tend to begin with a narrow problem statement and a small ask. A large, vague intention has no corresponding office, so there is nothing to select, and selection is what makes the frame hold.

The nine Presidents are named often enough to seem like the obvious starting point. They are not, unless the situation genuinely spans a large domain such as a legal proceeding, a public reputation, or a dispute involving multiple parties.

Six Stages, Fixed Shape

Every session in this protocol runs the same six stages in the same order. Preparation, in which you write the sheet, draw the sigil, and set the frame in a single sentence. Approach, in which you settle attention and hold the intention as a question rather than a demand. Invocation, the shortest stage and the one most often extended, where the name is spoken and the sigil is presented. Observation, in which you note what changes, including nothing. Withdrawal, which is a formal closing rather than an abandonment. Recording, completed the same day while memory is fresh.

The shape matters more than the content. Because the stages are fixed, the variable is the working itself, and any difference in the outcome can be attributed to the variable rather than to a difference in method. That is the whole reason for writing it down.

Keep observation as a distinct stage even when there is nothing to report. A session with no distinguishable effect is data, and treating it as a null result rather than a failure is what allows the file to stay honest.

The Kings and Dukes Carry Most of the Work

Of the seventy-two, twenty-seven are Kings and thirty-six are Dukes, and it is in these fifty-three that the specific, learnable offices live. A duke whose office concerns discovery, for instance, is doing something structurally different from a king whose office concerns rank and confirmation, and the difference will show up in how the working has to be framed.

Money work is where the office column pays off, because the catalogue separates several kinds of financial problem that get merged in conversation. One figure is named as the spirit of wealth outright. A duke is recorded as granting riches, which is a slightly different office and occasionally the better match. A king is recorded as granting dignities and confirmations, meaning that money here is a side effect of standing rather than the objective itself.

The three get conflated constantly, and conflating them costs you the match. A request for money when what is actually wanted is recognition has been sent to the wrong office, and the response will be proportionate to the wrong request. Take the stated office from the source text before the session rather than from a modern paraphrase, because the paraphrase is what flattened the distinction in the first place.

Family Groupings and Why They Matter

Beneath the ranking the spirits sort into twos, threes and fours, and those groupings earn their place. Working with two members of a family in sequence, or with a family as a set, produces a more coherent result than invoking unrelated figures in the hope that something lands.

The logic is straightforward once seen. A family represents a related cluster of offices, so a problem that touches several aspects of the same situation can be handled by figures whose powers are already complementary. Selecting across families for the sake of variety does the opposite, since the authorities being borrowed have nothing to do with one another.

A practical rule that has held up better than most: pick one family, read all three entries, choose the one or two whose offices fit, and leave the rest.

Timing the Window

Choose the window from the situation rather than from the calendar of correspondences, and write the reasoning into the sheet. Dates carry associations that can be used, but a window chosen for a symbolic reason is a session about symbolism, and it should be labelled as one.

For work that depends on an institution, a court, a landlord, an employer or a public body, the working day matters more than the planetary hour. For work that concerns a person or a pattern of behaviour, a private hour with no audience does more than any public timestamp.

If the timing is chosen purely by a table, say so in the sheet. Mixed motives produce muddy records, and the point of the sheet is to make the record legible later.

What to Do When Nothing Distinguishes Itself

Some sessions produce no distinguishable effect at all, and this happens more often than the literature suggests. The correct response is to log it and adjust the frame, not to increase the intensity.

Intensity is the default move and it is almost always wrong. Escalating a working that produced no signal tends to produce more of the same, at higher cost, and it is the mechanism behind the tradition's own advice that a short daily practice beats an escalating one. A shorter working with a narrower ask will often produce a result where a larger one produced nothing.

Escalation is worth doing in exactly one case, when the intention was genuinely unclear. Then the fix is to sharpen the intention rather than to add candles, hours or repetitions.

Recording While It Is Still Ambiguous

The same day, before interpretation has had time to tidy itself. Write what happened in order, then write what you think it meant, then write what you would predict if the frame were correct. The three are separate entries and they should not be allowed to merge, because merging them is how a log turns into a collection of confirmations.

Over a dozen sessions the file becomes more valuable than any book on the subject, because it is the only record that reflects your own pattern of response. Some people find that particular figures produce reliable results and others do not. That is discoverable in a log and invisible in a catalogue.

Revise the sheet after a month. Note which parts of the protocol you actually followed and which you skipped, since the skipped parts are where the unexplained variance lives.

Withdrawal and Closing

Closing is a stage, not an afterthought, and the reason is structural. A working that ends without a formal close tends to be re-run mentally later, which amounts to a second, unrecorded session with none of the six stages. That is a common source of drifting contact that practitioners experience as ambient and cannot account for.

The close is short: state that the working is complete, thank the office, dismiss the frame, and then do something else on purpose. The last part matters more than it sounds. Attention that is deliberately redirected does not reattach on its own, and this is why a session followed immediately by bed often behaves differently from the same session followed by an hour of scrolling.

If a session has gone badly, close it anyway. An unfinished working is the thing worth avoiding here.

Where This Protocol Does Not Help

This structure is fitted to problems with an identifiable shape: a decision, an obstacle, a person, a document, a habit. It is not fitted to grief, to illness, to bereavement or to anything where the person involved is not the intended beneficiary. In those cases the appropriate response is care, and occasionally professional help, and no arrangement of borrowed authority changes that.

It is also not fitted to the question of whether any of this is real. The protocol is designed to produce trackable results in your own record, and it is deliberately silent on the metaphysical claim. Someone who wants the claim settled should not come here for it; the file either accumulates differences or it does not.

And it is not fitted to a desire to be impressive. The sessions this protocol produces are not dramatic. They are small, well documented and mostly dull, which is the condition under which the method stays honest.

Frequently Asked Questions

What are the 9 Presidents of Goetia?

Taken in printed order, the Presidents run Bael, Agares, Vassago, Samigina, then Marbas, Valefor, Aamon, Belphegor and Saleos. Each carries a rank, a described visible form and a stated area of power in the source text, and the offices are broad rather than narrow, which is why they are reserved for work that genuinely spans a large domain. A President is not a stronger spirit than a King in any sense that matters for your purposes; the rank describes the breadth of the authority being borrowed and the preparation the tradition asks for.

Is the word demon doing any real work here?

The historical record offers catalogue entries and nothing more; not one of them shows any sign of agency beyond the printed page. That does not make them useless. As a symbolic vocabulary they are unusually well documented, which is exactly what a working needs: a figure with a defined office, a defined form and a defined set of powers gives attention somewhere specific to go, and most people get further with a good figure than with an unbounded intention.

Which demon gives money?

Three figures cover financial ground and they do not cover it in the same way. Mammon holds the office of wealth outright, so it is the direct match for a straightforward request. Clauneck is a duke described as granting riches, which is a broader phrasing and sometimes the better fit. Paimon grants dignities and confirmations of favour, so it belongs to cases where money is the route to recognition rather than the aim. Whichever you pick, take the wording from the source text and not from a summary, because the summaries tend to describe all three as money spirits and lose the distinction that made the choice worth making.

How long should a first Goetia working take?

Thirty minutes of actual session time, with an hour of preparation and twenty minutes of recording around it. The tradition's own advice points the same way, since a short daily practice is considered healthier than an escalating one. Longer sessions are not forbidden, but they are the standard route to the ambient drift that practitioners struggle to explain, because attention that stays open after an unstructured close tends to reopen on its own later.

What if nothing happens during a session?

Log it as a null result and leave the intensity alone. Then sharpen the intention, since an unclear ask is the most common cause, and run a shorter second working with a narrower target. Escalating a session that produced no signal usually produces no signal at higher cost, and a log that records the nulls honestly is worth considerably more than one that only keeps the successes.