Three Real Goetic Workings, With the Test That Keeps Them Honest
Most published workings are real in the sense that somebody performed them and something changed. That is a low bar. The three below clear a higher one: each is built on a specific clause of the Ars Goetia rather than a mood, each carries a number you can write down, each names the outcome that would have made you stop, and each names in advance the ordinary explanation that could imitate success. One of the three, the Furfur material, is included precisely because it is the one most likely to be fooling you.
Why "real" is the hard part of a Goetic working
There are three separate claims hiding inside the word real, and most published workings collapse them into one. The first claim is that the request was well formed: you asked a coherent question of something that plausibly answers questions. The second is that the method was sound: the actions taken were the ones a competent practitioner would take, whether or not a spirit was involved. The third is that the outcome was attributable: what happened afterwards happened because of what you did.
The first two claims are cheap. Almost every working you will read in a grimoire clears both. The method is usually a competent method, because the authors were competent magicians, and the petition is usually well formed, because a badly formed petition is obvious to the people writing it. What you rarely find is the third claim, stated in a form that could turn out to be false.
That is the whole difference between a working and a wish. A wish has a request. A working has a request, a method, a measurement, and a stopping rule written before the first session. Remove the last two and you have not removed the magic, but you have removed everything that would let you find out whether there was any.
The three workings below are built that way on purpose. Two of them will probably produce an effect that has nothing supernatural in it. One of them will probably do something, and the design is set up so that you can tell which.
Purson, the treasurer who arrives silent
Purson appears in modern editions of the Ars Goetia as number forty, and in the 1604 Oxford manuscript as number forty-five. As with the rest of the catalogue, position tells you almost nothing; the description is the entry. His office is unusual in one respect that matters here: he is a treasurer, and the description is written around accounting.
He is a king of the east, a president, called to command thirty legions. He teaches arts and sciences, and the profits of all manner of merchandise come to him. He shows up crowned, in the shape of an angel, holding the bills of everything he has gained. He is of great fame, and he is patient. He speaks only when asked, and what he says is worth listening to.
The clause that the working uses is the last one. He does not perform. He holds the bills, he waits to be asked, and he is not chatty. A method built on that entry should not be a method where you talk to him at length, and it should not be one where you name yourself out loud more than once.
The working is this. Pick a real commercial situation in the next three months: a fee you are undercharging for, a quote you are about to submit, a debt you are owed, a supplier whose terms you want improved. Something with a number attached to it that currently goes the wrong way. Build a sigil for the outcome. Do the session in silence, or as close to it as your life allows. Write down three figures before the session, one number you need for it to be worth doing, and three figures after.
Silence is the part practitioners tend to drop, because it is the part that is uncomfortable. It is also the entire reason to work with Purson rather than with a more conversational entry. If you want a spirit that is good at arriving and bad at small talk, the method should have the same shape.
The number that has to move before you will believe anything
Here is the false positive for the Purson working, stated before you run it. You are about to go into a negotiation where somebody has a constraint you have not figured out. The working will help you work out the constraint, because you will be paying attention for the first time in weeks, because you wrote the number down, and because you have a sigil and a session behind you. You will then read that as the spirit's doing.
The measurement that separates the two is not the first session. It is the fourth one. Run the working four times, on four different counterparties, in the same industry if you can manage it, so that the social learning is not the variable. Keep the sigil. Change the person.
If the win rate climbs across the four, something is happening that is not just you getting better at reading people. If it stays flat and you simply win more than you used to, the honest conclusion is that the sigil was decoration and the practice of writing the number down was the working. That conclusion is worth having. Most ritual literature will never let you have it, because most ritual literature has no way to state it.
The second measurement is your own preparation. Before each session, write the number you need and one sentence on what you will do if you get it. If you cannot write the sentence, the number is not a goal, it is a mood, and a mood will move for all the wrong reasons.
Crocell, the working that puts your mouth to work
Crocell is number forty-eight in modern editions. He is a duke who comes in the habit and order of a Franciscan, and the description spends most of its words on labour. He is a hard worker, he teaches the arts and sciences, and he can make the learned flow from the lips of those he teaches. He is honest, and he will not command things unless he is asked to.
Read the clause again and notice where the teaching sits. The learned flows from the lips of those he teaches. The teaching is inside the promise. The learning is not. If you want a fluency working built on this entry, the structure of the method has to include the teaching, because that is what the entry says the spirit is for.
The working: pick a subject you need to speak in fluently, and something you will actually have to say in it within a fortnight. A meeting, an explanation to someone who depends on you, a subject on a form. Build a small sigil. Then teach it out loud, to a real person who has to understand it, four times in a fortnight, on four different days. Say the sigil before the first one, and not again.
The redundancy is the point of the design. You have asked for fluency and you have also produced the conditions under which fluency normally arrives, which is being asked to explain something to somebody who needs it. Those two things are separable, and the whole experiment is about separating them.
The false positive here is the easiest one to predict and the easiest to fall for. Speaking a subject out loud, repeatedly, to someone who needs it, works. It works because the mechanism is well documented and has nothing to do with spirits. Whatever you do, do not let that outcome get filed under the sigil.
A fortnight of teaching without the sigil, a fortnight with the sigil and no teaching
Crocell is the working where you can run the comparison honestly, so it is worth running it rather than assuming.
The design is two periods of a fortnight. In the first, no sigil, no session, nothing added. Teach the subject four times, on four days, to a person who has to understand it. In the second period, the sigil and the session at the start, and no teaching at all. Then measure the thing that is actually supposed to have changed, which is not whether you know the material. It is how long it takes you to explain it, whether you have to look things up mid-sentence, and how often the person asks you to repeat yourself.
Those three measures are the ones a person outside your head can score. Ask the person at the end of each session, once, on a scale of one to five: how clear was that, how much did you have to repeat yourself. Do not ask after every repetition, because the asking changes the teaching and you are trying to measure the teaching.
If the sigil period shows a real difference on those numbers, you have something worth investigating, and the next step is to vary the sigil while keeping the no-teaching structure, which tells you whether the sigil is doing anything or whether the expectation of a session is.
If both periods are the same, you have learned something useful anyway. You have a subject you can now speak, a method for learning it that takes a fortnight, and a working that costs nothing and asks no spirit for anything. That is a better outcome than a sigil you are still trying to interpret.
Furfur, the working you are most likely to get wrong
Furfur is number twenty-six, a duke, and the entry is unusual in how much of it is a warning. He appears in the form of a white cat. Or in the form of a man with the head, eyes, and hands of a cat, the rest human, and he speaks in a hoarse voice. He is the deceiver, and he is too fair of face. He is good at causing love between a man and a woman, and he will appear in the woman's form, and take the woman's clothes. He is very liberal with lies and much given to boasting.
Two clauses decide how you work with this entry. He is given to lies and boasting, which is a description of a mode of operation rather than a moral judgment, and it should be taken as operational information. And the third time he appears, he will hurt those who hear him. That is not a figure of speech for poor manners, and there is no reading of it that makes it one.
The working, then, is about words. Pick a conversation that genuinely has to happen and that you are dreading, where the words are the actual mechanism rather than the wrapper around one. Write them, in order, the way you will say them. Say them. Nothing more.
Then write your refusal in advance: the sentence you will use if the other person escalates, if you are offered something you did not agree to, or if you feel the pull to keep going past the end of what you prepared. Most of what goes wrong with this working is not the spirit. It is you, on a good run, deciding that the thing is working and extending it past the point you had agreed with yourself.
The three things they share that most Goetic material leaves out
Each of these is built on a clause you can point at, not on an impression. Purson is silent and holds the bills, so the working is silent and produces a bill. Crocell teaches and makes the learned flow from the lips of those he teaches, so the working involves the teaching. Furfur lies and boasts and harms the third time, so the working is short, in daylight, with a prepared exit. If you cannot name the clause your method came from, you did not have a method, you had a habit that happened to work sometimes.
Each has a purpose narrow enough to be wrong. The Purson working can come back with the wrong number. The Crocell working can come back with a subject you already spoke fine. The Furfur working can come back with a conversation that was never going to go well. Narrow purposes are what make it possible to be wrong.
Each has a stopping rule, written before the first session. That is the part almost every piece of writing on this subject omits, and it is the part that decides whether you are practising or collecting evidence about yourself. A practice with no stopping rule does not fail, it just continues.
And each has its ordinary explanation named in advance. You get better at reading people. You learn a subject by teaching it. You rehearse a conversation and it lands. None of those are shameful outcomes. They are simply outcomes, and they belong in the record next to the rest of it.
The stop conditions, written down before you start
For Purson: two consecutive sessions in which the number does not move, and you have written down what the counterparty said. For Crocell: one fortnight in which the three measures do not change, measured by the person listening. For Furfur: one conversation in which you say something other than what you wrote down and cannot account for the difference.
Set a review date as well. Choose it now, write it down, and do not move it later. The interval that works is the one you will not renegotiate when the result is disappointing, which means the interval has to be short enough that moving it would be obvious.
Then there is the override, which applies to all three and takes precedence over results. If the practice has started producing arguments at home, sleep you are not recovering, or spending you did not intend, stop and re-ground. This is not a moral failure and it is not a sign that the working was strong. It is a sign that the working has become the most important thing in your week, and anything that does that will eventually cost more than it returns.
The override is the only part of this that most readers will not do, and it is the part that keeps a practice from becoming a problem. Write it now while you are reading, not later.
How to tell a real working from a well-told one
Four questions separate the two, and you can ask all four of any ritual in any book you own.
What clause is it built on? If the answer is a feeling, an atmosphere, or the practitioner is favoured, there is no clause. There is a mood with a costume on.
What number? Any working should produce a quantity or a count or a yes or no that somebody else could check. A working that produces a feeling has no measurement, and a working with no measurement cannot be wrong.
What was the pre-committed rule? Every honest working was written down before the result arrived. If the rule only exists in the chapter where the result is good, the rule is not a rule.
What result would have made you stop? This is the sharpest one. If the answer is that nothing would have, then the working is a practice with an open ending, and an open ending is indistinguishable from not having looked.
A working that fails the first two is folklore. One that fails the third or fourth is folklore wearing a lab coat. That distinction is worth keeping, because the first kind is charming and the second kind wastes months.
Where each of the three sits against the rest of a practice
Purson belongs next to sigil craft and next to bookkeeping. It has a number in it, and if your practice has no place where a number appears, this working will feel strange. That is not a warning sign, it is just the shape of the entry.
Crocell belongs next to an active study practice and specifically not next to a solitary one. The clause requires teaching, so a practice that is entirely alone cannot host it well. If you are working mostly in isolation, this is the working to leave for later.
Furfur is the one that should not run late, should not run repeatedly in a short period, and should not run alongside an active dream practice. If your work involves lucid dreaming, astral work, or anything that trains you to hold an image in mind while you sleep, then a working whose mechanism is rehearsed words is a bad neighbour. Do not run a make them say what I want entity next to a practice that trains you to open your own mind on purpose.
That is not a doctrinal objection. It is a scheduling one. The two practices interfere with each other, and you would have no way of telling which of them was the problem.
What to build first, if you are starting from nothing
Start with Crocell. It has the lowest falsehood rate of the three, because the method that does the work, teaching out loud, is free and you already know it works. You are not gambling on the sigil. You are measuring whether the sigil adds anything to a method you could run anyway. That is the right first experiment, because the correct answer, nothing added, is still a good outcome.
Then Purson, if you have a real commercial situation inside the next three months. The four-counterparty design only makes sense when you can actually get four sessions, and you cannot if there is nothing to negotiate. Without a real number, the Purson working is decoration on a ritual.
Furfur last, and only with the written exit, and only for a conversation that has to happen. There is no version of this working that is a warm-up, and a person running it as a warm-up is the most likely person to be hurt by the third time.
The ethics of asking for a number and for boldness
The Purson working is for money you are owed, money you are undercharging for, or terms you have no reason to be embarrassed to ask for. It is not for extracting a figure from someone who has already said no, and the difference is not a matter of degree. If the other person has declined, the working has left its territory and the outcome you are asking for is not a bargain, it is pressure with a sigil on it.
The Furfur working is the harder one to keep clean, because rehearsing words until they land is a thing you do to a person rather than to a spirit. The person on the other end has no sigil and no part in the method, and they are the one who gets the consequences. Ask yourself what the outcome is that you could not ask for directly, and if the answer is nothing, the working is not needed.
And the arithmetic one, which is the least mystical. Money spent on sigil supplies, on course fees, on candles, on apps for this is a real amount by the end of a year. If you are spending that money while the numbers you came for are not moving, you are funding a practice, not running a working, and the honest thing is to price the practice as a practice and decide whether you want it.
None of this makes the workings illegitimate. It makes them small, and it is better that they are small. A working for a fee you have earned, a conversation you were avoiding, and a subject you needed to learn are all things a person can want honestly.
Frequently Asked Questions
Do these three workings work without the sigil?
In two of the three, most of what happens will happen without it. The Crocell design is built around the assumption that teaching out loud does the work, which is why the control period exists. The Purson design assumes you will get better at reading people once you are paying attention. Only the Furfur working has a mechanism, rehearsed words, that is genuinely yours. The experiments are set up to find this out rather than to assume otherwise.
Which of the three is the strongest?
Strength is not the useful question, because the three do different things and the only common measure would be how much you believe in them. The useful question is which is most falsifiable. Crocell is, because you can run a fortnight with the sigil and no teaching and compare it against a fortnight with neither, using a score somebody else gives you. If you want to know whether sigils do anything at all, that is the one to run first.
Is the Furfur working dangerous?
The Ars Goetia entry says that the third time he appears he will hurt those who hear him, and there is no reading of that as a figure of speech that I find convincing. Practically, the risk is ordinary and behavioural rather than supernatural: rehearsing a conversation until it works, then extending it past what you agreed with yourself, is how people damage relationships over money. Run it in daylight, once, with the written exit in your pocket, and do not run it repeatedly. Treat the entry's warning as a scheduling instruction.
Can I run all three at once?
No. Pick one, run it to the stopping rule you wrote in advance, and then pick another. Running three at once means you have three stopping rules and no way of telling which one fired, and it also means the practice has taken over the shape of your week before you know whether any of it works. Two of the three need a specific real situation to exist, and one situation at a time is usually all you have.
How long should I give a working before I decide?
Write the review date before the first session and do not move it. For the Purson working the natural unit is four sessions, because that is the number of counterparties the design needs. For Crocell it is two fortnights, one per period, because a shorter period cannot separate the conditions. For Furfur it is one conversation, because there is no second one worth having. If the result is disappointing and you want to extend, that is usually the answer rather than the data.