ABRAXAS

Four claims do most of the work in the belief and paradigm material, and they are of four different kinds. One is a licence. One is a warning about your own cognition. One is a small real effect. One is a method for keeping yourself honest. Treating them as a single doctrine is the reason the area feels confused, and treating any one of them as the whole thing is why the area feels like it fails. A table with a row per claim, a column for what each does not license, and a column for what would change your mind is enough to make the whole thing usable.

Why a table rather than a position

The belief material is usually presented as a position you hold, and holding it is treated as the practice. That is the problem. A position can be held indefinitely without ever being examined, and the four claims in this area are of such different kinds that holding them as one thing guarantees confusion about which one is doing the work in any given moment.

A table is more useful than a position because it forces the separation. Each claim gets its own row. Each row has a column for what it supports, a column for what it cannot be used for, and a column for the observation that would change your mind. The last column is the one that matters, because it converts the whole area from something you believe into something you are checking.

The parabola, in one row

The parabola is the early theoretical move that gave this material its shape. It says magic has no reality of its own, and serves whatever end it is pointed at. It was a tool for suspending judgement, and it was genuinely useful for that, because it let practitioners work with a symbol without having to first settle whether the symbol corresponded to anything.

What it licenses is a great deal of practical freedom. What it does not license is the conclusion that therefore all aims are equivalent, which is a very common and very flat misreading. If the parabola is right then the mechanism is not doing the work, which means the aim, the attention, and the follow-through are the whole of it, and that makes the quality of the aim more important rather than less.

The confirmation engine, in one row

Confirmation bias is among the best-established findings in psychology, and this area is a particularly efficient machine for producing it. Any practice that rewards noticing successes and does not require counting failures will appear effective regardless of mechanism. The mechanism is irrelevant to the appearance, which is what makes the appearance so convincing.

The row therefore has a single job, which is to be a warning rather than a claim. It says that your impression of whether a practice works is untrustworthy, and it does not say the practice does not work. Those are different statements and collapsing them produces a scepticism that is as badly calibrated as the credulity it is correcting. What would change your mind about this row is nothing, because it is a description of a general process rather than a claim about you.

Expectancy, in one row

Expectancy effects have real measured influence on reported pain and on a range of physiological outcomes. This is a genuine and limited effect, and it is documented in ordinary medicine by people with ordinary equipment. It is worth taking seriously, and it is worth being clear about the ceiling, because the ceiling is where a lot of trouble comes from.

The ceiling is that expectancy is a small effect on outcomes that are themselves sensitive to expectation. It moves reported pain. It moves some physiological measures. It does not cure disease, and the version of the claim that says otherwise is a different claim, made elsewhere, by people who are not making this one. This row supports modest conclusions about modest effects and refuses the large ones, and the refusal is the useful part.

The updating rule, in one row

The fourth element is a method rather than a belief. Record what happens, and when the record disagrees with your expectation, update toward the record rather than defending the expectation. This is straightforwardly a Bayesian approach and it is entirely compatible with a person who is sincerely committed to an occult practice.

The reason it belongs in the table is that it is the only row that is falsifiable by your own conduct. The other three describe something, and you can be wrong about the description. This one asks you to do something, and you can visibly fail to do it. That makes it the load-bearing element in practice, and it is also the one most often abandoned, because the update is uncomfortable and the defensible version is comfortable.

What each row does not license

Row one does not license aimless practice. Row two does not license ignoring outcomes, since it is a warning about them rather than a dismissal. Row three does not license expecting a physical cure. Row four does not license the version of record-keeping that only confirms, which is the failure mode of the row that is supposed to prevent exactly that.

Almost all of the incoherence in this material comes from borrowing a licence from one row and a claim from another. The parabola is used to justify a practice, the confirmation row is ignored while the practice is evaluated by impression, the expectancy row is inflated to cover a claim it cannot support, and the updating row is quietly replaced by an anecdote. Separating the rows is the whole discipline.

The column that decides the case

The deciding column is the one headed by the question: what observation would change your mind. Fill it in for every row and the table changes character. A row with an empty cell is a belief. A row with a specific observation in it is a claim you are testing, and the difference is not semantic, because the filled cell gives you something to go and look for.

This is the single change that makes the difference between belief as tool and belief as commitment. With a filled cell, the practice generates information, and the information can go against you, and a practice that can generate information against itself is a practice that can teach you something. With empty cells, the practice generates confirmations, and confirmations teach nothing regardless of how sincere the person holding them is.

Building the table for a claim of your own

Take any aim of yours and put it through the same four rows. What kind of claim is it? What cognitive mechanism would make it look effective if it were not? What measurable outcome is expectancy likely to touch here and not there? And what is the updating rule, written before the aim starts, including what counts as evidence against.

It takes twenty minutes and it will usually deflate the aim slightly, which is the correct outcome. Aims that survive this exercise are narrower, more specific, and more checkable than the versions people usually carry around. The exercise is not designed to kill enthusiasm. It is designed to convert enthusiasm into a procedure, and a narrow checkable aim is far more likely to produce something than a broad one that cannot.

Where the table has an empty cell

There is a row this table cannot fill, and pretending otherwise is the failure the table exists to prevent. Nothing in these four rows supports the claim that the practice affects anything outside the practitioner. All four describe internal processes, and when people write the fifth row they are writing something none of the material reaches.

The empty cell is not a defeat. It is the boundary of the available evidence, and a table with one clearly empty cell is more honest and more useful than a table with five full ones, because the full ones are false. The correct response is to run the practice for its internal effects, to keep the record honestly, and to let the external claim stay unfunded until something turns up that fills the cell.

Using the table against your own aim

Now point it at something you are already doing, and the exercise turns unpleasant in a useful way. Write the aim as a sentence someone could check. Then write the sentence you would use if the aim had failed. Those two sentences are rarely the same, and the difference is the entire problem, because the second one is what a confirmation engine generates.

If the failure sentence is impossible to write, the aim cannot fail, and a practice that cannot fail cannot teach you anything. The most common example in this area is an aim stated as an improvement in some general quality, which can absorb any outcome at all. Rewrite it as an observable behaviour with a date attached. The rewrite takes five minutes and it is the difference between a working that teaches and a working that flatters.

Where the table stops being useful

It cannot settle whether any of this is true about the world, and it is worth being blunt that no version of it can. It can sort your claims by kind, tell you which ones are load-bearing, identify the boundary of the available evidence, and give you a procedure that will detect your own bias. That is a lot, and it is the lot that is available.

It also cannot make the practice worth doing for everyone. Some people find the discipline cold and find better traction in a practice held more loosely, and the table is not an argument against them. It is an argument against using a table-shaped practice as if it were a belief system, because the failure mode there is specific and it is the most common serious error in this whole area.

The card worth carrying

One page, four rows, three columns. Parabola as licence, with aimlessness excluded. Confirmation as a warning about your own impressions, with no claim that the practice fails. Expectancy as a small real effect with a hard ceiling. And the updating rule as the only falsifiable element, with the failure sentence written before the aim starts. Plus the empty cell for anything external.

That is the whole of the usable material and it fits on a card. The rest of what circulates is either elaboration of one of the four rows or a claim that belongs in the cell you have to leave empty. Keeping the card visible is more useful than any amount of further reading, because the problem in this area is not that people lack information. It is that they lack a way of sorting what they already believe.

Frequently Asked Questions

Does belief in magic make it work?

Belief is not established as a cause, and the open-label placebo research complicates the simple version of the claim by showing that measurable physiological effects can occur without belief in the treatment. What expectancy reliably touches is reported outcomes and some physiological measures, which is a real and limited effect. The rest is a separate question and the four claims in this area do not settle it.

Is the parabola just a way of avoiding belief?

It was designed to suspend judgement, and suspending judgement is a different move from avoiding evidence. The useful version lets you work with a symbol without settling a metaphysical question first, which frees attention for the things that can be checked. The failure version is a general permission that never has to face a result, and the difference between the two shows up in whether you wrote down what would have counted against you.

How do I avoid fooling myself with a record?

Decide what would count as failure before the aim starts, and write that sentence down. Then keep the record open before the first attempt rather than after, so you are not writing as someone who already knows how it went. The test of a useful record is whether it contains at least one entry you would rather not have written. A record with no such entries is not neutral, it is a list of confirmations.

Is it possible to be sincere about an occult practice and still be objective?

Yes, and the table is designed for that person. The method is to write down in advance what observation would change your mind, and to let the record outrank your preference. That is an ordinary methodological commitment and it has no requirement that you abandon your commitment, only that you stop defending the record when it is inconvenient. Most of the incoherence in this area comes from people who sincerely want both things and never notice the contradiction.

What if my record shows the practice did nothing?

Then the row is the most useful thing you have produced, because it is the only one of the four that can come out negative. Write down what you would do differently, and note whether the aim was too broad to fail, too small to notice, or simply wrong. A practice that generates negative results is functioning as a practice. The alternative is a record that only ever contains successes, and that record cannot tell you anything at all.