Music as Magickal Technology

I set up a musical working in the belief that a particular track would carry it, and I ran it for eight months badly. The failure was not dramatic. Nobody broke anything, nothing went wrong that I could point at afterwards, and the practice simply stopped doing whatever it had been doing. What follows is an account of where the effect went, why it went there, and what I ended up with instead, which is less interesting than what I started with and considerably more useful. The short version is that the track I loved was the least effective carrier I owned, and the track I found tedious was the one that held the practice together for months. I did not expect that, I argued against it twice, and the field notes are unambiguous about where the argument went. What follows is not an argument for dull music. It is a description of a variable I had been treating as decoration, plus the three substitutions that actually moved the result.

The working, and what it was supposed to do

The aim was a state change, not an outcome. I wanted to be able to sit down, run a short sequence, and arrive at a different place than I left, reliably enough that I could build a practice on top of it. That is a modest ask and it is the ask that music is actually good for, which the anthropological record supports: musical ritual is a structural component across most traditions that have been examined, and the physiological coupling to tempo is one of the more dependable non-pharmacological state-shifters available.

I had a track in mind for this. A specific piece, about eleven minutes, which I had used for years for unrelated purposes and which reliably produced a particular feeling within about ninety seconds of starting. That was the theory. Familiarity plus a reliable response plus a ritual frame equals a carrier. The theory is not unreasonable. It is just wrong in a specific way, and the way it is wrong took me most of a year to accept.

The month the soundtrack turned into wallpaper

By the second month I noticed that I was humming the piece during the rest of the day, and that the piece had stopped producing its effect by the third repetition inside the working itself. The reliable ninety-second response had become a reliable ninety seconds of nothing in particular, because a stimulus that has been paired with a context many times stops producing the state it used to produce. It was still pleasant. It was no longer doing work.

This is the part of musical ritual that the anthropology documents and that people using music ceremonially are usually not thinking about. The music is a carrier, and a carrier that has been used too recently is not carrying. The mechanism is uncontroversial and the effect is entirely predictable once you have thought about it, which is the problem: nobody thinks about it, because the music is the part of the practice that feels like atmosphere rather than technique.

The track that did nothing at all

For two months of the first year I used a binaural track, generated at a specific frequency difference, on the theory that the entrainment would carry me. The entrainment claim has real support: neural oscillation does tend to align with external rhythm, and that is a reasonably studied phenomenon. The behavioural consequences are modest, and the specific claims made about binaural frequencies and cognition have not replicated well in controlled work.

My own record matched the literature exactly, which is the least satisfying possible outcome. Nothing dramatic. No state change I could not get from sitting quietly. No harm either. Just eleven minutes of a sound that was technically doing something at the level of physiology and functionally doing nothing at the level of practice. I kept it for two months out of momentum rather than because it was working, which is itself a data point about how easily a ritual component becomes inert while everyone assumes it is still doing its job.

Tempo, and the part that actually coupled

What survived scrutiny was the crudest variable: beat rate. When the piece I was using sat in a moderate range and had a clear pulse, the state change came reliably within about three minutes. When the same sequence was run over material with no discernible beat, it took longer and was less certain. That is the whole finding, and it is smaller than I wanted.

The interesting part is what this implies about attention. A clear beat is a scaffolding for a monotonous task, and the task here was staying in one place for a defined period. The music was not producing a trance state. It was removing the decision about how fast to go, which is a decision I make badly when I am trying to concentrate. Anything with a pulse will do. Almost anything at a moderate tempo, played continuously, will do. The elaborate apparatus was solving a problem I did not have.

Keeping the same piece for longer than felt necessary

I changed the track roughly every two weeks for the first half year, on the theory that variety kept things fresh. The field notes say the opposite. The runs that produced a reliable state change were the runs with an unchanged piece for a month or more, including the runs where the piece had gone stale. Freshness was not doing anything. Continuity was.

This took a while to believe because it contradicts the wallpapering finding, and both findings are true. The piece stops producing its own effect within a session or two, which is what wallpapering means. And the same piece, kept for a month, produces a more reliable working, because it has become a fixed cue rather than a stimulus. Those are different mechanisms and they pull in opposite directions. The resolution is that the piece is a cue, not a drug, and a cue works by being boring enough to be predictable.

Volume, and why loud is not the variable

I spent a period of the year trying to run the practice at various volumes, treating loudness as a lever. The low end did nothing. The very high end produced a state I would describe as activated rather than settled, which is the wrong direction entirely. The useful range turned out to be a level low enough that I could still hear my own voice reciting over it without strain.

The masking function is what matters here, and it is better documented than the entrainment claims. The anthropological record describes music in ritual doing a job that speech cannot do, which is covering a voice that must be heard without being understood, whether that voice is a set of instructions or a person trying to be heard from another room. That job is done at moderate volume and destroyed at high volume, because at high volume you are aware of the music and at moderate volume you are aware of the recitation.

The headphones, and the honest verdict

For the binaural work, headphones are not optional, since the effect depends on each ear receiving a slightly different frequency. I used them, and I got the effect I expected, which was that I could hear the two tones separately and nothing else happened. Without headphones the same file produces a mild flutter or nothing at all.

So the verdict on the whole binaural line is: the perceptual effect is real, tightly constrained, and worth having, and the consequences I was hoping for are not supported by anything I could measure. I would not spend money on it again. I would also not tell somebody else not to, because the pleasantness of the perceptual effect is not nothing, and a practice that holds a regular schedule has value independent of what the state change is supposed to mean.

Three months of swapping tracks

For one quarter I kept a spreadsheet, which is the correct response to a practice I could not evaluate any other way. Fourteen tracks, each run for a fortnight, with a simple record: minutes to the first clear state change, whether the state change held, and whether I had stopped noticing the music by minute five.

The three tracks that performed best shared almost nothing except a clear beat in a moderate range. One was a piece I had never chosen for anything. One was instrumental, which I had assumed mattered. One was, in fairness, quite good, and it was the only one of the fourteen I would still choose today. So the honest summary is that the variable is tempo and continuity, the track is close to irrelevant, and the reason I did not know this for a year is that I had framed it as a question about music when it was a question about scheduling.

Four findings, in descending order of confidence

Four findings, in descending order of how confident I am. First, continuity of the piece predicted reliability better than any property of the piece itself. Second, a clear moderate beat predicted the state change better than frequency or volume. Third, familiarity within a session killed the effect, and familiarity across months built it. Fourth, the elaborate apparatus made no measurable difference to any of the above.

None of these findings is dramatic, and I want to be careful not to oversell them into a theory. They describe a scheduling practice that worked for one person over one year. What they do support is a conclusion about priorities: if you are building a musical ritual, spend the effort on when and how long rather than on what, and be suspicious of any account that requires a purchase to work. I never bought anything, which means the year produced no receipts, and that is the single most useful thing about it.

The rule I ended up with

One piece. A beat you can tap, in a moderate range. Volume low enough to talk over. Unchanged for a month before you judge it. Same time of day. Same length. That is the whole protocol and it fits in a sentence, which is a sign that the apparatus it replaces was mostly decoration.

The part I resisted most was the month of continuity, because it felt like it should not be necessary. It is necessary. A cue that changes is not a cue. The tradition this material comes from does not have this problem, because the drummer is the drummer and the pattern is the pattern, and continuity is built into the form. Reproducing that with recorded material requires a rule about not changing the recording, which nobody states because it seems too obvious to write down.

Someone else's music, and the ethics

A working built on a commercial recording uses somebody's work as a load-bearing component of your practice. I do not think that needs to be avoided, and I certainly did not avoid it for a year, but it should be a choice rather than an oversight. The recording is not a ritual object. It is a convenience that happens to be owned by someone else, and it may be unavailable, reissued, or altered in a way that changes the tempo you calibrated to.

The part I have become more careful about is the notion of a piece being especially suited. That gets said often enough to be a soft form of claiming, and the person who made the recording did not make it for a magical sequence. Where the practice is devotional the appropriation question is sharper, because the music sits inside a religious framing the maker may not share. My own working is non-devotional and my rules are simple: a piece I did not commission, used at a level where I could replace it with a metronome, never presented as having any particular spiritual property. The metronome test is not necessary. It is just honest.

What I would set up now

A single consistent piece with a pulse, a fixed window, a fixed length, and a written record of the only two numbers I actually use: minutes to the first clear shift, and whether it held. Thirty days before any judgement. Written down at the start, because the version of myself that starts a practice confidently is not the version that reports on it honestly, and the gap between those two is where every finding in this account came from.

I would also start with the state change and not with an aim. Eight months of aiming at a productive session and getting one about a third of the time was not a failure of the music. It was a failure of the aim. The music did its small job reliably and I kept asking it to do a large one, and the fix cost nothing and was available in month two. That is the part I would tell someone starting, and the part I would tell my year-one self, and the only part of this I am confident enough to state as general.

Frequently Asked Questions

Do binaural beats actually work for magic?

The perceptual effect is real and requires headphones, since it depends on each ear receiving a slightly different frequency. The entrainment and cognitive claims attached to specific frequencies have not replicated well in controlled studies, and the behavioural consequences of entrainment in general are modest. In my own record the pleasantness was consistent and the state change was absent. Worth using for enjoyment, not worth building a practice around on the strength of a claim about attention.

Does the music have to be spiritual or symbolic?

No, and working out that this is not required was the main practical result of a year of notes. The physiological part of the effect is tempo and continuity, neither of which is symbolic. What makes it ritual is that you use it in a fixed place, at a fixed time, for a fixed purpose, and that it is not otherwise in use. That last condition is the one that carries the most weight, since a piece you already use for other things is not a cue.

How long should I keep the same track?

At least a month before you judge it, unchanged. Continuity predicted reliability better than any property of the piece did, which is the opposite of what variety-seeking intuition suggests. Note that this is about the piece across weeks and the effect dying within a single session, which are different mechanisms. A piece that has gone flat inside a run is doing its job as a cue. A piece that has gone flat across a month is a decision you have not made yet.

Is it a problem to use commercial music in a ritual?

Not a serious one if the practice is non-devotional, but make it a choice rather than an oversight. You are using someone's work as a scheduling device, and you could replace it with a metronome at any time, which is the test I apply. It becomes a real problem when the music is presented as having a property the maker never put there, and particularly so inside a devotional frame where the maker may not share the framing at all.

What if I want silence instead?

Then use silence, on the same terms: fixed place, fixed time, fixed length, written record. Silence has a clear pulse in the sense that your own attention supplies one, and it avoids the carpeting effect entirely. What you lose is the masking, which matters if something needs to be recited that you do not want to interrupt yourself on. What you keep is the scheduling, which is most of what was actually working.