The Ideomotor EffectPendulums · Dowsing · Talking boards · Automatic writing · Muscle testing

How to read this page

The first half teaches these tools the way a good practitioner would teach them: fully, warmly, with the real method and none of it held back. The second half hands you the thinking tools. You need the first half to understand the second. Somebody who has never held a pendulum properly has no idea what is actually being explained.

What these tools are for

Every tool on this page does the same job in a different costume: it takes something too small to see and makes it big enough to read. A weight on a thread. A rod balanced on a fingertip. A planchette gliding on felt. A hand resting lightly on another hand. In each case a tiny movement gets converted into a large, obvious, legible one.

That is a real and useful function, and it is worth being precise about it before anything else. These are not receivers. They are amplifiers. The whole question of this page, and it is a fair question, seriously asked and seriously answered, is what exactly is being amplified.

People come to these tools for four reasons, and they are not the same reason:

• To decide. Yes or no, this or that, stay or go. A question with two doors and no obvious way to pick one.
• To listen. To something felt as guidance: a guide, an ancestor, a deceased person, the divine, the higher self, depending entirely on the tradition.
• To locate. Water, a lost object, an energy line, a sore spot on a body.
• To surface. To find out what you actually think, underneath what you are supposed to think. This is the one most people underestimate, and by the end of this page it may be the one you take most seriously.

Hold on to those four. They come back at the end, because the honest answer to “does this work?” is different for each one.

The pendulum, taught properly

A pendulum is a weight on a chain or a thread, usually four to eight inches long, held so it can swing freely. The weight can be crystal, brass, wood, a wedding ring, a washer on a shoelace. Transmitted Traditions differ on materials and every one of them will tell you their material matters. Physics says what matters is that the weight is heavy enough to hang straight and light enough to move easily.

Setting your yes and your no

You do not read a pendulum from a chart. You establish its language first, every session, and this step is not optional, a practitioner who skips it is guessing.

• Steady it. Hold the chain between thumb and forefinger, elbow unsupported or lightly resting, and let the weight come to a stop.
• Ask for yes. Say, aloud or internally, show me yes. Wait. Do not help it. Something will emerge: a front-to-back swing, a side-to-side, a clockwise circle. Whatever it does, that is your yes for this session.
• Ask for no. Same again. It will usually pick something distinctly different.
• Ask for “I won't answer that.” Most beginners never establish this one, and it is the most important of the three. A tool with only yes and no will give you a yes or a no to a question that deserves neither.
• Calibrate against known truth. Ask three things you already know the answer to: your own name, the day of the week, whether you are sitting down. If the answers come back wrong, stop. The session is not usable today. This is basic practice hygiene and it is skipped constantly.

Asking well

The quality of a pendulum session is almost entirely the quality of the questions. Vague in, vague out.

• One clause, no conjunctions. “Should I take the job and move cities?” is two questions wearing a coat.
• Present tense, present conditions. “Is this the right time to raise it with her?” not “Will it go well?” The first is answerable; the second asks the weight to see the future.
• Never ask a question you are not willing to hear both answers to. Practitioners across every tradition say this, and they are right for reasons that will become very clear in Part II.
• Never ask it a medical question. Not diagnosis, not dosage, not whether to stop a medication. No serious teacher in any lineage endorses this, and Part II shows exactly what happens when people do it anyway.

Session hygiene

Transmitted Most living traditions that use a pendulum share a common discipline, and it is good discipline regardless of what you believe is happening:

• Settle first. A few slow breaths. Agitated hands make agitated answers, every practitioner notices this, and Part II explains precisely why it is true.
• A short session. Five or six questions. Fatigue degrades the readout badly.
• Never re-ask to get a better answer. If you find yourself asking a fourth time, the session is over and you already know what you wanted to hear. That noticing is itself worth more than the answer.
• Write it down. Question, answer, date. A practitioner with a notebook can check their own record. A practitioner without one is relying on memory, and memory keeps the hits and quietly discards the misses.
• Close deliberately. Thank whatever you addressed, still the weight, put it away. Closing matters for the same reason a ritual beginning matters: it marks the state as bounded rather than ambient.

The talking board, taught properly

A board printed with the alphabet, the numbers, YES, NO and GOODBYE, and a small pointer, the planchette, that two or more people rest their fingertips on. The pointer moves; letters get spelled; a message assembles.

Whatever you believe about the mechanism, boards are the tool on this page most likely to frighten someone, and the traditional practice around them is almost entirely about containment. Transmitted Here is that practice, taught in full:

• Decide who is in the room, and why. Boards are group tools, and a group is a psychological amplifier before it is anything else. Nobody who is frightened, drunk, grieving acutely, or being dared into it should have their fingers on the planchette.
• Name a single questioner. One voice asks. Everyone else is quiet. Crosstalk turns the session into a committee.
• Open plainly. State aloud what you are inviting and what you are not. Traditions vary in wording; what they share is that the invitation is specific and limited.
• Agree the stop rule before you start. Anyone at the table can end the session, at any moment, without justifying it, and the others do not argue. This one rule prevents most of the harm boards actually cause.
• Do not ask for proof of identity or for anything about death. The traditional prohibition is on asking when someone will die. It is a good rule for reasons that hold up under any explanation you prefer.
• Close every single time. Move the planchette to GOODBYE deliberately, say the session is finished, lift hands, put the board away. A session that is never closed is a session the participants keep half-open in their heads for days.

The traditional rules around boards are, almost without exception, rules about the people at the table. That is a clue, and it is worth holding on to.

Automatic writing, taught properly

Pen on paper, hand moving, editor off. The practitioner writes continuously without steering, judging, or stopping to read. Words arrive; sentences form; sometimes pages of it.

Transmitted The method is simple and the discipline is where the skill lives:

• Set a timer. Eight to twelve minutes. Long enough to get past the self-conscious phase, short enough that you stop while it is still coherent.
• Do not lift the pen and do not re-read. Re-reading mid-flow reinstates the editor, which is the one thing the technique exists to bypass.
• Write through the nonsense. The first minute is usually junk. Junk is part of the method, not a sign of failure.
• Ask one question, or none. Both are legitimate. Open sessions tend to produce more material; directed sessions tend to produce more usable material.
• Read it a day later, not immediately. This is the single strongest piece of advice in the whole technique. Material read while still warm is read with the same state that produced it.

Automatic writing has a serious creative pedigree quite apart from any spiritual claim. Attested The Surrealists used it deliberately as a compositional method, entirely aware that the source was themselves. It worked because the source was themselves. Bypassing the internal editor is a real technique with real output, and it does not require anybody else to be in the room.

Rods, and muscle testing, as they are actually practised

Dowsing rods

Two L-shaped rods held loosely in closed fists, long arms forward, free to swing horizontally, or a single forked branch held under light outward tension. The practitioner walks a search pattern. Crossing rods, or a dipping branch, marks the spot. Transmitted Dowsing has a long living tradition for water, for lost objects, for burials, and in some lineages for what are described as energy lines in the land.

Good dowsing practice, as taught: define the target precisely before you begin, walk a grid rather than wandering, mark every hit without editing, and cross the site again from a different bearing. That last step is the tradition's own quality control, and Part II shows why it matters more than any other instruction on this page.

Muscle testing

Contemporary The practitioner presses down on the client's outstretched arm while the client holds a substance, or while a question is posed. An arm that holds firm reads as yes, or as a good match; an arm that gives way reads as no, or as something the body is rejecting. It is used to select supplements, identify foods, and in some settings to make claims about illness.

It is taught as a body-listening technique: the theory is that the body knows before the mind does, and that the arm is the readout. That framing has a real insight buried in it, which Part II takes seriously. The specific practice of using it to answer medical questions is where this page will not follow, and §13 explains exactly why.

The discipline these traditions already have

Read back over §02 to §05 and notice what the practices have in common. Settle before you start. Calibrate against something you already know. Ask one clean question. Do not re-ask to get a better answer. Cross the site from a second bearing. Read it a day later. Write it down so you can check yourself.

These are not superstitions. They are error-correction procedures, arrived at by practitioners over generations, entirely empirically, because sessions run without them go wrong in recognisable ways. A tradition that tells you to check the answer against a known truth has noticed something true about the tool.

Part II will tell you what the tool is. It will not tell you those rules were foolish. It will tell you they were better than the people who invented them realised.

The mechanism, and three beautiful experiments

Thinking about a movement produces tiny muscular movements in the direction of that thought. You cannot feel them. You cannot introspect them. You cannot switch them off by deciding to. And any device that magnifies small motions, a weight on a thread, a rod on a fingertip, a pointer on felt, converts them into something large and obvious. That is the whole of it, and it was worked out by three exceptionally careful people in the space of twenty years.

Chevreul, 1833, brace the hand and the answer stops

Attested Michel Eugène Chevreul was one of the most distinguished chemists in France. He took the “magic pendulum” seriously enough to investigate it properly, and published his findings as an open letter to the physicist André-Marie Ampère in 1833, later expanding them into a full book on divining rods and pendulums in 1854.

He found two things. The swing diminished sharply when the arm was supported so the hand could not contribute movement. And the swing was sight-dependent. It changed when the operator could not see the pendulum. He concluded that imperceptible muscular activity started the motion, and visual feedback amplified it. Then he noticed the detail that matters most: once a person understood what was happening, they could no longer reproduce the movement at will.

Carpenter, 1852, the effect gets a name

Attested The English physiologist William Benjamin Carpenter presented a paper at the Royal Institution on 12 March 1852 titled On the Influence of Suggestion in Modifying and Directing Muscular Movement, Independently of Volition. He called the phenomenon ideo-motor action: an idea producing a movement, with no act of will in between. The paper was written partly in response to the talking-board and table-turning crazes of the day.

Faraday, 1853, the finest experiment on this page

Attested Michael Faraday was pressed to investigate table-turning, the parlour phenomenon in which a group rested their hands on a table and the table moved. He agreed reluctantly, and worked with sitters he described as thoroughly honourable and reliably successful at producing the effect. They insisted the table was pushing their hands.

First he ruled out any exotic force, testing whether various materials placed between hand and tabletop interrupted it. Nothing did. Then he built the thing that settled the question: a stack of thin cards, lightly cemented to one another, placed between the hands and the table, the topmost slightly overlapping so any slippage between layers would be recorded.

The logic is exact. If the table drove the hands, the upper cards would lag behind. If the hands drove the table, the upper cards would slide ahead.

The cards slid ahead. The hands were moving the table.

Then Faraday did the thing that matters most. He rigged a simple indicator, a light lever with a pointer, so that the sitters could see their own unconscious sideways pressure in real time. With the pointer in front of them, the table stopped moving. He published the result in The Times on 30 June 1853 and in the scientific press the same year, concluding that tables move by unconscious muscular action and by nothing else.

The signature: it stops the moment you can watch it

The effect vanishes when the operator can observe it. That single sentence is the fingerprint of ideomotor action, and it is why every honest test of these tools works by removing what the operator knows.

Chevreul found that people who understood the mechanism could no longer produce the movement. Faraday found that sitters who could see their own pressure stopped producing it. This is not a quirk. It is diagnostic. A genuinely external force does not care whether you are looking at a dial.

It also explains something that otherwise looks like bad faith. Practitioners often say the tool stops working under scrutiny, or in the presence of doubt, or when tested. They are reporting something completely real. They have observed the effect accurately. The disagreement is not about the observation; it is only about what the observation means.

The talking board is a patented parlour game

Attested The board has no ancient lineage. It has a patent number. Elijah Bond filed on 28 May 1890 and the patent was granted on 10 February 1891, number 446,054, assigned to Charles Kennard and William Maupin. The Kennard Novelty Company had incorporated in Baltimore the previous October, expressly to manufacture and sell them. The boards sold for about a dollar, as family entertainment, alongside chess sets.

Attested It has been a toy-company product ever since. The Fuld family sold the rights to Parker Brothers in 1966; Parker Brothers was acquired by Hasbro in 1991, which manufactures the board today. The sinister reputation is largely downstream of The Exorcist in 1973, by which point the board had been a mainstream boxed game for over eighty years.

There is a nice detail in the origin story. Contested The patent office is said to have demanded proof the device worked before granting protection, and to have been satisfied when the board spelled out the examiner's name. Accounts of who actually invented the board conflict badly. Kennard, Bond, and several former employees all later claimed it in the Baltimore press, and a cabinet maker named E. C. Reiche, who died in 1899, is arguably the strongest candidate. Treat the charming version as folklore.

Why the board is so persuasive

With several hands on the planchette, nobody feels themselves pushing. Each person feels the others doing it. The group produces text that no individual consciously authored, which is precisely and exactly the experience of a presence in the room. The board is a machine for manufacturing that experience, and it is extremely good at it. That is not a slur on the people at the table. It is a description of a very effective piece of design.

The rotation logic

If a presence outside the participants is guiding the pointer, the message should survive the participants losing track of where the letters are. Blindfold everyone and quietly turn the board, and the pointer should still find the right letters, because the presence can see the board and the participants cannot.

And the genuinely interesting finding

Studied Hélène Gauchou, Ronald Rensink and Sidney Fels ran a careful experiment at the University of British Columbia, published in Consciousness and Cognition in 2012. Participants answered general-knowledge yes/no questions twice: once by speaking, once on a board. Each was paired with a confederate posing as another participant; once the real participant was blindfolded, the confederate quietly took their hands off the planchette. The participant was moving it alone and had no idea.

For questions the participants believed they knew, both methods performed similarly. For questions they believed they were guessing, spoken answers ran at chance, 50%, while board answers came in around 65%. Implicit knowledge, expressed through the ideomotor channel, that the person could not reach by asking themselves directly.

Studied The honest caveat, and it matters: a 2021 study substituting a Chevreul pendulum for the board found no such advantage, guessed questions came out at chance in both conditions. So the effect is real but not robustly general, and it should not be inflated into “the board knows things.” What it supports is narrower and still remarkable: under the right conditions, ideomotor movement can express knowledge that verbal report cannot reach.

Dowsing under test

Studied The most serious investigation of dowsing ever conducted took place in a two-storey barn near Munich in 1987 and 1988, funded by the German federal government to the tune of around 400,000 Deutschmarks. Roughly 500 dowsers were screened and the best 43 selected. Water was pumped through a movable pipe on the ground floor; the pipe was repositioned before each trial; the dowser worked on the floor above, trying to locate it along a ten-metre line. Over 800 trials were run.

The overall result was chance. The great majority of the pre-selected, extensively tested dowsers showed no ability whatsoever.

Contested The research team, led by physicist Hans-Dieter Betz, argued that a handful of individuals had performed so well that a real core phenomenon could be considered empirically demonstrated. Jim Enright, a physiologist at the University of California, reanalysed the data and concluded the standout results were exactly what you would expect from chance across that many trials, and that the original analysis used non-standard statistical methods fitted to the data after the fact. Betz called Enright's approach crude and insensitive and cited a third-party reanalysis supporting his conclusion. Enright published a final account in Skeptical Inquirer in 1999.

You should know that this dispute exists rather than be told it was settled cleanly. What is not in dispute is Enright's own summary of the study: that it was the most extensive and careful test of dowsing ever attempted, conducted in a sympathetic atmosphere on a hand-picked group, and that it is unlikely anyone will ever do a bigger one.

The pattern repeats across every tool on this page: the effect is there until the moment the operator's knowledge is removed.

⚠ Where dowsing killed people

The operators at those checkpoints were not frauds. They were soldiers and police officers holding a device that swung in their hands, doing their jobs, watching it respond. Ideomotor self-deception, sold with confidence, put explosives through security cordons. Any page that treats dowsing as a harmless quirk is not being serious with you.

⚠ Facilitated Communication

If you read one section of this library, read this one. It is also the section where the framing has to be exactly right, so let us set it first.

Attested In Facilitated Communication, a facilitator physically supports the hand, wrist or arm of a non-speaking person while they type or point at letters. From the late 1980s it produced fluent, articulate, emotionally complex messages from people who had never communicated in words. It was received as a miracle. Families wept. It was covered as one of the great breakthroughs in disability.

Studied Then it was tested, and the test was almost embarrassingly simple. Show the facilitator and the disabled person different images, or ask a question only one of them has heard. This is called message-passing. When facilitator and person saw the same thing, answers were correct. When they saw different things, the typed answer matched what the facilitator saw.

A systematic review of the authorship evidence published in Augmentative and Alternative Communication in 2014 concluded there was unequivocal evidence of facilitator control. An updated review covering 2014 to 2018 found no new authorship studies at all and no evidence that the messages are authored by the disabled person, an empty review, conducted to inform the American Speech-Language-Hearing Association's revised position statement.

Attested The facilitators were overwhelmingly not frauds. They were parents, teachers, aides and therapists who loved these children, believed completely in what they were doing, and were entirely convinced they were not moving the hand. Many refused to accept the results even after watching the tests. That is the ideomotor effect operating at full scale, in a person acting from love.

This is why the page exists. Ideomotor action can produce not just yes and no, but fluent, grammatical, sustained, emotionally rich narrative, authored by a sincere person who is certain they are not authoring it.

Sit with the consequence. Channelled material, automatic writing, spirit messages, trance speech, intuitive downloads: the mechanism to produce all of it, in complete good faith, already exists and has been demonstrated in controlled conditions. Not one dishonest person is required anywhere in the room.

⚠ Muscle testing and applied kinesiology

Contemporary Applied kinesiology was developed by the chiropractor George J. Goodheart Jr. in 1964, from his observation that a patient's weak muscle strengthened after he worked on it. He built it out into a diagnostic system linking specific muscles to specific organs, nutrients and meridians.

Studied It fails under blinding. A 1988 study in the Journal of the American Dietetic Association had three experienced practitioners assess nutrient status in athletes; the practitioners disagreed with each other, failed to match laboratory assays, and could not reliably detect real differences in muscle strength. The conclusion was that using the method to evaluate nutrient status is no more useful than random guessing. Other blinded trials have failed to distinguish an active substance from a placebo. Reviews consistently find performance at chance.

The mechanism is ideomotor action plus expectancy, running on both sides at once. The practitioner presses a fraction harder, or a fraction longer, without knowing it. The client resists a fraction less. Neither of them is lying and neither of them can feel it. Two people, in good faith, producing a result together and both experiencing it as something the body did on its own.

The steelman, and it is stronger than people expect

The tool works. It simply is not listening to anything outside the room. The pendulum tells you the truth: about you. And if you want to know what you actually think, underneath the answer you are supposed to give, that is a real and valuable thing to own.

The only failure mode is believing the answer came from somewhere else, because at that moment you will obey it instead of examining it.

Your unconscious is not your enemy

Your body moves without asking you. Your hands know things before you do. Something in you is always answering.

That is not a haunting. It is not an intrusion. It is you.

The ideomotor effect is not evidence that you are being fooled by your own biology. It is evidence that a great deal of you is working below the waterline all the time, and that occasionally it finds a channel to speak through. Finding out that the voice was yours is not a loss. Most people, once they sit with it, find it is the opposite.

The danger was never the unconscious. The danger begins the moment you decide it is someone else, because at that instant it stops being material you can examine and becomes an authority you must obey.

Which is the same sentence as the shadow work page, arrived at from the other direction: what you will not own, you will meet as something outside you, and then you will do what it says.

⚠ Risk

Test it yourself, at your own kitchen table

You do not have to take anyone's word for any of this, including ours. This is the one page in the library where you can settle the question yourself in about ten minutes, with things you already own.

• The pendulum brace test. Hold the pendulum and ask something you know the answer to. Watch it swing. Now brace your forearm and wrist firmly against a solid table so the hand genuinely cannot move. Ask again.
• The board rotation test. Play normally. Then blindfold everyone at the table and have a non-participant quietly rotate the board 180 degrees. Ask a question. Read out what gets spelled.
• The muscle test. Have someone test you with a substance. Then have them test you again with the substance concealed in identical opaque containers, shuffled by a third person, so that neither of you knows which is which.

Nobody has to argue with you. You can just look.

And notice what you are actually learning if the answer comes out the way the last two centuries suggest it will. Not that you were foolish. That your body knows things, moves without consulting you, and can produce entire conversations you would swear you had nothing to do with. That is an astonishing fact about being a person, and it happens to be true.

What a new seeker should actually ask

Provenance

Every claim above, with its source and its status. Primary sources are named so you can go and check us.

✦ For practitioners

Nothing here asks you to put the pendulum down. It asks you to change one sentence in how you describe what you are doing, and to add three habits.

• Say “this surfaces what I know” rather than “this tells me.” It is more accurate, it is more defensible, and it protects the person in front of you from obeying a movement.
• Keep a hit-and-miss log. Date, question, answer, and what actually happened. Most practitioners have never done this, and it is the single fastest way to find out what your practice is really doing.
• Build in a blind step. Have someone else set up a condition you cannot know. If your practice survives that, you have something worth knowing. If it does not, you have learned where the value actually lives, and §14 says it is still substantial.
• Refuse medical questions out loud, every time. Not with an apology. Refusing them in front of a client teaches them a standard they will carry to the next practitioner, who may not refuse.
• Watch for the re-ask. In yourself first. The fourth attempt at a question is data about you, not about the answer.

✦ For those guiding others

• Never let a tool make a decision for someone else. A pendulum held over another person's life is your unconscious answering their question. Say that plainly.
• Handle board fear directly and kindly. Someone frightened after a session needs the mechanism explained gently and completely, it is usually a profound relief, and needs to be asked, without drama, how they are sleeping and whether anything else is frightening them right now.
• Do not use these tools to answer questions about abuse, safety, paternity, or crime. The Facilitated Communication cases are what that looks like when it goes wrong, and the harm falls on people who were never in the room.
• If a client brings you a muscle-test diagnosis, do not simply contradict it. Ask what they were told, ask whether a doctor has looked at it, and make the referral easy to accept rather than shaming.
• Teach the mechanism before you teach the tool. A student who understands ideomotor action from day one can use a pendulum for the rest of their life without ever handing it authority.