Do Spells Actually Work? An Honest Answer From People Who Sell Ritual Work

The question hides three different questions

"Do spells work?" gets answered badly because it is really three questions wearing one coat, and the honest answer differs for each.

The first is causal: can a rite performed in a room change an event outside that room — make a specific person return, a court rule your way, money arrive? The second is psychological: does performing a ritual change the person who performs it, in ways that show up in how they feel and what they then do? The third is practical: given the first two answers, is paying someone for ritual work a reasonable thing to do or a waste of money?

Most writing on this subject blurs the three deliberately. A seller who wants your money will answer the second question with evidence and let you hear it as an answer to the first. A skeptic who wants to dismiss the whole field will answer the first and let you hear it as an answer to the second. We are going to take them in order and keep them apart.

What happened when researchers tested the causal claim directly

The closest thing to a controlled test of "someone performs a rite on your behalf, at a distance, for your benefit" is not a study of witchcraft. It is a study of prayer — structurally the same claim, and far better funded than anything our field has ever attracted.

The Study of the Therapeutic Effects of Intercessory Prayer, published in the American Heart Journal in 2006, enrolled 1,802 coronary artery bypass patients across six hospitals. Patients were randomized into three groups: some received intercessory prayer and were told they might or might not; some received none and were told the same; and a third group received prayer and were told with certainty that they would. Congregations prayed for named patients by first name and last initial.

The result was clean and negative. Complications within thirty days occurred in 52% of the prayed-for uncertain group and 51% of the not-prayed-for uncertain group — statistically indistinguishable. Mortality did not differ. The one statistically significant finding ran the wrong way for the hypothesis: the group told with certainty that they were being prayed for had more complications, 58.6%, a difference the authors could not explain and did not claim as a real effect of prayer.

That is the strongest available evidence on the causal question, and it points one direction. The related literature does not rescue it. When a Cornell psychologist published experiments in 2011 reporting that people could be influenced by future events, three independent replication attempts published in PLOS ONE in 2012 found nothing. This is what a null result looks like when a field keeps looking.

So: if you are asking whether a ritual can reach across distance and move an event, the honest answer is that nobody has shown it, and the best-designed attempt to show it failed. We are not going to tell you otherwise in order to sell you something. If a practitioner tells you otherwise, you have learned something important about that practitioner — and our guide to the red flags of a fake spell caster covers the rest of the pattern.

The "science says rituals work" story has a problem too

Here is where we part company with most of our own industry, which loves to cite psychology as proof that rituals are scientifically validated.

The single most-quoted paper in that genre was titled "Don't stop believing: Rituals improve performance by decreasing anxiety," published in Organizational Behavior and Human Decision Processes in 2016. You have probably read its findings summarized in a wellness article without knowing the source. It was retracted in November 2024. Its companion study on ritual and grief, published in 2014, has not been retracted, but on the public living document where its co-authors record what data they can still verify, the surviving lead author states that he does not have the data needed to reproduce the published results, and that the co-author now at the center of a data-integrity investigation collected the data for all four studies.

We are pointing at this because a great deal of "ritual is proven to reduce anxiety" content on the internet traces back to those two papers, and it has not caught up. Any practitioner still citing them as proof either has not checked or is counting on you not to.

What the surviving evidence does support

Strip out the retracted work and something smaller but more honest remains, and it comes mostly from anthropology rather than business schools.

In 2015, a team publishing in Current Biology showed the relationship running in the direction people rarely mention: inducing anxiety in a laboratory produced spontaneous ritualized behavior — action that became more repetitive and more rigid. Humans reach for ritual under pressure without being taught to. That finding says something real about what rituals are for, and it does not depend on anyone's contested dataset.

The reverse question — whether performing a ritual then lowers the anxiety — has a weaker and more interesting answer. A pre-registered experiment with 268 participants, published in Scientific Reports in 2022, had people perform either a ritualized action or a non-ritualized one after an anxiety-inducing task. The ritual group did show greater reductions in both self-reported anxiety and physiological arousal. But the authors were candid that the confidence intervals still contained zero: the effect was real in direction and imprecisely estimated in size. It sharpened considerably in the subgroup who had actually been made highly anxious.

That is an unglamorous result, and it is the honest one: a modest effect, clearest for people who genuinely need it, stronger for practices that are familiar than for rituals invented on the spot in a lab.

The most useful adjacent finding is about knowing. In a 2021 trial published in Pain, 262 people with irritable bowel syndrome were given either a placebo they were openly told was a placebo, a blinded pill, or nothing. Roughly 70% of both pill groups improved meaningfully, against 54% of the no-treatment group — and the open-label group, who knew exactly what they were taking, did as well as those who did not. Deception turned out not to be the active ingredient.

Applied here, that is the most defensible claim anyone in our field can make, and it is narrower than the pitch you usually hear: a deliberate practice can do something for you even when you understand precisely how it functions. You do not have to believe a candle has power for lighting one on purpose, at a chosen time, for a stated reason, to do work on you.

What ritual does not do

Three limits, stated plainly, because most of the damage in this field happens where these get fudged.

It does not override another person's will. Nothing in the evidence, and nothing in the responsible end of the tradition, supports the idea that a rite makes a named human being love you, return to you, or call. Work aimed at a specific person's choices is where ritual practice stops being a practice and becomes a fantasy of control sold to someone in pain.

It is not treatment. Ritual is not a substitute for medical or psychological care, and no honest practitioner will position it as one. If something is wrong with your body or your mind, the candle is not the intervention.

It does not come with an outcome. This is the one that separates the trade from the fraud. A practitioner can promise the work, the craft and the documentation. Nobody can promise the result, and a guarantee in this field is not a sign of skill or confidence — it is the single most reliable marker that you are being worked.

Why "it worked" is such an unreliable report

Suppose you do a ritual and something good follows. What can you actually conclude? Less than it feels like, and the reasons are worth knowing before you spend money.

Most requests are made about situations that were already in motion and had some chance of resolving anyway. Most are worded loosely enough — "things get better," "clarity," "an opening" — that almost any subsequent week contains something that fits. And once you have made an effort toward an outcome, you begin noticing evidence for it, which is not a character flaw but standard human perception. Add the fact that the stories which circulate are overwhelmingly the successes, since nobody posts about the ritual that was followed by an ordinary Tuesday, and you have a reporting system that manufactures confirmation on its own.

There is also a mundane confound that almost nobody names. Doing a ritual about a problem generally involves sitting down, deciding what you actually want, saying it in words, and committing to it. People frequently behave differently afterward — they make the call, end the situation, apply for the thing. If the outcome changes, the rite and the behavior change are hopelessly entangled, and the behavior change is the one with an obvious mechanism.

What you can genuinely check	What you cannot check, whatever anyone says
That the work was done, on a stated date, with described materials and wording	That the work caused any event that followed it
That the practitioner is a findable, named person with a real business behind them	That an unnamed "master" has powers other practitioners lack
That the price was fixed in advance and did not grow	That a higher price buys a stronger effect
That your own situation was understood rather than templated	That a curse was present and has now been removed
How you felt and what you did in the weeks afterward	What would have happened if you had done nothing

How to test it on yourself, cheaply and fairly

You do not need to resolve the metaphysics to find out whether this is useful to you. You need to stop grading yourself on a curve. The method costs nothing and takes ten minutes.

So — do spells work?

No, in the sense most people mean when they type the question at two in the morning. No rite reliably reaches out and rearranges the world on your behalf, nobody has demonstrated that it can, and the largest study ever built to look for it came back empty. If that is what you are shopping for, no honest person can sell it to you, and the people who will are counting on your worst week.

Yes, in a smaller and more durable sense. A ritual is a structured way of deciding what you want, saying it deliberately, marking that you have done so, and acting from there — and the evidence that structured practice does something for the person performing it is modest but real, and does not require you to believe anything in particular. That is why people have paid specialists for this work for as long as there have been specialists, and it is what we actually sell: craft, attention, a rite built for one situation, and a written record of what was done.

What we will not sell you is the first answer dressed up as the second. If that makes the offer less exciting than the one three tabs over promising your ex back in seven days, that is the trade. The exciting one is not true.

Frequently asked questions

Is there any scientific proof that spells work?

Not for the causal claim. No controlled study has shown that a ritual changes an external event, and the largest relevant trial — 1,802 cardiac surgery patients randomized to receive intercessory prayer or not — found no difference in complications or mortality. There is modest evidence that performing ritualized action reduces anxiety in the person performing it, but the effect is small and imprecisely estimated.

If spells don't cause outcomes, why pay for ritual work at all?

For the same reason people pay for a well-run ceremony, a good coach or a structured process: the value is in the craft, the attention and having something built for your actual situation rather than a template. That is a defensible service. It is not a mechanism for controlling events, and should never be priced or sold as one. We have written separately about what ritual work actually costs and which price patterns signal fraud.

Does it still work if I don't believe in it?

Belief appears to matter less than people assume. In a 2021 trial, patients who were openly told their pill was a placebo improved as much as those given it blind — roughly 70% in both groups versus 54% with no treatment. Applied to ritual, that suggests a deliberate practice can do something for you even when you understand exactly how it functions.

How do I know whether a ritual worked?

Decide in advance. Write one specific result sentence and the date you will check it before you begin, and note what you expect to do differently. Vague wording and deadlines chosen after the fact are why almost every ritual seems to have worked a little. A criterion set beforehand is the only version of this test that tells you anything.

Can a spell make someone come back to me?

No, and anyone promising this is exploiting you. Nothing in the evidence supports influence over another person's choices, and the responsible end of the tradition does not claim it. Work of this kind is better aimed at your own clarity and what you do next — which is the part that demonstrably can change.

Sources & further reading

• Benson, H. et al.: "Study of the Therapeutic Effects of Intercessory Prayer (STEP) in cardiac bypass patients" — American Heart Journal 151(4), 2006
• Elsevier: Retraction notice to "Don't stop believing: Rituals improve performance by decreasing anxiety" — Organizational Behavior and Human Decision Processes, November 2024
• Lang, M. et al.: "Effects of Anxiety on Spontaneous Ritualized Behavior" — Current Biology 25(14), 2015
• Lang, M. et al.: "Effects of predictable behavioral patterns on anxiety dynamics" — Scientific Reports 12, 2022 (pre-registered, n=268)
• Lembo, A., Kaptchuk, T. J. et al.: open-label placebo randomized controlled trial in irritable bowel syndrome — Pain 162(9), 2021 (n=262)
• Ritchie, S. J., Wiseman, R. & French, C. C.: "Failing the Future: Three Unsuccessful Attempts to Replicate Bem's 'Retroactive Facilitation of Recall' Effect" — PLOS ONE 7(3), 2012
• Many Co-Authors Project: data-availability statement for "Rituals Alleviate Grieving for Loved Ones, Lovers, and Lotteries" (2014) — accessed September 2026