Psychokinesis

← Encyclopedia

Psychokinesis is the claim that mind can act on matter directly — that intention alone can tip a falling die, bend a spoon, or pull a stream of random numbers off its course. Every curse and blessing presupposes something like it. The modern word names that presupposition restated as a hypothesis; what happened when it entered the laboratory is a record kept in numbers.

“Telekinesis” was coined in 1890 by the Russian psychical researcher Alexander Aksakof, translating the German Fernwirkung for the séance phenomena of his day. “Psychokinesis” is usually credited to the American publisher Henry Holt in his 1914 On the Cosmic Relations, though the word is attested as early as 1904. J. B. Rhine adopted Holt’s term at Duke, abbreviated it PK, and made it the umbrella, with telekinesis kept for visible movement. The field splits the claim into macro-PK — effects the naked eye can see — and micro-PK, statistical deviations only analysis can find. The subject’s whole evidentiary history is a migration from the first to the second.

As Rhine told the story, a young gambler visited the new Duke laboratory claiming he could sometimes make dice fall as he wished. The demonstration was unimpressive; the method was not — dice come with chance baselines built in. From 1934 he set subjects willing dice totals above seven and reported a small persistent edge — 5.53 hits per twelve-throw run where chance predicts five — then re-ran the tests willing low totals, against the dice’s own pitting bias; the edge survived. He withheld publication until 1943, then summarized twenty experiments, resting his case on hit rates declining down the record sheet — a pattern, he argued, no artifact or fraud should produce. The skeptical reply came from the record sheets: a 1952 Yale study filmed subjects scoring their own throws and found believers making tally errors favoring the effect, disbelievers the opposite — and the same pattern in Rhine’s records, then the strongest evidence on offer.

The dice literature received its full accounting in 1991, when Dean Radin and Diane Ferrari meta-analyzed 148 studies — some 2.5 million throws — and claimed odds against chance of 10⁷⁰ to one. Restricted to the 69 best-controlled studies, the odds fell to about a million to one; weighting every study by quality cut the effect size in half. Both camps accept the figures and read them oppositely: proponents point to the residue that survives the weighting, critics to the slope — an effect that shrinks as controls tighten is what an artifact looks like.

The macro-PK record is older and stranger. Daniel Dunglas Home, the most celebrated physical medium of the Victorian age, was reported in 1868 to have floated out of one third-storey window and in at another; the three witnesses agreed on almost nothing — address, ledges, height, lighting — and Home had stood on the same ledge days before. William Crookes tested him in the early 1870s with a spring balance and a caged accordion and declared a psychic force proven, but the light was dim, the accordion played under a dining table, and the magician John Nevil Maskelyne asked “why, in the name of common sense, have it underneath the table?” Eusapia Palladino was caught so often it became her distinction — Cambridge 1895, Antoniadi 1898, Columbia 1910 with magicians watching her freed left foot do the levitating — and rehabilitated each time: defenders pointed to better-controlled sittings where nothing was caught; critics answered that the investigators had missed it. Her 1908 Naples sittings remain the proponents’ best exhibit; the rebuttal — semi-darkness, fatigue, committed investigators — sits on record beside them.

The pattern repeated on television. Uri Geller brought metal-bending to a world audience in the early 1970s and was tested at Stanford Research Institute in 1973 — the resulting Nature paper claimed only picture-drawing perception, and SRI never validated the bending; Ray Hyman, reviewing for the government, called the protocols “sloppy and inadequate.” When Johnny Carson — an ex-magician, advised by James Randi — supplied his own props on the Tonight Show, Geller produced nothing for twenty-two minutes — and the failure backfired, multiplying his bookings. Spoon-bending itself is standard conjuring — misdirection, fatigued metal, the pre-bent spoon — and when Randi planted two teenage magicians in a parapsychology lab in 1979, the researchers took their tricks for the real thing. Standing prize money for one spoon bent under control was never collected; even academic parapsychology, by Caroline Watt’s 2016 accounting, mostly finds the macro-PK evidence uncompelling.

Micro-PK was the retreat to higher ground. From the late 1960s the physicist Helmut Schmidt had subjects try to bias generators driven by radioactive decay, reporting one- or two-percent deviations over enormous trial counts; critics granted him the field’s best methods and still found the work unrepeatable — he worked alone, as both experimenter and subject. The Princeton Engineering Anomalies Research laboratory, founded in 1979 by Robert Jahn, then Dean of Engineering, ran the paradigm for twenty-eight years — millions of trials, aggregate effects of fractions of a percent. The aggregate had troubles: baseline runs that did not vary as statistics require, and one operator — presumed a staff member — supplying fifteen percent of all trials and half the total effect. A consortium including PEAR itself failed to reproduce the results; the laboratory closed in February 2007, ending micro-PK’s academic life in the United States.

The meta-analytic endgame came the year before. Bösch, Steinkamp, and Boller, surveying 380 RNG studies in Psychological Bulletin in 2006, found a statistically significant but very small overall effect, extreme heterogeneity, and effect sizes inversely related to sample size — a combination their simulations showed publication bias alone could produce. Wilson and Shadish judged it “nearly impossible to draw any conclusions” from the corpus, and told researchers to produce larger effects or say what would make them accept the null. Radin and colleagues replied that experimenters run more trials precisely when effects are small, which manufactures the inverse relation, and that selective reporting on the required scale is implausible. The exchange ended with both sides claiming the same data — the field’s cleanest statement: the effect is tiny, the studies heterogeneous, and the dispute is whether the file drawer can carry it.

The boldest surviving claim is the Global Consciousness Project, begun in 1998 by PEAR’s Roger Nelson: dozens of hardware generators worldwide, tested against pre-registered windows around world events. The project’s formal result — five hundred events through 2015, a cumulative deviation of 7.3 standard deviations, odds against chance beyond a trillion to one — is granted even by its sharpest internal critic. May and Spottiswoode found the flagship September 11 deviation existed only inside the chosen time window; Scargle objected that the result will not be accepted until Bayesian and frequentist analyses agree. The deepest cut came from inside: after seventeen years, project analyst Peter Bancel concluded the data point not to a global mind but to an effect “associated with persons directly engaged with the experiment”; Nelson has answered, defending field-like structure; the exchange is unresolved.

The physics objection is not that psychokinesis is unexplained but that the room it would need is already occupied. As Sean Carroll spelled out in 2008, only two long-range forces can move a spoon; electromagnetic fields strong enough to bend metal would be trivially detectable — and brains cannot produce them — while any new force must be shorter-ranged than a millimeter or weaker than gravity to have escaped existing experiments. The casino argument needs no field theory: if intention moved dice, casino earnings would drift from chance; they do not. A National Research Council panel in 1988, reviewing the field for the Army, found no scientific justification in 130 years of research. Proponents answer from quantum mechanics — Schmidt theorized PK as the observer biasing quantum collapse; later theory retreats to non-signaling mind–matter correlations that cannot, by design, carry a message, and so, critics note, cannot be refuted by failed replication. The theoretical migration — force, then bias, then correlation — parallels the evidential one, from levitating tables to dice edges to flickers in a random stream. Every step keeps the claim alive, and every step leaves it harder to test.

Seen whole — the verdict that follows is interpretive, the site’s own — the subject has a strange integrity. Mind acting on matter is the oldest magical claim there is — the Hermetic tradition formalized it as the operator’s will at work on the cosmos — and psychokinesis is that claim walked under the statistician’s lamp. The lamp has been steady. From Home’s parlors to Rhine’s dice to PEAR’s diodes, the effect grew smaller as the watching grew better, until it was a fourth-decimal whisper carried half by one operator, defensible only by argument over a file drawer. No single exposure settled the question, and none needed to.

→ Related: J B Rhine · Remote Viewing