Telepathy

← Encyclopedia

Telepathy is the claim that one mind can reach another directly: that a thought, a feeling, the moment of a death can pass between two people with no word spoken and no signal sent. The wish underneath it is old and entirely human — to be known without saying; to learn, across any distance, that someone loved is in danger. The word itself is young, and it was built to sound like science.

It was coined in 1882 by Frederic W. H. Myers, a classical scholar and a founder of the Society for Psychical Research, formed in London that February to investigate mesmeric and spiritualist claims in a scientific spirit. The society’s first committee on the subject bore the plainer name Thought-Transference; Myers’s Greek compound — tele, at a distance; patheia, feeling — soon displaced “thought-reading.” The relabelling was deliberate, and it worked: it detached the idea from the séance room and dressed it for the laboratory. It also carried a hope critics later named openly: if living minds could touch without the body’s channels, minds outliving the body became easier to think. Eric Dingwall judged, looking back, that establishing telepathy, not open-ended experiment, had been the society’s real aim.

The first evidence arrived fast and collapsed almost as fast. The Creery sisters, a clergyman’s daughters whose guessing games filled the society’s first Proceedings, were caught signalling when their powers waned, and confessed to having sometimes cheated. The hypnotist G. A. Smith and the journalist Douglas Blackburn did better still, until Blackburn confessed in 1908 that “the whole of those alleged experiments were bogus.” Smith denied it to the end, and the confession’s details have been challenged; either way the affair became a standing lesson in how experimenters are fooled. The stage knew things the laboratory was slow to learn: the era’s celebrated “thought-readers” proved on examination to be reading muscles, not minds.

The society’s most ambitious early work was collection rather than experiment. Phantasms of the Living (1886), written mainly by Edmund Gurney with Myers and Frank Podmore, analysed 701 first-hand accounts of apparitions — many of them “crisis apparitions,” the figure of someone later found to have been dying or in danger at that hour — and read them as hallucinations telepathically generated by the distant mind. The Census of Hallucinations followed in 1894: some seventeen thousand people canvassed, roughly one in ten affirming a vivid waking hallucination, and a committee calculation that the death-coincidences exceeded chance. The objections came at once and have never really changed: C. S. Peirce held that no scientific conclusion can rest on anecdote, and others noted that the cases lacked contemporaneous record and that memory, exaggeration and deceit had nowhere been excluded. The shape of the next century is already visible here: a serious effort, a striking residue, and a critique that the controls were not equal to the claim.

The next move was from stories to statistics. In the 1930s the centre of the field shifted to J. B. Rhine’s laboratory at Duke, where card-guessing runs against a five-symbol deck made “extrasensory perception” a household phrase. That arc is quickly told, because it repeated the first one at scale: the early high scores are now discounted — sensory leakage and loose handling could account for them — the high scorers vanished as Rhine tightened his procedures, and the field eventually abandoned card-guessing altogether. A generation later the Maimonides Medical Center in Brooklyn ran a dream-telepathy programme (1964–1978): sleepers were woken from REM sleep to report their dreams while a sender concentrated on an art print, blind judges matching transcripts to targets. Some series were striking; none was independently replicated, a thinness even sympathetic reviewers concede. Sleep laboratories were also expensive, and the cheap successor became the battleground.

The successor was the ganzfeld. A receiver rests in mild sensory deprivation — halved ping-pong balls over the eyes, red light, noise in the headphones — and free-associates aloud for half an hour while a sender elsewhere views a randomly chosen image; the session’s target is then picked from four candidates, so chance is one in four. By 1982 some forty studies existed — forty-two by one count, forty-seven by another — and Charles Honorton argued they were sufficient evidence for psi. Ray Hyman disagreed, and in 1985 the two published duelling meta-analyses of the same database. Honorton found a hugely significant effect and no link between a study’s quality and its outcome; Hyman found that every study exhibited at least one of twelve methodological flaws, and that over half had failed to guard against sensory leakage. What happened next is rarer than either finding. In 1986 the two issued a joint communiqué: there is, they agreed, “an overall significant effect that cannot be reasonably explained by selective reporting or multiple analysis”; they continued to differ over whether it was evidence for psi; and the verdict would await new experiments, by more investigators, to stricter standards they specified together. Two adversaries had looked at the same data, refused to agree, and refused equally to pretend the disagreement could not be stated exactly. The exchange remains a model of adversarial honesty, and part of why the ganzfeld is still the case everyone argues about.

The new experiments came. Honorton’s own automated studies — 329 sessions between 1982 and 1989 — hit at roughly 32 per cent against the 25 expected, and were published with the social psychologist Daryl Bem in Psychological Bulletin in 1994 as evidence of a replicable anomaly; Honorton died at forty-six, nine days before the paper was accepted. Hyman replied that the significance came entirely from the new video-clip targets, the static pictures scoring at chance, and that one laboratory under one leader is not independent replication. Doubt reached backward too: protocol violations reported at Carl Sargent’s Cambridge laboratory touched a database in which Sargent’s and Honorton’s labs had supplied half the significant studies. In 1999 Julie Milton and Richard Wiseman meta-analysed thirty post-communiqué studies from seven independent laboratories and found nothing: the ganzfeld “does not at present offer a replicable method for producing ESP in the laboratory.” Proponents answered that the selection had excluded the successes. In 2010 Lance Storm and colleagues analysed twenty-nine newer studies and found 32 per cent again, at odds against chance of roughly fifty million to one; Hyman answered that the consistency had been manufactured by trimming outliers and merging databases, and that a meta-analysis is not a confirmation.

The argument now runs at the top of the literature. Etzel Cardeña argued in American Psychologist in 2018 that the cumulative evidence for psi is comparable to that for established phenomena in psychology and cannot be explained by fraud or selective reporting; Arthur Reber and James Alcock answered in the same journal that the claims “cannot be true” — pigs cannot fly — the data therefore flawed whether or not the flaw can be located. Each side gives up a different symmetry: one starts from the evidence, the other from the priors. Behind the priors stands the question of mechanism: the one physically plausible carrier, electromagnetism, was measured in telepathy tests in 1979 and found orders of magnitude too weak. Mainstream psychology’s settled verdict — delivered by a National Research Council review in 1988, which after 130 years of research found no scientific justification for the phenomenon, and unmoved since — treats the residue as the signature of selective publication and flexible analysis, with ordinary psychology — underestimated coincidence, confirmation bias — supplying the experiences. The field’s defenders concede an awkward regularity — the ganzfeld works extraordinarily well for a few experimenters and hardly at all for the rest.

All the while the word ran ahead of its evidence. Upton Sinclair’s Mental Radio (1930) reported his wife reproducing target drawings under conditions later critics judged uncontrolled; Einstein contributed a guarded preface — “We have no right to rule out a priori the possibility of telepathy.” Freud addressed it in four papers. Science fiction made telepathy its most durable paranormal ability — the hunted telepath-elite of Slan, the telepath-integrated society of The Demolished Man — until cyberspace gave fiction a more plausible wiring for mind-to-mind contact and the trope quietly receded. The folklore is older than the word and outlives every negative study. Twins were credited with a strange sympathy in John Wesley’s journal a century before Myers and in Dumas’s Corsican Brothers (1844); about a third of twins still report the connection, though controlled tests have found nothing; and in 2024 a podcast claiming telepathy in nonspeaking autistic children topped the American and British charts for weeks, resting on facilitated-communication methods discredited in the 1990s. The pattern repeats at every scale: the spectacular claim, then the cueing critique.

Read at this distance — and this is a reading, not a record — telepathy looks less like a discovery than a translation. What Myers did in 1882 was take an old idea, the sympathy of souls, the mesmeric rapport of operator and subject, and re-dress it for a statistical age: communion became transmission, spirits became signal. Historians of the coinage have traced exactly that movement. What no one designed is what the word became afterward — the longest controlled experiment ever run on an esoteric claim, fourteen decades of effects that appear and are then squeezed, retreating as the controls tighten or, as defenders insist, persisting a few stubborn points above chance. Either a small real signal has survived every control yet devised, or nothing has, and the controls themselves were the discovery. Both readings are still in print, in the same journals, answering each other.

→ Related: Mesmerism · J B Rhine · Modern Ganzfeld Experimental Parapsychology