Apparitions and hauntings

What it is

Two related but distinct claims.

An apparition is a perceived figure or presence with no physical source. The classic research category is the crisis apparition — a figure of someone seen at or near the moment of their death elsewhere, before the percipient could have known.

A haunting is recurrent anomalous experience tied to a place rather than a person: footsteps, figures, cold, movement of objects, and above all a strong sense of presence.

Poltergeist cases — recurrent spontaneous psychokinesis in the research literature — are distinguished by being tied to a person, usually an adolescent, and by involving movement of objects rather than figures.

Where it came from

Reports are universal and ancient. The systematic study is Victorian.

The Census of Hallucinations (1894) is the founding piece of work: the Society for Psychical Research surveyed 17,000 people, of whom around 10 per cent reported having experienced a vivid hallucination while apparently awake and in normal health. The investigators — Sidgwick, Myers, Podmore and colleagues — then examined the subset of crisis apparitions and argued the coincidence rate exceeded chance. Their statistical method has been criticised ever since, but the central finding, that ordinary healthy people commonly have vivid non-veridical perceptions, has held up completely and is now uncontroversial in psychiatry.

Phantasms of the Living (1886) collected several hundred such cases in detail.

How it is used

As evidence for survival; and, in the modern popular form, as entertainment — ghost-hunting television has made the apparatus of investigation into a leisure activity, with instruments used in ways their manufacturers would not endorse.

What we can and cannot say

We can say the experiences are extremely common and are not signs of illness. Perceiving a presence, hearing one's name, or seeing a figure — particularly in the hypnagogic and hypnopompic states at the edges of sleep, in sleep paralysis, in bereavement, in sensory deprivation, in grief, in exhaustion — happens to a large minority of the population. Bereavement hallucinations are especially frequent: studies of widows and widowers have found that a substantial proportion, in some samples a majority, perceive their spouse after death, and this is regarded as a normal feature of grief rather than a symptom.

We can say the sense of presence has a partial neurology. It can be induced by stimulating the temporo-parietal junction — the same region implicated in out-of-body experiences — and Olaf Blanke's group produced it in healthy volunteers in 2014 using a robotic system that introduced a delay between a participant's own movement and the touch they felt, so that their own agency was misattributed to someone standing behind them.

We can report the environmental hypotheses honestly, including the one that failed.

Infrasound — sound below about 20 Hz, at the edge of hearing — has been proposed as a cause, after Vic Tandy traced an apparition in his laboratory to a fan producing a standing wave near 19 Hz. Richard Wiseman's controlled study at Hampton Court found some association between reported experiences and physical variables including air movement and lighting, but the infrasound explanation on its own has not been established, and reviews find the evidence thin.

Electromagnetic fields. Michael Persinger's "God helmet" work, applying weak complex magnetic fields to the temporal lobes, reported induced sensed-presence experiences. A Swedish replication by Granqvist and colleagues in 2005 found the effect was explained by suggestibility, not by the magnetic field — participants who scored high on suggestibility reported presences whether the field was on or off. That is a clean negative result and it is the current state of the question.

Carbon monoxide is a real and documented cause: the 1921 case of a family reporting apparitions, hearing voices and feeling presences in a house with a faulty furnace is a genuine medical case report, and CO poisoning remains the one explanation on this list that can kill you. Anyone experiencing unexplained presences, headaches and lethargy in a specific building should have a CO alarm before anything else.

Mould — the Stachybotrys hypothesis, that toxic fungal exposure in old buildings produces the symptoms — is frequently repeated and is not well supported. It is a plausible-sounding idea in search of evidence.

We can say expectation drives the results. Wiseman's Hampton Court and Edinburgh Vaults studies found that people reported more experiences in areas they had been told were haunted, regardless of which areas those actually were.

We can say the modern investigation industry has no methodology. EMF meters respond to wiring. Temperature fluctuations occur in old buildings. "Electronic voice phenomena" is pareidolia applied to noise, and the effect gets stronger once you are told what to listen for.

We cannot say that every case has been accounted for. What we can say is that the experience is common, largely explicable, and that the one case in this article with a confirmed cause was a broken furnace.

Further reading

• Sidgwick et al., Report on the Census of Hallucinations (1894).
• Granqvist et al. (2005), on the failed God helmet replication.
• Richard Wiseman's Hampton Court and Edinburgh Vaults studies.