Premonitions

The teaching

As J. W. Dunne's An Experiment with Time (1927) teaches it.

Foreboding and foreknowing

A premonition is a sense that something is about to happen, usually something bad, with no reason that can be given for it. Precognition is the name for knowing an event before it happens, when it could not have been inferred. Premonitions come in many forms: a dream that seems to come true, a waking image, a dread that settles for no cause, a sudden reluctance to board a train or open a letter. J. W. Dunne's teaching concerns the first of these, the dream, and gives a method for testing it.

The second sight

Long before Dunne, the best-known foreknowing in Britain was the second sight of the Scottish Highlands and Islands. Martin Martin, who recorded it among the islanders, called it “a singular faculty of seeing an otherwise invisible object.” The vision took the seer suddenly: the eyelids lifted and the eyes stared until the thing vanished, and while it lasted the seer saw and thought of nothing else.

The seers read their visions by settled signs. A shroud seen about a living person foretold death; if it reached no higher than the middle, death was not looked for within the year, and as it rose toward the head, death was near. A woman seen standing at a man's left hand would be his wife. A seat seen empty while someone sat in it foretold that person's death. The hour of the vision gave the hour of fulfilment, so that a vision at noon would commonly come to pass that same day. Martin found that the faculty did not descend in families, could not be acquired by any compact, and was a burden to some who had it.

Dunne's claim

Dunne, who kept careful records of his own dreams for years, came to a startling conclusion: that all dreams, everyone's, are made of images of past experience and images of future experience blended together in roughly equal proportions. We do not notice the future half, he taught, because we never look for it, and because a dream's links to the future are as faint and trivial as its links to the past.

He held, too, that what a dream foresees is not an event itself but the dreamer's own coming experience of it: a report the dreamer will read, a scene the dreamer will see. In his best-known case a dream of a volcanic disaster matched not the disaster as it happened but his own later reading of the newspaper report of it, down to a figure he misread.

The experiment

Dunne's method can be followed exactly. Keep a notebook and pencil under the pillow. Immediately on waking, before even opening your eyes, set yourself to remember the vanishing dream. Usually one small incident comes. Do not chase the rest: fix your attention on that one incident and its details, and a larger section of the dream returns, often bringing with it an isolated incident from an earlier dream. Gather as many such incidents as you can, and jot each down in a word or two.

Then take the incidents one by one, recover the story around each, and write a brief outline; finally write the whole record out in full. Note details, as many as possible, and be especially careful with anything that would seem unusual if it happened in waking life, since that is where evidence is most likely to come. Until the record is finished, think of nothing else. Do not trust memory: write. And if you are sure you dreamed nothing, recall instead what you were thinking when you first woke, and the dream it came from will begin to return.

Seen apart from meant

Dunne warns that the dreaming mind tacks false explanations onto what it sees, so the record must keep the appearance apart from the interpretation. In one of his dreams a shower of sparks flew past his ears, which the dream explained as people throwing cigarette ends; he wrote down only the explanation, and so nearly missed the match the next day, when sparks from a log he was blowing with bellows streamed past his face. Write first what was seen, then what it seemed to mean.

Reading the record

Read the records over from the beginning at the end of each day. Dunne's rule for a match is strict. A single ordinary resemblance proves nothing, since a long dream and a long day are bound to share something by chance. What counts is what he called a “double event”: a match in two or more particular details at once, such as a pile of sixpences falling off a red book. He set two days as the limit for a match, stretched only for very unusual incidents, and advised choosing nights before a journey or some other break in routine, since the unusual is easiest to recognise.

His most important rule concerns the mind itself: “The waking mind refuses point-blank to accept the association between the dream and the subsequent event.” It reads straight past the very thing it is looking for. His remedy, which he likened to holding a page up to a mirror, is to pretend that the records are of dreams you are going to have tonight, and to look through the day just past for anything that could have caused them. This, he was careful to say, is a device for noticing and not for judging; the judging comes afterwards, on the strength of the details alone.

Serial time

Dunne explained his results by a theory he called serialism. Time, he held, has more than one dimension. The self that lives moment by moment along ordinary time is observed by a self in a second time, for whom past and future lie open together, and that observer by another, in an endless series. Awake, attention is held to the travelling present moment; asleep, the second observer's attention ranges along the line of time, forward as easily as back, and the dream gathers what it finds there.

Keeping the record fair

A record of forebodings is fair only if every entry is kept. Every foreboding goes in the book before its outcome is known, dated, with what is expected spelled out plainly and a time by which it should happen; the misses are counted with the hits; and a match is judged by its details, not by its feeling. Kept this way, the record shows the true proportion of hits to misses, which memory alone never does. This is Dunne's own discipline carried to its end, and it is the house's way with every premonition.

Living with a foreboding

No foreboding should decide what matters: it is no reason to call off a journey that counts, or to frighten another person with a warning. Where it points at something ordinary that can be checked, a sound from the engine, a window left open, a worry about one's health, check it as on any other day, because the check is sensible in itself. If dread settles and will not lift, take it to a doctor or someone you trust, since persistent dread is a burden that can be eased. And if a loss came after a dream that seemed to foretell it, the dream caused nothing: the grief is ordinary grief, and deserves ordinary care.