Back to An Experiment With Time Table
of Contents
THE EXPERIMENT (1917)
The
following is from J.W. Dunne's original Experiment published in 1927.
Farther down the page is the New Experiment, a revision to the original, which
was published in the appendix of the 3rd edition of An Experiment with
Time. To skip to the new, revised experiment please click here.
J.W. Dunne
feared being labeled a freak. He would have much rather have discovered
himself to be a 'medium' because with that label, he'd at least have
company. Little did J.W. Dunne know that he was neither clairvoyant nor a
freak. Many people share his extraordinary fault in relation to reality,
something so uniquely wrong that it compels them to perceive, at rare
intervals, large blocks of otherwise perfectly normal personal experience
displaced from their proper positions in Time (i.e. precognition). The
unfortunate thing about precognition is that it is generally only known and
believed by those whom are effected directly by it. Since J.W. Dunne was
able to convince himself that his precognitions were valid, he saw that there
was a remote possibility that, by employing this piece of curiously acquired
knowledge as a guide, he might be able to discover some overlooked peculiarity
in the structure of Time itself.
Time had
always, by science, been treated as a forth dimension. What J.W. Dunne's
precognitive dreams had shown was a displacement in that
dimension. He cared not whether Time were a form of thought, or an aspect
of reality, or compoundable with Space. What J.W. Dunne set out to
discover was how it got mixed.
In January of 1917, J.W. Dunne was at Guy's Hospital and
had a dream. In this dream, he was reading about 'combination
locks'. The next day, he was reading a book and came across a reference
to such a lock.
A few days
later the great Silvertown explosion occurred, shaking the whole building,
breaking windows, and causing nurses to extinguish the lights, on supposition
that Zeppelins were overhead. Such an experience is sure to make one
dream. In fact, J.W. Dunne did dream about this event, but on the wrong
night ---the night before the experience. After the disaster, J.W. Dunne
told a fellow convalescent of this precognitive experience. This other
person also had dreamed the same dream.
Taken
separately, each of these dreams appeared wild and extreme; but
considered together, they justified a little closer attention.
The dream
about the combination lock shows that dream pre-images are connected, not only
with exciting and dramatic events, but also with trivial matters. Exactly
as dream images of past events are connected just as often with unimportant
events as well as experiences that are more striking. Because the subject
matter of precognitive dreams is not always dramatic, it is quite possible that
people have failed to notice their connection with subsequent related
events.
What about
the de-ja-vu, the curious feeling which everyone has now and then experienced,
when suddenly one has a fleeting, disturbing conviction that something which is
happening has happened before? Could this be simply that they had
dreamed the event before hand and subsequently forgotten about it?
What about
those occasions when, receiving an unexpected letter or telephone call from a
friend whom one has not heard from for a long time, one recollects having
dreamed of them during the previous night?
What about
all the dreams which, having been forgotten, are suddenly, for no apparent
reason, recalled later in the day? What is the association which
recalls them?
What if
one was to make the unscientific leap that these phenomena were not abnormal,
but normal? This train of thought is similar to the
irrational rationalism described by Robert Anton Wilson in his book, "The
New Inquisition" ---but this is a different matter. Assuming that
this were the case then dreams in general, all dreams, everybody's dreams were
composed of images of past experience and images of future experience
blended together in equal proportions. It might be possible that the
universe was, after all, really stretched out in Time, and that the 'future'
part was inexplicably missing was due to a purely mentally imposed barrier
which existed only when we were awake. But, this abrupt recoil is
illogical and could not possibly be the explanation ---or could
it? If so, the real question would be: What was the barrier
which, in certain circumstances, debarred the individual from that proper and
comprehensive view.
If
precognitive dreams were a inherently a normal thing, not just in the
individual but in time itself, then if one could devise an experiment to
overcome the two initial difficulties of remembering and associating,
the phenomena may prove to be directly observable by a very large number of
people. The difficulty of remembering is easily overcome by means such as
keeping a dream journal by one's bedside. However, the difficulty of
associating dreams to their related chronologically definite past and/or future
event may prove an insurmountable task. Dreams are simply too complex to
allow such connections to be accurately traced on a regular basis.
* * * *
Once J.W. Dunne finally convinced himself there may be merit to his
outrageous normality hypothesis, he decided to test it out on
himself. According to this theory, he should have precognitive dreams,
not just every year or so but, at all intervals, unknown to himself. In
the past, J.W. Dunne only remembered a dream perhaps one night out of
ten. He knew that 'dreamless sleep' is an illusion of memory.
Therefore, he found tricks to exercise his recall ability (there are numerous
techniques to do so). Each morning J.W. Dunne recorded his dreams in a
journal. He then set out to trace the correlations between dream content
and waking events, particularly those events which occurred within two days
before or after the dream. But, the duration between the dream and the
waking event might be extended in ratio to the oddity and unusualness of the
incident; that would be a matter of judgment.
Because
the possibility of satisfactory identification would depend mainly upon
unusualness in the incident, the worst time to choose for the experiment would
be the period when one was leading a mundane life with each day being very
similar to the day before. The foregoing describes the conditions
that J.W. Dunne laid down for the test:
The
dreaming mind is a master-hand at tacking false interpretations on to
everything it perceives. For this reason, the record of the dream should
describe as separate facts, (a) the actual appearance of what is see, and (b)
the interpretation given to that appearance . . .
The
waking mind refuses point-blank to accept the association between the dream and
the subsequent event. For it, the
association is the wrong way around . . . The result is that, on reading
over the record at the end of the succeeding day (or two days), one is apt
to read straight on through the very thing one is looking for, without even
noticing its connection with the waking event.
The best way to circumvent this initial failure is to read
the dreams while pretending to one's self that the records one is reading are
records of what one will dream during the coming night. Then, one should
look over the events of the past day and see what events might legitimately be
regarded as the causes of those dreams. This is not meant to assist one
in judging the events, merely to enable one to notice them. If one is to
attempt judging those correlations, one should do so with no thought to Time
and order.
The
chances against the effects being coincidence must be statistically based upon
these two factors:
(1) The oddity of the individual effects.
(2) The frequency of their occurrence.
After conducting
this test on himself, J.W. Dunne concluded that the nature of the correlations
between his dreams and corresponding future events were so statistically
unlikely that they could not be described as pure coincidence. Even
though this conclusion entailed a partial collapse of the classical theory of
Time, J.W. Dunne was bound to postulate precognition as a working hypothesis
and thus scientifically possible. The nature of Time apparently allowed
the observer a four-dimensional outlook on the universe. This does not
sound so far fetched if one is to take into consideration that if modern
science believes the universe to have at least four dimensions, then it can not
dispute the possibility of observers in the world being similarly four-dimensional.
However, if each of us were four-dimensional observers, this would entail that
everyone possessed precognitive faculties ---This does not entail in turn that
everybody utilizes these abilities. If the ability is widely
distributed among individuals then all J.W. Dunne must do to prove his
theory is take average people and help them recall dreams and realize their
inherent abilities.
If the
theory of normality is right and the faculty which dreams of future events is
the same faculty which dreams of the past, then we can not expect resemblances
to the future to be any more striking than resemblances to the past.
J.W. Dunne
then proceeded to test his theory on other people, many of whom at first
claimed to rarely, if ever, recall dreams (many examples are listed in his
book). The end result is that his theory of normalcy held true with every
individual whom he experimented with.
* * * *
Why only in dreams? It is believed that one may cross the barrier
between memories of past and memories of future occurrences by other means such
as concentration and meditation. J.W. Dunne did not explore this question
in great detail.
THE NEW EXPERIMENT (1943)
The following is from an appendix of J.W. Dunne's 3rd edition of An Experiment With Time in which J.W. Dunne proposes a new, revised experiment. To jump back up to the original 1927 experiment please click here
.
The
primary drawback of the 1917 experiment was that it demands a great deal of
time at the one period of the day when virtually nobody has time to
spare. To recall and record the details of dreams which have occurred can
take as long as twenty-five to forty minutes. One could go to bed early
and set their alarm clock to wake them up, but few would care to undergo such
measures in order to conduct this experiment on themselves.
Another problem is that many people find the nature of the experiment itself to cause mental fatigue. It simply requires lots of concentration and determination.
The net result of these problems is that most people who start out enthusiastic tend to drop out on the third day and others are satisfied with quitting after they have obtained the first minor result.
Clearly there must be a means for limiting the number of nights required of each experimenter. Unfortunately, this is not easy as the first three days are more or less merely training for the experiment. During these first few days, one is building up recall ability and learning what sorts of things to look for that might constitute a congruity with waking events (past or future). Therefore, the real experiment should not begin until the forth day.
Suppose we were to get 1,400 people to experiment for 3 nights, producing 3500 records. Can we assert that this is the same as 300 people experimenting for 14 nights? Either way, an average may be calculated, but the results would be very different. In the first place, the best results (those obtained between the eighth and fourteenth days) would be missing in the mass experiment done for only 3 days. Furthermore, since recall and recording skills are developed over time, the mass 3 day experiment would yield results very different than the 14 day experiments on the average man.
Regardless of the experiment size and duration, the goal is to discover the
proportion of effects suggesting precognition to similar effects suggesting
retrospection. Only when we realize this will we have a new
scientific experiment.
* * * *
Given: P = Past, F = Future;
If the
theory of normalcy is proven, we should expect to discover among the
F-resemblances a number of coincidental resemblances which have occurred among
the P-resemblances. Of 1000 discovered P-resemblances, we saw that 100
were probably coincidences. We should then expect, by ordinary laws of
chance, to discover 100 F-resemblances of similar value; and this is precisely
the number that J.W. Dunne found. It follows that there are not the
smallest grounds for supposing that those F-resemblances involved anything
beyond expectable coincidence. The results imagined would, therefore,
support the old theory of dreams (see top of page).
Given also: i = coincidental
resemblances, a = all resemblances, n = number of
P-resemblances, N = number of F-resemblances
i/a = ratio of coincidences to all resemblances.
N = (probably) n x i/a
For example, if N = (probably) 500 x 1/100 = 5 then the old theory of dreams is right.
We can of course read this formula as:
n = a N
This experiment has the following advantages:
(1) The number of records made is immaterial and
may vary with different individuals taking part in the same experiment.
(2) The period, divided equally into past and
future, in which results are sought for, may vary with the different
individuals.
One
difficulty arises which must be dealt with before proceeding further. A
person dreaming of a past scene or person writes down afterwards, 'saw so-and
so', or, 'was at such-and-such a place.' and one must assume that this also is
recollection if one did not see the scene or person after the dream.
* * * *
Members of
the Society of Psychical Research suggested that J.W. Dunne raise the question
of the effects of the Age factor, pointing out that youth should render ideal
for the experiment. This assumption was made based on the proportion of
the individual’s life which was yet to have passed. Therefore,
there was a possibility that youths would have a higher proportion of
precognitive dreams than the elderly. While this assumption has yet to be
proven, J.W. Dunne did round up a group of 22 volunteers from the University of
Oxford, each of whom was asked to make 21 records (only 14 was actually
needed). Because of the poor scheduling of this experiment, all but two
students dropped out of the experiments due to stress related to final
examinations on campus. This left only 2 students who completed the
experiment. Therefore the results are inconclusive. The period
searched for waking incidents extended for two-and-a-half months both before
and after each dream. The content of these dreams and their corresponding
waking events are detailed in the appendix of the third edition of J.W. Dunne's
book, "An Experiment with Time."
Later,
in a less controlled experiment, 6
_______________________________________________
|Subject|#Records| P-Resemblance| F-Resemblance|
| | |Good|Mod|Indif|Good|Mod|Indif|
|_______|________|____|___|_____|____|___|_____|
| A | 6 | 0 | 1 | 1 | 1 | 0 | 2 |
| B | 9 | 0 | 0 | 0 | 0 | 0 | 0 |
| C | 12 | 0 | 1 | 0 | 0 | 0 | 0 |
| D | 7 | 0 | 0 | 0 | 0 | 2 | 0 |
| E | 21 | 3 | 5 | 0 | 3 | 3 | 4 |
| F | 16 | 0 | 0 | 0 | 0 | 0 | 0 |
| J.W.D.| 17 | 2 | 1 | 0 | 1 | 1 | 3 |
|_______|________|____|___|_____|____|___|_____|
|Totals | 88 | 5 | 8 | 1 | 5 | 6 | 9 |
|_______|________|____|___|_____|____|___|_____|
For the classical theory of dreams to hold true, the P-resemblances should have substantially outnumbered the above recorded F-resemblances. There is some hope that by extending the experiment further that the P-resemblances would be increased to some extent. But, the F-resemblances have a far enough head start so as to show little chance of being overtaken by the P-resemblances. It should be noted that this experiment has too small of a sample set to be considered valid statistical proof of J.W. Dunne's theory of normalcy. However, this test set does show that there is a need for further studies to be conducted on these phenomena.
Proceed to Temporal Endurance and Temporal Flow
