Some months
ago, I came across the neologism "retrodict," used
in the sense of describing the causes and nature of past events
for which there is insufficient direct evidence. The word stuck
with me because it aptly captures something of the situation
facing historians as they seek to understand and explain the
past. The term, obviously, draws on the word "predict" and suggests that the relation of historians to the past is
analogous to that of those who predict the future. The analogy
may not be exact given the difference in our relationship to
past and future, but the concept is still a useful one. Predicting
the future is chancy because of our lack of data about it. We
have to infer what we think is going to happen on the
basis of what is happening. The future hasn't happened
yet, so there is no evidence that comes to us out of the future.
The past, on the other hand, has happened, and while we do not
have direct access to the past (it being, after all, past),
we do have a plentitude of records produced in the past.
The differences
between our knowledge of past and future, however, may not be
as great as we think they are. First, we do actually have some "evidence" about the future. Experience and common
sense teach us that certain present actions will more or less
certainly lead to future consequences. When we see those actions
taking place, we have a good idea what will happen. We worry,
for example, about the consequences hard drinking or smoking
will have for a friend. We know when the seasons will change.
We know what to expect when we fly overseas. Second, we also
have contemporary records of the future, strange as it seems
when put that way. We schedule the future, sometimes down to
the second, and generally what we actually do next Monday resembles
the record we already have of next Monday in our appointments
dairy.
If we
stop to think about it, the past is not all that different from
the future. Nearly all of it is as shadowy and unknown as the
future because it has passed from living memory. If we are asked
what we had for breakfast yesterday, we can generally give an
answer, which will almost certainly be correct. But, if we are
asked what we had for breakfast on 17 April 1979, we are at
a complete loss—normally. We might be able to answer the
question, however, if we have some reason to
2
remember that particular breakfast
or if, for some reason, we have a record of what we ate that
morning. Suppose we stopped and thought hard about the question.
We might be able to "retrodict" what we probably ate
by remembering where we lived in April '79, what our schedule
was like, and what our eating habits were like then. We might
be able to answer, "I almost certainly had breakfast cereal
that morning. I always ate cornflakes for breakfast back then."
That is what is meant by "retrodicting" the past.
Past
and future share one essential quality. They no not exist in
the present. We must, therefore, use existing records to infer
their nature. Futurologists predict what will happen, and historians
retrodict what did happen. It may be that the historians have
a somewhat easier time of it, if only because the past is unchangeable
while the future has yet to exist at all. Still, as the above
breakfast example suggests, the study of the past is largely
a study of what probably or possibility happened. We do know
something of what happened in the past, but our knowing is always
incomplete to the point that calling the past "unchangeable"
is somewhat misleading. Our understanding of the past certainly
changes. Historians, in sum, must necessarily "retrodict" the past.
What
does this mean for our knowledge of the past? Does it mean
that we don't know much more about the past than about the
future? No, that's not the case. We have libraries full of
books telling us what happened in the past. Does it mean that
our knowledge of the past is no more certain than our knowledge
of the future? The answer to this question is more difficult.
For one thing, much of what we think we know about
the past is faulty for several reasons. Many influential actions
taken in the past went unrecorded. We can only infer their
taking place because of their consequences. The records we
do have tell only a part of the story, give only incomplete
details. Sometimes those records are misleading, intentionally
or otherwise. We know a lot more about the past than we do
the future, but it is not clear that our knowledge of the
past is all that much more secure than our knowledge of the
future.
Is,
then, our retrodictive knowledge of the past trustworthy?
The answer to this pressing question is, I think, "Yes,
No, and Maybe." The prior question that the critics of
the historian's craft invariably fail to ask is, "Is
human knowledge of even the present trustworthy?"
How much trust, that is, can we put in our ability to know
3
the world around us in all of its
dimensions, past,
present, and future? Is it not
true that largely our knowledge of what is happening around
us in the present is as sketchy as our knowledge of the past
and future? We have to guess at what people are thinking.
We have to try to make sense out of other peoples' actions,
even when they don't seem to make much sense. Important decisions
are made in shadowy places and even though we are affected
we don't know who made those decisions or why. It is one of
our most common experiences that we have to make important
decisions on the basis of insufficient information. It is
a fact that we have to
intradict the present nearly
as much (or just as much?) as historians
retrodict the past and futurologists
predict the future.
Thus,
we have to "dict" our way through all three time
dimensions, past, present, and future. Of the three dimensions,
the only one that is not in and of itself fluid is the past.
It was what it was. In some cases, then, what we retrodict
about the past is actually more trustworthy than what we predict
about the future or intradict regarding the present. Our understanding
of the past changes, certainly, as we gain new data and discover
new perspectives. But, when one considers all of the difficulties
we face in knowing what is "really going on" in
the present, retrodiction seems easier and maybe as trustworthy.
Herb Swanson
Ban Dok Daeng
September 2002
4
