It is a grave mistake to theorize before one has data

Is it a "grave mistake" to theorize without data, as the speaker contends? I agree insofar as
to theorize before collecting sufficient data is to risk tainting the process of collecting and
interpreting further data. However, in a sense the speaker begs the question, by overlooking
the fact that every theory requires some data to begin with. Moreover, the claim unfairly
ignores equally grave consequences of waiting to theorize until we obtain too much data.

In one important respect I agree with the speaker's contention. A theory conjured up without
the benefit of data amounts to little more that the theorist's hopes and desires--what he or she
wants to be true and not be true. Accordingly, this theorist will tend to seek out evidence that
supports the theory, and overlook or avoid evidence that refutes it. One telling historical
example involves theories about the center of the Universe. Understandably, we ego-driven
humans would prefer that the universe revolve around us. Early theories presumed so for this
reason, and subsequent observations that ran contrary to this ego-driven theory were ignored,
while the observers were scorned and even vilified.

By theorizing before collecting data the theorist also runs that risk of interpreting that data in
a manner which makes it appear to lend more credence to the theory than it actually does.
Consider the theory that the Earth is flat. Any person with a clear view of the horizon must
agree in all honesty that the evidence does not support the theory. Yet prior to Newtonian
physics the notion of a spherical Earth was so unsettling to people that they interpreted the
arc-shaped horizon as evidence of a convex, yet nevertheless "flattish," Earth.

Despite the merits of the speaker's claim, I find it problematic in two crucial respects. First,
common sense informs me that it is impossible to theorize in the first place without at least
some data. How can theorizing without data be dangerous, as the speaker con tends, if it is not
even possible? While a theory based purely on fantasy might ultimately be born out by
empirical observation, it is equally possible that it won't. Thus without prior data a theory is not
worth our time or attention. Secondly, the speaker's claim overlooks the inverse problem: the
danger of continuing to acquire data without venturing a theory based on that data. To
postpone theorizing until all the data is in might be to postpone it forever. The danger lies in the
reasons we theorize and test our theories: to solve society's problems and to make the world a
better place to live. Unless we act timely based on our data we render ourselves impotent. For
example, governments tend to respond to urgent social problems by establishing agencies to
collect data and think-tanks to theorize about causes and solutions. These agencies and
think-tanks serve no purpose unless they admit that they will never have all the data and that
no theory is foolproof, and unless timely action is taken based on the best theory currendy
available--before the problem overwhelms us.

To sum up, I agree with the speaker insofar as a theory based on no data is not a theory but
mere whimsy and fancy, and insofar as by theorizing first we tend to distort the extent to which
data collected thereafter supports our own theory. Nevertheless, we put ourselves in equal
peril by mistaking data for knowledge and progress, which require us not only to theorize but
also to act upon our theories with some useful end in mind.