BELIEVE   ME   NOT!    - -     A   SKEPTICs   GUIDE  

next up previous
Next: The Perturbation Paradigm Up: Falling Bodies Previous: Calculating Trajectories

The Scientific Method

One often hears that "the modern Scientific Method" can be traced back to Galileo, who first prescribed the panacea of "Observe, Hypothesize, Experiment and Confirm." This is complete nonsense.6.19

First of all, people have been doing more or less the same thing since before the Dawn of Recorded History;6.20 Galileo just grabbed the headlines when there was first Good Press to get! He was a hero, true, in that he championed the arrogance of thinking for oneself against formidable odds and outlined a procedure for doing it successfully (i.e. getting away with it) for which we all are in his debt. But he could hardly claim a patent on the idea.

Second, Galileo's Scientific Method, like his Mechanics, was an idealization of an imperfect experimental reality. As discussed earlier, we cannot Observe without relying upon our repertoire of models through which we interpret our sense data; the phrase, "Seeing is believing," betrays a profound naiveté if we consider carefully what we know about the retina, the optic nerve and the visual cortex. We may Hypothesize freely, but only the most righteous scientists are actually honest about when their hypotheses were formed - before or after the experiment!6.21 The one part of Galileo's prescription that we truly took to heart was the exhortation to Experiment - i.e. to go directly to Nature with our questions about "what will happen if we...?" Asking such questions in a form that Nature will deign to answer unambiguously is a profound art indeed; a lifetime is too short to learn it in. Finally, Galileo can be considered charmingly naive in his expectation that Experimentation will be able to Confirm any Hypothesis. As Karl Popper has pointed out, there is no logical basis upon which any "general explanatory theory" can be proven correct by any finite number of experiments; the best we can hope for is a Conjecture which is "not yet Refuted" by the evidence, and this is impressive only if there is a lot of non-contradictory evidence!

So the revised version of the "Scientific Method" should read something like this:

1.
Based on a lifetime of experience, form a Hunch.
2.
Using a trained analytical mind, refine the Hunch into a well-posed Hypothesis.6.22
3.
Think of a few Consequences of the Hypothesis that lead to Predictions that can be tested by Experiment.6.23
4.
Perform a Gedankenexperiment6.24 to visualize the results you should expect to get under different circumstances.
5.
Design a real Experiment, if possible, to produce the most clear and unambiguous results6.25 possible.
6.
Descend to the level of grubby sociopoliticoeconomic reality to seek funding, recruit personnel, fight battles for priority, coordinate with engineers, construct several versions of the apparatus (all but the last of which do not work), tinker with balky equipment, coax plausible results out of partially recorded data, argue with collaborators about procedure and interpretation, etc., for as long as it takes to get the Experiment done [which may exceed your lifespan in certain disciplines].
7.
Publish a Result (or Results) - often determined by "consensus" [i.e. politics] among Collaborators - and let the Community decide what it means.
8.
Go back to Step 1, if you did not already do so earlier.
Of course, these are the rules for a Professional Scientist; if you are content to remain an Amateur, the Scientific Method is a little simpler:

\fbox{ ~
{\sl Think for yourself. }
\rule[-3mm]{0pt}{9mm}
}

In all the above arguments, there is an implicit assumption that we usually do not discuss: namely, that there is an "external" Real World independent of our perceptions and models that behaves the way it does regardless of our expectations or observations - that we can, at least in spirit, set ourselves apart from The World as mere observers of its behaviour. Even in Classical Mechanics this is an obvious idealization, but perhaps a conscionable one. In Quantum Mechanics (as we shall see) this basic view of the Experimenter as Observer is challenged at its roots! Nevertheless there are things we can do which seem like Observations and which we will have to use to "pull ourselves up by the bootstraps" if we are to even grasp what Quantum Mechanics has to tell us. So, for the time being, I encourage you to steep yourself in the traditional æsthetic of Experimental Science and try to be as "objective" and "non-interfering" as possible in making (or imagining) your Experimental Observations.


next up previous
Next: The Perturbation Paradigm Up: Falling Bodies Previous: Calculating Trajectories
Jess H. Brewer
1998-09-06