When is it science?

There’s a lot of talk these days about people being “anti-science.”  The problem is, a lot of people making those claims either are a bit unclear on the idea of what science is or know full well what it is but are hoping you don’t.  Just because someone calls something science doesn’t mean that it actually is.

First off, science is not a collection of “facts”.  It’s not a set of conclusions.  And it most certainly is not ultimate Truth, forever and ever, amen.

Science is a method.  And the core of that method can be summed up in one simple question:

“How would we know if we were wrong?”

The late Richard Feynman described it this way:


First, we guess what we think our new law will be.  Then we calculate what must happen if that law is right.  Then we compare the result of that calculation with experiment.

And here’s the most important part.  If the calculation from our guess does not match experiment, it’s wrong.  Period.  Yes, there can be experimental error.  Yes, if the data is variable sometimes just from chance you’ll get a result that is atypical.  But once you account for those, once you’ve gotten your measurements nailed down precisely enough  to differentiate from your calculated result, once you’ve got enough measured data for the statistics to say whether it matches calculated results or not, then if they do not match, they’re wrong.  Period.

It doesn’t matter how “common sense” your proposed law of nature/theory/hypothesis (various terms which science uses to label proposed explanations of how the world works) is.  Doesn’t matter how much you want it to be true.  Doesn’t matter how good, or bad, the results will be for you.  Doesn’t matter how many people, how many scientists, say it’s true.  If it doesn’t match experiment, it’s wrong.

The only reason, the only reason to accept or reject some scientific law/theory/hypothesis is whether or not  it agrees with experiment. And any such law/theory/hypothesis is always subject to being amended, or outright rejected, as further data comes along.  Science is never settled.

Let me give you an example.  Back in the early days of optics as a science there were two schools of thought on what the nature of light might be.  One was the “corpuscular” theory, that held that light consisted of really small particles that bright objects emitted.  The other was the wave theory, that light consisted of waves, like sound.  Now, waves and particles behave differently in certain circumstances.  In particular, waves will tend to diffract and interfere and particles will not.

Someone looked at that diffraction and did the math and found that in certain circumstances light, if it were a wave, would behave in ways that was just patently absurd.  In particular it was found that in some very specific circumstances the shadow of a small object illuminated by a point source light of a single wavelength on a screen behind it, certain combinations of size of object, distance to the screen, and the wavelength of light, the shadow would contain a bright spot in its center.  Contrariwise, light shining through an aperture would have a dark spot near the center of the light spot.  This, of course, was completely ridiculous so of course light had to be a particle.

The science was settled.

Then, someone actually found a combination of object and screen distance, paired with monochromatic light (a sodium flame was useful for this, it’s two spectral lines are close enough that it can be treated as a single wavelength for the purposes of many experiments).  And the bright spot in the shadow, the dark spot in the light disk, was there.  Once this was seen, it was utterly clear that light had to be a wave.  Couldn’t be anything else.  Only waves act like that, produce the diffraction and interference that would make that happen.

The science was settled.

And then, once more, experiments started finding oddities.  We learned that the light had to be “transverse” waves rather than “longitudinal waves”:



Bu that led to some puzzling aspects.  If it was a transverse waves, what was “waving”?  Transverse waves aren’t carried through a liquid or gas, but only through a solid (the ocean waves you see on the shore are a different phenomenon and can only happen when there’s an interface between two materials).  Furthermore, experiments in interferometry had given us the wavelengths of light–very, very short wavelengths–and the speed of light suggested that whatever material was “waving” had to be very stiff indeed.  This led to the conclusion that the Universe was filled with something both extremely tenuous but also extremely stiff to allow light to pass through it.  But this material wasn’t dragging on the planets as they circled the sun so it had to be infinitely elastic.

Then folk started finding out other things.  They discovered that light didn’t quite, or didn’t always, act like a wave.  The photoelectic effect, the “ultraviolet catastrophy” of black body radiation (you heat something and it glows, but for a wave, the higher frequencies should carry most of the energy so that instead of glowing red, or even white, most of the energy should be in ultraviolet, x-rays, and gamma–but it wasn’t).

The science was becoming unsettled.

Then a certain Swiss Patent Clerk (I won’t keep you in suspense; it was Albert Einstein) suggested that light was waves that came in discrete “packets” called quanta.  Under certain circumstances they behaved as waves.  Under others, as particles.  This was the foundation of what is now called Quantum Physics.

And the science is settled.

This Time For Sure.

Or until someone else comes along to unsettle it with some experimental results that just don’t fit.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s