Hypothesis and Theory

Just as my discussion of “TRUTH” was not intended to compete with Aristotle or modern thinkers. My explanation of “hypothesis” is only intended to clarify the use of the word in science.

A hypothesis is a possible explanation for something, for some phenomenon. The word is derived from the Greek word for “to suppose.” For example, if I say that unemployment is caused by foreign trade, that is a hypothesis. If I say that the cure is higher tariffs, that’s also a hypothesis.

In everyday use, people often confuse “hypothesis” with “theory.” The words mean very different things. While a hypothesis is a proposed explanation for something, a theory, in scientific use, is a complete and rigorous system of ideas that explains a large class of observations.

For instance, the theory of Newtonian mechanics fills whole books and includes gobs of equations that, when used properly according to the theory, explain why tops spin and planes fly and what will happen if we launch a rocket of a certain weight, thrust, shape, etc.

Theories have a domain of phenomena that they explain. So you don’t use the theory of Newtonian mechanics to explain magnetism or optics, although you might try just to see how things work out.

Sometimes there are multiple theories to explain the same phenomena and we use which ever one is simpler. So, in optics, we can predict how a lens system will work using the theory of geometric optics. But we can also explain it using wave theory. Which one we choose depends on the situation and what’s simpler. Geometric optics is simple enough for a youngster to understand and it gives pretty good predictions for simple lenses. But, if the lenses are coated or compound lenses or if they have more complex shapes, geometric optics is difficult  to use, if useful at all. However, wave optics, though much more complicated mathematically, is more flexible for a complicated system and is the right choice to predict the performance of a complex, real-world lens system.

Does that make one theory “right” and the other theory “wrong?” No. Because the “truth” of the matter, to refer to my last post, is that scientists can explain that the two theories are just different formulations, different mathematical structures, that are really related at some level. More importantly, in the domain where both theories can be used, they do not contradict each other. If they did, one or both would have to be modified or rejected.

Einstein extended the theory of Newtonian mechanics by asking what happens when objects travel at near the speed of light. Contrary to popular belief, he did not invalidate Newtonian mechanics, he just showed how you had to change the math for high-speed objects. His theory of relativity devolves into Newton’s theory at slow speeds. So, just as in the optics example, we use the simpler Newtonian concepts for everyday calculations and Einstein’s mathematics when we are examining space travel and atomic physics. And, just as in the optics example, the two theories do not produce contradictory results at slow speeds where they both apply.

When skeptics say that “well, evolution is just a theory,” they are misusing the term “theory.”  They are seriously confusing “theory” with “hypothesis.” I may deal with evolution another time, but the point is, that evolution is a very, very large body of ideas with rigorous connections to each other, not a passing thought, a hypothesis, about a single phenomenon.

Sometimes your hypothesis is related to a theory, often one you don’t even identify. So when one politician proposes more government spending to end a recession and another proposes tax cuts, their proposals don’t just stand alone but have vastly different pedigrees, vastly different economic theories behind them.

The proposal to increase government spending is a prescription based on Keynes. The proposal to cut taxes is based on neoclassical economic theory and supply-side economics. You cannot realistically argue the prescriptions unless you are willing to argue the basic theories. That’s difficult because first, you have to be conversant with both theories and, second, you have to find clear experimental evidence to refute or support a theory. That’s not easy in the social sciences where ruling out extraneous effects is difficult or impossible.

So if I argue for the neoclassical tax cut and you argue for the Keynesian government spending we will almost never come to a common conclusion regarding which theory is correct because the evidence tends to be so murky, except for my theory, of course. Take another look at Richard Feynman’s video in the last post.

That last crack wasn’t just a joke. Because the biggest problem we have when arguing competing theories is that big stakes are riding on which theory wins. If my neoclassical ideas win, whole barge-fulls of bureaucrats lose their jobs. Politicians have trouble buying votes, and some economic departments close. That means that a dispassionate search for truth is almost impossible unless we are scrupulously honest and careful. But it’s pretty hard to find a dispassionate, unbiased, bureaucrat or politician.

Finally, if you want to test a hypothesis you can do an experiment. But you can also refer to a well-understood theory. If I claim to be able to jump a river with a motorcycle I can test my hypothesis by building a ramp and trying the jump. The results may be inconclusive. But if I do a little math based on Newtonian mechanics I might find that I need to travel at 800 mph to make the jump. That’s more conclusive. I can still try the experiment but I would run the bike by remote control based on the calculations.

I’ll go into some of the theories I’ve mentioned another time but, just for now, keep the notions of hypothesis versus theory clear. You’ll need to for the next post.

Advertisements

One Response to “Hypothesis and Theory”

  1. Obamanomics 1 « Liberty Physics Says:

    […] demonstrates a disparity between a forecast and an actual result, or, put another way, between a hypothesis of what the economy needed and an experiment. It clearly demonstrates that the hypotheses was […]

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


%d bloggers like this: