Common Misconceptions about Science Basics and the Multiple Methods of Science

July 27th, 2011

Continuing to think about science this week…

A lot of people think they know what science is, and some of them are quite smart and generally well informed, but not on this topic.  Let me give an a common misconception:

Many people think that there is a progress of scientific knowledge that progresses from hypothesis to theory to Law.  So, a theory is more likely to be true than a hypothesis, but until it’s really rock solid the theory doesn’t graduate to “Law” status.  Well, that’s just screwy.  A hypothesis is a generally a testable explanation for an observed phenomenon (e.g., the gravity that makes things fall on Earth makes planets move in their orbits).  A theory, or a “Scientific Theory,” is a well-developed hypothesis that has passed so many tests in so many situations that it’s about as close to the truth as we’re likely to get — at least pending a push into new realms (e.g., Einstein’s theory of General Relativity which continues to pass exacting tests and superseded Newtonian gravitational theory which fails in the limit of large masses).  A “Law” is just a mathematical relationship that works very well to describe a physical relationship, but may not be correct in every situation (e.g. Newton’s Law of Gravitation or the Ideal Gas Law).

OK, let me get to something a little more subtle.  A lot of people think that there is something called “the scientific method” and there’s only one such method that is scientific.  It goes something like this:

1. Make observations.

2. Make a hypothesis regarding relationships noted in the observations.

3. Carry out an experiment in a lab to test the hypothesis (usually where all factors but one under investigation are held constant).

4. If the hypothesis hasn’t been falsified, continue to test it with more experiments.  If it has been falsified, modify it or develop a new testable hypothesis.

And so on…

In reality, there are entire scientific fields in which this is impossible to do, or nearly so.  Astronomy is one of them.  Astronomy is what we call observational science, primarily, and the subfield of laboratory astrophysics isn’t a huge part of it.  I can’t run an experiment on a quasar to figure out how it works.  I can only watch it.  Oh, I can watch it in ways so strange and obscure that I might as well be a wizard, but I’m really just watching it, collecting light in one way or another.  The closest I can do to an experiment is testing a hypothesis by making some new observations that might might challenge it.  It feels like an experiment, but again, I’m just collecting some light and can’t control my subject at all.

Astronomy and many other fields also use computational methods.  Computer models have helped us understand a lot of things better, and are an invaluable tool in science.  With a computer model, you can vary parameters and see how the system behaves.  When the computer model matches observations and is based on understood physics, it often has good predictive power.  This is one way we study stellar interiors and evolution, for instance.  I’ve started hearing criticism of climatology in some quarters based on the notion that “writing a computer program” isn’t scientific.  That’s crap.  You test the models against reality, adjust them, improve them, and at some point their consistency and predictive power usually gets really good and you’ve got something akin to a traditional scientific theory, all without actually conducting an experiment in a lab.

Another area where it’s hard to do proper experiments is on human subjects.  Ideally a scientist could conduct proper experiments (double blind, randomized studies), but there are often ethical concerns.  In these cases sociology and medicine can start to resemble astronomy.  You get to watch, and hope you can draw some meaningful conclusions later.  Often it’s super complicated and the results are of limited value, which is why contradictory news stories come out all the time.

I don’t think science is really something with a single, clear method.  It is rather a system of rigorously testing ideas through observation and experiment, keeping ideas that don’t fail and have supporting evidence, and gaining trust in them over time.  There are a lot of quality-control elements like peer review, open publication of methods and data, consideration of uncertainties on measurements, and the like.

Some people seem to regard science as just another belief system, no different than religion or political persuasion.  Those people are silly and misguided.  Sure, there are assumptions in science and they are something like this: reality has consistent rules  and these can be explored through measurement and observation.  Since there’s a lot of evidence to support that statement, I’m not just choosing to follow a belief system, I’m letting the way things actually seem to work guide my beliefs.  Some of the silly people thinks that this neglects the supernatural, and it does.  If the supernatural has any measurable effect in our reality, it can be measured and studied by science, and if it has no effects that can be observed and measured, it doesn’t exist in any meaningful way and can’t have any effects on us.

So what’s the point of science?  I think there are misconceptions about that, too.  Some seem to think that it’s about making our lives easier, or making better technology to make more money, or to attack their own personal biases (religion, politics).  It’s just about developing reliable knowledge.  That’s it.  Knowledge is valuable.  Knowledge is power.  We’ve done things good and bad with our scientific knowledge, but not pursuing science and scientific knowledge is settling for ignorance, or worse: misconceptions, superstitions, and outright lies.

To conclude, science is messier than many people realize, but there’s a system of reviewing and testing that reduces mistakes and biases and eventually leads to better answers with ever increasing reliability.  Common sense, revealed knowledge, and wild-ass guesses are simply not always reliable.

Ask for the science, and don’t settle for anything less.

Share/Bookmark

Tags:

You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.