Wednesday, August 5, 2009

Falsifiability: Crucial To Science

I want to follow up to a comment I made on Nick's post.

An incredibly important concept in "what makes a good scientific theory" is falsifiability. When I first heard about falsifiability I thought it was interesting but not until recently did I realize how much majesty the concept holds.

First what is falsifiability?:
Falsifiability... is the logical possibility that an assertion can be shown false by an observation or a physical experiment... Falsifiability is an important concept in science and the philosophy of science...

Some philosophers... most notably Karl Popper, have asserted that a... theory is scientific only if it is falsifiable...

According to Popper, falsifiability is a general notion of creditability, even though he admitted that falsification is only method by which scientific theories may be formulated, criticized or refuted at all.
As you go through life you will find many explanations people give for various physical phenomena are fundamentally no different from Nick's "elves cause the lights to work" example. Ultimately, an "elves theorist" cannot find a definite way to falsify the elves theory and so the theory gets thrown out and replaced by the simplest falsifiable alternative: the theory of electromagnetism.

And, we as a human race are better of for it!

Anyways, good scientific theories, the ones that time has shown are very valuable to the human all have one thing in common: they make well defined predictions that are testable and hence the theory is falsifiable.

You can lean with confidence on such theories because, if they were false, somebody could demonstrate it.


  1. Interestingly, religious faith is not falsifiable. It is true-ifiable. Instead of claiming that it has stood up to tests X, Y, and Z, faith is proven true by personal experiences. In fact it makes me nervous when well-meaning people try to "prove" religious beliefs by claiming scientific tests could have shown religious faith to be false but failed. Faith doesn't work that way. Science starts with buts of truth and slowly generalizes until they fit together (e.g. electricity and magnetism pre-Maxwell). Religion starts with the general principle (e.g. God lives or the Book of Mormon is true) and then applys it to specific situations.

  2. On a different note, this is one of the areas where intelligent design fails since no one has ever come up with a way to show that something didn't design the universe.

  3. Nick I agree, with two footnotes:

    1. It is okay for religions to not be falsifiable becuase, for the most part, they aren't scientific thories trying to explain the physical world. For example, morality can never be explained by a scientific theory so theories or morality don't need to be fasifiable. Furthermore, theories of morality have great value for obvious reasons so a theory does not need to be scientific to have value.

    2. To the extent someone's religion is used as a scientific theory, to that degree it should be held to the same standard as any other scientific theory.

  4. These comments are obviously my own opinion, not that of anyone else. I find great value in the LDS church for many reasons maybe I will blog about later. I do, however, hold all scientific statements held by LDS people to the same standards I use for evaluating physical scientific theories I deal with as a scientist. (Lets just say this means many scientific claims by LDS people meat the chopping block. Especially the ones that are not falsifiable.)

  5. PS. For any readers in this position, the LDS church is great and I suggest you join. It will bless your own life as well as you family and posterity and is the right thing to do. (But when you join, still realize the best explanation for primordial fluctuations that led to large scale structure is still the theory of inflation. etc...)


To add a link to text:
<a href="URL">Text</a>