The University of Waikato - Te Whare Wānanga o Waikato
Faculty of Science and Engineering - Te Mātauranga Pūtaiao me te Pūkaha
Waikato Home Waikato Home > Science & Engineering > BioBlog
Staff + Student Login

how to argue against a scientific theory...

... and also, how not to. This is an excellent essay by the Sensuous Curmudgeon. I'll list his key points here, but you really should go over and read the whole thing.

What to do:

  • You could present a fact (a verifiable fact) that contradicts it (this, of course, means that you have to be sure you understand the theory that you are trying to contradict).
  • Or you could put forward your own testable theory - but it will have to be at least as good as the existing theory at explaining all the available evidence.

What not to do:

  • Make an argument from personal incredulity.
  • Argue that the theory fails because there are things that scientists don't know yet - no scientist would claim otherwise.
  • Say something's wrong when all the evidence suggests that it's correct.
  • Argue that the theory is wrong because some scientists have done something fraudulent in obtaining their results.
  • Make ad hominem attacks on the scientists, rather than engaging with their ideas.
  • Present as evidence the opinions of people who don't work in the field in question.
  • Object to the theory because you don't like its consequences.
  • Claim that the scientist later renounced his theory (a claim commonly - & incorrectly - made about Darwin). Even if true, it doesn't invalidate the theory.

In other words - an excellent checklist if you are weighing up a the arguments about a particular scientific theory :-) Again, my thanks to the Curmudgeon.

| | Comments (2)
Share via Email Share this on Twitter Share this on Google+ Share this on Facebook

2 Comments

Alison: I like your list: its nice and clear. I hope you refer to it in debates elsewhere and that your students take it on board.

If it interests any of your students, its strikes a chord as I'm in the middle of considering how to test an idea I have that the current model of how a particular DNA-binding protein bind DNA may be wrong. Its a model rather than a "theory", but the issues are much the same. I have one nice trump card against the existing model, which almost fills the criteria of the first point of "what to do", but its probably not enough on its own to bring down the existing model in others' eyes.

I have my own model already, so to work on the second point of "what to do", I get to show if its consistent with the experimental data and to devise tests of my model to try show it holds up to what I hope are (most of) the points of criticism. To offset getting overly keen on my own model, I prefer to spend as much time as I can trying to defeat my model, on the basis that if I can try and try defeat my own model, but it still stands, then I may have something worth showing others.

Others get to judge if I've got it right, of course!

"... in debates elsewhere..." - I suspect that some of them are a lost cause (thinking of some of Ken's guests); too much sophistry :-)

But I will happily use you as a Good Example, if I may!

Leave a comment