Trench Warfare - Studies vs. “Results.”
Saturday, November 17, 2007 12:39What’s more important - theory or practice?
Just because it “works on paper”, does it “work?” And just because “it works”, does it “work?”
Just what the heck am I talking about?
I’m referring to the so-called Diet Debate on CNN’s The Larry King Show that Gary Taubes, Mehmet Oz, Jillian Michaels, and Andrew Weil engaged in. It was funny enough to watch on the sheer merits of the participants interactions, but one particularly compelling moment arose when Jillian Michaels and Gary Taubes butted heads:
(from the CNN transcript):
“MICHAELS: Gary, if you can show me — Gary, if you can show me one person you have taken 100 pounds off, then maybe we can apply your theory…You are a scientist. You appreciate the fact that science must be applied for a theory to be proven true. Your theory falls short when applied practically because I’ve applied it.“
She’s wrong, by the way. First off, it’s not “Taubes’ theory” - if she actually read the book she’d have seen that it was a review and commentary of all the major dietary studies of the past 70 years or so; all of the research had been done by folks much smarter than her or Taubes. Secondly, his so-called theory (that exercising to lose weight fails), far from falling short, seems to pan out far more often than not (I’m sure all of you know someone - maybe it’s you - that failed to lose any appreciable body fat through exercise, particularly exercising by itself).
All that aside: What’s more important - theory or practice?
The answer: Both. Without science, there’s no intelligent practice - just a lot of random groping in the dark for answers without much rhyme or reason. And in order for a theory to hold true, it has to be demonstrable in real-world situations (as well as exhibiting repeatability in scientific studies). By actively testing theories (through rigorous studies) and applying them, we can come to new, better solutions to our problems. Conversely, if a theory doesn’t bring about real-world results, it should be discarded as faulty.
What’s arguably more important than observing the results of our experiments and applications of theory is how we interpret those results. Witness the second part of the exchange between Taubes and Michaels:
“TAUBES: I’m not a diet doctor here. I’m just trying to say…If you look at the actual evidence when people do clinical trials. Again, she changed a lot of things. She’s changed her diet. She changed the way she ate. She exercised. All of those things might have had an effect.“ (Emphasis mine)
This is why you can’t solely rely on “real-world applications” - you can’t discern the impact any one factor has on results! Is your result an effect of the intervention, or was it something else? This is why performing studies that attempt to isolate single variables is so important. By determining which variables have the greatest impact on results, it prevents us from doing applying things that at best, don’t help us towards our goals, and at worst, detract from our results.
Who knows, if Jillian Michaels implements Gary Taubes’ so-called theories, she might experience even greater success with her clientle.

