Monday, December 26, 2011

Are What We Call "Causes" Just Stories That Help Us Make Sense Of Things?

I've been saying "I don't know" a lot more.  The more I read, the less I feel I know, and the more confused I get. I'll admit. And the more amazed I am at people who know.  But Jonah Leher, in his article:

Trials And Errors: Why Science Is Failing Us , Wired, 16 December 2011

... says that I'm probably not alone. That trying to understand all the minutia may end up confusing me:
"We assume that more information will make it easier to find the cause. ... All those extra details end up confusing us; the more we know, the less we seem to understand."
Why? Because:
"The variables cannot be isolated. Such situations require that we understand every [Leher's emphasis] interaction before we can reliably understand any of them."
I visualize this as a fabric. When you pull one thread, you perhaps unintentionally tug or distort hundreds of other threads. The whole is not simply the sum of its parts - a fact that modern science best not lose sight of:
"This assumption - that understanding a system’s constituent parts means we also understand the causes within the system ... defines modern science. In general, we believe that the so-called problem of causation can be cured by more information, by our ceaseless accumulation of facts. Scientists refer to this process as reductionism."
He gives some examples:
  • The notable costly failure of Pfizer's heart-saving drug torcetrapib. Torcetrapib was supposed to raise HDL and lower LDL. It did, but it also raised blood pressure (the unintended thread tug).

  • We thought the cause of low back pain was disc abnormalities, which we could view in MRI and which could be rectified with surgery. But was it the surgery or the person's own healing mechanisms that relieved pain, since "about 90% of people with back pain [get] better within six weeks"? A 1994 NEJM study suggested it was probably the person's own healing: "the discovery of a bulge or protrusion on an MRI scan in a patient with low back pain may frequently be coincidental."
So, are what we call "causes" (as opposed to correlations) just stories that help us make sense of things? I bet that's often the case, even though some "causes" are indefatigable, like, say, that a deficiency of vitamin C causes scurvy.
Thanks to Shaun for the article.
Thanks to xkcd for the comic.


caulfieldkid said...


I think you could really get into philosophy. It's worthy of note that he cites Hume several times.

This really goes back to epistemology. I'm beginning to notice that science and philosophy are close relatives. It's no secret, but I just haven't really thought much about it in the past.


Laurie Endicott Thomas said...

Kant dealt with this problem in his Prolegomena to Any Future Metaphysics.

If you want to understand how epidemiologists establish causality, look at this article:

Bix said...

"I'm beginning to notice that science and philosophy are close relatives."

I'm beginning to see that too.

Setting a broken bone or giving fluids for dehydration ... that's pretty straightforward medicine to me. Prescribing a drug for, say, porous bone, coronary plaque, or depression ... that's almost philosophical medicine!