Monday, August 13, 2012

How can we tell?

Just a brief post today but a link to something serious to think about:
How are we to know what's 'true' even in science?
We rely on honest and complete reporting, and we even then tolerate routine dissembling and exaggeration or slanting of discussion slanted in an author's preferred direction.  But the facts must not be misrepresented!

We argue that science is self-correcting because results can be replicated.  That's in principle, however, and in practice often not possible.  How could we confirm the Large Hadron Collider's data?  Or what about a GWAS based on 10 international studies that included 250,000 diabetics and 500,000 unaffected controls?

Still, one way or another, important facts do get checked in various ways.  But here, too, we assume honest reporting.

Here is a link to a report by the Union of Concerned Scientists, on how corporate and other political interests corrupt science.  You'll not be surprised by some of it, but may be by other parts of this report.  Of course the UCS is itself not entirely unbiased, and we must take that into account.

We've posted on some of the abuses in this report, but the issues go beyond more or less formal manipulation and corruption.  Many fields have work that cannot be replicated easily if at all, and research resources would not stand much repetition: they're scarce enough as it is.  Even the discussion of a fossil site and interpretation is very difficult to check fully.

When scientists can be corrupted by financial support for their work, not to mention direct bribes, ghost-writing and so on, or when careers depend on 'positive' or dramatic findings, or when huge industries depend on a particular kind of result--or when policy that affects them does--how can we expect other than that there will be some corruption?

But we also have to be wary that we don't define 'corruption' as 'not agreeing with us'!  That some science is corrupted in various ways by bias is correct, and the more subtle or political the worse it is, but bias exists on all sides.   Not to defend dinosaural objectors to the truth of evolution, but we must as scientists accept that deniers, for example who resist some climate change explanations, can sometimes be legitimate scientists who think the 'in' view is itself being pushed for political reasons.  After all, we all have points of view.

What can we do about it?  Will reports such as the UCS one, which documents undoubted abuses for political and commercial reasons, make any difference?

No comments: