Tuesday, 13 September 2011

Science - from maximum to minimum honesty

*

A major difference between real science (as it was) and scientific research (as it is now) can be stated in the form that real scientists aimed to be as honest as possible, while scientific researchers do not allow their honesty to fall below a minimum level.

*

Real scientists were striving to be as honest as possible - constrained by the self-discipline, time and effort; other people's attention; the demands of bosses and referees and so on.

Modern scientific researchers strive to retain a core of essential truth in their communications - but in no sense do they try to be as honest as possible.

The appreciate that the system requires them to be dishonest, within limits, in order to get jobs, promotions, publications, funding (even in order to retain their status in situations where disagreement with the mainstream is regarded as not so much wrong as unethical).

*

Naturally, this means that real scientists were more honest that scientific researchers... but did this matter to the scientific process?

Was the difference significant?

Does it make any difference whether the average scientist is nowadays, say, 90 percent truthful when in the past the same person would have been 97 percent truthful?

*

Well maybe it does make a difference: maybe it makes all the difference in the world: the difference between science that works and science that does not work.

*

Maybe the combination of numerous persons work done with 97 percent honesty  is still mostly honest, whereas the combinations of work done as 90 percent accuracy has dipped below the level at which it is useful - the proportion of noise to signal overwhelms the specific content?

Maybe, too, once humans beings abandon the iron law of truth, and instead of striving to be 100 percent honest, they begin to allow a certain 'minimum' proportion of dishonesty (with respect to 'inessentials' - merely as a means to the end of necessary career or institutional success)

- but maybe, once you begin using dishonesty expediently, there is no reason to stop at any particular point, no reason to keep the dishonesty minimal; and many reasons incrementally to ramp-up the proportion of expedient dishonesty until...

Yes, I think that's how real science works; why scientific research does not work.

*

9 comments:

Anonymous said...

Outside of a few politically charged fields, would you say this is really true for the *average* scientist?

Bruce Charlton said...

For sure - it is the exceptional and rare scientist that is as honest as they can be all the time - including research grant applications, research plans etc (full honesty used to be the norm, at least in Britain, just a few decades ago). Most scientists will say what it takes to get a job, make career progress, get promotion etc.

The Crow said...

It is a lonely, lonely place to be, when you decide to become completely honest.
Robinson Crusoe, in the midst of Babylon.

Bruce Charlton said...

@Crow - yes, in mainstream society, honest people are regarded with great suspicion, as 'loose cannons' - who knows *what* they might say? Of course, they are not regarded as being honest, but as either crazy persons who should be locked up or at least secluded, or as evil trouble-makers pursuing some kind of attention-seeking agenda.

So far as I can see, the only famous scientists who are honest (which does not mean they are correct, but that they tell it as they see it) are regarded as 'mavericks' and either depicted as mad (e.g. Rupert Sheldrake) or evil (e.g. Peter Duesberg, James Watson). Or at most they are quoted very selectively (e.g. James Lovelock - who is honoured as the father of 'global warming' but the that he says warming is due to methane not CO2, and that we are doomed and human action cannot prevent mass destruction are ignored).

My point is not that these people are necessarily right, or completely right, but that they are basically much more honest than the usual, and they are sidelined.

The Crow said...

Has it ever occurred to you, Bruce, that almost nobody knows anything about anything?
That for all the bluster, apparent confidence, qualifications, and standing, almost nobody actually has a clue?
I hope I'm wrong, I really do, but I often get the sneaking feeling that this must actually be the case.

Anonymous said...

That for all the bluster, apparent confidence, qualifications, and standing, almost nobody actually has a clue?

I used to feel that way until I read Quine's "Two Dogmas of Empiricism" and found out why logical positivism failed.

It's worth the time to read that paper, at very least.

The Continental Op, Ph.D. said...

Bruce Charlton gives it to us straight, insight stacked on insight.

Quine is a formal philosopher.

Need I say more?

Anonymous said...

'Quine is a formal philosopher.

Need I say more?'


Yes. You need to say a lot more, unless you think that identifying someone as a philosopher invalidates everything that person has written.

You could start with an explanation of whether Quine's notion of "underdetermination" is better or worse than Duhem's "confirmation holism." (I would argue that Quine is better.)

Then you could go on to explain whether you think Quine's work supports or detracts from The Crow's notion that no one has a clue.

For an encore, you could show how Quine puts the original post in context.

Nick said...

- but maybe, once you begin using dishonesty expediently, there is no reason to stop at any particular point, no reason to keep the dishonesty minimal; and many reasons incrementally to ramp-up the proportion of expedient dishonesty until...

Your post applies to to so much more than scientific rigor and is spot on. There is a quite interesting book worth a read on the topic of confirmation bias and how we justify our behaviors, -- 'Mistakes Were Made (but not by me), by Tavris and Aronson. An excellent read that spells goes far beyond confirmation bias, and talks about how people make slow, incremental steps to get to a behavior that allows them to side step responsibility for their actions.