Wednesday 6 February 2008

Lemma 3: science is objective

I made the distinction in the previous post between 'subjective' and 'objective' when judging the merit of an idea. A valid question is whether a judgement made by humans can ever be truly objective. Humans are fallible and they have selfish agendas and moral beliefs and all manner of prejudices and tastes. An individual scientist can clearly make errors of judgement and can be influenced by his or her own beliefs. Why does an assertion that has been scientifically tested have more value than one that has not?

The answer is in the title. Individual scientists are not necessarily objective, but science and the scientific method are. There is no magic to the scientific method: you come up with an idea, you present that idea in an understandable and testable form, you test the idea against observation and you allow anybody else in the world to test it as well. Science is a group activity open to anybody who has the capacity to learn the language and the methods. There are no sacred cows: any idea can be challenged.

Now ... the scientific community (like society in general) is fairly conservative. When somebody comes along with an observation or a model that challenges the current consensus, it will be greeted with scepticism. But the history of science contains a huge number of minor and major revolutions. Better ideas always win in the end. Our current model of the universe is quite clearly better than it was a century ago.

A conspiracy theorist could argue that the entire scientific community is either deliberately or accidentally misleading itself. But there is nothing unique about scientists and nothing in the scientific method to exclude a specific group or belief. This argument is essentially saying that the entire human race is misleading itself. If you think that science has made a grave mistake, then politely knock on the door and put them right. Copernicus did it. Einstein did it. It is not a closed shop.

Are scientists 100% objective? No. Is every scientific conclusion 100% objective? No.

Are ideas that have been posed and tested via the scientific method more objective than those that have not: absolutely yes.

12 comments:

Moon GT said...

Hello.
I've been meaning to post some comments here for a while, but I just got round to getting one of these here new-fangled Google accounts... some of the stuff I want to comment on is way down the page now so I might have to wait until the subject comes up again.

On this particular post though, it reminds me of a time I was at a mad hippy festival (for some reason) a couple of years back, and a man had set up a tent full of a load of stuff about aliens and how scientists were lying and deceiving us because what they were saying was contrary to some inscriptions on the pyramids or something, which aliens had obviously done because they looked like spaceships. I asked him how he knew that it wasn't actually the aliens who were trying to deceive us, to which he responded, in shock, with "...well, that's just ridiculous!"

It occurs to me that aliens would have far more reason to deceive us than would our own best minds, if they wanted, for instance, to enslave humanity at some point in the future. It also occurs to me that we only get our idea of what spaceships look like from science FICTION, which has also in turn probably been influenced by ancient artifacts.

I don't know what any of this has got to do with the existence of God, but it's interesting to see how some people's minds work, and that they will find a justification that seems logical to themselves for pretty much anything they want to believe. It's also important that when one thing looks like another, it does not follow that A resulted from B any more than that B resulted from A, or that both resulted from C, or that it was just a coincidence. Newspaper science columnists ought to be more aware of this sort of thing as well.

Rich Lancashire said...

Haha, top science by pyramid boy.

The scientific method is objective; whatever its weaknesses, it's pretty ruthlessly objective. That scientists aren't, though, might be a bigger stumbling block to scientific understanding and advancement than this article gives credit.

Scientist buy in to paradigms, they're what makes science a workable proposition. The data that come out are objective, and interpreted semi-subjectively, but what may be more important is the data that's not collected, or even controlled as the paradigm deems it unnecessary.

Of course, science has the grace to admit it's wrong when it finds out, just saying that scientific understanding is objective data subjectively collected and interpreted.

Sam said...

I just read an interesting article on here that's relevant to paradigms and paradigm shifts.

There's also another 'flaw' in (some?) human beings that balances the subjectivity and conservatism: people quite like to upset the apple cart. It is at least as satisfying to find a hole in a solution as it is to add more evidence in support of it.

In extreme cases this leads to nutjobs like the chap I mentioned in another comment who thinks he's undermined the last two centuries of physics. But in moderation it also leads irritating new generations of scientists to challenge the conventions of their predecessors.

Rich Lancashire said...

I read Kuhn recently, I'd recommend it; it's a good book that's been misused by a lot of subjectivist po-mo intellectual milksops to defend things that it really wasn't saying.

I've also just read this about an anomaly that really isn't. Maybe I should get over to timecube to bust a few paradigms, just for balance.

Robin Johnson said...

I got as far as "seemingly contradicting Darwin's evolutionary theory" and had to stop. The trouble is this makes me a big fat dogmatist, of course.

Moon GT said...

The thing about something that "seemingly contradicts Darwin's evolutionary theory" is that it's pretty much equivalent to saying "I can't immediately see how that feature evolved". It doesn't mean it didn't though, you just haven't worked it out yet. The world is full of puzzles like this, and when Darwinism was new, there were loads of things it couldn't immediately explain, and they had to sit down and think about it for a hundred years or so, and they're still not finished.

The thing is, this sort of dismissal leads one to wonder about just how falsifiable evolution is, if anything that apparently contradicts it can be brushed aside with an "we haven't finished yet!" Supposing one day someone stumbles across a fossil of an animal that actually WAS created directly by God. How would we know? They'd just keep on expecting they would find its ancestor one day, or come up with some other explanation that wasn't to do with God at all. Or, more likely, they'd arbitrarily ascribe it a speculative ancestor on the basis that it looked a bit similar, which is pretty much what they do, and then something else turns up and they have to fiddle it round a bit. We don't have a complete lineage of everything right back to the first microbe, but we accept that there is one, because it's the only naturalistic explanation there is that makes any sense at all.

Put short, the only alternative to the claim that all life on earth evolved naturally (i.e. creationism) is in principle unverifiable, which means the counter-claim is in principle unfalsifiable. But I don't think that necessarily makes it invalid.

Moon GT said...

Another issue with the objectivity of science comes into play when research is funded by an organisation with a specific agenda. In particular, the pharmaceutical and food industries aren't really interested in finding out if their products are safe or not, they want to prove that a particular drug or additive they have concocted is safe so that they can sell it. Some people are skeptical of the idea that any intelligent person would actually DO that sort of thing, but I think that's naive. That sort of money seems to turn ordinarily decent folk into borderline psychopaths.

I'm not saying the actual scientists who conduct the studies are in the wrong, but they can only work with what they're given, and how their reports are edited after they submit them is out of their hands. This isn't really a complaint against science though, it's more that just because something is presented as scientific by an authority doesn't necessarily mean that it is. At the softer end of science (if it can be called such), government agencies use all sorts of statistical tricks to manipulate public opinion.

Am I paranoid? Maybe. But it's pretty easy to trick the average man in the street with figures and technobabble, which is especially potent when anyone who does understand science or statistics and points out how rubbish a particular study is gets immediately labelled by the mob as a paranoid conspiracy theorist.

Rich Lancashire said...

Haldane gave "fossil rabbits in the Precambrian" as a criterion for the falsifiability of evolution.

The discussion is a good example of the paradigm theory. Anomalies are looked at, explained if possible, if not they rattle around for a while. People like Robin dismiss them. :) Eventually someone comes up with a new theory that accounts for the old theory *and* the anomalies. Repeat to fade...

As yet, there are no animal remains found millions of years out of their expected strata, though.

One problem with the "missing link" argument is that finding a missing link only introduces extra missing links between your original two data and the missing link... the better a job palaeontologists do, the more missing links there are.

Sam said...

Put short, the only alternative to the claim that all life on earth evolved naturally (i.e. creationism) is in principle unverifiable, which means the counter-claim is in principle unfalsifiable. But I don't think that necessarily makes it invalid.

I'm not sure I follow this. The fact that one alternative is unverifiable doesn't imply that another alternative is unfalsifiable. Evolution is not a perfect logical "NOT" of creationism.

The broad idea of "evolution" is the sum of a number of interacting models, each of which is falsifiable:

* DNA contains the code that tells living cells what to do
* random mutation of genes occurs
* natural selection within populations causes evolution of species

etc etc

Arguably, the word "random" could be regarded as a logical NOT of "intelligently designed", but other than that all the constituent parts are falsifiable, aren't they?

Moon GT said...

Those three things are falsifiable. The theory of evolution as an ongoing process is falsifiable. A statement such as "X evolved from Y" is falsifiable. What isn't falsifiable is the claim that ALL life on earth is the result of natural evolution. It's like the "there are no non-black ravens" problem, only in this case it's complicated by the fact that there's no definite way to demonstrate that any particular raven is non-black.

Finding a rabbit fossil in Precambrian rock isn't really satisfactory either. It's a bit specific, for one thing, and isn't based on any reasoning as to why God would have put a rabbit there, or anywhere else (indeed, creationists hold to a much more recent origin of rabbits, not an older one). It's the informal fallacy of denying the antecedant. If evolution is true, then fossils are in certain places. If evolution is not true, it does not then follow that the fossils would be in different places.

In any case, supposing someone actually DID find a rabbit fossil in Precambrian rock, what do you think would happen? Claims of "hoax" and such things as "rabbit fell down a natural fissure which then closed", and "there must be a rational explanation".

Rich Lancashire said...

If fossils were found out of the expected places, then evolution is not true. That's falsifiability, not "if evolution is untrue then fossils must be found out of their expected places". That's makes it scientific; not necessarily true, but scientific. Which is a good place to start.

A counterexample is (should be) enough. Or rather, a few counterexamples. In this case, one on its own can be explained away, unlike, say, the precession of Mercury, and you can't have a useful system of universal enquiry that is destroyed by a rabbit falling down a hole, or a malicious human with an agenda.

But as a general case, it holds that remains are found in period-specific strata. It's not positive proof of evolution, of course; evolution fits the data as a theory with the minimum extraneous explanation.

Whether *all* life has evolved is a question of genesis, not evolution. There's an awful lot of evidence to support the claim that animals existed a long time ago, and some of their descendants adapted to what we see around us.

Moon GT said...

Ok, such a counterexample would "falsify" the theory, but I still don't think it's really a fair argument. Firstly, it's not a comprehensive test, only looking for one specific exception rather than the failure of the hypothesis in general. It's a bit like suggesting Newton's theory of gravity is falsifiable by finding an apple that falls upwards - indeed, it would falsify it if one was to observe such a thing. Secondly, it's not a test you can actually do. You can't do a test to find out if there are or are not any rabbits in precambrian rock, or if there are any apples that fall upwards, you just have to rely on one turning up. The lack of one observed so far doesn't imply their absence.

Newton's theory of gravity is falsified by the procession of mercury and the degree of bending of light around stars. In these cases though, we knew what to look for because general relativity actually predicts those effects. (In fact, the procession of mercury was known before but theorised to be caused by another planet closer to the sun, which they named Vulcan.) So in this case, the falsification of one theory was the validation of the newer theory - general relativity predicted where newtonian gravity would fail, so they knew what to look for. Evolution doesn't have that kind of competition. However, it's precursor did. A couple of hundred years ago, and dating back to the ancient greeks, the scientific thinking was that animals "spontaneously generated" in a variety of different ways. I shall quote now from a 1981 book I have entitled "Everyman's Scientific Facts and Feats" by Magnus Pike and Patrick Moore.

"Even van Helmont (1577-1644), a distinguished scientist working in Brussels, who made substantial progress in the study of plant nutrition, could seriously write down the recipe for the creation of mice: thus 'If a dirty undergarment is squeezed ubti the mouth of a vessel containing wheat, within a few days a ferment drained from the garments and transformed by the smell of the grain encrusts the wheat itself with its own skin, and turns it into mice.'"

This is quite easy to falsify in a controlled experiment.

That animals existed when science thinks they did I will not dispute. I'm not actually a creationist.