haggholm: (Default)
[personal profile] haggholm

The other day, someone told me that I treated science as a god replacement—in fact, two people who more or less know me had apparently been talking about me in those terms. I’m still trying to figure out whether that’s a good thing or not, apart from the blind faith connotations inherent in the religious terms (to my ears, at least).

The key is, of course, to nail down exactly what one means by science. I do not think that any scientific fact should ever be held as gospel—many are solid beyond the shadow of a doubt, but new facts, new lights, may cast new shadows. Obviously, I do not wish to deify any scientist. There are many scientists whose minds and accomplishments I admire (and while some, like Sir Isaac Newton, were awful human beings, others, like Charles Darwin, seem to have been very good people morally as well as intellectually), but they are or were human and fallible. Not only am I unconcerned by ad hominem arguments (Newton's being a right bastard doesn’t detract from his intellectual achievements, or make his theory of universal gravitation any less useful); I also am not bothered by any scientist being shown to have been wrong on any particular point.

This may not always be obvious in conversation and debate, of course. I’m rather fond of putting it so: When I am convinced that something is sufficiently probable that the margin of error, though never zero, is negligible for practical purposes—e.g. I am so convinced of the truths of gravity, electromagnetism, evolution, etc. that I see no reason to take the null hypotheses into consideration—I choose to say that I know it is so, because it is shorter, more forceful, and more pithy than I am convinced that the probability of it being so is so high as to render the null hypothesis negligibly small for practical considerations, even though the latter more accurately describes the stance I try to take.

There is also the lamentable fact that I’m stubborn and like to argue and have an all-too-human tendency to take a position with overtones of fiendish advocacy: If you take a position that I disagree with, I may for the sake of argument take an opposite position more extreme than I truly credit.

That aside, what’s left is not scientific fact, nor scientific practitioners, but science itself. Thus, either people who look at me and cry God replacement! misunderstand me and think that I deify scientists or scientific facts (which may be my fault as easily as it may be theirs), or it is science itself that is the issue at stake. It may well be the latter. But science is not a set of facts, or of people, though it needs these things: Science is a method and process of discovering truth based on logic, empirical observation, mathematics, statistics, probability, and naturalism. (This is not to say that science is by definition incapable of discovering supernatural phenomena—if any exist, they will be found precisely where science finds anomalies that cannot be accounted for under the assumption that these rules all hold.) It is the only method we have for discovering the real truth of the universe to any degree superior to our own brains, which, for all that they are marvellous pattern-recognising deduction machines, are also prone to finding false patterns, pareidolia, conflating correlation with causation, and many other errors that we should expect from an animal that pays a much more dire prize for false negatives than for false positives, and whose heuristics have limited powers.

I asked someone recently in a rather pointed way whether she felt that intuition was as good a guide to truth as scientific inquiry. She replied (you’re going to think I’m crazy, but…) that she does, because scientific models are always constrained by prevailing cultural and intellectual paradigms. In some areas of research, this may have a point, but I think it misses the main point entirely, because this is precisely what science is meant to avoid. If you can point out a way in which a study—any study—has a risk of being less than objective due to such biases, you have not poked a hole in the scientific method in general, but rather identified another bias for high-quality scientific inquiry to correct for—along with logical fallacies like conflating correlation with causation, confirmation bias, observer effects, the Hawthorne effect, regressions to the mean, placebo effects in medicine, and myriad other cognitive quirks that we have to isolate.

But the highest goal of science is to find objective truth, and I don’t care what you say—unless you subscribe to solipsism (in which case I won’t even bother with you), there are objectively verifiable and falsifiable tests. To paraphrase Richard Dawkins, aeroplanes built to scientific specifications fly, while (cargo-cult) aeroplanes built to religious specifications do not. Differentiating between flying and not flying is not subject to cultural bias, and nor are other good scientific facts—and the hard sciences abound with them. (Our differences in opinion may have been conflated artificially due to my background in the natural science and this conversation partner’s background in humanities and softer sciences—the observer’s bias is much more likely to influence observations of people’s behaviour than observations of the behaviour of atoms, I should think.)

And let’s face it—ultimately, you have to reduce reality to a set of objective truths unless you want to descend into solipsism. What I wish I had said (but thought of too late) might be something like this: Intuition cannot be trusted because it cannot be tested. You have a feeling that something is true; well, I have a feeling that it isn’t. Only objective testing can settle it; otherwise there are no facts, but only opinions. I might also point out that I have a feeling that your idea that intuition is as good as science is dead wrong

This, then, is what I think: Any scientific fact I hold up may be wrong (though some are fantastically unlikely to be so). Every scientist who ever lived, and every scientist who will ever live, is and will be fallible. But the scientific method is the only viable way of ascertaining truth beyond personal experience, and personal experience can be notoriously misleading; and any valid criticism of the scientific method as currently practiced will never tear it down, but only—at most—show how our current practice can improve.

I am open to challenges on this point.

I also feel—and this may well be controversial—that any field of inquiry that is irrevocably biased by cultural norms, etc., is not strictly scientific at all: I will not call it pseudo-science, but perhaps almost-science (or soft science…). This is not to take anything away from its practitioners: Strict science is ultimately the most accurate method for finding truth, but it is not therefore always the most practical (c.f. the old hypothetical example of the folly of having children empirically test their parents’ claims that the river teems with crocodiles). If your model purports anything beyond strict physical measurements, it has strayed from the field of hard science where this sort of reliability is possible.


haggholm: (Default)
Petter Häggholm

April 2016

345 6789

Most Popular Tags