What is science, continued

In the previous post with the same title, I wrote that I consider science to be best defined as the use of empirical data to critically test ideas, regardless of what specifically the ideas are about or how exactly the test looks like. The important thing is to work with observations from the world around us (as opposed to abstract concepts, for example) and to ask yourself how you would know if your idea could be disproved. If you don't do the former but the latter, you are doing non-science but you may still be dealing with a legitimate academic endeavor such as philosophy. If you do the former but not the latter, you are a pseudo-scientist.

It could be asked whether my definition is not a bit too lax - after all, aren't we all using observations to test our ideas? And indeed that has been part of the controversy between people who favor a wide circumscription of science (like myself) and those who favor a narrow one. For example, developmental biologist PZ Myers made the point that we are all daily surrounded by trivial mysteries that are best solved using basically the same approach as a scientist, stressing that "science is simply a process for examining the world, and anyone can do it, even if you don’t have a lab coat." In a similar vein, evolutionary biologist Jerry Coyne once compared science to plumbing because plumbers also use empirical observations and hypotheses in their work, prompting ridicule from philosopher/biologist Massimo Pigliucci. However, I do not consider the equivalence of scientific practice and everyday commonsense problem solving to be an absurd conclusion, as Pigliucci seems to think, but instead an interesting observation.

Think about it. We all behave as scientists under the following circumstances: when it really matters that we get things right and when dealing with an issue for which we do not have any ideological or religious preconceptions. When you have lost your key, what do you do? You think, "maybe it is still in the jacket", and check the pockets. "Ah, maybe it is on the kitchen table, let's check there." And so on. In other words, you formulate hypotheses and test them using empirical data, one after the other. That is different only in degree of complexity but not in principle from hypothesis driven scientific research. As another example, you also build models of other people's personalities and update them against empirical evidence. For example, if the neighbors' son has always been friendly and kind to you, you will get the idea that he is a decent person. If you then look out of the window one day and see him torturing a kitten you will update your model of his personality in a way that is different only in being less quantitative but not different in principle from a scientist using Bayesian analyses*.

Note what you would not do, by the way. You would not sit in a chair for the next six weeks and pray for the key to magically turn up. You would not assume that the neighbors' son is so cruel because he is suddenly possessed by a demon. Or at least I hope you wouldn't, and I assume you would consider somebody who behaved in that fashion to be rather odd.

Yes, finding your key is not professional science, but it would be hard to argue that it is something different in principle. It is to science what adding up today's expenses at the markets with pen and paper is to a professional mathematician calculating the minimum insurance premium her company has to demand to still turn a profit - you can hardly say that the latter is calculation but the former isn't because the pro is using some special mathematician-sauce in her work that you don't.

And I find it really important that people are aware of this equivalence between science and what they, themselves, do every day when they need to work things out. Too many have this idea that they use common sense to figure things out in their own lives while the scientists in their ivory towers use some weird magic egg-head sauce that is very different from common sense. Wrong. Laypeople use empirical data, hypothesis testing, Bayesian reasoning and parsimony, and scientists use the exact same tools. The only difference is that laypeople more often use qualitative reasoning where scientists are expected to work quantitatively, with exact numbers, and that scientists deal professionally with questions that one needs a lot more training to know how to even possibly address than finding the lost key.

Still, science is basically just a formalized variant of what we all do every day when we honestly need to figure something out. So our classification of ways to form beliefs about the world should not look like this:

In our daily lives: common sense
In science: special obscure magic egg-head approach that is silly because no normal person can understand it

But instead like this:

In our daily lives when it counts, and in science: reason, evidence, parsimony
When we are lazy or biased: gut feelings, faith, appeal to tradition, argumentum ad populum, etc.

One reason why I find this realization so important is because too many people, when faced with scientific results they do not want to accept, attempt to argue from postmodernism or from the problem of induction. Ah, they will say, in the end it is all just opinion or faith. You have yours, I have mine, let us agree to disagree, why are you so intolerant? Or, if they have read just a bit but not enough philosophy: Why should I trust science when inductive reasoning cannot be justified from deductive reasoning?

That is all intellectually dishonest because the same people making these arguments trust the very same methods used by the scientists whose results they are dismissing in their own lives, every day. The postmodernist, religious apologist or ideologue will look before they cross the road, trusting their eyes to examine evidence from the physical universe. They will trust the laws of nature to remain constant and people to be predictable in their behavior. They will actively search for their keys instead of praying for them to materialize magically in front of them or instead of contemplating that the continued existence of the key is in doubt because inductive reasoning cannot be justified or some such nonsense.

A straight and unbroken path leads from this rational, empirical stance towards approaching dangerous roads and lost keys to the methods scientists use to study evolution, the climate, vaccinations, antibiotics and any other issue, and if the believer and the ideologue could just rid themselves of their dogma and ideology and follow that path with intellectual honesty and consistency, they would necessarily arrive at the same conclusions as other scientists before them. Similarly, and as argued here before, if all our knowledge were obliterated, humans could walk the same path again and rediscover all the same knowledge because it really is knowledge about things that really are so and not just belief, opinion or ideology.

Understanding that this is so should fill us with awe: there is a universe to be discovered out there, and you and I and everybody has the same chance of discovering a bit of it as any scientist if only we use our reason. Unfortunately, all too many people think that science is something special and different from normal reasoning. In reality, it is when we each attempt to exempt our own cherished beliefs from critical examination that we do something literally special that is different in kind from normal/scientific reasoning: special pleading.

Footnote

*) One of the more tedious rejoinders to the demand to apply evidence-based and rational thinking consistently to all aspects of life is "would you use science to decide if you love your partner?" The question is besides the point because personal tastes and preferences are not knowledge, and nobody claims they are. The right question to ask would be: "would you use science to figure out if your partner loves you?", and the answer would be yes, for the soft and wide definition of science as used in this post. You know if somebody loves you by observing their behavior (i.e. from empirical evidence) and inferring to the most parsimonious explanation. For example, if they say they don't want anything to do with you, that counts as evidence against the hypothesis that they love you, and you should consider to stop stalking them already.