Getting at what Science is exactly, has not exactly been easy. I introduced Early Wittgenstein last entry so let me start with him again, but with an idea from his later period (but keep in mind the point in my last blog: there are things science can’t touch. This blog is about the things it can). One thing science is, is a language game. A way to speak about the world. It has a vocabulary, a grammar, a way of discoursing. The language is played with logic, rational discourse, and certain ways and traditions of presenting findings to the world. This game can be played by anyone and it often is. Playing the language game however is not what makes it science.
The task science takes on with its discourse is about finding a way to represent reality. It offers models of the world in forms that span everything from mathematical models to paper and glue models of DNA’s double helix. This is part of the language game, but is more formal and more demanding. How these representations represent reality has been an ongoing challenge in science. Newton’s laws and Einstein’s equations found their representation in mathematical models. Chemists have used plastic balls and sticks. Geologists have used maps. What unites these ideas is that in one way or another these models allow us access into insights about reality, this usually by providing simplifying ways of reducing complexity in useful ways. A good representation can go a long way in cutting trough the messiness of the real world and offering explanatory and predictive power. New techniques for building these representations is one of the activities of science, for example, in the last 20 years computer simulation as become an important kind of representation.
These are the two main goals of science (1) Figure out why things are as they are; and (2) be able to predict the way things will be. To be a good representation it as to conform to certain values that science embraces: coherence, repeatability, universality, causal or statistical explanation, predictability, beauty, etc.
Once it has a representation, whether it is a complex mathematical model or a verbal story, science plays the real world against these models in a back and forth between that assumed real-world and our representations: Model–reality-back to model-back to reality.
This is a refining process and is accomplished by experimentation: in which everything you can control is controlled in order to hone in on certain processes; and observation: the patient collection of facts about the way the world works.
This is also a messy project, so messy that some philosophers have defined science as ‘that which scientists do.’ There is a great deal of truth in this. Becoming a scientist requires an apprenticeship, an introduction to a culture of discovery that comes largely through institutions such as universities. For example, learning a particular research tradition, such as molecular biology means time in labs, hands-on training that really goes beyond anything you could read in a book. There are recipes but as any good cook knows it’s the secrets not in the cook book that make you a chef or anyone could do it. This has led some cultural relativists to argue that science just is these traditions. It is hard to break from these traditions and it sometimes takes massive data, blindingly insightful models, and new people to do so. This is an acknowledged weakness of science.
Science is a Darwinian selection of the best ideas. Everything scientists do is peer-reviewed. What this means is that no one gets to make claims without a severe critique by others in that research tradition. If you offer a mathematical proof, other mathematicians pour over your offering looking for errors or mistakes. If you are an ecologist, other ecologists pour over your research methods, your statistical analysis, and your comparison of what else has been done in the field. Peer-review means that only about 2 to 25% of the time is a paper selected for publication. With the stated 2% coming from high-end highly competitive journals, which are notoriously rigorous in their evaluation and in choosing what they select, and the 25% in more regional and specialized areas. Journals have reputations. Getting scientific work published is an exhausting demanding enterprise.
One of the most important activities of scientists is then communicating these ideas with people with other backgrounds including other scientists in other research traditions, decision makers, and the general public. This is one area scientists often drop the ball.
So what does all this buy us? Does science lead to infallible truth? Does it give us undisputable views of reality?
No. It tells us how to bet. That’s all. It gives you the odds that things are a certain way. The more that particular and continuing studies suggest a certain thing is true, the more you should update your probabilities for making wagers on that thing being the case. For example, on the wager that DNA is the carrier of genetic information I’d lay odds of about ten million to one in favor. Sure things are still being sorted out, such as the recent work on epigenetic inheritance of some things like methylation signature (if that means anything to you), but enough studies have shown that DNA is where inheritance lies that to bet against it would be just plain foolish.
Science lays out the odds. The philosopher Hume showed that you really cannot get at objective certainty in anything. And science is a long way from providing certainty. This opens certain vulnerabilities to its claims. One can raise suspicious about any claim. This is how the tobacco industry maintained for years that the evidence just wasn’t in for a connection between smoking and lung cancer. They had a strong interest in keeping doubts about the connection from the public and if you set your criteria for accepting any scientific finding high enough, the science can always be doubted. Even today one could claim, if one wanted, that the evidence just isn’t in on smoking and cancer. You could say, “I just don’t find it compelling.” But that’s not the way to bet. Even something as obvious as the claim that DNA is the carrier of inheritance can be doubted. These are the unhealthy suspicions I talked about last blog.
The expectation in some people’s minds, that science is about proving things, can lead them astray. When evidence seems to weigh against some pet idea, its time to update your probabilities, not dig a drench and throw up an embankment. To ignore the evidence because you can claim that it has not been proved, is a breach of what science is about. When someone falls back on Humean skepticism to maintain their believe (disbelief) in an idea that science is accumulating strong evidence against (for) and thereby arguing that the science is still out on a particular matter can be dangerous. By certain criteria the science will always be out.
Science then gives you a sense of where to lay your money. And its successes speak for themselves about just how good it is at laying out those odds. When you bet against well-established findings in science, your playing risky odds. Sometimes that pays off, usually it doesn’t. That’s the nature of gambling. That’s your prerogative, but it’s not wise. People are really, really bad at probability thinking (Hence places like Las Vegas) and those who figure out the odds are positioned for much better decision making than those who aren’t. This is true of governments as well as individuals.
So let’s spin the wheel and see where the bouncing ball lands on Climate Change. Next time.
I think science is not always a good bet.
But Gary. My last post argued that all the things you offered to defend in your post (Resurrection, Second Coming, etc.) aren’t even in the Science game. As you pointed out a couple of posts ago over at your site (hit the link in his comment) these are matters of Faith and you didn’t get them from Science. I agree with you on that!. That is my point too. As I’ve argued in my previous post, these two ways of knowing deal with different subject matter. You didn’t learn these things from science at all and Science has no stake in that game.
On the other hand where did you learn your Astronomy? Where did your physician learn her medicine? Do you believe DNA carries the code for inheritance? Don’t you want to get these things as right as you can? For the objective facts of the universe science is the best bet we have. Once again, take a peek at the Middle Ages for a good view of life without Science.
[...] What is science? asks a practicing LDS biologist. It is a language game and a way of speaking about the world. It devises models of reality and tests them against data from the assumed real world. It provides probability rather than certainty: “It gives you the odds that things are a certain way.” But “its successes speak for themselves” and “[w]hen you bet against well-established findings in science, you’re playing risky odds.” So … bet on Darwin. Unless you think science is not always a good bet. [...]
Steve, you are right. Life without science would be a bad idea. And I agree that, as a tool for learning about the present universe, science is a good bet. But I believe in “the more sure word of prophecy” (2 Pet. 1:19) when it comes to learning about the future universe. It’s that simple.
Thanks for the essay. I may borrow some of your thoughts and examples next time someone tries to tell me science is invalid. Especially when it comes to global warming.
Science is the greatest cultural achievement of humankind. There is NO reason for religionists to fear it or to denigrate it. Besides, I think God is a scientist. Oh, well, that is MY opinion.
I just had a heart scan. Maybe unlike R. Gary, I am betting my life on science.
I think that is a good point, S. Faux, “God is a scientist.”
Why can we not learn about the future universe just as we are learning about the past and present, through science? Obviously there was a process in how this world was created. And it was not a short process. It did not take six days, or six thousand years, but billions of years. This does not mean that God did not create the universe, it simply means that it took longer then what the Old Testament tells us.
And again, there was a process. A very long process, a scientific process. Matter formed, a world was created, plants and animals, (yes, through evolution) formed, and changed and grew. Just because we have found evidence of this through scientific methods, does not take God out of the picture.
Now, R. Gary, if we can see that through studying the past, and present, think of how much we can learn about our future.
What are the current odds in Vegas on GW? I’d like to place a bet.
I think you lack an important distinction between experimental and theoretical science.
I agree with your conclusions if you are talking about science that is experimental in nature (if I drop a stapler, then chances are it will hit the ground). I disagree with you if you also include theories that are heavily based on models or mathematics (e.g. Big Bang, String Theory, Climate Change, Evolution). I think it is pretty safe to bet against the conclusions of these types of theories. We have just seen too many cases in history where a small error in a theory entirely changes the conclusions.
All experimental data are “theory-laden.” You don’t collect data without a theory behind it. I take it you would include relativity and quantum mechanics on your list? Both of which are heavily based on models and mathematics.
Absolutely, Relativity and Quantum Mechanics are definitely on the list. In fact, there at least three or four interpretations of Quantum Mechanics that I am aware of. These interpretations are based on the same experimental data, but have very different overall conclusions about how nature functions. We scientists just choose the one that we like the best.
Yes, experiments are generally set up with some theory in mind, but the experiments are king. The problem is that the same set of experimental data can potentially support hundreds of theories, especially when the data are sparse.
In my opinion, we as a society have an enormous problem with the separation of experiment and theory. One being fact and the other just educated guessing.
Your separation of theory and experiments is utterly confusing me. To me it sounds like someone saying, “I like dance, as long as there is no movement in it.”
Can you give me an example of theory-free experiments?
What makes a good theory is that it suggests experiments that confront it. If an experiment supports hundreds of theories its a bad experiment. Part of the purpose of an experiment is to confirm and test some theory. That’s in part what makes it science.
All experiments are theory-free.
We may be misunderstanding each other. You may be talking about the influence that theories have on deciding which types of experiments are most useful. I am saying that the outcomes of experiments are always theory-free. When we experiment, Nature tells us the facts; it could care less what theory we had in mind when we did the experiment.
Experiments tell us the facts. Those facts can always be used to support multiple theories. I gave Quantum Mechanics as an example where the whole wealth of experiments done in that field can still support multiple theories equally.
It is similar to the children that play the game Hot and Cold. The experiments tell us if we are getting warmer or colder. The theories are just educated guesses of where we should experiment next.
We are speaking in different places because I’m saying that experiments are never theory-free. There is no way to interpret an experiment except in the light of theory. That’s the experiment’s purpose.
There is no such thing as a fact except in the light of some interpretive framework.
I agree with you. The main point I am trying to make is that conclusions of theories are not very reliable while the outcomes of experiments may very well be.
Even though theories and experiments are closely related, there is a fundamental disconnect between them. This is evident in the fact that multiple theories can use the same set of experiments as a basis, but give entirely different views on nature.
[...] People often confuse this with randomness and I’ve tried to sort this out a bit (here, here, and here), but what I want to focus on now is unpacking the difference between teleology and [...]