My Thoughts on Science (& a Bit on Religion, too)
I always have such a tough time explaining how I think to other people. No one ever seems to be able to follow my train of thought no matter how hard I try to explain myself, and I consistently end up thinking that maybe there's some part of me that is crazy in a way that is obvious to others but is completely invisible to me.

As I write this, I think I am undergoing a process that is wholly indescribable in the limited vocabulary I now possess (though my rss subscription to Erin McKean is slowly fixing this problem). I am, I think, falling in love with science, even as I abhor further and further the idea of 'science' that I held as a child. I'll try to explain.

As a child, I was taught the scientific method: observe, hypothesize, test, conclude, repeat. It was presented to me as a way of approaching truth in our world. Through science, we get closer and closer to the true reality, and although at any given point in history we are incorrect on the specifics of how things work, at each successive stage we get closer and closer to what is real. (I.e., in Newton's day we had strong knowledge of classical physics, which was correct to a certain degree of fine-tuning; then came quantum physics which gives identical solutions as classical physics in the aggregate, but gave better knowledge when the fine tuning was tuned ever finer.)

I really did trust in this way of thinking for a long while, and I know many (if not most) current scientists agree with this conception of what science is. It is, after all, what was taught to me in school, and what I read in the science for laymen books of my childhood.

But today, as a philosopher, I find such a conception to be slightly naive. Science, in my eyes, may or may not be building ever closer to the real theory of reality. But that's not the point. The point is to find what works, and to go with it.

Scientists do science by taking showers in the morning, sleeping at night, or meditating in the afternoon. Great discoveries are found through inductive reasoning, which they then reconstruct deductively in order to write a paper that will convince fellow scientists. The art of science is very different from the linear logical argument you'll find in most peer-reviewed papers. When I first realized this, I thought it a crisis for science, because it seemed to really contradict the idea of science as it was taught to me as a child. The true scientific method differed so much from what I recall from my old science textbooks.

But in the end none of this matters. What matters in science is not how an idea is arrived at. What matters is that the idea is tested and thrown out if it disagrees with experiment.

In essence, science is a negatively defined discipline.

What makes science great is that you can take a testable hypthesis, test it, and when it breaks you get rid of it in some fashion. Historically, we've 'gotten rid of' old theories by keeping them mostly intact and adding in a few bulges and denting in a few kinks so that they continue to correspond with experiment. Eventually, the theory becomes so convoluted that it is deemed better to start with a fairly clean theory that looks similar to the old one, but has a few fundamental differences. But the point is that none of this matters. The important bit is that, over time, we eventually prove such and such wrong and move on to something else. Because we're human, we happen to pick a new thing to replace it that is as similar as possible to the old idea. But it really could be whatever we wanted, so long as it, too, fit the data we have available. In fact, the only theoretical benefit of having a current theory that is favored over others is that we can then, as a group, all simultaneously work on figuring out why this one idea is wrong. (As opposed to, say, each scientist holding a different theory that they then attempt to disprove. Also, it should be mentioned that a nontheoretical benefit to having a single favored theory is that it makes iteasier to communicate with each other--another is that it allows nonscientists to have a worldview from which engineers can come up with useful contraptions and the like.)

Science, in this higher level negative sense of the word, is a beautiful, beautiful thing. But the science taught in public school in Alabama is ugly and dogmatic; it is arbitrary rule-making for no good reason. If you want to be a truly good physicist, you need to NOT learn the model, but to knowthe DATA. Models are good as a memorization aide. Models are good as a starting point to create hypotheses. But ALL that matters is the data.

I say this even though models have had some particularly good hits in the past. Newton models gravity, and reality turns out to follow it to a degree of accuracy far beyond the data Newton originally had access to, simply because he went with the solution that seemed most elegant, and nature just so happened to comply. Another example is the atom, modeled as a physical item rather than just a set of properties that could easily have fit the data we had back then just as well. Then further properties of the hypthesized atom were derived which then corresponded to nature once again, showing that the model was in fact closer to reality than my earlier paragraphs seem to imply they usually are.

But in science, as in life, it is often the case that we recognize the hits while forgetting the misses. Especially in science, since we get so used to recording both hits and misses (at least in nonparapsychology studies) that we forget that we aren't recording data when models break down the way we do when we break down those models. And the truth is that most models do in fact break down with nature not being so kind as she was with the gravity model or the atom model.

Anyway, my point is that whatever models we now have, no matter what it is (evolution, gravity, black holes, stars, blood circulation) is most likely fundamentally WRONG. Just think about it: the data we have right now on, say, blood circulation, is pretty extensive. The data is (as far as I know) very good, and the current model we have does an extremely good job of intuitively explaining this data in a way that makes logical sense. But there are an infinite number of alternative models that might also fit those same exact data points, and most of them are extremely dissimilar to the current model of blood circulation that we now hold as gospel.

Scientifically, in the old sense of the word, it makes sense to me why we should keep the model we now have. It makes teaching students easier, it gives nonscientists (medical doctors, for example) a starting point from which to come up with useful devices that may save our lives, and it even gives scientists a base from which they can start hypothesizing things that seems to spring up as ideas merely because they're starting with the idea of the model as truth.

But philosophically, the odds that 'the model we currently are using is even remotely similar to whatever reality is' are effectively zero. We use the model we now use because it is convenient; not because it is true. Nor even because it is close to truth. Science tells us NOTHING of what is true, nor even of what is close to truth. It merely tells us what isn't true.

This negatively-centered view of science is something I adore. It is something I am falling more and more in love with with each passing day. But I must stress again that it is very different from the 'science' of my childhood, which I view as exceedingly dogmatic and inane. So when I say that I am falling in love with science, this is the science I mean.

So, too, does this argument extend to religion. Truly, I am agnostic. But I am agnostic not because I cannot pass judgment on God's existence, but because I cannot pass judgment on anything at all! Even in science, where I implied that at least science disproved hypotheses even when it failed to ever prove anything in the least way, these disproofs are (to me) only the slightest of obstacles. There are a great many premises which must be first granted before you can even start to disprove hyptheses. My skepticism is kept at bay when it comes to negatively-centered science only in short bursts, and only by the fact that science has such an INSANEly good track record. Just the fact that I'm typing this on a laptop is incredibly god anecdotal evidence for the success of science. And it is anecdotal evidence like the inventions that surround our lives every day that compete against my skepticism against the so-called 'truth' of disproving hypotheses in negatively-centered science.

Anyway, that should make it clearwhy I am an agnostic. Really, I could not be anything but an agnostic, given the philosophical views I hold.

BUT: atheism draws me in the same way science does, but in an even stronger way, the way the anticipation of sex makes me crave women.

I'll steal an old atheist argument just to prove my point. Think of how utterly absurd it would be to say Poseidon is a true god. Think of how utterly obvious it is to be atheist with regard to Poseidon. That is how I feel with regard to the personal God of the West. The feeling of atheism is perhaps the strongest feeling I have on a regular basis. It is stronger, in fact, than any sexual or hunger urge I've ever experienced. (Maybe this is because I've never experienced true starvation, but this is immaterial to my point.) Yet as strong as those feelings of atheism are, they cannot stand up to my feelings when it comes to philosophical skepticism. I am an agnostic first, and an atheist second--even though I think my feelings on atheism are far stronger than most religious people's views on their god.



If you've read this far, then I congratulate you. Now the question is: am I just mad? Was the above completely retarded in your eyes? If so, please just tell me. I can handle being told I'm wrong, especially when I fear that that is indeed the case.

I only ask because most people don't agree with what I wrote above. In fact, most scientists I know don't agree with these things. Even when it comes to philosophers of science, I find very few people who write on themes similar to the above. Most phil of sci types tend to think the negatively-centered view of science is falsified by the fact that that's not how scientists really think. So forgive me if I fear that my opinions on these matters may be flawed in some way that I don't yet fully recognize.

Of course, all of this tirade may just be because it is Valentine's day, and I have absolutely nothing to do except write a lengthy nerdy article like this. Not to mention the lengthy public political article I wrote at eric.herboso.com earlier today. But I can't be entirely full of self-pity; at least Mary is in town. We haven't gone anywhere yet (she just flew in a few hours ago), but I plan on taking her out to see the sights of DC. I just hope I can pull her away from the religious crap like the nat'l cathedral* and catholic university long enough to see the int'l spy museum. That place sounds awesome, and I haven't yet visited.

(*: Not to say that the nat'l cathedral isn't architecturally interesting, but I think my two previous visits were more than enough for me.)


(PS: I did not even get into the idea that reality might not even have a form that any model, no matter how sophisticated, might not be able to accurately conform to the possible dataset. This concept is rather scary, and is rather difficult to comprehend at first, but the idea is that while there are an infinite number of conceivable models, there may in fact not be a model in this infinite set that corresponds to the particular data that the universe gives us through experiments. I.e., when you make a one-to-one correspondence between 'best theory' and 'current dataset', you may (not only) run out of theories before you run out of datastets (, but also run into datasets ordinally larger than any individual model can account for. See transfinity on wikipedia for details.)