A Realistic View of Science

Now that we have taken a look at the “ideal” description of the scientific method, it’s time to look at a more realistic (some may say “cynical”) example of how science works in the real world. It should be pointed out that most scientists genuinely do wish to adhere to the idealistic version of the scientific method. That is, they want to set up their four step loop: observation, induction, deduction, and experimentation. However, for virtually every interesting experiment, there is a big problem to getting this loop started.

Money.

It takes money to run experiments. Often, it takes massive amounts of money just to observe your experiments, especially when you deal in fields such as physics (anyone care to pay out of pocket for their own particle accelerator?). But even if you take a simple Darwinist experiment, where you observe finches in the Galapagos Islands, the scientist still must travel to the island with some kind of instruments to record data (even if just a pencil and paper), and he or she must live there with all the expenses of food and shelter that would normally be required.

So unless the scientist is independently wealthy, he relies upon OPM (Other People’s Money). And just like opium, OPM is an addictive drug. Experiments can run over-budget, and scientists are generally curious people in the first place so if there are extra funds they will find a way to put them to use (ever wonder what a loaf of bread would look like after it's put into a vacuum?—trust me, some scientist knows). In other words, you can never get enough of OPM.

The result is that, to get an experiment done, it takes more than just the ideal four-step loop. A scientist must first start by begging for money. This is often done in the form of grant proposals. Barring that, a scientist can also decide to go work for a corporation with a vested interest in his or her field. For example, a geologist may go work for an oil company, even if he’s a vegan green New Ager, simply because the company has funds to conduct experiments loosely based on his degree; or a biologist may go work for a pharmaceutical company, even if she voted for Obama and wants universal health care coverage extended to pets, again simply because the pharmaceutical company has money to fund research. But no matter how it’s done, the scientist must first secure funding.

Funding is often extremely limited in scope and duration. But it is also simple reality that apart from a few wealthy eccentrics, the money is given to scientists as an investment. Those who give the money expect something in return for it. When it comes to corporations, obviously the owners want to increase their profits—so if you can find a cheaper way to get oil, the oil companies will fund that research; if you find medicine that prevents heart attacks, the pharmaceutical companies will fund that. For other grants, there are a few cases where people give money for “pure” science, but many come from non-profit organizations with a vested interest in proving something (to use an example I personally am aware of, although for reasons you will see later I must keep it general, a non-profit once paid an archaeology team to determine which tribe of Natives arrived at a Central American site first).

Once the scientist secures this OPM, he can then begin to do the four-step loop of “ideal” science until the funds run out. At which point, he must ask for more OPM or else the science is “done.” The net result is that scientific method ends up working like this:

1. Get OPM
2. Observe
3. Hypothesize
4. Predict
5. Experiment
6. Return to step 2 until OPM = $0.
7. Return to step 1.

Naturally, while the above loop doesn’t have an end, there will be an end to science beyond just running out of money. For instance, the pharmaceutical companies look for effective drugs, and they only do this loop until it either becomes cost prohibitive or else they succeed at finding an effective drug they can sell at profit. This opens up a new dynamic, because each experiment is finite but you can always do new experiments if you keep your nose clean.

If you have a good relationship with someone who will give you OPM, it is more likely that you will receive OPM in the future. This means that the scientist, who does not wish to spend his time getting funds—he wants to do science—is given an incentive to keep the person writing him checks happy. For corporations, this means that scientists want to keep their employer happy; for those who live by grants, they want those who give grants to continue to give those grants to them.

This puts high subjective pressure on the scientists. Science is supposed to be objective, but when scientists know they have to keep certain people happy in order to continue getting OPM just so they can do science in the first place, it’s easy to “misread” an experiment. And sometimes, it’s done intentionally. For instance, the non-profit who paid for an archaeological dig to see which Native tribe first got to a specific Central American site paid for the research because they wanted to prove a specific tribe got there before another. The archaeologists found evidence that indicated the other tribe got there first. But the published reports of the experiment said otherwise. Because scientists know which side their bread is buttered on.

(Full disclosure: obviously, this example is hearsay, but comes from a source I trust. That doesn’t mean you have to, of course.)

Scientists are no more or less human than a Wall Street stockbroker or a banker or a lawyer or your next door neighbor. They are no more likely to “do the right thing even if it costs all their funding” than would anyone else. Yes, there are some who do so (just as there are some in every profession) but there are also some who take shortcuts, fudge the data a bit, and make sure the experiments come out in the way that will benefit them.

The “corrective” for this is repeatability. Science, to be science, ought to be repeatable by others so that this can expose biases (hidden or otherwise) in the methods the scientists use.

But what happens when the same group, or a likeminded group, funds both the original experiments and the repeats? Since funding is coming from the same, or similar, source, each group of scientists has an incentive to please the person giving them OPM. And when the opposite occurs—that is, when someone who opposes the first group’s ideology funds a counter-experiment to disprove it—it is ridiculed as being biased. The net result is that science becomes politicized, and he who controls the most OPM controls the results of science.

While OPM has the most impact on the objectivity of science, it is not the only thing to impact it. In my previous post, I mentioned that science is theory-laden. That is to say, people have to have a kind of structure already in place in order to do science. The fact is that while there are a few times when people discover something serendipitously, most discoveries will occur because someone is specifically looking for something. A simple example demonstrates this. If you were asked to read a news story and then, after you finished and the story taken away, were asked “How many proper names were in that story?” you would probably have a difficult time answering; yet if you were told “I want you to read this story and tell me how many proper names are in it” before you read it, you would easily be able to keep track of that information because you are actively looking for it.

But how do you determine beforehand what information is or isn’t important in a scientific experiment? It can be quite difficult. For example, suppose you were doing an experiment into what made the best baseball player and you found that the best (however you wish to define that) baseball players tended to have been born in the summer months. How would you know that this correlation is relevant rather than accidental? This is especially insidious because one can hypothesize many different reasons why the timing of one’s birth could be relevant: babies born in the summer were gestating through winter, and perhaps the extra stress of the mother going through cold contributes someway to their development; the Earth is located at a different place in its orbit during the summer months; babies born in the summer are born in warmer weather, so maybe they start out more active; etc. All these possibilities, however, are based on the first assumption that it is actually relevant what month a player was born.

You can see from that example how it is easy to get lost down many bunny trails once dealing with information that may, or may not, be correlated to your specific experiment. What you look for and what you discard as being irrelevant depend highly upon what your theory already leads you to believe.

And what is true for the individual scientists is likewise true for groups of scientists, especially as we relate back to funding. If those who control the purse strings of your experiments are convinced that the month you’re born in determines your athletic ability, then they will not pay you if you disagree with that thesis. But since that thesis seems harmless enough—how could it really impact anything? you may wonder—then even if you don’t hold to it yourself, it’s easy enough to say you do. Soon enough, you have scientific consensus established and anyone who disagrees is, by definition, unscientific for having rejected a consensus that was derived by OPM, not science.

This makes it all the more important for scientists to be clear about the sources of their funding (to disclose possible conflicts of interest) and for us to know what effects it may have on their research. And as we shall see in my next post, this is especially true for any scientific endeavor that claims “the science is settled.”