Science is by definition a very conservative discipline, and scientists are extremely hesitant to say that they are certain about anything. In fact, one of the rules of science is that you must always allow for the possibility that additional data will force you to let go of an idea that you were really sure about and quite possibly very fond of. Scientists take this rule very seriously — so seriously, in fact, that before they get their diplomas and lab coats, they have to pinky-swear that they'll follow it.
Suppose that you want to know whether a particular coin is a fair coin, that is, as likely to come up heads as tails when you flip it. So you flip the coin a few times, and it always comes up heads. The fact that you got heads in your first few flips may lead you to suspect that the coin isn't fair, but because you flipped it only a few times, you can't say for sure. What you have is a suspicion. What you need is more information. So you proceed to flip the coin all day long, and the next couple thousand times you toss it, the coin always comes up heads. With a few thousand flips under your belt and a head coming up each time, you conclude that the coin isn't fair. This process seems straightforward, and the conclusion seems obvious — to anyone who isn't a scientist.
But a scientist in the exact same position and with the exact same data would not say that the coin isn't fair. She would cite a very, very, very low probability that it's a fair coin, or she'd say that she's 99.99999 percent sure that the coin isn't fair. The reason for the difference is the precision with which scientists state what they know. After all, it is possible that a fair coin could come up heads that many times in a row; that possibility just isn't very big.
To which the layperson may reasonably respond, "So you're not really sure!" What the scientist is really saying is that she is sure — within a reasonable doubt. (If that explanation's not good enough for you, consider the shaky ground on which it puts the entire American justice system, which also relies on the standard of reasonable doubt.)
You say toMAYto; I say toMAHto: The language of science
Do scientists ever seem as though they speak their own language? In a way, they do. You probably recognize some words as scientific terms — words like nucleotide, paedomorphosis, heterozygosity, and others, which seem to be little more than Latin and Greek roots randomly strung together. But other terms — ones that mean one thing to laypeople and something else to scientists — are a bit trickier. The subtle differences in the way scientists and nonscientists use the same words are often sources of confusion.
You won't find a better example of this situation than the word theory. In scientific terms, a theory is a hypothesis that has overwhelming support — in essence, an idea that's been proved. In layman's terms, theory typically means best guess. See the problem?
To help you understand the scientific meanings of science's three most important words (the fourth is funding), I offer these definitions:
^ Fact: Something you can observe or measure.
^ Hypothesis: A working idea or set of ideas resulting from observations and measurements. The hypothesis serves to guide future investigations. It gives scientists suggestions about what facts they should try to collect next.
^ Theory: A conceptual framework, tested repeatedly but not rejected, that explains the facts, observations, and measurements, and makes accurate predictions of how the system will behave in the future.
The facts of evolution, as I show throughout this book, are clearly established. The current theories of how evolution functions are solidly supported as well. But the linguistic difficulties in communication are such that people continue to ask scientists, "But it's just a theory, right?"
Scientists don't start talking about something as a theory until it has overwhelming support. How do they get that support? Through scientific investigation. A coin toss is a good example of the process of scientific investigation:
1. Start with some observations about the natural world.
In the example of the coin toss, you observe that the first few tosses always end up heads.
2. Formulate a hypothesis.
Your hypothesis after flipping head after head? That the coin isn't fair.
The hypothesis serves as the scientist's starting point; maybe it's right, and maybe it's wrong. They key is to do enough testing to find out.
3. Gather additional data to test this hypothesis.
In the coin example, you gather additional data by tossing the coin several thousand more times.
Repeating one type of test ad infinitum — exhaustively flipping a coin, for example — and getting a particular result isn't good enough. First, you have to address and eliminate any other factors that could affect the test results. Maybe the coin is fine, but something about the person flipping it isn't quite right. Or maybe the coin is fine, but some other factor — like an air vent blowing air over the researcher's head — keeps it from landing heads or tails half of the time. You need to have the coin flipped by a bunch of other researchers in other parts of the room to make sure they get the same result.
4. As your data accumulates, it either supports your hypothesis, or it forces you to revise or abandon the hypothesis.
In the coin-toss example, heads continue to come up, thus lending support to the original hypothesis that the coin isn't fair.
5. At the point where an overwhelming amount of information starts to accumulate in support of the hypothesis, the hypothesis is elevated to a theory.
The hypothesis must get tested and tested to the best of everyone's ability before it arrives in the exalted land of theory. And even a theory is only one good experiment away from being rejected, which is one of the fundamental components of the scientific method: that the ideas scientists come up with must be falsifiable. That is, scientists must be able to imagine some set of results that would cause them to reject the theory; then they must see that over and over again, they never get the expected results. This process always sounds somewhat backward to nonscientists, but that's just the way scientists do things. We scientists never say that an idea is true; what we say is that, even after our best efforts, we've been unable to show that it is false. Then it's high fives all around, and we go grab a beer.
Was this article helpful?