Topics Science

Question Some people define a set of propositions as science only if they make testable (or perhaps falsifiable) predictions, and those preditions are verified. Is that a good working definition of science? If not, how do philosophers distinguish scientific claims about the world from non-scientific claims? (This question comes up in the current controversy over whether Intelligent Design is science.)

Accepted:
November 21, 2005

Comments

Scientific theories cannot be proven or disproven. For one thing, some of the data on which the proof or disproof is supposed to be based may itself be incorrect. (Francis Crick, of double-helix fame, supposedly said that if you theory fits all your data then you know its false, because you know that some of your data is false.) But even if all your data are true, scientific theories go beyond those data (that's their point) and so cannot be proven from them. And theories can not in general be disproven by data either. Although a theory may be used to make a prediction that is then found to be false, the prediction almost never followed from the theory alone, but only from the theory along with various additional assumptions (the instruments are working properly, there are no disturbing forces acting on the experiment, etc.), so when the prediction fails, you do not know for sure whether to blame the theory or one of the other assumptions.

Nevetheless, there is something to the idea that scientific claims should be in some way responsive to empirical evidence. But notice that science is not the only endeavor that is resp0nsive to empirical evidence, unless you have a very wide definition of science. Just think of the study of history, or the practice of cooking.

So it is not easy to find a feature that's found in all and only science. Where does that leave Intelligent Design? Its proponents appeal to empirical evidence. Some would say that it is still not science because it appeals to a supernatural entity and science is about the natural world. But proponents of Intelligent Design would probably deny that their theory is itself appealing to a supernatural entity (even if they believe in the existence of one): the intelligence could be natural, just extraterrestial.

At the end of the day, whether a particular theory should count as science is not the most important question anyway. It is not as important as the question of whether, science or not, we have good reason to believe the theory.

Just to amplify on the excellent point in Peter's last paragraph.The reason Intelligent Design (ID) shouldn't be taught in science classisn't that it's not science. (That debate leads to all sorts ofdreadful philosophical attempts to demarcate science from non-science.)The reason ID shouldn't be taught in science class is because it's alousy account of the phenomena it seeks to explain, a lousy accountwe're in no way committed to as it has a far superior competitor. Thereis religion lurking here but it's locus is often misidentified. Thereligion isn't in the claims of ID. Rather, the religion is in themotivation for pushing a lousy account into the curriculum.

Ithink there are two reasons why people shy away from this way ofputting the matter. First, if you call ID "lousy science", then itseems you've allowed the ID people a foot in the door, by acceptingthat their account is science. Science vs. non-science seems like amuch clearer and sharper dichotomy than better vs. worse science.(People are always tempted to look for simple, dichotomous ways ofputting things. But in the process they trip themselves up, becausethat's usually not the way the world is.) We should resist all this:let's just say that any collection of assertions about the workings ofthe natural world, however kooky they might be, can be called"science". Fine, ID is a science then (as is astrology, as are theaccounts of the Ancient Greeks). But what has a claim to being taughtin the science classroom isn't all science, but rather the best science, the bestclaims about the workings of the natural world. ID shouldn't be taughtin the science classroom any more than Ptolemaic astronomy and forexactly the same reason: they are both poor accounts of the phenomenaand both much improved upon by other accounts.

Thesecondreason has to do with politics and the local legal situation in theUnited States. The courts have had something to say about the constitutional guarantees of the separation of church and state. They've had nothing to say about theconstitutionality of teaching bad science. Hence, if you wish to usethe courts to stop school boards from introducing ID into thecurriculum, it seems you've got to make sure that ID isn't science. Idon't know the legal situation well enough to say, but can it beargued that the separation of church andstate is being violated not because ID is a religious doctrine butrather because its injection into the public educational curriculum ismotivated entirely byreligious considerations?