Article Review:The Truth Doesn't Explain Much

From: Introductory Readings in the Philosophy of Science

SCHA 250/2 aa

Jeremy Chapman

3726223

Cartwright's article The Truth Doesn't Explain Much states that the aim of science is to explain natural phenomena. She goes further to show that scientific theories have two different and distinct functions: showing us what is true in nature, and giving us a means of explaining it. Cartwright argues that the two are often intertwined and many think that explanation of an occurrence is a by-product of finding what is true in nature. The problem with the theories of such people as Hempel, Lambert & Britten, Salmon and Hanson is that they confuse the idea of true in nature with the explanation of the phenomena. Cartwright thinks that this is because they assume that laws are true, and therefore, any explanation that comes from laws must somehow capture the truth in nature. She says that this mistake is fostered by the covering-law model of explanation.

The covering-law model of explanation as formulated by Hempel assumes that laws are true and that with the laws, some logic and some probability theory, one can know which factors can explain which others. In other words, a law covers all possible phenomena. Whereas much of the criticism of Hempel's model was that it took in too much, Cartwright rejects his model on the account that it lets in too little. She says that with a covering-law model, we cannot explain such things as the role of DNA in the inheritance of genetic characteristics. We cannot explain these phenomena with a covering-law model because we do not have a law that covers them. Most phenomena that have good scientific explanations are not covered by any laws. Cartwright states that many phenomena are covered at best by ceteris paribus generalizations. These generalizations only hold under specific conditions, such as a perfect vacuum or at absolute zero. Here is the crux of Cartwright's argument against laws as truth. Since these ideal initial conditions are rare at best, these laws apply to a very small number of phenomena indeed. Ceteris paribus generalizations, read without the "ceteris paribus" modifier, are false. Since they are false, the covering-law model can no longer use them, for they are acknowledged to be false. Therefore, most of the time we have a law that claims to cover, but cannot explain because it is acknowledged to be fake, or we have a law that does not cover. Cartwright sees this as refuting the covering-law model.

Ceteris Paribus Laws

Cartwright here shows how there are no exceptionless laws. In fact, she says that every theory ever proposed in Physics, even at a time when it was popular, was known to be deficient in some specific and detailed way. The problem is that some laws get treated as if they were exceptionless, and others don't even though they are "on the books". As an example, she uses Snell's law.

Snell's Law: At an interface between dielectric media, there is also a refracted ray in the second medium, lying in the plane of incidence, making an angle t, with the normal, and obeying Snell's Law: sin/sint = n2/n1 where v1 and v2 are the velocities of propagation in the two media, and n1=(c/v1), n2=(c/v2) are the indices of refraction.

This theory, it is revealed later in the text, applies only to a specific situation where the media's optical properties are isotropic. In anisotropic media there will be two transmitted waves. This is an example of a ceteris paribus law, holding only in special circumstances. Cartwright goes on to suppose that the unrefined Snell's law is not to be understood as a universal law, but rather as some type of statistical law. We could say For the most part ..., but this doesn't work either. Most media are optically anisotropic, and therefore would produce two rays. Cartwright's point is that if ceteris paribus laws are to be true laws, they cannot be identified with statistical laws.

When Laws are Scarce

In this section Cartwright asks why do we keep such laws as Snell's law on the books when we know that it is both false and that it has a more accurate refinement available. The reasons for this have to do with the task of explaining. The specification of factors that are explanatorily relevant to other factors is science's task more than laying out the laws of nature. Even when the laws of nature are known, we must decide which factors to cite in our explanation. Cartwright states that one thing ceteris paribus laws express is our explanatory commitments, or what kind of explanations are permitted. For Snell's law, to leave it on the books is like signalling that the same explanation can be given for some anisotropic media. The explanation that is derived from the ideal situation is employed even when the ideal conditions are not met. So, Cartwrights says that we may use these laws to get close to the correct explanation, but often we have no law that deals with what happens in less than ideal conditions. Covering-law theorists, in Cartwright's view, will say that ceteris paribus explanations are stand-ins for true laws that are not yet known. When a false law is used, a covering-law theorist would assume that we are making a bet about what form the true law would take. Cartwright has two major problems with this idea. First, whereas covering-law theorists tend to think that the universe is well-ordered and that there is a law to cover each case, Cartwright sees natural object in the same way she perceives people in societies. Specific laws and general principles do not dictate their behaviour. What happens most of the time is not dictated by any law. Secondly, the elliptical explanations are not explanations, they are sometimes vague assurances that there are explanations to be had. The law that is supposed to be in a correct D-N explanation is not a law from a theory, it is untestable. Finally, the laws of nature are not enough to tell us what explanations can be given at that time. So, covering-law theorists make a decision to wager about the existence of unknown laws. The grounds for accepting these unknown laws are only as good as our reasons for adopting the corresponding explanatory strategy.

When Laws Conflict

Here Cartwright maintain that there are not enough covering laws to go around. There are many theories about what happens within specific scientific domains, and yet very little theory about what happens when these domains intersect. For instance, a ceteris paribus law would stats that adding salt to water decreases the cooking time of potatoes, whereas taking the water to a higher altitude increases the cooking time. We must be more specific, and say that when the altitude remains constant and salt is added to water this decreases the cooking time; while increasing the altitude at a fixed saline content increases it. Yet neither of these tells what happens if we both add salt and increase the altitude. Cartwright points out that most people would assume that there is a precise answer about what would happen, even if it is not common knowledge. But this is not always the case, most phenomena involve a complex combination of causes, and the general laws to describe what happens in these complex cases is not always available. For example, both quantum mechanics and relativity are advanced theories, there is no satisfactory theory of relativistic quantum mechanics. Therefore, we see that at the intersection of theories, laws are hard to find.

When Explanations Can Be Given Away

So far, Cartwright has argued simply that covering laws are wanting and that ceteris paribus laws are not true laws. The argument now is that ceteris paribus laws have a fundamental explanatory power. Her example uses her camellias: camellias like rich soil, so they were planted in manure, but they do not like high temperatures, and the manure was warm. Many camellias died, so cartwright could say that it was because she planted them in hot soil. Although one cannot be sure that this is the correct answer, since she made a reasonable effort to eliminate other sources of discomfort for the plants, we may have confidence in our explanation of their death. This example has no true covering law, it is simply an explanation; and it is the job of science to tell us which explanations are permissible. Cartwright's thesis is thus: if the world is not a deterministic system, then the job of how to explain will still be there when the descriptive part of science is over with. If it is possible to know all the facts about camellias that there are to know, one would still not know how to explain what happened in the garden.

In conclusion, Cartwrights says that most scientific explanations are ceteris paribus laws. If these laws are read literally, they are false, and deemed false even in the context of use. She says that trying to find unifying laws will be hard in a universe that may be varied and diverse. In short, the deterministic view of her contemporaries leads to false views of explanation and the acceptance of ceteris paribus laws which are deemed false by their own users.

Text Consulted

Essay: Nancy Cartwright The Truth Doesn't Explain Much

In the book: Edited by E.D. Klemke, Robert Hollinger, David Wyss Rudge Introductory Readings in the Philosophy of Science.

Prometheus Books, New York 1998