Philosophy of Science- Okasha, Chapter 3

In chapter 3 in Philosophy of Science by Samir Okasha, the author begins to with the idea of can science be explained or what exactly is scientific explanation? One theory is discussed in length, that of Carl Hempel’s covering law model of explanation. According to Hempel, “…scientific explanations are usually given in response to what he called ‘explanation-seeking why questions.’ i.e. why is the earth not perfectly spherical?’ (p.41). These questions often times have a logical structure to them, ‘i.e. a set of premisses followed by a conclusion. The conclusion states that the phenomenon that needs explaining actually occurs, and the premisses tell us why the conclusion is true.’ (p.41). By dissecting Hempel’s law, we find that first, the premisses should entail the conclusion, meaning the argument should be a deductive one. Second, the premisses should all be true and lastly, the premisses should consist of at least one general law (p.41). Let us apply this to an example:

Okasha explains: Suppose I want to know why the plant on my desk has died. It could be explained as follows, there is poor lighting in my study, no sunlight has been reaching the plant, sunlight is necessary for my plant to grow (photosynthesis) thus the death of my plant must be due to lack of sunlight (p.42).

This example fits Hempel’s law. However, Hempel’s law does have a few problems. For one, it does not solve the problem of symmetry. One popular example is the flagpole problem. Say you notice a flagpole casting a shadow 20m long and someone asks you why the shadow is 20m long (an explanation-seeking why question). Our answer would be, “light rays from the sun are hitting the flagpole, which is exactly 15 meters high. The angle of elevation of the sun is 37 degrees. Since light travels in straight lines, a simple trigonometric calculation shows that the flagpole will cast a shadow 20 meters long. This sounds like a good explanation and fits the covering law model, however, if we try to switch around the explanandum (the phenomenon to be explained) it no longer makes sense. Suppose instead we ask why the flagpole is 15 meters high. The length of the shadow and the height of the sun have nothing to do with why the flagpole is 15 meters high.

Another issue with Hempel’s law is the problem of irrelevance. John has been taking birth control pills for a year and has not become pregnant. Although we know birth control pills help prevent pregnancy, is the best explanation as to why John has not become pregnant because he has been taking the pills? Obvious answer would be no, John has not become pregnant because John is a male. Although the covering law of explanation is applied, it does not take into consideration the relevance of the information given. 

So, although Carl Hempel tries to bring some sort of order to defining what is scientific explanation, there are some clear problems with his theory. Hempel’s law relies too heavily on successful prediction rather than true causal explanation. Correlation or covariance does not automatically ensure a true conclusion. Thus Hempel’s theory cannot be relied upon fully.