Problems In The Philosophy Of Science

There has been much conversation lately in my life about the problems in the Philosophy of Science, and I thought I'd explain some of the things that I've discovered.

1. The basic problems of empiricism. An a posteriori belief is a belief based upon our experience. Most people have had times when their senses seemed to deceive them; e.g., Daniel wrote a post once about driving at night and having his senses deceived. He thought he saw a person on the side of the road, and slammed on the breaks. He looked again and saw only a trash can. He has then come to the belief that some a posteriori beliefs are mistaken: either there was a person or there was not. How, then does Daniel seek to rectify this problem? He can get out of the car, of course, and look around to see if there is a person hiding behind the trash can, or he may reason to the best conclusion. Surely, such an isolated incident isn't enough to give cause to question one's senses.

In fact, there is a general principle that many philosopher's use that states: believe your senses unless you have explicit reasons at a particular instance not to believe your senses; in that case, you might question that particular instance. They claim that even if your senses deceive you occasionally, there is no reason to completely reject them. However, how do we perceive a witness in court that has perjured himself? Do we not cast doubt on everything this perjurer says? But there is a more technical reason for rejecting the reliability of one's senses.

Our situation seems as follows: we accept the proposition that some a posteriori beliefs are mistaken (for this is why we even have a problem to begin with); we have a questionable a posteriori belief that we are testing to determine whether it is mistaken or not (was there a person or just a trash can on the side of the road?); and we seek to test this questionable a posteriori belief with another a posteriori belief. Do you not see the vicious circle? We are testing questionable a posteriori beliefs with other a posteriori beliefs whose credibilty are themselves in doubt (because we accept the principle that some--we don't know which ones--a posteriori beliefs are mistaken). This isn't just a small problem of trusting our senses excepting those times when we are consciously deceived; perhaps those times when we think we are deceived are the verdical beliefs, whereas the vast majority of our life we are being deceived. Who can tell?

2. The nature of scientific explanation. With what I've studied so far, I've not been able to discern any consensus on what makes an explanation scientific. The starting point for contemporary philosophers of science is with a man named Hempel. He believed that something is scientifically explained by setting up the discussion in a deductive syllogism. As one of your premises, you start with a law or laws of nature; the second premise would be your initial conditions (the conditions related to the event to be explained). Your conclusion would then be the event to be explained. This would work as both a predictive method and an explanatory method. For instance, imagine that there is a flag pole that casts a shadow of a particular length at a certain time of the day. What is the explanation of this phenomenon?

Hempel would assert that one should assert some laws of optics and the like, and then assert the initial conditions: namely, the degree of the height of the sun in the sky, the height of the flag pole, and etc. With this, one may deduce the length of the shadow and thus explain why the shadow is the length that it is. But, is there a problem here? In fact, there is. Let us assume the same situation, the same laws asserted and the same initial conditions, except that we exchante the height of the flag pole for the length of the shadow. Now, from this syllogism we can deduce the height of the flag pole; so now the length of the shadow explains the height of the flag pole! Even more queer is if we accept the same laws, but the initial conditions now include the height of the flag pole and the length of the shadow; we can now explain the degree of the height of the sun in the sky!

Well, suffice it to say, no one really holds to Hempel's account today. One of the leading possibillites to replace Hempel is Salmon's statistical relevance model (S-R). This states that a scientific explanation simply asserts every statistically relevant factor in regards to the explanadum. For instance, why did the mayor get paresis? (Paresis is a disease that results from untreated syphilis, however it occurs in only small percentage of cases of untreated syphilis). The S-R response is simply that the mayor had syphilis. There is no other statistically relevant factor; this has been scientifically explained. However, how do we explain why the mayor got paresis whereas his wife (who also had untreated syphilis) did not? Salmon cannot offer an explanation.

Bas C. van Fraaseen (which is a rad name) offers an account that seems radically different from any previous account. He states that an explanation is simply an answer to a question; the important step is finding which question is being asked. For example, if I were to ask "Why did Mrs. S shoot Mr. S?" I could be asked several questions:

1. Why did Mrs. S shoot Mr. S? - Why Mrs. S did it and not someone else.
2. Why did Mrs. S shoot Mr. S? - Why Mrs. S shot him instead of kissing him.
3. Why did Mrs. S shoot Mr. S? - Why Mrs. S shot Mr. S and not the maid.

In order to determine what question is being asked, one first names a topic: P(k). For an example, let us take question 1. The topic is something like "Mrs. S." Next, we need a contrast class: X. The contrast class X would be something like: Mrs. S; the main; the butler; the mail man; the mayor who has paresis; etc. The contrast class is basically saying: "Why Mrs. S and not any other member of X; so, why did Mrs. S, and not the maid (for example), shoot Mr. S?" The last part of van Fraassen's explanatory account is the relevance relation. It it this relevance relation that "picks" out the topic from all the other options in the contrast class. Thus, it seems, a scientific explanation is a story and we need to supply the missing part of the story in order to give an explanation.

Some have objected to van Fraassen's account because he does not give any real criteria for determining what a relevance relation must be. In fact, he says that a relevance relation must not pick out only the topic from the contrast class, but must just weed the selection pool down. One counter example given to his account of explanation is the question "Why is the sky blue?" The topic is "blue" and the contrast class might be: blue, white, red, yellow and etc. The relevance relation could be: Because the grass is green. This satisfies all requirements of van Fraassen's account. It seems to fail.

Well, I'm not going to go on about explanation, but my point is this: there is no consensus as to what makes an explanation scientific versus pseudo-scientific. This seems a problem.

3. The nature of causality. While many philosophers of science have sought, because of the inherent problems, to remove causality from their theories; however, some still insisnt on using the idea of causality. Briefly let me state this: in asking what caused something I am asking for the necessary and sufficient conditions that brought about the event in question. Suppose, for instance, I ask why the match lit when I struck it on a matchbook. What are the necessary and sufficient conditions for this to happen? A necessary condition is a condition that needs to obtain in order for an event to take place (e.g., it is necessary that a mother be a woman). A sufficient condition is a condition that if it obtains the event in question takes place (e.g., one might say that a sufficient condition for the ground being wet is for it to be raining). A necessary condition need not be sufficient; e.g., our example concerning a mother: while a mother must be a woman, being a woman does not necessitate that one is a mother. A sufficient condition may not be necessary; e.g., it does not need to be raining in order for the ground to be wet, there could be sprinklers.

A set of conditions that are both necessary and sufficient for a given event seem to be a cause of that event. So, let us harken back to the match that lit. What caused this match to burn? What are the necessary and sufficient conditions? Did striking the match cause it to burn? Is striking a match necessary and/or sufficient? If a match is placed in a chamber and the temperature is raised to the temperature at which paper kindles, the match will combust without the aid of being struck. Thus, being struck is not necessary. If a match is placed in a chamber without oxygen, and is struck repeatedly, the match will not burn. Striking a match is not sufficient. Does striking a match then play any part in causing the match to burn? We want to say, "yes", but our reason may lead us to say "no". This is similar to Zeno's paradoxes of motion.

What is causality? Hume's example is of billiard balls. When we see one billiard ball hit a second billiard ball, the second ball starts to move. Did the first one cause the second one to move? All we see is the first hitting the second, and then the second moving. How is this causality? Last night I turned off the lights in my living room, and a few moments later it started to snow outside. Surely this is causality. We might object that the lights had nothing to do with the snow; but how do we know this? What do the billiard balls have to do with each other? Hume answered that because of spatial contiguity and temporal succession we come to expect this regularity and then because of a hasty generalization we call it "causality." Can we escape this argument? Not all scientific explanations rely on the notion of causality (some reject it outright), but some do; certain popular science thinks that causality is important. Any Christian who believes in the cosmological argument must believe in causality; how can one prove it empirically?

4. The nature of laws. A professor of mine has stated to be that his friends in the physics world admit that physicists have not discovered one law of nature. (This, of course, distinguishes between laws of nature and laws of science). More interesting (at least to me) is not whether physicists have found any laws of nature, but whether there is even such a thing as a law of nature to be found. What is a law? There are several views which seek to explain the nature of laws, and they all, seemingly, have fatal flaws. One account which is sometimes call the naive regularity view by its opponents states that a law is a law if it can be stated in a true universal conditional. For instance, Newton's first law is something like: "If the net impressed force on an object is zero, then the object moves inertially." This can be stated (just for you Tim) in the predicate logic form of: for all (x)(Fx->Gx). In English: for all x, if the net impressed force on x is zero, then x moves inertially.

Now, this is a true statement because the conditional is true. Why is it true? Not because the antecedent and consequent are both true, but because the antecedent is false. If the antecedent is false, it does not matter whether the consequent is true or not; the conditional is true. There is no object (according to modern physics) that has a net impressed force of zero. So, let us consider an example: If the next impressed force on the moon is zero, then the moon moves inertially. Now, the net impressed force on the moon is not zero, therefore it doesn't matter if the moon actually moves inertially or not; the conditional is true, just not true of anything in reality. One may show the absurdity of this by giving another "law": If the net impressed force on the moon is zero, then the moon does NOT move inertially. This, also, is a law. Want to get more absurd? If the net impressed force on the moon is zero, then the moon is made of cheese. This is a true conditional; a law of nature. Well, perhaps these aren't laws of nature, but how does one distinguish between a law of nature and a pseudo-law?

There are other accounts of laws such as modal necessity. Modal necessity states that something is a necessary truth if and only if it is true in all possible worlds. However, the laws of nature are not true in all possible worlds (although, personally, I don't like the notion of possible worlds), for certainly the force of gravity could have varied inversely as the cube of the distance and not the square. It is conceivable. However, proponent might wish to say that the laws of nature are not logically necessary but nomologically necessary. This reduces to the truism that the laws of nature are true in all possible worlds where they are true. Another solution must be found, but there is no agreement as to the solution. My prof stated our problem in the following technical language: "We're screwed."

5. Realism versus anti-realism. I shall comment on briefly on this issue in the Philosophy of Science. Realism states that it is the aim of science to literally describe the universe. That is, if one is a realist, one need not think that a particular theory actually describes the world, but one believes that it is attempting to do so; when one accepts a particular theory one accepts it conditionally that it is attempting to describe reality. The anti-realist states that it is not the goal of science to literally describe reality. There are two main types of anti-realists: those who think that the theories in science do indeed explain reality, but only by taking the scientific theory metaphorically (or only after demythologizing it); and those who think that scientific theories ought to be taken literally, but not take them as true. Thus, if science asserts the existence of elections the two anti-realists will take this assertion as follows: 1. this is just a metaphor or mythology for stating something true about the universe 2. this theory really does assert that there are electrons, but just because I am committed to the theory does not mean that I think that this claim is true.

Is the purpose of science to discover or to invent? This is a quesiton that even pagan philosophers argue over. It seems that far too often Christians assume that science is meant to discover.

6. Lastly, there is one figure in the Philosophy of Science that I'm wanting to read up on; his name is Feyerabend. I suggest looking at something like Wikipedia to get a very brief introduction to some of the "radical" things he has to say. Is he right?