Neo-Naive-Baconianism?

The philosophy of science has a long history. But it's modern history has it as a goal to understand the underlying logic of science, the scientific method, and scientific discovery.

The modern history of the philosophy of science starts with Francis Bacon. In what has now been dubbed "Naive Baconianism" the theory went that scientists amassed data and a theory somehow emerged about the data.

This was subsequently replaced by Popper's falsificationism where the claim is that scientists ever attempt to falsify their theories by proposing falsifying hypotheses and then testing them, ultimately either offering corroborating evidence or a falsifying instance.

Another competitor for a theory of how science works is Kuhn who said that science works in paradigm shifts. That is, there is a research program and scientists work with it until something better comes along.

Fayerabend also offered an alternative, so did Lakaos.

But what we are seeing now is the emergence of a new way of looking at the logic of science - the petabyte method. We take a massive amount of data - more than we ever could have handled previously and see what data emerges from that. The method allows for all sorts of unorthodox results, and much of it is largely independent of the paradigm that a given science works with. Of course the computer models still make assumptions about the data, but they are able to crunch the numbers in a way that is irrespective of what the numbers mean. It is very "structuralist" or "Hilbertian" in that way. I suspect that philosophers of science will have to start rethinking their views on Bacon, as scientists increasingly rely on huge data sets and computationally intense formulations of results.

More on this to come.

Review of Ladyman's Understanding Philosophy of Science

James Ladyman's Understanding Philosophy of Science is a pretty good beginning text in the philosophy of science. It is very well organized, and as straight forward as can be. There are two main sections. The first deals with the scientific method, the second deals with the question of realism.

The first section goes through the various stages in the philosophy, especially the question of induction, from the rather naive position of Francis Bacon through falsificationism and the theory of scientific revolutions.

The second section is about the question of realism, especially the problems created by underdetermination, explanation, and inference to the best explanation. There is a very good focus on van Fraassen and the nature of constructive empiricism.

This book is an excellent choice for a first introduction to the main questions in contemporary philosophy of science.

Review of Wesley Salmon's Four Decades of Scientific Explanation

Four Decades of Scientific Explanation Is an excellent overview of the history of scientific explanation as seen from the standpoint of one of its most significant contributors.

Since the late nineteenth century is is generally agreed that science provides explanations of the phenomena in the natural world. But exactly what does an explanation look like? What form does an explanation take? How do we recognize an explanation and distinguish it from pseudo-explanations and the like?

Salmon's history covers the discussion from Hempel to the state of the art in 1989 when the book was written. He covers all the major theories and all the major questions. The book is highly readable, and definitely a worthwhile investment for those who want to understand this corner of the philosophy of science.

Review of Garfinkel's Forms of Explanation

Alan Garfinkel's Forms of Explanation is a bit dated and I am not sure what to make of it.

Chapters 1 and 2 are certainly worthwhile reading. They offer the beginnings of an interesting discussion of contrastive explanations and one account of what it is to be a reductive explanation. There are actually many more kinds than Garfinkel describes, but no matter. His is interesting enough. Chapter 3 peters out to some warmed over naive Marxism/Rawlsian stuff. There for some reason he pulls a fast one and pretty much tells you he is going to conflate explanations and justifications, and you should live with it. I suppose one can't argue with that.

Chapter 4 he builds up to a position that can probably be paraphrased as "given that we live in a society (with relations between individuals) you can't explain the relations in society by appealing only to individuals.

The book ends with what we'd now call a pragmatic account of explanation. Garfinkel was probably not aware of the work by people like Achinstein on this very notion, or he just didn't bother mentioning it. And the real statement of pragmatic explanations was published about a year before by van Fraassen, and he was most likely not aware of that either.

I am hesitant to recommend this, but the book is about the question of explanation in the philosophy of the social sciences. There is not all that much out there on this topic and you can do worse, but perhaps skipping chapter three would make it less annoying to read.

What is a planet?

We all live on a planet, and most of us can name most of the other planets in our solar system. But what is a planet? This is in some sense a real philosophical question. That is, it is a question that does not depend on the scientific facts of the matter, but rather on a set of definitions, and a "conceptual analysis" of what planet is. Again, it is not a scientific question, but a philosophical one.

It is good to see philosophy of science making the news, and it is good to see that astronomers are managing this question just fine without philosophers of science, and have a tentative definition, albeit not without its own troubles. This should be fun to watch.

As far as I know, there is little literature on the philosophy of astronomy, and hopefully this will change things.

The debate hinges on the following points: There are thousands of things orbiting our sun. They are of various sizes. They are of varied distances from the sun, and they have different types of orbits. They have different shapes too. They are effected by different celestial forces.

It will be important to doa few things. First, the number of planets, whatever it amounts to, is kept small. It would be too much of a mess to call everything a planet, and it would seem to miss the point. Second, everything we end up calling a planet has to have certain features in common. If not, we cannot have a real definition. Third, the traditional planets should be retained as planets. We do not want Earth to suddenly not be a planet because of our new definition.

Some earlier proposed definitions had eight planets, and excluded Pluto. And here is where politics infringes on science. Standard earlier definitions exclude Pluto as a planet . But Pluto is the only planet (of the three that have been discovered) that were discoverred by an American. Americans have a lot of say in science. American scientists, and Americans in general will want to make sure that there is at least one planet that is forever associated with American science. So we can be assured that as long as the US is a superpower, Pluto will be a planet.

Again, ultimately this is a philosophical question, one that nothing serious hinges on it. Some debates in science are much more significant, and the outcomes do make a difference, Some debates in the philosophy of science are serious and make a difference in setting a scientific framework, like the debate between cladists and pheneticists in biology.

We will likely realize in the long run that no two celestial bodies have that much in common, and we will use extensional definitions of planets instead of intensional definitions. We will just call a planet anything that we have been calling planet, and anything that the term planet catches on for, instead of trying to come up with a clear definition of what a planet is. If philosophy of sience has taught us anything, it is that a definition that includes all the cases we want, and excludes all the cases we don't want is never forthcoming. Nature is just not that clean.

Review of van Fraassen's The Scientific Image

Bas Van Fraassen's The Scientific Image is a very important work in contemporary philosophy of Science. It is the definitive statement of non-realism in science. The central claim is that the central claims of science are 1) empirical adequacy and 2) that a theory must fit the observable phenomena. There are three main theses that are given that show how a workable non-realist philosophy of science can handle the important issues in the philosophy of science. The three things are 1) there is a relation between theories and the empirical world, 2) there is no absolute notion of explanation, and 3) there is an empirical account of probability. The book is quite difficult and only for the serious student of philosophy of science.

While I tend toward realism in science, it does present strong reasons to take the non-realist position seriously. It is a real serious work, and I strongly endorse it. You must keep in mind though that this is not a popular account, and is not for the faint of heart. It is a very serious work in the philosophy of science.

While a real elaboration of the book is too long to go in to here, I thought I would just give a quick outline of the part of the book which deals with scientific explanation. This is likely to be the most enduring part of the book, though I the whole thing is worth reading and preserving.

An explanation is an answer to a why-question. So a theory of explanations must be a theory of why-questions. The why-question Q expressed by an interrogative in a given context will be determined by three factors: 1) The topic, 2)the contrast class, and 3)the relevance relation.

Furthermore what we need to understand this theory is 1) the contextual factors involved in the background of why-questions. (Those will delineate the set of answers) And 2) A way to evaluate the answers we get.

To evaluate the answers we look to other theories of explanation. For example, we can look to see if the answer addresses the rise in the probabilities of events happening to explain why they happened. This of course has all the shortcomings of other theories of explanations.

The upshot is that there is no right account of explanation. Rather, it depends on what the context is. Sometimes the length of the shadow really does explain the height of the tower. Sometimes the laws of physics and trigonometry explain the length of the shadow, and sometimes the length of the shadow is explained with reference to who van Fraassen's father slept with. Presumably this all goes to counter van Fraassen's realism, in the sense that there cannot be one account of explanation because there is more than one account of reality.