As I’m trawling through more of Popper’s ‘Logic of Scientific Discovery’, I’ve come to learn more about this work than was putatively understood about it. As previously stated in my post on Popper, this work is far more than simply the ‘falsification thesis’ as normally construed. Falsification also is realised in a number of innovative ways (such as the ‘dimensionality’ of scientific sentences described previously). Falsification is also important in considering revisability.

Simplicity

Popper comes forward with a notion of simplicity, which goes something like: the more general a claim is and the more it can capture with a shorter expression, the better. This basically captures what simplicity is. This might sound obvious, but the contrasting position of conventionalism poses a situation where a thesis which is initially a generalised formulation may find ways of being contradicted and undermined, and in order for a body of theory to survive such empirical challenges it must introduce ‘auxillary statements’ to introduce consistency.

Comparing theories

If one is to compare one theoretical system with another, the factors for deciding one over the other concerns, what accounts for more truth and which does not. A simpler theory over one with too many auxillary hypotheses would be better because of the greater explanatory power of a simpler theory using less to explain just as much, or more. Auxillary hypotheses are also increasingly suspect as the greater number of caveats introduced are exactly introduced to avoid being contradicted by the real world. That is not to say that a real life theory may need auxillary hypotheses, after all, a physical theory needs to explain. These comparative factors are merely between idealised cases.

Probability considerations

I must admit this is a part I least understand. Popper introduces a set of probabilistic concerns that would establish a credence of a theory, some of the constraints are fairly non contraversial: a proposition should not conflict with other true statements, a claim that contradicts a true one should have a lower probability. Probability is introduced as having logical constraints. In this section Popper addresses the work of Richard Von Mises and Maynard Keynes in their work on probability. My initial suspicion was that logical constraints on probability were more about what is a factor in discounting the likelyhood of an event, rather than a positive thesis on how to construct probabilities. I perhaps retract this initial thought in the consideration of introducing two formal axioms of probability, which seem more logical than mathematical. I must make a note of connecting why people hold to much disdain of logical approaches to probability (and Carnap I add as a correlated philosopher on this issue), to the more contemporamous methods of probability in philosophy. Popper seems to introduce a distinction between ‘objective and subjective’ probabilities and seems to say that while objective approaches are better (as he so construes it), subjective approaches can be useful as well in certain cases.

I need to read more on probability. I am quite confounded in general about probability, but I see it as interesting that Popper introduces probability in an almost systematic way in his logic of science. Perhaps probability (or belief credence calculus) is an essential part to a system of science. We’ve come a long way from Kansas Hume in this issue.