Our Epistemology So Far

This is a historical post

Well, we’ve covered a lot of ground in past posts. The problem with this ‘Reason as a Guide to Reality’ series of posts is that they build on concepts from past posts. It’s easy to get lost in all the concepts. So let’s do a quick review of past ideas and build up the principles of finding truth/knowledge (i.e. Epistemology) that we’ve determined so far.

First, everything we thought we knew about science turned out to be false. Namely, science is not specifically about predictionnor reductionism, nor holismnor observationnor falsification. All of those ideas are important to science, but they do not delineate a boundary for science.

Second, science is not justified by inductive thinking. The past does not determine the future. Instead, science (and all knowledge actually) is justified based on being our best explanations so far. No other justification is necessary and no other justification is possible.

Third, what constitutes an explanation is not ‘reducibility’ in some physical sense, but rather reducibility to an algorithm. (Though this alone is not enough either.) That is to say, all things in the physical universe that can be understood are understood through being able to convert them to computation. (At least so far.) 

Fourth, the primary means of gaining new knowledge is through coming up with conjectures (theories) and then over time refuting them. This forms a ‘natural selection’ or ‘survival of the fittest’ of ideas and explanations of reality, thus guaranteeing scientific progress that gets closer and closer to reality.

Fifth, refuting of a theory must always be in favor of a new and better theory. We do not refute theories by merely pointing out explanation gaps in existing ones but instead by finding a better one that explains more. A ‘rejectionist’ approach that only points to problems in the existing theory will never dislodge the prevailing theory on it’s own.

Leave a Reply

Your email address will not be published. Required fields are marked *