After the skeptic

So, we can rather safely conclude that we cannot be certain about anything. I’m not absolutely certain we can’t. A particular fact which I am unsure whether it can be certain or not is, ‘Something exists, have existed, or will’, which seems rather solid. But the question itself is very peculiar, so I wonder if it at all can be properly comprehended. And it might be possible to construct thoroughly bizarre models in which this is not true, despite the phænomena we so commonly perceive – I do not know, however ridiculous it sounds.

But just because we cannot be absolutely certain about anything, it does not follow that we can only be equally certain about everything. And furthermore, radical skepticism is a very unproductive position, one which we cannot really take seriously. It is paper doubt, not real and living doubt.

But from a position of absolute agnosticism, how do we proceed?

I would suggest to make a few tentative assumptions about how we can gain knowledge. However, care must be taken at this stage. There are several kinds of assumptions – there are working assumptions, methodological assumptions, axiomata, et cetera. Here I mean a methodological one. We can test a working assumption (“I assume that there is vacuum in this tube now”), but we cannot test a methodological one (“I assume my perceptions not to be systematically misleading”). We might always reach a point where a methodological assumption may become testable; nothing is not open for future revision.

Once we have a sufficiently strong groundwork of assumptions, we can from them deduce new truths, which will hold given the assumptions. We can construct arguments. The strength and necessity of arguments lies in that we can only be wrong in the concluson if we are wrong in an assumption. Certainity becomes inherent in the system.

For each assumption we make, there is a risk, even if likely vanishingly small, we may be wrong. This is true for all assumptions. What we then need to do is be parsimonious, that is, raze with Occam’s razor. Never assume more things than you need to make your model, explanation or theory work.

Which assumptions we initially make may fall as it does; with any luck they prove themselves fruitful or not. To assume logic, to assume we are not systematically misled in thought and perception, to assume that we need the ability to transfer knowledge and not acquiesce in mystic epistemic individualism, and to assume some kind of coherency in the reports we give and the reports people think they hear or read from us are examples of good starting points. Now, what is important are a few assumptions we should not make:

1. Realism

Realism is the position that our theories describe real things. In many ways, this may seem as if it is the natural conclusion via Occam’s razor, but in fact it is not. To assume that our theories describe real things is positing a new set of entities. We do not need them to make our theory work. To posit that strings exist when this is an untestable proposition, when we can just as well say that the maths perfectly describe the real observable events, is to make an unjustified, uneconomical assumption. They could exist or they could not. We do not need them, however.

As Sokal and Bricmont wrote in Intellectual Impostures:

This is rather like asking whether, given a finite set of points, there is a unique curve that passes through these points. Clearly the answer is no: there are infinitely many curves passing through any given finite set of points. Similarly, there is always a large (even infinite) number of theories compatible with the data – and this, whatever the data and whatever the number. (p.80)

This is not to say that realism is useless. Having a realist interpretation has often been very useful for experimenters, made theories more relatable and easy to understand, and perhaps more beautiful. One can perhaps argue about them at a level of very low, but still present, certainity. Materialism and dualism both do seem like bigger stretches than does neutral monism. But it is important to be aware that strictly speaking, they are all less rigorous and solid positions than antirealism.

2. Common Sense

Common sense is a very common method of estimating truth. It’s, er, in the name. Regardless, it is not good to utilise this, for the simple reason it is not common enough, and often at odds with fact even when it is. I doubt it is common sense that the denial of the antecedent in implicature means the implicature is true regardless if the consequent is (that is,  the sentence “If bananas are a type of worm, then the Prime Minister of Sweden is Fredrik Reinfeldt” is true). Nor would it be common sense to say that 0.999… = 1, when in fact it is. Many more examples may be shown of common sense erring.

Merely for the fact it feels right or it is common people think it right, it does not follow it is sensible to believe it.

3. Intuition

Intuition is hard to account for. It’s hard to give any reason it should be right; it merely feels that way. It may be – but it constitutes a poor argument for the matter. It varies among people and certain types of it may be said to be common sense of one variant or another. The veracity of an intuition is not good to judge on basis of it being one, but rather on its own merits. Some intuitions may be useful, but as a general area of source of knowledge, it seems to be poor rationality to accept it. After all, intuition is not transferable and it only makes sense to the person who has it.

Again. Merely for the fact it feels right or it is common people think it right, it does not follow it is sensible to believe it.

A note about procedure of knowledge: One must make a distinction between a source of knowledge and a justification of it. What I am concerned about here is justification. As far as reaching new facts goes, any method could and should be used. One can never know from where we may get new facts to incorporate. A mathematician may very well intuitively feel that a certain answer to whether the Riemann hypothesis is correct; then the intuition would be the source of that knowledge. But it would not be justified until he produced a proof.