Operational knowledge and credibility
Knowledge, as defined by the Stanford Dictionary of Philosophy, is justified true belief. This definition, unfortunately, is of no operational use. Firstly, there is no proposition which can be regarded as being absolutely true. You cannot prove that you are not a brain in a vat. Secondly, the word 'justified' applies only to the process by which knowledge has been acquired, not to the uses to which knowledge is to be put. Given that the point of having knowledge consists in being able to apply it, the definition allows us to define something as being knowledge, even though it should never be applied. For example, supposing that Donald Trump is justified in believing that climate change is a myth, we intuit that this justification does not meet the extraordinary standards which are required to apply that knowledge, given that the lives of billions of people may be on the line.
Given these problems, we have decided to give up on the whole idea of knowledge, and use the concept of credibility instead. This we define as follows:
- a proposition is foundationally credible if it may be applied justifiably in order to infer new knowledge, without recourse to independent investigation of the credibility of that new knowledge;
- a proposition is operationally credible if it may be used justifiably in order to act, without recourse to investigation of the credibility of the proposition;
- a proposition is tentatively credible if it may justifiably be used to consider courses of action, which will be examined on their own merits before being put into action;
- otherwise a proposition is not credible.
In the above definition, the word 'justifiably' should be taken to mean that the use made of the information would tend to be considered to be reasonable by independent parties that are acquainted with the context of that use, including the damage expectation associated with the eventuality that the information is in fact incorrect.
In the following sections of this lemma, we shall describe how the concept of credibility can be operationalised.
A new instrument
Mainstream science has no way of assessing credibility with sufficient rigor to enable us to assert that a particular explanation is more credible than others, let alone assert that this explanation is necessarily more credible than all other possible explanations.
The Occam method, in combination with QO, does allow us to make such assertions. This provides us with a new instrument with which to assess credibility, based on an adage which Conan Doyle put into the mouth of Sherlock Holmes: "Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth." The Occam method applies a variation on this theme, namely:
An explanation is credible, no matter how high its absolute Occam score is, if it can be shown that any other explanation must have a higher absolute Occam score. The extent of the credibility depends on the number of notches on the Occam scale of all other explanations put together, as follows:
- a foundationally credible explanation is at least 3 notches better than all other explanations put together
- an operationally credible explanation is at least 2 notches better than all other explanations put together
- a tentatively credible explanation is no more than 1 notch worse than all other explanations put together
This rule enables QO to go further than science and philosophy generally dares to go. That is possible, firstly, because QO provides a scale on which credibility can be assessed. Where mainstream science and philosophy compare rival explanations, neither of which can be proven, and declare the contest a draw, QO chooses a winner. In this manner, QO can reject the explanation that there just happened to be an infinite, all good, all powerful God who created everything, on the grounds that this explanation has a higher absolute Occam score than QO. It is possible, secondly, because QO provides means to discuss explanations generically, avoiding the need to itemize them. Itemizing all possible explanations is notoriously difficult, due to the difficulty of defining the word 'possible' and the challenge of producing an exhaustive list of explanations. That is why QO can be applied successfully in many situations in which the Sherlock Holmes version is of little use.
As we have shown in the lemma 'There is something, rather than nothing', QO has a lower absolute Occam score than all other possible explanations together. It is therefore credible, despite being a complex paradox. Conversely, no other explanation can be considered credible. QO may be ridiculous, but every other possible explanation is substantially more ridiculous.
Note that the rule is purely concerned with credibility, and not with proof. Proof is in any event impossible, so that is not much of a loss. The notion of proof also leads to fallacious arguments, in which a compound statement is only to be regarded as proven if each of its constituent arguments is proven. This is analogous to the argument that a party balloon filled with air cannot have a firm form, because neither the empty balloon nor the air which is pumped into it are firm. Just as the firmness of the party balloon is not due to the strength of the balloon but to the weakness of its environment, so too is QO made credible by the lack of credible alternatives rather than any inherent virtue in QO itself. That is as it must be, because the notion that something arose out of absolute nothingness is by definition incredible.