Weakness of Regularities of Nature Theory

The regularities of nature theory (RNT) is not, however, without some substantial flaws. The first problem of the RNT is what Ayer refers to as Vacuous Laws (VL.) The VL problem is a logical problem with universal existentials, or generalisations. Consider the following

(1)

All A's are B's; logically:

(1')

It is false that there is an A that is not a B; logically:

Both, (1) and (1') are equal. (1) states that for every x if x is A then x leads to B. (2) states that it is not the case that exists any x such that x has the property A and does not lead to B. The problem with this is that if there does not exist in any world x then it becomes true that (1'). In other words, because laws are universal generalisations, then it becomes automatically true (automatically true law) just because there are no instances of to negate the law. In plain English, one might be able to say that ``all Martian have four knees'' as a law just because Martian don't exist!

The way to patch this problem is to say that there must exist an x. With this ad hoc patch, however, one sees that it's too restrictive. Much like the Gettier problem in epistemology, it becomes to much of a restricted set of qualifiers. Scientifically recognised laws like Newton's first law of motion would be weeded out under this new patch.

Another problem with the RNT is in the Generalisation of Fact. For example, if you observe continually a fact x occurring under certain conditions y by body b then you would generalise that a similar body b' will produce the same fact x when in the same conditions y. This is a problem because if say on all islands you see that when you plant sunflower seeds, and get sunflower plants, then RNT logic would state that it's a law that you get sunflower plants on all islands, even if you plant walnut seeds.

To patch these problems Ayer exposes his Epistemic Regularity Theory (ERT) which states the following

(2)

Laws = universal truth + x

The x in this case can be one or more of the following:

1. Willingness to make predictions with the generalisation
2. our acceptance of the generalisation
3. the role that the generalisations play in the whole of the scientific domain.
4. our recognition that the generalisation explains its instances (in the sunflower analogy, for example, the generalisation doesn't explain instances of itself.)