Unaided and aided reasoning

Humans reason about many things. Some things are more complex than other things. The more complex a thing is, the more probably it is that one will reason wrongly about it. For simple things the probability of unaided reasoning reasoning wrongly is not high. For complex things the probability of unaided reasoning making a fallacy is high.(Making a fallacy is used interchangeably with reasoning wrongly.) By unaided reasoning I mean informal, non-explicit, intuitive like reasoning. The more aided a reasoning is, the more tools are applied to help it. Such tools are e.g. clarity, explicitness and logic.

Clarity could be to make distinctions between two meanings of a word/phrase. It would also be to avoid using ambiguous words/phrases and stick to non-ambiguous ones (or relatively non-ambiguous ones). Clarity is also to make it clear when one is using two terms or phrases synonymously/interchangeably as I just did before, and also to note when one uses two words as antonyms of each other. Consider this example of that: If a man is fat, then he is stupid. Peter is smart. Therefore, Peter is not fat. Here it is plausibly interpreted that “smart” should be taken to mean “not-stupid”, i.e. as an antonym. If done, then the argument is valid.

Explicitness is to type down the argument. Much reasoning happens in the mind without writing it down at all. It is also explicitness to write down more premises of an argument, so called hidden premises. Most of the time when we present arguments, both verbally and in written text, we do not write down all the premises and we have to guess or interpreted what the hidden premises are. Consider this argument: Since Peter is smart, then he is not stupid. This is best interpreted as: If Peter is smart, then he is not stupid. Peter is smart. Thus, Peter is not stupid. Sometimes we accept an argument that is based on a premise (or premises) we would not accept if we had noticed it/them or had been more conscious of it/them.

For a reasoning to be aided of logic would be to translate the argument from normal language into a formal language, or rephrase the propositions such that they are closer to the standard translations of formal logic. Logical aid would also be to learn logic. The fact that one knows logic helps one reason even when formal logic is not used because it helps one being more aware of potential fallacies. Learning about the various fallacies would also be a form of logical aid.

A general skepticism about reasoning

It is perhaps clear from some of my remarks about that I think one ought to apply a general skepticism about reasoning. We can strengthen this skepticism by considering how many times we have reasoned wrongly in the past. I will bet that for any person it has happened an innumerable number of times. From this we may infer with good inductive certainty that we will probably reason wrongly in the future too. We may also observe that the fallacies in reasoning often happen with unaided reasoning. Indeed the more unaided it is, the more often it goes wrong (i.e. a fallacy is committed). This is hardly surprising.

I shall give an example of unaided reasoning going wrong. The modal fallacy strikes me as the best example of unaided reasoning going wrong. The modal fallacy very often happens when people try to reason about something complex, that is, the consistency of foreknowledge and free will. This issue involves both temporal concepts and alethic concepts. Explaining the fallacy would take us too far away from the subject of this essay and thus I will leave it unexplained, but see N. Swartz for an explanation.1

Some issues are so complex and strange that the probability of our unaided reasoning not making a fallacy is so little, that we ought not to bother trying with it at all. This is the case with much of mathematics, and most people accept this. Nearly no one tries to reason informally about differential or integral math, and there is a good reason not to. It would not work very well. It would be a waste of time. A similar case is complex issues about or involving infinities. Though here the situation is different. Some people do try to use unaided reasoning to think about it, and not surprising the results are not good. They should have used the special kind of mathematics that has been developed to deal with it. I hold that these mathematical cases serve as a good analogy for various philosophical problems. The problems are so complex that it would, and it is, a waste of time to try to reason about such issues without help, that is, unaided.

Determining how complex an issue is

Suppose now that you’ve accepted that it is a waste of time to attempt to reason unaided about some philosophical problems. How do we determine which issues are too complex to be handled by unaided reasoning? Suppose we use this essay as an example. We can determine what level of general skepticism we ought to have about it by considering two things a) how complex the issue is, and b) how unaided the reasoning in this essay is.

How might we go about answering the first question? One good idea is to look at the issue and try to identify modal terms in it. Modal terms are terms such as “ought”, “can”, “necessary” etc. There are a few of these in this article but only a few. The presence of a few of these gives a weak reason to be skeptical about the reasoning in the essay.

How about the second question? We ought to keep the things I mentioned before in mind. Is there any formal logic in the essay? No. So there is no help from this to be had. Does the author know logic and fallacies? Yes. This is some aid. Are the arguments presented in clear language? Yes, to a good degree. How complete are the arguments formulated? To a medium degree. So, we can conclude that the reasoning presented in the essay is somewhat aided.

Now we have to consider the answer to the first and the second question together. Is the issue complex enough and the reasoning unaided enough to warrant extreme skepticism? No. I submit that the issue is complex enough to warrant a medium level of skepticism but that the aids used reduce the warranted level of skepticism a bit. It seems unnecessary to require formal logic to be used to trust the reasoning in this essay.

Similar reasoning to the above can and should be applied to any form of argumentation.

Aiding our reasoning

Supposing that you have accepted my reasoning above. Suppose now that we encounter a really complex issue. We now see that our unaided reasoning about it would not be very useful because we ought to have a high level of skepticism about it. To avoid this skepticism, how might we aid our reasoning? Basically we can do the things I mentioned in the beginning: We might improve clarity by avoiding ambiguous words and phrases. We might improve explicitness by writing out the complete arguments. We might get help from logic by formalizing the arguments and checking that they are formally valid. (Or something equivalent in the case of inductive arguments.) We might learn of the various modal logics to help us think about issues involving modalities. This last one I think is very important, cf. the example of the modal fallacy.

Consider the issue of examining whether a being being omnipotent implies that that being is eternally omnipotent.2 This issue involves lots of modalities and different kinds too. There is both the temporal kinds and the alethic kinds. I submit that this issue is so complex, that our unaided reasoning about it is close to useless; If we ought to reason about it, then we ought to reason highly aided about it. I also think that that issue is too complex for us to properly reason about it without formal logic. I base this on observed discussions concerning this issue and similar issues. The discussions almost never get anywhere without the aid of formal logic or at the very least the knowledge of propositional, predicate and the various needed modal logics.


Leave a Reply