13

What, specifically, is the problem with circular logic/reasoning? I think it's invalid. For some reason, everyone knows that it's wrong to use, but is there anything more to it than that?

Additionally, where can I find journal articles or papers that discuss this issue?

  • In the field of logic, a circular argument is often called a tautology. You might have more luck in your research efforts if you search for that term. – Cody Gray Apr 22 '12 at 9:28
  • 4
    @CodyGray Sorry, but you comment is misleading. A tautology in logic is not a circular argument and is certainly not invalid! See en.wikipedia.org/wiki/Tautology_(logic) . The OP is asking about en.wikipedia.org/wiki/Begging_the_question in argumentation theory. These are two almost completely different meanings of "tautology". It was exactly for cases like this that I proposed meta.philosophy.stackexchange.com/q/325/1582. – DBK Apr 22 '12 at 9:58
  • @DBK: You're right, mea culpa. I was definitely thinking of argumentation theory, rather than formal logic. And it's not exactly clear which one this question is asking about, or I'd suggest to retag it. Might be worth posting an answer that makes the distinction clear. :-) – Cody Gray Apr 22 '12 at 10:01
  • When using logic as a tool for 'making arguments' then it behaves as a closed system. It needs a clear input and an output. A beginning and an end. A proposition and a conclusion. One needs to draw a distinction between a circular and a recursive argument. I suspect a philosopher would still call that bad. A computer scientist would call it an algorithm. If it converges. Eventually. If circular reasoning was a "problem" Descartes wouldn't have be allowed to get past the "I" in "I think therefore I am" because it's recursive ;) Cogito ergo sum A => B Generalises to a transfer function: f(A)=B A – Todor Genov Aug 12 '18 at 17:17
  • "For some reason, everyone knows that it's wrong to use" - unfortunately, no. Otherwise you would not ask this question. – rus9384 Aug 18 '18 at 16:58
11

The answer to your question depends on a clarification of the concepts of reasoning and logic, and on a determination of what counts as a "problem" with relation to these concepts. It is probably most helpful if we begin by demonstrating how your question should be dispensed with according to basic conventions of philosophical terminology; thus, I'll try to explain why circular reasoning is a fallacious form of deductive reason. To this end, let's take Aristotle's definition of deductive logic as our basis:

A deduction is a discourse [logos] in which, certain things having been stated, something other than what is stated follows of necessity from their being so. (Prior Analytics I.1, 24b)

Given a certain set of premises, deductive inference should allow one to draw conclusions which are "something other than" the statements with which one begins.

According to modern logical jargon, validity is a property of an argument, such that an argument is said to be valid when its conclusions follow from its premises. According to this narrow definition, "Circular reasoning", "begging the question," or (to be latin about it) "petitio principii" is in fact a valid form of argument, but only in a trivial and vacuous sense. Such arguments only restate what they had assumed from the outset; thus undermining the very purpose of a logical argument.

It is easy to see both the validity and the vacuity of such a syllogism if we examine its simplest form:

A is B. (Premise)
Therefore, A is B. (Conclusion which is "circular" because it only restates what was already assumed.)

An argument is valid just so long as the conclusion follows from the premises, and the above syllogism obviously satisfies this condition. However, an argument which begs the question is still considered to be fallacious because it cannot be used to prove or demonstrate anything.

Searching google scholar for "begging the question" returns numerous articles. You may also want to consider taking a basic course in formal logic or reading an introductory book on the subject.

There are different conceptions of reason and logic according to which "circularity" of a certain sort is not problematic. Hegel, for instance, often talks in terms of circular movements of consciousness and reason, but this arises from a vastly different understanding of reason than we've discussed above. You might also want to investigate the idea of the "Hermeneutic Circle", which develops a logic of interpretive reflexivity founded on a divergent view of truth.

As far as I know, every careful and earnest thinker recognizes that, within the purview of deductive logic, circular reasoning is a sign of carelessness and is completely useless. Those that embrace some kind of circular reasoning do so only as a result of careful and deeply considered (which is not to say "accurate" or "correct") reconceptualization of the basic terms of philosophy.

|improve this answer|||||
  • 3
    I would upvote, but this is troubling: "Ultimately, such arguments only conclude what they had assumed from the outset and, as such, they can neither offer support for the conclusions inferred nor satisfy the purpose of a deduction: to prove that certain premises necessarily entail a certain conclusion." They do exactly that! That's the whole point of an argument being logically valid in propositional logic! (Consider the basic syllogism in modus ponens: A,B; A, therefore B.) – DBK Apr 22 '12 at 12:26
  • @DBK Thanks for pointing that out. I have removed the offending statement. – Shon Apr 23 '12 at 15:21
  • 1
    Of course "real" circular reasoning won't be that obvious. You will have some long, hard to understand proof, and the premise that makes it circular is not shown clearly and obvious at the start, but will get sneaked in somewhere in the middle where hopefully the reader won't spot it. The reader is made to believe that the conclusion follows from the premises listed at the beginning of the proof, while an additional premise is hidden away. – gnasher729 Nov 4 '15 at 11:08
2

Create a system called M, based on a set of axioms A and a logic L. After choosing the axioms your system is already finished. Now it's up to you to discover what you've created. However for every logic L you pick based on the equality principle, there exists a method of circular reasoning of which its output is its input (both equally arbitrary). Hence the discovery isn't possible and the method is useless.

Using useful deductive reasoning on the other hand, you're able to discover. That is to say, you're able to arbitrarily rearrange the facts available in your system to form other arbitrary structures. With deductions(=arrangements) which follow directly from your axioms being the least arbitrary. As you base your deductions which follow from your previous arbitrary deductions and interpretations of your axioms, the arbitrarity of your deductions increases.

|improve this answer|||||

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.