Epictetus, Discourses 1.7.3

May we say then that this faculty too must be acquired by him who is to behave with good sense in discussion, and who is himself to prove each point in his demonstration and to follow the demonstrations of others, and to avoid being led astray by sophistical arguments, posing as demonstrations? 

 

Thus, it comes about that we are led to think it really necessary to discuss and to practice the arguments and moods which are conclusive.

 

But note this: there are cases where we have granted the premises properly, and such and such a conclusion follows which, though it follows, is none the less false. What then is it fitting for me to do? Must I accept the false conclusion? How can I do that? Must I say I was wrong in granting the premises?

 

“No, you may not do this either.”

 

That it does not follow from the premises granted?

 

“No, you may not do this.”

 

What then is one to do in these circumstances? May we not say that just as in order to be in debt it is not enough merely to borrow, but one must remain a borrower and not have paid off the loan, so in order to be bound to admit an inference it is not enough to have granted the premises, but one must abide by having granted them?

 

My college, which prided itself on its intellectual rigor, offered just one introductory logic course for undergraduates, and it was only an elective. Even philosophy majors were not required to take it, such that we had these ridiculous scenes where students passionately debated about the merits of Kant or Hegel, without knowing how to go about forming a basic syllogism. 

 

In a moment of frustration during a class, I once told someone that it seemed like his argument lacked a middle term. I can still remember the silence in the room, and being stared at with deeply confused looks. In hindsight, I should probably have kept my mouth shut. 

 

Yes, we can pick up a good portion of our reasoning skills informally; we do much the same when we learn a language, but that does not excuse us from a degree of deliberate reflection upon grammar. 

 

The fact remains that if I wish to live well, I must also take up the responsibility of learning to think well, and if I wish to think well, I must work diligently on the habit of sound demonstration. 

 

The danger in failing to do so is not merely an academic matter, since it directly affects the exercise of my daily practice. As a creature of intellect and will, I am called to understand why I should act as I do, and I will only be throwing myself to the wolves if I allow my choices to be determined by the many careless or manipulative words that surround me. 

 

If I had a dollar for every flawed or misleading argument I have heard or read, I would be a rich man. Given the choice, however, I would prefer to have the power within me to refute and avoid such errors; the money would do me no good without the wisdom to know what counts. 

 

This will not come easily, and I will make many mistakes along the way. What am I to do, for instance, if I follow a train of thought through to its end, but I find that the conclusion is clearly incorrect? As with a math problem, or with any attempt at a task in life, I go back and check my work. 

 

Let’s say that the logic actually does turn out to be valid, and yet I remain stuck with a solution that can’t be the case. Surely, it must be a problem with the premises, and so I should go about questioning whether they are true. 

 

And the problem will be that I now appear to be changing my premises to fit the conclusion, which is the exact reverse order of how an inference should proceed. This is why scientists will frown upon altering the evidence to support the desired results. 

 

It may sound like a purely technical difficulty, but it gets to the heart of what it means to make progress in my reasoning, of moving forward instead of backward. 

 

Once I have proposed the premises, is it permissible to go back and modify them, or am I absolutely bound to them? A certain conclusion does indeed necessarily follow from those premises, but could a conflicting set of premises lead to me to a very different conclusion? At first, this may sound an awful lot like cheating, of cleaning things up after the fact. 

 

This example may be silly, but it can perhaps serve to illustrate the point. Consider the following syllogism:

 

All mammals are cold-blooded

All dogs are mammals

–––––––––––––––––––––––––––-

All dogs are cold-blooded

 

We instinctively see that this doesn’t work, and we quickly recognize what needs to be fixed. To “go back” and do so reflects the fact that our thinking is not static, that situations change, and, most commonly, that our perception of things grows and becomes clearer with greater awareness. The rules of inference are not violated by working through these many possibilities. 

 

If I do accept the premises, I must then indeed accept the conclusion, and yet nothing should hinder me from correcting a previous error, or from reconsidering the premises that I accept. This is still a part of moving forward. 

 

Just because I have borrowed does not mean that I must always be in debt, and just because I have speculated on one option does not mean that I am irrevocably bound to it. 

Written in 10/2000