Aristotle on Necessity and a Rethinking of the Term “Objective”

I spend a good amount of time writing about philosophy on this blog, and in particular I write a lot about the question of epistemic access and its ramifications for attempts at making what I have generally referred to as objective knowledge claims.  I have recently been questioned concerning my use of the word objective and in the course of thinking through how I would respond to these questions came across an interesting distinction made by Aristotle that I think may capture better what I mean in my arguments.

The distinction (shown to me by Professor John Hare) has to do with the concept of necessity, which Aristotle breaks down into two types:  simple and hypothetical.  Hypothetical necessity involves an “if” clause:  “it is necessary to have a if you are to have b.”  Simple necessity does not involve such an “if” clause and simply states that “it is necessary to have a.”  For Aristotle, mathematics is an example of simple necessity.  The rules of mathematics simply are.  2+2 is always 4, there is no conditionality to this.  a+b = b+a, this isn’t hypothetical, it necessarily is.  On the other hand, Aristotle thinks that nature and art are examples of hypothetical necessity.  The growth of a tree to reach a certain height isn’t necessary in and of itself, but for it to happen it is necessary for other conditions to exist first (such as a particular climate and the right soil conditions).

I want to apply this distinction to my thoughts about epistemic access and suggest that is is a better way of understanding what I have attempted to convey than the term objective.  What I have been meaning by objective knowledge claims can I think be seen as analogous to Aristotle’s notion of simple necessity.  In other words, this could be taken to mean that the proposition in question is claimed to be necessarily known in the same way that mathematical proposition carries necessity.  My claim is that in reality, all such propositions are hypothetical.

So for example, lets take a famous syllogism:

a) All humans are mortal.

b) Socrates is a human.

c) Therefore, Socrates is mortal.

This syllogism, as written, is intended to convey simple necessity.  Statements (a) and (b) are taken to be given and therefore statement (c) is necessarily true.  What I want to suggest is that in even the most basic of syllogisms like this, there is always an implied conditional.  So I would write this syllogism as follows:

a*)  If all humans are mortal

b*)  And if Socrates is a human

c*)  Then it follows that Socrates is mortal.

The importance of this “hypothetical form” is the acknowledgement that the premises are not givens.  I can imagine a counterargument to the first premise (perhaps taking the resurrected Jesus as its example) or a counterargument to the second premise (perhaps claiming that Socrates is not human but a mythical figure in Plato’s mind), both of which would undermine the argument.

Now it might be suggested that really what we have done here is introduce a new syllogism and that the old one still stands.  My response is that the old syllogism stands  only in so far as it represents a mathematical formula.  So the syllogism:

x)  A = B.

y)  C = A.

z)  Therefore, C = B.

is valid because as a mathematical formula it displays simple necessity.  The moment we introduce content to the statement (and specifically, content we do not intend to be merely a variable like the letters in an algebraic equation), however, I think we are entering into statements which can only display hypothetical necessity.  Socrates is only necessarily mortal in the context of our syllogism if it is indeed the case that all humans are mortal and that he is actually a human.  It may turn out that Socrates is indeed mortal, but not for any reason given in the syllogism (hence the fallacies of affirming the consequence and denying the antecedent), but in the context of this syllogism the necessity of Socrates’ mortality only follows if both the mortality of all humans and the humanness of Socrates can be demonstrated.

Now why is any of this important? 

My contention has been for a long time that there is a gap in our epistemic access which limits us to a kind of perspectivism.  In particular, this gap has to do, I think, with our ability to extrapolate from our immediate experiences using the mechanism of reason and come to broader, metaphysical conclusions.

When we make knowledge claims, we are assuming truth is attached to the propositions in question.  To claim that a rational argument leads to a conclusion that counts as a knowledge claim about metaphysics requires then that the argument (both its premises and conclusion) be true. All such arguments can be either in the hypothetical form or the simple form we have discussed.  In light of the distinction we have drawn from Aristotle in this post, the claim I am making might be put this way: we lack the epistemic access to make rational arguments in the simple form and can only ever make them in the hypothetical form.  Thus, there is a degree of conditionality to all such arguments and their conclusions.  The extent to which this conditionality is fatal to the cause of philosophy is debatable.  At the very least, however, it seems to limit our knowledge (defined in a traditional sense) to our immediate experience and make all other propositions subject to the realm of belief.  To me the realm of belief is extremely important and not something to be looked down on, but to others the prospect of metaphysical propositions not being able to count as knowledge is extremely problematic.  But more on that for later discussions…


7 thoughts on “Aristotle on Necessity and a Rethinking of the Term “Objective”

  1. The problem is that hypothetical premises like this don’t tell us anything interesting. Take your example:

    a*) If all humans are mortal

    b*) And if Socrates is a human

    c*) Then it follows that Socrates is mortal.

    This argument tells us nothing interesting. Arguments are supposed to persuade people to believe something. Sometimes we are rationally required to believe something is true. The premises tell us why. This does not tell me why I should believe that Socrates is mortal.

    Consider arguments actually used by scientists. They argue that evolution is true. Your hypothetical formulation of any such arguments would only be “if xyz, then evolution is true.” That would entirely fail to persuade anyone that evolution is true or is even likely true. And yet science denialism is irrational.

    1. Thanks for the comment!

      Your critique wasn’t one I had considered before now, so this may be a bit half-baked, but here goes:

      To start at the end of your comment, there may be some scientists who wouldn’t make their arguments this way, but I think most would accept the “if xyz, then evolution is true” formula and argue that we have compelling evidence for “xyz.” That’s exactly what inductive argumentation does: it attempts to mount the evidence together for a conclusion rather than attempting to deduce the conclusion from taken-for-granted premises. To rephrase my claim, I’m suggesting that almost all arguments are actually inductive in nature (even when written as deductions).

      A second response I might give is to say that there is a distinction between rhetorical persuasiveness and epistemic limitation. That distinction might be reflected in normal language: no one speaks or argues using the hypothetical form that I have proposed. What I’m suggesting is that embedded in the way we normally speak is the hypothetical form I’ve outlined. We assume those hypotheticals to be true in most normal discourse. But (at least so I’m claiming) if I were to prove to you either that Socrates was not human or that all humans are not mortal (or even if I were to produce a significant counter-argument to those claims that you had to refute), then the actual conditional nature of the premises of the argument would be revealed.

      I’m not advocating abandoning the way we normally conduct discourse. My hypothetical form is definitely a lot more cumbersome. My point here is to point out the way such embedded conditionality can serve to conceptualize our epistemic limitations.

      1. I agree that deductive arguments often require inductive evidence for each premise. Is that what you are mainly getting at? How we ordinarily speak seems important to assure us that the conclusion should actually be believed. Something certainly needs to be said about that.

        1. Keep in mind, what I’m primarily arguing has to do with epistemic limitations. I’m using Aristotle’s concept of necessity as a way of fleshing out what those limitations look like in terms of how logical reasoning works with real content.

          So yes, I am getting at the need for inductive evidence for the premises of deductive arguments. But only as a means to an end.

          As for how we ordinarily speak, I think that such language “works”- read “is compelling”- because we believe that such inductive evidence exists for a given premise and therefore we buy the premise.

          To give an example where this might be easier to see (since this issue has been in the news a lot lately):

          1) Good public policy will save lives.
          2) Gun control is good public policy.
          3) Therefore, gun control will save lives.

          I can make inductive arguments for each of the premises of this argument, and if you find those inductive arguments compelling, then you’ll buy the larger argument and its conclusion. I could also make inductive arguments/counter-arguments that would challenge each of these premises and if you bought those then you would claim the larger argument is rubbish.

          Now, in a way, what I’ve said isn’t actually very interesting. We only buy the arguments whose premises we have reason to believe. Duh!

          But that illustrates the point I’m getting at: especially in the case of a particularly controversial topic/claim, whose to say whether the premises are true or not? At the end of the day, we have an epistemic gap that prevents us knowing with certainty that those premises are true. What ends up happening in common discourse is that we make a judgment call about what premises we find most believable.

          So what I’m saying on the “meta-level” of epistemic analysis is this: implied in the larger argument is a conditionality: if the premises are true, the conclusion follows. On the other hand, common discourse works like this: if I buy the premises, I will buy the conclusion. But buying the premises is not the same thing as the premises being true, it simply means that we believe them to be true. That little epistemic gap between what we find most convincing and what we actually know to be true is what I’m getting at in this post, and that’s where I think the implied conditionality of even our deductive arguments comes from.

What do you think? I would love to hear from you, please share your thoughts. Just remember to be respectful of others.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s