Agency is the hypothesis that there are agents that act — that some events in the universe are instances of doing rather than mere happening. The distinction is this: Things that are done, are done by someone; but things that happen, just happen. They can be ascribed only to the impersonal “it” we use in sentences such as “It is raining.”
(If we hypothesize a single agent only, one whose act brings about all that happens, then we have the Aristotelian idea of the First Cause or Unmoved Mover. If we suppose that there are many agents, and that we ourselves are among their number, then the concept of agency is more-or-less contiguous with that of “free will.”)
In a fully deterministic universe, there is no doing, only happening. Each successive state of the universe is caused by the previous state, which in turn was caused by a previous state, and so on ad infinitum. Everything is caused, but nothing is caused by anything. Everything is passive, and nothing is active — which is what it means to say that such a universe contains no agents and no agency. When, say, an arrow hits a target, the arrow does not act; it merely responds passively to the action of the bow — the (scare-quoted) “action” of the bow, rather, since the bow is in fact just as passive as the arrow. The bow merely responds passively to the “action” of the archer — which, in a fully deterministic universe, is no less a passive response to preceding conditions than are the “actions” of the bow and arrow. Causation without agency — what I have in the past characterized as a universe that is “all dominoes and no fingers” — is incoherent, which is the whole point of the Aristotelian/Thomistic “First Cause” argument. It makes no sense to say that everything is caused unless it is ultimately caused by something.
An “action” which is caused, then, is no action but rather a mere happening (that is, a “passion” in the older sense of the word, where passion is to passive as action is to active). But — and here’s the rub — an action cannot be considered uncaused, either, since it must be done by an agent. If, as my Pragmatist namesake has said, an act be “a sheer novelty, that comes not from me, the previous me, but ex nihilo, and simply tacks itself on to me, how can I, the previous I, be responsible?” (James is here coming at agency from the “free will” angle. For the “First Cause” version, replace his first-person pronouns with “God.”)
I have for decades been aware of this antinomy at the heart of the concept of agency, and in the past I have considered it a conclusive argument against that concept. Action cannot be caused, and it cannot be uncaused — but that exhausts the logical possibilities. Therefore, there is no action and no agency, no “free will” and no First Cause. All is happening. As I have put it in the past, free will (I was not at that time using the more precise term agency) is not impossible the way miracles are impossible, but impossible the way square circles are impossible. In other words, it is self-contradictory and therefore not-even-false. And there, for years, I let the matter rest.
Somewhere in Plato — or perhaps only in my confabulating imagination, since I can’t seem to track down the passage in question — Socrates encounters a pair of Sophists who can prove the existence of anything they wish by showing that its non-existence is self-contradictory. (If memory serves, these were the same fellows who gleefully proved that someone had beaten his father from the fact that he beaten his dog. The dog was his, you see, and also happened to be the father of some puppies, and was therefore his father.)
Their argument ran more or less along these lines. Suppose you assert that unicorns do not exist. We can then reason as follows:
- If X does not exist, then X is nothing at all.
- If X is nothing at all, then X is not a unicorn.
- Therefore (from 1 and 2), if unicorns do not exist, unicorns are not unicorns.
- But “unicorns are not unicorns” is self-contradictory. Unicorns, by definition, are unicorns.
- Therefore, unicorns exist.
It’s not immediately obvious what is wrong with this chain of reasoning, but it is immediately obvious that something is wrong with it. “Unicorns do not exist” may or may not be a true statement, but it is clearly meaningful, clearly not in the “not-even-false” category. We can easily understand what is meant by it, while we cannot understand what is meant by a truly self-contradictory statement such as “Some circles are square.” Therefore, the correct (and natural) reaction to the Sophists’ argument is not “Well, darn my socks, I guess unicorns do exist!” but rather “That’s a clever paradox. I wonder what’s wrong with it.”
Thinking about what’s wrong with it may perhaps lead us to the important conclusion that existence is not a predicate, that “Unicorns do not exist” does not literally mean “There are some unicorns which don’t exist” (i.e., “There exists an x such that x is a unicorn and x doesn’t exist,” which is indeed self-contradictory) but is rather an idiomatic way of saying “There aren’t any unicorns” (“There does not exist any x such that x is a unicorn”).
My point, though, is not about the details of the solution, but rather about the correct initial reaction to the argument. Even supposing I cannot find a single flaw in the Sophists’ logic, I am still right to reject its conclusion, assume that something just must be wrong with it, and do my best to figure out exactly what. I may not be sure whether unicorns exist or not, but I am sure that both propositions are meaningful, and that any argument that says otherwise is flawed.
The same, then, goes for agency (or “free will”). No matter how cogently it can be argued that it is a meaningless concept of the “colorless green ideas” genus, the fact remains that everyone easily and naturally does understand what it means, at least until that understanding is argued out of them. To rest content with an obviously sophistical proof (even a very good sophistical proof) that there can be no agency is to betray the fundamental insincerity and lack of seriousness which is the mark of the sophist as opposed to the philosopher — the wisdom-hobbyist as opposed to the true lover of wisdom.
I seem to be particular susceptible to this sort of sophistical trap. Aside from the agency issue, I also once staked out the ridiculous position that time does not elapse, and that the idea of time elapsing is meaningless and not-even-false, due to my inability to frame that concept in non-self-contradictory terms. Only after running across an alternative model of time that allowed me to make sense of its elapsing did I recant that obviously goofy position — but it shouldn’t have taken that. I should have rejected it from the start, adequate model or no, and then continued actively trying to understand what “time elapsing” could mean. So that’s what I intend to do with agency: insist that, whatever the sophist on my shoulder may be saying about unicorns, it jolly well is a meaningful concept, and that I will keep chewing on it in the hope of someday making some progress in figuring it out.