I have quoted from Bernard Williams’ Shame & Necessity before, and drawn on his notion of “supernatural necessity”. To recap: the “supernatural” here is just a label to catch on to something that was operative, according to Williams, in ancient Greek notions of fate and necessity; but the label itself is not really meant to be informative: there is nothing interesting to draw from it about ancient vs. modern notions of “nature”, and about what might be above or beyond it (130-131).
One of the prime examples Williams cites as cases of supernatural necessity, in his sense of the term, is the story of Oedipus (and Sophocles’ play), where it comes in by means of the device of an oracular pronouncement.
[W]e are told that a certain thing will happen whatever we do, although it is just the kind of thing we might hope to avoid by action. Moreover, if efforts to avoid the outcome helped in fact to bring it about, this is a reliable sign, after the event, that the supernatural has been at work. This is what happened with Oedipus; this was the situation with the appointment in Samarra. (141)
The context in which I discussed this earlier (where I labeled it, for my own purposes, as “intimation of an inevitability”) was the implied theory of ghosts (as exhibited in the plot of Vertigo). But once more, this is something that can be fruitfully looked at in the terms I have deployed more recently, namely: interesting coincidences and local nomic constraints.
Not surprisingly, given the title of his book, necessity is a central term for Williams, and what he calls “supernatural” necessity is only one form of it. He doesn’t reflect explicitly on it, but Williams, too, brings out the character of necessity in terms of counterfactual probing. (Part of the reason why Williams does not reflect on it seems to be that he is intent on tying the notion of necessity to that of enslavement, of “being in someone else’s power”, rather than the more precise, if technical, modal notions available in his day. That predisposition leads to several questionable claims, in my view, in the final lecture of Shame & Necessity. But I’m not going to go into that issue here.)
Faced with the prediction that their infant son would grow up to kill his father, Jocasta and Laius decided, rather than keeping the baby at home, to give him to a servant with the instructions to kill him. […] Can we say that if he had stayed at home, he would still have grown up to kill his father? Perhaps we can: all that the supernatural necessity required was that in some way or another Oedipus should come to kill his father, and if his parents had kept him at home, then there would have been a route to the killing that started from his being kept at home. But there could be another picture. On this, we would rather say that if the baby Oedipus had been kept at home, then he would not have grown up to kill his father; but since it was necessary, in this supernatural sense, that he kill his father, it must have been necessary that he not be kept at home. And this implies, in turn, that Jocasta and Laius could not have kept him at home: either any decision they made to keep him there would have been ineffective or they could not have made that decision. (142-143)
It should be immediately obvious how similar this reasoning is to the counterfactual considerations we have used to describe the local nomic constraint that structures interesting coincidences, such as in the death of Damocles case. In that spirit, we might add to the scenarios Williams lists another one: that in which a prediction is made and the action to prevent it in fact is effective — no patricide and incest happen. In that scenario, we’d have no story. We’d never hear about it, since there’s nothing noteworthy about that case, and certainly nothing that would induce us to tell and re-tell the tale, preserve it throughout the millennia, and construct philosophical reflections around it.
Apart from that, there is a strikingly similar set of considerations around the idea of time travel, in the so-called “grandfather paradox” which has received its classical formulation in a paper by David Lewis. (In an earlier essay, I have tried to elucidate this situation with a triplet of notions called story knots, perspective fusion, and frustration of the will. The latter is of course identical with what Williams refers to here as “ineffective decisions”.) In a world in which time travel is both possible and technically feasible (in the way we know it from countless stories and films), a traveler might attempt to go back in time and shoot his grandfather. But since the time traveler exists and his personal history necessarily includes his ancestral chain, that’s an event that could not have happened in his own past (to which he must travel for the attempt). So the shooting can’t have been effective.
You know, of course, roughly how the story of Tim must go on if it is to be consistent: he somehow fails. Since Tim didn’t kill Grandfather in the “original” 1921, consistency demands that neither does he kill Grandfather in the “new” 1921. Why not? For some commonplace reason. Perhaps some noise distracts him at the last moment, perhaps he misses despite all his target practice, perhaps his nerve fails, perhaps he even feels a pang of unaccustomed mercy.
David Lewis, “The paradoxes of time travel”, 148.
Thus the notion of time travel, together with an altogether natural requirement of consistency, produces a local nomic constraint: if a time traveler goes “back” to live through an event that has happened “before”, then that event must happen “now” (experienced from the traveler’s perspective) in the same way as it has happened “earlier”, since “now” and “earlier” refer, in fact (in the unitary objective spatiotemporal system) to one and the same, i.e. identical event. A time traveler cannot change the past, as it is usually put, simply “because it has already happened”. A traveler, looking at it from a new (additional) perspective, might be tempted to try; but those attempts will not produce any change, their effects, so to speak, are already factored into what happened. Countless stories and films (Twelve Monkeys, say) explore this very constellation.
Now certainly, all this is rather well-trodden ground, and I don’t mean these cursory comparisons to demonstrate any new or original thought. They do show, I think, pace Williams (146-147; see also 131) that none of this has anything to do in particular with Sophocles’ “authorial power”. (Williams means “power” here not in an admiring sense, referring to great ability, but in the sense of an author’s capacity to coerce or manipulate audiences’ perspectives.) For clearly, all the plot elements which Williams discusses appear already in the received Oedipus myth; and the Samarra story, which Williams mentions himself (see the quote above), was not written by Sophocles either. His claim, I suspect, says more about the fixation of late 20th century intellectuals with power games than it helps us understand this specific notion of necessity.
Moreover, if the same kind of necessity appears in movie plots such as those of Vertigo and Twelve Monkeys, as well as in the logician’s exploration of the paradoxes of time travel, then it is hard to believe that we’re dealing here with an idea that was only fully intelligible in ancient Greek times, and is no longer “for us”; that “this kind of necessity, unlike others that I have discussed, is not part of our world” (132). Quite the contrary: we’re looking at an idea that is just as intelligible to us “moderns” as it was in ancient times, and which was just as alive in the collective imagination and its cultural products as of Williams’ 20th century as it was in the 5th century BCE.
