by Michael Frayn
(According to some sympathisers, the reason why drivers on the motorways failed to slow down in thick fog recently, and so crashed into each other in multiple collisions of up to thirty vehicles, was simply because the authorities had failed to provide illuminated signs explaining that the fog was fog. This is a situation on which Wittgenstein made one or two helpful remarks in a previously unpublished section of Philosophical Investigations.)
694. Someone says, with every sign of bewilderment (wrinkled forehead, widened eyes, an anxious set to the mouth): "I do not know there is fog on the road unless it is accompanied by an illuminated sign saying 'fog'."
When we hear this, we feel dizzy. We experience the sort of sensations that go with meeting an old friend one believed was dead. I want to say: "But this is the man philosophers are always telling us about! This is the man who does not understand--the man who goes on asking for explanations after everything has been explained!"
(A sort of Socratic Oliver Twist. Compare the feelings one would have on meeting Oliver Twist in the flesh. "And now I want you to meet Oliver Twist."--"But...!")
695. Now I feel a different sort of excitement. I see in a flash a thought forming as it were before my mind's eye--"This is at last the sort of situation which philosophers have always waited for--the sort of situation in which one as a philosopher can offer practical help!"
696. Imagine that the motorist said: "The trouble is, I can't see the fog for the fog." We might understand this as a request for practical information, and try to answer it by showing him the definition of "fog" in the dictionary. To this he might reply: "I can't see 'fog' for the fog." We respond by putting the dictionary an inch in front of his eyes. Now he says: "I can't see the fog for 'fog'."
697. At this point a philosopher might want to say: "He sees the fog but does not perceive its fogginess." Ask yourself what could possibly be the object of saying this.
698. Now the man says: "I can see the fog perfectly well, but I don't know that it's fog." I feel an urge to say: "Yet you know it's fog that you don't know to be fog!" (The deceptively normal air of paradoxes.) One can imagine his replying: "Naturally--it looks like fog." Or, if he is familiar with philosophical language: "Of course--I know that I am having fog-like sensations." And if one asked him what he meant by that, perhaps he would say: "It looks like what I see in places where I should know what I was seeing if it were labelled 'fog'."
699. Now the feeling of dizziness vanishes. We feel we want to say: "Now it seems more like a dull throbbing behind the eyes."
700. Of course, one is familiar with the experience of seeing something ambiguous. "Now it is the Taj Mahal--now it is fog." And one can imagine having a procedural rule that anything ambiguous should be treated as the Taj Mahal unless we see that it is labelled "fog."
701. The motorist replies: "What sort of rule is this? Surely the best guarantee I can have that the fog is fog is if I fail to see the sign saying 'fog' because of the fog." --One can imagine uses for this rule. For example, to lure people to their deaths.
702. Still the man seems uneasy. "To be sure that the fog is fog because it is labelled 'fog', I must first be sure that 'fog' is 'fog'. Now, supposing, without its being perceptible to the naked eye, the top of the 'o' were slightly open. How am I to be certain that it could not be accepted as a 'u', so that the word was not 'fog' at all but 'fug'? Or how can I be certain that the first letter is really 'f' and not a grossly deformed but still meaningful 'b'?"
So now we have to have a label for "fog"! And another label for the label of "fog"!
703. But we are not yet out of the wood! (Or, as one might say, out of the fog.) The motorist might object: "I still cannot understand. I see that the fog is labelled 'fog', and that 'fog' is labelled '"fog"', and so forth. But how am I to know that 'fog' means fog, or that '"fog"' means 'fog'?"
So we must qualify still further. We must expand "fog" to read "'fog', where 'fog' means fog."
704. Now imagine the motorist's face. Imagine that the doubtful expression remains. Imagine that he says: "But how do I know that the expression '"fog", where "fog" means fog' means '"fog", where "fog" means fog'"?
705. What sort of game are we playing here? What sort of language are we using? I am tempted to ask, what sort of man am I being used by? I have a certain feeling that goes with grating teeth, a frown, flushed cheeks. I want to say: "My offer of help is being abused."
706. One might try to provide the man with a mental picture, a working model of his position--as it were a map to enable him to get his bearings. I might say: "You are in a complete mental fog about the whole business." He seizes on this eagerly. He goes through the motions of assenting--nodding his head, pursing his lips, saying: "Yes, yes, that's it exactly. I am in a complete mental fog."
Now one asks: "But how do you know it's a mental fog you're in?"
707. At once he cries: "NOW I see! I see that I don't know I'm in a mental fog at all! I need an illuminated mental sign saying 'mental fog'."
708. If a lion could speak, it would not understand itself.