In recent weeks, I have begun to pay more and more attention to the elevators (or lifts) of our university building. The department is located on the top floor, which is why I use these machines pretty much every day, at least twice. Like the whole building, they are very modern and effective, very high-tech, (post)industrial. They would seem to be a consistent part of a well-oiled mechanistic system (the machinery of higher learning? the infrastructure of degree production?)

But there are cracks in any system. There is a seeming incoherence - not a flaw but an illogical element - in the functions of the elevators.

When you want to use the elevator, you press a button. At the narrower end of the building, there are two lifts, at the other end (close to the main doors) three. With the former, pressing a button will send a signal to the whole system: either of the two elevators may come to pick you up. But where there are three, the one on the left can only be called using a button close to it. The other two share their signals. This is perhaps slightly strange and confusing but I am ready to believe there is a rational explanation for it. Odd numbers often create some asymmetry, and there may be situations when you want to be able to call a certain elevator instead of any elevator - say, when you're moving large things between first and fifth floor.

So that's fine. The real problem has to do with the sound signals. There are two different signals built into the system: a short one (a single note) and a long one (two notes), and both announce the arrival of an elevator at the floor where a button has been pushed. And the problem, in short, is that I haven't been able to find any logical explanation for the co-existence of these two sounds. I cannot tell why, at any given moment, the system utters one instead of the other. I've had several theories, none of which seems to hold water. Apparently the signal is not determined by the direction in which the elevator is moving (up/down) or whether it's occupied or not. Once I had an idea it might have something to do with whether or not the elevator already has another destination - whether another button has been pushed by the time it arrives. In any case, the system sometimes makes fairly complex decisions: there can be many commands given almost simultaneously on the six floors and inside the two elevators.

For some time, I thought the short sound basically means the elevator was already (almost) there, and the long one that it came from two or more floors away. A signal coming within a few seconds always seemed to be the short one. But then this hypothesis was also shattered, even though I still think there's some kind of vague correlation between the sound and the distance travelled by the car after the button call. In any case, that's not good enough. There would have to be a consistent principle with a purpose, since the elevators have been designed by engineers. The machines are not, I'm afraid, works of sound art illustrating the random character of social utterances. Even if we wish they were.

In linguistic terms, the problem is that the contexts in which the two sounds occur seem to overlap. They seem to defy the principle of complementary distribution. If I remember anything from the course 12 years ago that included basic phonology, this would imply that the two sounds are not separate phonemes in the Liftese language. In other words, they do not create a meaning contrast. This, by the way, would cut the number of available meanings in that language by fifty per cent. If this is true, we could simply think of one of the two signals as a meaningless mutation, a pointless error, like the blurring of sounds in intoxicated speech.

I have to say I prefer to express the question in a more literary way: Where is the figure in the carpet?

Fredric Jameson has written quite a bit about the difficulty of understanding the totality of the system we all live in: the human effort to map cognitively. I think this is relevant to the problem of elevator sounds - not because the elevators are somehow fundamentally postmodern but because Jameson also writes about how we compensate for incomprehension, how we deal with the possible failure of cognitive mapping. Conspiracy theories, he says, are the poor person's cognitive maps. In other words, if we don't understand what elevators say to us, we can start developing theories about secret engineer societies who aim at driving us all crazy. For now, I don't even want to comment on the other obvious point of relevance, the ever more topical idea of the typical academic worker as the "poor person" of the community (relatively powerless, temporarily employed, uncertain of his/her professional future).

Perhaps I've only paid attention to all this because I am, after all, under temporary contract. Perhaps the permanent ("tenured") members of the academic community can afford to ignore signals like this, to disregard them as irrelevant, nothing but a drunken elevator's stutter. For Jameson, significantly, the process of cognitive mapping is a type of class consciousness.

The positive side, funnily enough, stems from the idea of incomprehension as productive of conspiracy theories. Failure to understand creates narratives. I'm always for narratives. I'm not at all sure I want to be able to discern the figure in the carpet. 19:48