Hofstadter, Dennett - The Mind's I

Rediscovering the Mind 44Be shown that in a very fundamental sense the system itself doesnot “know” which eigenstate it is in, and that it decides – atrandom – only at the moment the observer’s hand is put in to“test the water,” so to speak. The system, up till the moment ofobservation, acts as if it were not in an eigenstate. For allpractical purposes, for all theoretical purposes – in fact forall purposes—the system is not in an eigenstate.You can imagine doing a lot of experiments on the watecoming out of a quantum water faucet to determine if its isactually hot or actually cold without sticking your hand in(we’re of course assuming that there are no telltale clues suchas steam). For example, run your washing machine on the waterfrom the faucet. Still, you won’t know if your wool sweater hasshrunk or not until the moment you open the washing machine (ameasurement made by a conscious observer). Make some tea withwater from the faucet. Still, you won’t know if you’ve got icedtea or not, until you taste it (interaction with a consciousobserver again). Attach a recording thermometer just under thewater faucet. Until you yourself see the reading on thethermometer or the ink marks on its record, you can’t know thetemperature. You can’t be any surer that the ink is on the paperthan you are that the water has a definite temperature. Thecritical point here is that the sweater and the tea and thethermometer, not having conscious-observer status themselves,have to play along with the gag and, just as the water did,enter their own superpositions of states – shrunk and nonshrunk,iced-tea-and-hot-tea, ink-high-and ink-low.This may sound as if it has nothing to do with physics perse but merely with ancient philosophical conundrums such as“Does a tree in a forest make a noise when it falls if there’sno one there to hear it?” But the quantum-mechanical twist onsuch riddles is that there are observation consequences that arediametrically opposite to the consequences that would occur ifa seemingly mixed state were in reality always a trueeigenstate, merely hiding its identity from observers until themoment of measurement. In crude terms, a stream of maybe-hotmaybe-coldwater would act differently from a stream of waterthat is actually hot or actually cold, because the twoalternatives “interfere” with each other in the sense ofoverlapping waves (as when part of a speedboat’s wakemomentarily cancels another part reflected of a jetty, or when askipped rock’s successive bounces send out ripples thatcrisscross and create shimmering patterns on a still lakesurface). It turns out that such interference effects are onlystatistical, so the effect would become manifest only after alarge number of sweater-washings or tea-makings. Interestedreaders should consult the beautiful exposition of thisdifference in The character of Physical law by Richard Feynman.