Tuesday, March 10, 2009

The Illusion of Knowledge

"You have no responsibility to live up to what other people think you ought to accomplish. I have no responsibility to be like they expect me to be: it's their mistake, not my failing."

I have no illusions about the state of my knowledge of physics. Unfortunately, the same cannot be said of others. Like my friend, who emailed me today with the following question - "Can you explain this to me?"

Umm...

My fault to be sure, for having mentioned quantum physics in conjunction with what I was working on, to the chagrin of actual physicists everywhere. But I love a challenge, and I'm feeling feisty. (Just yesterday I learned how to read a basic space-time diagram of particle interactions. Not to be confused with a Feynman diagram, apparently, though I'm not entirely sure why.) Relatively speaking, I've got a better chance of being able to explain this article than most people, so why not...

Naturally, the preferred method for explaining such things is to refer the questioner to someone who has already explained it better than you could. (Preferably with pictures.) But where's the fun in that?

The first thing to do is to understand Hardy's paradox. In the absence of an article from our usual source - the almighty Wikipedia - we are forced to stray into press releases and blog postings to get our bearings. Hardy's paradox comes from a thought experiment that applies the fundamental tenet of quantum theory - an unobserved particle existing in a superposition of all possible positions - to a particle-antiparticle collision. Hardy reasoned that the attempts to create such a collision (see picture) might result in the particle and antiparticle disturbing each other without actually annihilating each other (as they are required to do by definition) due to their respective half-in half-out quantum states of being. (Curious minds stop to ponder what 'disturb but not annihilate' looks like...)

Hardy's design was previously thought to be untestable, as attempting to measure this 'disturbance' was itself a disturbance. That is, until the advent of interaction-free measurement or weak measurement, which itself violates a basic tenet of quantum physics - that the measurement of quantum systems (systems in a superposition of possible states) fundamentally alters those systems causing them to collapse "back to some kind of normality" (a single state). This kind of 'weak' measurement utilizes a measurement interval which is smaller than the inherent level of uncertainty about the properties of the particle. This means that you don't really know what you've got for any single measurement, but in theory you are able to deduce things from the average of such measurements repeated many times.

Your article reports on a modified test of Hardy's paradox, which used photons instead of particles and antiparticles. (Photons are their own antiparticles.) The claim is that physicists were able measure the system without really measuring it, and can therefore draw conclusions about the real (quantum) state of reality. Actually, this same experiment has been done twice by different groups/labs using the same techniques. And the weirdness is that they found regions which had fewer than zero particles in them. "Fewer than zero particles being present usually means that you have antiparticles instead." But photons are their own antiparticle, so what's going on? The analogy is made to Hardy's improbable hypothetical outcome of particle and antiparticles which disturb but fail to annihilate one another. But other than a shared sense of weirdness - "It looks impossible. But then I realised it was the only way to see it. It's beautiful." (here) - I'm not sure how the analogy applies.

Mind you, I don't have the source articles, so there's a good chance I'm missing something. But as far as I can tell, the point is basically that "there is a way to carry out experiments on the counter-intuitive predictions of quantum theory without destroying all the interesting results" and that "there are extraordinary things within ordinary quantum mechanics."

(That was actually fun! Bring it on!)

Now if you were asking if I can explain what it means, or how it fits with my idea... (sigh)

2 comments:

Unknown said...

Haha. I have the highest regard for your INTUITION about physics...and am not guilty for it. Actually, I should have specified the bit that bugged me. I don't get how we can say with any certainty that 'weak measure' doesn't disturb(yes, I clicked the link and read)... Is there math out there for it? Have we come far enough in Quantum Physics to be able to quantify such a thing? I don't think I ever understood the mechanism of the observation/disturbance scenario, so much as took it for fact (like I know how computer hardware parts go together but have no CLUE what's up with the circuits within them)and maybe that is my problem. Maybe those who understood it to begin with would roll their eyes, exchange knowing glances and mutter 'dolt.'

N said...

Yeah, I'm starting to be bugged by how any knowledge about the system - even if it is acquired indirectly through the averaging of these 'weak' measurements - can NOT have the effect of altering the system. Especially given the ideas I've put forward...

Regarding the math - that's your forte, not mine. ;) Lev Vaidman has a series of articles on interaction-free measurement in arxiv.org. Maybe later today, after some more coffee, I'll dig into those...

Regarding 'observation' as 'disturbance' - I think there are two frames of reference for that. 1) An 'observation' is not able to capture a particle in its smeared state, but rather only in a 'collapsed' single state of being. Making an 'observation' is supposed to cause this 'disturbance'/change in the state of the particle. 2) 'Observing' or 'measuring' a particular property of a particle - such as position - 'disturbs' the complementary property - momentum. To capture a particle's position with any precision, you need to impede its momentum. To me this always seemed like a reference problem - we define momentum in terms of position - so, of course knowing both at a single moment from a single measurement is impossible. This inability to know both is the 'uncertainty' that was referenced (I think).

And, hey - if 'dolt' is the worst thing I'm ever called, that's just fine. ;)