There's a kind of tradition in theoretical physics of wacky "what if" papers. The whole "wormhole" thing is an example of this in action-- somebody noticed that the structure of General Relativity would allow you to make tunnel-like structures between points in space, and then asked what you would need to make such a structure. The answer turns out to include "a negative energy density," which is impossible, as far as we know, but that hasn't stopped people from creating an entire cottage industry around papers about wormholes.
(I don't know if other sciences do this-- are there biologists who make a career out of speculative papers on what life would be like if it were based on silicon rather than carbon?)
As a fan of SF, I'm all right with this sort of speculative stuff, but it continues to strike me as odd when I run across these ideas being taken seriously. For example, there's the Dyson sphere search mentioned at Cosmic Variance the other day. Dyson spheres are another fanciful construct, based on the idea that a sufficiently powerful civilization might decide to harness solar energy more effectively by completely enclosing their star with a shell of matter. If you used the mass of Jupiter, and a radius twice that of the Earth's orbit, you'd get a shell a few meters thick.
Far-fetched as this sounds, people are actually looking for these things. If you work through the consequences of making a Dyson sphere, it turns out that the whole thing ought to radiate in the infrared, so a group at Fermilab is poring over infrared telescope data, looking for objects with the appropriate signature. They haven't found any, but they are looking.
There are times, though, when "what-if" papers tip over into annoying me.
More precisely, it's not the papers themselves that annoy me, it's the coverage of them. Take, for example, this Physics Buzz item on a scheme that would allow eavesdropping on quantum cryptography.
Quantum cryptography is a technique that uses entangled particles to generate an encryption key that is shared by two people (traditionally, Alice and Bob) who want to exchange secret information. The security of the key is guaranteed by quantum mechanics-- there is no way for an eavesdropper ("Eve," a deplorable bit of twee nomenclature) to intercept the key without being detected, thanks to the way quantum measurement works. A scheme to break this would involve some sort of major revision to our understanding of quantum physics.
So, the paper being reported at Physics Buzz is big news, right? Well, not so much. There is a scheme being presented that would let Eve intercept messages undetectably, that much is true. However, it requires her to travel back in time, through a wormhole.
This isn't remotely a practical scheme, it's "what if" cubed. If you could make a wormhole, you could travel back in time, and if you could travel back in time, you could potentially eavesdrop on quantum cryptography. And if I had quickness and jumping ability, I'd be in the NBA.
Now, to their credit, the folks at Physics Buzz do report this with the appropriate qualifiers, even in the headline. The Science News piece they based it on is a little less restrained (and also includes a really dopey explanation of cryptography involving birds). And I fully expect to see a New Scientist article proclaiming the end of quantum theory as we know it, based on taking the Science News piece and stripping out all the cautions and caveats.
"What if" papers are fun and all, but when they start getting taken too seriously, it rapidly becomes annoying.
Physical Science
Comments
Heck, if you've got wormholes and time travel, there's got to be simpler approaches to eavesdropping than cracking the cryptography...
Posted by: Dunc | December 4, 2008 11:02 AM
If you have wormholes and time travel, you should go back in time to compromise both Alice and Bob via post-hypnotic suggestion.
Posted by: abb3w | December 4, 2008 1:51 PM
What's worse is that we are not even sure that quantum theory would definitely work like this in the presence of closed time-like curves. It's just a suggestion of David Deutsch that looks to be consistent, and allows you to work out a lot of nice quantum information things. I'm not criticizing the paper per se, since it is interesting to work out the consequences of the proposal, but it's not even on as sure a footing as wormholes. We know that wormholes exist as solutions to general relativity, but you'd really need a full theory of quantum gravity to be sure that quantum theory works this way in the presence of closed timelike curves.
Posted by: Matt Leifer | December 4, 2008 2:03 PM
Hey, we have dedicated the whole advent season to What if's :-)
Posted by: Bee | December 4, 2008 2:07 PM
At least the Dyson sphere search has the spinoff of revealing more about Mira-type stars (since these seem to be pretty good "Dyson-sphere impostors")
Posted by: andy | December 4, 2008 3:54 PM
If you could go back in time, why would you need to eavesdrop at all.
1) wait for the thing to happen
2) go back in time
3) ...
4) profit!
Posted by: rpsms | December 4, 2008 5:01 PM
QKD is easy to jam. Just listen! or better yet cut the fiber! :-)
Posted by: perry | December 4, 2008 9:26 PM
As for life using silicon instead of carbon... I present this sentence from the Wikipedia article on silane (the silicon analogue of methane).
"At room temperature, silane is a gas, and is pyrophoric -- it undergoes spontaneous combustion in air, without the need for external ignition."
Posted by: Unistrut | December 4, 2008 10:11 PM
I think you are being a bit unfair to this paper. The takeaway point is that a previously absolutely unbreakable system is not so (even if the situation to break it is a bit ridiculous).
Posted by: Joshua Zelinsky | December 4, 2008 11:20 PM
There's an element of "what if?" to a lot of fundamental results. Carnot's work is based on "What if we had a perfectly reversible heat engine?" There is no such thing: The isothermal parts of the process have to happen infinitely slowly, and the adiabatic parts have to happen infinitely fast and yet without friction. Still, he got a hell of a result from it. I just wrote a paper on "What if we had an algorithm that could always solve such-and-such problem?" and used it to derive an upper limit on how quickly a certain task in optics can be accomplished. Of course, there is no algorithm that can always solve the problem I'm interested in, but there are some that can solve it in a wide enough range of cases to make it worth asking what the limit is.
Yeah, the paper that's ticking you off is well outside that realm, but the point is that "what if?" isn't inherently useless.
Come to think of it, quantum computation began with people saying "What if we could actually manipulate individual 2-level systems with a degree of control not yet achieved experimentally, maintaining quantum coherence throughout the process?" (Or at least not achieved at the time.)
Posted by: Alex | December 5, 2008 2:07 AM
>>The answer turns out to include "a negative energy density," which is impossible,
Isn't this in face NOT impossible if the whole "dark energy" business pans out? Doesn't the cosmological constant basically amount to a negative energy density in space?
I'm not saying you can scoop up a bunch of 'dark energy' and use it to prop your wormhole open as a PRACTICAL matter, but I don't think the very idea is necessarily nonsense.
Posted by: dr. dave | December 5, 2008 4:09 PM