Ooo, a thread on quantum mechanics! I have lots to say on the subject.
You know, it just occurred to me that the spren should definitely "collapse" to a defined value, Realmatically. We know that the Physical Realm limits the ways the power of creation can be accessed. For Allomancy, it's into sixteen metals. This "limiting" is sort of how in quantum mechanics, when a particle is subject to a potential well, it can't receive
any energy value; its energies are discrete.
So for spren, something in the Cognitive or higher Realm, it seems to me that the Physical Realm can limit those higher Realms in similar ways. In this case, the shape of spren itself--its actual presence in the Physical Realm--is limited.
Let's talk some quantum mechanics.
So, if you observe it, the photons act the same, uniformally. This implies that when unobserved/measured, the particles take boths sides at the same time, effectively splitting and then reforming on the other side, to make the fancy pattern.
This phenomenon has never been explained (as it defies all laws of physics, as do all the laws of quantum mechanics).
Now, hold up. "Defies all laws of physics" is a much stronger statement than is actually happening here. What's happening here is that, before "measurement" (or an "event", if the word measurement isn't your bag of tea) a particle has a wavefunction which is a distribution of possibilities. This wavefunction can interfere with other wavefunctions, because it's a wave, and waves can interfere with one another. The only difference is these waves are matter waves.
After measurement comes the
wavefunction collapse. That wavefunction--what was formerly a distribution--"collapses" to the single value. The wavefunction is no longer a distribution. or rather, an infinitely narrow distribution, centered at the measured value. Thus, in the instance of the double-slit experiment, in the unmeasured case, those wavefunctions interfere with each other, which gives you those cool patterns. But in the measured case, the photons' wavefunctions have collapsed and have an incredibly narrow distribution. That means they simply aren't wide enough to give you any interference patterns. The double-slit experiment is one of the great experiments which shows that wavefunction collapse is a real, observable thing. You can also do similar things with particles decaying, and measuring one of the created particles' spin before the other.
While it is true that physicists do not understand why wavefunction collapse happens, nor what constitutes a "measurement", wavefunction collapse fits very nicely in the framework of quantum mechanics. You need some mechanism to describe the probability interpretation of quantum mechanics and how you can actually get well-defined values from experiment, which is what you'd expect classically. This isn't the
only interpretation of quantum mechanics--there are others, like the many-worlds interpretation--but you do need such a mechanism to explain how you can actually measure things and be sure you actually got a true measurement.
Thing is, quantum mechanics wasn't made one individual. As the preface in my Quantum Mechanics book says,
Unlike Newton's mechanics, or Maxwell's electrodynamics, or Einstein's relativity, quantum theory was not created--or even definitively packaged--by one individual, and it retains to this day some of the scars of its exhilarating youth. There is no general consensus as to what its fundamental principles are, how it should be taught, or what it really "means." Every competent physicist can "do" quantum mechanics, but the stories we tell ourselves about what we are doing are as various as the tales of Scheherazade, and almost as implausible.
(Griffiths is an excellent writer; I would recommend his textbooks to anyone, but he can get very mathy. But if you want a nice physics textbook to read, Griffiths' particle physics text is good, and you can get through the first five chapters with virtually no math knowledge)
Why am I saying this? Just because I take issue you saying that wavefunction collapse "defies all laws of quantum mechanics."
That's not entirely accurate.
Sorry. I don't mean to break out Physicist Chaos (wow, that sounds like an awesome supervillain, doesn't it?). I totally agree with your point about the idea that Brandon got this spren idea from the concept of wavefunction collapse. But! I will correct your quantum mechanics if I must
For you non-physics people, what is happening in wavefunction collapse is that, prior to measurement, a particle does not have a specific position (or, well, we need not just say "position." It could be a number of properties that you are trying to measure, like angular momentum). It instead has a distribution of
possible positions, which are described by the particle's wavefunction. Once you measure,
now the particle has a defined position. In a sense, you have defined a property onto the particle--a position--where it did not have one previously. Weird, huh?
I feel like this attribute of spren (jokingly called the Geranid's Certainty Principle) is an antithesis to Heisenberg's Uncertainty Principle. The Uncertainty Principle is simply that the more that you know about a particle's position, the less you know about its momentum.
Geranid's Certainty Principle is that the more you know (and have recorded) about an object's size, the more the spren adheres to that recorded size.
Can we name it something other than Geranid's Certainty Principle? Comparing it the Uncertainty Principle is most definitely a false analogy, because this "certainty principle" has much to do with measurements defining a property on a spren that it didn't really have before.
The Uncertainty Principle is only cursorily related to wavefunction collapse. What the Uncertainly Principle says, precisely, that there is a lower bound to the product of the variance of the distribution of position and the variance of the distribution of momentum of a particle. This doesn't mean that you can't precisely know one of those properties, it just means you can't know
both position and momentum to infinite precision. Actually, there's more than one Uncertainty Principle, and you can get one
for any two observables whose operators don't commute (specifically, the first relation in that link. The Energy-time uncertainty principle is unrelated), but that doesn't much matter here.
While the Uncertainty Principle is a direct consequence of the statistical interpretation of quantum mechanics, and thus has some relation to the idea of wavefunction collapse, what the Uncertainty Principle is more about is dealing with distributions of particles.
However, with spren--call them spren mechanics, or something spiffy like that--it's clear they have more relation to wavefunction collapse, in that measurement bestows a defined property to what you are measuring. The Uncertainty Principle is independent of that. We should name it not the "Certainty Principle," but something like "sprenfunction collapse," because this property of spren deals much more with collapse than whether or not you can simultaneously find observables.
tl;dr: "Ow ow Chaos hurts my brain, please go away Chaos, and apparently we should name this property of spren to something with 'collapse' in its name, even though 'Certainty Principle' sounds hilarious. Owwww!!!"
