My question is based upon this video here:


From 9:00-10:00

Max Tegmark says an isolated galaxy with an observer inside will not evolve according to Schrodinger's equation. I don't understand this. From the perspective of anyone outside the galaxy... the galaxy system including its observers will evolve according to Schrodinger's equation won't it? I thought this was independent of any interpretations.

Why does he say this?


Tegmark is claiming that the scientist who evolved inside of the otherwise isolated Andromeda system qualifies as an observer; and since there is an observer, the system as a whole is being observed, and hence the state function will collapse randomly -- which will interrupt it's unitary evolution.

Tegmark raised his hand for "those who believe in the Many Worlds" interpretation; they specifically deny the "random collapse" of the state function; instead this is supposed to generate one of the Many Worlds; the result is that each of the superposed states evolves unitarily at all times; apparently there will be Tegmark's world, and the rest of us who remain in the superposed state along with my entangled photons in the lab.

Your question is: "Why does he say this?"; he is setting up a straw man argument, which he systematically demolishes during the next portion of his talk.

There are many problems with his line of argumentation, beginning with the assumption that there can be a state function for the cosmos, or for any really large system. Certainly no one has ever conducted an experiment which shows such behavior, and standard methods imply that the statistically mixed state is the best that one can do.

As one moves further back in time, as cosmologists are wont to do, he must approach the Big Bang; and then of course there is a lot of unknown physics. Perhaps one of the interpretations provides an answer there.

The fundamental problem is that the discussion is about the meaning of a measurement; the use of the word observation implies (to some) a sentient observer. The propensity in the early days of quantum mechanics to appeal to philosophical studies to further certain arguments is the true background to this entire question of "interpretations". As a student, you want to focus on how to do the calculations -- and everybody agrees on how to do the calculations.

  • 1
    $\begingroup$ I used to read Schroedinger and Heisenberg when I was young and I have to agree that both had a very "philosophical" streak, which I mistook for a relevant part of the physics history of the development of quantum mechanics. In hindsight I have to admit that I was thoroughly fooled by these writings. The actual history of quantum mechanics shows an early departure from philosophy towards an actual scientific understanding of the measurement process (e.g. by Mott) and towards relativistic systems, which have a different ontology altogether, and that is how it should be taught today, but isn't. $\endgroup$ – CuriousOne Mar 13 '16 at 22:04
  • $\begingroup$ So suppose instead of Andromeda... it's just some observer A performing a measurement on some object. And there's some observer B separated from A and the object. And B is not performing any measurements on A or the object. From B's perspective "A+object" evolves according to the Schrodinger equation regardless of what A does with the object right? $\endgroup$ – Ameet Sharma Mar 13 '16 at 22:09
  • $\begingroup$ @CuriousOne: amen to that! For more on Max Tegmark, a professor of physics at MIT, see his "crazy science" page: space.mit.edu/home/tegmark/crazy.html $\endgroup$ – Peter Diehr Mar 13 '16 at 22:12
  • 1
    $\begingroup$ @AmeetSharma: I don't know if classical objects have a quantum state; I would expect that the best one could ever find would be a collection of mixed states; the analysis in terms of measurables and boundary conditions is beyond the ability of quantum statistical mechanics. $\endgroup$ – Peter Diehr Mar 13 '16 at 22:14
  • $\begingroup$ @PeterDiehr: No offense... but MIT is known for selecting their teaching and research staff for the "drama" they can create rather than the actual quality of their research. :-) $\endgroup$ – CuriousOne Mar 13 '16 at 22:15

OK... that's got to be one of the more messed up "explanations" that I have seen. For one thing, he starts with a toy version of quantum mechanics that is not (and never was) interpreted that way by physicists who actually have experience with quantum systems. I think Peter Diehr has a point by calling it a straw man. I would totally agree with him on that.

To be perfectly clear: An isolated quantum system will evolve. How it will evolve is exactly what the Schroedinger equation tells us (for certain non-relativistic systems that is).

Whenever we perform a STRONG measurement on the system, we set it to one of the possible eigenstates of the measurement operator. Somewhat more precisely, we set it to a state that is close to one of the eigenstates of a hypothetical measurement operator that is being approximated by our actual, physical measurement device. Whenever we perform that measurement, we also change the system from an isolated one to an open one because we have to introduce an actual physical interaction between the system and our measurement device. It is not possible to speak of a measured system as "closed" or "isolated".

Now, the thing about these strong measurements is that they couple a small quantum system to a much larger measurement device, i.e. the measurement device (aka "observer") has many more possible microscopic states than the observed device. This was always understood (however implicitly) this way. Sometimes people forget about that and then bad stuff happens.

An observer who can do strong measurements on the wave function of the galaxy would have to be much larger than the galaxy and it would have to couple strongly to the entire wavefunction of the galaxy. A small internal part of the galaxy is not such a "strong" observer in the sense of quantum mechanics. Will it change the entire wavefunction of the galaxy to the same value all the time? No. How would that be possible? The observer could never possibly absorb or supply the entire energy/momentum/angular momentum that it would take. And such a "puny" internal observer is NOT what is being meant by measurement/observer in the original version of non-relativistic quantum mechanics.

  • $\begingroup$ But the wavefunction is observer dependent right? My point is... Regardless of whether this observer A collapses the wavefunction of the galaxy from his perspective.... some other observer B separated from "A+galaxy" is right to say the wavefunction of "A+galaxy" evolves according to the Schrodinger equation, regardless of what measurements A does, as long as B performs no measurements? $\endgroup$ – Ameet Sharma Mar 13 '16 at 22:14
  • 1
    $\begingroup$ @AmeetSharma: The wave function is a mathematical construct that tells ONE observer what will happen to a quantum system after a certain amount of time if it's being left alone (i.e. when it is isolated). One can't have two observers messing with the same quantum system independently. That's logically impossible because of the definition of "isolated". There is no such thing as wave function collapse. That's not a physical phenomenon but simply a different name for the Born rule, which was basically an ad-hoc term for what happens during a strong measurement. $\endgroup$ – CuriousOne Mar 13 '16 at 22:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.