Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Many Worlds Theory

  1. Dec 29, 2003 #1
    Greetings forum experts.

    What is the Many Worlds theory? And, is there any evidence to suggest that it is wholly or partially correct?

  2. jcsd
  3. Dec 29, 2003 #2
    Scientific American Cover Feature
    May 2003 = Must read for you. (Plus check out my recent post at Mkaku.org's forums (hosted by this site) in the Theory of everything Forum).

  4. Dec 29, 2003 #3
    Thanks. I will do that. Appreciate the direction.
  5. Dec 29, 2003 #4
    You bet!
    Glad I could help...

  6. Dec 30, 2003 #5


    User Avatar

    "Many physicists tried to explain the equations, Gordan said. each explanation failed for one reason or another. Then in 1957, a physicist named Hugh Everett proposed a daring new explanation. Everett claimed that our universe was one of an infinite number of universes existing side by side. each of these universes was constantly splitting. Everett called thsi the MANY WORLDS interpretation of quantum mechanics. His explination was consistant with the quantum equations, but physicists found it hard to accept though no one has ever shown it wrong. Everett had no patience with his colleages' objections. He insisted it was true whether you liked it or not.

    Gordan took out a piece of paper and started drawing. Very simple experiment, been done for two hundred years. Set up two walls one in front of another, with the first wall having a single verticle slit in it

    Now you shine a light at the slit. On the wall behind, you'll see a white line from light coming in through the slit.

    Gordan continued to sketch, Now, say insted of one slit you have two. Shine a light, and on the wall you see-

    Two vertical lines.

    No, You'll see 5 lines of light and 6 bars of wall going black, white, black, white, black, white, black, white, black, white, black.

    If you shine your light through four slits you get half as many bars as before, because every other bar goes black.

    (yeah I still don't get it)

    The light passing through the slits acts like two waves that overlap. In some places they add to each other and some places the cancel each other out. That makes a patern of alternating light and dark on the wall. What'w wrong with this is that this is a 200 year old explination, but since Einstein, we know that light consists of photons. Photons couldn't make this patern

    Particles aren't simple as the way you have described them. Paricles can interfere with one another. In this case, the photons in the beam of light are interfering with one another to produce the same patern.

    That seems logical, after all a beam of light is zillions of photons. It's not hard to imagine they world interact in some fashion. Nuh Uhh. One way to find out if that is true is to make a beam of light so weak only one photon comes out at a time, and put very sensetive detectors, so sensitive they can sense one photon. Now, there can't be any interaction with other photons. So the photons come through, one at a time. The detectors record where the photons land. What we see, is that the individual photons land only in certain places and never others. They only land where the bars of light were on the wall. They behave like they did in the beam of light. They are coming in one at a time. There are no photons to interere with them. Yet something still is interfering with them because they are making the usual interference pattern. So: What is interfering with a single photon. It has to be other photons , but where are they? We have detectors and they don't detect any photons. So where are they?

    What he is saying is that single photon interference proves that reality is much greater that just what we see in our universe.The interference is happening, but we can't see any cause for it in our universe. Therefore, the interfering photons must be in other universes. That proves other universes exist. Sometimes they interfere with our universe.

    Femember, within the multiverse the universes are constantly splitting which means that many other universes are similar to ours. And the similar ones interact. When we make a beam of light in our universe, beams of light are simultaneously made in many similar universes and the photons from those other universes."

    The experiment in the passage has been done many times over the hundreds of years. All that was true, I belive it.
  7. Jan 2, 2004 #6
    Problems with many worlds

    OK, here is a brief summary of the main reasons why many worlds has not been univerally accepted by physicists. I do not mean to suggest that these problems are not resolvable in the interpretation and indeed many attempts have been made. Nevertheless, the situation is not quite a simple as it is presented in popular accounts of the theory?

    1) Where does the probability rule come from? If all possible universes exist in the wavefunction then why do we experience the outcomes of experiments according to the standard QM probability rule (Born's rule)?

    2) Why does the universe split according to the basis in which we choose to measure in our experiments? In standard QM, there is nothing special about a particular basis, so this problem seems to indicate a priveliged role for measurements, which many worlds theorists are keen to remove from the theory?

    3) In classical physics there is nothing to rule out that all possible universes exist, as in the many worlds interpretation. However, we are likely to regard this statement as an empty one, since it would have no effect on the physics we can do here in this universe. How does the many-worlds theory differ from this empty statement?
  8. Jan 13, 2004 #7
    Wierd thing

    The wierd thing about multiple world theory is that everything that could hppen in a second realy happens in each of the world.
    i read a book that claimed that time travel is possible using the multiple world theory. as there are chances of worlds which are still in 14th centuary.

  9. Jan 13, 2004 #8


    User Avatar
    Science Advisor

    I have allways has one major problem with the Everret Interpretation, it is alluded to in point #1 of Slyboy's post. The first rule of the Many Worlds view is that events in one universe cannot have an effect on events in another. That is why when we perform the "Schrodinger's cat" experiment, we do not open the box and see two cats; one dead and one alive.

    The Everett Interpretation is an attempt to explain experimental results seen and measured many times by people in this universe. The dark and light bands produced by the double slit experiment have been measured by equipment, and seen by the naked eye. This proves that they cannot be the effect of some cause that is outside our universe.
  10. Jan 13, 2004 #9
    I am very interested in the subject of other universes. I am currently reading timeline which has to do with quantum mechanics and w.e. So i thought id check up on it. I was also wondering if it is possible to make a light faint enough so that only one atom passes through at the time.
  11. Jan 13, 2004 #10
    I Don't want to look for the details of a particular experiment that did that, I'm pretty sure it has been done, regardless, I assume that your related question is whether the interference pattern would still show up if only one photon waws emited at a time.... Yes it would, each 'atom' of light interferes with itself (as difficult as that may be to imagine)..

  12. Jan 13, 2004 #11
    What would be some consequences of the many world theory?

    Why is this theory so hard to believe?
    I know that slyboy had some points, but can you explain them again? I didn't really understand what was meant.
  13. Jan 14, 2004 #12
    Problems with many worlds (at great lenght)

    Strictly speaking there are no 'consequences' of the many world theory that aren't given by the standard quantum formalism. We simply have to judge whether it is a simpler or more elegant interpretation of QM. Many interpretations contain features that are hard for physicists to swallow. In the case of many worlds, it is the vast multiplicity of seemingly unobservable universes. The interpretation has to provide significant benefits in clarity if we are going to accept this.

    Having said this, many-worlds has proven to be useful in that the development of the first interesting quantum algorithm was an attempt by David Deutsch to show that many worlds must be true. We don't need many worlds to understand the algorithm, but this highlights how a diversity of approaches to QM can yield real results.

    I don't think it is hard to believe; no more so than many of the other interpretations of QM. Some physicists believe that it is obviously the only way to interpret QM, notably David Deutsch. Many in the Quantum Computing community seem to agree on some level, but there is nowhere near universal acceptance amongst physicists as a whole.

    OK, here is an attempt to explain them as simply as possible.

    1) Probability rule: Many worlds says that whenever a quantum decision is made (for simplicity lets say that there are always two possible choices) the universe splits into two universes. For example, one where a particle goes to the left and one where it goes to the right.

    However, QM predicts not only the possible alternatives, but also a probability associated to it. For example, there might be a 75% probability of the particle going left and a 25% probability of going right. How do we account for this if there are only two universes?

    One way of doing this would be to say there are actually 3 universes where the particle goes left and 1 where it goes right. The probabilities are then given by the relative number of universes corresponding to each choice. This goes beyond the standard many-worlds approach originally proposed by Everett.

    The problem with doing this is that the probabilities could conceivably be any number you like, so we might need a continuous infinity of universes for each quantum decision.

    Other alternatives for getting quantum probabilities from many-worlds have been proposed, notably a decision theoretical approach by Deutsch. However, these also have their own problems.

    2) The basis problem (this one is a bit more technical I'm afraid)

    Imagine that we have a detector that tells us whether a particle goes left or right and there is a 50/50 chance of each possibility. One possible wavefunction for this would be:

    |particle left>|detector left> + |particle right>|detector right>

    Many worlds says that the two terms in this wavefunction represent a splitting across two different universes. However, in QM there is no special significance to this decomposition and we might equally well write:

    (|particle left> + |particle right>)(|detector left> + |detector right>) + (|particle left> - |particle right>)(|detector left> - |detector right>)

    This could be interpreted in a many worlds fasion by saying that we have two universes each of which has a different superposition for the particle and for the detector.

    This may seem arbitrary, but it is not as stupid as it seems. One example that highlights this is the spin states of particles. In this case, (|left> + |right>) is often the same as the state |up>.

    Many worlds therefore has to provide an explanation of why one possible splitting occurs and not any other. Attempts to do this are usually based on decoherence, which is to do with the fact that the detector is a macroscopic object and thus its |left> and |right> states couple to the world differently. These explanations are not very convincing to me at the moment.

    3) My third problem is not really that serious on a technical level. I'm just saying that we could describe classical physics in the same way as the many-worlds interpretation of QM. The fact is that we don't, because there are no benefits to introducing a multiplicity of universes in this case. Is the phenomenon of quantum interference really any more of a reason to postulate all these extra universes? I don't think so, but it would be a boring world if we all agreed.
  14. Jan 20, 2004 #13
    can anyone explain the device that is used to emit just one photon?

    bet ya cant eat just one. First, let me just be up front by disclosing that I have a huge problem with the efficacy of these interferometry models. The polarisers, the material in the instrument, the light emmitters, the detectors...all of it introducing unknown and in many cases, unmeasurable effects on particle collisions.

    Its clear to me that these experiments have been put together with significant flaws for the last 30 years. Just because you only detect one photon does NOT mean that only one was emitted. And there is no way to have two detectors for the same particle. you would never know if there were originally two photons and one you never got or just the one, experimentally it is the same result and there is a major flaw in the great interference experiment. Even if you had two detectors for the same particle, you would never know if it was functional when the particle arrived.

    This is a fundamental flaw of fourth order interferometry and also a result of reducing the real world to theoreticals again and again until it is assumed that the approximation is "close enough". Which is appallingly sloppy science for a group of people who are very concerned that their theories be nice and tidy.

    In all of these experiments, there could be interfering particles which are not photons. There are also problems on the detector end of the apparatus. Measured resolution of one photon does NOT mean that every photon will be detected. It simply means that a photon was detected, not that all photons were detected. They come up with this equipment based on a timing scheme in which unless more than one photon is imaged during a certain interval after the laser being fired, it is assumed that there must have been no other photons. How would you know,for example, if your detector had a flaw that prevented a second late particle from being recorded if it arrived within a certain resolution period of the detector material? You wouldnt.

    This is analagous to getting a letter in the mail and assuming that I sent you only one letter. Perhaps one was lost in transit, or it was delivered but I simply overlooked it.

    Also consider the real world experiment of shining even coherent monochromatic light at a slit. Usually people talk about decoherence in the instrument. But the slit media itself ruins the integrity of coherence. Photons bounce off the edges of the solid material, refracting and reflecting every which way, because as everyone knows, there are no perfectly homogenous surface materials and anything three dimensional that has an edge is going to have protrusions at the atomic if not molecular levels. You might start with the best light you think you can attenuate, but you are going to get diffraction or diffusion patterns simply because you have imperfect surfaces on the edges of the slit. Of course there are interference patterns from this. This is classical EM; you dont need QT to predict the banded shadowing.

    So who is to say that you are not simply seeing the results of an imperfect experiment? Single photon interference is quite simply an untestable theory. Unfortunately much of QT has been focussed on interpreting 3 decades of flawed data but no one has ever been able to repair this experiment. The Hong-Ou-Mandel device introduces the same macroscopic flaws as every other interferometer does. Wasting all this time obsessing over SPI is simply mental masturbatiion and an impediment to understanding the paradoxes of QT.

    The fact that theorists have now come up with the multi-verse to explain what they see as quantum behavior of entangled pairs is completely at odds with Quantum Teleportation. Does anyone see the inconsistency here? Why would a separated sister particle change its spin in our universe when according to the multi-verse, its changing there? What would happen to our particle if someone in the multiverse changed their particle at the same time? I dont like the implications for symmetry here...some particles are symmetrical here,others in the multi-verse? That makes no sense to me at all.

    I think the Many Worlds theory stinks. Its far too elaborate of a description to meet the test of Occams Razor. Why go to these elaborate lengths simply to chase down problems in QT? Lets start with the pesky photon and find out whats really going on there before we start adding 26 dimensions to the theory.
  15. Jan 20, 2004 #14
    No one can doubt that the many worlds theory exists. Every time we sit down at our computers there is the possibility of entering a realm of infinte possibilities. Yet from which direction do we approach this "reality" is it from our perceptions of the world as we see it (i.e physics ) or is it from a completely different perspective (i.e., that as we imagine it. On this question , I feel , rests the whole of the future of physics, and of our future.
  16. Jan 21, 2004 #15
    Hello All:
    For those that believe in this theory, I am curious: What are the limitations of the many worlds theory? Infinite universes are spawning as I type this because I am sitting here typing instead of getting a cup of water or going out. So many universes, in fact, that it almost makes sense to ask what isn't possible? If we can't narrow it down this way, then the theory simply says that anything is possible. Can your limitations(of new universes) stop my immortality from happening (always wondered about the quantum immortality deal)?
    *I don't have a stand on whether I believe it or not.. just very curious!
    Last edited: Jan 21, 2004
  17. Feb 2, 2004 #16
    Re: Problems with many worlds (at great lenght)

    I'm not sure; what about Deutsch's quantum AI thought experiment, or approaches such as this?

    What problems are these? (This is a nonrhetorical question, i.e. I'm not implying there aren't any problems.) Are there major problems that haven't been clarified by e.g. Wallace?

    What about approaches such as this?

    An orthodox Everettist might say that he's not the one postulating universes, you're the one postulating them away. I'm inclined to agree with this: unless it's necessary to augment the Everett interpretation with extra structures (a fundamental probability measure, or something), it seems to me to be clearly simpler than other interpretations. It claims there is more stuff, but the universe doesn't seem to care about minimizing stuff, it seems to care about being simple (i.e. easy to specify).

    I don't understand enough to make a firm judgement on the probability and preferred basis issues (which is why I'm only a tentative "believer") but Occam is a really awful argument against many-worlds. If that's what Occam means, you might as well use it as an argument for solipsism or geocentrism.

    Also bad arguments are those of the form "many-worlds has no observational consequences, so it sucks", or even "many-worlds is not falsifiable, so it's false". These apply equally well to all other interpretations, which is why it's a philosophical issue. If the existence of many worlds is untestable, then so is the existence of only one world.
  18. Feb 26, 2004 #17
    I'm not too familiar with these experiments, but I have seen an argument by Deutsch which says that many-worlds predicts that you could recohere the universes with sufficiently advanced technology. This might invalidate some interpretations, but it seems clear to me that any interpretation that says QM is in principle applicaple to any system of any size has no problems with this.

    There are approaches one can take to these problems. My point was that the arguments have not been universally accepted by physicists, mainly because most of them don't read articles by people like Wallace. This is, of course, no reason to doubt that the arguments might be correct. I am not an expert on his work myself, but I do remember falling asleep in an incomprehensible seminar he gave on the subject once. He is actually visiting my institute at the end of this month and I will try understand his approach while he is here.

    I am generally fairly agnostic about quantum interpretations, but I do think that approaches such as Fuch's are worth considering as alternatives that are possibly simpler than many worlds.
  19. Feb 26, 2004 #18
    Occam is a GREAT argument against many worlds. Its a complexity test. Why should 20-something dimensions be more effective than say, 3?

    Why would the universe be better served by creating a separate physical instance of itself for every fork in the possible chain of events? To me, many worlds implies that every time one thing occurs that represents determination, the universe creates every possible variant of the original event representing every possible outcome.

    That defies explanation of purpose. After all, one would hardly argue that Schrodingers cat experiment is invalid because there are worlds represented where the cat is both live and dead, regardless of the determination made in the original universe.

    If you were going to proceed fully down the many worlds path, the logical conclusion is that somewhere there is a world in which things fall up, constantly, another in which a tossed coin is always heads, and in which light is always polarized one direction. This defies our mathematical understanding of physical processes as well as mundane ones involving social or historical events. How many permutations would you have to have to allow for this kind of construction? I think the answer, in terms of dimensions, approaches infinity, since there will be a separate world based on every particle interaction that ever occurs.
  20. Feb 26, 2004 #19
    "Approaches" infinity is kind of an understatement. I'd say more like far closer to infinity than we could possibly ever imagine, even just in one's own lifetime. No one denies the number of realities required for the Many Worlds theory.

    My question is in regards to preserving energy. How can energy be preserved in a system where two realities are created from one?
  21. Feb 26, 2004 #20
    Yah, its got many contradictions that you would encounter in classic thermodynamics. Personally I think the idea probably began as an intellectual exercise regarding quantum indeterminancy in string theory and somehow got extrapolated to a macroscopic level.

    However it came about, its certainly a very troubling theory from many angles, not the least of which is the thermodynamic entanglements.

    I think its very poor science. Taking Everett seriously leads to many logical inconsistencies. Chasing after unification from a mathematical standpoint, has led to with all these untestable hypotheses. More bothersome is the fact that Everett sincerely believed that wavefunctions exist thermodynamically independent of observation. If Everett is correct, you should be able to control the entire universe with just your thoughts and your mind...the will can literally determine reality. This is a problem for me. Reality, or at least, our view of part of it, exists without us to observe it.

    Copenhagen AND Everett are both incorrect. The effects they are trying to describe do not exist macroscopically. They may be nice ideas at the quantum level, but at our scale, cease to exist. Much as EM forces dissipate with distance...quantum effects are all but gone by the time you have gotten anywhere near even the molecular level.

    but, on the other hand, untestable theories represent job security...its a nice gig for a theoretician.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook