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

Why can't we use quantum bit entanglement for FTL information?

  1. Aug 9, 2010 #1
    I don't understand. Why can't we use quantum entanglement by the means of altering a state of an entangled electron so we can send an instantaneous bit of information to anywhere in the universe infinitely fast by the other electron changing its quantum state? We can can in theory transmit information faster than c.

    Why exactly can't we do this?
  2. jcsd
  3. Aug 9, 2010 #2
    Try this at home with a friend,

    Consider a yes or no message to be sent. Then take a coin and flip it, hold what it lands on up to your eyes and the other side up to your friends. This is like an entangled state that collapses, if you see head your friend will see tails (or vise versa) and you each know what the other sees. Now using this coin alone see if you can transmit a yes or no message to your friend.
  4. Aug 10, 2010 #3
    Okay, I see the point. Is there anyway we could choose what state the electron goes into somehow so then we could transmit a bit?
  5. Aug 10, 2010 #4
  6. Aug 10, 2010 #5
    Okay, what about sequencing when the electron has its wave function destroyed. It doesn't matter what the outcome is, it just depends on the timing of when the function is destroyed to transmit information indirectly. For example, collapsing 3 electron functions in 10 seconds means a specific bit say "yes" of information and collapsing 6 in ten seconds means "no". How can we not transmit information this way? Is this possible?
  7. Aug 10, 2010 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    By looking at my coin, how can I tell if you've looked at your coin yet to "destroy" the uncertainty?
  8. Aug 11, 2010 #7


    User Avatar
    Science Advisor
    Gold Member

    Because acting on one half of an entangled pair has no causal effect on the other half. There is no (reason to suppose a) FTL effect, not even one which is hidden.

    The non-locality issues in Bell/EPR thought experiments are red herrings. The critical issue is that quantum systems do not have objective (classical) states of reality.

    Entanglement is simply correlation. What is different in the quantum case is that the assertion of correlation (like any meaningful physical assertion) implies a measurement has been made (think of preparing the system as measuring a number of systems until one with the desired trait is found). This entangling measurement does not commute with the critical measurements of half the pair we take and thus prior to that second measurement each half does not have its own definite "state". Thus the correlation is meaningful in and of itself but can't be expressed in the classical way of:
    "Either system A is in state x and so is system B, or"
    "system A is in state y and so is system B, or "
    "system A is in state z and so is system B,"

    Each of these statements, when said along with "A and B are entangled so as to be perfectly correlated" is just as wrong as saying a particle has momentum p and has position x or has momentum p and position x'."

    I think part of the problem is the use of the term "entanglement" which intuitively hints at a causal connection, i.e. some string connecting the two halves. Translate "entanglement" to "quantum correlation" and you'll be less likely to incorrectly introduce a classical bias.

    Quantum systems are not just tiny classical systems.
  9. Aug 11, 2010 #8
    Okay, I understand what you mean by both particles still don't have definite states objectively, but wouldn't the particles show any objective sign whatsoever of the collapse of the function?

    That's all we need, just some sign in change of measurement or momentum. Then we could use sequencing of particles to indirectly transmit information like I said above.
  10. Aug 11, 2010 #9
    Wouldn't altering a quantum position change any sign of the position or momentum? One or the other more or less. We could use special measurement tools.
  11. Aug 11, 2010 #10
    Im not sure what you mean. But like Hurkyl says, you cannot tell if the other person looked by looking yourself.
  12. Aug 11, 2010 #11
    You don't need to check. I just want to know if you entangle 2 electrons and seperate them across the universe, could you break one and the other will act instantly by the means of any measurement. It doesn't matter what it does, just as long as it does something. Once this happens you can take more electrons and use a pattern of when the electrons behave in a certain way to transmit information indirectly.
  13. Aug 11, 2010 #12

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Zeromodz, what would it take to convince you that the answer is "no"?
  14. Aug 11, 2010 #13
    jambaugh just complicated the issue by spamming the thread with an unnecessary personal interpretation of QM.

    The original answer by Academic is sufficient: ie there is no known way of deterministically selecting a quantum state so there is no known way of transmitting information via entanglement.

    And you can't tell whether a measurement was made on an entangled partner by the result of measuring the other one, so zeromodz's other suggestion of 'just as long as it does something' won't work either.
  15. Aug 11, 2010 #14
    It doesn't play an alarm or raise a flag or anything. It just carries on being an electron. Nothing comes of it until you make a measurement. When you do, you get a random result. Can you tell if I also obtained a result on my half?

    Worse, it doesn't matter when each side makes a measurement. There is no "first". They just agree. If they are far apart and the measurements made near the same time, there is no universal agreement over the ordering! It depends on the observer, and there is no time ordering between them.

    If you still can't understand, it might do to work through a more concrete example. E.g. you are on a star ship, in deep space. There is an electron sitting in a trap in the safe. Has it "done something"? No, and it won't. At some point you decide to pass it through a polarizer and measure the result. What can you tell from that?
  16. Aug 12, 2010 #15


    User Avatar
    Science Advisor
    Gold Member

    Special relativity challenges the classical ideas of simultaneity. Depending upon the conditions, we could easily have observers that could say that group A measured before group B and another set of observers that say group B was first. Even with these conflicting views, they would all have to be correct. So as JDługosz states, there is no "first."

    Not only that, but you have to measure the system and the simple act of doing this collapses the state. You cannot monitor the system passively on your end and tell when the other group measured the system themselves.
  17. Aug 12, 2010 #16


    User Avatar
    Science Advisor
    Gold Member

    Firstly my "personal" interpretation is the orthodox one. It just doesn't get as much Hollywood airtime since it doesn't predict such ridiculous effects as FTL signals or infinite parallel dimensions that makes for good SciFi plot elements or fantastic article titles in the popular press.

    Secondly the quickest way to prove a theory (with multiple interpretations) doesn't in anyway predict an effect is to demonstrate that a valid interpretation of that theory excludes that effect. (Similar to demonstrating that say the parallel postulate is not a result of the other postulates by demonstrating a valid model satisfying the other postulates but definitely invalidating the parallel postulate).

    Thirdly and finally, I am not initiating the invocation of interpretations here. The whole FTL business comes from a ("mis")interpretation of the QM. I'm just piping in with classic CI for equal time.

    To zeromodz,
    To make it clear, "not having an objective state" implies no observation of an objective state. Without such observation there is no possibility of observing a change. "any objective sign" as I understand your meaning must mean an observable sign.

    Also The wave-function is initially (in the sequence of definition) a representation of our knowledge about the system. It takes an act of (mis-)interpretation to also give it meaning as a representation of a state of reality. In the first case the collapse is just a representation of the change in our knowledge about the system. Only if you add the "reality" do you assert that this collapse is actually occurring "out there" in the same instantaneous way that it is occurring on paper (and thus get instantaneous (meta)physical effects instead of just instantaneous updates in predictions.)

    Without that ("mis")interpretation one can only say that the wave-function collapse is expressing the same sort of discontinuity as we see in a classical probability distribution collapse once one invokes new information. One is making the transition to a conditional probability. "Give we observed X we no longer look at Pr(A) but Pr(A|X)." e.g. "Once we know the number 4 is drawn in the lotto, all the tickets without a 4 instantaneously become worthless."

    Now you can choose an alternative to the minimal interpretation I describe provide it consistent as far as empirical predictions goes. But given it is consistent with QM and given it doesn't predict beyond the predictions of QM it cannot predict phenomena inconsistent with the orthodox Copenhagen interpretation. Since that interpretation is consistent with relativistic causal locality and with QM there is
    a.) no way QM explicitly predicts an observable FTL causal event (of which a FTL signaling device would certainly be an example) and
    b.) no way an FTL causal event is consistent with both QM and relativity.

    This does not preclude an alternative theory: "QM plus a a relativity violating theory" which allows FTL signals. But my point is that adding QM in no way adds to our expectation of FTL effects. Speculating about "hyperwave radios" and "warp drives" are independent of the assertions of QM excepting that they should be consistent with QM if you believe in QM. But it is simply speculating both classically or quantum mechanically that relativity is wrong and not inferring anything at all from QM itself.

    Ask if there is any way we might sent an FTL signal. Why invoke QM per se?

    P.S. Actually, technically Relativity doesn't preclude FTL causality as such. It simply asserts that FTL causal signals as seen by one observer are backward in time signals as seen by another. You cannot invoke FTL without invoking Time Travel and remain consistent with relativity. It is sufficient to resolve the causal loop paradoxes one generates by asserting backward-in-time communication to retain both relativity and the hoped for FTL causation. But without some empirical examples such belief is an act of faith and such speculation is just "out there".
  18. Aug 13, 2010 #17
    By "orthodox one" I think you mean "the oldest one" (it's hard to tell because you just posted vague stuff about correlations), I don't think there's anything "orthodox" about it, it was accepted mostly because the mainstream scientists actually producing results weren't really bothered by the interpretation problem since it doesn't add anything useful to the work they were developing.

    This is still the case really, since the QM + unification project doesn't immediately require a resolution to the interpretation problem.

    In fact the interpretation issue will probably fall out rather easily once the really difficult work of building a mathematical model of microscopic reality and cosmology is complete. Unfortunately this is taking humankind a long time and in the the interim we have to put up with unhelpful would-be philosophers espousing their underwhelming ideas on what it might be like (yawn).

    You have no idea how FTL causality might be operating in QM, it may be just an apparent FTL effect due to holographic projection into the 3 dimensions we observe, or there may be a tachyon field generating a bohm type pilot wave, or there may be other dimensional spaces in which signalling can take place etc etc. The uncertainty principle may fall out of a simple resolution issue due to finite 'pixels' on a remote holographic screen, quantum randomness may occur at singularities in the microscopic dynamics or at dimensional boundaries where new degrees of freedom suddenly become available etc etc

    Of course it may just be a case of matched correlations due to some unknown mechanism or model of reality, but you certainly don't know that.

    You see there is little point answering a question about FTL signalling by appealing to "orthodox" interpretations or similar, the answer to the question is, that as far as we know there is no way of deterministically selecting a quantum state, so we can't transfer FTL information. This is a scientific observation which explains the state of our current knowledge.
    Last edited: Aug 13, 2010
  19. Aug 13, 2010 #18
    A personal thought experiment on classical entanglement:
    Consider a sock draw with N pairs of different colored socks, with each pair clipped together. Alice closes her eyes, takes out a pair, unclips the socks, and puts one in each of two waiting brown lunch bags. She then hands one bag to Bob, who carries it across town on the bus. They are now poised for an EPR type experiment. Can useful information be transmitted instantaneously in this arrangement?
    I think it should be obvious that the answer is no.
  20. Aug 13, 2010 #19


    User Avatar
    Science Advisor
    Gold Member

    I mean "orthodox" one because it is the most widely accepted one. This is the (orthodox) definition of "orthodox" whether any one person thinks it should be or not. http://wordnetweb.princeton.edu/perl/webwn?s=orthodoxy"
    I see no problem...except for those demanding a particular metaphysical interpretation to satisfy their own need to retain an old classical mindset. The operational interpretation is already unambiguously given in the mapping from formal mathematics to physical operations in the laboratory. CI simply makes that explicit which is why it has become the orthodox interp.

    Don't hold your breath. Your supposed need for a "mathematical model of microscopic reality..." is itself interpretationially driven and represents a pre-QM bias towards classical worldviews. What we DO need is a synthesis of the operational, predictive principles of both QM and GR. Recent attempts (Field Theories and String/Brane theories and even Loop QG) have been motivated by this same classical bias. [I'll happily elaborate in another thread or private message if you like] Their lack of success (in incorporating GR) is thus not unexpected (in hindsight).
    See below [Reparse my ...]
    It IS due to matched correlations due to a known mechanism, the prior interactions of the two half-systems. There is no mystery until you try to reconcile the QM prediction with a realist's interpretation.
    Re-parse my last post more carefully. I demonstrated that there IS an answer to the question. It just isn't the answer you want. Given a valid interpretation exists which denies FTL signaling then this demonstrates quite clearly that there is no prediction of FTL signaling within QM itself. As I said QM doesn't exclude the possibility but neither does it suggest it. The question itself doesn't come up unless you wish to --without any empirical or theoretical support-- posit that FTL signaling occurs in nature. Without such an assertion there is no problem to resolve and no need to reconcile FTL effects with QM, anymore than there is need to reconcile QM with unicorn levitation.
    Last edited by a moderator: Apr 25, 2017
  21. Aug 13, 2010 #20
    I can't work out if you're differentiation between FTL signalling and FTL causality, the latter may exist without the former since you may have conservation laws requiring FTL causality (conservation of spin with entangled photons) but have no way of using that for signalling (since neither spin state can be deterministically selected)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook