Talk:Novikov self-consistency principle

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

An extension for Consistency Doctrine[edit]

This is original research - don't delete it - this is a talk page. Just for the record.

So the exception to the "can't change the past" rule, is where the past is ambiguous (example: Bible prophecy, memory, perception, etc.) and that ambiguity might be used as a "buffer" (as in Chemistry) at some future juncture where a dominant teleological outcome exists. In that case the past might be changed sufficiently, having no other consequence, until the "tipping point." That's it.

If you think this is nuts, based on the typical apologetic level of non-theoretical criticism, you wouldn't pass the class entrance test, and you should recuse yourself from making comments.. If you have something constructive to add, or a question to ask, please do. We are dealing with unknown reality here. If it makes you uncomfortable, don't take it as an excuse to foul the drinking water

--Xgenei (talk) 01:35, 20 August 2017 (UTC) — Preceding unsigned comment added by Xgenei (talkcontribs) 01:30, 20 August 2017 (UTC)[reply]

Is this the same as claiming that if multiple pasts are consistent with the present, a time traveler can "choose" among them, ensuring that the present is unaffected? That's interesting! Shankar Sivarajan (talk) 03:58, 29 June 2018 (UTC)[reply]
I had a similar thought on this. The idea being that a traveler still has the free will to make changes as they see fit, but if it disagrees with any first hand knowledge (ie. memories) then there has to be an explanation as to why their memories are inconsistent. That reason would likely come down to the highest probability explanation, which could be some kind of mental illness, hallucination, etc. In this way the less a traveler knows about the future they are affecting better. If you meet a future version of yourself, do your best to forget any details about the encounter, that way when you get to the other side of the interaction you aren't trying to perform what you remember, you're interacting normally and less likely to fall out of the interaction you remember. Otherwise any inconsistencies in your memory will likely translates to real issues with your brain function. — Preceding unsigned comment added by 218.101.54.25 (talk) 23:52, 28 July 2020 (UTC)[reply]

Relation to Uncaused Paradox[edit]

I was wondering if any research into how this concept relates with the concept of an "uncaused paradox" has ever been done by anyone notable. By "uncaused paradox" I mean something similar to the following:

- I receive instructions on building a time machine from the future - I follow the instructions and build a time machine - I go to the past and deliver the instructions to my past self

Simply saying that that has a zero chance of happening seems to be a little axiomatic in this case. The reason for it not happening, would be that it would be a paradox. But "paradox" could be defined in a way consistent with both the Novikov principle and the uncaused event simply by stating axiomatically that a universe where the first event occurs exists; in this way, the second and third events would have to exist, and no "paradox" would have occured, despite the time machine itself having no history or origin. Or to put it another way, the fact that the time machine has no history or origin would, pardon me for indulging, paradoxically be considered to not be a paradox by the model Novikov's principle presupposes.

Or to put the concept I'm trying to address in Lostie terms, "Whatever happened, happened" would neatly avoid addressing this sort of paradox at all in a "not even wrong" fashion.

I'm trying to do a decent enough summary to enable googling, but I'm probably not doing the idea justice. My googling attempts were simply in vain, and if anyone has done anything, it would make a really great addition to the article. And, of course, while it would probably abuse the purpose of the talk page to discuss this kind of stuff right here for fun, I'd nonetheless be game. — Preceding unsigned comment added by 173.22.35.171 (talk) 01:56, 19 January 2013 (UTC)[reply]

There seem to be two, maybe three distinct, separate uncaused paradoxes. Distinguishing them seems to be important. But first, lets review Polchinski's caused-paradox. First, someone manufactures a billiard ball. Then they place it in a certain special spot. At that spot, it gets hit by a billiard ball from the future, at exactly the right angle to cause it to travel through a wormhole, travel back in time, and hit itself. This now much older ball then skeeters off somewhere, and eventually end up in a garbage dump. The point here is that although there were some (mathematical) closed timelike loops, near the trajectory that the billiard travelled. Note that no actual atoms are forever trapped moving along the CTC: the billard atoms are, in a sense, "different" and older, when they strike the younger ball. In particular, there is no paradox involving the (entropic) aging and decay of the plastic materials the ball is made from.
Now, onto the uncaused paradoxes. First, could there ever be an atom, caught forever in a CTC? The paradox there would be "Where did the atom come from?" the answer to which seems to be "it was always there". How about a molecule? a set of molecules? The problem there would be: "why that molecule?" and "why that particular arrangement of molecules?" and "where did that arrangement come from?" The last suggests that its impossible for a billiard ball to be infinitely trapped: as it would have no manufacturer, and the laws of entropy would be violated, as it would never age and decay. So much for atoms travelling on a CTC. How about photons (photons are bosons, atoms are fermions)? How about information? In your original example, it was bits of information - classical bits, not quantum bits, circulating round and round. Where did those bits "come from", as there was no original "manufacturer" for them? What is the likelihood that the bits would arrange themselves in exactly the pattern that they take? Note that this resembles the "uncaused" billiard-ball: it has no manufacturer, so who painted the number on its side? The number carries "classical information", but what to do with that information? Is that information a blue-print to build something? Classical information can be copied, "cloned", quantum information cannot (the no-cloning theorem). I conclude that its impossible (not just unlikely) to have "uncaused" atoms circulating on a CTC, but also "uncaused" information as well, as there is no one to "create" that information. This still leaves open the possibility of a single qubit travelling on a CTC. The no-cloning theorem and the no teleportation theorem seems to imply that there is no paradox for just a single qubit on a CTC. But, if there are two or more qubits on a CTC, then they carry classical information, and the uncaused-paradox seems to result.
Thus, conclude: there are no uncaused quantum field excitations traveling in a CTC. This leaves open the "uncaused" quantum field vacuum state traveling in a CTC ... Hmmm. 67.198.37.16 (talk) 15:16, 21 March 2016 (UTC)[reply]

Probability 0 doesn't mean impossibility[edit]

The opening paragraph implies that if an event has probability zero then it cannot occur. A counterexample is choosing the number "3" out of all of the integers. The probability is zero but choosing "3" is not at all impossible. —Preceding unsigned comment added by 18.248.5.194 (talk) 04:42, 19 July 2010 (UTC)[reply]

The probability is 1/Infinity which approaches 0, but is not 0. Probability 0 does indeed imply impossibility (under the provided knowledge-base). --IO Device (talk) 09:22, 8 August 2011 (UTC)[reply]
It's actually forbidden by the rules of probability to have a uniform probability distribution on the integers (i.e. for any two possible integers, you have an equal probability to pick either), as this would lead to paradoxes--for example, suppose someone picks two integers in this way, one after the other, but doesn't tell you which ones he got until you ask. Then if you ask what the absolute value of the first integer chosen was and get an answer, you should conclude there's a probability 1 the second integer must have had a larger absolute value, since whatever answer N he gives you, there are an infinite number of integers which have a larger absolute value than that, and only a finite number whose absolute value is smaller than or equal to N. But by the same logic, if you instead first ask what the absolute value of the second integer chosen was, you should conclude with probability 1 that the first integer must have had a larger absolute value. For a similar sort of paradox that can arise from imagining a uniform probability distribution on the set of all integers, see the two-envelope paradox.Hypnosifl (talk) 21:15, 26 February 2012 (UTC)[reply]

The original commenter is correct, and IO Device is wrong. Probability 0 is not the same as impossible. See http://en.wikipedia.org/wiki/Almost_surely . — Preceding unsigned comment added by 65.254.127.187 (talk) 21:54, 16 September 2012 (UTC)[reply]

Yes, I also agree, IO is wrong. Probability of zero does not imply the impossible event. The wording of this should be corrected. — Preceding unsigned comment added by 114.244.190.98 (talk) 07:18, 30 December 2012 (UTC)[reply]

Clarify[edit]

this article need more clarification i think. does it simply means that it time is unchangeabe and you cannot go back in time and marry you mother because you haven't done it already? No paradoxis are possible beacause everything has already happened?

then it should say so. --Alexandre Van de Sande 17:19, 5 Aug 2004 (UTC)

Novikov didn't use an example ("Rather than consider the usual models for such a paradox, such as the grandfather paradox..."), he used a mathematical/quantam model. Salasks 21:29, Aug 5, 2004 (UTC)

This quote from the article makes no sense: "if an event exists that could give rise to a paradox, then the probability of that event happening is zero"

Let's say that a time traveller has gone back in time to when his grandfather, a paratrooper in WWII, is going to parachute down into occupied France during the Normandy invasion. He knows that his grandfather is going to survive the war and go on to father his father, but the time traveller is perverse and so he arranges a minefield in the area his grandfather's going to land. There are many different scenarios for how this could play out, with a certain probability of his grandfather landing on any given square foot of land and then subsequently stepping on any given square foot of land. However, according to the Novikov principle, the probability of him stepping on one of the patches where there's a landmine is exactly zero - there's no chance of it ever happening, because it didn't happen, and if it did then it would "change history" (actually, there's a small chance he could step on a mine, but if that happens the mine will turn out to be a dud with 100% probability). Novikov proved this mathematically, using certain assumptions of how physics worked and a simpler model involving billiard-ball-like particles. Bryan 16:05, 18 Jun 2005 (UTC)
My problem with this proposition is that even if the landmines don't blow up his grandfather, the time traveller will have changed infinite things in the past by his mere presence. The same deal with the "historic fire" example in the article: the traveller could bump into people on the street, distract people, leave his footprints, etc. which would alter the outcome of the future, perhaps dramatically. The only way the self-consistency principle can be true, as I see it, is that the universe maintains self-consistency by not allowing time travel at all. 65.11.216.245 08:55, 15 March 2006 (UTC)[reply]
But the point of the principle is that none of those things is different from what "originally happened." There always were footprints there, the time traveller just never knew about them until he made them himself. The distractions are what caused people to do the things that ultimately led to the time traveller's version of the present. Etc. Novikov applies to everything that happens, both big and small. Bryan 17:12, 15 March 2006 (UTC)[reply]
But the only reason you can say those things originally happened is that the time-traveller didn't know about them. So someone who doesn't know the Titanic sank could go back in time and board the Titanic for some other reason (e.g., a cruise), notice it was starting to sink, and find a way to stabilize it. I suppose that when he returned to the present, as soon as he learns that the Titanic sank he'd teleport to a likely sleeping place and realize he was only dreaming? :) 166.137.101.169 (talk) 04:23, 13 July 2014 (UTC)Collin237[reply]

Philosophy[edit]

Some philosophers like Richard Hanley claim that the self-consistency principle is just non-contradiction which philosophers have used for thousands of years. gren 05:02, 18 Jun 2005 (UTC)

Flaw[edit]

This principle has a major flaw: one travelling into the future to pick up some object and take it back shall get the object, whatever it may be (including non-adherence to physical laws), because relatively to them, the people giving one the object have already recieved it from him when one returned.

I don't get your post but, if the NSCP is correct then whatever it is you're talking about won't happen (it might happen but then cancel itself out to achive an equilibrum), even if you have to be crushed by cement prior to doing whatever you said, if (and only if) what you said was a paradox. If it wasn't then the law does not apply. I might be terribly wrong, but this is how I understand it. -- Hexagon1 15:42, 15 January 2006 (UTC)

Another Flaw[edit]

The "time loop logic" program must be impossible, as it provides an easy way to create a machine that can fabricate any desired object from nothing. To wit:

  1. Define criteria C for a desired object. A 10,000 lb. diamond, for example.
  2. Allocate a communication channel c.
  3. Receive an object X from the future on channel c.
  4. Test that X meets the criteria C.
    1. If so, send X backwards in time on channel c.
    2. If not, transform object X into a new object Y (e.g. by attaching a "REJECTED" sticker to it) and send it backwards in time on channel c. Note that this results in a paradox, as the object received in step 3 above is not the same as that sent in this step.

The resolution to this paradox is clear: Nothing will ever be received at step 3. 71.178.183.147 (talk) 05:58, 21 October 2009 (UTC)[reply]

Your machine would not work, but it is not because it would otherwise create something from nothing (see my comment further down). Conservation of energy/conservation of mass does not necessarily apply in this situation, and even if it does, it can be accounted for in other ways, like a release of energy in the future and an absorption of heat energy in the past equivalent to the mass of the object (future goes BOOM for massive objects, but it could still work). LlubNek (talk) 10:22, 3 January 2011 (UTC)[reply]
Aside from choosing an impossible object, what is the paradox that you find here? Highly beneficial, it is, but I see no reason that it should be impossible.
Remember, you only receive an object in step 3 if it does in fact meet your set of criteria C. So if you pick something that is possible, like a piece of paper with your name written on it, then there is no paradox. You receive the object in step 3 and it meets the criteria in step 4.
If you pick something that is impossible, then you receive nothing in step 3, and so you have nothing to test in step 4. There's only a paradox when you both receive something and that something does not meet the criteria. Which, according to self-consistency principle, cannot happen as it would create a paradox. —Preceding unsigned comment added by 163.191.202.2 (talk) 22:38, 20 November 2009 (UTC)[reply]
You can attempt to send back a default or "randomly generated" object if you receive no object from the future. But the probability of the system somehow failing would likely always be higher than the probability of receiving just the right object. Also, if you're sending the SAME object, not just the same type of object, then the object itself has aged. It is not the same object which was received from the future, and will therefore produce a paradox (or rather, cannot be sent back successfully).
Likewise, the information version presented in the article wouldn't work either, since you still need some way of handling not receiving anything. You could have it set F to a random number if it times out waiting on channel c, but you would need to add an actual test for a prime N, and, for large values of N, it may still be more likely for the system to fail than for it to produce a correct result. LlubNek (talk) 10:06, 3 January 2011 (UTC)[reply]
I don't think that this "logic" could ever work. Imagine following scenario: You open a chronochannel, then you receive a boolean number - 1 or 0 - from the future. Then you send boolean opposite (0 for 1, 1 for 0) back. Or a random number if you received nothing at all. The probability of this stuff running perfectly is zero, so work it will not. But if it won't work, then this "chronochannel" does not have 100% reliability (quite a logical thought, too - just what information channel has 100% reliability?) And if chronochannel does not work 100% sure, then this paradox-based logic does not work at all: for, if there is even a slightest chance that you will send one thing in the future and receive another one in the past, then you can't rule out the possibility of a scenario where you receive the wrong answer, see that it is wrong, send back another answer, and still receive the aforementioned wrong one in the past. Even if the probability of this happening is really small, by the way of how conditional probability works, this "small" probability can quickly rise up to 100%, especially in the case of complex problems.
Or is there a flaw in my logic? —Preceding unsigned comment added by 91.78.2.69 (talk) 00:35, 30 August 2010 (UTC)[reply]
Sounds like an argument for the many-worlds interpretation. There are two timelines, and both happen. You've build a kind-of mobius strip or double-cover. 67.198.37.16 (talk) 11:04, 21 March 2016 (UTC)[reply]
The key is this: You're doing the same thing as saying, "This sentence is false." Just as self-referential sentences can result in paradoxes, so too would a self-referential equation.
sent value = opposite of received value
received value = sent value
received value = opposite of received value
The point is that with self-consistency in effect, you cannot change the past, and thus you cannot send back a different value than that which you received - even if you decide or discover that the value you received is not the value you would prefer to have received. 163.191.202.2 (talk) 15:59, 23 September 2010 (UTC)PedanticTwit[reply]
You can send back a different value, but because there is only one self-consistent universe in this model, you can only receive one value. You can't receive a different value in the next iteration of the loop because there is no next iteration, just the same iteration. In other words, either the transmitter in the future or the receiver in the past may fail (or both), but the value your system sees is fixed. LlubNek (talk) 10:31, 3 January 2011 (UTC)[reply]
I agree that this time logic has a fatal flaw. There's so many unanswered questions, like at what F do you start? 2? How would you know that the lowest prime factor would be returned? Isn't the principle of induction being broken here? Otherwise what's the point of incrimenting F by one if the number is not a prime factor? Think of it this way, what if you do recieve a prime factor, then either it must have been sent when another F was recieved that was not a prime factor (F-1) or you have immediately skipped to a prime factor. Which prime factor did you get? How can your initial condition be the number you recieve? What's to prevent the number you recieve from ALWAYS being an abort number, 0 or 1? Or what if you recieve no number at all, or garbage data -- a letter? It seems to me that this model assumes that the prinicple of induction still holds enough to provide you with a prime factor. That because you have the F+1 incrimentation, and you assume that the recieved inital F is <= the "correct" F, that you'll receive a prime factor at all. Isn't it just as logically consistant that the sun would explode? I turn on my prime factorization machine and in step 2 the sun explodes, destroying my machine and preventing any illogical number from ever being received in step 3. Why is it that the incrimental step 4-2 leads to the desired outcome of a prime factor instead of any other consistant outcome, like sun-splodey? Usually induction would be the tool which makes step 4-2 gaurentee that all cases follow from the base case. But induction requires a starting point, that base case, which now can ONLY BE the F that is itself a prime factor so we've prevented step 4-2 from ever ocurring! How can we arrive at where we want F to be if we can never use the path that's necessary to get there? It's like saying, "I will use my time machine to teleport to the hardware store, or if I don't end up at the store, then I will push my time machine 1 foot closer to the store and try again. Because I could just push it all the way to the store, I never actually have to because I already did." 165.214.14.24 (talk) 16:26, 28 May 2012 (UTC)[reply]
There are tests to show whether or not a number is prime. — Preceding unsigned comment added by 24.19.171.138 (talk) 21:31, 4 August 2012 (UTC)[reply]
And who is guaranteed to perform such a test? A spooky spirit?? 166.137.101.169 (talk) 04:38, 13 July 2014 (UTC)Collin237[reply]

My interpretation[edit]

This principle is very straightforward if you look at the universe with certain assumptions:

  • there are a set of equations that are the laws of physics (some of which we may know, some which we may not), which describe how things work through spacetime in the universe
  • you assume counterfactual definiteness
  • the universe (or a universe) is just a solution of the above equations over all of space and time (like solutions of differential equations in math), from an external point of view

Consequences:

  • for some initial conditions, the solution may not be unique (there may be many equally plausible universes that satisfy the same initial conditions)
  • in any particular universe, there is only one thing which happens at any given place and time, which is predetermined; but they may not necessarily need to have a cause
  • by the nature of being a "solution", a universe must by definition be consistent; so that hypothetical universes containing "inconsistent" actions must somehow violate the laws of physics

The preceding unsigned comment was added by Spoon! (talk • contribs) .

Please sign your posts. +Hexagon1 (talk) 07:27, 4 March 2006 (UTC)[reply]

A presentist view which makes the principle redundant[edit]

The principle is based upon an assumption that if time travel were possible the time traveller could return to an earlier state of the universe. The following explanation shows that asumption to be wrong, and as a consequence makes the principle redundant. Every particle in the universe, including those that make up me, travels on a trajectory through spacetime. Suppose I was to travel back through time to a point in spacetime that used to be occupied by the atoms that comprised my grandfather (say the year 1906): I would find that the atoms which comprised my grandfather were not there, as they were 'still' in 2006 where I had left them at the start of my travels. The only atoms that travel back in time are the ones that constitute the body of the traveller- all of the other atoms in the universe continue on their spacetime trajectories as before, and so are not found by the traveller when he/she arrives in the past. Given this, none of the supposed paradoxes arise, and so the self-consistency principle (in this context at least) is redundant. This seems pretty self-evident to me- am I missing something?

DoctorDen 1 April 2006 17.44 GMT (not logged in).

Yes. You're missing the fact that particles travel through space, not spacetime. In spacetime, the particles occupy positions. The whole of spacetime is a single construct. "The atoms that comprised my grandfather in 1906" (where we implicitly understand "1906" as some exact point in time) have a definite constant position, that we assume our time traveller can reach. Those atoms have a trajectory that will show them in a different spatial position in 2006, but they do not "travel" from the point of someone who can oversee all time.
Your point of view would imply that the past doesn't exist (physically), and hence time travel as a whole is impossible. This is a fine way of looking at things, in principle, but if time travel is impossible (as it likely is) then obviously none of this discussion is relevant. In any case, this is more relevant to the time travel article. 81.58.51.131 08:24, 7 April 2006 (UTC)[reply]

Doctor Den says: Thank you- that is very helpful. Can you recommend any references that explain how atoms have a definite constant position in spacetime, as I would like to read more.

DoctorDen 7 April 2006 20.35 GMT (not logged in).

A good article to try might be block time, it's all about this sort of conceptual model. Bryan 22:18, 7 April 2006 (UTC)[reply]

No mention of quantum mechanics?[edit]

Arn't the strongest arguments for the self-consistency principle all based upon quantum mechanics? For example, if your billiard ball was a quantum particle, then its interfearing with itself merely makes those trajectories rediculously improbable? I don't know enough about it to add this intelligently, but please get some input from someone who knows about quantum mechanics. JeffBurdges 19:35, 20 June 2006 (UTC)[reply]

Seems that way to me as well, in classical (non-quantum) physics if you know the position + velocity of the billiard-ball precisely the system is entirely deterministic and there is no probalility involved at all; so talking about the probability of when happens when the billiard ball hits itslef makes no sense unless you invoke quantum mechanical ideas (i.e. the unceratintly principle). Tomgreeny 18:24, 5 September 2006 (UTC)[reply]

Part of what makes this approach so intriguing is that it clearly shows that macroscopic objects, moving according to simple Newtonian Dynamics, cease to behave deterministically if self-interaction via time-travel into the past is introduced. The collision itself is still entirely "classical", but since the position and time at which the collision occurs is dependent on the trajectory of the future ball, which in turn is dependent on the position and time of the collision, things become non-deterministic without resorting to any specific kind of non-classical physics.
TeraBlight 00:04, 20 September 2006 (UTC)[reply]

I don't get it.[edit]

So if someone would be to travel through time, their actions would have no effect?

But time travel itself affects the past -- you're taking something from the present and putting it in a different point in time, thus, for the context of the earlier point in time you travel to, introducing matter/energy/whatever that didn't exist there before, thus (as per chaos theory) possibly screwing up everything. Any successful travel through time would have an effect, otherwise it wouldn't be time travel.

With that in mind -- does the principle boil down to "time travel is only possible if it has no effect, thus time travel (the possibility of which necessitates having an effect) is not possible"?

The matter and energy of which the time traveller (and his clothing, etc) consists would (if the universe is a closed system) have been in use by something else at an earlier point in time and thus a re-allocation of matter and energy already existent at that earlier point in time to recreate the time traveller doesn't seem to be a viable alternative either. But if the time traveller can neither be composed of matter and energy that didn't exist, nor of that which already existed but was "in use" already, what does he consist of?

Even if the universe is not a temporally closed system (i.e. matter/energy can vanish and re-appear at future points in time), for something to vanish in the present and re-appear in the past, it would have to vanish before its original appearance unless it can't exist twice at the same point in time (i.e. if atom A was sent back in time from a point Z to a point X, it would vanish at point Z (now) and appear at point X (back then) as A', but have to vanish again as time advances closer to the point Y (somewhere before point Z) where it was originally introduced as A, otherwise A and A' (which should be identical) would exist at the same point in time, which would definitely be an alteration of past events and thus incompatible with the principle.

I'm assuming the principle isn't based on the simplification of the idea that the only thing that "matters" is the action/inaction of humans (as in the interpretation of Schroedinger's experiment that the cat is in an unknown state where it is both alive and dead until the human interacts with it -- which implies the human's perception determines the state of reality). If that is true, then the example from Twilight Zone where someone goes back in history to start the fire he travelled back in time to prevent is clearly contradictory as it introduces a paradox (a time travel having to have occurred in order for the time travel to occur) unless it actually changed the events (i.e. originally the fire would have been caused by something else), which would still be inconsistent with the principle which doesn't allow for any changes of the past.

I'm aware this post is getting a bit rant-y, but I'm seriously at a loss here. How does time travel NOT change the past, regardless of whether it has any consequences in the grand scheme of things or not? The very concept of travel backwards through time is that one appears at an earlier point of time, thus changing the past.

The only possibility I can think of for time travel not to change the past is a paradox in which the time travel happens because it has already occurred (i.e. the future representation (person A') of someone (person A) appears in the present and at some future point in time the same events cause person A to travel back in time that "already" caused person A' to go back in time, i.e. both of their time lines are identical), but in which there is no "other" time line in which the time traveller didn't already appear before travelling back in time (i.e. there is no time line in which person A travels back in time without person A' having travelled in time already, i.e. where person A doesn't eventually become person A', the time traveller).

All of this is obviously rendered null and void if the laws of physics are inconsistent and random stuff does appear out of thin air for no reason. — Ashmodai (talk · contribs) 18:14, 20 December 2006 (UTC)[reply]

That last "possibility" you offered sounds close. Here's a quote from the article:

In another example, taken from an episode of The Twilight Zone, first referenced by Henry James, a person travels back in time to discover the cause of a famous fire. While in the building where the fire started, he or she accidentally knocks over a kerosene lantern and causes a fire, the same fire that would inspire him or her, years later, to travel back in time.

So the person learned that the fire had existed well before time traveling, so that person was already a part of history and didn't change anything. I suggest rereading the "Potential implications for free will" section in the article. --Brandon Dilbeck 18:37, 20 December 2006 (UTC)[reply]
The key here is to realize that the fire could have occurred without the time traveller's intervention. Had he not travelled back in time, then something else would have caused the fire anyway. LlubNek (talk) 11:36, 3 January 2011 (UTC)[reply]
That's an incorrect reading of the principle--what you're talking about sounds more like a type of "fate" where certain important events are guaranteed to occur, but other more trivial events can change. The principle says that nothing can be changed by a time traveler, not the slightest detail. If you want to think about questions of "if things had gone a little differently", you're free to imagine a multiverse of totally independent, non-interacting parallel histories which differ from one another but each one internally allows time travel and obeys the Novikov principle (Stephen Hawking talks about an idea like this at the very end of this article, in the paragraph that begins "Instead, the universe has every single possible history..."), but in this case there would be no reason to imagine that if a major fire occurred in one of them, it would have to occur in others as well. Hypnosifl (talk) 21:26, 26 February 2012 (UTC)[reply]

Yeah, this doesn't make any sense. By: Ron Francis —Preceding unsigned comment added by 99.245.249.223 (talk) 01:41, 26 November 2008 (UTC)[reply]

"if someone would be to travel through time, their actions would have no effect?" -- Incorrect reading of the text. The correct conclusion would be "If someone were to travel through time, the effect their actions would have would be to cause events to come to pass the way they already did." - In other words, only an ontological paradox is possible, not a contradictory one. You could go into the past and attempt to kill your younger self. However, you could not succeed. BUT the event of an older you attempting to kill the younger you COULD potentially be the event that inspires you to travel back in time and try to kill yourself (i.e., while it may seem crazy, you could try and kill your younger self to prevent your older self from trying to kill your younger self, failing, but creating the pattern of causality that leads to your attempt. Eh... wibbly wobbly, timey wimey. Dodger (talk) 22:18, 13 May 2011 (UTC)[reply]

The self-consistency principle is wrong[edit]

There are so many problems with Novikov's self consistency principle. Time travel is unphysical [1]. Here are some of the serious problems of the self consistency principle:

1. The Polchinski paradox[2]: a billiard ball falls through a wormhole, travels back in time, collides with its younger self and prevents its falling in the wormhole in the first place. The supporters of time travel explain this by rejecting the initial conditions leading to this scenario. In classical physics there is no such mechanism, and regular quantum mechanics cannot enforce this either.

2. The self-consistency principle demands that free-will is only an illusion I. D. Novikov, Notion of the past & can we change it?, On Time.-Suppl. to EAS Newsletter 18.. This is wrong in so many ways. Leaving the physics arguments, let us observe that this is very similar with the utopia of totalitarian arguments The Open Society and Its Enemieswhere human rights (local physics) are trampled by the totalitarian state demands (global self-consistency condition).

3. With no claim of exactness, Godel’s second incompleteness theorem shows that an axiomatic system can prove it is consistent if and only if it is inconsistent. First let us clarify this statement. If a theory is inconsistent, then it can prove anything, like the sky is pink, and therefore it can certainly prove that itself is consistent and correct. The reverse is not trivial, and what Godel showed was that if a theory can arrogantly state about itself that is infallible (starting from its axioms, or original assumptions), then the theory is garbage. If Novikov’s self-consistency principle is part of the axioms of the Universe, then the Universe can be proven consistent, and thus it will be inconsistent. If on the other hand the principle is not part of the axioms of the Universe, than it is ad-hoc, no more than wishful thinking to avoid the paradoxes of time-travel. Therefore the self-consistency principle is either too arrogant (and wrong) or is just a statement with no physical consequences. This criticism is in regards to the tautology comment on the main argument.

There are also other very significant arguments against the principle (involving deeper knowledge of physics).{{[[Florin Moldoveanu 20:09, 28 December 2006 (UTC)]] 18:37, 28 December 2006 (UTC)}}[reply]

Talk pages are for discussing articles themselves on Wikipedia, and not for debating or assessing the topic of the article. You are of course welcome to add criticism to the article from a neutral point of view, as long as you keep in mind the no original research and conflict of interest policies.
Also, please keep in mind to sign your comments on talk pages with four tildes (~~~~) -- intgr 15:58, 27 December 2006 (UTC)[reply]
What you ought to do, Fmoldoveanu, is find some documentation (that you didn't write) and add that information into the article, adding citations to the source.
You'd also have to make sure that the average reader can understand the points you introduce. I can't understand arguments 2 and 3 you provided above (and an average reader shouldn't have to seriously look up every other word). --Brandon Dilbeck 18:56, 27 December 2006 (UTC)[reply]
To complete the trifecta, argument 1 is explicitly addressed in the article... --Sabik 11:01, 9 February 2007 (UTC)[reply]
Let me reply to the comment that argument 1 is addressed in the article. Here is a quate form the Rama and Sen paper linked at argument 1:
"We have described a scenario first in 1 +1 and then in 3+ 1 dimensional space time which leads to an irresolvable paradox in the presence of a time machine. No consistent classical solution exists for this scenario. Since the system is macroscopic with an action large compared to Planck’s quantum, it is unlikely that quantising the system will resolve the paradox. Moreover, for a macroscopic system with no consistent classical solution, it is not obvious how path integral quantisation can be carried out nor is a Hamiltonian method of quantisation available in the presence of a time machine. Forbidding by fiat all the troublesome initial conditions will solve the problem. However, no known physical law can enforce this kind of censorship."
Novikov's principle would demand that not all initial conditions are acceptable and this is not clear in the text in the main article. Along on the same lines, here is a fourth argument (there are even more distict arguments available, but not yet published in a peered-reviewed journal):
4. Why for a flat space-time manifold, the Nature always obeys the metric diag (-1,+1,+1,+1), and not say diag (-1,-1,+1,+1)? The second case would correspond with a Universe with 2 time dimensions where the ususal Minkowsky causality cone can be circumvented and closed causal loops would be possible. If Novikov's principle were true, then just by restricting all the initial conditions that would generate paradoxes, then that Universe would be paradox-free and consistent. However, Nature exhibits only one time dimension for a very good reason: there is no mechanism to restrict the "bad" initial conditions. A Universe with more than one time dimension, just like a Universe with a general relativity solution with closed time-like curves is always going to be paradoxical and would self-destruct. -Florin Moldoveanu 06:00, 10 February 2007 (UTC)[reply]
Wikipedia is really not the place to debate ideas such as this, but to respond to your points:
1. The whole point of the self-consistency condition is that it is an additional constraint on histories beyond existing local laws (classical or quantum), no one is saying you can derive it from these local laws, so your complaint that the laws "cannot enforce this" doesn't make sense.
2. All the known laws of physics say that free will is an illusion--they are all either deterministic or stochastic (containing a random element). Free will is not supposed to be the same as randomness--if you hook up an A.I. program to a random number generator this obviously doesn't give it "free will", as the random element is totally out of its control and has nothing to do with its preferences or thoughts or emotions.
But any physical principle by which a time machine is built necessarily goes far beyond what is known. Assuming that a time machine is possible abuses general relativity, by extending it beyond its known range of applicability. Once you've allowed such an abuse in your debate, your opponent can similarly abuse the Copenhagen Interpretation and claim that it supports free will.
Anyway, laws of physics are valid only if they have been experimentally verified. But the ability to experiment with a wide enough variety of cases to pin down a law-like relation requires free will, to ensure that the experimental setup isn't controlling our choices. So if free will is an illusion, then the laws of physics cannot be verified. But then there is no basis for rejecting the a priori belief that free will is real. The contradiction can easily be avoided simply by noting that there is no evidence we've finished discovering all the laws of physics. 166.137.101.169 (talk) 05:26, 13 July 2014 (UTC)Collin237[reply]
3.Your argument about Godel's incompleteness theorem is based on an overly vague notion of what the theorem says. It doesn't correspond to the nebulous english statement "an axiomatic system can prove it is consistent if and only if it is inconsistent", it is a rather precise mathematical statement about axiomatic systems whose theorems can be interpreted as representing statements about arithmetic, and it shows that if the axiomatic system is sound (meaning that it only generates theorems that correspond to true statements about arithmetic), then there must be a formula which corresponds to another true statement about arithmetic (one which can be seen as equivalent to a formal assertion that the axiomatic system is consistent) that the system can never prove or disprove (see the definition here). I don't know how you'd find a way of mapping statements of physical facts about a universe generated by certain laws of physics onto only true statements about arithmetic, but even if you could do such a thing, nothing would stop all the statements of physical facts from being consistent, Godel's theorem would just show there is some possible physical fact (like the statement that a particular physical system will eventually reach a particular configuration) corresponding to a true statement about arithmetic that will never actually occur in the universe. This isn't really any more problematic than the idea that if you run a certain computer program forever (one running a simulated universe, perhaps), you may be able to prove that it will never generate a certain output (the simulated universe will never arrive at a certain configuration of simulated matter, say, in much the same way that it can be proven that Conway's Game of Life can never evolve into certain Garden of Eden patterns). Hypnosifl (talk) 02:17, 8 March 2008 (UTC)[reply]
Also, in response to the paper by Rama and Sen that you linked to in argument 1 (though your link wasn't working correctly), see this paper by Krasnikov which shows how an apparent paradox involving two billiard balls similar to the one analyzed by Rama and Sen can be resolved if you introduce new billiard balls which exist only within the time travel region, whose world-lines form closed loops in that region (closed timelike curves). This is a bit like the strange situation with the watch in the movie Time and Again which is discussed here. So, no counterexample has been found to the principle that you can always find a self-consistent extension to any set of initial conditions outside the Cauchy horizon which is the boundary between the time travel region and the no-time-travel region, although the events inside the Cauchy horizon might have to be quite strange. Hypnosifl (talk) 23:22, 9 March 2008 (UTC)[reply]
Additional response to point #1: The argument that "supporters...reject the initial conditions" would be a valid one, except that in this case the hypothesis itself states that those initial conditions are impossible and that, instead, the only possible outcomes are ones in which either the ball misses entirely upon exiting the wormhole, thereby not diverting it's earlier self from its course or where the ball strikes itself in such a way as to make it strike itself in that way. To not reject the initial conditions would be in instantly agree with the objection, at least in the case of this argument. Accepting the initial conditions would be to accept that the principle is false immediately, as the principle itself clearly states that the conditions are not possible. Dodger (talk) 22:31, 13 May 2011 (UTC)[reply]
  • This is not a forum for discussing the self-consistency principle. This talk page is a place for discussing the writing and editing of the article about the self-consistency principle. Wryspy (talk) 15:52, 10 March 2008 (UTC)[reply]

Not very good article[edit]

This is a fairly crappy article, peppered with original research. Perhaps a philosopher who understands the subject of time travel could patch it up? Ben Finn 15:56, 16 April 2007 (UTC)[reply]

External links[edit]

The "The Logic, or otherwise, of time travel - in simple language" link is biased and offers no information. It only asks the basic questions about time travel paradoxes, and offers no answers or even theories. Bredd13 21:14, 8 July 2007 (UTC)[reply]

Agreed. Wryspy 23:02, 8 July 2007 (UTC)[reply]

Affect vs. change[edit]

"Horwich also argued that it was possible to affect the past, but not to change it." What is the difference? 86.164.186.238 (talk) 12:10, 20 January 2008 (UTC)[reply]

The idea is that you can influence the past, but any influence you had on history must have been part of history all along, so for example a time traveler could not successfully kidnap a historical figure who they knew historically had not been kidnapped (unless the historical record turned out to be wrong), but a time traveler who knew about the disappearance of Jimmy Hoffa might go back and kidnap him, taking him back to the future--in this case it must always have been true that Hoffa disappeared because he was kidnapped by a time traveler, there was no "original" history where he disappeared for some other reason, so history was not "changed" by the traveler. Hypnosifl (talk) 02:25, 8 March 2008 (UTC)[reply]

Fourth-Dimensional perspective[edit]

So taking an imaginary look at this in four dimensions, where one can see all of time like a strip of film: According to the NSCP, there is no paradox because the time travel is built into the film in the first place. Then events just happen. The film is already extant, sans paradox. Applejuicefool (talk) 15:06, 11 March 2008 (UTC)[reply]

Potential implications for free will[edit]

So you could travel back in time but you would be constrained from carrying out any event that would change the past. It's difficult to see exactly what form this constraint would take - if you went back with the intention of changing the past would you find your efforts thwarted at every turn, or would you simply lose the intention? Of course this principle might be upheld just as firmly if going back in time caused you to arrive dead, comatose, amnesiac or paralysed. But of course the most surefire way to uphold it would be for time travel to be impossible in the first place. Lee M (talk) 19:37, 21 March 2008 (UTC)[reply]

If you want to see an example of it, watch The Time Machine (2002). In it, the protagonist built a time machine to go back and prevent the death of his to-be finance. However, every time he prevented her death, she would die soon thereafter in a different way. It was later explained to him that without this impetus, he would have never built the time machine in the first place. So the only changes that he could make in history would be ones that would cause the chain of events that would cause him to build the time machine. Admittedly, having time travel be impossible would be the easier solution, but this principle attempts to provide a solution if time travel is possible. — Val42 (talk) 20:09, 21 March 2008 (UTC)[reply]
That movie doesn't fit the idea of the Novikov self-consistency principle though, since the protagonist does at least change the manner of his wife's death. The idea of the Novikov self-consistency principle is that there is a single fixed history which already includes all the influences of time travelers, not even the tiniest detail can be "changed". Hypnosifl (talk) 09:12, 25 March 2008 (UTC)[reply]
Okay, The Time Machine (2002) isn't a perfect fit, but it is much easier to follow than Twelve Monkeys. — Val42 (talk) 04:32, 26 March 2008 (UTC)[reply]
Well, how about Bill & Ted's Excellent Adventure? Both in the scene where they met their future selves (which was shown twice, from both points of view) and in the scene where they broke various historical figures out of jail by using helpful items (like the dad's keys, which had been missing since the start of the movie) that they planned to go back and plant in the right spots later using the time machine, the movie was actually a pretty good illustration of the idea of a single unchangeable timeline which already incorporated the actions of the time travelers. Hypnosifl (talk) 00:10, 28 March 2008 (UTC)[reply]
Okay, you make another good point. — Val42 (talk) 05:26, 28 March 2008 (UTC)[reply]
"preventing disasters from happening in the past and the potential paradoxes this may cause (notably the idea that preventing the disaster would remove the motive for the traveller to go back and prevent it...)" Of course, there's always Peabody and his boy Sherman. As I recall, they visited the past in order to visit some famous person, only to find there was a disaster preventing the person from doing whatever they were known for in the future. Peabody and Sherman then prevented the disaster, preserving our own time line.Mcswell (talk) 01:40, 9 October 2012 (UTC)[reply]

Heinlein[edit]

There should be a mention of Heinlein's short story "All you zombies..." imho.85.28.87.67 (talk) 19:53, 17 June 2009 (UTC)[reply]

There should also be mention of assorted Doctor Who episodes and shorts, particularly ones written by Stephen Moffat, including "Time Crash," "Time and Space," "Blink," and even "Doctor Who and the Curse of Fatal Death" Dodger (talk) 21:56, 13 May 2011 (UTC)[reply]

Indeed, one of the major themes of Doctor Who is the Blinovitch Limitation Effect (discussed on another wikipedia page).Kuitan (talk) 16:14, 19 September 2011 (UTC)[reply]

Closed timelike curves make quantum and classical computing equivalent[edit]

At some point the results from Aaronson and Watrous' refereed paper should be considered for the Time loop logic section. It extended Deutsch's result currently on the page and established that quantum computers are no more powerful than classical computers if CTCs are included in the model for computation (both can efficiently solve PSPACE problems). One potential link is here.--Adam Wolbach (talk) 04:25, 18 June 2009 (UTC)[reply]

Entropy[edit]

I find it strange that the article doesn't mention the problem of entropy, since that would seem to be a major issue with the Novikov self-consistency principle. If the billiard ball collides with its earlier version, then that must have some physical effect on the atomic makeup of the earlier ball, even if very slight. If you collided a billiard ball with another billiard ball a thousand times, would it crack? How about a billion times? How about an infinite amount of times? We can say positively that anything that experiences physical stress an infinite amount of times would be damaged at some point. Since this billiard ball is travelling back in time and colliding with itself an infinite (literally) number of times, the ball will at some "point" crack. Actually for a crack to occur after x amount of collisions, each individual collision must have some effect. For each collision the ball will be slightly more damaged internally. Therefore it's not the exact same ball that travels back for each iteration. Since the point of the principle is that it applies to a one-timeline model, there is actually only one iteration, a closed loop, so the billiard ball has be physically the exact 100% copy of the ball it's hitting. Since that is clearly not the case, the principle fails after just one iteration. 193.91.181.142 (talk) 10:50, 20 June 2009 (UTC) (Nick)[reply]

Damage to the balls, and any other entropy increase, is not inevitable, just overwhelmingly probable, on statistical grounds. Self-consistency requirements can force the entropy increase to be zero. In essence, this is because the number of ways of having an undamaged ball is vastly increased, so much so that it becomes overwhelmingly probable that a randomly chosen microstate will correspond to an undamaged ball.. This should really go on the page, but I don't have a cite immediately to hand. —Preceding unsigned comment added by 91.107.199.68 (talk) 07:24, 20 August 2009 (UTC)[reply]

You're both wrong. Entropy poses no problem to the principle, as there is only a single collision, and the two balls that collide are not required to be atom-for-atom identical. To see this, imagine that each time a ball is involved in a collision, it acquires one new crack. The sequence of events is then:

  1. A ball with zero cracks is moving towards the wormhole entrance.
  2. A ball with one crack appears from the wormhole exit.
  3. The two balls collide. Each one acquires a new crack.
  4. The ball with one crack (which had none a moment ago) disappears into the wormhole entrance.
  5. The ball with two cracks (which had one a moment ago) continues on its trajectory into the universe. It does not re-enter the wormhole.

71.178.183.147 (talk) 13:00, 21 October 2009 (UTC)[reply]

Think of elasticity and momentum instead of cracks. The ball is not perfectly elastic, as nothing is perfectly elastic. In it's collision with itself, some momentum is lost between the two versions of itself, so the ball enters the wormhole with a lower velocity than it left the other end with (which can't happen since there's only one universe in this model). This assumes that the ball was at rest before being hit by it's future self though. If it's already moving, then the momentum transferred by collision with it's future self is added to whatever momentum it already had, which means it's new momentum might be just right to enter the wormhole and leave at the appropriate velocity. In fact, it will be just right because it's already determined that the ball entered the wormhole (since the future ball appeared and the collision occurred). LlubNek (talk) 11:57, 3 January 2011 (UTC)[reply]
Right. The ball only experiences two collisions ever. Assume it's been struck with a cue to be travelling at 2mph. Upon being struck from the side to divert it, some energy is gained from the colission, and some is lost from the diversion, and now it is travelling at 1.5mph. It travels through the wormhole, come back out and strikes it's earlier self, transferring some energy and losing some, dropping to 1mph. It then bounces off th side and comes to rest. From the point of view of the ball, it merely gets struck by another ball out of nowhere, then it strikes a ball. Nothing else happens to it. Now, if the ball were at rest, and was struck by another ball with sufficient force to send it through the wormhole, and came out and struck its earlier self with sufficient force to send it through the wormhole and still come back out and strike itself, then there would then be a problem, But that's not this scenario. Dodger (talk) 22:09, 13 May 2011 (UTC)[reply]
Even that is not a problem: the last thing that happens before the collision might be that it rolls down a ramp, picking up speed. 67.198.37.16 (talk) 15:46, 21 March 2016 (UTC)[reply]

Existence of God[edit]

I suspect that we can deduce existence of God (defined as a function solving any mathematical problems in physical reality) from existence of time machine.

See this my Usenet message for details.

I however (yet) not found any way to prove this conjecture. VictorPorton (talk) 00:45, 18 February 2010 (UTC)[reply]

I think you are suggesting that the fact that the universe always stops a time traveler from creating a paradox (like killing his parents before he was conceived) requires some kind of "intelligence" on the part of the universe. If so, see this post of mine where I discuss a thought-experiment showing that a purely algorithmic process can produce self-consistent histories of the type supposed by Novikov's principle. Hypnosifl (talk) 05:45, 9 March 2010 (UTC)[reply]
The whole idea is rather pointless, as we have no proof of the Novikov self-consistency principle itself. Therefore attempting to deduce the existence of non-existence of an arbitrarily defined and conceptually vague thing such as deities based on a hypothesis with no foundation in anything except that it can be used as an argument to solve a problem that may or may not exist seems a bit absurd. Too many "if"s in a row. Dodger (talk) 22:00, 13 May 2011 (UTC)[reply]

Butterfly effect[edit]

There was this example in the article about a time traveler saving passengers of Titanic and replacing them with fake corpses, but I think it would violate the self-consistency principle.

The time traveler and his/her actions would cause a butterfly effect that would affect to nearly everything in the Earth and it's future light cone (more entropy) and therefore his/her actions would affect to the time traveler himself/herself.

So this example would actually break the self-consistency principle (if we accept the butterfly effect).

If the time traveler would be outside the Titanic's future light cone the operation would be possible, but the time traveler would know nothing about the incident.


(Sorry if there is any flaws in my English.)

--MrZalli (talk) 11:45, 3 May 2012 (UTC)[reply]

The butterfly effect need not produce an inconsistency. The future self is allowed to influence the past self in this theory. (The precise state of the future self travelling back has "already" been influenced by these effects.) But I agree that the Titanic example is rather confusing than enlightening, apart from being quite silly. --83.228.196.179 (talk) 20:54, 9 May 2012 (UTC)[reply]
"... need not ..." but can, right? E.g. in the "Fire time travel", the time traveler must not inform their younger self that they, themselves, caused the fire. Because then, they wouldn't travel back any more. Another example is the movie "12 Monkeys": The psychiatrist knows who spread the virus at the end of the movie, so she would instantly tell everybody and their dog about it, which would cause everything to change and cause inconsistencies (most of all that the scientists would not send somebody back to find out who spreads the virus). --Felix Tritschler (talk) 15:46, 4 April 2021 (UTC)[reply]

Time-loop logic and P/NP[edit]

I am confused by the sentence "Physicist David Deutsch showed in 1991 that this model of computation could solve NP problems in polynomial time [...]." Is this not completely obivous? You receive an answer, then you check it in polynomial time, if it was correct you send it back, if not you send back something different (which is an impossible outcome because of self-consistency). Or am I making a mistake here? --80.219.253.238 (talk) 22:37, 7 May 2012 (UTC)[reply]

(1) Unless someone published it, we cannot use it, as it would be original research.
(2) Your argument only applies if the problem always has a solution, such as (modified) factorization; given a number n, you send back the prime factorization of n (assuming "PRIME is P" is correct).
(3) Or, perhaps I should send the correct answer back in time? In any case, (1) is sufficient for a source being required. — Arthur Rubin (talk) 07:40, 29 May 2012 (UTC)[reply]
Well, but t does illustrate a problem, either with this article, or with Deutsch's work. Up above is a discussion of "uncaused effects" which asks three or four questions: can a single atom travel and be caught in a CTC? (if so, "where did it com from?") Can a quantum field excitation of any kind travel in a CTC? A qubit? Can a molecule? Molecules can encode classical information. so "where did that information come from?" The Deutsch thing is "can information travel in a CTC", and if so "where did it come from?" The point seems to be "since no one created that information that is traveling on the CTC, then it always there all along", and "if it was always there all along", then how do we know that this information records the solution to some particular problem, as opposed to the solution to some other unrelated problem? What is the meaning, the semantics, of the (classical, non-quantum) bits caught in the CTC, and how can we verify that they mean what Deutsch claims .. i.e.. that they are the solution to some particular NP problem? 67.198.37.16 (talk) 15:58, 21 March 2016 (UTC)[reply]

Yet Another Flaw[edit]

Suppose you have essentially complete knowledge of how something happened, and thus you went back in time to affect it. That is to say, all the examples given only prove that no one seems to know anything about the past for certain (for example, going back in time to meet Jesus only to end up being Jesus himself, etc). As such, if you KNOW FOR A FACT that something happened in a very specific way--like no one dropped trousers and stuck their bare-butt into JFK's face just before he got shot--then there's no way reality could end up being the same... there's no way history can continue on "unaltered".

But here inlies the rub: if the ONLY way for history to actually be altered is due to just KNOWING that it will be altered--as in, one's ignorance is the only thing that kept history from being altered in the stated examples--then I think this whole principle falls apart; the human mind alone is not that powerful as to be able to alter time itself. -- 71.141.121.26 (talk) 02:36, 14 June 2012 (UTC)[reply]

Ignorance isn't what stops you from changing the past--the idea is that there is no way for history to be altered whatsoever, it's deemed completely impossible by the principle. If you know every detail about how things went at a certain place and time in the past, and you had the idea of using your time machine to change things, then something would prevent you--perhaps you'd just change your mind, or perhaps you'd get a heart attack before you could turn on the time machine. If you imagine a scenario where you are able to change the past for whatever reason, then you are imagining a universe where the Novikov self-consistency principle doesn't apply. Hypnosifl (talk) 04:39, 19 June 2012 (UTC)[reply]

Another thing to keep in mind is that your memory is an unreliable record of the past. If you use known facts about the past (eye witness being the most reliable) to go back and take an action that clearly contradicts that record (assuming circumstances/the universe don't prevent it), then the NSCP would imply that your record of the past is wrong. Given that record is your brain then you should probably be worried. Hallucinations don't tend to come out of no where, so you might want to get checked for tumours. 155.63.200.55 (talk) 02:36, 30 October 2020 (UTC)[reply]

Request clarification[edit]

"then some event will prevent the execution of step 3 that receives the value F from the future. "

Outside of closed time loops, under NSCP, there is no way to affect the past from the future. Prior to any incoming time travel (the future affecting the past), the universe is perfectly consistent and in most normal computer architectures, there shouldn't be a way for an insoluble problem to cause a computer failure or similar improbable event.

Since NSCP is an assertion of "no new physics", just because we are about to feed Time Loop Logic an insoluble problem does not affect at all the probability of computer failure/spontaneous violations of thermodynamics 2nd law/events of extreme improbability (eg. freak cosmic rays). Prior to any time travel, the computer executing the algorithm on the given problem is not a special computer and normal computers do not spontaneously combust if given an insoluble problem.

Therefore, it seems obvious to me that the only thing that can prevent of execution of the algorithm will be an incoming time travel event that must get itself sent. Clearly, we are not about to engineer a computer system that will refuse to execute our algorithms, so it will be overwhelmingly likely to execute and result in paradox if something doesn't stop it and the only thing that can stop it is the time travelling event.

If need be, this event may indeed cause spontaneous violations of the 2nd law of thermodynamics or other exceedingly improbable events simply by being arranged "just so". But outside any time travel event with "just so" arrangements (which is allowed and even likely under Time Loop Logic), exceedingly improbable events remain just that. — Preceding unsigned comment added by 175.156.215.144 (talk) 16:41, 5 November 2012 (UTC)[reply]


I think you over-complicated the problem. The principle simply says that nature collapses locally only to globally consistent (no paradox) scenarios. In the case you're referring to ("N is prime, no satisfying factors F") the only globally consistent scenario is the one with an empty channel. I suppose, that what you described is a slightly different, however equivalent, realization of the machine: one could reprogram the machine and define an "error" symbol, which is never selected during normal execution, but always resend back in time. However, in this realization, the programmer would also have to define what happens, if no transmission is received, thus this is redundant. lessismore (talk) 19:36, 27 November 2012 (UTC)[reply]
The programmer can get the machine to generate a paradox when no transmission is received though (just send anything back). I was mostly referring to the claims that "something" will prevent the paradox. That "something" must also involve a time travel event in the cases when no time travel leads to something being sent. eg. if (received_from_future == nil) then send('1') And when designing such a machine, we do want to get an answer, so the 'nil' solution must be ruled out in the design anyway. — Preceding unsigned comment added by 115.132.121.26 (talk) 08:05, 28 November 2012 (UTC)[reply]

Infinite Improbability Drive?[edit]

If the Novikov self-consistency principle was valid, and using time loop logic, could an "infinite improbability drive" (apologies to Douglas Adams) be created? If so, would this not disprove the principle? Msiminski (talk) 22:57, 24 January 2013 (UTC)[reply]

A sentence from the article[edit]

The Novikov Principle does not allow a time traveler to change the past in any way at all, but it does allow them to affect past events in a way that produces no inconsistencies—for example, a time traveler could rescue people from a disaster, and replace them with realistic corpses if history recorded that bodies of victims had been found.

What it has to do with physics? Physics should not be concerned whether history recorded the corpses or not; it should not be concerned with collective knowledge of humanity. And I think I won't be far off if I say it should not be concerned with individual people's knowledge or opinions, either. - 89.110.17.174 (talk) 19:49, 4 August 2013 (UTC)[reply]

I agree that the Novikov principle is nothing to do with physics. It is just a conjecture. The reference to a presentist view, above, makes the principle entirely redundant. for a good explanation see http://essaydensushing.blogspot.co.uk/2013/10/bye-bye-dr-who.html 62.232.115.50 (talk) 05:52, 11 October 2013 (UTC)EDS[reply]

KISS[edit]

Isnt the simplest solution simply to have a generalized property that BLOCKS All Timetravel ???? Because if you allow some mechanism which blocks the formation of local individual paradoxes (article talks about either inherent modifications to the results of backward traveling actions (mutating the actions) or just not allowing those specific backward traversals of actions which WOULD create a paradox - and you must now account for subsequent/additional backward (time travel) effects which could put the original paradox back on course (or a whole infinite sequence of attempted corrections -- all materializing before/after/simultaneously with the first (paradox potentialed) time travel event).

The simplest solution by far, is to just not have ANY Timetravel be possible - a general blockage that now has no worries about infinite complications (and possible breaking of who knows how many actually real physical laws).

Was this not mostly anticipated by SF author James Hogan?[edit]

in his early 1980's book "Thrice upon a Time?" — Preceding unsigned comment added by 71.190.176.128 (talk) 01:45, 14 June 2015 (UTC)[reply]

What was anticipated? That there's a non-renormalizeable quantum mechanical solution to time travel through closed-timelike-curves? Or was it just a time travel story? IsaacAA (talk) 07:11, 14 June 2015 (UTC)[reply]
An older version of the page mentioned a number of fictional examples that assume a fixed self-consistent timeline, but this section was deleted as original research. Still, if you look at the older version you can see many stories that used this idea predating James Hogan's. Hypnosifl (talk) 22:24, 9 August 2015 (UTC)[reply]
I don't think "Thrice Upon a Time" meets the Principle. Each timeline has messages from other timelines, which presumably no longer "exist". — Arthur Rubin (talk) 03:57, 10 August 2015 (UTC)[reply]

Russian or soviet?[edit]

As this conjecture was written in the mid 80's, it's not better to reference Novikov as soviet and not russian? --Canopus49 - Replies here 00:59, 4 August 2015 (UTC)[reply]

Bill & Ted's Excellent Adventure: consistent or no?[edit]

I recently added "Bill & Ted's Excellent Adventure" to the "See Also" section of this page, and user: IsaacAA removed it, commenting: "Bill & Ted has 'timelines' that don't follow the single-self-consistent-history idea. Movie has nothing to do with the principle." But the movie is paradox-free, and there is no explicit mention of alternate timelines as far as I recall.

And elsewhere on this talk page, in the "Potential implications for free will" section, it is remarked: "Well, how about Bill & Ted's Excellent Adventure? Both in the scene where they met their future selves (which was shown twice, from both points of view) and in the scene where they broke various historical figures out of jail by using helpful items (like the dad's keys, which had been missing since the start of the movie) that they planned to go back and plant in the right spots later using the time machine, the movie was actually a pretty good illustration of the idea of a single unchangeable timeline which already incorporated the actions of the time travelers." Hypnosifl (talk) 00:10, 28 March 2008 (UTC)

Can anyone point to a sequence of events in the movie that would violate the Novikov self-consistency principle on its face (without reference to any hypothetical paradoxes that may have arisen from kidnapping a variety of historical figures and later presumably depositing them back in their own time), or is it logical to conclude that the movie is a most excellent portrayal of a self-contained, self-consistent, Cauchy Horizon-bounded universe which is ultimately harmonized across time and space by the righteous majesty of Wyld Stallyns music? — Preceding unsigned comment added by Zwitterbiel (talkcontribs) 00:08, 7 August 2015 (UTC)[reply]

Instead of arguing about the self-consistency of Bill & Ted, let's simply agree that the see-also section is for similar topics, not for examples. IsaacAA (talk) 00:19, 7 August 2015 (UTC)[reply]

Fair enough, but what I'm hearing now is that the Novikov self-consistency principle page itself is in dire need of an entire section on this most radical and historical adventure which is faithful to the laws of physics as this Russian time-traveling dude understood them. — Preceding unsigned comment added by Zwitterbiel (talkcontribs) 00:38, 7 August 2015 (UTC)[reply]

Experimental verification using CTC generated by thunderstorms[edit]

Hi, I might have experimental evidence that time travel also explains ball lightning. Recall some of the early conjecture by Feynman that suggests antimatter is matter traveling back in time. As thunderstorms seem to generate antimatter in quantity, this could be responsible as radio signals *might* be able to pass through a BL and back in time by hours or even days provided they are within range. Its also possible that if my hypothesis is correct it can be tested experimentally, if the spatial and temporal coordinates of an event is known then a signal can be sent back through time on purpose.

I also had some limited success (unable to replicate) generating a CTC on a small scale; the effect was strong enough to knock over a radio inside a sealed cupboard and other researchers suggest that larger scale effects could account for reports of strange activity near certain geographical areas. I'd be happy to share the experimental configuration but not sure if publishing it in an open journal is such a good idea for any number of reasons. As the person sent part of the experimental setup hasn't been seen since! — Preceding unsigned comment added by 185.3.100.3 (talk) 06:45, 6 September 2018 (UTC)[reply]


Violation of conservation of energy/thermodynamics flaw[edit]

This article does not explain how the NSCP tackles the law of conservation of energy / thermodynamics / entropy of the universe. Self consistency can not over ride these. If I have a machine that can send a message to the past and I use it to send a message to myself of 10 minutes ago. I suddenly receive a message from my future self and print it out. 10 minutes pass and the moment arrives where I should send the message to myself of 10 minutes ago (the message I had received). But I suddenly refuse to send the message. I do not send it. Yet I have the printout of the message my future self had sent back in my hands.

If I don't send it the NSCP suggest that due to self consistency that message will be sent of its own accord. This violates conservation of energy / entropy of the universe. 85.148.213.144 (talk) 21:31, 29 September 2021 (UTC)[reply]

Picture doesn't match description in article[edit]

The picture of Echeverria and Klinkhammer's resolution of Polchinski's billiard ball paradox, looks like simple moving the 'entrance' of the wormhole to the left so that it can 'catch' the ball whose path was altered by its past self.

But in the text it seems that the wormhole just stays where it always was and the reason that the billiard ball can still enter it despite being hit by its future self is that when the billiard ball exits the wormhole it does so under a different angle than in the first picture, allowing it to touch its younger self in a more gentle manner that does not prevent it from entering the wormhole, but just makes it enter the wormhole under a different angle, which then in turn explains why it had exited the wormhole under a different angle in the first place.

Which description of E & K's resolution (the one in the pictures or the one in the text) is the correct one? I hope the one in the text and not the one in the picture, because that would be much more interesting. But if so, could some one perhaps edit the picture to match the true solution? Octonion (talk) 22:15, 15 February 2023 (UTC)[reply]

Correct Interpretation of the Novikov self-consistency principle[edit]

The Novikov self-consistency principle offers a concise interpretation of the so-called paradoxes related to Closed timelike curve's, Problem of time, and Hugh Everett's many-worlds interpretation (MWI), Multiverse. The clever idea of Igor Dmitriyevich Novikov is the realization that the pseudo-contradictory logic of these paradoxes can be eliminated by incorporating event observability likelihoods, i.e., chance or probabilities. In a nutshell, we can certainly postulate the theoretical existence of paradoxical events, however the intrinsic event-inconsistencies (the paradox itself) suggest that they are measure-theoretically trivial. In other words, an event causing a paradox is almost surely of zero probability. Here are two examples of this idea:

  • Consider throwing a dart at a target that is square and measuring just the x-position of the dart landing point, . Of course, we know and love the rational number's - they are easy to do arithmetic with, they are dense in the reals, they have infinite (countable) cardinality, and given a certain level of precision, rational numbers can approximate any real number. Well, with all these perfect characteristics, relative to the richness of the real numbers, rational numbers are so few that they are practically negligible. The event is not observable, i.e., it's measure theoretically trivial, , its almost surely not going to be observed as an outcome of this dart experiment, no matter how many times we throw the dart at the square board. This of course assumes that our dartboard has infinite measuring precision, instead of a finite number of grid cells the way real dart disks are built precisely to discretize the outcomes. The relevance of this example to Novikov self-consistency principle, closed-time-curves, and all types of grandfather paradoxes should now be obvious: Realistic traveling back-and-forth in time, requires one to lend twice at one fixed point in spacetime (specific event). The likelihood of this event is zero, it's not going to be realistically observed, just like with the dart throwing experiment, where the probability of landing twice at the same x-position is almost-surely trivial.
  • Instead of considering time-travel, consider the intuitively simpler space-travel. Can you visit the same spatial location twice? The answer to this is - "you can't!" The event of visiting the (exact!) same spatial location in the universe multiple times is an event of measure zero. Of course, this does not contradict out consistent observations that we can go each day to work and at the end of the day find our way back home! Revisiting the "same spatial location" refers to the same location in Minkowski spacetime, not the earth surface or even the solar system. Multiple time dimensions and its simplest example of complex time (kime) provide another way to look at spacetime paths, including closed time trajectories. Spacekime ties together the longitudinal event-ordering (indexed by the complex time magnitude) and sampling via repeated random experiments (indexed by the complex time phase). This directly reflects the multiverse interpretation of the physical world and the Wheeler–DeWitt equation. In essence, each experiment, whether a real physical trial or a philosophical cognitive exercise (think of the Schrödinger's cat thought experiment), intrinsically generates stochastic outcomes. Some experiments have larger and some have smaller variability, but instantiating any observation fragments the 4D universe according to the specific observed outcomes, collapsing the universal wave function into one specific instance. Iwaterpolo (talk) 22:10, 3 July 2023 (UTC)[reply]

The Diagrams Are Wrong[edit]

The diagram captioned "Echeverria and Klinkhammer's resolution" incorrectly shows the wormhole as being moved.

It should instead show the angle of the ball as being changed 2600:1002:B033:C99:6304:2793:9226:625B (talk) 15:24, 30 September 2023 (UTC)[reply]