Talk:Axiom of Causality

Reversibility
It would be nice to see something in this article about the problem of the time reversibility of physics versus the time asymmetry of causation.

All known laws of physics are time reversible (or more strictly they have charge-particle-time symmetry - see Stephen Hawking's a Brief History of Time), meaning that reversing the velocities of every particle in a given system would result in a time reversed version of the system. If a baseball breaks a window, and we then reversed the velocities of all the particles in the window and the ball, we would see the window miraculously put itself back together and the ball fly back in the opposite direction.

The point of this is that it makes it much less obvious how causes can be said to be always in the past and never in the future. Imagine a collision taking place on a frictionless pool table (an analogy for the reversible motion of microscopic particles) after time t1 and before time t2. Our causal intuition says that we cannot change the position of the balls at time t2 without changing their positions or velocities at time t1, which is correct, but we also cannot change their positions at time t1 without also changing something at time t2. If the laws of physics are symmetrical in time then where does the temporal asymmetry of causation come from? From these considerations it appears to be something that must be carefully explained. The intuition that it is so self-evident that it can be taken as an axiom seems at odds with this.

This is an important issue which is often missed in both physics and philosophy. For one attempt at a resolution and a good review of the issue see Time's Arrow and Archimedies' Point by Huw Price.

80.47.211.87 14:33, 29 June 2007 (UTC) Time cannot be reversed, because it is an abstract concept and always flows forward by definition. We can make time appear to reverse by inverting the course of events such as in the ball and window example given above, such that it looks like we are "rewinding" time, however the time at which the reversal is completed will still be some point in the future after the reversal was begun. This can be proved by measuring the time it takes for the window to reform and the ball to go back to the person who threw it - this period of time will be some positive amount of seconds or minutes. It would certainly not be a negative amount of time, and therefore some positive amount of time has elapsed, taking us into the future.

Only a philosopher would ever analyze something so obviously straightforward! —Preceding unsigned comment added by Gcsnelgar (talk • contribs) 12:39, 4 December 2007 (UTC)


 * No, actually this is far from straightforward. It appears that nearly all physical processes are time reversable (i.e., have T symmetry.  A few incredibly rare processes such as certain kaon decays are NOT T symmetrical, but neither are they CP symmetrical, which is just as weird; however, they maintain CPT symmetry at least.)  You can look at a past event as the cause of a future event, or, oddly, as a future event as a cause of a past event, and both will make sense physically.  The Second Law of Thermodynamics appears to be the only law that gives a strong arrow of time, and why it does so is very unclear.


 * Gcsnelgar, time is not an abstract concept, and nor is it defined to flow forwards. Time is a physical dimension, and its definition is such.  Time DOES NOT FLOW, and indeed all moments in time are equally valid at once, in a sense.  You say that negative values of time make no sense, but in fact ALL values of time are entered as negative, and time is therefore sometimes referred to as a negative dimension.  Positive values of time could be used, too, which would simply be described as looking into the past rather than the future.  It also makes no sense to define time as pointing "forwards," but rather defining "forwards" (i.e., the future) as the direction to which time points.  This in no way explains why time points in any direction (although it clearly does).


 * Most people believe that The Second Law of Thermodynamics is simply definable based on the Axiom of Causality, so it is this axiom which ultimately decides the arrow of time. Essentially, the fact that time has an arrow at all entirely comes down to THIS AXIOM.  It is an incredibly important point which definitely should be mentioned here.


 * Other arrows of time might include the "cosmological arrow," by which the universe is always expanding forwards in time by definition, so, if you can imagine, when the universe begins to contract it will move backwards in time. In some theories this explains the Second Law of Thermodynamics, but not in most, so it is not commonly used.


 * And finally, I think it's funny that you say "only a philosopher would ever analyse anything so obviously straightforward!" when this is, in this case, primarily a physical question, rather than a metaphysical or "philosophical" one.Eebster the Great (talk) 23:55, 6 April 2008 (UTC)


 * Christ, get a life! — Preceding unsigned comment added by 85.240.50.25 (talk) 10:50, 27 February 2013 (UTC)

"All effects have causes" does not imply "All causes have only one possible effect"
"The Axiom of Causality is the proposition that everything in the universe has a cause and is thus an effect of that cause." [...] "If all causes have effects, and all effects logically follow the rules of the universe, then all events follow a theoretically predictable pattern, thus all future events have already been determined by past events."

There is a logical error here. If all effects have causes, that does not imply determinism (in any form). Determinism requires that there is only one possible effect for a given set of causes, which is not the same thing at all.

For example, a distributed message passing system follows the axiom of causality (at least after its initial state is bootstrapped): if a message is received, then it must have been sent; if a process is created then another process must have requested its creation, and so on. But the order in which messages are received and processes are created can be nondeterministic, without any contradiction. David-Sarah Hopwood (talk) 16:39, 6 February 2009 (UTC)

Reply: You are right that determinism does not follow only from the fact that all effects have causes. But when you consider that these effects follow a predictable* pattern, it follows that all events in the universe are determined by the universe's initial state.


 * here it is not required that a human can actually predict the pattern. If we had the right technology, we could predict the pattern. The fact that this technology does not yet exist does not change the fact that the events follow such a pattern. —Preceding unsigned comment added by 203.211.89.1 (talk) 17:58, 21 June 2009 (UTC)

Retrocausality
Shouldn't retrocausality be mentioned somewhere? — Preceding unsigned comment added by 2.227.66.211 (talk) 15:15, 13 January 2012 (UTC)