Well, t=10 is probably a time at which the universe was still quark soup, so you’re rather unlikely to show up as a conscious being then.
As for appearing at t=N vs t=N+1, the explanation I remember hearing has two parts: Firstly, if you check out your physical state at any given time, it will necessarily not contain memories of the future; memory formation generates a lot of entropy, and time tends to progress from low-entropy to high-entropy states. So regardless of what time you find yourself at, you’ll remember the past but not the future. Secondly, if we treat your personal sampling of time as random versus in a linear fashion, since at each point in your personal experience your memory satisfies this property you’ll perceive time as flowing linearly (or at least in a forwards fashion) at any point.
So basically, as near as I can tell, there’s no good way to tell the difference between any type of sampling from the time distribution; your subjective experience would be very similar in any case. You find yourself at t=N because you have to find yourself at some t, and t=N is about as likely for you to sample as any other.
This seems to repeat the confusion in the original post. In what sense are “you” sampling different times? What is this “you” doing the sampling? Is there some sort of disembodied consciousness flitting randomly from time slice to time slice?
This seems to repeat the confusion in the original post. In what sense are “you” sampling different times?
There is a difference between the standard “timeless physics” of the 4D block universe, and the specific kind of “timeless physics” advocated by Julian Barbour, which Eliezer was referencing in that post. Your explanation was a good reply within the context of the 4D block universe. And there, there is no issue of “sampling”.
However, under Barbour’s account, what exists is bigger than a single 4D block universe. What exists is a configuration space (called “Platonia”) in which each point represents a possible state for a 3D spatial universe. Conversely, every such possible state is represented by a point in Platonia. In addition, this configuration space supports a static complex scalar field (something like a stationary state solution to the Schrödinger equation in quantum mechanics). Using the Born rule, this scalar field can be interpreted as a fixed probability distribution assigning high probability to some regions in Platonia and low probability to others. Barbour does appeal to this probability distribution to explain why we never “find ourselves” in highly “improbable” configurations of the macroscopic universe.
For example, Platonia contains a point (i.e., a 3D universe) containing people just like us, except that they are looking up into the sky and seeing two suns where their memories say that there was only one sun moments before. That is, they are witnessing what appears to be a flagrant violation of the laws of physics. Barbour’s explanation for why we never have this experience is that such configurations get practically no probability mass from the complex scalar field.
So, there is a sort of sampling going on in Barbour’s account. He would admit, I think, that this “sampling” is just as mysterious as the Born rule is in the usual many-worlds interpretation of quantum mechanics.
I was referring to the set of all experiences that identify as being XiXiDu as “you” for simplicity’s sake; the sampling is the selection of a particular timeslice to experience (ie, XiXiDu was presumably experiencing t=N when he wrote this).
Maybe it would make more sense to frame this differently; the laws of physics dictate that XiXiDu will experience conscious thought at times t=a, a+1...b (assuming consciousness is non-magical and is a result of physics), so those timeslices contain conscious experiences by an entity self-identifying as XiXiDu.
As I understand it, timeless physics predicts that they will all experience simultaneously (so to speak), so there will be b-a instances of XiXiDu running at the same time, giving a 1/(b-a) chance that a given instance will be experiencing t=N.
Well, t=10 is probably a time at which the universe was still quark soup, so you’re rather unlikely to show up as a conscious being then.
As for appearing at t=N vs t=N+1, the explanation I remember hearing has two parts: Firstly, if you check out your physical state at any given time, it will necessarily not contain memories of the future; memory formation generates a lot of entropy, and time tends to progress from low-entropy to high-entropy states. So regardless of what time you find yourself at, you’ll remember the past but not the future. Secondly, if we treat your personal sampling of time as random versus in a linear fashion, since at each point in your personal experience your memory satisfies this property you’ll perceive time as flowing linearly (or at least in a forwards fashion) at any point.
So basically, as near as I can tell, there’s no good way to tell the difference between any type of sampling from the time distribution; your subjective experience would be very similar in any case. You find yourself at t=N because you have to find yourself at some t, and t=N is about as likely for you to sample as any other.
This seems to repeat the confusion in the original post. In what sense are “you” sampling different times? What is this “you” doing the sampling? Is there some sort of disembodied consciousness flitting randomly from time slice to time slice?
There is a difference between the standard “timeless physics” of the 4D block universe, and the specific kind of “timeless physics” advocated by Julian Barbour, which Eliezer was referencing in that post. Your explanation was a good reply within the context of the 4D block universe. And there, there is no issue of “sampling”.
However, under Barbour’s account, what exists is bigger than a single 4D block universe. What exists is a configuration space (called “Platonia”) in which each point represents a possible state for a 3D spatial universe. Conversely, every such possible state is represented by a point in Platonia. In addition, this configuration space supports a static complex scalar field (something like a stationary state solution to the Schrödinger equation in quantum mechanics). Using the Born rule, this scalar field can be interpreted as a fixed probability distribution assigning high probability to some regions in Platonia and low probability to others. Barbour does appeal to this probability distribution to explain why we never “find ourselves” in highly “improbable” configurations of the macroscopic universe.
For example, Platonia contains a point (i.e., a 3D universe) containing people just like us, except that they are looking up into the sky and seeing two suns where their memories say that there was only one sun moments before. That is, they are witnessing what appears to be a flagrant violation of the laws of physics. Barbour’s explanation for why we never have this experience is that such configurations get practically no probability mass from the complex scalar field.
So, there is a sort of sampling going on in Barbour’s account. He would admit, I think, that this “sampling” is just as mysterious as the Born rule is in the usual many-worlds interpretation of quantum mechanics.
I was referring to the set of all experiences that identify as being XiXiDu as “you” for simplicity’s sake; the sampling is the selection of a particular timeslice to experience (ie, XiXiDu was presumably experiencing t=N when he wrote this).
Maybe it would make more sense to frame this differently; the laws of physics dictate that XiXiDu will experience conscious thought at times t=a, a+1...b (assuming consciousness is non-magical and is a result of physics), so those timeslices contain conscious experiences by an entity self-identifying as XiXiDu. As I understand it, timeless physics predicts that they will all experience simultaneously (so to speak), so there will be b-a instances of XiXiDu running at the same time, giving a 1/(b-a) chance that a given instance will be experiencing t=N.