Constants & Variables

March 3, 2010 6:22 PM

Caltech physicist Sean Carroll recently tweeted that he was meeting up with Lost producers Damon Lindelof and Carlton Cuse. This was posted to the forums at Lostpedia, prompting immediate spoiler complaints ... so Carroll signs up and drops in to the thread to clear up the confusion, also offering some of his thoughts on the use of time travel in the show and referencing a longer blog post he wrote shortly before the start of the final season.
posted by mannequito (75 comments total)

The most recent episode had some really bad writing.

"HI SAYID WE WERE JUST LOOKING THROUGH YOUR BAG BECAUSE YOU KNOW THAT'S WHAT US LITTLE KIDS DO YOU SEE AND WE FOUND THIS VERY CONSPICUOUS PHOTOGRAPH YOU CARRY AT ALL TIMES OF YOUR BROTHER'S WIFE AND WE'RE SHARING IT WITH EVERYBODY VERY LOUDLY BECAUSE WE THINK IT'D REALLY MOVE THE PLOT FORWARD, OK?"
posted by defenestration at 6:37 PM on March 3, 2010 [12 favorites]

He does say that his visit wasn't related to the show, though.

From his blog post:

Traveling near the speed of light, or lingering in a powerful gravitational field, you will “move into the future faster” than someone floating freely in empty space.

I don't like it when time travel is explained in this way. I don't like it because it confused me for a long time. It makes it sound as if velocity is absolute. Relative to some frame of reference, we're all traveling near the speed of light. But this does not mean that "moving into the future faster" is relative to a frame of reference. Cases like the twins paradox makes this clear... the two spacetime points at which the twins' lives intersect are perfectly objective and frame-independent, as is the fact that one twin aged more than the other.
posted by painquale at 6:43 PM on March 3, 2010 [2 favorites]

Although me criticizing Sean Carroll's explanation of time travel is probably the height of stupidity and will lead to a smackdown.
posted by painquale at 6:44 PM on March 3, 2010

The most recent episode had some really bad writing.

"GO! LEAVE THIS PLACE! NEVER COME BACK!"

And yet, they still give Jeff Fahey at least one unbelievably awesome line every episode.


(okay, his lines aren't always great, but he always MAKES them great.)
posted by SmileyChewtrain at 6:46 PM on March 3, 2010

defenestration: “The most recent episode had some really bad writing.”

Almost as bad as the past six seasons.
posted by koeselitz at 6:49 PM on March 3, 2010 [3 favorites]

Almost as bad as the past six seasons.

So you think last night's episode was better written than every episode of the past *5 seasons? Does this even mean anything besides "Hey guys I don't like LOST!" ?
posted by Solon and Thanks at 7:02 PM on March 3, 2010 [9 favorites]

Solon and Thanks: “So you think last night's episode was better written than every episode of the past *5 seasons? Does this even mean anything besides "Hey guys I don't like LOST!" ?”

Sorry. Only slightly more, I guess. I just finished watching up through season 4 at the behest of my Lost-fanatic roommates, and I'm giving up. This thing just bores the trash out of me. I guess I've just gone way past saturation point.

And, well, yeah. I guess I can't speak to season 5, but 1-4 were about the same as last night's episode, to my ear. Really.
posted by koeselitz at 7:08 PM on March 3, 2010

I don't like it when time travel is explained in this way.

I pitty the fool that thinks there is a privileged point of rest!
posted by nola at 7:15 PM on March 3, 2010 [3 favorites]

Oh, then, fair enough. It's certainly not for everyone. I wasn't sure if it was one of those comments made by someone who watched 3 episodes.

Anyway, Sean Carroll's blog post is good. I'd be interested in hearing what he has to say about the alternate timeline (if that's what they are? I know there are theories otherwise) introduced this season.
posted by Solon and Thanks at 7:17 PM on March 3, 2010

By the way - in that blog post he links to Rules for Time Travelers. It's a great read, it eloquently explains things that are very difficult to explain. (The concept of meta-time is something that always really bothered me, but I had no way to talk about it.)
posted by Solon and Thanks at 7:22 PM on March 3, 2010 [1 favorite]

I don't like it when time travel is explained in this way.
--
I pitty the fool that thinks there is a privileged point of rest!

You guys have looked back at these primitive ideas and laughed. You just don't know it yet.
posted by The Deej at 7:24 PM on March 3, 2010 [5 favorites]

defenestration: "The most recent episode had some really bad writing."

koeselitz: Almost as bad as the past six seasons.

I've just started watching Lost on hulu, mostly to have something to talk about with my mom. I'll tell you, I wish I had $10 every time a conflict is resolved with a knock-out punch. One punch or a rifle stock to the head is the screenwriters' standard trick for moving the plot forward. While there are dozens of other weak areas with the two seasons that I've watched so far, these concussions are really starting to bother me. I keep thinking poor John and Jack, their brains must be mush and will be suffering from dementia by the fifth season. CAN THE SCREENWRITERS SAVE THE ISLAND FROM CHRONIC TRAUMATIC ENCEPHALOPATHY?! Really.

Anyway, Lostpedia is a little too dense and spoiler-rich for me. Can anyone point to any decent meta-analysis of Lost looking at the writers, producers, themes, and story structure and not so focused on the story minutia canon?
posted by peeedro at 7:25 PM on March 3, 2010 [6 favorites]

I only ever watched the first episode, but I wonder if someone who's followed the series could tell me if the duplicate star formations in the first seconds of the first night scene are ever seen again or explained in any way.
posted by Anything at 7:31 PM on March 3, 2010 [1 favorite]

The plot hits a singularity when the hatch imlodes and the story never really recovered.
posted by freshundz at 7:32 PM on March 3, 2010

"Lost" is stupid and unwatchable.
posted by BitterOldPunk at 7:33 PM on March 3, 2010 [7 favorites]

Can anyone point to any decent meta-analysis of Lost looking at the writers, producers, themes, and story structure and not so focused on the story minutia canon?

Yes. Alan Sepinwall writes quality episode-by-episode analyses, but he only started about halfway through season 2. You can read his thoughts/review/summary as you watch the episode, and you can even read some theories without spoilers in the 300+ comments his reivews garner.
Here is the link to the posts tagged as Lost (Season 2)
posted by milestogo at 7:37 PM on March 3, 2010 [1 favorite]

peeedro: “I've just started watching Lost on hulu, mostly to have something to talk about with my mom. I'll tell you, I wish I had $10 every time a conflict is resolved with a knock-out punch. One punch or a rifle stock to the head is the screenwriters' standard trick for moving the plot forward. While there are dozens of other weak areas with the two seasons that I've watched so far, these concussions are really starting to bother me. I keep thinking poor John and Jack, their brains must be mush and will be suffering from dementia by the fifth season. CAN THE SCREENWRITERS SAVE THE ISLAND FROM CHRONIC TRAUMATIC ENCEPHALOPATHY?! Really.”

What's interesting to me is that the show seems to have managed a method for the production of rarefied suspense. I'm not really a fan of J J Abrams' writing at all, though the plot points are at least mildly interesting; but what's most interesting is the entire ethos of the show, which seems truly intended to fit almost every suspenseful trope from every genre - not just science fiction and fantasy, but from all fiction in general. And that evocation of suspense becomes the theme of the show, as during every episode at least one something that can't possibly make sense is revealed. This was most obvious during the first season, I think (polar bear! DUN DUN DUN! other people on the island! DUN DUN DUN!) but it's really a general rule for every episode I've seen so far. This alternates somewhat, however, with a loose sort of mystical vibe which I think is intended to very briefly relieve the viewer before the plunge back into suspensefulness.

It's at least notable as fiction, anyway. And I can get being into it. I think watching a whole bunch of them all at once kind of ruined it for me; you have to have at least a bit of waiting in between for the suspense to kick in. Also, I've met a lot of people who've just gotten fed up with the endless suspense that seems to point nowhere, and I can understand where they're coming from; I got tired of that, as well.

Anyhow, sorry if this is a derail.
posted by koeselitz at 7:37 PM on March 3, 2010 [4 favorites]

Why comment in a thread if you're not going to contribute anything? Rather than thread shit if you don't like the show, you're free to talk about Sean Carroll, or time travel, or you know - move on.
posted by Solon and Thanks at 7:38 PM on March 3, 2010 [6 favorites]

Actually, my bad, those aren't real episode analyses yet. He only gets going in season 3.
posted by milestogo at 7:39 PM on March 3, 2010

Sorry, koeselitz, that was directed at BitterOldPunk, not you.
posted by Solon and Thanks at 7:39 PM on March 3, 2010

"Lost" is stupid and unwatchable.

Did you get lost on your way to Twitter?
posted by jessamyn at 7:55 PM on March 3, 2010 [43 favorites]

Can anyone point to any decent meta-analysis of Lost looking at the writers, producers, themes, and story structure and not so focused on the story minutia canon?

CAN I?

The House Next Door blog has been recapping each episode since the end of Season 2 - interesting takes on the story and themes, and spoiler free (this link takes you to the oldest set of entries):

http://www.slantmagazine.com/house/tag/lost/page/3/

The AV Club's Noel Murray has been recapping LOST since the beginning of Season 4. He's my favorite LOST blogger (deeply in love with the show, but clear-eyed about its flaws) and the discussion threads are the best LOST-talk on the internet (which you'd expect from that site, which has the best writing and discussion about any and all pop-culture):

http://www.avclub.com/tvclub/tvshow/lost,38/

For something a bit more out there (not Doc Jensen at Entertainment Weekly out there - just KIND of out there) - There was an English professor blogging about LOST from mid-season 3 ("Not In Portland") through last year (when his health problems intervened). He explores each episode by fleshing out references to various books, myths, etc. in light of the story and his entries were really fascinating. His earliest entries are here:

http://www.powells.com/blog/?author=104&paged=4
posted by moxiedoll at 8:16 PM on March 3, 2010 [10 favorites]

I hate the non-explanations throughout the show. Often explained by the old stand-by, "THERE'S NO TIME TO EXPLAIN!" And often, their oblique explanations don't make sense in retrospect.

Sample dialogue—

Juliet: Don't open that door, Jack!
Jack: Why shouldn't I...?
Juliet: Because if you do... we'll all die!
Jack: What does that mean? How will we die if I open this door?
Juliet: Just trust me, we will.
Jack: Seriously, why would we die if I opened this door right now? I'm giving you a chance to explain ONE THING clearly to me.
Juliet: There's no time, you just have to trust me!
Jack: I'm opening it...
Juliet: NO!

[Jack opens door, water floods in]

Jack: Um... why couldn't you have just said that we were underwater...?
Juliet: I thought it would be a more dramatic reveal. And I wanted to make a larger statement about trust and faith.
Jack: Fuck you.
posted by Joe Beese at 8:51 PM on March 3, 2010 [18 favorites]

Why comment in a thread if you're not going to contribute anything?

Seriously. Especially considering we just had a thread where all the people that don't watch Lost came in and be all "Lost is fucking stupid"? Do we really have to do that again?

And I don't understand the Gawker comment quote -- How is that sample dialogue? Is there a scene I'm not remembering? The only scenes in Lost I remember about people being underwater, they all knew it. Right? And if it's a metaphor, I simply don't get it. Please explain.
posted by inigo2 at 9:53 PM on March 3, 2010 [1 favorite]

It's that place where Jack was held at the beginning of season three whilst Sawyer and Kate were stuck in the bear cages because Ben wanted someone to operate on his cancer. The place where Juliet was in charge of bringing him sandwiches.
posted by tapeguy at 10:07 PM on March 3, 2010 [1 favorite]

inigo2: “And I don't understand the Gawker comment quote -- How is that sample dialogue? Is there a scene I'm not remembering? The only scenes in Lost I remember about people being underwater, they all knew it. Right? And if it's a metaphor, I simply don't get it. Please explain.”

Well, there are two things about the Gawker comment quote. First of all, yes, it's supposed to be lampooning stuff, and it's not actual dialogue from the show. Second, it's not supposed to be a mockery of the show itself; it's supposed to be a humorous presentation of the most annoying aspect of "The Others" as a group of people on the show: the fact that they seem to take a sort of pretentious delight in NEVER EXPLAINING ANYTHING THEY'RE DOING OR WHY. Which, yes, is obnoxious. I think any fan of the show can agree with that, and that's why "The Others" aren't exactly the big protagonists of the show. The comments in that thread were speaking approvingly about [MOUSEOVER TO REVEAL SPOILER]. Given that, I think it's pretty clear that the person who wrote that comment is a Lost fan, not a hater.
posted by koeselitz at 10:10 PM on March 3, 2010 [4 favorites]

I'm reading his new book right now and it's pretty fucking mind blowing. I can't recommend it enough.
posted by empath at 10:11 PM on March 3, 2010 [1 favorite]

I don't like it when time travel is explained in this way.

In his defense, it's complicated to explain. And he spends pages and pages and pages on it in his book, and goes over all that.
posted by empath at 10:15 PM on March 3, 2010

[MOUSEOVER TO REVEAL SPOILER]

Awesome use of HTML.
posted by grouse at 10:16 PM on March 3, 2010 [2 favorites]

So that lostpedia thread makes me think it would be awesome to invite Sean Carroll to Metafilter to chat/promote his book, in case anyone actually wants to do the metafilter interview thing I suggested a few days ago :)
posted by empath at 10:19 PM on March 3, 2010

Most fundamentally, you can’t go into the past and alter the future; there are no alternate histories or any such cheap ploys.

I love the blog post, but I don't get this. I mean, I get it from a determinist perspective, but it's one of those "impossible to know" things, you know? Not even like, "Is there a God" where faith and science and lack-of-proof and cosmology and ontology come into it, but more like, "If Dragons existed, who would the interact with leprechauns?" It's just so hypothetical that not only can we come to any conclusions about it, but even if time travel existed we'd be hard-pressed to find an experiment which could prove the hypothesis one way or another.

I had a thought experiment back in high school - as far as I know it's original but someone else has probably thought of the same type of thing - that I called the "transmigrant sould theory." Essentially the idea was, how do we know that when we sleep, our souls don't drift around in some ethereal game of musical chairs, awaking into a new body each morning, with all of that body's memories? We'd have no way of knowing, after all, just that we go to sleep and wake up.

Similarly, I love Lost, but half of the fun in Time Travel is exploiting the paradoxes. Which is part of why I love this season so much.
posted by Navelgazer at 10:24 PM on March 3, 2010 [1 favorite]

Thanks milestogo, that's pretty good. And whoa, moxiedoll that's good too.

And koeselitz, if that's a derail, I started it and I'll offer my apologies to the haters and threadsitters out there. Thanks for contributing more than "stupid and unwatchable."
posted by peeedro at 10:30 PM on March 3, 2010

Most fundamentally, you can’t go into the past and alter the future; there are no alternate histories or any such cheap ploys.

The reason why you can't do it has to do with the grandfather paradox and the telegraph paradox. But more importantly, it has to do with the fact that going back in time requires faster than light travel, which, based on all available evidence, is impossible.
posted by empath at 10:37 PM on March 3, 2010

Did Hurley die yet? 'Cos I'm pretty sure the whole show is going to be in his head, like Saint Elsewhere, in the minute or two he has as the plane plummets into the ocean.
posted by barnacles at 10:39 PM on March 3, 2010

grouse: “Awesome use of HTML.”

Thanks - and if anybody wants to know how to do that, here's how.
posted by koeselitz at 10:48 PM on March 3, 2010 [3 favorites]

(and that's why all your time travel worries will make "sense", I should say; dying moments don't NEED to make sense)
posted by barnacles at 10:48 PM on March 3, 2010 [1 favorite]

koeselitz: [MOUSEOVER TO REVEAL SPOILER]

grouse: Awesome use of HTML.

Well, uh, don't sell us short. The use of a non-linear timeline and time travel might reveal the genius of other posts in less obvious ways.

empath: But more importantly, it has to do with the fact that going back in time requires faster than light travel, which, based on all available evidence, is impossible.

Wait, is this the guy who traveled here from the future telling me this? Can I trust him? Is he making that up?
posted by peeedro at 10:51 PM on March 3, 2010

Well, Hurley - yeah. I don't really know what to say about most of the character development on Lost – that's not really the point of the show, I guess – but I've liked that character since I laid eyes on him, and I think he's been established well from the beginning. Might even be the only character that's really rounded out quite well. He's the reason I kept watching, anyway. Love that guy.
posted by koeselitz at 10:52 PM on March 3, 2010

I love the blog post, but I don't get this. I mean, I get it from a determinist perspective, but it's one of those "impossible to know" things, you know?

Time travel that involves changing the past is genuinely incoherent. I'm under the impression that most every serious physicist thinks that time travel is in principle possible but would not allow changes to be made to the past.
posted by painquale at 11:45 PM on March 3, 2010

Of course the universe could have some sort of strange metatime and we don't notice since we don't travel through time all that often. It would be a whole lot simpler if it didn't. But the universe is what it is, and sometimes it turns out to be pretty damn weird.
posted by Zalzidrax at 12:29 AM on March 4, 2010

We travel through time constantly. But just the one direction.
posted by empath at 12:35 AM on March 4, 2010 [1 favorite]

painquale, depends on how much you want to split hairs. You cannot alter your past (in the broadest sense - all the history of the universe as it played to cause your current state). Or, alternately, you cannot alter your past (same broad sense).

For example, let's say 50-years-in-the future painquale is just about to enter a machine that will, in fact, just disintegrate him at the sub-atomic level into pure entropy (his existence is not really a necessary condition). There's a non-null chance (never mind the amount of decimal places) of, right now, 2010, zillions of sub-atomic particles just tunneling all at the same time into the exact same configuration as 50-years-in-the-future painquale (supposing memory/thought process/sentience is a physical phenomenon). With a strong many-worlds interpretation, since that's possible (no matter how infinitesimally improbable), that's necessarily one component of the universe's wave function (that is, one "parallel universe"), so, there's definitely one branch where it just happened. If you ask that particular "time-traveling" painquale, he'll tell you that he sure is a time traveller, he disappeared from the future and appeared here. Never mind that there's no causal relation between new-painquale showing up and old painquale disappearing, in his mind it's solid. He can of course, go ahead and kill present day painquale, and it won't do shit, as there is no actual causal relation (as I said, that future where he thinks he came from may even have absolute zero chance of existing). Of course, he can decide to disintegrate himself again, and THIS pretericide-painquale configuration can again also just show up randomly 50 years later, in another infinitesimally improbable branch of this same branch of the universe (again, no causality violation here). Again, no time travel, just particles tunneling around. But 2060 pretericide-painquale of course has the whole causal relation in his mind. And no one around him will have any idea who he is (aside from being the guy who said he came from the future and killed painquale), which is exactly his expected outcome of a time travel. Success, for all he cares, and no actual causality laws broken*.

* Like a good pseudo-scientist, I completely ignored entropy in this example.
posted by qvantamon at 12:44 AM on March 4, 2010 [3 favorites]

Obligatory citation:

"I have a time machine at home. It only goes forward at regular speed. It's essentially a cardboard box and on the outside I wrote 'Time Machine' in Sharpie." (Demetri Martin)
posted by qvantamon at 12:47 AM on March 4, 2010 [1 favorite]

Right, but that's not time travel. Hopping into a parallel universe that is qualitatively just like the past is not hopping into the past.
posted by painquale at 12:51 AM on March 4, 2010

Of course time travel is impossible. What a silly idea. Now excuse me, I need to leave. CERN started with their shit again, need to find a large bird and some laxatives ASAP.
posted by Higgs Boson at 12:51 AM on March 4, 2010 [2 favorites]

The most recent episode had some really bad writing.

Compared to what? Have you watched TV recently? Even if some of the dialogue was sub-par, that episode was Henry V relative to the rest of broadcast television.
posted by Mayor Curley at 2:10 AM on March 4, 2010 [2 favorites]

Anyway, Lostpedia is a little too dense and spoiler-rich for me. Can anyone point to any decent meta-analysis of Lost looking at the writers, producers, themes, and story structure and not so focused on the story minutia canon?

I think your question remains unanswered, as many of the, quite good, analysis and recaps linked here don't go much further than pointing to details about the web of relationships and weird stuff that populates Lost. Like you can read at Gamasutra, what I wish would happen is that once the series finish, the producers write a gigantic postmortem for the series, including all the dirty stuff about writing a tv series like this. Like "Lost, the kitchen" or something like that, I would like to know what was the original idea (show me the napkin!), and how evolved. I mean, I would like to know everything about the creative and commercial and business process involved in Lost.
posted by samelborp at 3:07 AM on March 4, 2010

Right, but that's not time travel. Hopping into a parallel universe that is qualitatively just like the past is not hopping into the past.

Why not?

---

Also I was thinking. What if you had a machine that could reverse entropy, reverse the 'arrow of time'. You can do that on small scales if you increase entropy somewhere else, and use energy. So, for example take the archetypal "messy desk".

As time moves forward, the desk gets messier. Entropy! But obviously you can clean up your desk if you want to. So what if we called something "Time travel" but actually it would just put everything back the way it had originally been.

That supposes that quantum events are controlled by some deterministic mechanism below what we can see, and that we can actually calculate where everything had been in the past, and then move it.

So it would be a way of moving into the past by "undoing" the future. You would have to do this over the whole earth to work, but then of course all our space probes and other planets would look like they were in the wrong places, so you'd even need to go so far as to do it to the entire solar system.

But it seems to me if we had a computer larger then the solar system, more energy then the sun, and an ability to move every hypothetical sub-quantum particle in the solar system, we could do it.

I've read things where people say that time is only the arrow of time from thermodynamics, so it stands to reason if that's the case, and if we can locally reduce entropy, then we could also locally reverse time.
posted by delmoi at 3:28 AM on March 4, 2010

I work with a four year old who is obsessed with clocks and time right now and time itself is pretty damn weird when you really try to explain it. Trying to explain that time only has one speed and only moves forward and no, I can't make the clock go faster and no, the only answer I have is "BECAUSE THAT'S HOW IT IS" really makes you realize how little you know about the universe.

Or maybe that's just me.

Also, I love Lost and the only thing better than Sayid is Evil Sayid.
posted by grapefruitmoon at 5:34 AM on March 4, 2010 [2 favorites]

The reason why you can't do it has to do with the grandfather paradox

Philip J. Fry had a novel, if somewhat disgusting, solution to that particular paradox.
posted by DecemberBoy at 6:15 AM on March 4, 2010

http://en.wikipedia.org/wiki/Roswell_That_Ends_Well
posted by DecemberBoy at 6:17 AM on March 4, 2010

Choke on that, causality!
posted by adamdschneider at 8:02 AM on March 4, 2010

Delmoi, if you wanted to avoid the appearance of discrepancies, you'd need to do it to the whole universe (think of our solar system's position in the galaxy, light reaching us from other galaxies, etc.), which would require the ability to "pump" entropy into some other universe. Entropy is weird. I am so going to start reading my copy of Four Laws That Drive the Universe. Anyway, I would bet there is some natural law/mechanism preventing us from exercising such fine and total control.
posted by adamdschneider at 8:37 AM on March 4, 2010

Right, but that's not time travel. Hopping into a parallel universe that is qualitatively just like the past is not hopping into the past.

Why not?

Well, I guess you could say it's a kind of time travel. But going to a pastlike parallel spacetime and saying you've gone back in time is like going to Amish country and saying you've gone back in time, or going to a movie set of Tahiti and saying you've gone to Tahiti. Returning to a place involves returning to a particular point, not a different point that has a bunch of the same properties. I take it that real time travel involves returning to a spacetime point that is somewhere behind us in our light cone; parallel universes aren't in our light cone.

I've read things where people say that time is only the arrow of time from thermodynamics, so it stands to reason if that's the case, and if we can locally reduce entropy, then we could also locally reverse time.

I have heard people say this! You can get some cool sci-fi stories out of it. You might think that the arrow of time supervenes on the increasing entropy of the universe as a whole and not just local subpockets, though. But this is more about the perception of the arrow of time, anyway; if you're going to reduce statements about the direction of time to statements about entropy, there's a pretty good case to be made for saying that you don't think time has a direction at all, it's just that entropy explains why we think that time has a direction.

(That seems to be the big problem in explaining the flow of time and time travel anyway. Even if you do think that there is a privileged direction of time and the laws of physics are time-asymmetric, sticking a little arrow on our big block of spacetime doesn't really explain why we experience time as flowing in that direction.)
posted by painquale at 8:50 AM on March 4, 2010

I've read things where people say that time is only the arrow of time from thermodynamics, so it stands to reason if that's the case, and if we can locally reduce entropy, then we could also locally reverse time.

Sean Carroll actually discusses that in his book.

If you get the positions and momentum of every single particle in a given closed system, and reverse the momentums (and a few other changes, like switching all the particles to anti-particles, etc) then you would have effectively reversed time. However that only works in a closed system. As soon as it so much as touches a photon from the 'ordinary' universe, then the 2nd law of thermodynamics starts again and normal entropy is introduced.

A further problem is that if you lived in a part of the universe where entropy was locally decreasing, then you wouldn't be able to tell. Your own sense of time would reverse, and the future would just become your past.
posted by empath at 9:22 AM on March 4, 2010

Trying to explain that time only has one speed and only moves forward and no, I can't make the clock go faster and no, the only answer I have is "BECAUSE THAT'S HOW IT IS" really makes you realize how little you know about the universe.

You might want to read up on general relativity, because it sounds like you know even less than you think :) I can't imagine trying to explain THAT to a four year old, though.
posted by empath at 9:24 AM on March 4, 2010

A further problem is that if you lived in a part of the universe where entropy was locally decreasing, then you wouldn't be able to tell. Your own sense of time would reverse, and the future would just become your past.

That is, assuming that entropy in your body was also locally decreasing. If it were unchanged from its current state, then your body would introduce entropy into the system again.

The reason for that is that in any given collection of possible states for a closed system, the number of states where entropy is low is vanishingly small compared to the number of states where entropy is high. Any universe that's configured in such a way that entropy is always locally decreasing has to be one that is configured in a very particular way. Any change in the system introduced from a normal, entropy-increasing system (such as your body) will change the entire system to a normal entropy-increasing one.
posted by empath at 9:31 AM on March 4, 2010

A further problem is that if you lived in a part of the universe where entropy was locally decreasing

can you give an example of entropy decreasing?
posted by freshundz at 9:33 AM on March 4, 2010

can you give an example of entropy decreasing?

There aren't any. But it's theoretically imaginable. Just take any system in the universe, separate it from the rest of the universe, then reverse all the momentums, spins, etc. Then you have a closed system where entropy is locally decreasing.

All the laws of physics are time symmetrical, so if you run them backwards, everything works just as well. Since the initial state of the universe was low entropy, then if you reverse any part of that system, then you'll be resetting it so that entropy is decreasing. This only works as long as nothing interferes with the system. Any change will make it an entropy increasing system again.

If the initial state of the universe WASN'T low entropy, then time (probably) wouldn't have an arrow to begin with.

If you think about it, out of all the potential states for any given closed system, there are just as many possible states where entropy is always decreasing as there are systems where entropy is always increasing. (take any system where entropy is always increasing, and reverse it). Both of those states make up an incredibly small percentage compared to the number where entropy stays more or less static.

The fact that we live in a universe where entropy is constantly increasing is actually incredibly unlikely, and is due to the low entropy initial state of the big bang.

Of course Heisenberg's uncertainty principle makes finding all the states, etc, impossible anyway.
posted by empath at 9:50 AM on March 4, 2010

From what I understand, what you are proposing is impossible because according to the laws of physics the processes are irreversible.

The fact that we live in a universe where entropy is constantly increasing is actually incredibly unlikely

I dont know about "likely" but since increasing entropy is, I believe, an inevitable result of any process then it stands to reason that the universe would have to have low entropy to start with.
posted by freshundz at 10:02 AM on March 4, 2010

From what I understand, what you are proposing is impossible because according to the laws of physics the processes are irreversible.

The universe is CPT-symmetric. It would work just as well reversed.
posted by empath at 10:04 AM on March 4, 2010

I dont know about "likely" but since increasing entropy is, I believe, an inevitable result of any process then it stands to reason that the universe would have to have low entropy to start with.

You're kind of begging the question a bit. The low entropy state is unlikely, whenever it occured. The fact that it set the arrow of time is a result of that unlikely state existing.
posted by empath at 10:09 AM on March 4, 2010

[Hurley] might even be the only character that's really rounded out quite well.

Is this some sort of pun?
posted by inigo2 at 11:06 AM on March 4, 2010

I had a thought experiment back in high school - as far as I know it's original but someone else has probably thought of the same type of thing - that I called the "transmigrant sould theory." Essentially the idea was, how do we know that when we sleep, our souls don't drift around in some ethereal game of musical chairs, awaking into a new body each morning, with all of that body's memories? We'd have no way of knowing, after all, just that we go to sleep and wake up.

Did you come up with that idea after watching Dark City? Sleeeeep now!
posted by Thoughtcrime at 11:06 AM on March 4, 2010

The low entropy state is unlikely, whenever it occured. The fact that it set the arrow of time is a result of that unlikely state existing.

It's unlikely that a high entropy state will become a low entropy one, not that a low entropy state exists at all. The likelihood of the initial condition of the universe being low entropy is something that we currently have no way of answering, and that might not even make sense.
posted by painquale at 11:41 AM on March 4, 2010

It's unlikely that a high entropy state will become a low entropy one, not that a low entropy state exists at all.

To use an example from Sean's book -- imagine 2 boxes connected by a hole. Now imagine there are 2000 gas atoms inside. For the sake of simplification, let's say that the only information we care about is whether the gas atoms are in one box or the other.

The lowest entropy state is that all 2000 atoms on one side. Out of all the possible states, there's only 1 in which that case is true. There are 2000 cases where there is 1 atom on 1 side and 1999 on the other side. There are 2000^2 cases where there are 2 atoms on one side, and it goes up exponentially as the atoms divide more evenly.

If the initial state of that 'universe' was random, then one in which all 2000 atoms were on one side, that's almost unimaginably improbable.

The second law of thermodynamics works at all depends on low entropy states being improbable, not the other way around.
posted by empath at 11:53 AM on March 4, 2010

The likelihood of the initial condition of the universe being low entropy is something that we currently have no way of answering, and that might not even make sense.

Entropy is almost by definition the probability of a certain state existing. If we know the entropy of the early universe (and we do), then we know the probability that it existed.
posted by empath at 11:56 AM on March 4, 2010

For further reference: Statistical Mechanics
posted by empath at 11:59 AM on March 4, 2010

If the initial state of that 'universe' was random, then one in which all 2000 atoms were on one side, that's almost unimaginably improbable.

But we have no way of knowing whether this is true. This is an instance of the base rate fallacy. There's nothing in statistical mechanics that tells us that that's the least likely initial condition.

There's a big difference between asking questions about why laws, regularities, the initial condition, the cosmological constant, etc. are the way they are, and questions about what is implied when all those other things are fixed. Everything we know about probability and statistical mechanics applies to the latter. Sometimes people use the fact that the universe's laws are just so to argue for things like the existence of god (isn't it crazy that, out of all the possible laws that could have existed, the ones that support life are the ones that exist?). There are anthropic concerns here, but apart from that, these arguments rely on the same base rate fallacy. We have no idea how many possible states of initial conditions there could have been, how the laws could have been different, or the relative likelihood of variations on this score. Perhaps the set of laws that we have are the most likely ones there could have been... or necessarily the only ones.
posted by painquale at 1:04 PM on March 4, 2010 [1 favorite]

But we have no way of knowing whether this is true.

IF the initial microstate of the universe was randomly determined, then that it was such a low entropy microstate is wildly improbable. I really don't think that's even logically arguable. I'm suggesting that the state of the universe at the time of the big bang was either not randomly determined or that given an infinite amount of space time, the existence such an anomalous area becomes inevitable. (see Boltzmann's brain)
posted by empath at 2:45 PM on March 4, 2010

IF the initial microstate of the universe was randomly determined, then that it was such a low entropy microstate is wildly improbable.

If there are many possible initial conditions and all possible initial states have equal prior probabilities, then yeah, I agree with you. But we don't have any reason to think that either of those are true, and I don't think we can assume anything one way or another. Did the universe have to blast out of a big bang, or could it have started some other way? I don't know how we'd start answering a question like that, and I think a lot of physicists might even shrug it off as a pseudoquestion.

given an infinite amount of space time, the existence such an anomalous area becomes inevitable.

I dunno, I don't think we can make any assumptions about whether an infinitely large area will inevitably lead to anything in particular. Pi might contain one thousand 9s in a row or it might not; the fact that it's non-terminating and non-repeating doesn't mean that it needs to contain every possible sequence of digits.
posted by painquale at 5:01 PM on March 4, 2010

I can almost guarantee that there is a thousand 9s in a row, but you'd have to go up to about 10^1000 digits to be guaranteed to find it, assuming pi is normally distributed, which it seems to be. If you go up to 10^8 digits, you can find strings of 8 repeating digits in a row, just as you'd expect.
posted by empath at 5:13 PM on March 4, 2010

I thought it was an open problem whether pi contains all finite sequences? Anyway, I didn't have to choose pi: I just wanted to make the point that irrationals needn't contain all finite sequences.
posted by painquale at 5:24 PM on March 4, 2010

Incidentally, the reason that I think that this is interesting is that people have only recently begun to appreciate our lack of knowledge of the prior probabilities of the conditions of the universe. (There's been a bit of a Bayesian revolution in philosophy of science as of late.) it turns out that in light of this fact, a whole bunch of shopworn arguments in the philosophy of science get refuted.

1) It attacks the fine-tuning argument for the existence of god that I mentioned earlier. We can't know whether the likelihood of the cosmological constant being what actually is is infinitesimally small, so we don't need to invoke god to explain it.

2) It attacks certain objections against Humeanism about laws. Realists about laws claim that the initial state of the universe and laws of nature are fundamental (and all subsequent states get generated from these), and Humeans claim that the fundamental thing is a big multi-dimensional spacetime block, and laws are expressions of regularities within this block. An objection to Humeanism is that it seems like an extraordinary coincidence the spacetime block is structured in such a patterned way that it allows for us to sum up its structure with a theory involving so few laws. However, the Bayesian rebuttal states that there's no reason to think it's an extraordinary coincidence that the universe has this structure, because there's no reason to think it has lots of other metaphysical contenders. And the realist is in the same boat: there's no reason to think that a block universe structured just so is any less likely than the initial condition and the laws being just so.

3) It attacks the conflation of the unlikelihood of the existence of a low entropy state and the unlikelihood of a high entropy state becoming a low entropy state, as we've been discussing in this thread.

4) It attacks the no-miracles argument in favor of scientific realism. The no-miracles argument works like this: take any of our mature scientific theories. Look how well it works! Look at all the predictions it makes! It would be a miracle for it to work so well if the theory were false. And we don't want to countenance miracles. So, the theory is (at least approximately) true. The problem here is the sentence "It would be a miracle for it to work so well if the theory were false." This is another instance of the base rate fallacy. We have no idea how many false theories would work well. Maybe many would; maybe it wouldn't take a miracle at all.

5) It (purportedly) attacks the pessimistic induction argument against scientific realism. The pessimistic induction argument is a sort of counterpoint to the no-miracles argument which points out that nearly every successful scientific theory we've posited in the past has turned out to be false, so we should inductively expect our current theory to be false too. I admit that I don't really get the base rate fallacy underlying this argument, but the point has been made. It (and the previous) are argued for here.
posted by painquale at 7:44 PM on March 4, 2010

Just wanted to note that Sean Carroll the Caltech physicist should not be confused with Sean Carroll the Wisconsin biologist, who has also written some very good popular science books.

[Side question: does anyone remember the webcomic which played on that? I think it involved a suggestion of a science fantasy league, akin to sports fantasy league. One person drafted "Sean Carroll" and the other asked which Sean Carroll he meant. I thought it was an XKCD strip but there's a good chance I'm wrong about that. I've been searching for it and haven't been able to find it.]
posted by DevilsAdvocate at 10:34 AM on March 5, 2010