Maybe I misundertand something, but I find Greg Egan's treatment of the simultaneity problem between distant clones in Diaspora problematic, strangely at odds with the overall technically detailed rigorous style of the book. Consider this in Chapter 11:

Elena had chose not to wake if any other versions of her had already encountered life. Whatever fate befell each of the remaining ships, every other version of him would have to live without her.

But this is surely unattainable, given the clones are tens of light years dispersed into different directions! Let's say clone #1 just encountered life, and clone #2 seventy light years away is about to encounter life very soon. There's no way to prevent #2 from waking because by the time message of encountering life from #1 arrived 70 years later, it would be too late.

Special relativity 101 tells us in cases like these, there will be some frame of reference from whose perspective #1 encounter life first, and another frame of reference from whose perspective #2 encounter life first. There's no objective claim to which clone encountered life first, and both clones will end up waking! Given fragmentation of identity is a major theme and pathos of the story, I find it unlikely that GE is just sweeping this basic limitation under the rug for (in)convenience.

I hope I'm not being too serious, because this is a wonderfully super-technical serious story ;-P


There's no indication in the novel how Elena's scheme was implemented but absent superluminal signaling, her various exoselves would likely adopt the following strategy.

Sort the Carter-Zimmerman polis expeditions by estimated time of arrival. Each exoself in each polis waits for news from all the trips with earlier arrival times before waking their copy of Elena. If after all the returns come in there is no news of a life discovery from an expedition with an earlier arrival time, then that exoself can wake its copy of Elena if life is discovered on that particular expedition.

  • I think this strategy works if every ship has predestined destinations, which seems to be the case, at least for the first 1 or 2 millennium. – Eric Feb 20 '16 at 2:34
  • Besides, I think this strategy may fail if each polis uses a different frame of reference, if by "earlier arrival times" you mean earlier with respect to local times of different polises. Two polises may disagree about who arrived earlier by their own standards since they are moving at some fraction of c relative to each other. – Eric Feb 20 '16 at 2:49
  • 1
    In the Glossary at the end, UTC is "extended across interstellar distances by use of a reference frame at rest with respect to the sun". – Eric Feb 20 '16 at 2:59
  • 1
    Each ship would have flight plan, so you can sort the trips by expected duration before anyone leaves. Also, using the flight plans each ship could slew their clock ticks to keep faith with UTC, adjusting for what special relativity was doing to their measurements. As long as they do that, there is an objective time standard. But really just computing the flight times before everyone leaves should be enough. – Kyle Jones Feb 20 '16 at 2:59
  • Now I've come to think to wait may be the only way to implement Elena's scheme, to prevent another doppelganger from waking up. Better be a polis citizen than a flesher in this case because they can rush ;-P – Eric Feb 20 '16 at 3:14

The crafts were going significantly slower than the speed of light, so I think it's plausible that for nearly all pairs of crafts going to two destinations A and B, if the first craft sent a signal at the moment it arrived at A, and the signal moved at the speed of light, then the signal would reach B before the other craft did.

We can estimate the typical speed of their interstellar craft--in Chapter 10 we see a clone of Paolo Venetti about to leave Earth on 31 December 3999, and in Chapter 11 we see him arrive at the star Vega on 10 September 4309, so the trip took about 310 years, rounded off. And according to the wikipedia article on Vega, this star is about 25 light-years from Earth, so the average speed must have been 25/310 times the speed of light, or about 8%.

Now, suppose that another craft had gone in the opposite direction on 31 December 3999, to a star 30 light-years from Earth. Then at 8% the speed of light, it would arrive 375 years after it left, or 65 years after the first craft reached Vega. But the distance between this star and Vega would be 30 + 25 = 55 light years, so if the craft that arrived at Vega immediately sent a signal towards this other star, then despite their lying in exactly opposite directions, the signal would still beat the second craft to the star by 10 years.

Only if two craft happened to be going to star systems almost exactly the same distance from Earth would you get a situation where this wouldn't work--and the wiggle room in "almost exactly" would be largest if the two systems happened to lie in opposite directions from Earth, if they were closer to the same direction, the match in distances would have to be even more exact. For the case of stars in opposite directions, it's easy enough to figure out a formula--if one star is a distance D1 and the other a larger distance D2, then the difference in time between the two craft arriving will be (D2 - D1)/0.08c, while the time for a light signal to go from one to the other will be (D1 + D2)/c. In this case, the largest value for the difference in distance, (D2 - D1) that would allow for the second craft to arrive at the same moment or earlier than the light signal would be when (D1 + D2)/c = (D2 - D1)/0.08c, and a little algebra shows this means D2/D1 = (1 + 0.08)/(1 - 0.08) = 1.174. So if the two stars are in opposite directions, the second star's distance must be less than 1.174 times the distance of the first in order to prevent the light signal from beating the second craft, and if they are not in opposite directions the maximum possible difference in distance would be even less.

  • Nice calculation. I remember somewhere in the book it says the ships are traveling at about 10% speed of light. But I think there's enough room in the 17.4% distance to accommodate a lot of stars! So this bet is unlikely to work. – Eric Feb 20 '16 at 1:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.