Do the laws of physics permit any chance for a Type III civilization on the Kardashev scale to halt or even reverse the expansion of the universe, or bring the galaxies back closer together, assuming they had a few billion years to accomplish it? Would it have to be a Type IV on the extended scale?

It does not have to be a biological civilization where individuals are mortal. It could be an self-aware AI civilization or any other kind that doesn't have to deal with internal conflicts or fluctuations in public opinion or priorities or anything else that would be distracting.

We assume that heat death of the universe is inevitable because of entropy, but we also know that energy can't be created or destroyed - so, it here any way to prevent the heat death of the universe by stopping the galaxies from drifting apart, even a few of them, if not all? All stars eventually burn out, but perhaps the loss (dissipation) of their energy could be prevented by enclosing them in hermetically sealed Dyson spheres?

  • 4
    $\begingroup$ You're dealing with two distinct problems here. First is stopping the expansion of the universe, and the second is reversing entropy. They're from different branches of physics, and solving one doesn't have any impact on the other. If you were to somehow slow down the expansion of the universe, the heat death would still occur. $\endgroup$ Jun 9, 2020 at 19:36
  • $\begingroup$ According to physics, and assuming that "heat death" theory is correct, an advanced civilization can postpone, but not prevent the inevitable. $\endgroup$
    – Alexander
    Jun 9, 2020 at 20:33
  • $\begingroup$ is your version of universe truly a "closed" system because like it or not that's entropy $\endgroup$
    – user6760
    Jun 10, 2020 at 0:02
  • $\begingroup$ It is possible, but only if you harvest suffering of adolescent girls as they transform into eldritich abominations. $\endgroup$
    – Euphoric
    Jun 10, 2020 at 5:23
  • $\begingroup$ I think you'd have to go all the way to Type V. Type IV has mastery over the entire energy output of the Universe, but even this is finite and will eventually run out due to entropy. The only way you could possibly replenish it would be to take energy from another universe, and at that point you're talking about Type V. $\endgroup$ Jun 10, 2020 at 14:01

2 Answers 2


This is certainly a question that's been thought about a lot. Hooper (2018) suggested that a Type III civilization could quickly travel outwards from their home and build Dyson spheres around stars in other galaxy clusters. The ideal range is stars of $0.2M_{\odot}<M<1M_{\odot}$, as they live long enough to be useful but are massive enough to not produce insignificant amounts of energy. Hooper calculated that, assuming a Dyson sphere efficiency of $\eta=1$, the civilization could capture and bring back stars currently up to $d=65\text{ Mpc}$ away, assuming they travel outwards at a speed of $v=0.1c$. This would make much of the Virgo Supercluster accessible.

Some shortcomings:

  1. This is only a plan for mitigating the expansion, not ending it. It would dramatically increase the amount of material that a civilization could keep gravitationally bound after other galaxy clusters begin to move away, but it is only effective over a finite volume of space.
  2. It doesn't affect non-baryonic matter (dark matter and dark energy) or objects like gas clouds - in other words, the vast majority of the universe. We can only get a select subset of stars and anything in orbit around them. This is the same issue you run into when trying to move an entire galaxy when just moving stars.$^{\dagger}$

On the other hand, actually stopping the expansion of the universe would require changing the energy density of the universe at all points in space. The mean density would need to be brought below the critical density, and we can't simply conjure matter or energy out of thin air (thin vacuum?) - it's a law of physics. A civilization can only mitigate the effects of the expansion by increasing the matter available to them as time goes to infinity - and as Hooper showed, that can be significant, possibly increasing the mass of the Local Group by three orders of magnitude! But in an infinite universe, the problem seems unsolvable.

$^{\dagger}$For the case of Shkadov thrusters, Hooper notes that efficiencies would be much too low for this to be a feasible route in our scenario.

  • 1
    $\begingroup$ That was very informative, thank you! It seems to me that dark matter and dark energy pose the biggest problem. $\endgroup$
    – Jinjinov
    Jun 9, 2020 at 20:20
  • 1
    $\begingroup$ @Jinjinov Agreed! Let’s destroy them so they aren’t a problem anymore! ;-) $\endgroup$
    – SRM
    Jun 10, 2020 at 0:00

General relativity posits that the expansion of the universe is driven by a cosmological constant, a uniform distribution of energy everywhere in the universe; I have no idea how a civilization might attempt to influence this energy density, nor does anyone else. Dragging a few galaxies into mutual gravitational influence (so that they are not driven apart) seems plenty doable for a civilization with enough Shkadov thrusters and willpower; after all, the Milky Way/Andromeda child galaxy isn't projected to be ripped apart, so all you'd need to do is repeat that with as many galaxies as you can grab before the more distant ones get too far away to return from.

That doesn't actually prevent heat death, though.

If the universe suddenly just stopped expanding one day, the natural physical progression of its contents would still be towards thermodynamic equilibrium. The stars still burn out, anything hot equalizes with the temperature of its surroundings, the universe runs out of usable heat gradients for civilizations to extract work from, and it eventually enters a state of uniform temperature and maximum entropy.

A Dyson sphere cannot be perfectly insulated and, barring some extraordinary far-future energy storage technology, will leak energy in heat. Even if you have magical infinite-duration batteries, if a civilization wants to do anything with this energy at all, it has to run it through some processes which will be less than perfectly efficient (the second law of thermodynamics limits heat engine efficiency based on temperature; theoretically it'd be perfect at absolute zero, but you can't reach that). And if you're not using the energy for something, why do you want to keep it around?

  • $\begingroup$ So, it comes down to creating a perfectly insulated system that does not leak heat. Such a system could use the energy with less than perfect efficiency, providing that it would capture the heat that wasn't used for work. $\endgroup$
    – Jinjinov
    Jun 9, 2020 at 20:17
  • 2
    $\begingroup$ I'm not aware that it's feasible to capture and reuse waste heat infinitely. The Dyson sphere approach to this is to build a Matrioshka brain — a sequence of shells outside the first one that catch everything the shell inside radiates — but the outermost shell in a Matrioshka brain still has to leak at some point, when the gradient between its temperature and that of the background universe is too small to usefully extract work with any more shells. $\endgroup$
    – parasoup
    Jun 9, 2020 at 20:21

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .