The Dyson Dilemma states that every civilization of a great enough development stage will want to capture all of the energy from its home star and surround it with a Dyson Swarm that uses solar panels to capture every last square meter of sunlight. Such a structure should be detectable within 1 kpc and yet our current sky surveys have found bupkis.

Building millions of solar panels in space and dealing with the cosmic rays, solar flares and micrometeorites seems like a real hassle. Having your entire civilization in one location also seems like it would make it vulnerable to interstellar warfare, rogue nanobots, and so on...

On our current technological development timeline, we will have nuclear fusion power well before any Dyson Swarm gets up and running (Decades vs Centuries). The Sun is also extremely inefficient running at 10^4 times slower than any man-made fusion process. It's so slow that, per unit volume, the sun puts out about the same energy as your average compost heap.

A civilization that grows large enough to need a Dyson swarm is going to keep growing to the point where its star's output is no longer fast enough, and it will need to access the hydrogen directly (Starlifting) in order to power all of its fusion reactors.

So why would any alien civilization ever build these Dyson Swarms?

Dyson Swarm - Source: https://commons.wikimedia.org/wiki/File:Dyson_swarm.png


Our sun produces something like $3.8 \times 10^{26}$ Watts.

That requires something in the neighborhood of 600 million tonnes of Hydrogen per second.

Getting a similar power from fusion would require a similar consumption of Hydrogen. Even for a culture that could build a Dyson sphere, that's a lot of Hydrogen to get if you are not using a star to do it. Are there ways to get it? Oh, probably. Are they as easy as would be cuddling up to a star? Probably not. One might imagine scooping up interstellar hydrogen, for example. Or vacuuming nearby solar systems.

As to meteorites, they would certainly scour the system they started in. They would want the mass, if nothing else. They might need to scour some nearby systems as well. They might even need to harvest mass from nearby stars, and do some nuclear synthesis.

As to cosmic rays, they will have a huge device from which to project any protective measures. Cosmic rays can be deflected with magnetic fields, for example. One expects there are other ways of dealing with radiation from space that would be discovered.

A Dyson sphere might well be easy to detect during construction, especially as the star's brightness starts to do wacky things. After all, Tabby's Star was pretty exciting for a while. It seems that natural explanations have been found. But for a while, it was tempting to hypothesize that we might have exciting "neighbors."

Once it's constructed it might be a lot harder to detect. It will radiate the same amount of energy but at a much lower temperature than the star inside. I'm not an astronomer, so I could be completely wrong here.

Plus, when you have $3.8 \times 10^{26}$ Watts to play with, you might be able to construct some exciting defense systems. Imagine they could direct some appreciable fraction of their star's solar wind into a beam, just as an example.

Having a culture with a huge number of citizens in one place has interesting possibilities. Imagine, for example, the computer network they could construct. A Dyson sphere with the orbit of the Earth has an area of $2.8 \times 10^{17}$ square km. If you had even one fairly modest CPU in each of those square km, you'd have a truly enormous parallel computer. It's quite difficult to project what you could do with such a system. But it would certainly not be boring.


Your question is like asking "why build an oil power plant with all the related hassle, when we can chop wood and light a fire?".

The answer is: the order of magnitude of the produced energy. A star emits Petawatts of energy, while a fusion power plant can produce Megawatts, several order of magnitude less.

And a star saves the hassle of harvesting all the fuel, since gravity already did the job.

  • 8
    $\begingroup$ Also, OP is assuming nuclear fusion WILL be feasible and viable. That's not given at all. I see the question more like "Why would an advanced civilization mine minerals from comets when it can use alchemy to transmute dirt into any mineral?" $\endgroup$ – Teleporting Goat Jun 20 '19 at 13:37
  • 3
    $\begingroup$ @TeleportingGoat Well, assuming nuclear fusion will be viable is not much of a stretch. However, the Sun still has an upper hand - it uses pure hydrogen. That's something we don't even have serious plans for, since it relies on the weak force converting protons into neutrons, and the weak force is, well... weak. Designing a nuclear reactor that can efficiently use protium fusion is quite a bit more complicated (the CNO cycle, which uses C, N and O as catalysts, would work, but that's far beyond our current technology). Until then, hydrogen fusion is limited by breeding. $\endgroup$ – Luaan Jun 21 '19 at 7:14

To power nuclear fusion, you need hydrogen. The overwhelmingly largest hydrogen reservoir in any solar system, and conveniently an already working fusion reactor is its star.

In other words, there is far more energy to collect from a star than you could ever hope to generate in reactors.

  • 1
    $\begingroup$ If you take the scientific limits further, dubbed Stellar Engineering if I'm right, you could start slicing up the stars to smaller, manageable sizes. "Easier" to use them as battery after overcoming the initial hurdle. $\endgroup$ – Lupus Jun 19 '19 at 15:45
  • 4
    $\begingroup$ @Benjamin not if you want to produce a compareable power output. $\endgroup$ – ths Jun 19 '19 at 16:40
  • 1
    $\begingroup$ @Benjamin The assumption that Dyson makes is that our civilization's energy needs will continue to expand geometrically along with our population and that inevitably we will require more energy than our planet can provide. $\endgroup$ – Mike Nichols Jun 19 '19 at 16:42
  • 2
    $\begingroup$ @Benjamin You just posted an argument that no civilization would risk such a great investment in a project, and yet you think given the arguments we have today over water rights and the need for water world-wide that a civilization would consume it's most irreplaceable resource for electricity? THS has simply and eloquently hit the nail on the head. Save the planet, use the star. $\endgroup$ – JBH Jun 19 '19 at 16:57
  • 1
    $\begingroup$ @Benjamin you are assuming magic, i.e. that the civilization can take ordinary water and make its hydrogen fuse with the same efficiency as a deuteron + tritium reaction, without the disadvantages of a star’s fusion, while being able to use the released energy flawlessly with a mechanism that for some reasons can’t be applied to the energy released by a star’s fusion, despite having an identical form. Well, under that premise, there might be indeed no reasons to build a Dyson sphere. $\endgroup$ – Holger Jun 20 '19 at 16:48

So, I'm going to frame challenge this, and say "no alien civilization entirely surrounds their stars with solar collectors."

Piecewise assembly

Here's the thing about a Dyson Swarm: It's not a single massive project that returns nothing until it's complete.

All you need is solar-orbiting structures that (also) collect solar power, and some sort of orbital traffic control. As long as people keep building orbital structures, the power collected by the system increases, and eventually you get a Dyson swarm. No one ever decides one day "we should make a Dyson swarm." Instead they decide to make one (or some other reasonable number) of solar-orbiting structures. They quickly get the benefit of that structure - whether it's a habitat, research structure, manufacturing hub, etc.

But then what?

This continues until either the unused mass in the system vanishes or the civilization kills itself. Once the unused mass in the system is gone, no more structures will be put around the star. Sure, going to another star is possible, but why bother to ship the mass back? Power is power, regardless of where you get it. Interstellar shipping is expensive in terms of energy. And, as you said, having your entire civilization in one place is a bad idea.

Therefore, advanced alien civilizations put partial Dyson swarms around every star they can get their mitts on. No star is ever fully surrounded.

  • $\begingroup$ Interstellar shipping is expensive in more than energy. What’s the average distance between stars? Fifty light-years? $\endgroup$ – WGroleau Jun 20 '19 at 5:45
  • 3
    $\begingroup$ "why bother to ship the mass back?" because there are advantages to a dense civilisation. Sure, it's really expensive but it's not that different to building expensive skyscrapers in a city centre and commuting when you could relocate your office and workers to the countryside. $\endgroup$ – Robin Bennett Jun 20 '19 at 9:22
  • 2
    $\begingroup$ Yeah, but when it's so expensive in energy and time to ship material back (that asteroid you're sending back to Sol could be used as part of your Von Newman economy) that it isn't worth it. This is shipping material to make a new skyscraper across the Atlantic ocean... using rafts. $\endgroup$ – ltmauve Jun 20 '19 at 17:20
  • 1
    $\begingroup$ "This continues until either the unused mass in the system vanishes or the civilization kills itself." Those aren't the only possible terminal conditions: the furthest it can continue is "until all the available mass is in use, OR all the radiant energy is being captured". Actually, if you run out of usable mass first, you can all start moving your structures into closer orbits to capture more energy, so full star utilization is the more "hard" upper limit. $\endgroup$ – dgould Jun 20 '19 at 19:18
  • $\begingroup$ @WGroleau It depends what scale you look at, and where. If you're asking "across the entire observable universe", it's much lower than just across the Milky Way, which in turn will be significantly less dense than just across the Milky Way galactic center. $\endgroup$ – Fund Monica's Lawsuit Jun 21 '19 at 17:25

The commonly given reason for building any version of a Dyson Sphere is because the civilization needs an amount of energy comparable to the output of their star. A Dyson sphere captures a significant fraction of that ouput, and does it more or less passively.

I would posit that a civilization could not create a Dyson sphere without first having fusion (both for electrical/thermal power, and for space propulsion), because of the tremendous amount of matter that needs to be moved around. There comes a time, however, when the materials (copper or superconductor precursors, for instance) to make fusion devices become too scarce, while (presumably) common structural materials are easier to source, and those to make solar collectors are also common.

If all the high quality conductors in the Earth were tied up in fusion generators, it would still be possible to collect more energy from solar emissions, if one could cover enough of the sky -- and eventually, if our civilization lasts long enough, we'll need that energy badly enough to begin construction of a Dyson sphere.



Your civilization has built up its system. It can meet its energy needs through fusion and more esoteric sources. But the sun is aging and as it ages, it becomes brighter.


The civilization does not need all that energy. Really does not need it - it is throwing things out of balance and threatening the habitability of terraformed planets closer to the star.

The answer is to soak up the extra. The Dyson sphere is a defensive maneuver to regulate the output of the star. It is analogous to damming a river - it is nice to have the river by your town but it is also potentially destructive. Its power output must be controlled and channeled.

The Dyson sphere reradiates output to simulate the younger star, and captures and converts the excess. The question of what to do with excess energy is an interesting one. Maybe they convert it back to matter.

  • 1
    $\begingroup$ I enjoy the idea of converting it back to matter -- it could even be a self-building station (or several of them) which, over time, gradually expand to soak up more and more of the excess energy. If you're writing a book off of this, that could be a plot hook. Maybe someone forty thousand years ago miscalculated the rate of stellar decay and now too much energy is being absorbed because too much surface area is covered. $\endgroup$ – Fund Monica's Lawsuit Jun 21 '19 at 17:27

Psychological Reasons

Why do people climb mountains? It is dangerous and dirty. There are more safe and efficient ways to get exercise. There are more safe and efficient ways to get to the top.

Why do people buy Lamborghinis? A Toyota Corolla costs substantially less, carries more, is more comfortable, and uses less fuel.

Why do we keep building more, stronger weapons? We can already equip our armies and then some. We can already wipe out all life a few times over.

They build a Dyson swarm for the same reason they climb a mountain. Because they want to challenge themselves. Because they want to look out and say 'I did that, and not everyone could. Or would.'

They build a Dyson swarm for the same reason they buy a Lamborghini. To show off. To make a statement that they have resources to burn. That cool and awesome are worth more to them than efficiency.

They build a Dyson swarm for the same reason they keep building better weapons. To intimidate their enemies. To say 'Look at this. Do you really want to mess with us when we have this?'

  • 2
    $\begingroup$ Resorting to psychology, data from mankind is of very limited value as aliens would be ... alien. We have no idea whether what we call psychology is a function of any characteristic in this universe or universal by nature. So we do not know whether conceptualizations like ''intimidation', 'challenge', 'warfare' are common amongst sentient/intelligent beings or a one-shot occurrence (in fact we do not even know whether civilization is singular). Imho this answer is very anthropocentric ( which applies to the question as well ). $\endgroup$ – collapsar Jun 21 '19 at 11:59

A Dyson sphere's warmth isn't about how hot the sun is, it is about how hot the civilization is. The fact it gets its heat from a star is immaterial -- replace the star with pocket fusion reactors within the 1 AU sized shell, and you get the same thermal profile for the structure as a whole.

Using Energy is generating Entropy, and Entropy leaks as Heat carried by Energy. What matters is how "big" the civilization is (in surface area) and how much Energy it uses.

We are currently small enough (surface of a planet) and poor enough (0.7 K-scale) that we don't show up thermally really. But as we upgrade, unless we inflate, we'll show up as a hot spot.

That sky survey looked for hot spots. There is no Dyson-sphere scale civilization, in surface area and energy budget, within 1 kiloparsec. It doesn't matter if the energy comes from Fusion or from a Star.

For a civilization to be within that range, it has to be significantly poorer than a star's energy budget, or significantly larger in surface area than a 1 AU ball.


It's not about the power, it's about the land-mass. The vast majority of a planet is unusable materials deep underground. If you strip-mined the Earth all the way down to the core, you could stretch it out into a dyson structure that would have hundreds of millions of times the surface area of a spherical planet.


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.