Why is the night sky not bright?

Our recent publication of the article Crisis in Cosmology continues to stimulate debate and comment. Here we have a reader from Italy commenting on the Olbers paradox which states that if galaxies and stars had existed for an infinite amount of time then the whole sky would look bright. So why is the night sky dark? Harry Nielsen comments.

Our recent publication of the article Crisis in Cosmology continues to stimulate debate and comment. Here we have a reader from Italy commenting on the Olbers paradox which states that if galaxies and stars had existed for an infinite amount of time then the whole sky would look bright. So why is the night sky dark? Harry Nielsen comments.

Dear comrades,

I agree with Harry Nielsen's reply to some criticisms of Alex Nichols on the article “Crisis in Cosmology”. But I feel there is one point that does deserve a more decisive counter-argument: the Olbers' paradox.

Heinrich Olbers stated that if galaxies and stars had existed for an infinite amount of time then the whole sky would look bright, not dark [at night]. Heinrich Olbers was right (at least if the Universe is not expanding fast enough and if it does not have a fractal shape).

Indeed, the Universe looks bright  just not as bright as the Sun, because the Sun and the other stars are much warmer than the average temperature of the Universe. Every point of the sky is bright, at the typical wavelength of the Cosmic Background Radiation. It's just a wavelength our eyes cannot see: "an observer having his eyes sensitive to the Planck radiation at 3 K would see that the night sky is bright, as expected by Heinrich Olbers." (Paul Marmet, The 3K Microwave Background and the Olbers Paradox).

Nevertheless, plasma cosmology does not state that galaxies and stars have existed for an infinite amount of time. Eric J Lerner explained in his famous book “The Big Bang Never Happened” that stars came into existence at a certain stage of the development of the (infinitely old) Universe, something like 20 billion years ago. The Universe has always existed; galaxies and stars have not.

"The light has been turned on" in the Universe with the start of the nuclear age (fusion inside the stars), but we can't see the light of stars more than 20 billion light-years away. When we look so far we just see what was happening “before” the nuclear age, and those were much darker ages dominated by gravitation and electromagnetism.

If the pace of physical reactions transforming energy is getting faster and faster (and not slower and slower as the Big Bangers postulate), looking around us we would not see an infinitely intense light coming from every point: we would see a lot of light coming from nearer light-emitting spots and darkness when we look at places farther away  everything covered by a halo of light resembling a 3K blackbody radiation, coming from the interstellar medium (mainly, plasma). Well, in fact that is what we actually see!

The infiniteness of the Universe in space and time is not paradoxical, particularly if we imagine it as an evolving universe, not a stationary one.

Mauro Vanetti, December 8, 2005

Pavia (Italy)

Could the universality of fractals in nature be a consequence of the dialectics of nature?

It's a very interesting comment from the comrade, and it's true that it could resolve Olbers' paradox - the question of why the night sky is not bright. Perhaps also what he suggests could have happened more than once, and in more than one place? If matter and hence time and space are infinite then the universe has already passed though an infinite number of states. Several of those states, an infinite number in fact, must also have had stars and galaxies that produced light. Following this line of thought, any event that has happened must in fact already have happened before, an infinite number of times, in an infinite number of places, and will happen again. The whole of history is pre-determined and we have no free will......

Here we've stumbled into a trap - the contradiction of infinity that philosophers and mathematicians have grappled with for thousands of years, from Zeno the Greek to the present day. A simple solution is to deny that the universe has existed for an infinite amount of time, as mainstream cosmologists have done. Ironically they draw this conclusion from the singularity in Einstein's equations - an infinity occupying zero time and space that supposedly exists at the origin of the universe. Then with suitable embellishments that adjust the theory to fit a few of the facts they claim to predict the development of the universe.

In another part of the physics department, perhaps across the university quadrangle, they would find physicists who have also faced the contradiction of infinity but with more success than the cosmologists. Through computer simulations these physicists have been able to show that complex non-linear systems - and a universe of infinite extent is a good candidate for this - can evolve and change for an infinite period of time without ever repeating. The earth's weather, for example, never repeats exactly. It is impossible to predict but it is not random; effect follows cause in every part of the system and in theory for infinite time. As it changes it can come close to states it has been in before; over an infinite period of time it may come infinitesimally close, but it will never repeat.

Physicists draw diagrams of “phase space” to analyse this behaviour. These are graphs that show the different states the system goes through; the path that the system goes along in phase space describes how the system is changing over time. In complex systems the patterns made by these paths are fractals - there is more and more detail at finer and finer scales. The system may come infinitesimally close to a position in phase space that it has been at previously but it will never come to exactly the same position. The universe can evolve for an infinite period of time but it will never repeat. History is not pre-determined, and is not determined mechanically at all, but is determined dialectically as a result of the many interactions between earth's billions of inhabitants. But it is not random; the underlying physical process is the development of production, now constrained and held back by the capitalist form of production, but it is this which pushes society forward in the same way that the sun's energy drives the weather.

It's possible, therefore, that the lights could have come on in the universe 20 billion or so years ago, as the comrade suggests, and prior to that time there was no light. As he says, this is consistent with a view of the universe that is infinite in time but which is evolving and not static. When we look far away we see what was happening earlier, and, as he says, it's possible “those were much darker ages dominated by gravitation and electromagnetism”. Perhaps there was a “change of state” on a cosmic scale, analogous to the change that occurs when ice forms from water, and the stars and galaxies then appeared.

But when we are trying to understand what has happened in the universe, we need to stay close to the evidence, and to what we know, if we are to avoid writing science fiction. Other than Olbers' paradox perhaps, is there any evidence that this is what has happened? Other than the idea of an infinite but continually evolving universe is there any known physics (meaning physics that has been established and tested) that could explain how or why this might happen? The idea does not seem to sit too well with Hannes Alfven's view that we should first try to use the physics that we know in order to explain the universe before inventing new physics. And is it a co-incidence that we are almost exactly repeating the words of the Christian myth of creation - “let there be light”?

The idea that the paradox can be resolved by a fractal distribution of the stars is discussed by the mathematician Benoit Mandelbrot in his book “The Fractal Geometry of Nature”. Since Mandelbrot's work in the 1960's there has been an increasing awareness of how prevalent fractals are in nature, and in society. They can describe rock layers in the earth, leaves on trees, river beds, income distributions, even the coast of Britain. (A famous paper from Mandelbrot is titled “How long is the coast of Britain?”, and he concludes that because coastlines are fractal the coast of Britain is infinitely long. Having walked part of the coastal path I can believe it.)

If the stars and galaxies have fractal geometry then we could expect clusters of galaxies, then clusters of clusters and so on. Similarly there would be voids in which there would be almost no stars or galaxies, some of which could come together to make bigger voids and then still bigger voids and so on. Mandelbrot points out that in that situation it is possible to find “a large proportion of directions that go to infinity without encountering any star”. So the sky will be dark at night.

The plasma physicists say that scattering of light by plasma clouds causes the background radiation. Strictly speaking this is different from the subject of Olbers' paradox, which talks about light from stars and not light from scattering. Supporters of the Big Bang would still argue though that the observed level of the radiation implies that the universe cannot be infinite in time because (quoting from Alex Nichol's letter) “the effect of this endless scattering of light, x-rays and infra-red radiation would be a much higher background temperature than is evident in the CMB today.” 

Scattering by fractals (infinite scattering by infinite numbers of discontinuities at all scales) is a difficult topic and, as was mentioned in the reply to Alex Nichols, it has only recently been looked at in fields such as solid-state physics or geophysics where there are similar scattering problems. There does not seem to be anything in the scientific literature about Olbers' paradox and scattering by fractal clouds of plasma. Given the nature of fractals, it seems possible that a similar statement might apply in that case for the scattered light as for the direct light, but this is only a guess.

Possible support for the fractal idea as a way of resolving the paradox is that the universality of fractals in nature could be a consequence of the dialectics of nature. The scaling behaviour of fractals in nature, which links processes at many different scales, must be a symptom of the underlying general laws of change and motion - a symptom of the universality of the dialectical laws of change and development. Widely different processes can occur at different scales (the erosion of a small pebble from the sea-shore compared to the formation of bays and deltas), but a common geometry emerges because of the common dialectical behaviour.

Take the rocks in a cliff face for example. From a kilometre away it is possible to see a number of rock layers, and walking towards the cliff more layers will become visible. But the pattern of the layering (which could be quantified using suitable measurements) is the same at all scales, from the very large scale when viewing the layering at a large distance down to the scale of a few millimetres when close-up. (This is a controversial subject amongst geologists but does appear to be supported by the evidence.)  How can this happen, when the small scale layering in the rocks was produced by completely different physical processes compared to the large layers?

This is another example of the complex dynamical problems that physicists have begun to study in the last 20 years or so. What the physicists are finding is what Hegel discovered several centuries ago - that changes in nature occur according to the general dialectical laws of motion. The small layers in the cliff face might be caused by changes in sedimentation rates; tectonic processes might cause the larger layers. But these different processes all produce layering with similar patterns because all these processes obey common dialectical laws of motion. Punctuated equilibria occur - at all scales. There is an accumulation of quantitative changes that produces sudden qualitative changes - at all scales. Necessity is expressed by chance - at all scales.  

This is a paragraph from the article “Crisis in Cosmology” that mentions this:

“The reality of most physical phenomena is that there are many interlinked features, all of which interact. And increasingly physicists are beginning to find that it is the interaction that is more important than the details of the physical laws. Transitions occur that are determined by the complexity itself regardless of the details of the physics; this is why a fractal model can work over many scales and different physical mechanisms. The physical axioms and their deductions not only become inadequate, they become irrelevant. More general laws emerge - the dialectical laws of quantity and quality, of the union of opposites and the negation of the negation.”

Having said all this, there is only a limited amount of evidence that the universe might in fact have a fractal structure. There were some presentations at the “Crisis in Cosmology” conference on this topic, but the results were not convincing. Time will tell.


Harry Nielsen

Appendix - Olbers' paradox

Olbers' paradox says that if the stars are uniformly distributed the sky will be lit nearly uniformly, day and the night, to a brightness similar to that of the Sun:

“When the light emitted by a star is proportional to its surface area, the amount of light reaching an observer at a distance R is proportional to 1/R2, but the star's apparent surface is itself proportional to 1/R2. Thus, the apparent ratio of light to spherical angle is independent of R. Also when the distribution of stars in the universe is uniform, almost any direction intersects some star. Therefore, the sky is uniformly bright, and seems ablaze. (The moons disc would form an exceptional dark domain, at least in the absence of atmospheric diffusion.)” From Mandelbrot, “The Fractal Geometry of Nature”.

The quote the comrade gives from the article by Paul Marmet appears to be wrong in saying that because of the cosmic background radiation “the night sky is bright as expected by Heinrich Olbers”. The background radiation in fact is much weaker than predicted by the paradox.

See also: