Why does the universe expand?

Five hundred years ago, Nicolaus Copernicus proposed that the Earth might be one of several planets orbiting the Sun, rather than the centre of the universe. He compared the geocentric model to a monstrous form assembled from parts of different bodies, like the Creature Mary Shelley brought to life three centuries later in Frankenstein — each part appearing human on its own, but as a whole a grotesque patchwork. The geocentric model was built to directly match the sky: where a planet paused against the background stars and looped into retrograde, a dial was added so the planet would pause in its orbit around the Earth and loop backwards a while before resuming its normal course. In contrast, Copernicus recognised that the outer planets — Mars, Jupiter, and Saturn were known at the time — might enter retrograde loops due to parallax, their apparent positions shifting relative to the background stars as our orbit brings us near and then we pass them on our way around the Sun. Copernicus had no idea of the physics that Isaac Newton or Albert Einstein would eventually use to explain planetary motion. Nor did he imagine this motion resulted from the same phenomenon that causes apples to fall from trees, or paths of light to bend around the Sun. His model ended up with as many knobs and dials as the geocentric system due to his use of circles rather than ellipses to describe orbits. Even so, Copernicus recognised that a Sun-centred theory afforded the possibility that it might eventually explain why phenomena like retrograde motion should appear as they do to us, despite the planets’ motion being continuously in one direction only. And in doing so, he paved the way for others like Newton and Einstein, who later fleshed out both the underlying concepts and formal mathematical descriptions of a solar system in which planetary motion is expected to appear with all the complexity we observe. In hindsight, the discovery Copernicus’s proposal prompted was that broken symmetries — our off-centre perspective from a planet orbiting the Sun, and the non-uniform motions of all planets including ours — would complicate appearances within an ontological framework that is nonetheless simpler. In a similar sense, it may be argued that the standard cosmological model today — which gives an accurate description of phenomena but is nevertheless an amalgam of ad hoc patches, each inserted to unnaturally force the evolution of a universe that is otherwise expected to be different from the way it appears — bears a closer resemblance to Frankenstein’s monster than it does the simple explanations of planetary motion given by Newton and Einstein. For despite all the dials and knobs that have been added to ensure the standard model does directly resemble appearances, after a century of development it still affords no explanation of why our universe should be expected to expand, as it appears to do. Why should our universe expand? In the Copernican tradition, we ought to ask why our universe should expand. The standard cosmological model affords no such explanation. It is based on a principle, famously promoted by Einstein together with his colleague Willem de Sitter, that characterises the universe as expanding in spite of a tendency to decelerate because it is filled with everything we see. This is important: according to the basic Einstein-de Sitter framework for describing cosmic expansion, all the galaxies and light we see across the universe are thought to work against the universe’s expansion, slowing it down; and anything driving expansion is an ad hoc dial we’ve added so that base model fits appearances better than it naturally should. In fact, this tendency for light and matter to slow cosmic expansion mathematically blows up to an infinite amount at the Big Bang. Therefore, the model’s only “explanation” for why our universe even could be expanding today is that it began with such a tremendous rate that momentum carried it against its natural tendency towards the opposite. For this reason, a century ago British astronomer Arthur Stanley Eddington complained of the Einstein-de Sitter model that would dominate twentieth century cosmology, “One cannot deny the possibility, but it is difficult to see what mental satisfaction such a theory is supposed to afford.” Much like its Ptolemaic predecessor, this model has since been augmented with various features allowing it to fit the data with impressive accuracy. First, there is an inflationary epoch, thought to have occurred a moment after the Big Bang, which would drive a fleeting period of exponential expansion and erase several tensions the Einstein-de Sitter model otherwise leaves unresolved — though leaving the initial expansion problem untouched. Then for a long while the universe is thought to have decelerated, its slowing rate driven primarily by radiation in the early universe followed later by matter, in good alignment with Einstein-de Sitter. Finally, after several billion years a component we’ve come to call dark energy, which does tend to drive expansion, is thought to have become significant enough that the expansion rate eventually began to accelerate. In comparison with the Ptolemaic model, both inflation and dark energy are similar to the eccentric and equant: later patches, added to a universe filled with the stuff we observe directly, that enables us to describe the universe we see in spite of the fact that without these patches the more basic physics suggests the universe should not evolve as it appears to do. A bare Einstein-de Sitter universe should not appear the same in different regions of the sky that could not have interacted before the times we now see, due to the finite speed of light. And such a universe should not appear spatially flat, as ours appears to be. Inflation is the dial we use to fix both of these problems. An Einstein-de Sitter universe also should never come to expand at an accelerating rate as we’ve observed — let alone expand to start with. Dark energy fixes the former issue, but its effect is null at the Big Bang and only gradually becomes significant over billions of years, so it cannot explain why the universe should ever have expanded in the beginning. And inflation can only happen within an already existing, expanding universe — so invoking it as the primary cause would be tautology. More recently, two separate cracks have opened in the dark energy patch. There is a persistent and growing tension between the expansion rate measured from the early universe and the rate measured from the late universe, which a cosmological constant does not reconcile. And independently, large surveys of galaxy clustering and supernovae have been read as favouring a dark energy that weakens over time rather than holding steady. So astronomers have proposed evolving dark energy models, adding an evolution dial to a source of repulsion that was never explained in the first place. When an ad hoc patch needs its own ad hoc patch before a model that fundamentally abhors the phenomenon it is intended to describe can be brought in line, that should be a strong sign that the base model is wrong. Symmetry breaking and physics In light of the problems the standard cosmological model has with reconciling the evidence and providing an explanation for the world we seem to live in — the growing number of knobs and dials to recover a phenomenologically accurate description of a universe that is nonetheless fundamentally expected to be different — we ought to take a page from Copernicus and ask what symmetries we may be assuming are fundamental, which our reality may in fact essentially break. We should ask what appearances we see that may not directly represent the world that is, but which may instead only appear as such because our place in the universe is not central, so our perspective is owed in part to a broken symmetry. This is not idle speculation. In essence, physics is an exercise in recognising the various forms in which symmetries are broken in nature. By this, I mean generally any phenomenon that removes a degree of symmetry within the natural world. For example, because the Sun spins, it is wider around its equator. The Sun breaks one dimension of symmetry by spinning around an axis, and causes an equatorial bulge we can measure. And by measuring the Sun’s rotational rate and the size of its equatorial bulge, we can estimate other physical properties that are more difficult to measure, like its density profile. Or imagine that the Earth was at the centre of everything, and we were orbited only by the Sun which moved in a perfect circle around us, and that the sky was perfectly uniform with no randomly scattered stars across it. In this case, the only broken symmetry we could reference in our sky would be the Sun itself. We would still have day and night. The sky would still be brighter as we look closer to the Sun. In this case, we could build a model to describe the Sun as orbiting the Earth once a day, at a fixed distance from Earth. But with no other symmetry breaking to worry about, we could equivalently describe the Earth as spinning around once a day while the Sun remains fixed in place. In fact, if all else were the same but it was really the Earth orbiting a fixed Sun, still in a perfect circle, we could not tell the difference from the moving Sun picture. Due to unbroken symmetry, either description could be used regardless of what is really going on.