12

A popular misconception in the layman public is that the Big Bang was some sort "explosion" at a single point of space, where originally all matter was concentrated and then it "exploded" outwards. This is of course different from the modern general-relativity understanding of reality, which is that it is space itself which expands - not the content of the space moving, and the Big Bang did not start at a single point, but everywhere.

My question is - what experimental evidence do we have that can convince people that the explosion model can't be right.

Note that I'm not asking why GR guarantees that the space-expanding is the correct model, not the "explosion". I know that. I also know that GR has a lot of experimental evidence for its correctness at least in smaller scales. Rather I'm asking which evidence we have from astronomy, CMB measurements, or whatever, showing directly that the "explosion" model simply cannot be a valid explanation of the universe's history.

Qmechanic
  • 201,751
Nadav Har'El
  • 2,614
  • 4
    Note to potential answerers: inferred faster-than-light recession rates are not evidence against an "explosion from a point" model. Neither is the fact that the observable universe is larger than the age of the universe times the speed of light. Both of these happen in a very simple "explosion from a point" model. – Sten Jun 13 '23 at 09:11
  • 5
    As a more general point, I don't think it's possible to say the Universe did not originate from something like a point, because there are indeed viable models where it did (namely in the context of eternal inflation, where the Universe is taken to be our own non-inflating bubble). And I should note also that this is not in conflict with the GR description. "Expanding space" shouldn't be taken too literally; the difference between that and "everything is flying apart" is just a coordinate transformation. – Sten Jun 13 '23 at 09:16
  • 3
    Why the insistence on experimental evidence? (I'm smelling religious resistance, in which case no amount of evidence will suffice: if you want to believe, you'll just keep moving the goalposts on the required amount of evidence.) – RonJohn Jun 14 '23 at 05:38
  • 1
    No religious issues at all :-) I'm asking about experimental (or observational) evidence because I know how to explain space expansion using GR (the Friedmann equations, and so on), the GR model works well - but I started thinking how we know the "explosion" model doesn't work well also. For example Susskind's "the theoretical minimum" video lectures on cosmology start by looking at expansion of the universe as traditional movement in space, without GR, and he gets pretty far with this analogy. – Nadav Har'El Jun 14 '23 at 06:46
  • 1
    A rubber sheet is a 2D surface embedded in flat 3D space. However, the theory of GR is a theory of intrinsic curvature: our 4D spacetime isn't some "sheet" inside a higher dimensional space. Instead, we define curvature by how geodesics converge/diverge (tidal forces and sums of triangle angles). There is no "outside" space, so it loses meaning to say the big bang exploded "from" anywhere. – Kevin Kostlan Jun 14 '23 at 07:34
  • @KevinKostlan There are very general embedding theorems that allow any sufficiently smooth manifold to be embedded in a sufficiently high dimensional flat space. In this case it's at least ten dimensions, I believe (which "magically" also appears in string theory, if I am not mistaken). This is the general relativistic version of flat earth... the desperate (and uncalled for) need to explain everything in terminology that was borrowed from Euclid. The natural way of explaining the universe is, of course, in terms of itself, in which case there is nothing "more straight" than a geodesic. – FlatterMann Jun 14 '23 at 08:04
  • 1
    @NadavHar'El We have to explain more than just the Hubble flow in cosmology. Primordial isotope concentrations, for instance, depend on the thermodynamics of the early universe. Now a "simple explosion" won't do. We have to make sure that "the cake bakes for just the right amount of time at the right temperature". I am sure there is a way to fit an explosion model just the right way... but can it be done with a six parameter fit from first principles like in the current concordance cosmology? I would doubt that until somebody proves me wrong. – FlatterMann Jun 14 '23 at 08:09
  • @Sten I thought expanding space implies that objects are not flying apart, but rather the space itself is expanding, which would mean that eventually it could (unless the expansion stops) expand so much that it overpowers other forces - imagine that the expansion happens so fast that the distance between two atoms grows faster than they could attract to each other. It could also mean that the speed of light was relatively higher in a "tighter" universe. – rus9384 Jun 14 '23 at 08:31
  • @FlatterMann Primordial element abundances just require that (1) the universe was radiation dominated and (2) a particular density of asymmetric baryons. (1) is automatic if the "explosion" temperature exceeds a few MeV. (2) needs tuning, but not more than we need anyway. More generally it's accurate that you would need to fine-tune the "explosion" to get the Universe we observe -- but is that worse than the fine tuning that we need anyway? Indeed if the "explosion" were followed by inflation, it's not clear to me that there is any difference between this and the conventional paradigm. – Sten Jun 14 '23 at 09:57
  • Fundamentally inflation is supposed to erase any evidence of whatever arbitrarily messy configuration was present beforehand. That should apply to an explosion too. – Sten Jun 14 '23 at 10:01
  • @rus9384 It's a common misconception that expansion of space is a fundamental physical phenomenon. See my answer to "Does space expand?". I also have links to further reading in my answer to "How do we know we're not getting bigger?". – Sten Jun 14 '23 at 10:14
  • @rus9384 Further accurate answers about expanding space on Stack Exchange include https://physics.stackexchange.com/a/674077/180843 https://physics.stackexchange.com/a/719453/180843 https://physics.stackexchange.com/a/601323/180843 – Sten Jun 14 '23 at 10:23
  • @Sten I don't see any fine-tuning in the Lambda-CDM model, certainly not at a level that tunes the different eras. The parameters are self-consistent fits to the data (i.e. measurements) and the physical assumptions about cross sections etc. are all taken from experiments that were made 13 billion years after the fact. The universe seems to have stayed true to itself over this stretch of time. The only "fine-tuning" that you will find is in inflation models, which I reject for exactly the same reason. They don't explain anything. – FlatterMann Jun 14 '23 at 17:31
  • @FlatterMann Primordial spectral amplitude of $A_s=2\times 10^{-9}$ and spectral index of $n_s-1=-0.04$. Baryon-to-photon ratio of $10^{-9}$. $6$ times as much dark as baryonic matter. Dark energy density of $10^{-123}$ (in natural units). Where do these numbers come from? --- And what do you mean by tuning different eras? With an explosion, you wouldn't need to tune different eras. You'd be dealing with the same problem of setting the initial conditions as you have otherwise. – Sten Jun 14 '23 at 17:47
  • @Sten From measurements. The only criticism you can have here is that the model might be wrong, but then you have to present a model that uses even fewer parameters and fits even better. Explosion doesn't do that. It's a bad name anyway. What you are talking about is more like a "matter fountain". We have a modal for that: a black hole in its final death throes. The problem with that is that they go out in a whimper. The total amount of energy that gets released "explosively" is more on the order of a large nuclear weapon than that of a universe. – FlatterMann Jun 14 '23 at 18:02
  • @FlatterMann I'm confused then. If you're fine with tuning a "standard Big Bang" (whatever that is) to reproduce measurements, what's the issue with tuning an explosion? Or do I misunderstand you? The discussion started with primordial element abundances. That's set by the initial baryon-to-photon ratio. Whatever mechanism one proposes needs to specify that. – Sten Jun 14 '23 at 18:04
  • @Sten That's the point. We aren't fine-tuning. We are simply taking the standard theory that was confirmed at the scale of the solar system (Mercury perihelion and planetary radar data) and neutron stars and black hole mergers, we are extrapolating it to the entire universe and see if it can explain what we see with very few free parameters that are all being measured by the model. It does very well. Unless you can perform that trick with another model, the orthodox thing to do is to stick to Lamda-CDM. Structurally the far better model than explosion is CCC, anyway. – FlatterMann Jun 14 '23 at 18:09
  • @FlatterMann LCDM doesn't specify the origin though. It's compatible with a range of scenarios. You could make an "explosion" origin reproduce LCDM. (To be clear, by "explosion" I again mean universe originating at a point within a larger space.) Again it's unclear to me what others take to be so special about the explosion origin, compared to others. – Sten Jun 14 '23 at 18:15
  • Let us continue this discussion in chat. – FlatterMann Jun 15 '23 at 00:17

5 Answers5

14

I am not going to add to the conventional list of differences between standard cosmology and the explosion picture, but I will say a few things about the general problem with these kinds of discussions in the public.

The first fundamental problem with the question is that it contains at least two fallacies. One is the false dichotomy fallacy that there are only two models and that therefor one of them has to be right and the other one has to be wrong. That is completely false. We simply do not know what the universe really looks like outside of the observable universe, hence any statement about its shape is local and therefor incomplete. If we take this limit to our knowledge seriously, then it should be obvious that we have to allow for an infinite variety of cosmological models that may or may not turn out to fall into some broad equivalence classes upon further analysis.

I don't know of anybody who has done a logically self-consistent analysis of the global situation so far (and I doubt we have nearly enough knowledge about the necessary theory, yet, to even perform such an analysis), hence I can't tell you what the result of such a broader approach would be. We can be assured, though, that it contains far more than just these two trivial cases.

In all likelihood we can assume that even the local expansion model is merely an approximation, so even if you manage to convince the layman that "expansion it is", did you really educate them about proper scientific thought or did you merely enforce an orthodoxy that stems from a rather limited view of the problem that was predominantly popular in the 1960s and maybe 1970s? I would suggest it's the latter. Personally I do not believe that this is a service either to the public or to science PR. It would be far better to educate them that cosmology is a discipline that keeps an open mind and that is aware of the approximative nature of statements about the cosmos. If we do that, then it becomes much easier to motivate active areas of research like inflation and more daring models like cyclical conformal cosmology etc..

The second major fallacy is the implicit assumption that both you and the layman understand the dynamics of explosions and that only the layman is struggling with the phenomenology of expansions. I would instead claim that neither of you know enough about the dynamics of explosions and that this lack of knowledge also applies to the majority of cosmologists. Those who do know more are most likely the ones who are also actively working on supernova physics and they happen to know that the phenomenology of real explosion is neither homogeneous nor isotropic. It is a turbulent and messy process that leaves a complicated aftermath. In other words: the cosmic explosion imagery is mostly based on a cartoon version of an explosion rather than on physically observed processes that show instabilities and asymmetries. In a serious scientific context explosion cosmology would therefor have to explain the detailed mechanism by which this special explosion would have yielded a CMB with a homogeneity of one part in 100,000 when no other explosion ever observed is even remotely that "smooth".

Incidentally this problem also exists for expansion and it is, to the best of my knowledge, still not solved satisfactorily. Inflation is one attempt to solve it, but you can find serious arguments in the literature (at least serious in my eyes as a cosmology layman) that indicate that inflation might actually be the wrong approach entirely. To me this is yet another reason to keep an open mind and to explain the problems in cosmology ("What are the observations that we have to explain?") rather than to force a toy model on the public that may or may not explain these observations well enough.

You are, by the way, not the only one who struggles with physics communication. Take Steven Weinberg, who is probably one of the smartest physicists ever. He wrote a beautiful book called "The First Three Minutes" that, to a large extent, gets most of this right... but it ultimately turned out to be wrong about the big picture completely. Weinberg preferred, if I remember correctly, a cyclical big bang model in that book. That model has been completely ruled out by now.

So what are you supposed to do? I would say, try to keep an open mind and motivate to your audience why and how you are doing that. That, if anything, is the best approach to science IMHO. We are trying to get answers rather than trying to give them.

FlatterMann
  • 2,968
  • 1
    A fundamental problem with trying to analyze things like the Big Bang is that physical laws describe an idealized version of the Universe which is generally close enough to reality to make reasonably accurate predictions in most typical situations, but sometimes fail to predict how things will behave in atypical situations. Given that the Big Bang would be about the most atypical situation imaginable, it would hardly be surprising if estimates of how things would behave are so far off as to make any estimates about initial conditions meaningless. – supercat Jun 13 '23 at 22:30
  • @supercat Even the assumption that the big bang has special initial conditions is already problematic. Imagine that we are out at sea in a storm. Does the storm end at the horizon? No. But is there blue sky somewhere else? Obviously. We just can't see the remainder of the ocean. That is what the universe is: a mostly invisible ocean. To make special assumptions about the "beginning" of "our" slice of the visible universe is a very 20th century way of thinking about it, already. It's stuck between horizons and false choices. We need to cultivate longer vision, even in lay people. – FlatterMann Jun 13 '23 at 23:01
  • It didn't read to me like they were saying the Big Bang necessarily had special initial conditions -- more that our current understanding of physical laws is woefully incomplete, and thus we can't accurately model such a massive and massively complex series of interactions. I think that meshes quite well with your (excellent) answer. – Matthew Read Jun 14 '23 at 15:55
  • 2
    I guess to use an analogy, trying to use the existing laws of physics to analyze the big bang may be like trying to use Boyle's gas laws to predict how gases will behave at extremely higher pressures or lower temperatures than have been observed. Trying to predict how CO2 would behave at 2,000psi when all experiments involving the substance have involved pressures below 200psi. – supercat Jun 14 '23 at 16:26
  • 1
    @MatthewRead There are people who like to call our universe "fine tuned", which is what I was referring to. The cheap solution to that is the anthropic principle, which I dislike a lot because it has no explanatory power. I agree with you. We are still very far away from understanding what is really going on. I mentioned Weinberg because I grew up with a different cosmology than the one we have today. I have witnessed the difference that additional data can make to our understanding of the universe. I expect at least one more fundamental shift in the field before we "get it right". – FlatterMann Jun 14 '23 at 17:22
  • @supercat "Multiply it by infinity, and take it to the depth of forever, and you will still have barely a glimpse of what I'm talking about.". If we could attribute these words from the movie "Joe Black" to nature, then maybe that is what she would be telling us. The physical conditions are certainly far beyond my imagination, even though I understand exponential notation just fine. – FlatterMann Jun 14 '23 at 17:26
5

The evidence is in the distribution of velocity and the distribution of matter as a function of position, evolving over time.

The observed distribution (at the large scale) is like a swelling loaf of bread or a solid undergoing thermal expansion: each part moves away from its neighbouring parts with the same relative speed, for a given separation at any given time, no matter which region you examine. Also the matter is distributed uniformly over space at any given time.

For an explosion into space from a point, the region outside the expanding ball of debris is empty, and the velocity distribution within the ball of debris is typically not like the one outlined in the previous paragraph. So this differs from what is observed in both these respects.

The empirical evidence has not so far ruled out that the universe may have a small positive average curvature and a finite volume. It might also have a finite volume even if the average curvature is zero or negative. These aspects are not known. But we do know that the mathematics of General Relativity allow these possibilities. I mention it because if the total volume of space is finite but changing with time, then clearly the space itself is changing, not just the stuff in it.

We cannot directly observe the farthest regions so in the end we don't know for sure that the matter content does not fall away to zero outside some region, but that seems less plausible and less simple than the standard scenario where space is filled up roughly uniformly.

Andrew Steane
  • 58,183
  • 3
    The velocity distribution is a tricky point. For example, if we simplify the "explosion from a point" by neglecting self-gravity, a particle is displaced today by an amount proportional to its velocity. That is, we get automatically the kind of expansion that we observe (namely Hubble's law). – Sten Jun 13 '23 at 10:57
2

If the Universe originated from a single event in spacetime, then that event lies in our causal past, and the worldline of every particle in the Universe crosses that event. Therefore, every particle's worldline would enter our causal past. We would in principle be able to observe every particle in the Universe.

Does every particle lie in our observable universe? Well, naturally we cannot make a definitive statement about whether there are particles that we can't observe. Also, there is a practical limit to how far we can see, because the early universe was opaque.

Still, we have established observationally that the Universe was dominated by radiation (relativistic particles) from a time of about 1 second up to a time of around 52000 years. We do not know what happened before 1 second, but the simplest assumption is that the Universe was also radiation dominated before then. If radiation domination indeed extends indefinitely into the past, then the "true" observable universe is only about 1 percent larger in radius than the portion of the universe that we can practically observe (which terminates at the surface of last scattering of the cosmic microwave background).

Therefore, if the Universe originated from a point, and if nothing else very exotic happened before a time of about 1 second, then we can currently observe about 99% of the distance to the edge of the Universe's mass distribution. That is not really plausible; given that we see no large-scale inhomogeneity, it would imply an incredibly specific distribution of initial momenta (to produce a mass distribution that is uniform right up to the edge) and that we reside within 1% of the true center of the Universe.

Of course, maybe something else exotic did happen (e.g. inflation) that would lead to our "practically observable" universe being much smaller than our "observable in principle" universe (defined by the particle horizon).

So to conclude, we cannot say for certain that the Big Bang was not an explosion in space. But we can be pretty sure that it was not just an explosion in space.

Sten
  • 4,851
  • 1
  • 13
  • 30
  • 1
    "We have established ... up to a time of around 52000 years"-- of course, the relevant evidence (primordial abundances, CMB anisotropies, and lss) is compelling, but it's also indirect. My understanding is that the inference relies on the dynamics of GR's perturbative FLRW cosmology. Meanwhile, the "exploding universe" seems to be more of a vague idea than a specific and complete cosmological model. Am I mistaken here, and do you know of any references discussing such inferences in that context? Are the dynamics of the radiation (even its density scaling) actually constrained in this picture? – jawheele Jun 14 '23 at 15:58
  • @jawheele I'll just clarify that I'm not viewing an explosion as an alternative to the GR description, because I think that's a mistaken premise (as noted in the comments on the question). I accept perturbative FLRW as applicable because (1) we find GR to be accurate and (2) we observe homogeneity. Not to dismiss your concern -- it's certainly fair to ask how theory-laden an inference is. – Sten Jun 14 '23 at 16:13
  • Why would the radiation era extend indefinitely into the past? What does "past" and the kind of world line causality that applies to a transparent universe with strong energy density gradients (clocks) even mean in a system that does not have such local clocks and in which "communication" is blocked by the background? Strictly speaking none of the ideas of relativity apply in such a scenario. We are simply using familiar language to describe a causally unfamiliar situation. The conceptual problems here run very, very deep. – FlatterMann Jun 14 '23 at 18:13
  • @FlatterMann No reason other than that it's the simplest assumption. The point is that if you suggest that a localized explosion created the hot plasma of the early universe, you might run into issues. But through more creative scenarios, there's not necessarily any issue with a localized origin. – Sten Jun 14 '23 at 18:19
  • 1
    (And relativity of course still applies in situations where light can't stream freely. Why wouldn't it? Do you argue that relativity does not apply within opaque everyday objects? Regardless, neutrinos and gravitational waves can still stream freely through the plasma and can carry causal information.) – Sten Jun 14 '23 at 18:27
  • Relativity doesn't apply because now there is a special background. We are quite familiar with that situation: the acoustic Doppler effect is different from the relativistic Doppler effect, for instance. – FlatterMann Jun 15 '23 at 00:11
  • @FlatterMann Relativity is valid irrespective of what field configurations are present. It's the very purpose of general relativity to account for how the structure of spacetime is connected to the field configurations. (I assume we're still talking about mainstream physics.) – Sten Jun 15 '23 at 00:31
  • Two systems that interact through acoustic waves do not follow the laws of relativity, i.e. the physics depends on the relative velocities to the background and not just the relative distance and velocity between the systems. We also don't have energy and momentum conservation because there are constant exchanges with the background. Things in motion don't stay in motion. It becomes Aristotelian rather than Newtonian/Galilean physics. – FlatterMann Jun 15 '23 at 00:53
  • @FlatterMann They do follow the laws of relativity. Are you suggesting that they don't follow some acoustic version of relativity? That's certainly true, but they still obey relativity itself. Relativity is just about the geometrical structure of spacetime; it's not fundamentally about light propagation. (The "speed of light" is a geometrical constant, not actually a function of what the light is doing.) – Sten Jun 15 '23 at 00:57
  • Let us continue this discussion in chat. – FlatterMann Jun 15 '23 at 00:58
1

There can be several approaches to this. But my point would be to compare Hubble law of receding galaxies ("debris of explosion") and true explosion in a spherical charge, like grenade covered by Gurney equations.

Hubble law states, that galaxy receding speed is :

$$ v = H_0~D \tag 1,$$

while grenade shell receding initial velocity is (assuming charge mass is a lot greater than shell mass) :

$$ v_0 \approx 1.3 ~\mathcal G \tag 2,$$

Where $\mathcal G$ is the Gurney constant for a given explosive. It depends on how effective explosive is and is related to a potential energy per mass unit of explosive which gets converted into shell kinetic energy.

Main differences between those two explosions is that grenade shell after acquiring initial maximum speed $v_0$,- starts to loose kinetic energy with the distance covered in the environment due to air drag force and similar reasons. Galaxy, on the contrary as (1) shows, with the distance covered $D$ just increases it's speed. Hence, universe is in ongoing and accelerated "explosion", for which we don't even have an analogy in real-life explosions, because every of it ends fast enough and with decreasing shell speed over time.

Other also important difference is that grenade spatially has center of detonation, because shell particles which are closer to the center of explosion - will have greater initial receding speeds and particles which are further away form it,- will have lower escape speeds. Hence, if you would detect grenade debris speeds with slow-motion camera or other instrument - you could find out debris velocities gradient and following that,- find out coordinates $x,y,z$ where explosion in space has started. In contrary, universe has no such property,- there is no "special" direction in outer space where galaxies would not commit to Hubble law (except some disturbances from local galaxy group gravitation, which may "override" global law). Or if you like - whatever galaxy habitant will deduce the same Hubble law and some may try to falsely make a conclusion that everything is receding from them. Hence, every point in the universe is "center of BigBang" and has participated in the early universe all historical events.

Agnius Vasiliauskas
  • 13,467
  • When I mentioned "explosion" I didn't mean literally an explosion of an explosive, and obviously there wouldn't be any air drag of any of the "exploding" matter. I had in mind more a point with a huge amount of matter with random velocity distribution (due to quantum fluctuations), leading some matter to fly out faster than other matter, or something like that. – Nadav Har'El Jun 13 '23 at 13:43
  • 2
    IMHO, matter explosion detonated from a single point in space, does not differ much from a spherical shell explosion in explosives, hence my analysis should be valid as well for denying such analogy as you wanted. If there's no drag or any resistance,- matter should fly apart with constant speed,- not dependent on distance covered. This is different from Hubble law, which states galaxies speed gradient. So analysis still valid. – Agnius Vasiliauskas Jun 13 '23 at 13:57
  • 4
    If gravity is neglected, Hubble's law necessarily holds when the matter originates from a point, no matter what the velocity distribution is. This is because particles' present-day positions are set by their velocities. What would happen instead, if you had a highly concentrated distribution of initial speeds, is that the matter today would be spatially concentrated in a ring around the initial point, as opposed to the homogeneous distribution that we observe. (Observers in the ring would see an inhomogeneous distribution that nevertheless obeys Hubble's law.) – Sten Jun 13 '23 at 14:02
  • 1
    Sorry, hollow shell, not ring. – Sten Jun 13 '23 at 14:08
  • Agree about hollow shell. Similar debris inhomogeneous distribution should be found on the ground after grenade or any spherical charge explosion. Hence, it also could be used for disproving shell model of universe "explosion". – Agnius Vasiliauskas Jun 13 '23 at 14:40
1

The CMB is the best experimental evidence we have that there was not an explosion, but rather an expansion of spacetime. It is mostly isotropic: it comes from all directions at the same time.

Not in your question, but it also helps to imagine what such an "explosion" would look like: a scattering of mass from a central point. This does not make a lot of sense: what was there in that point before? What remnants would remain of that point? Why would there be a preferred location? The absence of any clustering of galaxies around a central point (as far as we can see) is also experimental evidence that it was space itself that was expanding, not just the mass contained.

alexfernandez
  • 119