[does space expand or are things just moving?]

[speedfreek]

Is it "expansion" that seems a bad picture, or expanding "space"?

[Ken G]

It all seems to come down to a problem with the analogies, that's true, I don't deny that, but this seems to be related ultimately with expansion itself seeming a bad picture of the underlying theory, and if expansion IS an interpretation of the theory, why do we base so much of modern cosmology in an interpretation that originates so much diatribe?

The answer to that is that we do not base any of modern cosmology on that interpretation. Modern cosmology is quite firmly based on GR + cosmological principle + inferred initial conditions, with a healthy dose of dark matter and dark energy stirred in to make it all hang together as a workable hypothesis for the history of our universe. Not a shred of that relies in any way on an interpretation involving "space itself expanding", nor do any modern cosmology papers require that interpretation to be correct. That interpretation is nothing more than a pedagogical picture, useful for both communicating ideas to laymen, and also for a language between scientists. Experts may disagree on how well it accomplishes those pedagogical goals, but none of it has anything to do with the field of cosmology as a research enterprise.

[Staticman]

The answer to that is that we do not base any of modern cosmology on that interpretation. Modern cosmology is quite firmly based on GR + cosmological principle + inferred initial conditions, with a healthy dose of dark matter and dark energy stirred in to make it all hang together as a workable hypothesis for the history of our universe. Not a shred of that relies in any way on an interpretation involving "space itself expanding", nor do any modern cosmology papers require that interpretation to be correct. That interpretation is nothing more than a pedagogical picture, useful for both communicating ideas to laymen, and also for a language between scientists. Experts may disagree on how well it accomplishes those pedagogical goals, but none of it has anything to do with the field of cosmology as a research enterprise.

I differ. The whole LCDM rests on the interpretation of redshift as some sort of expansion of "something ", you choose your preferred analogy of what it is that is expanding ( if nothing expands there can be no expansion). Not only that but the inferred initial conditions (BBT) you talk about are also derived from that interpretation (basically, the CMB and helium nucleosynthesis too,but we can leave these side for the moment since they need the redshift interpretation anyway) even the cosmological principle was restricted from the generalized Copernican principle to accomodate the interpretation of redshift as "space expansion" since that is the only way you get homogenous space slices in a dynamical universe.
So from your 3 ingredients recipe, only GR equations are not determined by the interpretation of redshift as some undefined stuff expanding.
So it seems we are indeed basing at least some of modern cosmology on that interpretation don't you think?

[Ken G]

I differ. The whole LCDM rests on the interpretation of redshift as some sort of expansion of "something ", you choose your preferred analogy of what it is that is expanding ( if nothing expands there can be no expansion).

No, that is simply untrue, you have reversed the logic. The model does not depend on its interpretation, it's a model defined by a set of equations that make predictions and explain observations. You can interpret it any way you like, the model doesn't care-- we only need interpretations to help us apply the model, and to give us a warm fuzzy feeling we understand the model, but different people may find they apply the model better using different interpretations. This is quite a routine state of affairs in physics, it is the area known as pedagogy and it's quite separate from the actual model predictions. It is an important part of science though-- we do need good pedagogies, and it's worth discussing what is or is not a good pedagogy, as long as we don't mistake that for a discussion about the goodness of the model itself.

Not only that but the inferred initial conditions (BBT) you talk about are also derived from that interpretation (basically, the CMB and helium nucleosynthesis too,but we can leave these side for the moment since they need the redshift interpretation anyway) even the cosmological principle was restricted from the generalized Copernican principle to accomodate the interpretation of redshift as "space expansion" since that is the only way you get homogenous space slices in a dynamical universe.

That is equally untrue. The inferred initial condition is also not derived from any interpretation, that is simply demonstrably false. The initial condition is derived from the observations applied to the mathematics of the model, that is just where it comes from. Interpretation is totally unnecessary, it's only used later on to help give us a picture of how it all hangs together. That is clearly true-- if we have observations, and we have mathematics that lets us extrapolate those observations back in time to get an initial condition, what more do we need? Nothing-- we need nothing more than GR + cosmological principle + observations, that's it (OK, with some dark matter and dark energy thrown in, to get consistency), it gives us the necessary initial conditions right away. The cosmological principle is indeed necessary, but has nothing to do with interpretation of the Big Bang as any expansion of space, it is simply an application of Occam's Razor, which we will adhere to until we decide we can no longer get away with it, which is just exactly how Occam's Razor is always used in science.

So from your 3 ingredients recipe, only GR equations are not determined by the interpretation of redshift as some undefined stuff expanding.
So it seems we are indeed basing at least some of modern cosmology on that interpretation don't you think?

No, that is not interpretation of a model, it is the model itself-- and it doesn't say squat about expanding space, the topic of this thread. The cosmological principle doesn't assert that "space expands the same everywhere", it asserts that the model is spatially homogeneous at each epoch of cosmic time, that's all perfectly well defined and interpretation-free, and is all you need to apply the GR to get the dynamics.

ETA: last line added.

[Staticman]

You can deny the facts as much as you want, they won't change. That the standard cosmology is supported by two main pilars-first, one assumption not derived from observation nor from any physical law but from philosophical considerations wich is called the cosmological principle , and second, the interpretation of redshift as cosmological expansion, it can be found in any Cosmology book. That's not good or bad, is just the way it is.
When you talk about observations here you mean redshift, don't you? And you have to interpret it some way to draw conclusions from that observation, and to fit them in some model, right?
I can't believe I have to explain such obvious things. Equations are not simply applied in abstract, they need boundary conditions, assumptions, observations interpreted a certain way. You don't really believe in the naive picture of physics that you defend in your post, at least I've seen you criticize it in several threads, physics need interpretations in order to make a single prediction.

[Staticman]

Of course it can be argued that to get the FRW metric all one needs besides the GR equations is the cosmological principle as a boundary condition. That is formally true. If we look at the history of cosmology we can see that Friedmann in 1924, before the 1929 paper by Hubble which originated the modern interpretation of redshift, published a model of expanding universe, but the fact is that it was dismissed and forgotten for years, because at the time no interpretation of the redshift that had been observed by Vesto Slipher at least since 1912 had been agreed by the scientific community. It was after the Hubble paper that Robertson formulated the FRW metric the way we know it, and it was inmediately accepted, because by then we had not only the observation of redshift but an interpretatio of the observation that allowed for a expanding model to be taken seriously:the Hubble law.

[Ken G]

You can deny the facts as much as you want, they won't change. That the standard cosmology is supported by two main pilars-first, one assumption not derived from observation nor from any physical law but from philosophical considerations wich is called the cosmological principle , and second, the interpretation of redshift as cosmological expansion, it can be found in any Cosmology book.

All I can say is that you do not understand the reason that the cosmological principle is used. It has nothing to do with philosophy, other than the simple philosophy that defines what science is (Occam's Razor). Simply put, we use the cosmological principle because we can. Period. If you think there is any other reason, you are just wrong.

When you talk about observations here you mean redshift, don't you?

Certainly, but also laboratory measurements of the fusion of light elements, and laboratory measurements of the behavior of light intensity, and laboratory and astronomical measurements of how gravity works, how motion works, how magnetic and electrostatic forces work, how causal connections work, relativity, thermodynamics, quantum mechanics, and a few more. You get the idea-- most of the entire body of modern physical understanding is brought to bear somewhere in validating the basic consistency of the story told by GR + the cosmological principle + observations.

And you have to interpret it some way to draw conclusions from that observation, and to fit them in some model, right?

No, you don't need to interpret them, that's the entire point. You need to predict them, the model predicts. The model is just what I said it was: a set of equations, which connect observations to an initial condition, all purely with mathematics, and the predictions must all be consistent with each other. There's no interpretation there anywhere, there's no step that says "take the observations and believe they mean X, and see if you like X." No, there's only "take the observations X and compare them to the predictions of theory Y." Pick up any paper in the cosmological literature that is testing the model, and it will look like the latter, not the former. Papers that invoke interpretations are pedagogy papers, not model testing papers. Model testing papers certainly may invoke a pedagogy to give themselves a language to carry their point, but at no place in that paper will the pedagogy be essential, as any other pedagogical interpretation that applies to the same model will yield the exact same results, because it is always the model that is being tested by observation, never the interpretation. That's science.

Equations are not simply applied in abstract, they need boundary conditions, assumptions, observations interpreted a certain way.

Now you claim observations have to be interpreted in a certain way? If I observe a Lyman alpha line redshifted by a factor of 5, I need to interpret that? I can't just assert that I am clearly seeing a Lyman alpha line redshifted by a factor of 5? That's not what interpretation means in this thread, it means giving a meaning to the redshift, rather than simply noticing the redshift. Of course I have to interpret the numbers I see, I have to interpret that I have eyeballs that see, I have to interpret that I have an instrument that measures, I have to interpret that the whole experience if not a figment of my imagination. None of that has the least bit to do with interpreting the meaning of "expanding space," which is the actual subject of this thread.

You don't really believe in the naive picture of physics that you defend in your post, at least I've seen you criticize it in several threads, physics need interpretations in order to make a single prediction.

No, I never said that it any of my posts. Instead, I defended the need to find a good pedagogy, I supported a central role of philosophy in physics, and I attacked the claim that many make that they are "shut up and calculate" types. But none of that means we need to interpret a model to test that model-- all of that had to do with the importance of interpretations in guiding us when we need to change a model, or apply it in different contexts where various different interpretations might prove more facile. I have always been clear, in my own head anyway, of the difference between a prediction of a model and an interpretation of a model. They are as different as apples and oranges, that's really the only thing the "shut up and calculate" camp gets right.

[Ken G]

Of course it can be argued that to get the FRW metric all one needs besides the GR equations is the cosmological principle as a boundary condition. That is formally true. If we look at the history of cosmology we can see that Friedmann in 1924, before the 1929 paper by Hubble which originated the modern interpretation of redshift, published a model of expanding universe, but the fact is that it was dismissed and forgotten for years, because at the time no interpretation of the redshift that had been observed by Vesto Slipher at least since 1912 had been agreed by the scientific community. It was after the Hubble paper that Robertson formulated the FRW metric the way we know it, and it was inmediately accepted, because by then we had not only the observation of redshift but an interpretatio of the observation that allowed for a expanding model to be taken seriously:the Hubble law.

No, it was because we understood that we had a set of observations that the model predicted. That's it, that's what was not previously recognized. If anything, interpretation was the problem, it was the stumbling block-- old interpretations of what motion was like in the universe had to get out of the way. Interpretation is to be cautious of, that's more or less the point of this thread.

[Staticman]

No, it was because we understood that we had a set of observations that the model predicted. That's it, that's what was not previously recognized. If anything, interpretation was the problem, it was the stumbling block-- old interpretations of what motion was like in the universe had to get out of the way. Interpretation is to be cautious of, that's more or less the point of this thread.

Ha ha, what model? the only models seriously considered previous to 1929 Hubble law where Einstein's and de Sitter's, none of them predicted the Hubble law observation. The model was built from that. Interpretations are to be cautious of is also my lemma, but you can't deny that we need them to begin with. there is no such thing as pure observations, numbers or measures don't mean anything unless interpreted in a certain frame or model.

[Staticman]

All I can say is that you do not understand the reason that the cosmological principle is used. It has nothing to do with philosophy, other than the simple philosophy that defines what science is (Occam's Razor). Simply put, we use the cosmological principle because we can. Period. If you think there is any other reason, you are just wrong.

If this is as scientific as you get on this point, no further discussion is needed, but normally scientific arguments are abit more elevated than this "cause we can. Period" type of reasoning. The Ockham razor is not correctly used here since you are not comparing the cosmological principle with any other to get the one that is more economic.

[Staticman]

A couple of references that might help anyone interested in understanding better what the Cosmological principle is and what isn't.

http://arxiv.org/abs/1006.0801
http://arxiv.org/abs/1012.5624

[Ken G]

The Ockham razor is not correctly used here since you are not comparing the cosmological principle with any other to get the one that is more economic.

Of course I am, I am comparing it to any model that invokes spatial variations instead of the cosmological principle. That's exactly the kind of comparisions that Occam involves. Now, as is also typical of Occam's Razor, its application involves intentional simplifications-- we know quite well that the cosmological principle is not correct, there will always be some spatial variation on any scale that we can observe. Again the purpose of Occam's Razor is to find useful simplifications, moreso than strictly correct ones. None of this is the least bit controversial, it's how science has been done for time immemorial. If you don't agree, then you are right I cannot force it to be true by saying it-- but who has time to demonstrate the point through the last few thousand years of scientific discovery? And the articles you cite don't say anything surprising about what the cosmological principle is, or isn't, they are about testing whether or not we can get away with it-- which is always the very next thing one does after applying Occam's Razor, this is all routine science.

[Staticman]

Of course I am, I am comparing it to any model that invokes spatial variations instead of the cosmological principle.

Ok, I should've got that.

That's exactly the kind of comparisions that Occam involves. Now, as is also typical of Occam's Razor, its application involves intentional simplifications-- we know quite well that the cosmological principle is not correct, there will always be some spatial variation on any scale that we can observe. Again the purpose of Occam's Razor is to find useful simplifications, moreso than strictly correct ones. None of this is the least bit controversial, it's how science has been done for time immemorial. If you don't agree

But I agree, none of what I've said I leads to thinking otherwise since we haven't discussed the Occam's razor before that I can recall, but the english language is ambiguous I guess.

And the articles you cite don't say anything surprising about what the cosmological principle is, or isn't, they are about testing whether or not we can get away with it-- which is always the very next thing one does after applying Occam's Razor, this is all routine science.

Glad you say this, then we can discuss other principle similar to the cosmological principle and decide which one of the two passes the razor with less gratuitous assumptions while being compatible with observations.
I'm refering to the Conditional cosmological principle. This one assumes the universe is statistically isotropic and homogenous, but doesn't assume spatial homogeneity. So you see it is a more simplifying principle while perfectly compatible with astrophysical observations thus far.

[Staticman]

Well, to settle why we choose the cosmological principle instead of other like the abovementioned, the reason is just that we hapen to have a model to fit the cosmological principle, the standard model, while we don't have any model that could fit the other principle. It's as simple as that, I guess that is a different way to say because we can, so Ken is right in the end, once again

[Ken G]

Glad you say this, then we can discuss other principle similar to the cosmological principle and decide which one of the two passes the razor with less gratuitous assumptions while being compatible with observations.

Yes, that should be the key goal.

I'm refering to the Conditional cosmological principle. This one assumes the universe is statistically isotropic and homogenous, but doesn't assume spatial homogeneity. So you see it is a more simplifying principle while perfectly compatible with astrophysical observations thus far.

And you are right we must constantly adjudicate these very similar possibilities, but we do so based on a larger context of what the goals are and what we can get away with (which is still an active area of research, as you show in those papers you cited). For a different example of how ubiquitous that kind of issue is in science, consider thermodynamics, where we talk about the density of a gas in a box, though we know that gases don't have constant densities. There are sound waves, statistical perturbations, and gravity in the box, all of which induce some variation. Does that invalidate the concept of the ideal gas law, because a box does not actually have a constant density? It all depends on what the context is, particularly on the desired level of precision vs. parsimony. We will always try to use the simplest model, but we don't marry it-- we'll keep an eye out for when it breaks down. Einstein said it well: "A theory is the more impressive the greater the simplicity of its premises is, the more different kinds of things it relates, and the more extended is its area of applicability." ( http://www.entish.org/wordpress/?p=638). The key point here is, we do not speak of which theory is more correct, or which is more absurd, we speak of which is more impressive, which can also be taken to mean more useful, or most commonly invoked in scientific literature.

[Spaceman Spiff]

The key point here is, we do not speak of which theory is more correct, or which is more absurd, we speak of which is more impressive, which can also be taken to mean more useful, or most commonly invoked in scientific literature.

(my bold)
Ken G --
That's one of the best distillations of the nature of scientific inquiry I've seen in a long time...