Nature of astrophysics (9) - Empiricism, Extrapolation, Etc

[Nereid]

As the story goes, when the apple fell on Newton's head, he was looking at the sky, at the Moon. It was at that moment (again, as the story goes) that he realised that what made apple fall is the same as what keeps the Moon going round the Earth! In Newton's hands, "the same" meant his law of universal gravitation, with its inverse square characteristic, and a constant ("big G").

But Newton did not live to see his proposal tested in the lab; Cavendish didn't 'weigh the Earth' until 1798. There's more; the inverse square part wasn't tested - empirically, using the Earth as one of the two masses - until the 20th century.

Late in the 19th century, several emission lines were discovered, in the spectra of what were, and still are, called planetary nebulae. Several generations later, with quantum mechanics firmly established, these lines (which are purely electromagnetic phenomena) were explained as 'forbidden transitions' of various ions of oxygen, nitrogen, and so on. However, few have ever been observed in any lab experiment.

There are many examples like this, in physics, especially when it's applied to astronomy; theories tested within a relatively narrow range (basically, what's possible in labs, here on Earth) are assumed to work far beyond that range. In other words, extrapolation is valid.

But is this always so? Just because General Relativity has passed every lab-based test to date, does that mean that gravitational lensing occurs? After all, it's just as much an extrapolation as deriving the orbit of a dwarf planet far beyond the Kuiper Belt!

And if it's OK to extrapolate various characteristics and behaviours of plasmas far beyond what's been tested in labs - using scaling laws - why not other characteristics and behaviours, say of excited atomic states?

Empiricism means that we cannot know how the universe works, much beyond the Earth's atmosphere, if by empiricism we mean that which can be reproduced in controlled experiments, objectively. After all, no one can produce a copy of the Sun, much less do controlled experiments on it, in a lab here on Earth. On the other hand, if we recognise that the universe is constantly doing experiments for us, the results of which we (plural) can observe, repeatedly, then hypotheses about extrapolations - be they about gravity, plasmas, ions, or whatever - can be tested ... and in that sense astrophysics is empirical.

What do you think?

[Nereid]

There's a post, by MrAmsterdam, which seems relevant to this thread.

Here it is:

This discussion seems to be quite old. Lets check our beloved wikipedia.

Empiricism
http://en.wikipedia.org/wiki/Philosophy ... Empiricism

Empiricism is a form of realism that denies that mathematics can be known a priori at all. It says that we discover mathematical facts by empirical research, just like facts in any of the other sciences. It is not one of the classical three positions advocated in the early 20th century, but primarily arose in the middle of the century. However, an important early proponent of a view like this was John Stuart Mill. Mill's view was widely criticized, because it makes statements like "2 + 2 = 4" come out as uncertain, contingent truths, which we can only learn by observing instances of two pairs coming together and forming a quartet.

Contemporary mathematical empiricism, formulated by Quine and Putnam, is primarily supported by the indispensability argument: mathematics is indispensable to all empirical sciences, and if we want to believe in the reality of the phenomena described by the sciences, we ought also believe in the reality of those entities required for this description. That is, since physics needs to talk about electrons to say why light bulbs behave as they do, then electrons must exist. Since physics needs to talk about numbers in offering any of its explanations, then numbers must exist. In keeping with Quine and Putnam's overall philosophies, this is a naturalistic argument. It argues for the existence of mathematical entities as the best explanation for experience, thus stripping mathematics of some of its distinctness from the other sciences.

Putnam strongly rejected the term "Platonist" as implying an overly-specific ontology that was not necessary to mathematical practice in any real sense. He advocated a form of "pure realism" that rejected mystical notions of truth and accepted much quasi-empiricism in mathematics. Putnam was involved in coining the term "pure realism" (see below).

The most important criticism of empirical views of mathematics is approximately the same as that raised against Mill. If mathematics is just as empirical as the other sciences, then this suggests that its results are just as fallible as theirs, and just as contingent. In Mill's case the empirical justification comes directly, while in Quine's case it comes indirectly, through the coherence of our scientific theory as a whole, i.e. consilience after E O Wilson. Quine suggests that mathematics seems completely certain because the role it plays in our web of belief is incredibly central, and that it would be extremely difficult for us to revise it, though not impossible.

For a philosophy of mathematics that attempts to overcome some of the shortcomings of Quine and Gödel's approaches by taking aspects of each see Penelope Maddy's Realism in Mathematics. Another example of a realist theory is the embodied mind theory (below). For a modern revision of mathematical empiricism see New Empiricism (below).

For experimental evidence suggesting that one-day-old babies can do elementary arithmetic, see Brian Butterworth.
-
Embodied mind theories

Embodied mind theories hold that mathematical thought is a natural outgrowth of the human cognitive apparatus which finds itself in our physical universe. For example, the abstract concept of number springs from the experience of counting discrete objects. It is held that mathematics is not universal and does not exist in any real sense, other than in human brains. Humans construct, but do not discover, mathematics.
-
New Empiricism

A more recent empiricism returns to the principle of the English empiricists of the 18th and 19th Centuries, in particular John Stuart Mill, who asserted that all knowledge comes to us from observation through the senses. This applies not only to matters of fact, but also to "relations of ideas," as Hume called them: the structures of logic which interpret, organize and abstract observations.

To this principle it adds a materialist connection: All the processes of logic which interpret, organize and abstract observations, are physical phenomena which take place in real time and physical space: namely, in the brains of human beings. Abstract objects, such as mathematical objects, are ideas, which in turn exist as electrical and chemical states of the billions of neurons in the human brain.
-
This second concept is reminiscent of the social constructivist approach, which holds that mathematics is produced by humans rather than being “discovered” from abstract, a priori truths. However, it differs sharply from the constructivist implication that humans arbitrarily construct mathematical principles that have no inherent truth but which instead are created on a conveniency basis. On the contrary, new empiricism shows how mathematics, although constructed by humans, follows rules and principles that will be agreed on by all who participate in the process, with the result that everyone practicing mathematics comes up with the same answer — except in those areas where there is philosophical disagreement on the meaning of fundamental concepts. This is because the new empiricism perceives this agreement as being a physical phenomenon.
-
For most mathematicians the empiricist principle that all knowledge comes from the senses contradicts a more basic principle: that mathematical propositions are true independent of the physical world. Everything about a mathematical proposition is independent of what appears to be the physical world. It all takes place in the mind. And the mind operates on infallible principles of deductive logic. It is not influenced by exterior inputs from the physical world, distorted by having to pass through the tentative, contingent universe of the senses. It all happens internally, so to say.
-
If all this is true, then where do the world senses come in? The early empiricists all stumbled over this point. Hume asserted that all knowledge comes from the senses, and then gave away the ballgame by excepting abstract propositions, which he called “relations of ideas.” These, he said, were absolutely true (although the mathematicians who thought them up, being human, might get them wrong). Mill, on the other hand, tried to deny that abstract ideas exist outside the physical world: all numbers, he said, “must be numbers of something: there are no such things as numbers in the abstract.” When we count to eight or add five and three we are really counting spoons or bumblebees. “All things possess quantity,” he said, so that propositions concerning numbers are propositions concerning “all things whatever.” But then in almost a contradiction of himself he went on to acknowledge that numerical and algebraic expressions are not necessarily attached to real world objects: they “do not excite in our minds ideas of any things in particular.” Mill’s low reputation as a philosopher of logic, and the low estate of empiricism in the century and a half following him, derives from this failed attempt to link abstract thoughts to the physical world, when it is obvious that abstraction consists precisely of separating the thought from its physical foundations.
-
The conundrum created by our certainty that abstract deductive propositions, if valid (i.e., if we can “prove” them), are true, exclusive of observation and testing in the physical world, gives rise to a further reflection...What if thoughts themselves, and the minds that create them, are physical objects, existing only in the physical world?

This would not reconcile the contradiction between our belief in the certainty of abstract deductions and the empiricist principle that knowledge comes from observation of individual instances. We know that Euler’s equation is true because every time a human mind derives the equation, it gets the same result, unless it has made a mistake, which can be acknowledged and corrected. We observe this phenomenon, and we extrapolate to the general proposition that it is always true. However, based on this rationale, one would still not be warranted in concluding that mathematics are purely empirical in nature.

This applies not only to physical principles, like the law of gravity, but to abstract phenomena that we observe only in human brains: in ours and in those of others.

All quotes are from the article ; http://en.wikipedia.org/wiki/Philosophy_of_mathematics

Ps. I saw a documentary once, where a university professor was waving his hand in front of his face and said;" there could be 9 dimensions in front of me, but there is no way for me to see these dimensions" ---

(Ill try really hard to find the documentary again)

[jjohnson]

It's hard to decide whether the Universe is the lab in which (at least some) sentient beings are running experiments, or whether we are the lab in which the Universe is running its experiments.

Almost all of the Universe is inaccessible to us, so there is a rational degree of uncertainty as to whether or not extrapolation is valid. How would we know for sure unless we could go there and observe directly? Assuming that extrapolating the rules that we think govern how, if not why, things work the way they do is an assumption of convenience or simplification. But if we observe phenomena relatively close by, from, say, our planetary system out to the stars in the local group, and they exhibit a set of very similar signatures with quantifiable differences and similarities, and then look farther and farther away, and observe the same electromagnetic and temporal behaviors, within a range of reasonable differences, it is not totally unwarranted for us to behave as if things behave fundamentally the same for as far as we can obtain useful data. It's just that we cannot know if we should be certain of this.

But then, no risk: no reward. If we assumed the opposite case, sort of as an exercise in hyperbolic logic, that nothing works elsewhere in the Universe in quite the same way as it does "locally", why bother? What would we prove by that? We obtain little data that says that stars in Andromeda or in galaxies imaged in Hubble's deepest field generate light and move around in certain patterns and give us readings that generally are similar to what we can observe with some regularity and precision"around here", but operate under quite different laws.

Rather than approaching exploring the Universe as an exercise in futility, it is not only more logical to assume it operates everywhere under the same set of rules (albeit under widely varying conditions), it is more efficient and would involve a lot less guesswork. We get enough of that right now, with our own singular set of laws.

So yes, as a general rule, until we obtain sensible observations with which to attach a different explanation as to how the Universe operates elsewhere, we will avoid getting ourselves bogged down and going nowhere, making up things we don't and possibly can't ever observe. Dance with the one that brung ya!

That said, the IEEE drily observes that the normal, observable condition of an incredibly high fraction of the observable matter in the Universe is deployed in the plasma state. Are they wrong? Have we obtained information from astronomical observations which counters that? If not, should most of the Universe off planet Earth not be approached with the idea that it works a lot like our solar home and its local group, and that there are exceptions to local phenomena such as supernovas and CRB's and quasars which, although far away and different, might be explained by the relatively familiar laws we extend out to their region? I hope so, but I can't know. I think that we will progress faster if we work at pinning down how the things we know work, work, and where they work, and when they don't, and see how our "picture" (it's always visual with beings who devote so much cortical real estate to the sensation and interpretation of a narrow band of EM energy) can be applied as universally as possible.

This is why I balk at the idea of dark matter, I suppose. It seems too conjured, too convenient, too intentionally defined as difficult to observe, to pass the common-sense smell test. Naturally, I would have balked at quantum mechanics, but local measurements and observations - experiments - have shown otherwise. It's at least plausible now. Maybe next generation, dark matter will be too, but I'm not betting the ranch on it.

Jim

[Biggins]

In my opinion, astrophysics is not a true science.
Let me qualify that statement. Like archeaology and other ancient historical disciplines, astrophysics bases its 'science' on observation only. We can dig up objects or gather spectrographic data on stars, but we can not manipulate conditions to test our hypotheses. We can only make further observations to see if our hypotheses are still valid.
Obviously there are times, especially in the initial stages of development, where an hypothesis will have to be modified based on further discoveries. But it should be clear that any discovery that required too big a change to the original hypothesis MUST be seen as a deficiency of that hypothesis and therefore it should be discarded, even if there is nothing to replace it with. If I can see a deficiency of the dating of an artifact via carbon dating, I need not supply a correct date for that artifact, I merely have to show that the current understaning is incorrect.

It is this point that I want to make. Astrophysics is about observation and extrapolation. Many hypotheses can be created that accurately describe and predict the observations. Mainstream astrophysics AND the EU proponents should be aware that we cannot prove anything, only predict and observe. The question is, who can predict better?
The answer is not clear. Thornhill has made some very good and correct predictions of comets, so I believe that the pendulum swings in his favour.
We must never forget that scientists, or at least good scientists should be first and foremost sceptics. Sceptical of everything - mainstream theories, data collection, data extrapolation, non-mainstream theories, me, you, the bloke from the pub, anyone wearing a lab-coat, TV evangelists, Nereid, Thornhill, Stephen Hawking, Phil Plait, Hannes Alfven, etc.
Unfortunately it is all too easy to be sceptical of non-conventional theories, and non-sceptical of theories that we hold dear. But we should not hold on to an idea becasuse it's all that we have. It is the height of stupidity to hold on to a theory becauseit's the only one we have. "I don't know" is an acceptable viewpoint and should not be underestimated. Most highly ignorant and highly intelligent people suffer from the same inability to show that they do not know something.

Sorry that the post is a little disjointed....

[Goldminer]

Biggins:

No,no,no . . . you are exactly not disjointed. You can sit around my fire any time. Right on!

I heard an acquaintance call it a B.S filter. He had a pair of ear muffs that just filtered out the B.S.
I just love the Stuffed Shirts. Always willing to explain what you don't understand (to you.)
A hundred times.
Even when you retort: Yes, I understand. Now, how come this or that?
"Well, if you have to ask that, you obviously don't understand . . . here, let me explain it to you . . ."
Snore

Seems everyone is going for the legend. Most folk just don't realize exactly what they have assumed, presumed.
Dark Matter? that's easy. Here let me tell you. It's a lot like dark matter.

On the other hand;
One time, a hundred years ago, I was trying to teach my son fractions by using the face of a clock. I had posed a problem to him, and he came up with some god-awful answer. I wanted to fold up my program right there. Well, I kept a straight face and asked him how he arrived at that. He made me proud. A perfectly logical explanation, one I could not have concocted. I just hadn't made a little tiny point clear to him. Once we had that straightened out, he was popping the answers to adding, subtracting ect those fractions off the face of that clock faster than I could pose'em.

Then there's those folks sit'n around the fire, look'in at the stars and talk'in to themselves . . .

[Aardwolf]

Just because General Relativity has passed every lab-based test to date, does that mean that gravitational lensing occurs?

Which lab-based tests are you refering to?

[Nereid]

Just because General Relativity has passed every lab-based test to date, does that mean that gravitational lensing occurs?

Which lab-based tests are you refering to?

Those covered, or at least cited, in Clifford Will's The Confrontation between General Relativity and Experiment, plus the ones reported subsequently, e.g. Müler et al. (2009).

If you're interested, we can go through these, one by one, in separate (new) threads in the Electric Universe section.

[Nereid]

Yet another excellent, long post by you Jim!

That said, the IEEE drily observes that the normal, observable condition of an incredibly high fraction of the observable matter in the Universe is deployed in the plasma state. Are they wrong?

Well, one problem a purely empirical (i.e. no extrapolation allowed) approach would have with bald statement is that it is totally dependent on extrapolating far, far, far beyond anything ever seen in any lab.

Have we obtained information from astronomical observations which counters that? If not, should most of the Universe off planet Earth not be approached with the idea that it works a lot like our solar home and its local group, and that there are exceptions to local phenomena such as supernovas and CRB's and quasars which, although far away and different, might be explained by the relatively familiar laws we extend out to their region? I hope so, but I can't know. I think that we will progress faster if we work at pinning down how the things we know work, work, and where they work, and when they don't, and see how our "picture" (it's always visual with beings who devote so much cortical real estate to the sensation and interpretation of a narrow band of EM energy) can be applied as universally as possible.

And the challenge has been taken up, by many, over many decades; to date the results are pretty bleak for this 'there's nothing but earthly phenomena, writ large' idea (to considerably over-simplify): all such quantitative models developed to date are inconsistent with the totality of relevant astronomical observations, sometimes spectacularly so. Of course someone, some day, may come along with a new model based on nothing but 'earthly phenomena, writ large' - a quantitative one of course - that is consistent; maybe a Thunderbolts Forum member will that someone?

This is why I balk at the idea of dark matter, I suppose. It seems too conjured, too convenient, too intentionally defined as difficult to observe, to pass the common-sense smell test. Naturally, I would have balked at quantum mechanics, but local measurements and observations - experiments - have shown otherwise. It's at least plausible now. Maybe next generation, dark matter will be too, but I'm not betting the ranch on it.

A good test then: what about white dwarf stars? neutron stars?

[Aardwolf]

Just because General Relativity has passed every lab-based test to date, does that mean that gravitational lensing occurs?

Which lab-based tests are you refering to?

Those covered, or at least cited, in Clifford Will's The Confrontation between General Relativity and Experiment, plus the ones reported subsequently, e.g. Müler et al. (2009).

If you're interested, we can go through these, one by one, in separate (new) threads in the Electric Universe section.

If you could just point out which are the lab-based tests.

[Nereid]

Here are some examples (by no means a comprehensive list):
* All the Eöt-Wash, and other short-range force, tests
* Pound-Rebka experiment and all later replications and extensions
* the Müller et al. (2009) experiment
* All the 'local clock' tests, of which Chou et al. (2010) is one of the most recent.

I don't know whether Gravity Probe B counts, for you, as a lab-based test. Going a bit further, there are all the lunar laser ranging tests, ...

[Nereid]

plus the ones reported subsequently, e.g. Müler et al. (2009).

This contains a typo; it's "Müller", not "Müler", and (2010); so "Müler et al. (2009)" -> "Müller et al. (2010)"

[Aardwolf]

Here are some examples (by no means a comprehensive list):
* All the Eöt-Wash, and other short-range force, tests
* Pound-Rebka experiment and all later replications and extensions
* the Müller et al. (2009) experiment
* All the 'local clock' tests, of which Chou et al. (2010) is one of the most recent.

I don't know whether Gravity Probe B counts, for you, as a lab-based test. Going a bit further, there are all the lunar laser ranging tests, ...

I not sure I agree the Pound-Rebka test is lab-based but I doubt whether they had the precision to measure what they reported, or that what they reported had anything to do with gravitational time dilation.

The Muller et al (2010) paper was not a test, it was a re-interpretaion of test data not meant for that purpose, and has already been disputed here.

Chou et al. (2010) is very recent so the jury's out at the moment but I would have reservations about using single clocks for comparison or measurement considering the tiny effect they are seeking in clocks that are prone to drift and error.

Do you have any further detail regarding Eöt-Wash tests as at the moment the comment "has passed every lab-based test to date" is looking a little tenuous at best. Especially if this is all we have after 100 years.

And no I would only count tests performed under laboratory conditions as lab-based.

[Nereid]

I not sure I agree the Pound-Rebka test is lab-based

Perhaps it would help if you defined "lab-based" unambiguously?

In any case, since then there have been lots of others, with considerably greater precision and accuracy.

but I doubt whether they had the precision to measure what they reported, or that what they reported had anything to do with gravitational time dilation.

Why not write up your doubts - in a suitably quantitative form - in a paper, and submit it for publication?

The Muller et al (2010) paper was not a test, it was a re-interpretaion of test data not meant for that purpose, and has already been disputed here.

Well, Müller et al. responded (link is to the arXiv preprint abstract; it's already published in Nature)

We stand by our result [H. Mueller et al., Nature 463, 926-929 (2010)]. The comment [P. Wolf et al., Nature 467, E1 (2010)] revisits an interesting issue that has been known for decades, the relationship between test of the universality of free fall and redshift experiments. However, it arrives at its conclusions by applying the laws of physics that are questioned by redshift experiments; this precludes the existence of measurable signals. Since this issue applies to all classical redshift tests as well as atom interferometry redshift tests, these experiments are equivalent in all aspects in question.

Chou et al. (2010) is very recent so the jury's out at the moment but I would have reservations about using single clocks for comparison or measurement considering the tiny effect they are seeking in clocks that are prone to drift and error.

It's only the latest (?) such paper; there's quite a few (>10?), going right back to the 1972 Hafele-Keating experiment.

Do you have any further detail regarding Eöt-Wash tests as at the moment the comment "has passed every lab-based test to date" is looking a little tenuous at best. Especially if this is all we have after 100 years.

This University of Washington website is a good place to start

[Aardwolf]

I not sure I agree the Pound-Rebka test is lab-based

Perhaps it would help if you defined "lab-based" unambiguously?

In any case, since then there have been lots of others, with considerably greater precision and accuracy.

Then they will be subject to the same issues regarding whether its applicable to interpret red-shift movements as differing rates of time, and that they are certain to be excluding all other possible effects on the red-shifts.

And lab-based = In a lab under lab conditions. I'm not sure why the term "lab-based test" could be ambiguous. Why refer to a lab if your not expecting a controlled experiment under laboratory conditions. Why else would you even use a lab? Would one use a lab and not have a controlled experiment? Why not just do it in you lounge at home and not waste valuable research resource.

Maybe you need to explain what you meant by the term?

but I doubt whether they had the precision to measure what they reported, or that what they reported had anything to do with gravitational time dilation.

Why not write up your doubts - in a suitably quantitative form - in a paper, and submit it for publication?

Why? I'm satisfied it's not concrete proof of anything. If one chooses to accept it as proof of time dilation then that's a matter of faith not science, as it is not certain that is what they are measuring. It's just one interpretation and as its not strictly lab-based under obvious lab-based recommendations it shouldn't be included as one of your "lab-based tests".

The Muller et al (2010) paper was not a test, it was a re-interpretaion of test data not meant for that purpose, and has already been disputed here.

Well, Müller et al. responded (link is to the arXiv preprint abstract; it's already published in Nature)

We stand by our result [H. Mueller et al., Nature 463, 926-929 (2010)]. The comment [P. Wolf et al., Nature 467, E1 (2010)] revisits an interesting issue that has been known for decades, the relationship between test of the universality of free fall and redshift experiments. However, it arrives at its conclusions by applying the laws of physics that are questioned by redshift experiments; this precludes the existence of measurable signals. Since this issue applies to all classical redshift tests as well as atom interferometry redshift tests, these experiments are equivalent in all aspects in question.

And no doubt the opponents stand by their rebuttal. Irrelevant anyway as it's not a test, just a mathematical re-interpretation. As such it shouldn't be included in your "lab-based test" collective.

Chou et al. (2010) is very recent so the jury's out at the moment but I would have reservations about using single clocks for comparison or measurement considering the tiny effect they are seeking in clocks that are prone to drift and error.

It's only the latest (?) such paper; there's quite a few (>10?), going right back to the 1972 Hafele-Keating experiment.

What are the others? Are they "lab-based"? Do they use single clocks for comparison? and it's quite worrysome that you even mention the Hafele-Keating "experiment". That's about as far away from a lab-based experiment as you can be. I would hesitate to even glorify it with reference as an experiment of any kind. PR stunt is the most favourable description I can think of.

Do you have any further detail regarding Eöt-Wash tests as at the moment the comment "has passed every lab-based test to date" is looking a little tenuous at best. Especially if this is all we have after 100 years.

This University of Washington website is a good place to start

Most (all?) of these have nothing to do with proving GR. Some mention dark matter. Do they have some in a lab somewhere?

Unless you can identify some specific papers your "all lab-based tests" is still looking sparse to say the least. All we have is potentially Chou et al. (2010) and I have my doubts about that. Your sweeping comment sounded like it had more weight than the detail actually suggests. Unless you have a far more relaxed view regarding what lab-based means.

[Nereid]

Aardwolf, a detailed examination of the confrontation of GR with experiment is really beyond the scope of this thread; would you like to start a new one, or add to an existing one, in the Electric Universe section? I'd be happy to comment further in such a thread.

This, however, is very relevant to this thread:

And lab-based = In a lab under lab conditions. I'm not sure why the term "lab-based test" could be ambiguous. Why refer to a lab if your not expecting a controlled experiment under laboratory conditions. Why else would you even use a lab? Would one use a lab and not have a controlled experiment?

What is "a controlled experiment"?

In the case of gravity - in the sense of the idea that it has to do with mass - control could mean different masses held at different distances, some version of the Cavendish experiement perhaps. There are, however, severe limitations to such (lab-based, controlled, experiments/tests); for example, to apply the results of any such to the Earth, say, or the galaxy SDSS J113428.35+002830.9, an extrapolation of many orders of magnitude is required. But even more than that, all labs are on, or very near, the surface of the Earth; even though there might, one day, be labs on the surface of Mars, say, or even billions of km from the nearest large solar system object, no one reading this post today (6th January, 2011) will be alive to read about labs doing Cavendish-type tests on gravity somewhere in SDSS J113428.35+002830.9.

Even such a test - conducted by an ET perhaps? - would still not be a fully controlled test! Why not? Because the value of G (assuming Newtonian gravity) cannot be controlled.

Going back to another example in the OP, forbidden lines in planetary nebulae. Many of those prominent in the spectra of such astronomical objects have never been produced in "a controlled experiment under laboratory conditions", yet their assignment - [OIII] 1s22s2p3 -> 1s22p4, say - is hardly ever (never?) challenged, and certainly there is no serious proposal for ever more exotic rock/water/air/etc samples to be tested, in the lab, in search of a new element (nebulium) that produces these lines.

Aardwolf, you seem to be saying that we cannot know how the universe works, much beyond the Earth's atmosphere, ever; are you?