Multiverse Evidence Explanation?

[Chalnoth]

Why ponder an effective theory that lacks observational evidence? It might be mathematically attractive, but, is that really sufficient?

Because it isn't remotely true?

Many multiverse ideas that have been proposed do have testable consequences. Case in point:
http://www.livescience.com/15530-multiverse-universe-eternal-inflation-test.html
(note: this was a well-done, study, but the statistics aren't strong enough to say that they actually found evidence of a bubble collision)

On the other side, many experimental ideas which are very much testable happen to include a multiverse as one of their features. That is to say, the fact that there is a multiverse doesn't itself have experimental consequences, but it is a natural conclusion from the testable consequences of the theory. Examples here would be the spontaneous symmetry breaking of the standard model, a discovery that our vacuum is metastable, the many worlds of quantum mechanics, and many more.

My main problem with this sort of issue is that some people seem to harp on the idea that somehow a theory isn't valid or reasonable if it happens to include a multiverse, with no reasonable justification given whatsoever for this extreme case of theoretical bias. The only reasonable conclusion from the fact that the multiverse itself isn't testable is, "Okay, we won't use that aspect of the theory to evaluate its merits. What else does this theory have to offer?" Instead we get, "OMG! Multiverse! Not science! Bad! Get that theory out of here!"

 

[aleph-aleph]

I think if we going to discuss about the topic, we better do it in mathematical way. Apart from that, you can watch the videos about M theory for more information.
http://www.youtube.com/watch?v=J4Hn7l7jDY8&list=PL029DEA05B111AA1D

 

[marcus]

BTW quotes from new book by science writer Jim Baggott here:
http://www.math.columbia.edu/~woit/wordpress/?p=6002

... The only reasonable conclusion from the fact that the multiverse itself isn't testable is, "Okay, we won't use that aspect of the theory to evaluate its merits. What else does this theory have to offer?" ...

That sounds comparatively reasonable. As long as you have a testable theory and don't TALK about the fact that it involves some non-zero probability of other spacetime regions having arisen with different characteristics. As long as that is considered scientifically IRRELEVANT and does not enter the discussion I don't see how anyone would object.

I think this is a ridiculous strawman exaggeration:

Instead we get, "OMG! Multiverse! Not science! Bad! Get that theory out of here!"

That sounds like irrational bias, which is not how i would characterize what I hear as objections to definitely unwise unscientific behavior. People may express disgust with the multi aspect being brought up as an EXCUSE or a COP-OUT. Traditionally science is supposed to explain this world that we see and live in and experience---to find sufficient reasons for how things are. And when a theory FAILS and predicts a whole landscape of different versions of the universe then rather than chuck the theory some people have been known to cling to the theory and give up on reality: they say a plethora of universes exists and the theory is right to predict a whole multitude--one of which we happen to live in. That is GIVING UP on the traditional program of empirical science.

Another cause for disgust is when multi-fantasy is used as a selling point to young people because it appeals to the imagination.

So I might disagree with other things you may have said, but I like this:
"Okay, we won't use that aspect of the theory to evaluate its merits. What else does this theory have to offer?"

If the multi aspect of a theory is treated as IRRELEVANT and does not enter the discussion because it is irrelevant--and if the theory is focused on explaining the world we see: why it is this way instead of some other way, without giving up or making excuses, then I don't see why any reasonable person would object.

The multi aspect would then just be an irrelevant detail. Under those circumstances why would today's critics of multi-babble (who in my experience are pretty reasonable!) bother to take issue?

 

[Mordred]

There are numerous multi verse theorem, unfortunately little evidence. Despite that its still viable.
When you think about it there is very few models of our beginnings that do not involve a previous universe. The only ones I can think of involve the universe from nothing.
Cyclic and bounce both involve a previous universe, for that matter numerous inflationary models lead to multiverse scenarios.

Poplowskiis universe inside a BH is another example of multiverse as is numerous string theory models.
In some ways its harder to avoid the multiverse scenario.

However one also needs to ask a key question. If one universe can form, then why can't another?.
Logically it only makes sense that there can be more than one universe. Granted this causes other questions. Such as what separates one universe from another. What lies in between? Do the same laws apply?

Quite frankly we do not know enough about our own universe to answer those questions. Perhaps a better understanding of our universes beginnings will answer that or perhaps understanding the multiverse will provide the answer to our beginnings.

Opinions naturally vary on multiverse theorem. After all their is little to no evidence. However discounting the possibility would also be a mistake as we also have no evidence that it doesn't exist.

Therefore its important to keep an open mind on the subject, however their are plenty of misleading pop media coverage on the subject that in some ways does more harm than good, it has the side benefict of preparing the public a grudging acceptance, in case the multiverse is proven. Without that aid research into multiverse theorem would suffer I would think. So in a limitted way pop media is doing science a favor albiet in a roundabout manner.

Just a little something to consider.

 

[Haelfix]

There is a lot of indirect evidence for the multiverse, more or less coming from the same data that confirms inflation to begin with. The whole idea wouldn't be take seriously by so many theoreticians if there wasn't, even if laymen don't realize it. Physicists don't like putting additional structure if they don't have too.

It is a very natural progression of taking the theory of inflation seriously and going through the mathematical and logical consequences. This isn't a new thing either, it was understood pretty much right from the beginning, some thirty years ago.

What's not being appreciated here is just how many hoops you have to put a theory through in order NOT to have a multiverse. Scalar fields like the inflaton are very strange animals mathematically, and they tend to require very delicate and special conditions in order for them to work their magic in the first place. This puts very stringent mathematical structures and restrictions on the space of possibilities, and when properly analyzed it seems so far that those that produce eternal inflation and/or multiverse like scenarios tend to have much nicer properties and are more well behaved than those that do not.

So when people say that it's a virtue that their fundamental theories lead to a multiverse, it is precisely this last point that they are referring too. Namely the much nicer phenomenology and fewer miracles required in order to have a working model.

Instead if you read threads on this board you get the impression that physicists are adding on additional superstructures for no reason other than to save their pet theories. That is nonsense.. It is precisely the opposite.. The pet theories are there to explain what would otherwise seem to be a miracle.

 

[gpussetto]

Gravitational waves detectors

I believe that besides the mathematical constructions our best hope to understand and ultimately detect multiverses (if they exist) is by gravitational waves detectors.

In string theory, for example, the graviton can pass though universes because it is a close loop. Of course, there is a huge need for experimental tests that unfortunately seem to be way below the capability needed to probe all these theories. For now.

I do not think though that LIGO and eLISA will be sensitive enough for the task but they will open the door for new experimental science in that direction.

 

[Chronos]

Quoting Steven Weinberg, a multiverse proponent, from 'Living in the Multiverse', http://arxiv.org/abs/hep-th/0511037: "There is also a less creditable reason for hostility to the idea of a multiverse, based on the fact that we will never be able to observe any subuniverses except our own. Livio and Rees and Tegmark have given thorough discussions of various other ingredients of accepted theories that we will never be able to observe, without our being led to reject these theories. The test of a physical theory is not that everything in it should be observable and every prediction it makes should be testable, but rather that enough is observable and enough predictions are testable to give us confidence that the theory is right."
This critique of 'multiverse hostility' is weak. What credible observational support exists for 'observables', or 'testable predictions' of any multiverse theory proposed to date? AFAIK, none. It remains a hypothesis, IMO.

 

[Mordred]

Interesting paper Chronos, let me ask one question. What observable and testable prediction showed that the higgs boson was correct? As far as I know there was none prior to its discovery, and yet it was predicted.
I'm not too familiar with the anthropic principle I simply never looked at it. However this paper does make several valid points. One of which you have mentioned. Whilst its true that the mathematics of various models show the possibility of subuniverses, none have been observed, nor has any evidence been found to support multiverse/subuniverse theorem. That does not mean that they do not exist. Merely that we have not found evidence of it.
Much like the higgs in that regard. Thus far we have also found no evidence that deny a multiverse. So quite frankly its still viable. The CMB may or may not offer clues and I am loath to follow any multiverse claims based upon the Planck data. For one numerous multiverse model claims jumped at the anistrophy found without taking the time to verify and provide corrolary data. For all we know it could have
been a calibration error or other
factor that wasn't accounted for.

As far as whether a multiverse exists? At this point there is no right or wrong answer we simply do not know.

 

[mfb]

Interesting paper Chronos, let me ask one question. What observable and testable prediction showed that the higgs boson was correct? As far as I know there was none prior to its discovery, and yet it was predicted.

Electroweak symmetry breaking was not a prediction, but still an implication of the Higgs mechanism (the Higgs mechanism was designed to do this).
Electroweak precision measurements even allowed to estimate (predict!) the approximate mass of the Higgs boson, and the discovery is in agreement with that prediction.

 

[Garth]

Of course we might point out again that the standard Model in particle physics has been observed and verified to high precision, the Higgs boson being the last piece of the jigsaw, whereas with the standard model in cosmology, the $\Lambda$CDM model, Inflation, DM and DE remain undiscovered in laboratory physics and the multiverse that Inflation may spawn may well be always unobservable. As I said in #5 "The problem is that any 'evidence' for the multiverse is likely to be able to be also explained by other less exotic causes."

As far as Weinberg's statement quoted by Chronos above (#42),

The test of a physical theory is not that everything in it should be observable and every prediction it makes should be testable, but rather that enough is observable and enough predictions are testable to give us confidence that the theory is right,

the same could have been said by proponents of the Ptolemaic theory in the 16th Century.

Garth

 

[Mordred]

Electroweak symmetry breaking was not a prediction, but still an implication of the Higgs mechanism (the Higgs mechanism was designed to do this).
Electroweak precision measurements even allowed to estimate (predict!) the approximate mass of the Higgs boson, and the discovery is in agreement with that prediction.

Thats good to know, I recall some heated forum debates on the Higgs prior to its discovery. This was prior to my studies so I was never clear on the Higgs development. Particle physics is something I'm still learning.

 

[Haelfix]

Epicycles were a perfectly valid theory, and it made perfect sense to believe them at the time. They retro predicted all data to within the error bars of the time, as well as made new predictions. A perfectly sensible theory at the end of the day.

Of corse, once Newton wrote down his laws epicycles were superfluous. A simpler, more powerful theory had become available.

Currently this does not exist for the inflationary paradigm. Consequently we take the internal framework seriously, absent any other explanation, and the simplest form of the laws tend to involve multiverses.

Incidentally, the notion that the Higgs boson is a similar analogy is a stretch for a number of reasons. Before 2010, there was A LOT of empirical evidence for the Higgs FIELD. In a sense three out of the four components had already been observed. Its like if you put a dollar in coins under three jars, and measure the first two.. In a sense the third is already observed under the assumption that the world obeys logic.

 

[skydivephil]

I don’t think the Higgs is comparable to the multiverse. As I understood it, the LHc was powerful enough to either rule the Higgs in or out. There was a do able experiment that could say one way or the other if the Higgs existed. BTW I should add this is my understanding based upon some lectures I attended at my Uni. If someone more knowledgeable than me wants to correct that understanding go ahead.
The problem with the multiverse is there is no such as do able experiment. Let’s consider the bubble collision test. IF it is not seen, that won’t invalidate eternal inflation. However if it seen and it stands up and is shown to be robust that will certainly change things..
What if it is not seen? then we go back to trying to increase our understanding of inflation, including whether it really happened at all, what’s its energy scale is and whether its eternal or not. Some of this understanding may come from theory, some from experiment.
What I don’t understand is why the black and white views? Science doesn’t always give us a clear yes or no answers. Science should allow for uncertainty. If the data comes back fully in favour or inflation and if theorists tell us inflation must be eternal but we don’t see the direct evidence of bubble collisions, then what? Well I do think it would be right to say the multiverse is elevated way beyond the status of wild speculation, religion or science fiction. But it still won’t enjoy the status of something easily measured, like the spectrum of a star. But why can’t we be okay with that? There are something s that are clearly science, others that are clearly not sicene, but there is a grey area in between.

The multiverse is perhaps implied by a theory that has not been definitely shown to be true but has a lot of good observations in favour of it. That theory is inflation. Inflation has a lot of evidence to back it up, but there will still be room to doubt it without gravity wave measurements. It’s also possible that Alan Guth and friends have made a mistake in the way they think the inflaton field evolves. The picture is uncertain, that’s the world we live in.

I also have to object to the comment that was made “the Λ CDM model, Inflation, DM and DE remain undiscovered in laboratory physics and the multiverse that Inflation may spawn may well be always unobservable.”
The universe is a lab and a telescope is nor more or less reliable than a particle accelerator. Those guys are doing things indirectly too, no one has ever seen a quark and it’s unlikely they ever will. LCDM is well verified by data.

 

[Mordred]

The Higgs I'll agree is a poor example. I was trying to show an example of theory to discovery.
As far as inflation goes Guths work isn't the only inflationary model that fits observational data. There is currently over 60 inflationary models that do match observational data.
His original model "old inflation" (false vacuum) has been replaced by chaotic inflation as well as slow roll inflation. However other inflation models do fit. These include higgs inflation (single scalar) hill inflation, natural inflation etc.
Judging from numerous articles the slow roll
approximation appears to me to be a standard that other inflation models are compared against. If I recall the slow roll also leads to bubble universes but I could be wrong on that.
I don't know if Higgs inflation
does nor on some of the others.

This inflationary review uses the slow roll as a comparision along with a fortran program ASPIC. Its extremely lengthy and technical but I find its handy as the various inflationary formulas are included. Makes life easy when your interested in finding the formula for say Hill inflation etc.
http://arxiv.org/abs/1303.3787

 

[mfb]

Well, we can use supersymmetry (SUSY) as an example:
It can be tested, but it cannot be falsified by current experiments. You can just find it, or restrict the parameter space.
It is even worse with string theory (which requires SUSY, by the way)... and both are serious theories.

 

[DennisN]

Well, we can use supersymmetry (SUSY) as an example:
It can be tested, but it cannot be falsified by current experiments. You can just find it, or restrict the parameter space.
It is even worse with string theory (which requires SUSY, by the way)... and both are serious theories.

LOL, I was about to make exactly that comparison, but I saw mfb had already said it .

 

[Chronos]

The Higgs particle was no more than a hypothesis, however well motivated, until observationally confirmed. I'm no multiverse fan because it's too easy to invoke the dumb luck clause in lieu of seriously addressing questions about the nature of our universe. That's no better than intelligent design, IMO.

 

[Mordred]

I have to agree on that, I'm also one that prefers answers to our own universe. As opposed to quoting multiverse as an answer to questions we cannot yet answer. I recall one in which Hawking tried to explain that information isn't lost in a BH. Instead the information is sent to a universe without a BH. I never did see the technical paper on that so it may be just pop media.

 

[skydivephil]

The Higgs particle was no more than a hypothesis, however well motivated, until observationally confirmed. I'm no multiverse fan because it's too easy to invoke the dumb luck clause in lieu of seriously addressing questions about the nature of our universe. That's no better than intelligent design, IMO.

Do you think inflationary cosmology itself is no better than ID?

 

[Garth]

Do you think inflationary cosmology itself is no better than ID?

Actually Chronos was talking about the multiverse conjecture.

ID says if that if you cannot explain anything then 'God did it'. Nothing is explained further.

There is a tendency to say that if you cannot explain anything cosmological (anthropic coincidences for example) then it happens in the multiverse (the ensemble of random selections of physical constants for example). Nothing is explained further.

You can't 'observe' God, although many find reasons for believing in God, just as you can't observe other universes, although many find reasons for believing in them.

I would say the multiverse conjecture is no better as an explanation than ID.

Just a thought.

Garth

 

[skydivephil]

But there's a big difference, god is an arbitrary conjecture. In this case the multiverse is not the conjecture, . It comes from inflationary cosmology. I agree inflation could be wrong and the chain between inflation and eternal inflation can also be wrong. But wrong or not it is not an arbitrary conjecture.
I agree people can over use the possbility of a mulitverse to explain thing that may have no connection to it. But to say its very existence is equivalent to believing in something silly like ID is going too far. Of course there are people that go toof ar on the other side, writing about the multiverse as if it's a well establilshed fact. That's going too far too. But extrapolating a reasonably well tested theory may be wrong, but its not the same as believing in fairy tales.

Some ideas are science and some are not. but there is a grey areaa in between and I think both sides in the multiverse debate should acknolwedge this.

 

[Mordred]

Unfortunately multiverse theorem is not restricted to inflationary causes,

http://arxiv.org/abs/1212.5150 loop Quantum multiverse

Poplowskii universes inside BH's

http://www.physics.indiana.edu/~nipoplaw/publications.html
http://www.physics.indiana.edu/~nipoplaw/articles.html

not to mention string theory

 

[Garth]

Or the many other oscillations of this universe in an oscillating big bounce scenario.

Universes, universes everywhere! Other universes pop up all over the place.

Show me one.

Garth

 

[Mordred]

lol no kidding I forgot a classic one

http://www.technologyreview.com/view/424073/multiverse-many-worlds-say-physicists/

 

[Chalnoth]

I have to agree on that, I'm also one that prefers answers to our own universe. As opposed to quoting multiverse as an answer to questions we cannot yet answer.

Then how do you deal with the fact that if we do live in a multiverse, then certain features of our universe might only possibly make sense if understood in that context?

To take an example, consider the cosmological constant. If we live in a multiverse where an extremely large number of different values of the cosmological constant are realized in different contexts, then there is nothing to explain: we see the value we do simply because life requires a small cosmological constant.

If this is true, then it simply will not ever be possible to understand the cosmological constant without referencing the multiverse.

And if you think this explanation isn't possibly testable, then clearly you haven't thought about the problem hard enough. The essential point is that this explanation requires reference to a fundamental theory which allows many different values of the cosmological constant to be realized. Its testability doesn't rely at all upon the direct measurement of those alternate universes, but instead in our ability to measure that more fundamental theory.

To provide a simple example, let's imagine that we had such a theory, and were able to deduce that it didn't allow any possible value of the cosmological constant, but only allowed certain, very specific values. Imagine, for example, that the theory predicted that the cosmological constant could only take values 0.5302, 0.7204, 0.9856, 2.3519, rendered in units of the current density fraction, and a bunch of other values both smaller and larger (but not inbetween these values). That would give us a testable result: we would know, for example, that the only feasible value would be the 0.7204 one. So if we measured very carefully, and found that the cosmological constant was $0.70 \pm 0.02$, then we would have ruled out this particular fundamental theory.

Granted, we would have to be extremely lucky to have a theory like this. String theory, for instance, predicts some $10^{400}$ possible vacua, which would mean that the spacing between the different cosmological constants for those vacua might easily be as tight as $10^{-280}$ in units of the current density fraction, vastly too small for us to ever conceivably measure. But the point is that we would be testing an entire theory, not just one small aspect of it (the multiverse). We're not there yet, but there is no good reason to reject a theory a priori just because it happens to include a multiverse. In fact, I would go so far as to say that given what we know today (which isn't much), we should strongly, strongly prefer such multiverse theories.

 

[Mordred]

Good example, I think you misunderstand me though my problem isn't so much in multiverse theories.
I can't think of any good arguments as to why the multiverse shouldn't exist. The problem I have is the tendency to explaining problems such as the information paradox by statements like

"the information isn't lost it goes to a universe without black holes"
"The wormhole opens into another universe that is why we can't find the evidence of white holes"
" that anistrophy in the CMB is evidence of an interaction between our universe and another universe"

Granted many of these postulates derive from models that are not necessarily mainstay, but their is a tendency to jump to the multiverse conclusion first as opposed to finding answers in the realm of our own universe. If it proves impossible to answer a question without a multiverse then I'm OK with it.
There is a practicality in modelling alternate universes, as one can adjust the laws to test various ideas out. However it should be clear when its specifically a modelling, as opposed to a statement of existence. Granted some of that is due to pop media but not always.

 

[Naty1]

Either it is eternal, or it is generic. If it is generic, then you have a real problem explaining why it happens once, and only once when we know that quantum mechanics is about probabilities.

This seems good logic. I have thought that either our universe is a ' custom made universe, uniquely suitable for us' or one of many universes, of all types and conditions only some of which are self sustaining and perhaps perpetuating others. Probability suggests to me the latter is more likely.

In fact, I would go so far as to say that given what we know today (which isn't much), we should strongly, strongly prefer such multiverse theories.

That might be a bit much, but I'd sure like to keep an open mind, especially given the poor record of consensus science over the ages.

 

[mfb]

especially given the poor record of consensus science over the ages.

Poor record? There are some notable exceptions, but in general I think science has been extremely successful.
Most new theories are just extensions of existing theories, modifying them in some special parts to make them more precise.

 

[Naty1]

mfb posts:

...Poor record?...Most new theories are just extensions of existing theories, modifying them in some special parts to make them more precise.

I know what you mean, but in general scientists are way too ' huffed and puffed' up about too many things...and my comment IS a bit sarcastic, but many of the 'revisions' are substantial...and I bet our views of a multiverse will be laughed at 100 years from now...I sure hope so...Here is a short list. I have omitted some famous quotes, real 'doozies' , from eminent scientists such as

"High speed train travel is impossible as air will be sucked out of people's lungs.'...

I was stunned when I tried to find ANY theory that started out, was comprehensive and correct. Even stunning relativity is bounded by quantum theory, and vice versa...
Earth is covered by a circular dome Ptolomy mistaken belief lasted 1500 years
Earth is Flat Erastothenes
Aristotelian physics Wrong for 2,000 years until Galilean physics
Earth is center of universe –Copernicus Galilelo Denied by Catholic Church
Milky way IS the universe 1920’s..Hubble showed Many galaxies
Travel over 30 MPH not possible Breathing not believed possible
Planetary oribits are circular Kepler Orbits are elliptical
Nothing left to Discover… .Wolfgang Pauli See all the stuff below….
Night sky is black due to interstellar dust Edgar Allan Poe (The writer!)
Atomic weights just a curiosity Mendelev created periodic table
Proton, neutrons elementary particles Composed of quarks
All matter came from big bang Iron and heavier stuff comes from stars
Electrons point particles More likely strings, waves
Alchemy is science Widespread transmutation not possible
Universe is common matter Vera Rubin: only 5% we know, rest we don’t!
uniform galactic rotation
Black Holes don’t exist unproven, but highly likely they do
Universe has always existed Big bang theory…universe had a start
Universe is static Hubble it's actually expanding

Space-time is continuous String , entropy, & quantum theory
spinfoam
Space has three dimensions String theory: 11 more likely (Ed Witten)
Lavoisier’s caloric theory heat is a form of energy
Thompson, Joule, Carnot
Laser fusion simple (1960) after 50 yrs, still trying
We know most mass, energy hardly know any: dark is 96% of universe

 

[marcus]

I would say the multiverse conjecture is no better as an explanation than ID.
...

That sounds fair, with emphasis on "as an explanation". If you had a theory with real testability and explanatory power but which just as a side effect also implied the existence of causally separate worlds that would seem unobjectionable since the multiverse feature would not be invoked to explain anything. Just a harmless byproduct. Or?

But there's a big difference, god is an arbitrary conjecture. In this case the multiverse is not the conjecture. It comes from inflationary cosmology...

I think statement needs some qualification, Skydive. It does not come from inflationary cosmology in general.

Here is a clean simple inflationary cosmology that does not drag any multiverse in with it:
http://arxiv.org/abs/1301.1264
I think it's preferable to the other inflation scenarios I've heard because it does not need any mysterious "quantum fluctuation" to get started, or any fine-tuning to get adequate efolds and graceful exit. No elaborate or far-fetched assumptions. No funny-shaped plateau potential to "roll down". Just the quantum corrected Friedman equation and a scalar field.

So multiverse is certainly not entailed by every inflation model. Though it is entailed by some which, in comparison with 1301.1264 look over-dressed and out-of-date. My two cents.

 

[Naty1]

Marcus:

Here is a clean simple inflationary cosmology that does not drag any multiverse in with it: Duration of inflation and conditions at the bounce as a prediction of effective isotropic loop quantum cosmolog

My oh my oh my!

So happy to see you coming around to 'bounce' ideas...I still recall when I posted a Steinhardt/Turok paper and you were...well...maybe just a tad skeptical!

[I seem to recall needing a few vodka and tonics that evening! [LOL] ]

Good to see and 'old dog learning new tricks'... much, much better than my wife says about me!

 

[mfb]

@Naty1:
Most of those things were bad science, not science, bad extrapolations or good approximations.
A quote from some scientist is not a scientific theory. There was never a theory "there is nothing left to discover", and alchemy (as an example) never had a theory "transmutation is possible", just the hope that it would be (well, it is, but not with chemical methods).

Electrons point particles More likely strings, waves

I think "More likely strings" is a very ... optimistic view for string theory. Waves and point particles are not a contradiction.Newton's laws of motion / classical mechanics: More than 300 years old, and every modern theory has this as limit of low velocities, small energies, large timescales and so on. We know that it is not exact, but it is an extremely good approximation in many setups.
Evolution: More than 150 years old, and still the most important theory in biology. More and more details were added, but the basic laws are still the same.
Ordinary matter is made out of atoms: While the idea is ~2000 years old, actual scientific experiments are newer. And the model is extremely successful.
Special relativity: More than 100 years old, survived thousands of tests without any flaw. It is the basis of every modern (fundamental) theory.
General relativity: Nearly 100 years old, survived thousands of tests. The fly-by anomaly is a bit odd, but I don't think this is new physics.
Quantum theory: Our modern world would look completely different without working applications of quantum mechanics.

By the way, your post would be easier to read with [noparse]

[/noparse]-tags around the "table".

 

[Naty1]

Most of those things were bad science, not science, bad extrapolations or good approximations.

You may think all science now is good science; I have my doubts.

But the 'good news' is we do make progress...as I think your comments imply.

 

[mfb]

Not all science, but I think the fraction improved significantly, at least in physics and related areas.

 

[Mordred]

In regards to my take on improvements that cosmology has taken, I'll tell a little story of my time on an older forum that eventually shutdown on space.com. Back in 2002, we used to get a slew of questions such as below.

1) what shape is the universe, is it flat, sphere, hyperbolic, klein bottle, donut or some other shape. the answer wasn't determined then. thank WMAP.
2) which hot big bang inflationary model is correct? unfortunately there is still debate on that but not nearly as bad.
3) which is more accurate the CDM model or the hot big bang model. unfortunately at this time CDM had numerous models, and was not nearly as fine tuned as now, also during this time DM and DE were still in hot debate.

needless to say in the past 10 years numerous advancements have come along way for which I am thankful.