Please forgive another complaint.
Dark Matter was first postulated by Zwicky in the 40s, but it was 30
years before it was widely accepted.
"Dark Energy" was postulated on the basis of limited supernova data,
but it was accepted almost immediately, with alternative explanations
virtually ignored.
The difference? Cosmologists did not want Dark Matter, but very much
wanted "Dark Energy".
Rob O.
>>>>> "RO" == Rob Oldershaw <rlolders…@amherst.edu> writes:
RO> Please forgive another complaint. Dark Matter was first
RO> postulated by Zwicky in the 40s, but it was 30 years before it was
RO> widely accepted.
Strictly, if you consider dark matter to be any matter that is
detected first by its gravitational influence, rather than its
electromagnetic emissions (or reflections), the idea of dark matter is
hundreds of years old. Neptune was suggested to exist on the basis of
its gravitational influence, not because it was seen through a telescope.
RO> "Dark Energy" was postulated on the basis of limited supernova
RO> data, but it was accepted almost immediately, with alternative
RO> explanations virtually ignored.
I note that no alternative explanations are suggested.
The literature is replete with various alternative explanations being
considered and then rejected because they do not fit the data. Among
the ones of which I am aware are dust extinction and luminosity
evolution of the Type Ia supernovae. Indeed, as just one example, the
abstract of one of the initial papers, Riess et al. (1998) includes
the following sentence: "We estimate the likely effect of several
sources of systematic error, including progenitor and metallicity
evolution, extinction, sample selection bias, local perturbations in
the expansion rate, gravitational lensing, and sample
contamination. Presently, none of these effects appear to reconcile
the data with Omega_Lambda = 0 and q_0 >= 0."
–
Lt. Lazio, HTML police | e-mail: jla…@patriot.net
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html
Joseph Lazio <jla…@adams.patriot.net> wrote in message <news:mt2.0-8920-1053474227@star.bris.ac.uk>…Lt. Lazio, HTML police …
> >>>>> "RO" == Rob Oldershaw <rlolders…@amherst.edu> writes:
> RO> Please forgive another complaint. Dark Matter was first
> RO> postulated by Zwicky in the 40s, but it was 30 years before it was
> RO> widely accepted.
> Strictly, if you consider dark matter to be any matter that is
> detected first by its gravitational influence …, the idea of dark matter is
> hundreds of years old. Neptune …
Usually I don’t argue with the HTML police, but I think the Lt. misses
the point here. I’m specifying the Dark Matter that dominates the
mass of the observable universe.
> RO> "Dark Energy" was postulated on the basis of limited supernova
> RO> data, but it was accepted almost immediately, with alternative
> RO> explanations virtually ignored.
> I note that no alternative explanations are suggested.
> The literature is replete with various alternative explanations being
> considered and then rejected because they do not fit the data.
I was wrong to say that alternative ideas (to Dark Energy) were
"ignored". But I do think that the alternative ideas were fighting
against an unjustifiable bias against them. My main argument is that
the DE concept was embraced at face value with unusual quickness.
Rob O.
rlolders…@amherst.edu (Rob Oldershaw) wrote:
> I was wrong to say that alternative ideas (to Dark Energy) were
> "ignored". But I do think that the alternative ideas were fighting
> against an unjustifiable bias against them. My main argument is that
> the DE concept was embraced at face value with unusual quickness.
Well, if you consider that "Dark Energy" is really only a new name for
the cosmological constant, it is a concept that has been around since
Einstein’s first cosmological solution to the EFE. (Really before
that, in Newtonian cosmological attempts).
For details, see John Earman, "Lambda, The Constant that Refuses to
Die" (Arch.Hist.Exact Sci., 55:189, 2001).
>>>>> "RO" == Rob Oldershaw <rlolders…@amherst.edu> writes:
RO> Joseph Lazio <jla…@adams.patriot.net> wrote in message
RO> <news:mt2.0-8920-1053474227@star.bris.ac.uk>…Lt. Lazio, HTML
RO> police …
>> >>>>> "RO" == Rob Oldershaw <rlolders…@amherst.edu> writes:
RO> Please forgive another complaint. Dark Matter was first
RO> postulated by Zwicky in the 40s, but it was 30 years before it was
RO> widely accepted.
>> Strictly, if you consider dark matter to be any matter that is
>> detected first by its gravitational influence …, the idea of dark
>> matter is hundreds of years old. Neptune …
RO> [...] I think the Lt. misses the point here. I’m specifying the
RO> Dark Matter that dominates the mass of the observable universe.
At the risk of descending into semantic nit-picking, some of the dark
matter that Zwicky detected must surely have been the hot X-ray gas
that fills the Coma cluster. Not all of it, of course, but some of
it. I think my point remains accurate: Dark matter is not a new idea.
RO> "Dark Energy" was postulated on the basis of limited supernova
RO> data, but it was accepted almost immediately, with alternative
RO> explanations virtually ignored.
>> I note that no alternative explanations are suggested.
>> The literature is replete with various alternative explanations
>> being considered and then rejected because they do not fit the
>> data.
RO> I was wrong to say that alternative ideas (…) were "ignored".
RO> But I do think that the alternative ideas were fighting against an
RO> unjustifiable bias against them. My main argument is that the DE
RO> concept was embraced at face value with unusual quickness.
I think one would have to check the literature/try to dig up talks by
the various SNe groups. I don’t think they jumped on the dark
energy/cosmological constant idea. They tried "obvious" alternative
ideas first, like intergalactic dust and source evolution. When one
rules out all other options, one has to adopt the remaining one, at
least until a better one is presented.
Given that I see no better (indeed no) options being presented, I’m
not sure what else I can say on this topic.
–
Lt. Lazio, HTML police | e-mail: jla…@patriot.net
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html
In article <mt2.0-20741-1054209…@star.bris.ac.uk>, Joseph Lazio
<jla…@adams.patriot.net> writes:
> RO> "Dark Energy" was postulated on the basis of limited supernova
> RO> data, but it was accepted almost immediately, with alternative
> RO> explanations virtually ignored.
> I think one would have to check the literature/try to dig up talks by
> the various SNe groups. I don’t think they jumped on the dark
> energy/cosmological constant idea. They tried "obvious" alternative
> ideas first, like intergalactic dust and source evolution. When one
> rules out all other options, one has to adopt the remaining one, at
> least until a better one is presented.
Correct. IIRC, their early papers talk about, in modern language,
measuring Omega. Lambda wasn’t mentioned. Interestingly, the first
sample they analysed contains what is now known to be a statistical
outlier. As a result, their first analysis was CONSISTENT WITH the
Einstein-de Sitter universe and this model looked even more likely than
the present "standard model". (Error bars were big, though, and of
course the present standard model wasn’t strongly ruled out.) More and
better data eventually forced a new conclusion.
Most if not all of the people involved certainly didn’t embrace Lambda
right away, but sought other explanations. Their papers produced a
whole lot of other papers with alternative explanations, but few if any
hold up. Of course, both the current paradigm and alternative
explanations make DIFFERENT predictions for higher redshift, so the
number of viable possibilities will be decreased in the future.
Also, the supernova results came from two different groups using
different samples, different analysis techniques etc.
You can be sure that the scepticism with which this was originally
confronted by most of the community led to everything being checked and
re-checked many times. Around the same time, it was possible to publish
really BAD science, even in Nature, if it concluded that the (then)
standard model was correct.
So, though many people are on the Lambda bandwagon now, most didn’t jump
on until it was almost too late.
hel…@astro.multiCLOTHESvax.de (Phillip Helbig—remove CLOTHES to reply) wrote in message <news:mt2.0-28028-1054311198@star.bris.ac.uk>…
- Hide quoted text — Show quoted text -
> Most if not all of the people involved certainly didn’t embrace Lambda
> right away, but sought other explanations. Their papers produced a
> whole lot of other papers with alternative explanations, but few if any
> hold up. Of course, both the current paradigm and alternative
> explanations make DIFFERENT predictions for higher redshift, so the
> number of viable possibilities will be decreased in the future.
> Also, the supernova results came from two different groups using
> different samples, different analysis techniques etc.
> You can be sure that the scepticism with which this was originally
> confronted by most of the community led to everything being checked and
> re-checked many times. Around the same time, it was possible to publish
> really BAD science, even in Nature, if it concluded that the (then)
> standard model was correct.
> So, though many people are on the Lambda bandwagon now, most didn’t jump
> on until it was almost too late.
Hmm, bandwagon. Maybe it’s not so much that the new DE scenario was
adopted so quickly that bothers me (though it does), but the extent to
which it became the consensus view. There is definitely a bandwagon
feeling to it.
Not so long ago we had a standard model with inflation that predicted
a critical density dominated by matter. Then comes the supernova data
and ‘oops, well, actually we have an energy dominated model with a
positive cosmological constant that causes accelerated expansion’.
What do we think, in objective moments, are the chances that the new
bandwagon model will hold up any better than the recently abandoned
one? Maybe we should just admit that things are still very uncertain
in cosmology. We should have open minds, not a bandwagon mentality.
We should be scientists, not "true believers".
Recently Gia Dvali of NYU and colleagues reported that a 5-d GR theory
postdicts accelerated expansion without a cosmological constant. A
5-d approach also has some potential for unfying GR and EM. Will this
idea get a decent hearing, or is it ‘something not needed because we
already have the correct explanation’?
In article <mt2.0-10042-1054467…@star.bris.ac.uk>,
Rob Oldershaw <rlolders…@amherst.edu> wrote:
>Hmm, bandwagon. Maybe it’s not so much that the new DE scenario was
>adopted so quickly that bothers me (though it does), but the extent to
>which it became the consensus view. There is definitely a bandwagon
>feeling to it.
[...]
>Recently Gia Dvali of NYU and colleagues reported that a 5-d GR theory
>postdicts accelerated expansion without a cosmological constant. A
>5-d approach also has some potential for unfying GR and EM. Will this
>idea get a decent hearing, or is it ‘something not needed because we
>already have the correct explanation’?
It seems rather harsh to criticise cosmologists for jumping on the
lambda bandwagon and at the same time to imply that they’re likely to
be closed-minded about its possible successors, don’t you think?
Martin
—
Martin Hardcastle Department of Physics, University of Bristol
A little learning is a dangerous thing; / Drink deep, or taste not the
Pierian spring; / There shallow draughts intoxicate the brain …
In article <mt2.0-10042-1054467…@star.bris.ac.uk>,
rlolders…@amherst.edu (Rob Oldershaw) writes:
> hel…@astro.multiCLOTHESvax.de (Phillip Helbig—remove CLOTHES to
reply) wrote in message <news:mt2.0-28028-1054311198@star.bris.ac.uk>…
> > So, though many people are on the Lambda bandwagon now, most didn’t jump
> > on until it was almost too late.
> Hmm, bandwagon. Maybe it’s not so much that the new DE scenario was
> adopted so quickly that bothers me (though it does), but the extent to
> which it became the consensus view. There is definitely a bandwagon
> feeling to it.
Though some people might have jumped on the bandwagon for the wrong
reasons, that alone does not mean that it is not going in the right
direction.
> Not so long ago we had a standard model with inflation that predicted
> a critical density dominated by matter. Then comes the supernova data
> and ‘oops, well, actually we have an energy dominated model with a
> positive cosmological constant that causes accelerated expansion’.
> What do we think, in objective moments, are the chances that the new
> bandwagon model will hold up any better than the recently abandoned
> one? Maybe we should just admit that things are still very uncertain
> in cosmology. We should have open minds, not a bandwagon mentality.
> We should be scientists, not "true believers".
There are two issues here. First, the real "prediction" of inflation is
a flat universe, i.e. Omega + lambda = 1. Without any other evidence in
favour of lambda, then the simplest case is Omega = 1. So that is what
one often heard, though it was never a real prediction. To be fair, I
do remember being at a conference, just a short time before the current
standard model became such, where a Very Famous Inflation Guy said
"Omega (matter) is 1, let’s move on to the next question". Even so, one
shouldn’t confuse the propaganda of certain people with the real
predictions of the theory. (Even if there were not evidence in favour
of lambda, another reason not to go with the simplest case would be
evidence against Omega being 1. Though the situation was a bit
confusing, I think that this WAS ignored a bit too much. While the
actual value of Omega might have been an open question, I think it’s
been clear for a long time that Omega = 1 is too high.) Second, with
"standard model" I meant just the values of Omega and lambda. I’m
willing to bet quite a bit that the current standard model in that sense
will stand up. It is going A LOT further to include some model of
inflation in this standard model.
> Recently Gia Dvali of NYU and colleagues reported that a 5-d GR theory
> postdicts accelerated expansion without a cosmological constant. A
> 5-d approach also has some potential for unfying GR and EM. Will this
> idea get a decent hearing, or is it ‘something not needed because we
> already have the correct explanation’?
Does it make any testable PREdictions?
Martin Hardcastle <m.hardcas…@bristol.ac.uk> wrote in message <news:mt2.0-10336-1054472032@star.bris.ac.uk>…
> It seems rather harsh to criticise cosmologists for jumping on the
> lambda bandwagon and at the same time to imply that they’re likely to
> be closed-minded about its possible successors, don’t you think?
I don’t quite get your point. Are you saying that "jumping on the
lambda bandwagon" shows an open-mindedness? To me a bandwagon
mentality combines undue credulity towards ONE idea (we could hardly
call this open-mindedness) with closedmindedness towards other ideas.
If you wonder whether it’s too "harsh" to criticize cosmologists (and
even more so science journalists) for two things in the same post, I
think not.
Rob
[Mod. note: quoted text trimmed -- mjh]
In article <mt2.0-16180-1054535…@star.bris.ac.uk>,
Rob Oldershaw <rlolders…@amherst.edu> wrote:
>I don’t quite get your point. Are you saying that "jumping on the
>lambda bandwagon" shows an open-mindedness?
I’m suggesting that the (evidence-based) shift over the past few years
from one `standard model’ to another is indeed evidence for
open-mindedness.
> To me a bandwagon
>mentality combines undue credulity towards ONE idea (we could hardly
>call this open-mindedness) with closedmindedness towards other ideas.
You’d have to have some evidence that there were in fact other ideas
that did as good a job at explaining the new observations but were
ignored for me to accept that `closedmindedness’ is at all appropriate
as a description here.
Martin
—
Martin Hardcastle Department of Physics, University of Bristol
A little learning is a dangerous thing; / Drink deep, or taste not the
Pierian spring; / There shallow draughts intoxicate the brain …
hel…@astro.multiCLOTHESvax.de (Phillip Helbig—remove CLOTHES to reply) wrote in message <news:mt2.0-16180-1054535295@star.bris.ac.uk>…
> In article <mt2.0-10042-1054467…@star.bris.ac.uk>,
> rlolders…@amherst.edu (Rob Oldershaw) writes:
> > hel…@astro.multiCLOTHESvax.de (Phillip Helbig—remove CLOTHES to
> reply) wrote in message <news:mt2.0-28028-1054311198@star.bris.ac.uk>…
> Though some people might have jumped on the bandwagon for the wrong
> reasons, that alone does not mean that it is not going in the right
> direction.
Of course! I fully agree.
> There are two issues here. First, the real "prediction" of inflation is
> a flat universe, i.e. Omega + lambda = 1. Without any other evidence in
> favour of lambda, then the simplest case is Omega = 1. So that is what
> one often heard, though it was never a real prediction. To be fair, I
> do remember being at a conference, just a short time before the current
> standard model became such, where a Very Famous Inflation Guy said
> "Omega (matter) is 1, let’s move on to the next question". Even so, one
> shouldn’t confuse the propaganda of certain people with the real
> predictions of the theory. (Even if there were not evidence in favour
> of lambda, another reason not to go with the simplest case would be
> evidence against Omega being 1. Though the situation was a bit
> confusing, I think that this WAS ignored a bit too much. While the
> actual value of Omega might have been an open question, I think it’s
> been clear for a long time that Omega = 1 is too high.)
All along, the leaders of the theoretical pack have insisted on a
"flat universe", no matter how strongly that was contradicted by
observations.
> Second, with "standard model" I meant just the values of Omega and lambda.
>I’m willing to bet quite a bit that the current standard model in
that sense
> will stand up.
2 approx. quotations. ’Those who do not learn the lessons of history
are doomed to repeat their mistakes’ – G.S.? ’Science begins and ends
with observations’ – A.E.
> > Recently Gia Dvali of NYU and colleagues reported that a 5-d GR theory
> > postdicts accelerated expansion without a cosmological constant. A
> > 5-d approach also has some potential for unfying GR and EM. Will this
> > idea get a decent hearing, or is it ‘something not needed because we
> > already have the correct explanation’?
> Does it make any testable PREdictions?
Yes, two good ones. 1. Subtle deviations in Earth/Moon motions.
Experiments are planned and will take 3-5 years. 2. A less steep
curve for SN Ia brightness vs distance than the curve in the Lambda
case. Data collection is underway, but could take 10 years. Science
is a slow process; that’s why the "rush to judgement" mode does not
show scientific wisdom.
In article <mt2.0-15773-1054668…@star.bris.ac.uk>,
wrote
someone whose name is lost in the nested quoting
| Recently Gia Dvali of NYU and colleagues reported that a 5-d GR theory
| postdicts accelerated expansion without a cosmological constant. A
| 5-d approach also has some potential for unfying GR and EM.
and someone else asked
# Does it make any testable PREdictions?
Rob Oldershaw <rlolders…@amherst.edu> then replied
>Yes, two good ones. 1. Subtle deviations in Earth/Moon motions.
>Experiments are planned and will take 3-5 years. [[...]]
Actually, experiments have been underway for 30+ years — the first
laser retroreflector on the moon was put there in 1969, and laser
ranging experiments have been ongoing since then.
The current accuracy of Earth-Moon distances measured by this technique
is on the order of 1-2 centimeters (!), so even very small effects
can be clearly seen in this data. (2cm / 3e5km is a bit better than
1 part in 10^10.) As discussed in the references given below, so far
no deviations from general relativity have been found.
@article
{
Dickey-etal-1994-Science-LRRR-general-review,
author = "J. O. Dickey and P. L. Bender and J. E. Faller
and X X Newhall and R. L. Ricklefs and J. G. Ries
and P. J. Shelus and C. Veillet and A. L. Whipple
and J. R. Wiant and J. G. Williams and C. F. Yoder",
title = "Lunar Laser Ranging:
A Continuing Legacy of the Apollo Program",
journal = "Science",
volume = 265, X-number = "FIXME",
pages = "482–490",
year = 1994, month = "July 22",
snote = "general review of binary pulsar tests of GR",
}
@article
{
Nordtvedt-1999-lunar-laser-ranging-vs-GR,
author = "Kenneth Nordtvedt",
title = "30 years of Lunar Laser Ranging
and the Gravitational Interaction",
journal = "Classical and Quantum Gravity",
volume = 16, number = "12A",
pages = "A101–A112",
year = 1999, month = "December",
note = "maybe available online at
abstract
http://www.iop.org/EJ/S/3/31/D3W0eKTDEk.rPjcuFN2vhg/abstract/
0264-9381/16/12A/305
full paper
http://www.iop.org/EJ/S/3/31/D3W0eKTDEk.rPjcuFN2vhg/article/0
264-9381/16/12A/305/q91b05.ps.gz
",
}
–
— "Jonathan Thornburg (remove -animal to reply)" <jth…@aei.mpg-zebra.de>
Max-Planck-Institut fuer Gravitationsphysik (Albert-Einstein-Institut),
Golm, Germany, "Old Europe" http://www.aei.mpg.de/~jthorn/home.html
"An eye for an eye only makes the whole world blind." — Mahatma Gandhi
Martin Hardcastle <m.hardcas…@bristol.ac.uk> wrote in message <news:mt2.0-7708-1054562637@star.bris.ac.uk>…
> I’m suggesting that the (evidence-based) shift over the past few years
> from one `standard model’ to another is indeed evidence for
> open-mindedness.
That’s a reasonable argument (that there was a considerable amount of
open-mindedness from, say, 1998 to 2000), but how many have retained
that open-mindedness and how many consider the issues "settled"?
> You’d have to have some evidence that there were in fact other ideas
> that did as good a job at explaining the new observations but were
> ignored for me to accept that `closedmindedness’ is at all appropriate
> as a description here.
Let me preface my answer by saying that I have no a priori reason for
rejecting accelerated expansion or a Lambda interpretation.
I have read papers that question the strength of the evidence for the
proposed brightness vs distance relationship (e.g., not enough high
quality data). I have also read papers that accept that relationship,
but suggest alternatives to the accelerated expansion hypothesis.
There are also papers that accept the relationship and accelerated
expansion, but offer alternatives to the Lambda hypothesis.
Before going to the trouble of citing these papers, and probably
engaging in a "fool’s errand", I would want to know how we measure
whether the alternatives to the prevailing ideas "do as good a job at
explaining". Surely not by the numbers of adherents ("The authority
of of a thousand is not worth the humble reasoning of a single
individual" – Galileo). The "because it all fits so well" arguments
do not convince me, given the multitude of skilled theoreticians and
the flexibilities of model-building. What objective measure, at this
very early stage of the game, would suffice? Even worse, how would
one prove that the alternatives were, or were not, ignored?
Rob
rlolders…@amherst.edu (Rob Oldershaw) wrote:
> > Does [5-d GR] make any testable PREdictions?
> Yes, two good ones. 1. Subtle deviations in Earth/Moon motions.
> Experiments are planned and will take 3-5 years. 2. A less steep
> curve for SN Ia brightness vs distance than the curve in the Lambda
> case. Data collection is underway, but could take 10 years. Science
> is a slow process; that’s why the "rush to judgement" mode does not
> show scientific wisdom.
Those Earth/Moon motion deviations had better be damned subtle! Of
course, since we have such good measurements of lunar orbit, I guess
that’s a good place to push the limits of GR.
The real question is whether or not this 5-D theory can account for
CMB observations. If not, then the lamba option seems to have more
going for it.
Rob Oldershaw wrote:
> All along, the leaders of the theoretical pack have insisted on a
> "flat universe", no matter how strongly that was contradicted by
> observations.
Not quite sure how one should take this… The inflation theorists
kept inisting that there should be a flat universe despite strong
observational evidence against them since no one believed in a
cosmological constant and the value or omega was stubbornly
measured as less than 1. After many years the observational
evidence now seems to confirm their prediction with the surprising
corallary that lambda > 0. Seems like an
excellent example of a strong prediction made by inflation that
is now being confirmed observationally! In another thread you’ve
been complaining about the lack of good predictions (versus
postdictions) made in cosmology. Here’s one!
Regards,
Tom McGlynn
In article <mt2.0-15773-1054668…@star.bris.ac.uk>,
rlolders…@amherst.edu (Rob Oldershaw) writes:
> > Though some people might have jumped on the bandwagon for the wrong
> > reasons, that alone does not mean that it is not going in the right
> > direction.
> Of course! I fully agree.
> > There are two issues here. First, the real "prediction" of inflation is
> > a flat universe, i.e. Omega + lambda = 1. Without any other evidence in
> > favour of lambda, then the simplest case is Omega = 1.
> All along, the leaders of the theoretical pack have insisted on a
> "flat universe", no matter how strongly that was contradicted by
> observations.
I don’t follow you here. Today, there are many observations and
combinations of observations which favour a flat universe. As far as I
know, there are none, and have never been any, which disfavour a flat
universe. Of course, if one ASSUMES that lambda is 0, then measures
Omega, one will conclude that the universe is not flat, but the problem
here is the assumption, not the measurement—flatness wasn’t measured,
Omega was, and only with the wrong assumption of a nonexistent
cosmological constant could one arrive at the conclusion that the
universe is not flat (or quite close to being flat).
> > Second, with "standard model" I meant just the values of Omega and lambda.
> > I’m willing to bet quite a bit that the current standard model in
> > that sens will stand up.
> 2 approx. quotations. ’Those who do not learn the lessons of history
> are doomed to repeat their mistakes’ – G.S.? ’Science begins and ends
> with observations’ – A.E.
You seem to be implying that WHATEVER observations there are, we
shouldn’t believe them since they might be revised tomorrow. If so,
then this is a misunderstanding of the way science works. See the
wonderful essay "The Relativity of Wrong" by Isaac Asimov in the book by
the same name:
http://homepage.mac.com/jhjenkins/Asimov/Books/Book378.html
> > Does it make any testable PREdictions?
> Yes, two good ones. 1. Subtle deviations in Earth/Moon motions.
> Experiments are planned and will take 3-5 years. 2. A less steep
> curve for SN Ia brightness vs distance than the curve in the Lambda
> case. Data collection is underway, but could take 10 years. Science
> is a slow process; that’s why the "rush to judgement" mode does not
> show scientific wisdom.
That’s good. However, in both cases I doubt that the current
observational situation is completely understood, since in both cases
there could be many sources of noise. Of course, if the m-z relation
turns out to be what this theory predicts, and it would be awkward to
explain it otherwise, then that’s a plus for this theory. However,
there are OTHER observations which point to the same value of lambda,
and such a theory would have to account for them as well.
On Wed, 4 Jun 2003 10:19:07 GMT, hel…@astro.multiCLOTHESvax.de
(Phillip Helbig—remove CLOTHES to reply) wrote:
>>universe. Of course, if one ASSUMES that lambda is 0, then measures
>Omega, one will conclude that the universe is not flat, but the problem
>here is the assumption, not the measurement—flatness wasn’t measured,
>Omega was, and only with the wrong assumption of a nonexistent
>cosmological constant could one arrive at the conclusion that the
>universe is not flat (or quite close to being flat).
What the supernova data measure is, more or less, what used to be
known as the "deceleration parameter". If you ignore priors (Bayesian
or otherwise) then with the data that were published in 1998-99:
a) A simple least squares (or minimum chi-square in usual astronomers’
jargon) fit indicates a strongly spatially *closed* universe. Nobody
believes that and nobody dwells on it, but it’s a fairly robust result
in the sense that it holds for both major groups’ data and continues
to hold with the addition of data from the small number of susequently
published high z supernovae.
b) The data are at worst marginally inconsistent with an open low
density universe.
Both of the high z supernovae teams calculated solutions in which they
*imposed* the constraint of spatial flatness and came up with values
for Omega_m close to the currently accepted value. Much of the claimed
statistical significance for the hypothesis that Lambda>0 came from
flat solutions. That’s fine as long as you believe that the universe
is exactly flat, but it’s not a test of flatness.
As far as "non-standard" models go, anyone with an alternative
cosmological model to peddle will claim to be able to fit the
supernova data. It’s not that hard to do, especially if you give
yourself at least a couple free parameters to play with.
Here’s one recent example from arxiv.org
<http://www.arxiv.org/abs/astro-ph/0302357>, and here’s another
<http://www.arxiv.org/abs/astro-ph?0303150>. Papers like these come
around every few weeks.
Michael Peck
In article <mt2.0-18619-1054721…@star.bris.ac.uk>,
rlolders…@amherst.edu (Rob Oldershaw) writes:
> Martin Hardcastle <m.hardcas…@bristol.ac.uk> wrote in message
> <news:mt2.0-7708-1054562637@star.bris.ac.uk>…
> > I’m suggesting that the (evidence-based) shift over the past few years
> > from one `standard model’ to another is indeed evidence for
> > open-mindedness.
> That’s a reasonable argument (that there was a considerable amount of
> open-mindedness from, say, 1998 to 2000), but how many have retained
> that open-mindedness and how many consider the issues "settled"?
I still don’t see your point. Some things, like beta decay, are
completely understood. There was a time when it wasn’t, and lots of
wild theories were kicking about, and people had to be very open-minded.
Now it’s understood. That doesn’t mean that they are close-minded. If
an issue is settled, then this does not imply that one is close-minded.
As Sagan said, extraordinary claims demand extraordinary evidence.
Similarly, I think it is perfectly OK to be open-minded about, say, the
mechanism of inflation, but not about the value of Omega.
> I have read papers that question the strength of the evidence for the
> proposed brightness vs distance relationship (e.g., not enough high
> quality data). I have also read papers that accept that relationship,
> but suggest alternatives to the accelerated expansion hypothesis.
> There are also papers that accept the relationship and accelerated
> expansion, but offer alternatives to the Lambda hypothesis.
I think the fact that all of these papers exist in the refereed
literature shows that there is no closed-mindedness in the community as
a whole.
> Before going to the trouble of citing these papers, and probably
> engaging in a "fool’s errand", I would want to know how we measure
> whether the alternatives to the prevailing ideas "do as good a job at
> explaining". Surely not by the numbers of adherents ("The authority
> of of a thousand is not worth the humble reasoning of a single
> individual" – Galileo). The "because it all fits so well" arguments
> do not convince me, given the multitude of skilled theoreticians and
> the flexibilities of model-building. What objective measure, at this
> very early stage of the game, would suffice? Even worse, how would
> one prove that the alternatives were, or were not, ignored?
Occam’s razor.
Let me point out that in the case of the m-z relation for type Ia
supernovae, all the alternative explanations came out AFTER the data
were around, and some of the explanations are rather ad-hoc. This in
itself is reason enough to demand that these alternate explanations not
only get the SN Ia data right—they were built to do that—but also
make other testable predictions. Also, keep in mind that ONE value of
lambda is compatible with the CMB results, the supernova stuff etc etc.
Again, Occam’s razor and the economy of hypotheses.
(To be sure, explanations like "evolution" were always possible, but as
far as I know no-one PREDICTED the observed m-z relation with an
evolution model before the data were in.)
In article <mt2.0-20167-1054740…@star.bris.ac.uk>, Michael Peck
<mpe…@ix.netcom.com> writes:
> On Wed, 4 Jun 2003 10:19:07 GMT, hel…@astro.multiCLOTHESvax.de
> (Phillip Helbig—remove CLOTHES to reply) wrote:
> >>universe. Of course, if one ASSUMES that lambda is 0, then measures
> >Omega, one will conclude that the universe is not flat, but the problem
> >here is the assumption, not the measurement—flatness wasn’t measured,
> >Omega was, and only with the wrong assumption of a nonexistent
> >cosmological constant could one arrive at the conclusion that the
> >universe is not flat (or quite close to being flat).
> What the supernova data measure is, more or less, what used to be
> known as the "deceleration parameter".
Wrong. The deceleration parameter q was interesting in the past because
it is the first interesting term in a series expansion at low redshift.
q = Omega/2 – lambda. The current SN Ia data are at such high redshift
that the deceleration parameter is meaningless in this context. If you
have to express things as a single number, they measure something closer
to Omega – lambda.
> If you ignore priors (Bayesian
> or otherwise) then with the data that were published in 1998-99:
> a) A simple least squares (or minimum chi-square in usual astronomers’
> jargon) fit indicates a strongly spatially *closed* universe.
You are contradicting yourself here. IF they measure q, then they
measure Omega/2 – lambda and, without additional information, one can’t
say whether this implies a spatially closed universe or not.
> Nobody
> believes that and nobody dwells on it, but it’s a fairly robust result
> in the sense that it holds for both major groups’ data and continues
> to hold with the addition of data from the small number of susequently
> published high z supernovae.
As I pointed out in another post in this thread, the combination of all
data, implying a close-to-flat universe, does not rule out a spatially
closed universe (with a large radius of curvature). The supernova data
actually prefer a spatially closed universe (with a smaller radius of
curvature) to a flat universe, though they don’t rule out the latter. I
don’t see how you can say that people ignore this—it’s right there in
the Omega-lambda plots in their papers.
> b) The data are at worst marginally inconsistent with an open low
> density universe.
> Both of the high z supernovae teams calculated solutions in which they
> *imposed* the constraint of spatial flatness and came up with values
> for Omega_m close to the currently accepted value. Much of the claimed
> statistical significance for the hypothesis that Lambda>0 came from
> flat solutions. That’s fine as long as you believe that the universe
> is exactly flat, but it’s not a test of flatness.
I agree. However, they ALSO present plots in the lambda-Omega plane.
These alone imply a positive cosmological constant for reasonable values
of Omega (say, > 0). True, one has to assume flatness to get a specific
Omega value. But the CMB data now show pretty strongly—with no hidden
assumptions—that this is the case. I would say that the fact that the
SN Ia data coupled with the CMB flatness measurement result in a value
for Omega consistent with more direct measurements shows that we are on
the right track.
If I’m a judge, and one witness says he knows where the accused was one
hour BEFORE 10:30 but can’t say where he was after that, and another
witness says that he knows where the accused was one hour after 10:00
but can’t say where he was before that, and both witnesses agree where
the accused was at the times indicated, then I would conclude that the
court knows exactly where the accused was between 10:00 and 10:30, even
though neither witness alone can prove this and even though each
witness’s statement, in itself, gives different constraints. But they
are consistent for a certain period of time, and the joint constraints
provide more information. The situation in cosmology at the moment is
more like having a third witness who says he saw the accused in the same
place at 10:15.
> As far as "non-standard" models go, anyone with an alternative
> cosmological model to peddle will claim to be able to fit the
> supernova data. It’s not that hard to do, especially if you give
> yourself at least a couple free parameters to play with.
> Here’s one recent example from arxiv.org
> <http://www.arxiv.org/abs/astro-ph/0302357>, and here’s another
> <http://www.arxiv.org/abs/astro-ph?0303150>. Papers like these come
> around every few weeks.
Right. But the goal is to fit ALL the data.
On Thu, 5 Jun 2003 09:49:48 GMT, hel…@astro.multiCLOTHESvax.de
(Phillip Helbig—remove CLOTHES to reply) wrote:
>In article <mt2.0-20167-1054740…@star.bris.ac.uk>, Michael Peck
><mpe…@ix.netcom.com> writes:
>> What the supernova data measure is, more or less, what used to be
>> known as the "deceleration parameter".
>Wrong. The deceleration parameter q was interesting in the past because
[...]
>have to express things as a single number, they measure something closer
>to Omega – lambda.
OK, I either misremembered the definition of the deceleration
parameter or correctly remembered a reference with a typo. But as long
as we’re discussing rough caricatures of what the data measure,
Perlmutter et al.’s subset C (which they took as their best fit) had
an unconstrained least squares solution for (Omega_m, Omega_l) of
(0.73, 1.32). The best flat solution was (0.28, 0.72). Connect the
dots and you get an equation of Omega_m-3/4 Omega_l ~= -0.26 which is
what they state in their abstract is the approximate relationship
being "constrained" (i.e. measured).
>> If you ignore priors (Bayesian
>> or otherwise) then with the data that were published in 1998-99:
>> a) A simple least squares (or minimum chi-square in usual astronomers’
>> jargon) fit indicates a strongly spatially *closed* universe.
>You are contradicting yourself here. IF they measure q, then they
>measure Omega/2 – lambda and, without additional information, one can’t
No I’m not. I’m quite familiar with what the data indicate because for
starters I’ve done the curve fitting exercise myself. Again,
Perlmutter et al.s unconstrained least squares fit for their subset C
gave (Omega_m, Omega_l) ~= (0.73, 1.32). If I’m not mistaken that is a
strongly spatially closed universe.
I’ve tried adding Riess et al.’s data, as well as SN 1997ff (the
supernova at z ~= 1.7) and a couple additional data points published
this year, and came up with an unconstrained fit of (.63, 1.29) and a
flat fit of (.28, .72). So, as I said the fit is fairly robust with
respect to sample selection.
>closed universe (with a large radius of curvature). The supernova data
>actually prefer a spatially closed universe (with a smaller radius of
>curvature) to a flat universe, though they don’t rule out the latter. I
>don’t see how you can say that people ignore this—it’s right there in
>the Omega-lambda plots in their papers.
There’s a semantic distinction between "nobody dwells on" and "people
ignore", so you are misrepresenting what I said. In fact though I
can’t find a direct mention of the unconstrained fit in the text of
Perlmutter et al.’s paper. It’s in a table and visible in plots of
joint confidence intervals, but it’s not an issue they dwell on.
Instead they discuss the supporting evidence (mostly preliminary at
the time) for a flat, low matter density universe.
A few people have taken seriously the idea of a slightly spatially
closed universe or nontrivial topology based primarily on WMAP data,
but I seriously doubt you’ll find anyone willing to entertain the idea
that Omega_tot ~ 2, which again is what a simple minded interpretation
of the SnIa data point to. The WMAP team’s best fit model had
Omega_tot ~= 1.02 as I recall.
>Omega value. But the CMB data now show pretty strongly—with no hidden
>assumptions—that this is the case. I would say that the fact that the
>SN Ia data coupled with the CMB flatness measurement result in a value
>for Omega consistent with more direct measurements shows that we are on
>the right track.
Yes, I agree. And part of the reason I question the emphasis on the
supernovae data is that it didn’t by itself make so compelling a case
for a cosmological constant that people would have instantly hopped on
the bandwagon had they not had other good reasons to do so.
>Right. But the goal is to fit ALL the data.
Yes, of course, and that’s why my comment was more or less dismissive.
People will do detailed statistical analyses of the supernova data
within the context of their alternative cosmologies, but I haven’t
seen anyone confront other cosmologically significant datasets in the
same level of detail. The supernova data is the easiest hurdle to
overcome, which is why people tend to tackle it first.
Michael Peck
- Hide quoted text — Show quoted text -
Tom McGlynn <t…@lheapop.gsfc.nasa.gov> wrote in message <news:mt2.0-18619-1054721919@star.bris.ac.uk>…
> Rob Oldershaw wrote:
> > All along, the leaders of the theoretical pack have insisted on a
> > "flat universe", no matter how strongly that was contradicted by
> > observations.
> Not quite sure how one should take this… The inflation theorists
> kept inisting that there should be a flat universe despite strong
> observational evidence against them since no one believed in a
> cosmological constant and the value or omega was stubbornly
> measured as less than 1. After many years the observational
> evidence now seems to confirm their prediction with the surprising
> corallary that lambda > 0. Seems like an
> excellent example of a strong prediction made by inflation that
> is now being confirmed observationally! In another thread you’ve
> been complaining about the lack of good predictions (versus
> postdictions) made in cosmology. Here’s one!
> Regards,
> Tom McGlynn
The "Consensus Model", like its predecessors over the last few
decades, looks highly manipulated to me, and fey. Theorists play with
their plastic models until they them make compatible with the most
recent observations (Inflation is always protected. If it cannot be,
the observations are belittled). Then they say, ‘Voila, Precision
Cosmology’. Predicting a "flat universe" is a valid prediction, but
it has not been "confirmed", at least not by objective scientific
standards. Egos and emotions have interfered with that desired
objectivity.
And by the way, where did all the magnetic monopoles go? A
conveniently forgotten primary prediction of Inflation that has failed
miserably.
Let me give my “ 2cents” on this subject and I hope experts can
correct me.a
- Hide quoted text — Show quoted text -
> Hmm, bandwagon. Maybe it’s not so much that the new DE scenario was
> adopted so quickly that bothers me (though it does), but the extent to
> which it became the consensus view. There is definitely a bandwagon
> feeling to it.
> Not so long ago we had a standard model with inflation that predicted
> a critical density dominated by matter. Then comes the supernova data
> and ‘oops, well, actually we have an energy dominated model with a
> positive cosmological constant that causes accelerated expansion’.
> What do we think, in objective moments, are the chances that the new
> bandwagon model will hold up any better than the recently abandoned
> one? Maybe we should just admit that things are still very uncertain
> in cosmology. We should have open minds, not a bandwagon mentality.
> We should be scientists, not "true believers".
> Recently Gia Dvali of NYU and colleagues reported that a 5-d GR theory
> postdicts accelerated expansion without a cosmological constant. A
> 5-d approach also has some potential for unfying GR and EM. Will this
> idea get a decent hearing, or is it ‘something not needed because we
> already have the correct explanation’?
even before the supernova results I thought there were some indicates
that the universe might be acceralting. More example Krauss & Turner
in 94/95 published the paper “cosmological constant is back” arguing
that Lambda must be 0 in order to reconcile the reconcile the age of
the universe with the ages of the oldest stars. Also I thought that R
Daly’s work with radio galaxies showed evidence for accelerating
universe and this was announced one year before the SN results. On the
flip side I had thought that earlier results from strong gravitational
lensing ruled out Lambda (don’t know at what % c.l.) can someone tell
me what the latest results indicate? anyhow so I don’t think that SN
results were a complete shock.
Melroy
[Mod. note: presumably the poster means Lambda > 0 in the above... --mjh]
In article <mt2.0-24156-1054806…@star.bris.ac.uk>,
hel…@astro.multiCLOTHESvax.de (Phillip Helbig—remove CLOTHES to
reply) writes:
> If I’m a judge, and one witness says he knows where the accused was one
> hour BEFORE 10:30 but can’t say where he was after that, and another
> witness says that he knows where the accused was one hour after 10:00
> but can’t say where he was before that, and both witnesses agree where
> the accused was at the times indicated, then I would conclude that the
> court knows exactly where the accused was between 10:00 and 10:30, even
> though neither witness alone can prove this and even though each
> witness’s statement, in itself, gives different constraints. But they
> are consistent for a certain period of time, and the joint constraints
> provide more information. The situation in cosmology at the moment is
> more like having a third witness who says he saw the accused in the same
> place at 10:15.
I’m a physicist, not a lawyer. Let me rephrase that! :-)
If I’m a judge, and one witness says he knows where the accused was at
10:30 but can’t say how long he had been there, and another witness says
that he knows where the accused was at 10:00 but can’t say how long he
remained there, and both witnesses agree where the accused was at the
times indicated, and it takes several hours to travel to where the
accused claims he was at 10:15, then I would conclude that the court can
be pretty sure where the accused was between 10:00 and 10:30, even
though neither witness alone can prove this and even though each
witness’s statement, in itself, gives different constraints. But they
are consistent for a certain period of time, and the joint constraints
provide more information. The situation in cosmology at the moment is
more like having a third witness who says he saw the accused in the same
place at 10:15.
In article <mt2.0-25488-1054830…@star.bris.ac.uk>, Michael Peck
<mpe…@ix.netcom.com> writes:
> OK, I either misremembered the definition of the deceleration
> parameter or correctly remembered a reference with a typo. But as long
> as we’re discussing rough caricatures of what the data measure,
> Perlmutter et al.’s subset C (which they took as their best fit) had
> an unconstrained least squares solution for (Omega_m, Omega_l) of
> (0.73, 1.32). The best flat solution was (0.28, 0.72). Connect the
> dots and you get an equation of Omega_m-3/4 Omega_l ~= -0.26 which is
> what they state in their abstract is the approximate relationship
> being "constrained" (i.e. measured).
OK, we agree here.
> >You are contradicting yourself here. IF they measure q, then they
> >measure Omega/2 – lambda and, without additional information, one can’t
> No I’m not. I’m quite familiar with what the data indicate because for
> starters I’ve done the curve fitting exercise myself. Again,
> Perlmutter et al.s unconstrained least squares fit for their subset C
> gave (Omega_m, Omega_l) ~= (0.73, 1.32). If I’m not mistaken that is a
> strongly spatially closed universe.
Yes, but as I said this is NOT because they "measured a certain value
for q and this implies a closed universe" but rather because this is the
best fit in the lambda-Omega plane.
The important question, however, is not "what is the best fit" t rather
"what region of the parameter space is allowed at a reasonable
confidence level". A lambda=0.7, Omega=0.3 model IS allowed. The SN Ia
data, by themselves, don’t speak for this model nor for a flat universe.
But no-one claims that they do.
There are several cosmological tests. They each produce regions of
allowed parameter space in the lambda-Omega plane. They ALL overlap in
a region which is compatible with all tests. That’s where our universe
is. There is no conflict. There would only be a conflict if a
particular test had an allowed region which didn’t overlap with the
overlap from the rest of the tests. But this is not the case. See my
(revised) remark about witnesses in the courtroom.
> >closed universe (with a large radius of curvature). The supernova data
> >actually prefer a spatially closed universe (with a smaller radius of
> >curvature) to a flat universe, though they don’t rule out the latter. I
> >don’t see how you can say that people ignore this—it’s right there in
> >the Omega-lambda plots in their papers.
> There’s a semantic distinction between "nobody dwells on" and "people
> ignore", so you are misrepresenting what I said. In fact though I
> can’t find a direct mention of the unconstrained fit in the text of
> Perlmutter et al.’s paper. It’s in a table and visible in plots of
> joint confidence intervals, but it’s not an issue they dwell on.
> Instead they discuss the supporting evidence (mostly preliminary at
> the time) for a flat, low matter density universe.
Well, their best-fit model is COMPLETELY RULED OUT by a variety of
cosmological tests (CMB anisotropy (Hi Ted!), "direct" measurements of
Omega) so there is no point in dwelling on it. One would expect, of
course, that a) as their sample size increases and b) as more and more
sources of error are understood that they will converge on the real
cosmological model. We’ll see. It would only be worth dwelling on if
the "standard model" was NOT compatible with their allowed region, but
it is.
Until recently, CMB data alone could only measure, roughly, Omega +
lambda. Thus, it was good to have another test which measured another
combination, since combining the two gives tighter constraints.
> A few people have taken seriously the idea of a slightly spatially
> closed universe or nontrivial topology based primarily on WMAP data,
> but I seriously doubt you’ll find anyone willing to entertain the idea
> that Omega_tot ~ 2, which again is what a simple minded interpretation
> of the SnIa data point to. The WMAP team’s best fit model had
> Omega_tot ~= 1.02 as I recall.
Yes, but there are lots of other reasons for rejecting Omega=2 or even
Omega=1. Why place so much emphasis on the SN Ia results? There are
other cosmological tests as well. (The SN Ia results are interesting
because they were the first good cosmological test which BY THEMSELVES
indicated lambda > 0.)
> Yes, I agree. And part of the reason I question the emphasis on the
> supernovae data is that it didn’t by itself make so compelling a case
> for a cosmological constant that people would have instantly hopped on
> the bandwagon had they not had other good reasons to do so.
Quite the opposite, as I said above. By themselves—and this is
rare—they DO indicate a positive cosmological constant. Combined with
flatness (from CMB) they indicate a low Omega, consistent with a host of
other results.
> Yes, of course, and that’s why my comment was more or less dismissive.
> People will do detailed statistical analyses of the supernova data
> within the context of their alternative cosmologies, but I haven’t
> seen anyone confront other cosmologically significant datasets in the
> same level of detail. The supernova data is the easiest hurdle to
> overcome, which is why people tend to tackle it first.
Also, since they indicate a positive cosmological constant by
themselves, they are an attractive target to shoot down for those who
would like to have a zero cosmological constant.