Galaxy cluster at z=1.4 challenges BBT

21 views
Skip to first unread message

r...@firstpr.com.au

unread,
Mar 3, 2005, 2:23:35 PM3/3/05
to
The Australian Broadcasting Corporation radio news this morning
carried an interview with Chris Mullis about a galaxy cluster,
discovered initially with X-rays and then confirmed
spectroscopically with the VLT.

http://www.astro.lsa.umich.edu/~cmullis/research/xmmuj2235/

Discovery of an X-ray-Luminous Galaxy Cluster at z=1.4
http://arxiv.org/abs/astro-ph/0503004

The interview should soon be available at:
http://www.abc.net.au/rn/talks/brkfast/

Chris Mullis et al. say their technique could be used to find
many such objects with relative ease.

. . . XMMUJ2235.3-2557 is likely more massive
than RDCS1252-29 (previously the most massive, distant
cluster known at z = 1.24).

They estimate the cluster is 9 billion light years away. In
the interview Chris Mullis indicated that he thought the cluster
must have begun forming 11 billion years ago. He referred to
the age of the Universe as being 13.7 billion years.

He indicated that this cluster is a major challenge to theories
of galaxy formation - which will need to be revised in order to
account for them forming and collecting themselves into clusters
so rapidly.

I think that a better approach would be to question the Big Bang
Theory. All we need to disprove it is a mechanism by which
light is redshifted 1 part in about 15 billion per year of
travel in the intergalactic plasma. See

http://astroneu.com/plasma-redshift-1/

for such theories and discussion of problems with the BBT and
some alternative theories, concerning:

Heating and acceleration of stellar coronae and winds.

How galaxy clusters do not resemble the shapes one would
expect to result from gravitational formation, but rather
the liquid between bubbles in a foam. I propose the void
IGM is heated to extreme temperatures by a plasma redshift
(I plan to reformulate this as sparse particle redshift)
of distant starlight, creating high enough pressures,
despite the very low density, to push galaxies (and their
more massive surrounding coronae) into the cluster or
supercluster shapes we observe.

Plasma (sparse particle) redshift occurring close to quasars -
so the Lyman forest is local to the quasar. This would also
explain the failure to find the transverse proximity effect
with a foreground quasar - a failure which directly challenges
the Doppler / expansion assumption about redshift on which the
Big Bang Theory is based.

A theory of dark matter in galactic halos consisting of black
dwarfs and their collision fragments. This would be
impossible if the galaxies are less than 14 billion years
old or so, since (according to conventional theories, which
I think are probably fine) stars would take too long to cool.
However, if we we abandon the BBT and consider that galaxies
are probably much older than this, with some as-yet unknown
source of matter/energy, then they could be old enough to
generate collapsed and cooled stars with a mass exceeding that
of the luminous stars. I propose how these would eventually
wind up in widely dispersed elliptical orbits around a spiral
galaxy - so explaining the long-standing problem of galactic
rotation curves.

Pointers to Jerry Jensen's critique of the conventional
analysis of supernova light curves. This conventional finding
of time dilation would need to be disproven in order to
abandon the BBT.

- Robin http://astroneu.com http://www.firstpr.com.au

jacob navia

unread,
Mar 11, 2005, 6:51:51 AM3/11/05
to
One more data point in the same direction:

http://www.universetoday.com/am/publish/galaxies_early_universe.html
What did the universe look like when it was only 2 to 3 billion years=20
old? Astronomers used to think it was a pretty simple place containing=20
relatively small, young star-forming galaxies. Researchers now are=20
realizing that the truth is not that simple. Even the early universe was=20
a wildly complex place. Studying the universe at this early stage is=20
important in understanding how the galaxies near us were assembled over=20
time.

Jiasheng Huang (Harvard-Smithsonian Center for Astrophysics) said, "It=20
looks like vegetable soup! We're detecting galaxies we never expected to=20
find, having a wide range of properties we never expected to see."

"It's becoming more and more clear that the young universe was a big zoo=20
with animals of all sorts," said Ivo Labb=EF=BF=BD (Observatories of the=20
Carnegie Institution of Washington), lead author on the study announcing=20
this result.

Using the Infrared Array Camera (IRAC) aboard NASA's Spitzer Space=20
Telescope, the astronomers searched for distant, red galaxies in the=20
Hubble Deep Field South-a region of the southern sky previously observed=20
by the Hubble Space Telescope.

Their search was successful. The IRAC images displayed about a dozen=20
very red galaxies lurking at distances of 10 to 12 billion light-years.=20
Those galaxies existed when the universe was only about 1/5 of its=20
present age of 14 billion years. Analysis showed that the galaxies=20
exhibit a large range of properties.

"Overall, we're seeing young galaxies with lots of dust, young galaxies=20
with no dust, old galaxies with lots of dust, and old galaxies with no=20
dust. There's as much variety in the early universe as we see around us=20
today," said Labb=EF=BF=BD.

The team was particularly surprised to find a curious breed of galaxy=20
never seen before at such an early stage in the universe--

*old, red galaxies that had stopped forming new stars altogether.*

Those galaxies had rapidly formed large numbers of stars much earlier in=20
the universe's history, raising the question of what caused them to=20
"die" so soon.

The unpredicted existence of such "red and dead" galaxies so early in=20
time challenges theorists who model galaxy formation.

"We're trying to understand how galaxies like the Milky Way assembled=20
and how they got to look the way they appear today," said Giovanni Fazio=20
(CfA), a co-author on the study. "Spitzer offers capabilities that=20
Hubble and other instruments don't, giving us a unique way to study very=20
distant galaxies that eventually became the galaxies we see around us now=
.."

The study will be published in an upcoming issue of The Astrophysical=20
Journal Letters.

This press release is being issued in conjunction with the Observatories=20
of the Carnegie Institution of Washington.

---------------------------------------------
Emphasis in
*old, red galaxies that had stopped forming new stars altogether.*
added by me.

How can an old galaxy form and die in only 2 Bill years?
Assuming a rotation rate identical to the milky way, it has
the time to make only 8 turns abd it is already dead and old...

Every months we have discoveries like this:
*There's as much variety in the early universe as we see around us today*

The scopes have arrived at the immediate neighborurhood of the
supposed big bang and there is not the slightest hint of a bang
to see.

jacob

Max Keon

unread,
Mar 14, 2005, 11:01:12 AM3/14/05
to
jacob navia wrote:
>
> One more data point in the same direction:

> http://www.universetoday.com/am/publish/galaxies_early_universe.html
> What did the universe look like when it was only 2 to 3 billion years

> old? Astronomers used to think it was a pretty simple place containing

> relatively small, young star-forming galaxies. Researchers now are

> realizing that the truth is not that simple. Even the early universe was

> a wildly complex place.

The complex evolutionary state of the "early" universe described
in your post certainly can't be logically justified within the
scope of the BBT. While researchers are out there realizing the
"truth", they should also realize that continually upgrading the
theory to incorporate emerging, and damning evidence, will only
serve to further embarrass the entire physics community.

One theory has so far passed every test, with flying colors.

-----

Max Keon

xant...@well.com

unread,
Mar 15, 2005, 4:13:39 AM3/15/05
to
jacob navia wrote:

> How can an old galaxy form and die in only 2 Bill
> years?

One can envision lots of mechanisms; all it takes is
something to sweep the galaxy clear of dust and gas
from which to create more stars.

A glancing collision with a larger galaxy could do
that.

The "big enough" jet from another galaxy's massive
black hole could perhaps do that, hosing away the
dust but leaving the existing stars.

Flying through a dust-thick extensive unconsolidated
cloud at relativistic relative velocities could
probably do that too; the stars would bully on
through, but the galactic dust between them would be
stopped in its tracks.

Once you stop forming stars, the blue ones die their
quick deaths, and soon only the longer-lived red
ones remain, and the galaxy looks "old" only because
it no longer has any surviving blue stars to make it
look "young". That is just as should be expected if
you take out the loose dust by _any_ mechanism.

There's nothing "contradictory to the big bang
theory" about finding a _few_ anamolous objects.

The universe is plenty big enough for a few highly
unlikely happenings nontheless to have occurred.

The more data we find, the more fractal-like the
universe seems, and fractals provide lots of room
for extremal cases.

Finding the anomalous objects to be the _prevalent_
types would certainly be worrisome to the BBT; any
theory which finds mostly exceptions to its
predictions hasn't long to live.

> Assuming a rotation rate identical to the milky

> way, it has the time to make only 8 turns a[n]d it


> is already dead and old...

Which is completely irrelevant to the issue.

> Every months we have discoveries like this:
> *There's as much variety in the early universe as
> we see around us today*

Yep; like every other theory of the real world,
things grow more interesting the better your ability
gets to resolve details in the data.

That doesn't necessarily invalidate the larger
theory.

What would be absolutely mind boggling would be if
the _opposite_ were the case, if all the new
instruments' resolving power were a waste of effort,
because nothing new or unexpected or interesting at
all were there to be seen.

> The scopes have arrived at the immediate

> neighborhood of the supposed big bang and there


> is not the slightest hint of a bang to see.

You mean besides the cosmic microwave background
radiation that already confirms the BBT to several
decimals of precision?

That _is_ the Big Bang, granted you aren't going to
see it in an optical telescope, which would be
looking among the wrong
wavelengths for the Big Bang in any case.

Were you looking for some _other_ Big Bang?

If so, why?

One more than suffices, I would think.

xanthian, amused that _every_ theory finds its
_inevitable_ gathering of naysayers.

jacob navia

unread,
Mar 15, 2005, 6:14:19 AM3/15/05
to
xant...@well.com wrote:
> jacob navia wrote:
>
>
>>How can an old galaxy form and die in only 2 Bill
>>years?
>
>
> One can envision lots of mechanisms; all it takes is
> something to sweep the galaxy clear of dust and gas
> from which to create more stars.
>
> A glancing collision with a larger galaxy could do
> that.
>
> The "big enough" jet from another galaxy's massive
> black hole could perhaps do that, hosing away the
> dust but leaving the existing stars.
>
> Flying through a dust-thick extensive unconsolidated
> cloud at relativistic relative velocities could
> probably do that too; the stars would bully on
> through, but the galactic dust between them would be
> stopped in its tracks.
>

Relativistic relative velocities???

One of the fastest moving galaxies (NGC 1427A ) is falling into the
Fornax cluster at ... 600 Km/sec.
(http://hubblesite.org/newscenter/newsdesk/archive/releases/2005/09)

To accelerate *A GALAXY* to relativistic speeds would require so
much energy that I can safely bet that there will never be an
observation of such an object. Besides, the high speed of the
galaxy should be *noticable* in its spectra, either in an increased
or decreased red/blue shift.

This looks like a desperate explanation. Yes; it is *possible* but...
is it likely?


> Once you stop forming stars, the blue ones die their
> quick deaths, and soon only the longer-lived red
> ones remain, and the galaxy looks "old" only because
> it no longer has any surviving blue stars to make it
> look "young". That is just as should be expected if
> you take out the loose dust by _any_ mechanism.
>

Just 2 Billion years?
A sun-like star lives 10 Billion years. Even if there weren't
any new star formations, sun-like stars should go on for quite
a while. Supposing this "encounter at relativistic speeds" takes
place 1 Bill years after the bang, we should see a lot of blue
stars 1 Billion years later, not enough to make the galaxy red.


> There's nothing "contradictory to the big bang
> theory" about finding a _few_ anamolous objects.
>

Sorry but this is *one* from many examples discovered.
Old galaxies with iron in it, galaxy clusters at
9 Billion years
(http://www.eso.org/outreach/press-rel/pr-2005/pr-04-05.html)
and *many* others.

> The universe is plenty big enough for a few highly
> unlikely happenings nontheless to have occurred.
>

Probably. The point is, the more "unlikely" events we find, the
more unlikely the theory becomes, that is my point. I am not
saying that this is 100% impossible to explain with BB theory, just
that BB theory becomes more and more unlikely as more facts are
known.

Ptolomeus rotating spheres model could ALWAYS accomodate new
observations by making a NEW sphere. But at some point people
just preferred the new model because it was simpler...

Now, the big problem here is that there isn't any Galileo
around :-)

> The more data we find, the more fractal-like the
> universe seems, and fractals provide lots of room
> for extremal cases.
>

> Finding the anomalous objects to be the _prevalent_
> types would certainly be worrisome to the BBT; any
> theory which finds mostly exceptions to its
> predictions hasn't long to live.
>

That's exactly my point.

>
>>Assuming a rotation rate identical to the milky
>>way, it has the time to make only 8 turns a[n]d it
>>is already dead and old...
>
>
> Which is completely irrelevant to the issue.
>
>

No. Galaxies are flat, and to get flat they have to
rotate for some time to flatten themselves isn't it?

>>Every months we have discoveries like this:
>>*There's as much variety in the early universe as
>>we see around us today*
>
>
> Yep; like every other theory of the real world,
> things grow more interesting the better your ability
> gets to resolve details in the data.
>

I agree

> That doesn't necessarily invalidate the larger
> theory.
>
> What would be absolutely mind boggling would be if
> the _opposite_ were the case, if all the new
> instruments' resolving power were a waste of effort,
> because nothing new or unexpected or interesting at
> all were there to be seen.
>
>
>>The scopes have arrived at the immediate
>>neighborhood of the supposed big bang and there
>>is not the slightest hint of a bang to see.
>
>
> You mean besides the cosmic microwave background
> radiation that already confirms the BBT to several
> decimals of precision?
>

There was a discussion in sci.astro about "overaveraging"
and the whole "wrinkles in the face of god"
story. I remain a sceptic about that. But yes, there
is no alternative explanation to the cosmic background.

The problem is that it could very well be that we just
do not know what the Cosmic Background *is*, and we see it
as we can: as a "BB " relic.

> That _is_ the Big Bang, granted you aren't going to
> see it in an optical telescope, which would be
> looking among the wrong
> wavelengths for the Big Bang in any case.
>

I am not so stupid to believe we could "see" the big bang.
Of course not. But its immediate neighborhood should have
*some* marks of such a "bang" having happened relatively shrtly,
i.e. 500 Mill years...

> Were you looking for some _other_ Big Bang?
>
> If so, why?
>
> One more than suffices, I would think.
>
> xanthian, amused that _every_ theory finds its
> _inevitable_ gathering of naysayers.

And, an established theory will find its inevitable
gathering of people that will stick to it no matter what.

jacob

xant...@well.com

unread,
Mar 15, 2005, 9:16:49 AM3/15/05
to
[Mod. note: please could all posters try to remain polite and focussed
on the scientific issues rather than on scoring points -- mjh]

jacob navia wrote:
> xant...@well.com wrote:
>> jacob navia wrote:

>>> How can an old galaxy form and die in only 2
>>> Bill years?

>> One can envision lots of mechanisms; all it takes
>> is something to sweep the galaxy clear of dust
>> and gas from which to create more stars.

>> A glancing collision with a larger galaxy could
>> do that.

>> The "big enough" jet from another galaxy's
>> massive black hole could perhaps do that, hosing
>> away the dust but leaving the existing stars.

>> Flying through a dust-thick extensive
>> unconsolidated cloud at relativistic relative
>> velocities could probably do that too; the stars
>> would bully on through, but the galactic dust
>> between them would be stopped in its tracks.

> Relativistic relative velocities???

Notice that I supplied _three_ mechanisms, and
rather than respond to the grist of any of them,
you wasted your effort jumping on a badly chosen
word. This isn't productive to understanding the
phenomena, at all.

> One of the fastest moving galaxies (NGC 1427A ) is
> falling into the Fornax cluster at ... 600 Km/sec.
> (http://hubblesite.org/newscenter/newsdesk/archive/releases/2005/09)

You'd have a lot less misunderstandings if you'd
resist jumping crowingly and unthinkingly on a
misused _word_, and work out the _math_ for
yourself. If a 600 kmps galaxy hits a 0 kps dust
cloud, a good guess would be that the galaxy's
internal dust falls behind at 300 kmps. How long
does it take to sweep such a galaxy clean of dust?

Well, a galaxy is roughly 10^18 km across, if I
haven't lost a decimal, and 300 kmps is roughly
10^10 km/year, so 10^8 years would suffice, only a
tenth of a billion: lots of time to have cleaned out
a galaxy 2 billion years old.

> To accelerate *A GALAXY* to relativistic speeds
> would require so much energy that I can safely bet
> that there will never be an observation of such an
> object. Besides, the high speed of the galaxy
> should be *noticable* in its spectra, either in an
> increased or decreased red/blue shift.

Yada, yada, yada... You're arguing with the hand.

> This looks like a desperate explanation. Yes; it
> is *possible* but... is it likely?

If you had bothered to work out the math for
yourself using the data _you_ supplied, you'd know
the answer is in the affirmative. This habit of
failing to do your own homework isn't helping you
understand the issues, at all.

>> Once you stop forming stars, the blue ones die
>> their quick deaths, and soon only the
>> longer-lived red ones remain, and the galaxy
>> looks "old" only because it no longer has any
>> surviving blue stars to make it look "young".
>> That is just as should be expected if you take
>> out the loose dust by _any_ mechanism.

> Just 2 Billion years?

> A sun-like star lives 10 Billion years. Even if
> there weren't any new star formations, sun-like
> stars should go on for quite a while. Supposing
> this "encounter at relativistic speeds" takes
> place 1 Bill years after the bang, we should see a
> lot of blue stars 1 Billion years later, not
> enough to make the galaxy red.

Give me strength. Go look up the lifetime for blue
stars, please. You are off by factors containing
multiple digits.

>> There's nothing "contradictory to the big bang

>> theory" about finding a _few_ anomalous objects.

> Sorry but this is *one* from many examples
> discovered. Old galaxies with iron in it, galaxy
> clusters at 9 Billion years
> (http://www.eso.org/outreach/press-rel/pr-2005/pr-04-05.html)
> and *many* others.

Yep, but lots of _single_, different, anomalies
aren't an issue; among "billions of billions" of
galaxies, even the most (within reason) unlikely
events have had time to happen "somewhere, once" by
purely statistical arguments based simply on the
huge number of objects the universe contains. It is
when they are happening "everywhere, often" that you
start to worry about your theory.

>> The universe is plenty big enough for a few
>> highly unlikely happenings nontheless to have
>> occurred.

> Probably. The point is, the more "unlikely" events
> we find, the more unlikely the theory becomes,
> that is my point.

No, the more _identical_ unlikely events we find,
the more danger those events present to the theory.

The more detailed our data investigating capability
becomes, the more _kinds of_ *unique* unlikely
events we can pull out of the data, and that means
there are lots more _unique_ anomalies reported with
every sensing device improvement; but that, again,
is merely an expected result.

Worry when some one contradicting _kind of_ anomaly
starts to dominate the findings, not when new kinds
of anomalies arrive with each sensor improvement.
That's exactly why we pay the big bucks for the
better sensors: so we can winkle the strange and
interesting stuff out of the mostly bland data.

> I am not saying that this is 100% impossible to
> explain with BB theory, just that BB theory
> becomes more and more unlikely as more facts are
> known.

Think about what you just wrote.

> But at some point people just preferred the new
> model because it was simpler...

So far as I know, the Big Bang Theory, with
Inflation, remains by far the most parsimoneous
explanation of the universe as we see it today, and
all the statistical flukes in the world aren't
contradicting that impression; numerous statistical
flukes are an _expected finding_ in such a large
data set.

> Now, the big problem here is that there isn't any
> Galileo around :-)

I think we need more pressingly to bring back
William of Occam.

>> The more data we find, the more fractal-like the
>> universe seems, and fractals provide lots of room
>> for extremal cases.

>> Finding the anomalous objects to be the _prevalent_
>> types would certainly be worrisome to the BBT; any
>> theory which finds mostly exceptions to its
>> predictions hasn't long to live.

> That's exactly my point.

But you fail to do the math, and so miss making your
point; gut arguments don't work well for arguing
where the numbers involved defy correct intuitions.

>>> Assuming a rotation rate identical to the milky
>>> way, it has the time to make only 8 turns a[n]d it
>>> is already dead and old...

> > Which is completely irrelevant to the issue.

> No. Galaxies are flat,

That is not true in general; spiral galaxies are
flat, elliptical galaxies are, surprise, elliptical.

At the distances where a whole galaxy takes up a
couple of pixels on your sensing device, deciding
whether a galaxy is a certain shape is
"interesting", and has to be based on its spectrum,
not its outline in your sensing device.

But in the cases under discussion, the spectral
anomalies are _already_ an issue, and make trusting
spectral analysis to determine really subtle stuff
like "is the spectral spread appropriate for an
elliptical galaxy, or for a spiral one", a dodgy
choice. That's the level of math I'll _happily_
defer to the experts.

> and to get flat they have to rotate for some time
> to flatten themselves isn't it?

Which is still irrelevant to the issue of why they
are blue light deprived. Most certainly, they don't
have to "flatten" to be galaxies-at-all.

>>> The scopes have arrived at the immediate
>>> neighborhood of the supposed big bang and there
>>> is not the slightest hint of a bang to see.

>> You mean besides the cosmic microwave background
>> radiation that already confirms the BBT to
>> several decimals of precision?

> There was a discussion in sci.astro about
> "overaveraging" and the whole "wrinkles in the
> face of god" story. I remain a sceptic about that.
> But yes, there is no alternative explanation to
> the cosmic background.

Then why on earth did you make your previous
statement?

> The problem is that it could very well be that we
> just do not know what the Cosmic Background *is*,
> and we see it as we can: as a "BB " relic.

Yep, that's this big problem with science, you get
this raw data, and then you come up with some
"cooked" interpretation of that data. That's where
we get the documents called "PhD theses".

While that interpretation retains good predictive
capability, it remains "the accepted theory". As
soon as it loses that ability, it lands "in the
dustbin of history". Right now, Big Bang theory is
in no danger at all of landing in the dustbin,
however much it annoys some folks religious or
math-deprived intuitive notions.

>> That _is_ the Big Bang, granted you aren't going
>> to see it in an optical telescope, which would be
>> looking among the wrong wavelengths for the Big
>> Bang in any case.

> I am not so stupid to believe we could "see" the
> big bang.

And yet you go right on to insist on exactly that:

> Of course not. But its immediate neighborhood
> should have *some* marks of such a "bang" having
> happened relatively shrtly, i.e. 500 Mill years...

You write like you expect to see the Big Bang from
some exterior vantage point. We're inside it, and
the cosmic background microwave radiation is there
to be seen in every direction we look. It precisely
satisfies your need for "some marks".

It helps to remember, too, that the Universe spent a
good while after the Big Bang, being totally opaque.

300,000 years sticks in my mind, and I refuse to
go look it up to confirm that. Astronomy is an
"also interesting" for me, not something that
dominates my intellectual life, so my
willingness to invest time into it is pretty
limited.

Between the point where it changed phase to be
transparent, and the point where gravity had enough
time to work to produce some stars, there's probably
a big bunch of "nothing at all to look at" out
there, so don't expect to be finding the
astronomical equivalent of "fossils of soft bodied
organisms" where there are none.

It's like the hole in the data before 10^-34 seconds
or whatever the figure is; some stuff is plain
impossible ever to "see"; _lack_ of data isn't much
persuasive in favor of _any_ theory.

> And, an established theory will find its
> inevitable gathering of people that will stick to
> it no matter what.

Right. The ones who win are the ones who do their
own homework. Pretending blue stars will last ten
billion years just because yellow ones do, rather
than bothering to look up the right answer, punches
some pretty large holes in your arguments from the
vantage of more clueful observers. That reduces both
any chance _you'll_ understand reality, and any
chance you'll prevail in pressing your version of
"how things went" on those who think _they_ do.

FWIW

xanthian.

Phillip Helbig---remove CLOTHES to reply

unread,
Mar 16, 2005, 7:05:54 AM3/16/05
to
In article <mt2.0-23739...@hercules.herts.ac.uk>, jacob navia
<ja...@jacob.remcomp.fr> writes:

> Ptolomeus rotating spheres model could ALWAYS accomodate new
> observations by making a NEW sphere. But at some point people
> just preferred the new model because it was simpler...

No. As Fourier pointed out, ANY periodic motion can be thought of as
consisting of a sum of sinusoidal motions of various periods, so in that
sense, yes (I bet Ptolemy didn't know he was doing Fourier synthesis).
However, this applies just to TRANSVERSE motions. Ptolemy's model
predicts completely different RADIAL motions than that of Copernicus or
Kepler so, as soon as you can measure the distance to a planet, you can
falsify Ptolemy's model.

As for the rest of the discussion, I think you need to define "big bang
theory" before going any further. A common mistake is to define "big
bang theory" to mean more than it actually does. Even if these
additional details really are falsified by some observations, it just
means that these additional details are falsified, not the "core" of the
big bang theory.

jacob navia

unread,
Mar 16, 2005, 7:05:24 AM3/16/05
to
xant...@well.com wrote:
> [Mod. note: please could all posters try to remain polite and focussed
> on the scientific issues rather than on scoring points -- mjh]
>

I agree with that.

>>Relativistic relative velocities???
>
>
> Notice that I supplied _three_ mechanisms, and
> rather than respond to the grist of any of them,
> you wasted your effort jumping on a badly chosen
> word. This isn't productive to understanding the
> phenomena, at all.
>

OK, OK.

Since I have *also* a badly choosen word (I spoke about "blue"
stars instead of just "bright" stars, I will acccept your point.
:-)

>
>>One of the fastest moving galaxies (NGC 1427A ) is
>>falling into the Fornax cluster at ... 600 Km/sec.
>>(http://hubblesite.org/newscenter/newsdesk/archive/releases/2005/09)
>
>
> You'd have a lot less misunderstandings if you'd
> resist jumping crowingly and unthinkingly on a
> misused _word_, and work out the _math_ for
> yourself. If a 600 kmps galaxy hits a 0 kps dust
> cloud, a good guess would be that the galaxy's
> internal dust falls behind at 300 kmps. How long
> does it take to sweep such a galaxy clean of dust?
>
> Well, a galaxy is roughly 10^18 km across, if I
> haven't lost a decimal, and 300 kmps is roughly
> 10^10 km/year, so 10^8 years would suffice, only a
> tenth of a billion: lots of time to have cleaned out
> a galaxy 2 billion years old.
>

OK, so your scenario is
t=0 Start of collision with non moving dust/gas cloud,
galaxy speed 600 Km/sec

t= 10^8 Million years. Galaxy loses all the dust
and material to form new stars, that stays behind
separating from the galaxy at 300Km /sec.

Supposing that the collision (t=0) is 1 000 mill
years after the BB, we have 900 million years left.

True, huge stars and many blue stars live only a few
million years, OK, so after 900 Mill years most of
them disappear. The galaxy loses its blue component
in its spectra. Main sequence stars are not affected at all,
and bright stars (those that live at least 1 Bill years)
are not affected at all.

My point is that only 900 Mill years after the crash,
most bright stars (stars slightly larger than the sun,
but still in the main sequence) should be around, and the
total looks of the galaxy should not be so red as observed.

{snip}

>
> While that interpretation retains good predictive
> capability, it remains "the accepted theory". As
> soon as it loses that ability, it lands "in the
> dustbin of history". Right now, Big Bang theory is
> in no danger at all of landing in the dustbin,
> however much it annoys some folks religious or
> math-deprived intuitive notions.
>

OK. OK I would invite you to this meeting, posted in this
same discussion group:
>>
This is an announcement for the 1st Crisis in Cosmology Conference
(CCC-I) which will be held in Portugal in the period 23-25 June 2005.
All information can be obtained from http://www.cosmology.info

Best regards,
Jose B. Almeida
>>

Please note that I am in *no way* related to that
but I think that's at last a step in a good direction.

>
>>>That _is_ the Big Bang, granted you aren't going
>>>to see it in an optical telescope, which would be
>>>looking among the wrong wavelengths for the Big
>>>Bang in any case.
>
>
>>I am not so stupid to believe we could "see" the
>>big bang.
>
>
> And yet you go right on to insist on exactly that:
>
>
>>Of course not. But its immediate neighborhood
>>should have *some* marks of such a "bang" having
>>happened relatively shrtly, i.e. 500 Mill years...
>
>
> You write like you expect to see the Big Bang from
> some exterior vantage point.

NO!
I just expect to see a different environment 500 million
years after such a bang, than what we actually see. I would
expect no galaxy clusters, nor galaxies with a lot of iron,

xant...@well.com

unread,
Mar 17, 2005, 4:49:11 AM3/17/05
to
jacob navia wrote:

> OK, so your scenario is

> t=0 Start of collision with non moving dust/gas
> cloud, galaxy speed 600 Km/sec

> t= 10^8 Million years. Galaxy loses all the dust
> and material to form new stars, that stays behind
> separating from the galaxy at 300Km /sec.

> Supposing that the collision (t=0) is 1 000 mill
> years after the BB, we have 900 million years
> left.

Well, no, whether consciously or not, you're still
manipulating the data to make things come out the
way you want.

If you are asking: "does the existence of 'red
galaxies' at BB + 2Gyears discredit BBT", you
have to be asking "can they happen _at all_ by any
mechanism consistent with BBT". Pushing the starting
point of the dust cleanout to BB + 1Gyears for no
particular reason is biasing the "yes/no" answer for
no particular reason. You don't get to do that.

You want to take the earliest possible time for a
galaxy to exist, assume, on statistical arguments,
that it has an extremal speed, run it through one or
the other "scraper" to denude it of dust and gas,
and ask "how early can that possibly happen"; before
deciding whether there should still be lots of
midsequence stars around to add their color to the
average light seen.

For one thing, a "partial scraper" operating at the
_same time_ as the stars were forming might have
biased them almost all to be small ones.

Essentially, in another scenario, all you need is
some unknown mechanism to make the vortices in the
consolidating stardust closer spaced and smaller;
I'm guessing there are many possible sources of
turbulance in the early universe; proximity to a
quasar might be one such.

> True, huge stars and many blue stars live only a
> few million years, OK, so after 900 Mill years
> most of them disappear. The galaxy loses its blue
> component in its spectra. Main sequence stars are
> not affected at all, and bright stars (those that
> live at least 1 Bill years) are not affected at
> all.

> My point is that only 900 Mill years after the
> crash, most bright stars (stars slightly larger
> than the sun, but still in the main sequence)
> should be around, and the total looks of the
> galaxy should not be so red as observed.

Well, except, again, that you've ripped an
additional 1Gyear off that 0.9 Gyear without any
explicit justification, and that's time to dim down
a bunch more stars.

Anyway, we can dispute that part until the moderator
grows bored with the discussion, but the larger
issue is that the universe is _complicated_, and
trying to use events occurring 2Gyears after the
big bang to discredit the big bang is a pretty dicey
approach. There's just too much time for catenations
of post-BB events we don't understand yet to have
mucked with the data, to claim that data from that
late in time is still clean enough for a backward
look that contradicts the plentiful much earlier
pro-BBT data we _do_ have. The cosmic background
radiation itself, even, IIUC, is a limited view only
back to that 300,000 (or whatever) years after the
event, when the universe first became transparent.

There's still time even by then for lots to have
happened that we don't understand as well as we
understand that the BB happened at all, to have
mucked about with the data _we_ get to see.

Pretty much, I think, you're forced, in looking at
the consolidated matter of the later universe, to
concede the BB, and get on with the task of trying
to understand all the googleplexes of unlikely
things that have occurred forever after that point,
and to trying to winkle out what _they_ were, and
why _they_ have had the effects they've had on what
we see.

Trying to argue against the big bang right now is
like trying to argue against evolution; the other
team has all the data on its side.

xanthian.

Max Keon

unread,
Mar 18, 2005, 7:50:46 AM3/18/05
to
jacob navia wrote:
>
> xant...@well.com wrote:
>> jacob navia wrote:
-----
-----

>>>The scopes have arrived at the immediate
>>>neighborhood of the supposed big bang and there
>>>is not the slightest hint of a bang to see.
>>
>>
>> You mean besides the cosmic microwave background
>> radiation that already confirms the BBT to several
>> decimals of precision?

> There was a discussion in sci.astro about "overaveraging"
> and the whole "wrinkles in the face of god"
> story. I remain a sceptic about that. But yes, there
> is no alternative explanation to the cosmic background.
>
> The problem is that it could very well be that we just
> do not know what the Cosmic Background *is*, and we see it
> as we can: as a "BB " relic.

Since the validity of the BB theory is very much in question, I
assume that arguments posed by alternative theories are now
open for discussion?

The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
is an extract from a theory which describes a universe that
originated from absolutely nothing, and it provides an alternative
explanation for the CMBR. But without some prior understanding of
the theory the link may not make much sense. To make things even
more difficult, the concept itself is almost incomprehendable
because there are virtually no parallels that can be drawn from our
understanding of how we fit into the structure of the universe that
can be compared with it.

Describing such a universe is no less difficult than it is to
comprehend, so don't expect too much if/when you visit. And spare
a thought for me.

-----

Max Keon

[Mod. note: Just in case people aren't aware of the policy,
`alternative theories' have always been up for discussion on s.a.r.,
but they should be discussed in a scientific (and polite!) way. A
descent to personalities (by either side) or arguments that blatantly
ignore the experimental evidence are likely to run foul of the
moderation policy -- mjh]

Bjoern Feuerbacher

unread,
Mar 19, 2005, 7:22:09 AM3/19/05
to
Max Keon wrote:
> jacob navia wrote:
>
>>xant...@well.com wrote:
>>
>>>jacob navia wrote:
>
> -----
> -----
>
>
>>>>The scopes have arrived at the immediate
>>>>neighborhood of the supposed big bang and there
>>>>is not the slightest hint of a bang to see.
>>>
>>>
>>>You mean besides the cosmic microwave background
>>>radiation that already confirms the BBT to several
>>>decimals of precision?
>
>
>>There was a discussion in sci.astro about "overaveraging"
>>and the whole "wrinkles in the face of god"
>>story. I remain a sceptic about that. But yes, there
>>is no alternative explanation to the cosmic background.
>>
>>The problem is that it could very well be that we just
>>do not know what the Cosmic Background *is*, and we see it
>>as we can: as a "BB " relic.
>
>
> Since the validity of the BB theory is very much in question,

Not by the actual scientists working in cosmology.


> I assume that arguments posed by alternative theories are now
> open for discussion?

Everyone is free everytime to propose alternative theories. See the
moderator's comment below.


> The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
> is an extract from a theory which describes a universe that
> originated from absolutely nothing, and it provides an alternative
> explanation for the CMBR.

Can it explain why the spectrum of the CMBR is such a nice blackbody,
without any spectral lines? Why its temperature changes with time in
accordance with the predictions of the BBT? The fact that if the CMBR
is assumed to have a cosmological origin, the parameters we derive
from it (Hubble parameter, density of dark energy etc.) are nicely
consistent with determinations using other methods? Why computer
simulations which study how the density fluctuations grow with time
produce the observed large-scale structure? The power spectrum (hint:
I don't talk about the blackbody spectrum) of the CMBR, especially the
acoustic peak? The Sunyaev-Zel'dovich effect? The integrated
Sachs-Wolfe effect


> But without some prior understanding of
> the theory the link may not make much sense.

If your theory can explain all the things listed above
(quantitatively), I'll look at it.


[snip]


> [Mod. note: Just in case people aren't aware of the policy,
> `alternative theories' have always been up for discussion on s.a.r.,
> but they should be discussed in a scientific (and polite!) way. A
> descent to personalities (by either side) or arguments that blatantly
> ignore the experimental evidence are likely to run foul of the
> moderation policy -- mjh]

That is a really good point: one should first be aware of the
experimental evidence before one starts proposing alternative theories.


Bye,
Bjoern

Bjoern Feuerbacher

unread,
Mar 21, 2005, 5:07:30 AM3/21/05
to
Bjoern Feuerbacher wrote:
> Max Keon wrote:

[snip]


>>The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
>>is an extract from a theory which describes a universe that
>>originated from absolutely nothing, and it provides an alternative
>>explanation for the CMBR.
>
>
> Can it explain why the spectrum of the CMBR is such a nice blackbody,
> without any spectral lines? Why its temperature changes with time in
> accordance with the predictions of the BBT? The fact that if the CMBR
> is assumed to have a cosmological origin, the parameters we derive
> from it (Hubble parameter, density of dark energy etc.) are nicely
> consistent with determinations using other methods? Why computer
> simulations which study how the density fluctuations grow with time
> produce the observed large-scale structure? The power spectrum (hint:
> I don't talk about the blackbody spectrum) of the CMBR, especially the
> acoustic peak? The Sunyaev-Zel'dovich effect? The integrated
> Sachs-Wolfe effect

Oh, and let's add the observed polarization of the CMBR.


[snip]

Bye,
Bjoern

Max Keon

unread,
Mar 23, 2005, 8:28:02 AM3/23/05
to
Bjoern Feuerbacher wrote:
>
> Max Keon wrote:
>> jacob navia wrote:
-----
-----

>>>There was a discussion in sci.astro about "overaveraging"
>>>and the whole "wrinkles in the face of god"
>>>story. I remain a sceptic about that. But yes, there
>>>is no alternative explanation to the cosmic background.
>>>
>>>The problem is that it could very well be that we just
>>>do not know what the Cosmic Background *is*, and we see it
>>>as we can: as a "BB " relic.
>>
>>
>> Since the validity of the BB theory is very much in question,

> Not by the actual scientists working in cosmology.

>> I assume that arguments posed by alternative theories are now
>> open for discussion?

> Everyone is free everytime to propose alternative theories. See the
> moderator's comment below.

>> The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
>> is an extract from a theory which describes a universe that
>> originated from absolutely nothing, and it provides an alternative
>> explanation for the CMBR.

> Can it explain why the spectrum of the CMBR is such a nice blackbody,
> without any spectral lines?

Yes.

> Why its temperature changes with time in accordance with the
> predictions of the BBT?

The BBT predicts a blackbody curve, but not the specific temperature
of course. My theory predicts a similar curve, and that has been
tweaked to the shape of the CMBR with a multiplier which indicates
the current state of evolution of the universe.

The CMBR paints the picture to which we all fit our theories.

> The fact that if the CMBR
> is assumed to have a cosmological origin, the parameters we derive
> from it (Hubble parameter, density of dark energy etc.) are nicely
> consistent with determinations using other methods?

Dark matter can certainly be explained, if it's required.

> Why computer
> simulations which study how the density fluctuations grow with time
> produce the observed large-scale structure? The power spectrum (hint:
> I don't talk about the blackbody spectrum) of the CMBR, especially the
> acoustic peak?

Every time I study the WMAP maps, all I can see is a well formed
universe that could have been there forever.

> The Sunyaev-Zel'dovich effect? The integrated
> Sachs-Wolfe effect

I wasn't aware of the Sachs-Wolf effect. Thanks.
But what's to explain? The zero origin universe works just fine.

What evidence supports that effect anyway? The assumption seem to
be that photons behave like matter when in gravitational potential
wells, that they can gain or lose energy, but by contracting or
extending their wavelengths. If a photon is moving through a
deepening potential well, it will exit the well with an extended
wavelength (I think). But that is clearly impossible. It would be
hard to explain where the trailing edge of a very long wavetrain
in the visible light spectrum might be stored while it's waiting
for the extended train length in front of it to exit the potential
well. Even if time slows in the deepening well and the light path
length increases, that path length will again shorten when the
wavetrain moves away from the well. Whatever is assumed to happen,
what is going to permanently alter? What experimental evidence
directly supports such a thing?

If the deepening potential well was moving away from an observer,
that effect may be noted. But that's not relevant to the CMBR, is
it?

>> But without some prior understanding of
>> the theory the link may not make much sense.

> If your theory can explain all the things listed above
> (quantitatively), I'll look at it.

> [snip]

>> [Mod. note: Just in case people aren't aware of the policy,
>> `alternative theories' have always been up for discussion on s.a.r.,
>> but they should be discussed in a scientific (and polite!) way. A
>> descent to personalities (by either side) or arguments that blatantly
>> ignore the experimental evidence are likely to run foul of the
>> moderation policy -- mjh]

> That is a really good point: one should first be aware of the
> experimental evidence before one starts proposing alternative theories.

One should also be aware that the evidence can be interpreted in
more ways than one. From the time of my initial encounter with the
zero origin universe (around 30 years ago) I've tested the theory
to the best of my ability against emerging evidence. The universe
seems to be falling into place very nicely. Even the electron and
positron, through experimental evidence, have emerged with amazing
precision to fill the role of the postulated components which I
initially labeled "absolute opposite stress characters".
----------

The polarization found in the CMBR that you refer to in your
follow-up post is a question I need to address. Could it be caused
by light bouncing around a rotation polarized universe (if it can
be termed thus)?

-----

Max Keon

Bjoern Feuerbacher

unread,
Mar 24, 2005, 5:52:56 AM3/24/05
to
Max Keon wrote:
> Bjoern Feuerbacher wrote:
>
>>Max Keon wrote:

[snip]


>>>The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
>>>is an extract from a theory which describes a universe that
>>>originated from absolutely nothing, and it provides an alternative
>>>explanation for the CMBR.
>
>
>>Can it explain why the spectrum of the CMBR is such a nice blackbody,
>>without any spectral lines?
>
>
> Yes.

Your link above goes to a page which mainly contains curves and not
many explanations, as far as I can see. Could you please explain here
shortly what the source of the CMBR is in your model, and why it has a
blackbody spectrum?

>>Why its temperature changes with time in accordance with the
>>predictions of the BBT?
>
>
> The BBT predicts a blackbody curve, but not the specific temperature
> of course.

Err, that was not my point. Read again what I actually wrote, please.
I did not talk about temperature - I talked about *changes* in
temperature.

See e.g. here:
<http://www.astro.ucla.edu/~wright/stdystat.htm#Tvsz>


[snip more irrelevant arguments]

>>The fact that if the CMBR
>>is assumed to have a cosmological origin, the parameters we derive
>>from it (Hubble parameter, density of dark energy etc.) are nicely
>>consistent with determinations using other methods?
>
>
> Dark matter can certainly be explained, if it's required.

That has nothing to do with my argument above. Try again, please.

>>Why computer
>>simulations which study how the density fluctuations grow with time
>>produce the observed large-scale structure? The power spectrum (hint:
>>I don't talk about the blackbody spectrum) of the CMBR, especially the
>>acoustic peak?
>
>
> Every time I study the WMAP maps, all I can see is a well formed
> universe that could have been there forever.

That has nothing to do with either of my two arguments above. Try
again, please.


>>The Sunyaev-Zel'dovich effect? The integrated
>>Sachs-Wolfe effect
>
>
> I wasn't aware of the Sachs-Wolf effect. Thanks.

But you were aware of the Sunyaev-Zel'dovich effect? If yes, could
you please outline how your model explains the observations?


> But what's to explain? The zero origin universe works just fine.

Well, then please show how your model explains these two effects.
Quantitatively.


> What evidence supports that effect anyway?

Which one? Sunyaev-Zel'dovich or integrated Sachs-Wolfe?

For the first one, see e.g. here:
<http://cfa-www.harvard.edu/~aas/tenmeter/sz.htm>

For the second, see e.g. here:
astro-ph/0307335


> The assumption seem to
> be that photons behave like matter when in gravitational potential
> wells, that they can gain or lose energy, but by contracting or
> extending their wavelengths.

Err, that is not an assumption. That has actually been experimentally
confirmed. Both in the lab and in astronomical observations.

> If a photon is moving through a
> deepening potential well, it will exit the well with an extended
> wavelength (I think). But that is clearly impossible.

Well, then why has this been observed?


> It would be
> hard to explain where the trailing edge of a very long wavetrain
> in the visible light spectrum might be stored while it's waiting
> for the extended train length in front of it to exit the potential
> well.

What makes you think that this trailing edge has to be stored
somewhere and has to wait?


> Even if time slows in the deepening well and the light path
> length increases, that path length will again shorten when the
> wavetrain moves away from the well.

Pardon? When the light moves away from the well, the path length
*inside the well* shortens? Sorry, I can't follow you here.

If you talk about the path length *outside* the well, then what
has that to do with the redshift occuring *inside* the well?

> Whatever is assumed to happen,
> what is going to permanently alter? What experimental evidence
> directly supports such a thing?

Try this, for starters:
<http://scienceworld.wolfram.com/biography/Pound.html>


> If the deepening potential well was moving away from an observer,
> that effect may be noted. But that's not relevant to the CMBR, is
> it?

No.


>>>But without some prior understanding of
>>>the theory the link may not make much sense.
>
>
>>If your theory can explain all the things listed above
>>(quantitatively), I'll look at it.

So far, you have addressed nothing but the very first point. And even
there, you did not bother to gave an explanation - you merely
asserted that your model explains that.

[snip]


>>That is a really good point: one should first be aware of the
>>experimental evidence before one starts proposing alternative theories.
>
>
> One should also be aware that the evidence can be interpreted in
> more ways than one.

As I already said: feel free to address the evidence. Quantitatively.

> From the time of my initial encounter with the
> zero origin universe (around 30 years ago) I've tested the theory
> to the best of my ability against emerging evidence.

You admitted yourself above that you weren't aware of some pieces
of evidence, and apparently misunderstood other pieces.


[snip more irrelevancies]


> The polarization found in the CMBR that you refer to in your
> follow-up post is a question I need to address.

<http://www-news.uchicago.edu/releases/02/020918.carlstrom.shtml>
<http://astro.uchicago.edu/dasi/polexpert/>


> Could it be caused
> by light bouncing around a rotation polarized universe (if it can
> be termed thus)?

Don't merely speculate. Address the results. Quantitatively.


Bye,
Bjoern

r...@firstpr.com.au

unread,
Mar 29, 2005, 11:21:22 AM3/29/05
to
For why mature galaxies are observed so soon after the Big
Bang, see Robert Karl Stonjek's Dark Time hypothesis:
"Article: Most distant galaxy cluster yet is revealed"
(sci.physics, 2 March).

I propose a non-BBT explanation for the Sunyaev-Zel'dovich
effect. But first . . .

Bjoern, what do you think of the failure to find evidence for
the transverse proximity effect with a foreground quasar?

http://astroneu.com/plasma-redshift-1/#TPE

The conventional view is that the quasars must be turning on
and off, or have very short lifetimes. I think a better
explanation is that the quasars are not located where the BBT
says they are - due to most of the redshift of their light,
including probably most or all of the Lyman alpha forest,
occurring in space near to them. My best guess is that this
occurs due to some kind of plasma redshift or sparse particle
redshift mechanism.

If the BBT is true, then the quasars are exactly where the
conventional researchers say they are, and therefore the
quasars must have very limited lifetimes in order to have not
ionized the neutral H in their vicinity, which these
researchers believe they observe in the Lyman alpha forest of
the background quasar. (The conventional researchers
generally reject the other two explanations: very narrow
quasar beaming and some kind of shielding effect, which is
much the same as beaming.) The researchers do not seem to
consider that these observations constitute a good challenge
to the theory that the redshift of quasars is due to
Doppler / expansion of the Universe. (I wrote to them about
this a year ago and got no response.)

Do you think quasars have such short lifetimes or such low
duty cycles as to not generally ionize neutral H in their
vicinity?

As far as I know, quasars were not generally considered to
have short lifetimes until this lack of TPE business arose.
If quasars are the same as, or cousins to, "radio galaxies"
then its hard to imagine them having such short lifetimes
since (according to BBT theories) these radio galaxies have
such huge lobes that they must have been running continually
for very long periods of time.

Here is a hypothesis I made up a year ago, regarding the CMB
and the Sunyaev-Zel'dovich effect. (See above URL.)

The CMB is produced by the graveyard of black dwarfs and
their collision fragments, produced from dead stars over
countless billennia (many galaxies are very old indeed -
this is a non-BBT theory). These spin out of the plane of
the spiral galaxy since they survive close encounters, which
would rip active stars apart due to tidal forces. This halo
of dead cold solid matter constitutes the dark matter which
explains galactic rotation curves. Over time (we have lots
of time . . . ) they attain the average radiative
temperature of the Universe, which is about the same
temperature as the CMB. (So far, this theory, or most of
was not first proposed by me - sorry I can't find the URL of
the site of the chap who proposed this a few years ago.)

To this model, I add redshift of the CMB as it passes
through the void IGM - for instance due to a plasma or
sparse particle redshift mechanism. By the way, I am
considering redshift mechanisms which do not necessarily
involve loss of energy - just the change in the short
impulse length em wave so that more quanta of lower energies
are delivered. (I reject the "photon" - one quantum of
energy lost to one quantum of energy received, without
interaction with the emr caused by other quanta - view of
electromagnetic radiation - but that's another story.)

In my hypothesis, CMB from galaxies beyond a nearer galaxy
(or galaxy cluster) will generally be redshifted compared to
the contribution of CMB coming from nearer galaxy's black
dwarf halo. Therefore we observe somewhat hotter CMB from
the vicinity of the nearby galaxy or cluster - which is my
understanding of the Sunyaev-Zel'dovich effect.

I haven't studies the Sachs-Wolfe effect. The supposed
precision of the BBT theory of CMB doesn't impress me or
many other critics. It can be easy to think of other
explanations - and then, with sufficient effort, to
fine-tune them to observations too.

Max Keon

unread,
Mar 29, 2005, 11:20:41 AM3/29/05
to
Bjoern Feuerbacher wrote:
>
>Max Keon wrote:
>>Bjoern Feuerbacher wrote:
>>>Max Keon wrote:

> [snip]

>>>>The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
>>>>is an extract from a theory which describes a universe that
>>>>originated from absolutely nothing, and it provides an alternative
>>>>explanation for the CMBR.
>>>
>>>
>>>Can it explain why the spectrum of the CMBR is such a nice blackbody,
>>>without any spectral lines?
>>
>>Yes.

> Your link above goes to a page which mainly contains curves and not
> many explanations, as far as I can see. Could you please explain here
> shortly what the source of the CMBR is in your model,

As I previously indicated, to "explain here shortly" is almost
impossible. But the rest of my reply may help.

> and why it has a blackbody spectrum?

It's based on temperature change of the universe throughout its
evolution from the zero origin. It has the spectrum of the CMBR,
just like your theory does.

>>>Why its temperature changes with time in accordance with the
>>>predictions of the BBT?
>>
>>The BBT predicts a blackbody curve, but not the specific temperature
>>of course.

> Err, that was not my point. Read again what I actually wrote, please.
> I did not talk about temperature - I talked about *changes* in
> temperature.

The temperature at the origin was zero. The universe is evolving.
Its temperature is increasing at a logarithmic rate, hence the
^1.12 adjustment to each (equally spaced relative to a fixed time
zone) curve generated from the Planck equation,
#=((8*pi*h*f^3)/(c^2*(EXP((h*f)/(k*t))-1))) ^1.12

The ^1.12 exponent is near enough to constant for the blackbody plot
of the universe that we can meaningfully comprehend. It would have
been infinitesimally greater than 1 for the plot at the origin. So
there's still a long way left for us to go.

> [snip more irrelevant arguments]

>>>The fact that if the CMBR
>>>is assumed to have a cosmological origin, the parameters we derive
>>>from it (Hubble parameter, density of dark energy etc.) are nicely
>>>consistent with determinations using other methods?
>>
>>Dark matter can certainly be explained, if it's required.

> That has nothing to do with my argument above. Try again, please.

That argument has nothing to do with a zero origin universe either.

>>>Why computer
>>>simulations which study how the density fluctuations grow with time
>>>produce the observed large-scale structure? The power spectrum (hint:
>>>I don't talk about the blackbody spectrum) of the CMBR, especially the
>>>acoustic peak?
>>
>>Every time I study the WMAP maps, all I can see is a well formed
>>universe that could have been there forever.

> That has nothing to do with either of my two arguments above. Try
> again, please.

But it has a lot to do with my argument. The all sky picture of the
universe from the zero origin is crystal clear. According to that
picture, matter is slowly clumping together, increasing the depth
of dimension, of space. The picture provides a remarkable insight
into how the matter content of the universe is evolving. The picture
at the very origin would have contained one infinitesimally minute
anisotropy within a completely black background. A universe with
zero anisotropy would not exist.

>>>The Sunyaev-Zel'dovich effect? The integrated
>>>Sachs-Wolfe effect
>>
>>I wasn't aware of the Sachs-Wolf effect. Thanks.

> But you were aware of the Sunyaev-Zel'dovich effect?

No, not until you mentioned it.
A quick search at the time found only this sentence;
"Fluctuations arising from the Sunnyaev-Zel'dovich (SZ) effect,
the up-scattering of the background spectrum by both the hot gas
surrounding galaxy clusters and the peculiar velocity of the
cluster, should be observable on spatial scales of around 3
arcminutes." (I've lost the link. I'll post it next time) According
to that sentence the effect has yet to be noted, or is already
factored in as a component within the anisotropy. It really doesn't
have any more relevance to my argument than the Sachs-Wolfe effect
though.

> If yes, could
> you please outline how your model explains the observations?

>>But what's to explain? The zero origin universe works just fine.

> Well, then please show how your model explains these two effects.
> Quantitatively.

>>What evidence supports that effect anyway?

> Which one? Sunyaev-Zel'dovich or integrated Sachs-Wolfe?
>
> For the first one, see e.g. here:
> <http://cfa-www.harvard.edu/~aas/tenmeter/sz.htm>
>
> For the second, see e.g. here:
> astro-ph/0307335

>>The assumption seem to
>>be that photons behave like matter when in gravitational potential
>>wells, that they can gain or lose energy, but by contracting or
>>extending their wavelengths.

> Err, that is not an assumption. That has actually been experimentally
> confirmed. Both in the lab and in astronomical observations.

>>If a photon is moving through a
>>deepening potential well, it will exit the well with an extended
>>wavelength (I think). But that is clearly impossible.

> Well, then why has this been observed?

The fact that a photon wavelength changes according to local
gravitational potential may have been confirmed, but not the
*assumption* that they gain or lose energy in the process.

Consider this; Two adjacent straight lengths of equally spaced
billiard balls, labeled (1) and (2), are set in motion along the
line of their pointing direction. Train (1) travels a straight line
through free space while train (2) is set to run the gauntlet of a
deepening gravitational potential well. Along the journey to the
deepest part of the well on (2)'s travels, space-time will be
stretching and will of course extend its train length. But because
the well is still deepening, (2)'s departure from the well will be
further restrained than if the well was constant. However, when (1)
and (2) are returned to the same space-time environment they will
still measure the same length. The additional restraining forces
applied by the deepening well are applied equally to each billiard
ball along train (2). Nothing will change.

Now replace the billiard balls with photons. Either the speed of
light in not isotropic over the train length, or the photons overlap
to accommodate their added wavelengths????

Not wishing to break from the subject, but the concept of photons
as particles has no place in the zero origin universe.

>>It would be
>>hard to explain where the trailing edge of a very long wavetrain
>>in the visible light spectrum might be stored while it's waiting
>>for the extended train length in front of it to exit the potential
>>well.

> What makes you think that this trailing edge has to be stored
> somewhere and has to wait?

-----
-----

>> The polarization found in the CMBR that you refer to in your
>> follow-up post is a question I need to address.
>
> <http://www-news.uchicago.edu/releases/02/020918.carlstrom.shtml>
> <http://astro.uchicago.edu/dasi/polexpert/>

>>Could it be caused
>>by light bouncing around a rotation polarized universe (if it can
>>be termed thus)?

> Don't merely speculate. Address the results. Quantitatively.

I'll need time of course.

-----

Max Keon

r...@firstpr.com.au

unread,
Mar 30, 2005, 3:59:07 AM3/30/05
to
Max, I looked at your page:

http://members.ozemail.com.au/~mkeon/the1-1a.html

and had the same experience I have with many contrarian
physics sites - too many things seemed to make no plausible
sense and I couldn't find a reason for looking at any of it
in sufficient detail to assemble a critique.

If, as I understand, your theory is different from that of
conventional Big Bang cosmology, and if you suggest yours is
a better theory, then I think you should be able to point
out which observations the BBT fails to properly explain,
and how yours offers a better explanation.

Can you list such observations? You don't need to explain
your theory - just present evidence that the BBT predicts
things different from what is observed. Or are you simply
arguing that your explanations are more elegant than the
BBT's - with exactly the same predictions?

Progress in science can involve simply disproving someone
else's theory. Its not necessary to have a better one -
though it is nice if you do.

Bjoern Feuerbacher

unread,
Mar 30, 2005, 3:59:58 AM3/30/05
to
Max Keon wrote:
> Bjoern Feuerbacher wrote:
>
>>Max Keon wrote:
>>
>>>Bjoern Feuerbacher wrote:
>>>
>>>>Max Keon wrote:
>
>
>>[snip]
>
>
>>>>>The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
>>>>>is an extract from a theory which describes a universe that
>>>>>originated from absolutely nothing, and it provides an alternative
>>>>>explanation for the CMBR.
>>>>
>>>>
>>>>Can it explain why the spectrum of the CMBR is such a nice blackbody,
>>>>without any spectral lines?
>>>
>>>Yes.
>
>
>>Your link above goes to a page which mainly contains curves and not
>>many explanations, as far as I can see. Could you please explain here
>>shortly what the source of the CMBR is in your model,
>
>
> As I previously indicated, to "explain here shortly" is almost
> impossible. But the rest of my reply may help.

We'll see.


>> and why it has a blackbody spectrum?
>
>
> It's based on temperature change of the universe throughout its
> evolution from the zero origin. It has the spectrum of the CMBR,
> just like your theory does.

How could "temperature change of the universe throughout its evolution
from the zero origin" explain the existence and the blackbody spectrum
of the CMBR?

>>>>Why its temperature changes with time in accordance with the
>>>>predictions of the BBT?
>>>
>>>The BBT predicts a blackbody curve, but not the specific temperature
>>>of course.
>
>
>>Err, that was not my point. Read again what I actually wrote, please.
>>I did not talk about temperature - I talked about *changes* in
>>temperature.
>
>
> The temperature at the origin was zero.

That's contrary to observations, which show that the temperature was
*greater* in the past. See the link shortly below.


> The universe is evolving.

Finally something we agree on.


> Its temperature is increasing at a logarithmic rate, hence the
> ^1.12 adjustment

How do you get from a logarithmic temperature increase to a factor ^1.12?


> to each (equally spaced relative to a fixed time
> zone) curve generated from the Planck equation,
> #=((8*pi*h*f^3)/(c^2*(EXP((h*f)/(k*t))-1))) ^1.12

In order to apply the Planck equation, you need something material
which is in thermal equilibrium. What is this in your model? In the
standard BB scenario, it was the plasma which filled the early universe.

BTW, the Planck curve to the power of 1.12 does not give a blackbody
curve again. You even have problems with the units there!


> The ^1.12 exponent is near enough to constant for the blackbody plot
> of the universe that we can meaningfully comprehend.

I have no clue what this is supposed to mean.


> It would have
> been infinitesimally greater than 1 for the plot at the origin.

Why?


> So there's still a long way left for us to go.
>
>
>>See e.g. here:
>><http://www.astro.ucla.edu/~wright/stdystat.htm#Tvsz>

I notice you did not bother to address this.

>>[snip more irrelevant arguments]
>
>
>>>>The fact that if the CMBR
>>>>is assumed to have a cosmological origin, the parameters we derive
>>>>from it (Hubble parameter, density of dark energy etc.) are nicely
>>>>consistent with determinations using other methods?
>>>
>>>Dark matter can certainly be explained, if it's required.
>>
>>That has nothing to do with my argument above. Try again, please.
>
>
> That argument has nothing to do with a zero origin universe either.

It is an argument about observational evidence for the BBT. So if you
claim that you can explain all the evidence which the BBT can explain,
you need to address this. Why don't you bother?

>>>>Why computer
>>>>simulations which study how the density fluctuations grow with time
>>>>produce the observed large-scale structure? The power spectrum (hint:
>>>>I don't talk about the blackbody spectrum) of the CMBR, especially the
>>>>acoustic peak?
>>>
>>>Every time I study the WMAP maps, all I can see is a well formed
>>>universe that could have been there forever.
>
>
>>That has nothing to do with either of my two arguments above. Try
>>again, please.
>
>
> But it has a lot to do with my argument.

So what? You claimed that you can explain all the evidence which the
BBT can explain. So why don't you address this?


> The all sky picture of the
> universe from the zero origin is crystal clear. According to that
> picture, matter is slowly clumping together,

That's the same as the BBT says.


> increasing the depth of dimension, of space.

That's incomprehensible.


> The picture provides a remarkable insight
> into how the matter content of the universe is evolving. The picture
> at the very origin would have contained one infinitesimally minute
> anisotropy

That's very close to what the BBT says.


> within a completely black background.

That is contrary to the observations.


> A universe with zero anisotropy would not exist.

Why not?

>>>>The Sunyaev-Zel'dovich effect? The integrated
>>>>Sachs-Wolfe effect
>>>
>>>I wasn't aware of the Sachs-Wolf effect. Thanks.
>
>
>>But you were aware of the Sunyaev-Zel'dovich effect?
>
>
> No, not until you mentioned it.

So, we have now at least three pieces of evidence for the BB
explanation of the CMBR which you were not aware of. And please
keep in mind that I am by no means an expert in cosmology
- just a physicist with a private interest in cosmology. You should
think about what this might imply about the amount of evidence
you are not aware of...


> A quick search at the time found only this sentence;
> "Fluctuations arising from the Sunnyaev-Zel'dovich (SZ) effect,
> the up-scattering of the background spectrum by both the hot gas
> surrounding galaxy clusters and the peculiar velocity of the
> cluster, should be observable on spatial scales of around 3
> arcminutes." (I've lost the link. I'll post it next time) According
> to that sentence the effect has yet to be noted, or is already
> factored in as a component within the anisotropy.

The quote you give above is outdated. Look at the link I provide
below. The effect *has* been observed.


> It really doesn't
> have any more relevance to my argument than the Sachs-Wolfe effect
> though.

Err, both are effects which are explained by the BB model for the
CMBR. So why do you think you can simply ignore these two effects?

>>If yes, could
>>you please outline how your model explains the observations?

I notice that you don't bother to do that.

>>>But what's to explain? The zero origin universe works just fine.
>
>
>>Well, then please show how your model explains these two effects.
>>Quantitatively.

I notice that you don't bother to do that.

>>>What evidence supports that effect anyway?
>
>
>>Which one? Sunyaev-Zel'dovich or integrated Sachs-Wolfe?
>>
>>For the first one, see e.g. here:
>><http://cfa-www.harvard.edu/~aas/tenmeter/sz.htm>

I notice that you choose to ignore that.


>>For the second, see e.g. here:
>>astro-ph/0307335

I notice that you choose to ignore that.


>>>The assumption seem to
>>>be that photons behave like matter when in gravitational potential
>>>wells, that they can gain or lose energy, but by contracting or
>>>extending their wavelengths.
>>
>>Err, that is not an assumption. That has actually been experimentally
>>confirmed. Both in the lab and in astronomical observations.
>
>
>>>If a photon is moving through a
>>>deepening potential well, it will exit the well with an extended
>>>wavelength (I think). But that is clearly impossible.
>
>
>>Well, then why has this been observed?
>
>
> The fact that a photon wavelength changes according to local
> gravitational potential may have been confirmed, but not the
> *assumption* that they gain or lose energy in the process.

So you disagree with E=hf? Or with f=c/lambda?

If you don't disagree with both, then you get E=hc/lambda, i.e.
every change in wavelength is equivalent to a change in energy.


> Consider this; Two adjacent straight lengths of equally spaced
> billiard balls, labeled (1) and (2), are set in motion along the
> line of their pointing direction.

That has little to do with photons and light.


> Train (1) travels a straight line
> through free space while train (2) is set to run the gauntlet of a
> deepening gravitational potential well. Along the journey to the
> deepest part of the well on (2)'s travels, space-time will be
> stretching and will of course extend its train length. But because
> the well is still deepening, (2)'s departure from the well will be
> further restrained than if the well was constant. However, when (1)
> and (2) are returned to the same space-time environment they will
> still measure the same length.

Why should they?


> The additional restraining forces
> applied by the deepening well are applied equally to each billiard
> ball along train (2). Nothing will change.

I can't follow your logic. What "restraining forces"?

> Now replace the billiard balls with photons.

That would be a false analogy.


> Either the speed of
> light in not isotropic over the train length, or the photons overlap
> to accommodate their added wavelengths????

"their" added wavelengths? Due to grammar, the "their" seems to refer
to the photons. But photons do not have wavelengths. Only
electromagnetic waves have wavelengths. So, what are you talking about?

> Not wishing to break from the subject, but the concept of photons
> as particles has no place in the zero origin universe.

Well, then how do you explain the photo effect and the Compton effect?
(quantitatively!)

[snip more of that]


>>>The polarization found in the CMBR that you refer to in your
>>>follow-up post is a question I need to address.
>>
>><http://www-news.uchicago.edu/releases/02/020918.carlstrom.shtml>
>><http://astro.uchicago.edu/dasi/polexpert/>
>
>
>>>Could it be caused
>>>by light bouncing around a rotation polarized universe (if it can
>>>be termed thus)?
>
>
>>Don't merely speculate. Address the results. Quantitatively.
>
>
> I'll need time of course.

While you are at it, you can also look at all the stuff you ignored
above.

Bye,
Bjoern

Bjoern Feuerbacher

unread,
Mar 30, 2005, 4:00:24 AM3/30/05
to
r...@firstpr.com.au wrote:
> For why mature galaxies are observed so soon after the Big
> Bang, see Robert Karl Stonjek's Dark Time hypothesis:
> "Article: Most distant galaxy cluster yet is revealed"
> (sci.physics, 2 March).
>
> I propose a non-BBT explanation for the Sunyaev-Zel'dovich
> effect. But first . . .
>
> Bjoern, what do you think of the failure to find evidence for
> the transverse proximity effect with a foreground quasar?
>
> http://astroneu.com/plasma-redshift-1/#TPE

I am not an astronomer, just a physicist with cosmology as his
"hobby", so I am not really qualified to comment on this.

But I would like to point out that
1) apparently only very few quasars were examined this far, and
we should wait for more data before jumping to conclusions, and
2) The second explanation offered by the researchers (the foreground
quasar's energy is beamed) looks quite sensible to me.

The author does not bother to explain why that second xplanation does
not work (as far as I can see); he merely claims that all three
explanations "are all highly unlikely, or at least at odds with
reasonable interpretations of other observations."


> The conventional view is that the quasars must be turning on
> and off, or have very short lifetimes.

Don't know about that.


> I think a better
> explanation is that the quasars are not located where the BBT
> says they are - due to most of the redshift of their light,
> including probably most or all of the Lyman alpha forest,
> occurring in space near to them. My best guess is that this
> occurs due to some kind of plasma redshift or sparse particle
> redshift mechanism.

When you can more than just guess, i.e. when you can provide
a quantitative explanation how this works, and how this explains
all the observed evidence, please send me a note.


> If the BBT is true, then the quasars are exactly where the
> conventional researchers say they are,

I wouldn't say that the two statements depend so strongly on
each other.


> and therefore the
> quasars must have very limited lifetimes in order to have not
> ionized the neutral H in their vicinity, which these
> researchers believe they observe in the Lyman alpha forest of
> the background quasar. (The conventional researchers
> generally reject the other two explanations: very narrow
> quasar beaming and some kind of shielding effect, which is
> much the same as beaming.) The researchers do not seem to
> consider that these observations constitute a good challenge
> to the theory that the redshift of quasars is due to
> Doppler / expansion of the Universe. (I wrote to them about
> this a year ago and got no response.)

As the author of that page, you seem to have a misconception
about cosmological redshift: cosmologists do *not* say think
that it is due to the Doppler effect.
<http://www.astronomycafe.net/cosm/expan.html>

Why the researchers do not consider that to be a challenge to
the BBT? Probably due to the simple fact that the BBT is very
well established and supported by observations - and before
one begins to question such a well-established theory, one
first looks for errors in other parts of one's assumptions.


> Do you think quasars have such short lifetimes or such low
> duty cycles as to not generally ionize neutral H in their
> vicinity?

I don't have enough knowledge of quasars to judge that.


[snip]


> Here is a hypothesis I made up a year ago, regarding the CMB
> and the Sunyaev-Zel'dovich effect. (See above URL.)
>
> The CMB is produced by the graveyard of black dwarfs and
> their collision fragments, produced from dead stars over
> countless billennia (many galaxies are very old indeed -

Why don't we see stars older than about 13 billion years then?


> this is a non-BBT theory). These spin out of the plane of
> the spiral galaxy since they survive close encounters, which
> would rip active stars apart due to tidal forces. This halo
> of dead cold solid matter constitutes the dark matter which
> explains galactic rotation curves.

If you can explain with this model the observed rotation
curves *quantitatively*, feel free to show your work.


> Over time (we have lots
> of time . . . ) they attain the average radiative
> temperature of the Universe, which is about the same
> temperature as the CMB.

What is the "average radiative temperature of the universe"?


> (So far, this theory, or most of
> was not first proposed by me - sorry I can't find the URL of
> the site of the chap who proposed this a few years ago.)
>
> To this model, I add redshift of the CMB as it passes
> through the void IGM - for instance due to a plasma or
> sparse particle redshift mechanism.

See my note above wrt guessing.

> By the way, I am
> considering redshift mechanisms which do not necessarily
> involve loss of energy - just the change in the short
> impulse length em wave so that more quanta of lower energies
> are delivered. (I reject the "photon" - one quantum of
> energy lost to one quantum of energy received, without
> interaction with the emr caused by other quanta - view of
> electromagnetic radiation - but that's another story.)

Feel free to explain the photo effect and the Compton effect.
Quantitatively.


> In my hypothesis, CMB from galaxies beyond a nearer galaxy
> (or galaxy cluster) will generally be redshifted compared to
> the contribution of CMB coming from nearer galaxy's black
> dwarf halo.

Why?


> Therefore we observe somewhat hotter CMB from
> the vicinity of the nearby galaxy or cluster - which is my
> understanding of the Sunyaev-Zel'dovich effect.

Feel free to come up with a quantitative explanation, instead
of just handwaving.


> I haven't studies the Sachs-Wolfe effect. The supposed
> precision of the BBT theory of CMB doesn't impress me or
> many other critics.

Interestingly, most of the critics are not aware of most of
the evidence...


> It can be easy to think of other explanations

Yes. Making up stories without bothering to do actual
quantitative checks is very easy indeed.


> - and then, with sufficient effort, to
> fine-tune them to observations too.

"fine-tune them to observations, too"? Please point out
what fine-tuning to observations was done in the BBT.


Bye,
Bjoern

r...@firstpr.com.au

unread,
Mar 30, 2005, 12:06:30 PM3/30/05
to
Bjoern,

It doesn't matter whether the redshift of distant objects
under the BBT is called "Doppler" or something else. The
point is that apart from a little motion which is relative
to nearby objects, and a little gravitational redshift, the
BBT says that there is no redshift mechanism other than the
expansion of the Universe.

Therefore, the foreground and background quasars in the
Quasar-Quasar Transverse Proximity Effect work:

http://astroneu.com/plasma-redshift-1/#TPE

are, according to the BBT, at distances which can be so
reliably estimated that researchers can be sure that a
specific portion of the Lyman alpha absorption in the
background quasar's spectrum occurred at the same distance
from Earth as that of the foreground quasar. This places
that section of the path from the background quasar at a
distance from the foreground quasar which can be directly
calculated from the angle between the quasars on the sky,
and the distance to the foreground quasar, which depends on
the cosmological parameters used to convert redshift to
distance.

Contrary to their expectations, the researchers find
absorption in those parts of the spectrum corresponding to
this locality. They expected an absence of absorption
due to the foreground quasar radiating UV approximately
anisotropically and thereby ionising any hydrogen in the
area.

If the BBT is correct, then the quasar redshifts must be
attributable only to their position in the expanding
Universe - so the researcher's estimates of their distance
must be accurate. So if the BBT is correct, we must
conclude either that the foreground quasar is very narrowly
beamed (either in its intrinsic pattern of radiation or due
to some kind of shielding arrangement) or that a short time
before it emitted the light we observe, it was not emitting
sufficient UV to ionize the local hydrogen. This time
corresponds to the distance between it and the path from
the background quasar.

The researchers find this a difficult choice - and in their
papers discount beaming and shielding, for reasons which
seem reasonable to me: the beaming would have to be
implausably narrow. Rather then question their theory
about the nature of the redshift (probably because they
consider their knowledge to be a fact, rather than a
theory), and therefore question their estimates of distances
to the quasars, they conclude that the foreground quasar
wasn't radiating at a time which would have altered
absorption in the background quasar's spectrum for the light
we observe today.

No matter whether they chose beaming/shielding or a short
lifetime for each quasar (perhaps including low duty-cycles
of on/off radiation) they have a major problem: all these
explanations involve the actual number of quasars being very
much larger than is usually estimated. Amongst other things,
revising this estimate of the abundance of quasars must
surely require some major revision of the now very detailed
interlocking network of quantitative theories which
constitute the current version of the Big Bang Theory.

My point is that these researchers, and it seems you too,
are pursuing a path of quasars being very narrow - in time
or beaming - for which there is no other obvious supporting
observations, when a much simpler explanation is that the
quasars are not located at the distances that the BBT says
they are.

If quasars are at distances closer than their BBT-predicted
redshift distances, then a bunch of other problems are
solved. For instance the rapid changes in output become
compatible with quasars of a smaller size and smaller output
once it can be admitted that quasars are closer than the
BBT says they are.

It only takes one piercing observation, correctly
interpreted, to disprove an entire theory. Its not like
in a democracy where opinions and numbers of votes matter.

The BBT predicts that the quasars and the neutral H is
exactly where these researchers think they are, but all
other observations indicate that quasars are not
exceedingly narrowly beamed and are not prone to having
short lifetimes.

Rather than question the BBT, the researchers - and you
too it seems - prefer to pursue a view of quasars which
is seems to be incompatible with theoretical
interpretations of a vast number of observations. While
I think these interpretations are badly skewed by over-
estimates of distance, I am not aware of any reason to
question the theories of quasars being big, long-lasting
and not narrowly beamed, at least in their UV radiation.

A lot of the problems with understanding quasars are due
to the BBT's insistence that they are at vast distances
due to their generally high redshift.

I suppose that if someone considers the BBT to be a
precise, well-developed, locked-together theory with few
observational challenges, then it would make sense to
be prepared to consider otherwise implausible things about
quasars. But other folk see the BBT as a big mess and can
tentatively imagine various other forms of redshift, which
would lead to theories with greater explanatory power.

It is not necessary for me to provide a new theory or any
quantitative material at all to make scientific progress.
All I need to do is disprove an existing theory. I say the
failure to find the TPE with a foreground quasar is a
disproof of the BBT.

It so happens that I am working on a redshift mechanism
which occurs in sparse plasmas or gasses. There is another
theory with comparable characteristics, by Ari Brynjolfsson.

http://arxiv.org/abs/astro-ph/0401420

However I don't have to provide a replacement theory to
disprove the BBT.

Since the BBT causes a lot of problems in explaining
quasars, since there are no satisfactory conventional
explanations for astrophysically crucial processes such
as coronal heating and stellar wind acceleration, and
since the BBT relies on the notion that electromagnetic
radiation is not changed at all in terms of the quanta
of energy it deposits as it travels for billions of years
in sparse plasma, I say it is a very vulnerable theory.

You used the word "quantitative" five times in your response.
No-one needs any new theory, with or without quantitative
predictions, in order to show that the BBT is broken.

You wrote:

> Why don't we see stars older than about 13 billion years
> then?

Our knowledge of the age of stars is entirely theoretical.
We do not observe the age of stars.

I don't know enough about this area to point to evidence for
stars being older than this, but perhaps someone can. Since
most workers in this field depend entirely on funding
arrangements which would be threatened if they questioned
the prevailing paradigm - or if they produced results which
were obviously wrong according to that paradigm - it would
not surprise me if they generally failed to theorise, or at
least publish, stellar theories which indicate some stars
are older than whatever the age of the Universe was
considered to be at the time.

> What is the "average radiative temperature of the
> universe"?

I am not sure. Googling "average temperature of the
universe" yields lots of references to 2.7k or 3k. I
haven't investigated beyond this. The question is, if you
put a largish black object out in the middle of
intergalactic space, or on the edge of a galaxy, or in the
midst of a galaxy cluster and let it sit there for a long
time, what will its temperature be?

I don't make any great claims about my black dwarf
hypothesis for galactic missing mass and the CMB. (Does
anyone know of other such proposals? I think I was mistaken
about the site http://hometown.aol.com/wmitch8493/myhomepage/
I thought mentioned such a theory.) All I was saying is
that its not hard to think of half-way plausible
explanations which do not involve the Big Bang. Since the
BBT has no explanation for the galactic missing mass
(without recourse to exotic physics) and since the black
dwarf hypothesis looks promising, I suggest it as part of an
alternative set of theories which explains things better
than the BBT.

Guessing is a perfectly valid activity when contemplating
new explanations for observations which currently have
little or no explanation. I don't claim to have a solid
theory - just one that may interest some folk.

> Feel free to explain the photo effect and the
> Compton effect.

I'll let you know when I work up a good critique of the
photon idea. While I know it is attractive for considering
X-rays and electrons, I think it is not at all adequate for
a number of other aspects of electromagnetic radiation. I
think that the assumption that emr involves "photons" which
start at one point and end (with the magical collapse of
the wavefunction at another point) and which do not
interfere with each other at all, is a terrible mistake.

> > In my hypothesis, CMB from galaxies beyond a nearer
> > galaxy (or galaxy cluster) will generally be
> > redshifted compared to the contribution of CMB coming
> > from nearer galaxy's black dwarf halo.

> Why?

By whatever sparse particle or plasma redshift mechanism I
propose is shifting the visible light of the galaxies and
quasars.

> > Therefore we observe somewhat hotter CMB from the
> > vicinity of the nearby galaxy or cluster - which is my
> > understanding of the Sunyaev-Zel'dovich effect.

> Feel free to come up with a quantitative explanation,
> instead of just handwaving.

Suggesting a hypothesis which explains the observations
in principle is a constructive contribution - not
handwaving. A hypothesis doesn't have to be quantitative
to warrant consideration and provoke people to think of
better ideas than I suggest.

If we consider spiral galaxies to be vastly older than the
BBT allows, then what do you think of the idea of their
collapsed stellar remnants being so robust in close
gravitational encounters (compared to active stars) that
they are flung out to orbits wider and wider and not
aligned with the galactic plane? There, over time, I think
they would occasionally collide and so have a much larger
surface area than if they remained intact black dwarfs.

I do not expect anyone to accept this hypothesis seriously.
Its just a suggestion to show that there are other
explanations for the observations which do not depend on
the BBT.

> Please point out what fine-tuning to observations was
> done in the BBT.

1 - The so-called "Hubble Constant":

H_0: The Incredible Shrinking Constant, 1925-1975
Virginia Trimble, PASP v.108, p.1073-1082

http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1996PASP..108.1073T

2 - When new observations show that mature galaxy clusters
found at redshifts which (according to the BBT) date
them as being not long after the BB, BBT supporters
suggest contorting their theories of galaxy formation
into ever shorter timeframes rather than question the
validity of the BBT.

3 - When no transverse proximity effect is found with a
foreground quasar, BBT supporters pursue a line of
quasar theory which is at odds with all previous
interpretations of other types of observations,
rather than question whether redshift is really as
reliable an indicator of distance as the BBT says it
is.

The first example is a series of quantitative revisions,
each probably approximately as confidently made as today's
"13.7 +/-0.2 Gigayear" estimate.

The latter two are qualitative examples. These are
instances of BBT supporters choosing to revise existing
theories in dramatic ways - to the point where many
objections can easily be made and where the revisions are
destructive of some probably sensible existing theories -
rather than question the veracity of the BBT's insistence
on how substantial redshift can only be caused by Doppler,
expansion or whatever you want to call it.

Bjoern Feuerbacher

unread,
Mar 31, 2005, 3:48:04 AM3/31/05
to
r...@firstpr.com.au wrote:
> Bjoern,
>

[snip to the point]

>>Please point out what fine-tuning to observations was
>>done in the BBT.
>
>
> 1 - The so-called "Hubble Constant":
>
> H_0: The Incredible Shrinking Constant, 1925-1975
> Virginia Trimble, PASP v.108, p.1073-1082
>
> http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1996PASP..108.1073T

That has nothing to do with "fine-tuning to observations".
In contrast, this is standard science: determining a parameter
from observations, where the parameter often changes as measurements
get better with time.


> 2 - When new observations show that mature galaxy clusters
> found at redshifts which (according to the BBT) date
> them as being not long after the BB, BBT supporters
> suggest contorting their theories of galaxy formation
> into ever shorter timeframes rather than question the
> validity of the BBT.

That again has nothing to do with "fine-tuning to observations".
And the reason here is simple: galaxy formation theories are plagued
with uncertainties, and the computer simulations done on them are
permitted to be inaccurate even by the very researchers working
on that. So it's quite natural to question these first instead
of questioning a well-established theory like the BBT.

> 3 - When no transverse proximity effect is found with a
> foreground quasar, BBT supporters pursue a line of
> quasar theory which is at odds with all previous
> interpretations of other types of observations,
> rather than question whether redshift is really as
> reliable an indicator of distance as the BBT says it
> is.

Essentially same comment as above.


> The first example is a series of quantitative revisions,
> each probably approximately as confidently made as today's
> "13.7 +/-0.2 Gigayear" estimate.
>
> The latter two are qualitative examples. These are
> instances of BBT supporters choosing to revise existing
> theories in dramatic ways

You conveniently ignore that people working on galaxy formation
freely admit that the current models are far from accurate,
and a lot of work has to be done on them. It's quite natural
to look for errors in uncertain theories first, don't you think?
Have you ever read an article on galaxy formation?


> - to the point where many
> objections can easily be made and where the revisions are
> destructive of some probably sensible existing theories -

Simply wrong for point 2. Don't know for point 3.

> rather than question the veracity of the BBT's insistence
> on how substantial redshift can only be caused by Doppler,
> expansion or whatever you want to call it.

Expansion. If you want to argue against the BBT, at least
use the right terms to describe it.


Bye,
Bjoern

Phillip Helbig---remove CLOTHES to reply

unread,
Mar 31, 2005, 3:47:24 AM3/31/05
to
In article <mt2.0-7327...@hercules.herts.ac.uk>,
"r...@firstpr.com.au" <r...@firstpr.com.au> writes:

> Bjoern, what do you think of the failure to find evidence for
> the transverse proximity effect with a foreground quasar?

There are observations of the proximity effect: Jakobsen et al., A&A
397, 891(2003).

> The conventional view is that the quasars must be turning on
> and off, or have very short lifetimes. I think a better
> explanation is that the quasars are not located where the BBT
> says they are - due to most of the redshift of their light,
> including probably most or all of the Lyman alpha forest,
> occurring in space near to them. My best guess is that this
> occurs due to some kind of plasma redshift or sparse particle
> redshift mechanism.

Note that in gravitational lens systems, where observations without any
redshifts indicate which is the foreground and which is the background
object, it invariably turns out that the background object has the
larger redshift.

> As far as I know, quasars were not generally considered to
> have short lifetimes until this lack of TPE business arose.

There might not have been any direct evidence for it, but how can there
be, when we've been observing QSOs for only a few decades.

> If quasars are the same as, or cousins to, "radio galaxies"
> then its hard to imagine them having such short lifetimes
> since (according to BBT theories) these radio galaxies have
> such huge lobes that they must have been running continually
> for very long periods of time.

Many QSOs ARE variable. Variability can lead to a lack of the TPE where
one would otherwise "expect" it. It doesn't mean the whole QSO turns
off forever.

Max Keon

unread,
Mar 31, 2005, 3:47:49 AM3/31/05
to
Bjoern Feuerbacher wrote:
>Max Keon wrote:
>>Bjoern Feuerbacher wrote:
>>>Max Keon wrote:

> [snip]

>>>>The contents of this link http://www.ozemail.com.au/~mkeon/cmb.html
>>>>is an extract from a theory which describes a universe that
>>>>originated from absolutely nothing, and it provides an alternative
>>>>explanation for the CMBR.
>>>
>>>
>>>Can it explain why the spectrum of the CMBR is such a nice blackbody,
>>>without any spectral lines?
>>
>>Yes.

> Your link above goes to a page which mainly contains curves and not
> many explanations, as far as I can see. Could you please explain here
> shortly what the source of the CMBR is in your model,

-----
-----

I wrote:
------------------


The temperature at the origin was zero. The universe is evolving.
Its temperature is increasing at a logarithmic rate, hence the
^1.12 adjustment to each (equally spaced relative to a fixed time
zone) curve generated from the Planck equation,
#=((8*pi*h*f^3)/(c^2*(EXP((h*f)/(k*t))-1))) ^1.12

------------------

"(equally spaced relative to a fixed time zone)" should read
"(equally spaced relative to the present)". A fixed time zone could
be anywhere, even close to the origin of the universe.

That error prompted me to check what was actually on the web page.
I was alarmed to find another three such errors, which would confuse
the hell out of anyone trying to understand it. I've listed the
offending part paragraphs below.

The remaining three black curves were each generated in three equally
separated stages in the evolution of the universe, from the origin.
^^^^^^^^^^^^^^^
Should read "from the present to the origin."


The next set of graphs is a re-run of this set, but includes twice
the number of (equally spaced from the origin) past blackbody
sources ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Should read "equally spaced from the present to the origin"


I could add any number of equally spaced curves, from the zero
origin to the present, ^^^^^^^^^^^^^
^^^^^^^^^^^^^^^^^^^^^
Should read "from the present to the zero origin"


This paragraph toward the end of the page
highlights why they are deemed errors.

But whatever the case, I cannot possibly perceive my true
circumstances of existence from where I am. I'm always the zero
point of my measuring stick to the universe. I measure 10 billion
light years to a past realm and then 20 billion light years to
another and conclude that the distance to the latter realm is twice
that to the former. But if the same measurement is taken from the
true zero mark, at the origin of the universe, the two distances
could be only infinitesimally less than identical.

Another thing I noticed while I was reading is the fact that I
hadn't adequately addressed the problem at hand. Even though the
curves were generated according to a kind of inverted logic from
the zero origin universe, when I wrote that page, it was clear to
me that words were the only problem, and I imagined that it would
be clear to anyone else. But now that my memory of the page content
has faded, I see now that it just leads to total confusion.

I thought I had found a number that would identify the current state
of evolution of the universe, and that became the priority. But what
I apparently found was more to do with the scale of the graph than
anything else.

I've stored a set of updated graphs at
http://www.ozemail.com.au/~mkeon/graphs.html
which were generated from a modified version of the Qbasic program
that generated the original graphs. The combined input curves are
first shifted to the hotter side of the CMB power spectrum peak,
then redshifted using the correct logic of the zero origin universe.

I'll update the page as soon as I can.
-----------

The link I didn't post:

"Fluctuations arising from the Sunnyaev-Zel'dovich (SZ) effect,
the up-scattering of the background spectrum by both the hot gas
surrounding galaxy clusters and the peculiar velocity of the
cluster, should be observable on spatial scales of around 3
arcminutes." (I've lost the link. I'll post it next time)

http://www.phys.unsw.edu.au/jacara/Papers/pdf/pasp97_mgb.pdf

-----

Max Keon

Ulf Torkelsson

unread,
Mar 31, 2005, 5:27:13 AM3/31/05
to

I cannot see a major problem at this stage in either
of the explanations. It appears today that essentially
all galaxies are harbouring a massive black hole in their
centre. Therefore it is likely that all these galaxies
have gone through one or more quasar phases in their youth.
It is well known that the quasar density is much higher at
high redshifts than in our local universe.

Let us know look at the arguments in favour of either
of the explanations of the lack of this proximity effect,
though I must admit that I do it without having looked at
the papers that report on this. Beaming would not be
surprising. Other kinds of active galaxies, in particular
Seyfert galaxies, show strong signs of that the black
holes are surrounded by dost tori on a scale of say 100 pc.
This dust torus is absorbing most of the optical and
ultraviolet radiation that is emitted in the plane of the
torus. For that reason we observe two kinds of Seyfert
galaxies, Seyfert 1s that we observe face on, so that we
see straight down to the region surrounding the black
hole in the centre, and Seyfert 2s that we see from the
side, that is through the dust torus. The Seyfert 2
galaxies are thus missing the ultraviolet radiation
and the broad spectral lines that are formed close to
the black hole. Such a dust torus in the quasar could
explain the missing proximity effect.

A short duty cycle may also explain the proximity
effect. If the quasar would only last for a few
million years at a time, which is a short time in
astronomy, then the quasar would have switched of
once its radiation would have ionised the surrounding
gas that would absorb light from the more faraway quasar.
I cannot see a serious problem with a quasar model
that turns on for a few million years, and then goes
back to a dormant stage for say five to ten million
years. The individual outbursts could then perhaps
be the result of the black hole disrupting and
swallowing a giant molecular cloud that comes too
close to it, but there can also be other mechanisms
that can explain these outbursts. The point is
that our knowledge of quasars is still sufficiently
incomplete that we cannot rule out a lot of
models for how they work.

>
> My point is that these researchers, and it seems you too,
> are pursuing a path of quasars being very narrow - in time
> or beaming - for which there is no other obvious supporting
> observations, when a much simpler explanation is that the
> quasars are not located at the distances that the BBT says
> they are.

But there are a host of other observations demonstrating
that the redshifts are a good distance estimator for the
quasar. During the last twenty five years we have observed
a number of gravitational lenses, in which a galaxy at an
intermediate redshift is producing multiple images of a
quasar with a larger redshift. For these systems it is
always the case that the lensed quasar has a significantly
larger redshift than the lensing galaxy, as it should be
if the redshift is a valid distance estimator.

When we observe the Lyman-alpha forest due to
absorption by intervening gas in the spectrum of a
quasar we always find that the Lyman-alpha
absorption lines have smaller redshifts than the
quasar, which once again is compatible with that they
are at a smaller distance from us than the quasar.


>
> If quasars are at distances closer than their BBT-predicted
> redshift distances, then a bunch of other problems are
> solved. For instance the rapid changes in output become
> compatible with quasars of a smaller size and smaller output
> once it can be admitted that quasars are closer than the
> BBT says they are.
>
> It only takes one piercing observation, correctly
> interpreted, to disprove an entire theory. Its not like
> in a democracy where opinions and numbers of votes matter.
>
> The BBT predicts that the quasars and the neutral H is
> exactly where these researchers think they are, but all
> other observations indicate that quasars are not
> exceedingly narrowly beamed and are not prone to having
> short lifetimes.
>
> Rather than question the BBT, the researchers - and you
> too it seems - prefer to pursue a view of quasars which
> is seems to be incompatible with theoretical
> interpretations of a vast number of observations. While
> I think these interpretations are badly skewed by over-
> estimates of distance, I am not aware of any reason to
> question the theories of quasars being big, long-lasting
> and not narrowly beamed, at least in their UV radiation.

The point here is that the big bang theory is a much
simpler and better understood theory supported by simple
observations, while our models for quasars are messy and
not always supported by the ambiguous observations that
we have of quasars, and by that I do not mean the
observations of the redshifts of the quasars, but rather
attempts to explain the features in the spectra of quasars.
Rather than throwing out the simple and well understood
theory, we prefer to think that there is something wrong
in the really messy model of the quasars.

[snipping the rest]

Ulf Torkelsson

Martin Hardcastle

unread,
Mar 31, 2005, 5:58:03 AM3/31/05
to
In article <mt2.0-14058...@hercules.herts.ac.uk>,

r...@firstpr.com.au <r...@firstpr.com.au> wrote:
>No matter whether they chose beaming/shielding or a short
>lifetime for each quasar (perhaps including low duty-cycles
>of on/off radiation) they have a major problem: all these
>explanations involve the actual number of quasars being very
>much larger than is usually estimated.

We discussed this in the newsgroup last year, and I pointed out then
that there really isn't a problem with short duty cycles for quasars.
Here are some further thoughts:

1. The radio galaxy observations that you rely on don't put a very strong
*direct* constraint on the ages of the underlying AGN (must be > 10^6
years in a few cases -- though the best estimates of the ages of those
systems are longer) and in any case there's a selection effect towards
long-lived systems in radio galaxies, which are selected on their
luminosity integrated over time.

2. Short duty cycles for quasars do *not* imply the `actual number of
quasars being very much larger than is usually estimated'. In fact,
they're entirely consistent with the consensus that all massive
galaxies host central massive black holes. In that scenario we might
well expect that every massive galaxy has a good chance of hosting a
number of AGN events in its lifetime (of varying duration and
luminosity -- these will depend on the availability of mass to fall
onto the central black hole and on the rate at which it can do so).

3. Even if point 2 were not correct, which it is, there is no way of
firing up `big bang theory' and predicting the `abundance of quasars'
as you imply. `Big bang theory' has *nothing to say* about the
abundance of quasars. Galaxy formation models need to produce the
required central supermassive black holes on a sensible timescale, and
that *is* an interesting problem, but we are a long way off having the
tools to go directly from the state of the universe on the largest
scales to the dynamics of the matter around a central AGN on scales
comparable to that of the solar system, and that's what we'd need to
do in order to use the properties of AGN to overthrow the entire
theory.

Martin
--
Martin Hardcastle
School of Physics, Astronomy and Mathematics, University of Hertfordshire, UK
Please replace the xxx.xxx.xxx in the header with star.herts.ac.uk to mail me

Max Keon

unread,
Mar 31, 2005, 9:11:49 AM3/31/05
to
Bjoern Feuerbacher wrote:

>
> r...@firstpr.com.au wrote:
>>
>> I haven't studies the Sachs-Wolfe effect. The supposed
>> precision of the BBT theory of CMB doesn't impress me or
>> many other critics.

> Interestingly, most of the critics are not aware of most of
> the evidence...

>> It can be easy to think of other explanations

> Yes. Making up stories without bothering to do actual
> quantitative checks is very easy indeed.

>> - and then, with sufficient effort, to
>> fine-tune them to observations too.

> "fine-tune them to observations, too"? Please point out
> what fine-tuning to observations was done in the BBT.

A background radiation was perhaps predicted, but not at precisely
2.73 K, was it! Doesn't that amount to fine-tuning to observation?
Not that I think there's a problem with that.

-----

Max Keon

Phillip Helbig---remove CLOTHES to reply

unread,
Mar 31, 2005, 10:28:34 AM3/31/05
to
In article <mt2.0-20025...@hercules.herts.ac.uk>, Max Keon
<mk...@ozemail.com.au> writes:

> A background radiation was perhaps predicted, but not at precisely
> 2.73 K, was it! Doesn't that amount to fine-tuning to observation?
> Not that I think there's a problem with that.

Gamow's back-of-the-envelope 1940s estimate was of the right order of
magnitude. The temparature just scales as T = T_0*(1+z). All you need
to know is the density of the universe today. The density at which the
CMB was formed is fixed, as is the temperature. Obviously the density
of the universe today cannot be calculated from first principles, since
it depends on the epoch of observation.

Bjoern Feuerbacher

unread,
Apr 1, 2005, 4:07:46 AM4/1/05
to
Max Keon wrote:
> Bjoern Feuerbacher wrote:
>
>>r...@firstpr.com.au wrote:
>>
>>>I haven't studies the Sachs-Wolfe effect. The supposed
>>>precision of the BBT theory of CMB doesn't impress me or
>>>many other critics.
>
>
>>Interestingly, most of the critics are not aware of most of
>>the evidence...
>
>
>>> It can be easy to think of other explanations
>
>
>>Yes. Making up stories without bothering to do actual
>>quantitative checks is very easy indeed.
>
>
>>>- and then, with sufficient effort, to
>>>fine-tune them to observations too.
>
>
>>"fine-tune them to observations, too"? Please point out
>>what fine-tuning to observations was done in the BBT.
>
>
> A background radiation was perhaps predicted, but not at precisely
> 2.73 K, was it!

Indeed. So what? Please point out when and where the BBT was
"fine-tuned" so that the value of 2.73 K came out. That implies
that the theory was twisted purposefully to make this value come
out. That would be news to me.


> Doesn't that amount to fine-tuning to observation?

No. That amounts to "determining parameters of a theory from
observations". And that's done for *every* theory in physics.


[snip]

Bye,
Bjoern

Max Keon

unread,
Apr 5, 2005, 7:06:32 AM4/5/05
to
Bjoern Feuerbacher wrote:
>
> Max Keon wrote:
>>
>> Bjoern Feuerbacher wrote:
>>>
>>>Your link above goes to a page which mainly contains curves and not
>>>many explanations, as far as I can see. Could you please explain here
>>>shortly what the source of the CMBR is in your model,
>>
>>
>> As I previously indicated, to "explain here shortly" is almost
>> impossible. But the rest of my reply may help.

> We'll see.

>>> and why it has a blackbody spectrum?
>>
>>
>> It's based on temperature change of the universe throughout its
>> evolution from the zero origin. It has the spectrum of the CMBR,
>> just like your theory does.

> How could "temperature change of the universe throughout its evolution
> from the zero origin" explain the existence and the blackbody spectrum
> of the CMBR?

As the universe evolves, its temperature rises. Looking back into
the past universe from the present (anytime), its combined
temperatures, right from the origin, are on display as background
radiation. There are no discrete stages of evolution of course. The
changing curve from the entire past, and the uniform temperature
curve generated in the present must all add up to equal 2.73K.
But the average temperature of the current universe alone must be
higher than when the rest of the background is added to it.

A point which I have never properly explained, anywhere, is how I
arrived at the equally spaced stages in the evolution of the
universe for my graph plots. Firstly, the universe is not expanding.
The noted redshift is due to a reduced speed of light in a lesser
evolved universe. Assuming that the current temperature of the
universe is 3.4K, the temperature at the halfway mark toward the
origin is 1.7K, where the speed of light will be halved, relative
to now.

For a four input stage graph plot that aligns with the CMBR, the
blackbody curves are, .85, 1.7, 2.55 and 3.4 K. But I don't know
what the true uniform temperature of the current universe is, and
that determines what other temperature graphs should be included
from the past. The curve is always much the same though.
-----
-----

>> The temperature at the origin was zero.

> That's contrary to observations, which show that the temperature was
> *greater* in the past. See the link shortly below.

>> The universe is evolving.

> Finally something we agree on.

>> Its temperature is increasing at a logarithmic rate, hence the
>> ^1.12 adjustment

> How do you get from a logarithmic temperature increase to a factor ^1.12?

And I've read it so many times!

>> to each (equally spaced relative to a fixed time
>> zone) curve generated from the Planck equation,
>> #=((8*pi*h*f^3)/(c^2*(EXP((h*f)/(k*t))-1))) ^1.12

> In order to apply the Planck equation, you need something material
> which is in thermal equilibrium. What is this in your model? In the
> standard BB scenario, it was the plasma which filled the early universe.

Thermal equilibrium is achieved over time. Redshift that extends
to light speed noted in the current universe has always been as it
is, right from the origin. At any stage of evolution, that picture
was the same. It's not possible for one clean spectral line to
emerge from that completely blended spectrum.

> BTW, the Planck curve to the power of 1.12 does not give a blackbody
> curve again. You even have problems with the units there!

That has become redundant.
-----
-----

>>>>Dark matter can certainly be explained, if it's required.
>>>
>>>That has nothing to do with my argument above. Try again, please.
>>
>>
>> That argument has nothing to do with a zero origin universe either.

> It is an argument about observational evidence for the BBT. So if you
> claim that you can explain all the evidence which the BBT can explain,
> you need to address this. Why don't you bother?

Why should I explain observational evidence for predictions of
the BBT when it has nothing whatever to do with the zero origin
universe? Anyway, most of the evidence that supports the BBT is
from distant sources. In this unbounded universe, if one searches
for long enough, seeking evidence of some effect which is predicted
by a theory, that evidence will probably be found. But the real
cause for what is observed may be entirely unrelated. Compiling much
of this type of evidence forms a solid foundation which is
unjustifiably hard to wedge apart.

Since the argument is to do with your rejection of the zero origin
universe, cluttering the post with BBT predictions is pointless, so
I've snipped what I consider irrelevant. You can put it all back
again if you want.
-----
-----

>> The all sky picture of the
>> universe from the zero origin is crystal clear. According to that
>> picture, matter is slowly clumping together,

> That's the same as the BBT says.

>> increasing the depth of dimension, of space.

> That's incomprehensible.

Exactly. That's what I've been trying to tell you all along.

>> The picture provides a remarkable insight
>> into how the matter content of the universe is evolving. The picture
>> at the very origin would have contained one infinitesimally minute
>> anisotropy

> That's very close to what the BBT says.

>> within a completely black background.

> That is contrary to the observations.

>> A universe with zero anisotropy would not exist.

> Why not?

Because there is nothing there. When you better understand the zero
origin universe you'll know why.
-----
-----

>>>>If a photon is moving through a
>>>>deepening potential well, it will exit the well with an extended
>>>>wavelength (I think). But that is clearly impossible.
>>>
>>>
>>>Well, then why has this been observed?
>>
>>
>> The fact that a photon wavelength changes according to local
>> gravitational potential may have been confirmed, but not the
>> *assumption* that they gain or lose energy in the process.

> So you disagree with E=hf? Or with f=c/lambda?
>
> If you don't disagree with both, then you get E=hc/lambda, i.e.
> every change in wavelength is equivalent to a change in energy.

This is not a scenario where those equations necessarily apply.
I'm not convinced that the wavelengths undergo permanent change.

>> Consider this; Two adjacent straight lengths of equally spaced
>> billiard balls, labeled (1) and (2), are set in motion along the
>> line of their pointing direction.

> That has little to do with photons and light.

>> Train (1) travels a straight line
>> through free space while train (2) is set to run the gauntlet of a
>> deepening gravitational potential well. Along the journey to the
>> deepest part of the well on (2)'s travels, space-time will be
>> stretching and will of course extend its train length. But because
>> the well is still deepening, (2)'s departure from the well will be
>> further restrained than if the well was constant. However, when (1)
>> and (2) are returned to the same space-time environment they will
>> still measure the same length.

> Why should they?

Every one of the billiard balls have been equally affected by the
deepening well, so when the train emerges to compare with train (1),
the distance between the balls must still be the same. But there is
one major difference. The speed of train (2) will have slowed.
Momentum is lost and that loss must be accounted for. The energy has
obviously been used up in restraining the increasing well depth.

If the balls are replaced with photons which simply follow the
changing geodesic path set up by the deepening well, it would be
a mind boggling challenge to explain why they would shift further
apart (especially if they don't have wavelength????) in order to
overcome a potential momentum change that can't possibly exist.

Are you quite sure that the Sachs-Wolfe effect is valid,,,, in any
circumstance?
-----
-----

>> Not wishing to break from the subject, but the concept of photons
>> as particles has no place in the zero origin universe.

> Well, then how do you explain the photo effect and the Compton effect?
> (quantitatively!)

There does seem to be a case for point source shafts of E/M
radiation. The reaction wavelength will of course remain as it was
created (relative to dimension along its travels) and so too will
the energy carried over the wavelength. But it can't be described
as a particle in the zero origin universe. Planck opened a Pandora's
Box when he did that, in my opinion.

-----

Max Keon

Robin Whittle

unread,
Apr 5, 2005, 7:05:31 AM4/5/05
to
Some people see the BBT as successful and useful - but I
don't. I see it as a huge ship doomed from the start -
but with passengers and crew so transfixed by the size
of the vessel, its long history and the good company they
are in to recognise how the theory fails to explain things
which really must be understood, if the theory is to be
regarded as being as reliable as the proponents seem to
think it is.

I don't see how anyone can take the BBT seriously, in
principle or especially in terms of these supposedly very
precise quantitative estimates of the Hubble "constant", the
"age" of the Universe (13.7 +/- 0.2 = 1.5%), when the BBT
proponents have no proper explanations for some phenomena
(or at least observations we reasonably conclude reflect
phenomena) which seem to be crucial to any understanding of
stars, galaxies and large-scale structure.

I have already mentioned the failure to find the Transverse
Proximity Effect with a foreground quasar. This is an acid
test of the BBT. If the BBT is true, and unless quasars
are much shorter lived, more intermittent or narrowly
beamed than any other observations indicate, then the
effect would be observed. The researchers fully expected
to find it, and they didn't. If they had, I would have
been highly inclined to abandon my critique of the BBT,
if this particular prediction was observed. This is a
quantitative prediction - about where exactly in a
spectrum some absorption will not occur. There's no room
in the BBT for the absorption to be found or not found at
any other part of the spectrum of the background quasar.

Finding this lack of absorption, in a number of objects,
would be so impressive.

The high redshift seemingly old galaxy clusters is likewise
another acid test - unless galaxy formation theory is
contorted into ever shorter periods of time.

Here are some other important phenomena / observations I
think the BBT proponents have so far failed to
satisfactorily explain:

The intergalactic medium (IGM) emitting X-rays which can
best be explained by extraordinarily high temperatures,
such as 440,000,000 Kelvin:

Field, G. B.; Perrenod, S. C. 1977
Constraints on a dense hot intergalactic medium.
ApJ vol. 215, Aug. 1, 1977, p. 717-722.

http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1977ApJ...215..717F

Marshall, F. E. et al. 1980
The diffuse X-ray background spectrum from 3 to 50 keV.
ApJ vol. 235, Jan. 1, 1980, p. 4-10.

http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1980ApJ...235....4M

I don't know of any conventional explanation for such high
temperatures. (My theory is that it is heated by starlight
etc. due to some redshift and/or scattering process which
is not yet properly recognised. It can't easily radiate
the energy, except by getting to such high temperatures,
because it is so sparse that the particles rarely get close
enough to emit bremsstrahlung.) Stars surfaces are only
a fraction of this temperature. We can't even explain
1 Mega Kelvin temperatures in our own Sun's corona - and
the most popular conventional explanations of that are
based on magnetic waves, which clearly can't work out
into the IGM, if only because it is such a lousy conductor
due to it being so thin.

Why galaxy clusters in no way resemble the shape of
gravitationally bound collapsing systems, such as
galaxies or our solar system.

Why the galaxy clusters often are stretched out in space
and resemble liquid squeezed into the gaps between
generally spherical bubbles. (I suggest that the void IGM
is so hot that it is of sufficient pressure, which is
probably very low, to corral the galaxies into the
smallish clusters.)

Why galaxies don't so often come close to each other.
(I figure that galaxies are exuding a corona which
pushes others away. Exactly how the mass of the galaxy is
coupled to this in an aerodynamic fashion, I am not sure,
but a rough guess is that most of the mass is in
black-dwarfs and their potentially numerous and relatively
small collision fragments, which would have a fair bit of
drag. I am not sure how anything could push a star around
to a significant degree, by gas pressure in the
surrounding medium, but maybe not much pushing is required.
Maybe none is required if the visible stars are
gravitationally bound to the larger mass of black-dwarf
fragments which are themselves coupled to the corona of
the galaxy.)

The extra mass in spiral galaxies which presumably causes
the observed visible stellar rotation curves.

The heating and acceleration of stellar coronae and
winds. http://astroneu.com/plasma-redshift-1/#Cranmer

Likewise the nature of solar spicules, the heating and
acceleration of prominences etc.

A whole bunch of things about quasars and AGN:

Why they vary so fast when according to the BBT they are
impossibly large, due to their supposedly high output,
based solely on their distance being according to the
BBT interpretation of redshift.

The nature of jets.

How, if as according to the BBT, there used to be lots
of quasars etc. why there aren't similarly massive
and luminous objects around the place today, such as
in the middle of galaxies.

The CMB. While the BBT has an explanation for the CMB, I
don't think it is the only possible explanation, as I have
written in previous messages in this thread.


I know its a big task to develop cosmological theories.
The BBT is fine as a theory, but I see so many problems
with it that I can't take it seriously.

Other folk don't seem to see or care about the problems
I think are significant - but to me, the BBT really looks
like a great overblown and entirely wrong theory which
will soon be discredited. The key, I think, is coming
up with a good in-principle - and yes Bjoern, Quantitative -
theory of the redshift we observe in stars, galaxies and
AGN.

I am on the case, but I think the first task is to
overcome the problems caused by thinking of
electromagnetic radiation and the quanta of energy which
result from it as involving independent "photons". Once
there is a good redshift theory - especially one we can
test in space or on Earth - then the only remaining task
to deal with is the BBT supporter's interpretation of
supernovae light curves, which are conventionally
understood to show time dilation. Jerry Jensen's critique
looks like a good starting point:

http://arxiv.org/abs/astro-ph/0404207

But it will be a lot of work getting the raw data and
reworking it, whilst paying close attention to all the
difficult questions of corrections and interpretation.

Paul F. Dietz

unread,
Apr 5, 2005, 7:48:59 AM4/5/05
to
the scienceMax Keon wrote:

>>How could "temperature change of the universe throughout its evolution
>>from the zero origin" explain the existence and the blackbody spectrum
>>of the CMBR?
>
>
> As the universe evolves, its temperature rises. Looking back into
> the past universe from the present (anytime), its combined
> temperatures, right from the origin, are on display as background
> radiation.

But the evolution of the universe doesn't cause it to have a temperature
in the sense you want it to, since the universe is not in thermal
equilibrium. More specifically, the energy being released in various
processes (stars, accretion) is nowhere near thermalized. You can't
get blackbody radiation from a nonthermal source.

Also, the evidence is that the temperature of the background radiation
has *declined* with time.

Paul

Bjoern Feuerbacher

unread,
Apr 5, 2005, 10:39:58 AM4/5/05
to
Max Keon wrote:
> Bjoern Feuerbacher wrote:
>
>>Max Keon wrote:
>>
>>>Bjoern Feuerbacher wrote:
>>>
>>>>Your link above goes to a page which mainly contains curves and not
>>>>many explanations, as far as I can see. Could you please explain here
>>>>shortly what the source of the CMBR is in your model,
>>>
>>>
>>>As I previously indicated, to "explain here shortly" is almost
>>>impossible. But the rest of my reply may help.
>
>
>>We'll see.
>
>
>>>> and why it has a blackbody spectrum?
>>>
>>>
>>>It's based on temperature change of the universe throughout its
>>>evolution from the zero origin. It has the spectrum of the CMBR,
>>>just like your theory does.
>
>
>>How could "temperature change of the universe throughout its evolution
>>from the zero origin" explain the existence and the blackbody spectrum
>>of the CMBR?

You haven't explained that below.

> As the universe evolves, its temperature rises.

How could it? Where should the energy come from?


> Looking back into
> the past universe from the present (anytime), its combined
> temperatures, right from the origin, are on display as background
> radiation.

What is "the temperature of the universe"? What exactly is the
source there?


> There are no discrete stages of evolution of course. The
> changing curve from the entire past, and the uniform temperature
> curve generated in the present must all add up to equal 2.73K.

So you say that different blackbody curves added up give again
a blackbody curve?


> But the average temperature of the current universe alone must be
> higher than when the rest of the background is added to it.
>
> A point which I have never properly explained, anywhere, is how I
> arrived at the equally spaced stages in the evolution of the
> universe for my graph plots. Firstly, the universe is not expanding.
> The noted redshift is due to a reduced speed of light in a lesser
> evolved universe.

Does this explain why redshift is *proportional* to distance?


> Assuming that the current temperature of the
> universe is 3.4K,

Why should one assume that?


> the temperature at the halfway mark toward the
> origin is 1.7K, where the speed of light will be halved, relative
> to now.

Why should the speed of light be proportional to the temperature
of the universe?

> For a four input stage graph plot that aligns with the CMBR, the
> blackbody curves are, .85, 1.7, 2.55 and 3.4 K. But I don't know
> what the true uniform temperature of the current universe is, and
> that determines what other temperature graphs should be included
> from the past. The curve is always much the same though.

Shouldn't one *integrate* over blackbody curves of different
temperature?


[snip]


>>>Its temperature is increasing at a logarithmic rate, hence the
>>> ^1.12 adjustment
>>
>>How do you get from a logarithmic temperature increase to a factor ^1.12?
>
>
> And I've read it so many times!

Huh? How is that supposed to answer my question?

>>>to each (equally spaced relative to a fixed time
>>>zone) curve generated from the Planck equation,
>>> #=((8*pi*h*f^3)/(c^2*(EXP((h*f)/(k*t))-1))) ^1.12
>
>
>>In order to apply the Planck equation, you need something material
>>which is in thermal equilibrium. What is this in your model? In the
>>standard BB scenario, it was the plasma which filled the early universe.
>
>
> Thermal equilibrium is achieved over time. Redshift that extends
> to light speed noted in the current universe has always been as it
> is, right from the origin. At any stage of evolution, that picture
> was the same. It's not possible for one clean spectral line to
> emerge from that completely blended spectrum.

You did not answer my question: *what* is at thermal equilibrium
in your model? What exactly is the *source* of the CMBR?


I notice that you snipped this link:
<http://www.astro.ucla.edu/~wright/stdystat.htm#Tvsz>
and entirely ignored my demand that you address this evidence.


[snip]

>>>>>Dark matter can certainly be explained, if it's required.
>>>>
>>>>That has nothing to do with my argument above. Try again, please.
>>>
>>>
>>>That argument has nothing to do with a zero origin universe either.
>>
>>It is an argument about observational evidence for the BBT. So if you
>>claim that you can explain all the evidence which the BBT can explain,
>>you need to address this. Why don't you bother?
>
>
> Why should I explain observational evidence for predictions of
> the BBT when it has nothing whatever to do with the zero origin
> universe?

I did not say that you should explain observational evidence
"for predictions of the BBT". I said essentially that you should
explain observational evidence which is simply there, and which the
BBT *can* explain. Don't you understand the difference?

My original argument was: if the CMBR is assumed to have a

cosmological origin, the parameters we derive from it (Hubble
parameter, density of dark energy etc.) are nicely

consistent with determinations using other methods.

You *still* have not explained how that could be possible if the
CMBR has *not* the origin the BBT proposes.

> Anyway, most of the evidence that supports the BBT is
> from distant sources. In this unbounded universe, if one searches
> for long enough, seeking evidence of some effect which is predicted
> by a theory, that evidence will probably be found.

But the argument above was not about "evidence of some effect which
is predicted by [the] theory". Don't you understand the argument,
or are you delibarately avoiding it?


> But the real
> cause for what is observed may be entirely unrelated.

"may". Indeed. You are free to present an alternative explanation.
So far, you haven't.


> Compiling much
> of this type of evidence forms a solid foundation which is
> unjustifiably hard to wedge apart.

There is nothing "unjustifiable" about that.


> Since the argument is to do with your rejection of the zero origin
> universe, cluttering the post with BBT predictions is pointless,

Again: the above was *not* about BBT predictions. Read the argument
again. Try to understand it this time. And then address it, instead
of trying to weasel out of it again.


> so I've snipped what I consider irrelevant.

In other words: you simply snip all evidence for the BBT which you
can't explain, and then claim that your model can explain all the
existent evidence. How convenient.


> You can put it all back again if you want.
> -----
> -----
>
>
>>>The all sky picture of the
>>>universe from the zero origin is crystal clear. According to that
>>>picture, matter is slowly clumping together,
>
>
>>That's the same as the BBT says.
>
>
>>>increasing the depth of dimension, of space.
>
>
>>That's incomprehensible.
>
>
> Exactly. That's what I've been trying to tell you all along.

Well, since it was you who coined the phrase "increasing the depth of
dimension, of space", and not actual cosmologists, then why do you
agree with me that this phrase is incomprehensible? If even you
yourself think it is incomprehensible, why did you coin it?

>>>The picture provides a remarkable insight
>>>into how the matter content of the universe is evolving. The picture
>>>at the very origin would have contained one infinitesimally minute
>>>anisotropy
>
>
>>That's very close to what the BBT says.
>
>
>>>within a completely black background.
>
>
>>That is contrary to the observations.

I notice you choose to ignore that.

>>>A universe with zero anisotropy would not exist.
>
>
>>Why not?
>
>
> Because there is nothing there.

"zero anisotropy" does not imply "nothing is there".


> When you better understand the zero origin universe you'll know why.

As long as you continue to simply ignore the evidence for the BBT,
I see no reason to try understanding your model.

For example, you apparently now also ignore the Sunyaev-Zel'dovich
effect and the integrated Sachs-Wolfe effect.

>>>>>If a photon is moving through a
>>>>>deepening potential well, it will exit the well with an extended
>>>>>wavelength (I think). But that is clearly impossible.
>>>>
>>>>
>>>>Well, then why has this been observed?
>>>
>>>
>>>The fact that a photon wavelength changes according to local
>>>gravitational potential may have been confirmed, but not the
>>>*assumption* that they gain or lose energy in the process.
>
>
>>So you disagree with E=hf? Or with f=c/lambda?
>>
>>If you don't disagree with both, then you get E=hc/lambda, i.e.
>>every change in wavelength is equivalent to a change in energy.
>
>
> This is not a scenario where those equations necessarily apply.

Which one of these two do not apply, and why not?

> I'm not convinced that the wavelengths undergo permanent change.

So you think they change only temporarily, and some time later
magically return to their original values, or what?

>>>Consider this; Two adjacent straight lengths of equally spaced
>>>billiard balls, labeled (1) and (2), are set in motion along the
>>>line of their pointing direction.
>
>
>>That has little to do with photons and light.

I notice you choose to ignore that.

>>>Train (1) travels a straight line
>>>through free space while train (2) is set to run the gauntlet of a
>>>deepening gravitational potential well. Along the journey to the
>>>deepest part of the well on (2)'s travels, space-time will be
>>>stretching and will of course extend its train length. But because
>>>the well is still deepening, (2)'s departure from the well will be
>>>further restrained than if the well was constant. However, when (1)
>>>and (2) are returned to the same space-time environment they will
>>>still measure the same length.

I'm not entirely sure what scenario you propose here. Does train (2)
go down the well and then up again, or only down?

>>Why should they?
>
>
> Every one of the billiard balls have been equally affected by the
> deepening well, so when the train emerges to compare with train (1),
> the distance between the balls must still be the same. But there is
> one major difference. The speed of train (2) will have slowed.
> Momentum is lost and that loss must be accounted for. The energy has
> obviously been used up in restraining the increasing well depth.

Sorry, I have no clue what you mean with the last sentence.


> If the balls are replaced with photons

Bad idea. Photons do not behave like billiard balls in most circumstances.


> which simply follow the
> changing geodesic path set up by the deepening well, it would be
> a mind boggling challenge to explain why they would shift further
> apart

No one says that the photons shift further apart. Every individual
photon loses some energy.


> (especially if they don't have wavelength????)

Photons indeed dont have wavelength. Only the corresponding
electromagnetic wave has.


> in order to
> overcome a potential momentum change that can't possibly exist.
>
> Are you quite sure that the Sachs-Wolfe effect is valid,,,, in any
> circumstance?

Yes.

>>>Not wishing to break from the subject, but the concept of photons
>>>as particles has no place in the zero origin universe.
>>
>>Well, then how do you explain the photo effect and the Compton effect?
>>(quantitatively!)
>
>
> There does seem to be a case for point source shafts of E/M
> radiation.

"point source shafts"???


> The reaction wavelength will of course remain as it was
> created (relative to dimension along its travels)

What is "reaction wavelength"?


> and so too will
> the energy carried over the wavelength. But it can't be described
> as a particle in the zero origin universe. Planck opened a Pandora's
> Box when he did that, in my opinion.

You did address neither the photo nor the Compton effect. Big surprise.

Bye,
Bjoern

Joseph Lazio

unread,
Apr 5, 2005, 7:43:02 PM4/5/05
to
>>>>> "RW" == Robin Whittle <r...@firstpr.com.au> writes:

RW> Some people see the BBT as successful and useful - but I don't.
RW> [...]

RW> I don't see how anyone can take the BBT seriously, in principle or
RW> especially in terms of these supposedly very precise quantitative
RW> estimates of the Hubble "constant", the "age" of the Universe
RW> (13.7 +/- 0.2 = 1.5%), when the BBT proponents have no proper
RW> explanations for some phenomena (...) which seem to be crucial to
RW> any understanding of stars, galaxies and large-scale structure.

Notably the BB model is a model for the evolution of the Universe. It
is only applicable on scales on which the density is relatively
uniform. That means it is clearly not applicable to galaxies.

RW> I have already mentioned the failure to find the Transverse
RW> Proximity Effect with a foreground quasar. [...]

I'll confess that I haven't been able to follow closely all of the
discussion on sci.astro.research in recent weeks, nonetheless, I don't
get it. The transverse proximity effect, at least in the one paper I
pulled up quickly (Schirber et al.) is described as being relevant on
scales of roughly 1 Mpc. Given that one needs to look on scales
larger than about 40 Mpc before the Universe approaches a uniform
density, how is the transverse proximity effect relevant?


RW> The high redshift seemingly old galaxy clusters is likewise
RW> another acid test - unless galaxy formation theory is contorted
RW> into ever shorter periods of time.

This issue seems to come up again and again. Many people (Bjoern,
myself, others) have posed the same question: This indicates that
either the BB model or galaxy formation models are incomplete. Why do
you choose the BB model? (Particularly given the relatively crude
level of galaxy formation models?)


RW> Here are some other important phenomena / observations I think the
RW> BBT proponents have so far failed to satisfactorily explain:

Note that most of the examples you cite on relatively small scales
with respect to those relevant to the BB model. Nonetheless ...

RW> The intergalactic medium (IGM) emitting X-rays which can best be
RW> explained by extraordinarily high temperatures, such as
RW> 440,000,000 Kelvin:
[...]

In fact predicted as a result of structure formation. See papers by
Dave and collaborators and Cen and collaborators.

RW> Why galaxy clusters in no way resemble the shape of
RW> gravitationally bound collapsing systems, such as galaxies or our
RW> solar system.

Because they are still forming?

RW> Why the galaxy clusters often are stretched out in space and
RW> resemble liquid squeezed into the gaps between generally spherical
RW> bubbles. (...)

See my comments about structure formation above.

RW> Why galaxies don't so often come close to each other. (...)

Given that there are any number of images of apparently interacting
galaxies and that the Milky Way and the Andromeda Galaxy are falling
together, I submit that this criticism is wrong.


RW> The extra mass in spiral galaxies which presumably causes the
RW> observed visible stellar rotation curves.

O.k., now we're really getting way below the BB model's region of
applicability. Again, for the record, the BB model is a model for the
evolution of the Universe, not for the formation of everything in it.


[...]
RW> How, if as according to the BBT, there used to be lots of quasars
RW> etc. why there aren't similarly massive and luminous objects
RW> around the place today, such as in the middle of galaxies.

Like the supermassive black hole at the center of the Milky Way?


RW> The CMB. While the BBT has an explanation for the CMB, I don't
RW> think it is the only possible explanation, as I have written in
RW> previous messages in this thread.

Of course the BB model isn't the only explanation for the CMB! One
can probably come up with an infinite number of explanations. The
question is what explanation is the most parsimonious yet consistent
with most of the data?

--
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

Max Keon

unread,
Apr 6, 2005, 5:32:44 AM4/6/05
to
I'm expanding on this part of my recent post, if I may.

>> The all sky picture of the
>> universe from the zero origin is crystal clear. According to that
>> picture, matter is slowly clumping together,

> That's the same as the BBT says.

>> increasing the depth of dimension, of space.

Bjoern Feuerbacher wrote:
> That's incomprehensible.

I replied:

Exactly. That's what I've been trying to tell you all along.

--------

I hope you're ready for this.
First of all, Pound an Rebka demonstrated, only, that the frequency
of the characteristic rays emitted from a radioactive iron sample
at the bottom of the tower was slower than when it was shifted to
the top of the tower. If both (top and bottom) measurements were
taken at the top, or at the bottom, of the tower the discrepancy
between them should still be exactly the same. The apparent redshift
caused by the climb from the bottom of the tower was then already
there when it began the climb.

If that has already been noted, or perhaps has gone un-noticed, it
confirms or can confirm the existence of the zero origin universe.
But the Mossbauer effect, which I know very little about, is
probably not capable of performing the required task.

The Pound and Rebka experiment was always my key evidence that the
depth of dimension varies according to local matter content. But
every attempt at describing this variable dimension, that nobody
can comprehend, naturally always ends up in a comprehension
nightmare. I've never succeeded in satisfactorily explaining this,
even to myself. The comprehension nightmare doesn't end here either.

I was hoping to use a failed Sachs-Wolfe effect as a springboard
to help with this latest, very brief, explanation.

Each characteristic wavelength emitted from its parent material
in a gravity well was created from some kind of charged particle
interaction. The interaction rate, and consequent wavelength
creation process, is slowed because the interactive components have
been stretched further apart. And there's no limit to how far the
stretch can go. That added depth of dimension does not exist to
an observer outside the well because it's created by an increased
speed of light, which takes it beyond the realm of the outsider's
existence rate. Neither the added dimension nor the increase in
the speed of light will be noted.

Matter collects together in a gravity well and thus becomes further
apart. How am I supposed to explain that?

In my web page description of the zero origin universe, I've
sidestepped this and other such questions to some degree, hence my
tendency to waffle on in those areas. I had no other choice at the
time, and probably still don't.

-----

Max Keon

Ulf Torkelsson

unread,
Apr 6, 2005, 5:32:05 AM4/6/05
to

If I remember correctly, someone did point to a paper
that reported seeing this lack of absorption. However,
what is more important here is to keep in mind that the
presence of the absorption can be seen as a problem for
any model that predicts that the redshift is a distance
estimator, and therefore it is also a problem for any
"tired light cosmology".


>
> The high redshift seemingly old galaxy clusters is likewise
> another acid test - unless galaxy formation theory is
> contorted into ever shorter periods of time.

Since our current understanding of galaxy formation
is still very primitive, it would not surprise me if
we find in the future that there is a mechanism that
forms galaxies faster than we currently think is possible.

Your information is seriously outdated. For the
last ten years it has been understood that most of
the X-ray background comes from obscured active
galaxies/quasars, and not from a diffuse intergalactic
gas.


>
> Why galaxy clusters in no way resemble the shape of
> gravitationally bound collapsing systems, such as
> galaxies or our solar system.

The universe is only 14 billion years old, so
the clusters have not had time to virialise,
settle down into the shape that you expect for
an old system.


>
> Why the galaxy clusters often are stretched out in space
> and resemble liquid squeezed into the gaps between
> generally spherical bubbles. (I suggest that the void IGM
> is so hot that it is of sufficient pressure, which is
> probably very low, to corral the galaxies into the
> smallish clusters.)

Due to the way the matter was distributed from the
beginning, the collapse may have proceeded faster in
one direction than in the other directions. That will
inevitably lead to that it will collapse into a
pancake shape.


>
> Why galaxies don't so often come close to each other.
> (I figure that galaxies are exuding a corona which
> pushes others away. Exactly how the mass of the galaxy is
> coupled to this in an aerodynamic fashion, I am not sure,
> but a rough guess is that most of the mass is in
> black-dwarfs and their potentially numerous and relatively
> small collision fragments, which would have a fair bit of
> drag. I am not sure how anything could push a star around
> to a significant degree, by gas pressure in the
> surrounding medium, but maybe not much pushing is required.
> Maybe none is required if the visible stars are
> gravitationally bound to the larger mass of black-dwarf
> fragments which are themselves coupled to the corona of
> the galaxy.)

As a matter of fact galaxies come close to each other.
We have observed plenty of interacting and even
colliding galaxies by now. Actually we now think
that the mergers of galaxies played an important role
in the early evolution of the galaxies.


>
> The extra mass in spiral galaxies which presumably causes
> the observed visible stellar rotation curves.

Yes, this is an interesting topic, and it has been
discussed in other threads on this newsgroup recently.


>
> The heating and acceleration of stellar coronae and
> winds. http://astroneu.com/plasma-redshift-1/#Cranmer
>
> Likewise the nature of solar spicules, the heating and
> acceleration of prominences etc.

These are very interesting problems in my opinion,
and many good researchers are working on them, but
they do not have any impact on cosmology.


>
> A whole bunch of things about quasars and AGN:
>
> Why they vary so fast when according to the BBT they are
> impossibly large, due to their supposedly high output,
> based solely on their distance being according to the
> BBT interpretation of redshift.

No, quasars are not impossibly large, though they
are very bright. As a matter of fact the black hole
which is driving the activity in the interior of the
quasar has a radius of around a billion kilometers,
and would fit inside our solar system
>
> The nature of jets.

Another interesting problem, which is completely
unrelated to the big bang theory.


>
> How, if as according to the BBT, there used to be lots
> of quasars etc. why there aren't similarly massive
> and luminous objects around the place today, such as
> in the middle of galaxies.

The centres of most galaxies seem to harbour
massive black holes, but today they are usually
quiescent. Supposedly because there is not enough
gas around them to fuel the activity. My best guess
would be that the remaining gas has been used up
in forming the stars of the galaxies.


>
> The CMB. While the BBT has an explanation for the CMB, I
> don't think it is the only possible explanation, as I have
> written in previous messages in this thread.

There may be other mechanisms, but they have a very
serious problem in forming a spectrum which is
sufficiently close to the theoretical black body
spectrum.


>
>
> I know its a big task to develop cosmological theories.
> The BBT is fine as a theory, but I see so many problems
> with it that I can't take it seriously.
>
> Other folk don't seem to see or care about the problems
> I think are significant - but to me, the BBT really looks
> like a great overblown and entirely wrong theory which
> will soon be discredited. The key, I think, is coming
> up with a good in-principle - and yes Bjoern, Quantitative -
> theory of the redshift we observe in stars, galaxies and
> AGN.

Well, other people know that the problems you
mention do not exist or are not related to the
big bang.


>
> I am on the case, but I think the first task is to
> overcome the problems caused by thinking of
> electromagnetic radiation and the quanta of energy which
> result from it as involving independent "photons".

Good luck, but such a theory will immediately
be in conflict with established experiments like
the photoelectric effect and Compton scattering.
If you cannot resolve these conflicts nobody will
pay any attention to your theory.

[snipping a short discussion on cosmological
supernovae]

Ulf Torkelsson

Phillip Helbig---remove CLOTHES to reply

unread,
Apr 6, 2005, 5:31:26 AM4/6/05
to
In article <mt2.0-14585...@hercules.herts.ac.uk>, "Robin
Whittle" <r...@firstpr.com.au> writes:

> Some people see the BBT as successful and useful - but I
> don't.

What you see doesn't matter; evidence does.

> I don't see how anyone can take the BBT seriously, in
> principle or especially in terms of these supposedly very
> precise quantitative estimates of the Hubble "constant",

Explain the quotes, please.

> the
> "age"

Explain the quotes, please.

> of the Universe (13.7 +/- 0.2 = 1.5%), when the BBT
> proponents have no proper explanations for some phenomena
> (or at least observations we reasonably conclude reflect
> phenomena) which seem to be crucial to any understanding of
> stars, galaxies and large-scale structure.

Tell me, precisely, the relationship between, on the one hand, stars,
galaxies and large-scale structure and, on the other hand, the big-bang
theory.

> I have already mentioned the failure to find the Transverse
> Proximity Effect with a foreground quasar.

I mentioned a paper describing an observation of it. Ignoring evidence
doesn't make you look any better. Here it is again: P. Jakobsen et al.,
A&A, 387, pp. 891--8 (2003). Tell us, exactly, about the mistakes in
this paper.

> This is an acid
> test of the BBT. If the BBT is true, and unless quasars
> are much shorter lived, more intermittent or narrowly
> beamed than any other observations indicate,

Tell us about the other observations which indicate a longer lifetime.

> The high redshift seemingly old galaxy clusters is likewise
> another acid test - unless galaxy formation theory is
> contorted into ever shorter periods of time.

Again, you are confusing the big-bang theory per se with
galaxy-formation theory. Maybe you don't understand what "theory"
means. Science isn't a doctrine or a collection of facts, it's a way of
thinking. Crossing the finish line is just a small part of running a
marathon, and some people would say it's not even the most important
part. Science evolves. There are always some things which are not
completely understood. You seem to think that science should offer
perfect explanations for everything with no refinement based on
observation. The interaction between theory and observation is a
crucial part of science, and usually occurs with no harm to the basic
ideas of the underlying theories. Also, you need to present an
alternative which is BETTER.

> I don't know of any conventional explanation for such high
> temperatures.

The lack of your knowledge is not a reason for other people to give up
the big-bang theory.

> (My theory is that it is heated by starlight
> etc. due to some redshift and/or scattering process which
> is not yet properly recognised.

Theory? Details, please. This sounds like idle speculation.

> Why galaxies don't so often come close to each other.

Rubbish. Interactions are common and the evidence observed.

> (I figure that galaxies are exuding a corona which
> pushes others away. Exactly how the mass of the galaxy is
> coupled to this in an aerodynamic fashion, I am not sure,

Again, speculation.

> The extra mass in spiral galaxies which presumably causes
> the observed visible stellar rotation curves.

Interesting, to be sure, but this has nothing to do with the big-bang
theory per se.

> Why they vary so fast when according to the BBT they are
> impossibly large,

Reference, please, for "impossibly large".

> How, if as according to the BBT, there used to be lots
> of quasars etc. why there aren't similarly massive
> and luminous objects around the place today, such as
> in the middle of galaxies.

According to biology, there used to be lots of dinosaurs etc. Why
aren't there similarly massive animals around the place today?

Bjoern Feuerbacher

unread,
Apr 6, 2005, 6:55:33 AM4/6/05
to
Robin Whittle wrote:

[snip most - other people have addressed that already well]


> Once there is a good redshift theory - especially one we can
> test in space or on Earth - then the only remaining task
> to deal with is the BBT supporter's interpretation of
> supernovae light curves, which are conventionally
> understood to show time dilation. Jerry Jensen's critique
> looks like a good starting point:
>
> http://arxiv.org/abs/astro-ph/0404207

You really think that is a good starting point?

Jerry Jensen
1) uses scatter plots for which he himself admits that the fit
to the line he uses isn't very good, and just looking at the graphs
shows nicely how bad it actually is.
2) claims that a Malmquist type II bias could account for the data.
But articles examining time dilation of SN light curves already
discuss such a bias and come to the conclusion that it can't explain
the data. See e.g. section 5.3 of astro-ph/0309368.
3) claims that SNIc could be mistaken for SNIa. But that possibility
is also taken into account by the SN researchers (see e.g. section 5.2
of astro-ph/0309368). And Mr. Jensen does not explain where all the
SNIa are if all or most of the observed SNs are really SNIc.
4) His note 1 (page 28) shows that he misunderstood both the results
of the SN study of Riess (2004) and the implications for the theory.
He claims that according to Riess et al, the universe has stopped
expanding some time ago (at the highest redshifts) and started to
contract. In fact, what Riess et al. found was that the universe was
always expanding, but in the beginning that expansion was decelerating
and only some billion years ago it began to accelerate. And that's
*exactly* what the Lambda CDM model predicts!

I did not bother to look at all of Jensen's article; but the four
points above are already enough to relegate it to the dustbin.

[snip]

Bye,
Bjoern

Bjoern Feuerbacher

unread,
Apr 6, 2005, 7:17:46 AM4/6/05
to
Max Keon wrote:
> I'm expanding on this part of my recent post, if I may.

I would prefer if you finally started to address all the evidence
and the arguments you keep ignoring.

>>>The all sky picture of the
>>>universe from the zero origin is crystal clear. According to that
>>>picture, matter is slowly clumping together,
>
>
>>That's the same as the BBT says.
>
>
>>>increasing the depth of dimension, of space.
>
>
> Bjoern Feuerbacher wrote:
>
>>That's incomprehensible.
>
>
> I replied:
> Exactly. That's what I've been trying to tell you all along.
> --------
>
> I hope you're ready for this.
> First of all, Pound an Rebka demonstrated, only, that the frequency
> of the characteristic rays emitted from a radioactive iron sample
> at the bottom of the tower was slower than when it was shifted to
> the top of the tower. If both (top and bottom) measurements were
> taken at the top, or at the bottom, of the tower the discrepancy
> between them should still be exactly the same.

Err, how on earth do you arrive at that conclusion???


> The apparent redshift
> caused by the climb from the bottom of the tower was then already
> there when it began the climb.

Incomprehensible.

> If that has already been noted, or perhaps has gone un-noticed, it
> confirms or can confirm the existence of the zero origin universe.
> But the Mossbauer effect, which I know very little about, is
> probably not capable of performing the required task.

Hint: knowing little about something can be cured by learning.


> The Pound and Rebka experiment was always my key evidence that the
> depth of dimension varies according to local matter content.

What on earth does "depth of dimension" mean?


> But every attempt at describing this variable dimension

What do you mean with "variable dimension"?


> that nobody
> can comprehend, naturally always ends up in a comprehension
> nightmare. I've never succeeded in satisfactorily explaining this,
> even to myself.

So you made up something which even you yourself can't understand?


> The comprehension nightmare doesn't end here either.
>
> I was hoping to use a failed Sachs-Wolfe effect as a springboard
> t