ethan@utastro.UUCP (Ethan Vishniac) (05/14/84)
[But why not talk about the BIG picture?]
I have been urged, privately and publicly, to continue this
discussion as long as I can without repeating myself. I'm going
to give this a shot. I assume that if I start repeating myself
too much I'll also hear from my friends about it.
Doubts about the standard model of cosmology arise from the
following points (as near as I can tell):
1) Singularites (such as the t=0 point in the standard model) are
estheticly repulsive and a sure sign that the theory has gone
astray.
2) The general theory of relativity is based on dubious assumptions
about physics and could easily be wrong when considering large
scale phenomena.
3) The picture of an expanding universe is philosophically repugnant
and the available facts can be otherwise explained.
4) The available facts can be otherwise explained so the history of the
universe is essentially unknown, with many equally likely possibilities
competing for attention.
My short responses are as follows:
1) This is also my opinion. However, this does not have any bearing that
I can see on the utility of the general relativity outside of the
domain where we know that quantum gravity must be invoked. Many
useful theories are incomplete in the sense that their domain of
applicability is limited. General relativity is clearly one since it
is not a quantum field theory. Nevertheless, after the first 10^-43
seconds (or rather at all times when the classical theory gives a
larger number for the age of the universe) the question can be reduced
to the question of what the correct classical theory of gravity is.
2) This is true. The only question is whether such an extended theory
can explain the observed large scale evolution (or state) of the
universe as well as GR.
3) I'm not interested in the first part of this one. The second part
is equivalent to point 4.
4) If there are other equally viable theories which can explain the
observed features of the universe I'm not aware of them (doesn't
mean they don't exist).
I'm going to list what I see as the main observational points. However I'd
like to address a few theoretical points first.
- The standard model of the universe is consistent with an infinite, or a
finite universe. All we can say is that the part of the universe we
can see is finite (approximately given by the speed of light times the
age of the universe). This is why the night sky is dark. (Redshift
of distant objects will not, by itself, accomplish this).
- The vacuum may or may not have an associated energy density. That
possibility can, by itself, be accommodated within GR. More complicated
proposals mentioned by D. Gwyn and others require a different (extended?)
theory.
The following observational points must be explained by any satisfactory
theory of cosmology.
1- Objects (like galaxies) are observed to have a redshift which is
proportional to their distance.
2- In all directions there exists a population of unusual objects called
quasars with large redshifts. At least some these objects show
unmistakable signs of really being as far away as their redshifts indicate.
These signs include the observation of distant galaxies and clusters of
galaxies centered on the quasar, and the observation of quasars that are
"lensed" by the gravitational fields of galaxies that are themselves very
distant. Also, quasars have a large space density at large distances and
a very small one nearby.
3- There is an isotropic background of microwave radiation in space. It has
a spectrum which is a perfect blackbody with a temperature of 2.7 K.
The upper limits on anisotropy are very small (depending on the angular
scale a few times 10^-5 to 10^-4). The only exception to this is a dipole
variation in the temperature which can be explained as due to the motion
of our galaxy through space (relative to the rest frame of the radiation
background) at 600 km/sec.
4- Stars have a metallicity (an abundance of elements heavier than Hydrogen
and Helium) which is related to their ages. The oldest stars are the
most nearly pure Hydrogen and Helium. This is clearly proof that our
galaxy is changing as it gets older. Nearby galaxies show similar trends.
The oldest stars appear to be the same age, i.e. we do not see any star
clusters that are clearly older than the oldest stars in our own galaxy.
5- The chemical abundances of elements of low atomic number in nearly
unprocessed material, in our own galaxy or in nearby galaxies appear
to reflect a primordial set of abundances. These abundances are
what one would expect from the standard model. They are inconsistent
with many proposed alternatives.
6- On the largest scales (certainly no greater than a 100 megaparsecs or so)
the distribution of galaxies in space appears to be relatively uniform.
That is to say that variations in density are no greater than a factor of
two in any particular region and possibly a whole lot less.
Now for some detailed comments on these points.
First, I don't think that point number one is in dispute here. However,
the explanation for it in the standard theory needs elaborating. The
redshift effect can be explained in the following three ways - all of
which are equivalent descriptions in the standard model: 1) the universe
is expanding and as each galaxy moves away from its neighbors the light
that reaches its neighbors is redshifted due to a Doppler effect; 2) the
universe is expanding and as light propagates through an expanding spacetime
its wavelength is stretched; 3) the universe is expanding and so light that
reaches us from a distant galaxy was emitted when the gravitational potential
of the universe was larger (in magnitude, the sign is negative) than it is
now so that the light we see has climbed out of a potential well and so is
redshifted by a gravitational effect. In the standard model all these are
equivalent descriptions (although some are more unsatisfactory than others
due to the picture that they evoke). Models that produce this effect without
expansion rely on "tired light" schemes that are used (to my mind) as a kind
of deus ex machina, invoked with no supporting evidence.
The second point has been disputed by Doug Gwyn. I think that reasonable
doubt could be said to exist during the 60's and (to a lesser extent) during
the 70's. Our observations have improved and I don't think any reasonable
doubt remains.
The discovery of the microwave background was one of the turning points
of modern cosmology. The perfect isotropy of the background was recognized
immediately as a strong argument for its extragalactic origin. The fact
that it adheres so closely to blackbody shape shows that it must have been
produced in a gas in thermal equilibrium. The present matter density of
the universe is much too small to produce such a background. The standard
explanation is that we live a "hot" universe with about 10^9 photons per
baryon. As the universe expanded the gas eventually cools to the point
where the loose ions in the gas collect electrons and becomes neutral.
When this happens the universe changes from an opaque gas (due to electron
scattering) to a transparent gas. This happens when the temperature is about
4000 K. Subsequently the light from this redshifts as it streams through
the universe so that the present observed temperature is less than 3K.
Less orthodox explanations involve generating the background from an early
epoch of star formation. Such an event would have to have happened when
the universe was much more dense than it is now. The large number of photons
per baryon in the universe is thought to be related to the matter/antimatter
asymmetry in the universe through processes that occurred at *very* large
temperatures (about 10^28 K). I am not aware of any explanation for the
background in a steady state universe. This is the major reason such
theories have died within the astronomical community.
The metallicity versus age relationship for stars is reasonable proof that
our galaxy, and those near us have evolved from nearly pure hydrogen and
helium (at some early time) to their present states. As near as can be
determined, the time this has taken has been the same for all galaxies
observed.
The fact that the universe seems to have contained a "primordial"
set of abundances that agree with the standard picture is an relatively
unappreciated strength of the model. These abundances were predicted at
a time when they were relatively uncertain, and so constitute a prediction
of amazing accuracy. These abundances do not follow automatically from
"any cataclysmic event". Of course, other models might be found to produce
this result. However, all of those I am aware of fall into the category
of "I don't know what will result so I will assume that my model *could*
give agreement with observation".
The nearly uniform distribution of matter on large scales is simply
mentioned as a confirmation (along with the microwave background) that
the central assumptions of homogeneity and isotropy used in the standard
model are reasonably well confirmed by observation.
Any alternative to the standard model of cosmology has to explain these
points. It should be clear that any theory which takes the "perfect
cosmological principal" (homogeneity in space *and* time) as its starting
point is going to have a tough time of it. One possible fix (from that
point of view) is to assume that the visible, finite universe is a kind of
fluctuation away from a set of stable conditions which existed within
our universe at temperatures comparable to the Planck temperature (at which
gravity must be quantized ~ 10^32 K). The "big bang" is the beginning of
that fluctuation. This suggestion has the "advantage (!?!)" of being so
divorced from observation as to be uncheckable. We could wait for the
regions exterior to our universe to restore equilibrium :-). However,
the boundary is probably advancing upon us at the speed of light. Life
as we know it will be impossible on the other side. (Could this be the
end of the net?).
"Just another Cosmic Cowboy"
Ethan Vishniac
{ut-sally,ut-ngp,kpno}!utastro!ethan
Department of Astronomy
University of Texas
Austin, Texas 78712gwyn@brl-vgr.ARPA (Doug Gwyn ) (05/15/84)
I too have been urged to keep the discussion going, but since I haven't worked in cosmology for a long time (my job title is now Computer Scientist), I am running out of things to say on the topic. My objections to the generally-accepted cosmology are primarily based on theoretical considerations as a long-time student and disciple of Einstein. I especially object to the use of classical General Relativity in the discussion of such things as Black Holes and cosmology. General Relativity is a very special case of a more general geometric fundamental theory called "unified field theory" (from the expectation that other fields than gravitation would be encompassed by the theory). Unified field theory is "classical" in the sense that it explicitly ignores conventional quantum theory, but that is entirely appropriate for its intended purpose, which is to get at the root causes of the universal structure known as laws of physics. There are obvious mechanisms for quantum effects to appear later in the development of the theory, so there is no need to arbitrarily stick them in at the beginning. Indeed, the hallmark of the unified field theory is its reluctance to incorporate ANY "ad hoc" physical principles; only things forced on us as necessary considerations are to be included in the setting up of the theory. This is clearly a radical departure from textbook physics, and so far as I have been able to determine, textbook writers have never understood the rules of this game or what its goals really are. Far from "complicating" General Relativity, the Einstein-Schr"odinger unified field theory REMOVES restrictions imposed for General Relativity (which becomes a VERY special case of the extended theory). There is a formal technique for measuring the degree to which a field theory constrains the subject fields; this shows General Relativity to be relatively highly constraining. Certainly it needs to be "relaxed" if it is going to include more phenomena than gravitation. The E-S unified field theory is based on: (a) a four-dimensional continuum (although this feature has little to do with determining the form of the theory and needs further explanation itself); (b) the principal affine connection of the natural tangent bundle; (c) precious little else. If E-S is invalid, then General Relativity should be even more so. One consequence of the removal of unnecessary restrictions is that the resulting field laws (NOT arbitrary! - see my thesis) are different IN CHARACTER from those of General Relativity, with the expected impact precisely in the areas of cosmology and the very small (nuclear scale). Back to cosmology: Why the night sky is dark (mistakenly called Olber's Paradox) is a very interesting question that is by no means settled. There are explanations besides the finite extent of the observable universe; see Mandelbrot for one such (based on recursive clustering). In any case, almost any sensible cosmological model is going to agree with the Hubble effect so this is not an important point for distinguishing between most cosmologies. The closed, bounded, perfect-cosmological-principle cosmology is actually simpler in any measurable sense of which I am aware than the expanding "big bang" universe. Although I am not convinced that the evidence that quasars are remote is conclusive, this question does not seem to be to bear directly on the question of which cosmology to choose. Nor do I have any problem with stellar or galactic ages. New stars are constantly being born (we have direct photographic evidence for this) and I see no reason that the aging process cannot have a counterbalancing birth process. I was never a "steady state universe" (`a la Hoyle) fan, but that serves as one attempt to show that this might actually happen. I should state that I have not been following the "first N seconds after the big bang" developments, so I do not know to what extent the current elemental abundances depend on the details of that theory of their formation. I do know that there are so many theoretical objections to a "big bang" that I would reserve it for a theory of last resort. The (rough) isotropy and homogeneity of the observed universe is just fine. It makes cosmological modeling easier. The last point of this posting is that THE 2.7oK ISOTROPIC BLACK- BODY BACKGROUND RADIATION ESTABLISHES A PARTICULAR FRAME OF REFERENCE AS "DISTINGUISHED". This is contrary even to the Special Theory of Relativity and needs a VERY skeptical examination. Unfortunately, even from the first detection of this phenomenon (which is known ONLY for this solar system) there was little adverse reaction to this suggestion. This, and the lack of objection to the idea of a universal time reference (which is implicit in the Big Bang model), make me think that our current crop of physicists have either not been well trained in the concepts of relativity, or that they do not understand their fundamental significance. To give one VERY SIMPLISTIC alternative explanation of the 2.7oK radiation, just to show that alternatives to the conventional idea that it is a remnant of the Big Bang are possible, consider that this region of space may be immersed in a gas (which it is) that has some source of energy being supplied to it by local objects (galactic magnetic field, for example). Then the gas would be expected to have a black-body spectrum and it would be expected to be stationary (relatively) with respect to the solar system. There is no need to postulate a UNIVERSAL privileged reference frame for this radiation, just a source tied to local objects such as our galaxy. Enough for now. I merely wish to urge people to keep an open mind with respect to such (difficult to test) theories. Just because the experts believe something does not mean that they have the right answer, as the history of science has shown.