Some Critiques of the Big Bang Cosmology
| Scientific Paper | |
|---|---|
| Title | Some Critiques of the Big Bang Cosmology |
| Read in full | Link to paper |
| Author(s) | Jean-Claude Pecker |
| Keywords | Cosmology, Big Bang |
| Published | 1997 |
| Journal | Journal of Astrophysics and Astronomy |
| Volume | 18 |
| No. of pages | 11 |
| Pages | 323-333 |
Read the full paper here
Abstract
Still more shocking than the metaphysical assumption of some initial singularity, is the constant insistence upon the so-called cosmological principle of "homogeneity" and "isotropy" of the Universe. Observations do contradict this principle. And to me, the inhomogeneous, fractal at least on a certain scale range, of the distribution of matter is in itself an important cosmological fact, hitherto almost neglected. Moreover difficultties as to the applicability of the second principle of thermodynamics, observations of abnormal redshifts, etc., are casting large doubts not only upon the standard cosmological models, but even on the interpretation of the observed redshift as due solely to a universal expansion.
Overview
Pecker's paper is a conference contribution — delivered to a meeting at which Halton Arp, Jayant Narlikar and Geoffrey Burbidge also spoke — setting out, from the standpoint of a senior professional astronomer at the Collège de France, why he had never accepted the Big Bang. It is not a paper proposing an alternative model. It is a catalogue of the assumptions he regards as unearned, arranged as five complaints: the initial singularity, the cosmological principle, the arbitrary selection of "significant" facts, the unexamined application of the second law of thermodynamics to the universe as a whole, and the accumulating evidence of anomalous Redshifts.
His central objection is methodological. The standard model, he argues, rests less on observation than on a demand for simplicity — "the argument of simplicity is a Pythagorician argument, which cannot convince" — and Occam's razor is being used not to prune hypotheses but to forbid them. He is careful to distinguish an "old big bang", which contained a genuine singularity and carried an obvious metaphysical charge, from a "new big bang" repaired by inflation and quantum-gravity speculation; the passage between the two, he says, happened "without noticing it really", and the new version is "full of scars and repairs, complexifications and perhaps improvements" but is by no means simple. He records, without endorsing it, that Pope Pius XII's 1951 address equating the "fiat lux" with the big bang shocked him in Rome and that such non-scientific arguments, though invalid, have had "a huge impact".
The argument
The singularity and the cosmological principle
Pecker traces his own doubts to 1957 and a dissenting unsigned footnote in L'Astronomie au jour le jour. Even granting Hubble's expansion, he insists, it is observed "now and here, nothing more"; extrapolating it backwards is a separate step. Einstein's 1917 solution introduced both a Cosmological Constant and the cosmological principle of homogeneity and isotropy, whose justification he calls frankly metaphysical — "the stability, infinity and simplicity of the universe". Setting Λ to zero yields the Friedmann solutions, which fit Hubble's law at the price of an infinite energy density at t = 0.
His objection is that the cosmological principle is simply false as an observational statement. Following de Vaucouleurs (1970), he reproduces a diagram of hierarchical structure running from neutron stars up to the largest superclusters, spanning some forty orders of magnitude in density, and cites Hannes Alfvén's judgement that the cellular structure of the universe was the main argument against the big bang. The standard reply — Weinberg's, that homogeneity means homogeneity of a "smeared-out" universe averaged over cells of 108 or 109 light-years — he finds unsatisfactory even if the 3 K background is genuinely cosmological.
He then asks what the general-relativistic solutions would look like for a hierarchical universe. We know how to solve the equations for constant density ρ, he says, but not for ρ = ρ(R). Intuitively, less dense regions expand more slowly, so the measured Hubble constant Hi from any given galaxy is an average over a locally varying H. If in addition the distance-versus-time curves of different structures cannot cross — a continuity requirement — then large structures never pass through a minimum size, and only the smallest could reach near-infinite density. The singularity, on this picture, cannot occur everywhere at once, which is exactly what the standard model requires.
Selecting the facts
The classical big bang is credited with three successes: the linearity of Hubble's law, the existence of an isotropic background, and the light-element abundances. Pecker challenges each. Segal and Nicoll (1986) argued for a non-linear redshift–distance relation. The background could be a local radiation field out to optical depth of order unity in the millimetric, unconnected to any decoupling epoch — and he notes that a 3 K value was predicted in the early 1950s by Finlay-Freundlich and by Max Born on a Tired Light basis, and earlier still by Eddington from the integrated radiation of stars. The helium and lithium abundances are measured only near our own Galaxy and may be local or altered by transport; he cites his own 1972 calculation that a rotating, flattened forming galaxy can expel hydrogen from its poles by Lyman-continuum radiation pressure while helium remains trapped, enriching the galaxy in helium by a non-negligible amount. Olbers' paradox, offered by some as a fourth success, "can be understood in practically all cosmologies" — the Charlier hierarchy was built for precisely that purpose — while the analogous gravitational paradox of Seeliger is easy in a hierarchical universe and hard in the big bang.
He dismisses the then-current controversies as beside the point. Because he sees no reason for the present expansion rate to encode an age, the conflict between globular cluster ages of 15 billion years and a Hubble time 1/H of about 10 billion years, and the whole "missing mass" debate, are for him "completely irrelevant debates".
Entropy and anomalous redshifts
The fourth section raises a question he says has received very little attention: whether the second law can be applied to the universe at all. No closed box can be constructed once gravitation is present, since "there is no wall against gravitation"; yet if the universe is all that exists, its entropy must increase. He notes Tolman's result that entropy growth would make successive cycles of an oscillating universe larger, and canvasses two escapes — Lukash and Novikov's disconnected mini-universes, or a locally well-defined Arrow of Time with some region of spacetime restoring "neguentropy".
The longest list is of anomalous redshifts: the solar limb effect exceeding the gravitational prediction (Adam, Roddier); excess redshift in light passing close to the Sun during eclipses; the dispersion of redshifts among active galaxies and quasars of equal apparent magnitude; systematic redshift differences between galaxies of different morphology in the same group; large-scale periodicity in quasar redshifts (Depaquit, Pecker and Vigier 1985); the smaller periodicities of Tifft; and Arp's discordant associations. Individual items may be spurious, he grants, but each must be dismissed on its own merits, and "a single observed undisputed fact of the sort we just mentioned" would establish the need for a second redshifting agent.
As to what that agent might be, he is deliberately noncommittal — emission-time redshift as Arp suggests, interaction with the particles of a Dirac vacuum, or a geometric effect of spacetime as proposed by Segal. He observes that a "tiring" mechanism would require a non-zero photon rest mass, experimentally bounded below 10−54 g, and argues that assuming it exactly zero is a stronger assumption than leaving it unknown. Had Occam's razor been applied that way in the past, he says, we might have set G = 0 or h = 0. He concedes one point: if the galaxy evolution between z = 5 and z = 0 apparently shown by Hubble Space Telescope observations is confirmed, that is "the strongest argument" in favour of local expansion.
Assessment
The paper's real value is as a statement of what a well-informed opponent of the standard model actually objected to in 1997, and several of its points are sound. The distinction between the "old" and "new" big bang is fair and rarely made so plainly: inflation was introduced to solve a problem (reconciling background isotropy with observed inhomogeneity) that the earlier theory could not handle, and the resulting model is not simple. His insistence that the cosmological principle was originally adopted for reasons of tractability and taste rather than observation is historically accurate. And his refusal to treat Occam's razor as an argument from evidence is a legitimate philosophical point.
On the numbers, most of what is checkable checks out for its date. A Hubble time of "around 10 billion years" corresponds to H0 ≈ 98 km/s/Mpc, the high value then contended for, and against globular cluster ages of 15 billion years the conflict he describes was real. It has since dissolved from both ends: H0 is now measured at 67–73 km/s/Mpc, giving 1/H0 ≈ 13.5–14.5 billion years, while Hipparcos parallaxes revised the cluster distance scale and brought the oldest cluster ages down to roughly 12–13 billion years. The argument he set aside as irrelevant was in fact answered. One figure does not survive checking: the photon mass bound. Ten to the minus fifty-four grams is about 6 × 10−22 eV/c2, which is far tighter than the direct laboratory and planetary-magnetic-field limits of the period and than the current Particle Data Group direct bound of about 10−18 eV (roughly 2 × 10−51 g); values that small come only from model-dependent galactic-field arguments. The number is quoted as though it were a straightforward experimental fact, and it is not. There is also a small internal slip: the de Vaucouleurs hierarchy is said in the text to cover "40 orders of magnitudes in size", whereas the figure caption gives the range as densities, from ρ = 1010 to 10−30.
The larger difficulties are with what has been measured since. The fractal-versus-homogeneous question, which Pecker treats as open, has been settled observationally in the direction he resisted: the 2dF and Sloan Digital Sky Surveys find the galaxy correlation function turning over to homogeneity above roughly 100 megaparsecs, and the baryon acoustic oscillation scale at 150 Mpc is measured as a standard ruler in exactly that regime. In the discussion following the talk Pecker himself concedes the point — "I do not object to homogeneity at 'some' scale" — but he then argues that hierarchy between the stellar and supergalactic scales is "enough to exclude a 'singularity' universe", which does not follow: the standard model requires homogeneity only in the averaged sense Weinberg described, and structure formation from small perturbations is what generates hierarchy at smaller scales.
The tired-light suggestion faces two measurements that the paper does not confront. The COBE FIRAS spectrum, already published in 1990, shows the background to be a blackbody to better than one part in 104 — a spectrum that scattering-based redshift mechanisms characteristically distort, and a serious obstacle to the "local radiation field to optical depth 1" proposal. And the light curves of Type Ia supernovae are observed to be stretched by exactly the factor (1 + z), which follows immediately from expansion and does not follow from photons losing energy en route. To this can be added the Tolman surface-brightness test, whose (1 + z)4 dimming has since been measured and favours expansion. Pecker cannot be faulted for the anisotropy results that came later, but the acoustic peak structure now measured by WMAP and Planck constitutes a quantitative success of the model he rejected that has no counterpart in any of the alternatives he canvasses.
Finally, the anomalous-redshift catalogue is presented as cumulative when its items are of very unequal weight, and Pecker offers no analysis of any of them here — he refers to his own earlier work and to Arp. His own strongest methodological principle cuts against him: he asks that each anomaly be assessed individually and in detail, which is precisely what a list of seven citations does not do. The redshift periodicities in particular did not survive larger samples. What the paper offers is a candid statement of dissent from a distinguished observer, not a competing calculation.