A Survey of Anomalous Redshifts

Scientific Paper
Title	A Survey of Anomalous Redshifts
Read in full	Link to paper
Author(s)	Hilton Ratcliffe
Keywords	redshift, quasars, galaxies, expansion, cosmology, anomalous, peculiar.
Published	2008
No. of pages	41

Read the full paper here

Abstract

One of the greatest challenges facing astrophysics is derivation of remoteness in cosmological objects. At large scales, it is almost entirely dependent upon the well-established Hubble relationship in spectral redshift. The comparison of galactic redshifts with distances arrived at by other means has yielded a useable curve to an acceptable confidence level, and the assumption of scale invariance allows the adoption of redshift as a standard calibration of cosmological distance. However, there have been several fields of study in observational astronomy that consistently give apparently anomalous results from ever-larger statistical samples, and would thus seem to require further careful investigation. This paper presents a review summary of recent independent work, primarily (for galaxies and proto-galaxies) by teams led by, respectively, D. G. Russell, M. Lopez-Corredoira, and H. C. Arp, and for galaxy clusters and large-scale structures, those of N. A. Bahcall, J.C. Jackson, and N. Kaiser. Included also are several other important contributions that will be fully cited in the text. The observational evidence is presented here per se without attempting theoretical conclusions or extrapolating the data to cosmology.

Overview

The PDF carries the title "A Review of Anomalous Redshift Data"; it is the same paper. Hilton Ratcliffe writes as an independent researcher, and the piece is deliberately a review rather than a theory: he collects observational results that sit badly with the assumption that spectral redshift is a reliable distance calibrator, and declines to draw a cosmological conclusion from them. "The observational evidence is presented here per se without attempting theoretical conclusions or extrapolating the data to cosmology." The scope is likewise limited by choice — morphological anomalies and redshift quantisation (the Karlsson effect) are named as categories of evidence but explicitly excluded from the survey.

The framing argument is that the Hubble relation was never as firmly grounded as its subsequent role suggests, and is least tested where it is most relied upon. Ratcliffe quotes Weinberg on Hubble's original data — that it is "perplexing how he could reach such a conclusion", the velocities being "almost uncorrelated with their distance" — and notes via Hoyle, Burbidge and Narlikar that Hubble in his 1953 George Darwin lecture "was still doubtful if the expansion was real." Since it had been recognised since 1936 that the systematic redshift does not operate within the Local Group, redshift-expansion could not be checked against parallax, the one distance method he regards as fully reliable. His summary: "the Hubble relationship fits best where it is tested least."

The evidence surveyed

What counts as anomalous

Ratcliffe sets out seven criteria, following Arp and the Burbidges' checklist of "properties of nearness, alignment, disturbances, connections". A redshift is anomalous if high-redshift objects cluster near the nuclei of nearby galaxies; if physical connections are seen between objects of very different redshift; if non-redshift indicators place high-redshift objects nearby; if radial alignment suggests ejection from a common origin; if absorption lines place a higher-redshift object in front of a lower-redshift one; and (excluded from the review) if morphology or quantisation contradict the redshift distance. He also frames the anomaly symmetrically: either the redshift is wrong, or the redshift is taken as the benchmark and doubt falls on everything else measured about the object.

Statistical scale

The first question he takes on is whether the associations are simply rare chance alignments — the standard objection to Arp's work. His answer is sample size. Fulton and Arp analysed roughly 118,000 galaxies and 25,000 quasars in the 2dF deep field, comparing counts within a 30′ circle around each galaxy against a concentric annulus of equal area. Bell and McDiarmid found six peaks in the redshift distribution of 46,400 SDSS quasars. López-Corredoira and colleagues worked with the Flesch & Hardcastle all-sky catalogue of 86,009 optical counterparts of radio/X-ray sources. The 2dFGRS lists about 250,000 galaxies. These are no longer the handfuls of the 1960s.

Specific associations

Several systems are given in detail. NGC 3516, which Arp called the "Rosetta Stone" of intrinsic redshift: five X-ray-emitting blue stellar objects within 12 arcmin of the Seyfert galaxy, distributed along its minor axis, with "a very good correlation between their redshift and their angular distance." AM 2230-284: a family of fourteen quasars around one compact parent galaxy, seven with positive and seven with negative velocity shifts, as the Narlikar–Arp ejection model predicts from momentum conservation. Arp adds an internal check — at the conventional redshift distance (z = 2.149, H₀ = 55) the cluster would be 4,456 Mpc away and 181 Mpc across, so expansion of space alone should imprint about 9,955 km s⁻¹ of front-to-back velocity, whereas only 3,600 km s⁻¹ is measured, most of that attributable to ejection.

Markarian 205 and NGC 4319, the bridge on the covers of Arp's books, which Sulentic defended again in 2007 against a Hubble Heritage press release: "You can see the narrow core in the connection, which HST is able to detect because of its excellent resolution. It is seen exactly where we found it in the earlier studies." NGC 7603, where López-Corredoira and Gutiérrez measured the two knots in the filament connecting the galaxy (z = 0.029) to its companion (z = 0.057) at z = 0.394 ± 0.002 and 0.245 ± 0.002 — four objects of four different redshifts on one filament. NGC 7319, where Galianni, Burbidge, Arp and others found a ULX of z = 2.114 eight arcsec from the nucleus of a galaxy at z = 0.022, tested its colour against 32 comparison QSOs from the Hewitt–Burbidge catalogue and found it if anything bluer than average rather than reddened by intervening dust. 3C 343.1, a galaxy at z = 0.344 joined by a radio bridge to a QSO at z = 0.750 within 0.25 arcsec, with an accidental-superposition probability Arp and the Burbidges put conservatively at ~1 × 10⁻⁸. And NEQ 3, where re-observation turned a two-redshift anomaly into a three-redshift one.

Russell's Tully-Fisher survey

The strand Ratcliffe treats as methodologically strongest is David G. Russell's, because it uses no ejection hypothesis and no morphological judgement — only the Tully-Fisher relation, "the second most reliable measure of remoteness at that scale". Calibrating in both B and I bands for Virgo Cluster spirals, Russell reported that the redshift distribution has a morphological dependence "inconsistent with a peculiar velocity interpretation"; that ScI-type galaxies carry a systematic excess redshift relative to H₀ = 72 km s⁻¹ Mpc⁻¹; and that pairs and groups agree well internally by TF while deviating strongly as groups from the Hubble prediction. Taken as velocities, the giant Sb galaxies approach the Milky Way at a mean −898 km s⁻¹ while the giant ScI galaxies recede at +824 km s⁻¹. Ratcliffe draws the sting: "the standard redshift interpretation of velocity would have us believe that galaxies migrate peculiarly by type! The notion of species-dependent universal expansion is an exceptionally strong argument against the Hubble Law." Three of the ScI/Seyfert galaxies (NGC 4321, 4535, 4536) have Cepheid distances, so inaccurate distances are an unlikely explanation. Russell's separate point about scatter is also reproduced: TF scatter is ~±0.35 mag observed in clusters but ±0.64 to ±0.84 mag when calibrated from Hubble distances for field galaxies, which would require either implausible intrinsic scatter or field velocity dispersions above 1000 km s⁻¹.

Clusters and large-scale structure

Two effects are treated as anomalies of a different kind, since both are accepted in the mainstream literature: Jackson's 1972 "Fingers of God", the radial elongation of clusters in redshift space from virial motions, and Kaiser's 1987 flattening from infall on larger scales. Ratcliffe's point is that both are admissions that redshift-mapped structure is not real space. He then cites Neta Bahcall on structures of 100–150 h⁻¹ Mpc: "Observations of galaxies and clusters, however, show inhomogeneities and structure on all scales studied so far … When does the Universe become homogeneous?" — and her judgement that structure on scales ≥ 30 h⁻¹ Mpc "cannot be matched by unadorned CDM models".

Sociology

Three short notes address the conditions of the research: the difficulty of obtaining telescope time (López-Corredoira and Gutiérrez observed about a dozen systems, half of which contained definite anomalies, and were refused time thereafter despite publishing in major journals), the pressure on publication (the NGC 7319 paper was accepted by ApJ only on condition that a background-object argument be inserted), and a flat rejection of conspiracy explanations: "As far as this author is aware, there exists no instance of global conspiracy, on any matter, and in any form whatsoever." He also notes that gravitational lensing has been offered as an alternative account of QSO-galaxy correlations, quoting López-Corredoira that it cannot work at all for the correlations with the brightest and nearest galaxies, which are the most significant ones.

Assessment

As a review this is well-judged in its selection and unusually disciplined in its ambitions. Ratcliffe does not propose a redshift mechanism, does not invoke tired light, and repeatedly stops short of the cosmological inference the material invites — his closing questions are questions, not claims. He is candid about what he has left out, and he chooses his emphasis well: Russell's Tully-Fisher work is the most persuasive item here precisely because it does not depend on the contested judgement of whether a filament or a bridge is real, and the morphology-dependent redshift excess is a genuine puzzle for a naive velocity reading. His observation that the anomalies "result always, and exclusively, from our comparison of the data with theoretical models" is the right framing for a review of this kind. Letting his sources speak at length, rather than paraphrasing them, is also to his credit given how often this literature is characterised second-hand.

The weaknesses are those of an advocate's brief. The probability figures quoted — 10⁻⁸ for 3C 343.1, 10⁻⁹ to 10⁻¹⁰ for the NGC 5985 configuration — are reproduced without independent scrutiny, and the a posteriori objection to them is treated as bad faith rather than as a statistical point. It is a real point: a configuration selected because it looks striking cannot be assigned the probability it would have had if specified in advance, and Arp's counter-question ("what is the combined probability of all these previous cases?") is rhetorical rather than a calculation. The paper never states a trials factor or a look-elsewhere correction for any of its cases. Similarly, the counting statistics from Fulton and Arp are described but their control for the well-established clustering of faint background sources behind bright foreground galaxies is not examined, and this is exactly where the mainstream lensing and selection-effect explanations bite.

Against established measurement, the review's silences are conspicuous. If quasars are nearby ejecta, their observed properties should show it, and the paper does not confront the strongest counter-evidence: quasar host galaxies are resolved at the redshift-implied distances and show the stellar populations and sizes of galaxies at those distances; quasar absorption-line systems (the Lyman-α forest) show intervening material at every redshift between the source and us, which is very hard to arrange for a nearby object; and the time dilation of quasar and supernova variability scales as (1 + z), most sharply demonstrated in the (1 + z) stretching of Type Ia supernova light curves. Ratcliffe's own argument about the superluminal expansion of quasar radio structure — presented as an absurdity of the distance interpretation — has a standard resolution in relativistic beaming that the paper does not mention. On the ULX cases, the systematic follow-up spectroscopy since 2004 has identified the great majority of ultraluminous X-ray sources as accreting binaries within their host galaxies, not background QSOs, which weakens the inference from the NGC 7319 case considerably.

There are smaller lapses of care. The paper asserts that "the calculated charge density of quasars is in some cases so high that it would appear that photons could not likely escape the interior" without a source or a calculation, and the remark that quasar luminosity at cosmological distance "enters the realm of metaphysics" is rhetoric standing in for the accretion-power argument it should be engaging. The Fingers of God and Kaiser effects are presented as anomalies when they are in fact quantitatively predicted consequences of peculiar velocities within the standard picture; treating a successful prediction as evidence against the model is a category error. And a review that excludes redshift quantisation — arguably the most testable of all the claims in this field, and the one most vulnerable to sample-selection critique — leaves out the item on which the case would most cleanly stand or fall.

What survives is worth taking seriously. The Russell morphology result, the NGC 7603 filament with four discordant redshifts on it, and Bahcall's own doubts about structure scales are not disposed of by any of the above, and Ratcliffe's central procedural claim — that a single well-verified exception would be enough to require re-examination of the distance ladder, and that such exceptions have been systematically denied the telescope time needed to verify or refute them — is a fair criticism of how the question has been handled. The paper is best read as it presents itself: a set of pointers for the reader who intends to check further, not a demonstration.

See also

• Hilton Ratcliffe
• Redshift
• Halton Arp
• David G Russell
• Margaret Burbidge
• Jack W Sulentic
• Quasar
• Cosmology
• Big Bang
• Tired Light
• Dark Matter
• Edwin Hubble
• Eric J Lerner
• Tom Van Flandern
• Oliver K Manuel
• Apeiron