The completely revolutionary threat to toppling the last of Big Bang's supporting pillars came not from outside mavericks or the fringes, but from among the respected ranks of the professionals. And from its reactions, it seems that the Establishment reserves its most savage ire for insiders who dare to question the received dogma by putting observation before theory and seeing the obvious when it's what the facts seem to say.
Halton Arp comes from a background of being one of America's most respected and productive observational astronomers, an old hand at the world-famous observatories in California and a familiar face at international conferences. Arp's Atlas of Peculiar Galaxies has become a standard reference source. Then, in the 1960s and '70s, "Chip" started finding excess densities of high-redshift quasars concentrated around low-redshift galaxies.
A large redshift is supposed to mean that an object is receding rapidly away from us; the larger the shift, the greater the recession velocity and the distance. With the largest shifts ever measured, quasars are by this reckoning the most distant objects known, located billions of light-years away. A galaxy showing a moderate shift might be thousands or millions of times less. But the recurring pattern of quasars lying conspicuously close to certain kinds of bright galaxies suggested some kind of association between them. Of course, chance alignments of background objects are bound to happen from time to time in a sky containing millions of galaxies. However, calculating how frequently they should occur was a routine statistical exercise, and what Arp was saying was that they were being found in significantly greater numbers than chance could account for. In other words, these objects were associated in some kind of way. A consistently recurring pattern was that the quasars appeared as pairs straddling a galaxy.
The first reactions from the orthodoxy were simply to reject the observations as being incorrectbecause they had to be. Then a theoretician named Claude Canizares suggested an explanation whereby the foreground galaxy acted as a "gravitational lens," magnifying and displacing the apparent position of a background quasar. According to Einstein's theory, light rays passing close to a massive body will be bent by its gravity (although, as discussed later in the section on relativity, other interpretations see it as regular optical refraction). So imagine a massive foreground galaxy perfectly aligned with a distant quasar as viewed from Earth. As envisaged by the lensing explanation, light from the quasar that would otherwise pass by around the galaxy is pulled inward into a conejust like light passing through a convex optical lensand focused in our vicinity. Viewed back along the line of sight, it would be seen ideally as a magnified ring of light surrounding the galaxy. Less than ideal conditions would yield just pieces of the ring, and where these happened to be diametrically opposed they would create the illusion of two quasars straddling the intervening galaxy. In other cases, where the alignment is less than perfect, the ring becomes a segment of arc to some greater or lesser degree, offset to one sidemaybe just a point. So quasar images are found close to galaxies in the sky more often than you'd expect.
But the locations didn't match fragmented parts of rings. So it became "microlensing" by small objects such as stars and even planets within galaxies. But for that to work, either the number of background quasars would need to increase sharply with faintness, whereas actual counts showed the number flattening off as they got fainter. Such a detail might sound trivial to the lay public, but it's the kind of thing that can have immense repercussions within specialist circles. When Arp submitted this fact to Astronomy and Astrophysics the editor refused to believe it until it was substantiated by an acknowledged lens theorist. When Arp complied with that condition, he was then challenged for his prediction as to how the counts of quasars should vary as a function of their apparent brightness. By this time Arp was becoming sure that regardless of the wrecking ball it would send through the whole cosmological edifice, the association was a real, physical one, and so the answer was pretty easy. If the quasars were associated with bright, nearby galaxies, they would be distributed in space the same way. And the fit between the curves showing quasar counts by apparent magnitude and luminous Sb spiral galaxies such as M31 and M81galaxies resembling our ownwas extraordinarily close, matching even the humps and minor nonlinearities. 59
Arp's paper detailing all this, giving five independent reasons why gravitational lensing could not account for the results and demonstrating that only physical association with the galaxies could explain the quasar counts, was published in 1990. 60 It should have been decisive. But four years later, papers were still reporting statistical associations of quasars with "foreground" galaxy clusters. Arp quotes the authors of one as stating, "We interpret this observation as being due to the statistical gravitational lensing of background QSO's [Quasi-Stellar Objects, i.e., quasars] by galaxy clusters. However, this . . . overdensity . . . cannot be accounted for in any cluster lensing model . . ." 61
You figure it out. The first part is obligatory, required by custom; the second part is unavoidable, demanded by the data. So I suppose the only answer is to acknowledge both with an Orwellian capacity to hold two contradictory statements and believe both of them. Arp's paper conclusively disproving lensing was not even referenced. Arp comments wearily, "As papers multiply exponentially one wonders whether the end of communication is near."
It's probably worth restating just what's at stake here. The whole modern-day picture of extragalactic astronomy has been built around the key assumption that the redshifts are Doppler effects and indicate recessional velocity. Since 1929, when Edwin Hubble formulated the law that redshift increases proportionally with distance, redshift has been the key to interpreting the size of the universe as well as being the prime evidence indicating it to be expanding from an initially compact object. If the redshifts have been misunderstood, then inferred distances can be wrong by a factor of from 10 to 100, and luminosities and masses wrong by factors up to 10,000. The founding premise to an academic, political, and social institution that has stood for three generations would be not just in error but catastrophically misconceived. It's not difficult to see why, to many, such a possibility would be literally inconceivable. As inconceivable as the thought once was that Ptolemy could have been wrong.
It began when Arp was studying the evolution of galaxies and found a consistent pattern showing pairs of radio sources sitting astride energetic, disturbed galaxies. It seemed that the sources had been ejected from the galaxies, and the ejection had caused the disturbance. This was in line with accepted thinking, for it had been acknowledged since 1948 that galaxies eject radio-emitting material in opposite directions. Then came the shock that time and time again the sources turned out to be quasars, often showing other attributes of matter in an excited state, such as X-ray emissions and optical emission lines of highly energized atoms. And the galaxies they appeared to have been ejected from were not vastly distant from our own, but close by.
These associations had been accumulating since the late sixties, but in that time another kind of pattern made itself known also. A small group of Arp's less conformist colleagues, who even if perhaps not sharing his convictions totally, remained sufficiently open-minded to be sympathetic. From time to time one of them would present observational data showing another pair of radio or X-ray sources straddling a relatively nearby low-redshift galaxy which coincided with the optical images of Blue Stellar Objectsquasar candidates. To confirm that they were quasars required allocation of observation time to check their spectra for extreme quasar redshifts. At that point a dance of evasion would begin of refusals to look through the telescopesliterally. The requests would be turned down or ignored, even when they came from such figures as the director of the X-Ray Institute. When resourceful observers cut corners and made their own arrangements, and their findings were eventually submitted for publication, hostile referees would mount delaying tactics in the form of finicky fussing over detail or petty objections that could hold things up for years.
In the 1950s, the American astronomer Karl Seyfert had discovered a class of energetic galaxies characterized by having a sharp, brilliant nucleus with an emission line spectrum signifying that large amounts of energy were being released there. Arp found their association with quasar pairs to be so strong that it could almost be said to be a predictable attribute of Seyfert galaxies. Spectroscopically, quasars look like pieces of Seyfert nuclei. One of the most active nearby spiral galaxies, known by the catalog reference NGC4258, has a Seyfert nucleus from which the French astronomer G. Courtès, in 1961, discovered a pair of proto-spiral arms emerging, consisting of glowing gaseous matter also emitting the "synchrotron" radiation of high-energy electrons spiraling in magnetic fields. An X-ray astronomer called Wolfgang Piestch established that the arms of gas led like rocket trails to a pair of X-ray sources coinciding perfectly with two Blue Stellar Objects. When the ritual of obstructionism to obtain the spectra of the BSOs ensued, Margaret Burbridge, a Briton with over fifty years of observational experience, bypassed the regular channels to make the measurement herself using the relatively small 3-meter reflector telescope on Mount Hamilton outside San Jose in California, and confirmed them to be quasars. Arp put the probability of such a chance pairing as being less than 1 in 2.5 million.
His paper giving all the calculations deemed to be scientifically necessary, along with four other examples each with a chance of being coincidental that was less than one in a million, was not even rejectedjust put on indefinite hold and never acted upon since. When the number of examples continued growing, as did Arp's persistence, his tenure was suddenly terminated and he was denied further access to the major American observatories. After facing censorship from the journals and ferocious personal attacks in public by prestigious figures at conferences, he left the U.S. in 1984 to join the Max-Planck-Institut für Astrophysik in Germany, who he says have been cooperative and hospitable.
A new generation of high-resolution telescopes and more-sensitive instruments produced further examples of gaseous bridges emitting in the X-ray bands, connecting the quasars to their source galaxies. The configurations could be seen as a composite, physically connected object. But the response of those trained to the orthodox view was not to see them. They were dismissed as artifacts of random noise or instrument errors. I've witnessed this personally. On mentioning Arp's work to a recent astrophysics graduate I was cut off with, "Those are just background noise," although I hadn't mentioned bridges. I asked him if he'd seen any of the pictures. He replied stonily, "I haven't read anything of Arp's, but I have read the critics." Whence, knowing the approved answers is presumably all that is needed. Shades of the Scholastics.
In 1990, the Max-Planck-Institut für Extraterrestrische Physik (MPE) launched the X-ray telescope ROSAT (Röntgen Observatory Satellite Telescope), which was later used to look for a filament connecting the violently disrupted spiral galaxy NGC4319 to the quasarlike object Markarian 205, whose association had been disputed since 1971. Although the prime aim failed (Arp thinks the connection is probably too old now to show up at the energies searched for), it did reveal two new X-ray filaments coming out of Mark205 and leading to point-like X-ray sources. So the high redshift, quasarlike Seyfert ejected from the low redshift spiral was itself ejecting a pair of yet-higher-redshift sources, which turned out to be quasars.
The NGC4319-Mark205 connection was subsequently established by a high-school teacher, when the NASA announced a program making 10 percent of the time on the orbiting Hubble Space Telescope available to the community of amateur astronomers. It seems that the amateur communityfor whom Halton Arp has an extremely high regardhad taken a great interest in his work and were arranging more investigations of nearby quasar connections, drawing their subject matter mainly from Arp's 1987 book, Quasars, Redshifts, and Controversies, which the NASA committees that allocated observation time had been avoiding like the plague. After another amateur used his assigned time for a spectroscopic study of an Arp connecting filament, the Space Telescope Science Institute suspended the amateur program on the grounds that it was "too great a strain on its expert personnel." No doubt.
On this basis, quasars turn out to be young, energetic, high-redshift objects ejected recently, typically from Seyfert galaxies of lower-redshiftin fact, high-resolution X-ray images of the Seyfert galaxy NGC4151 show clearly proto-quasars forming in its nucleus prior to being ejected.
The quasars are not very luminous but grow in brightness as they age and evolve. The enormous brightness that's conventionally attributed to them arises from incorrectly assigned distances that place them on the edge of the observable universe. Arp found that on charts showing quasar positions, pairing the quasars by redshift almost always leads to finding a cataloged Seyfert close to the center point between them.
The process can be taken further. The Seyferts in turn usually occur in matched pairs about some larger, still-lower-redshift galaxy from which they appear to have been originally ejected. This yields a cascade in which large, older galaxies have ejected younger material that has formed into younger companion galaxies around it. The younger galaxies in turn eject material as quasars, which evolve through a sequences of stages eventually into regular galaxies. Corresponding to the age hierarchy at every step is the hierarchy of redshifts reducing as the associated objects become older. Such cascades lead back to massive central spiral galaxies whose advanced age is marked by their large populations of old, red stars. Typically they are found with smaller companion galaxies at the ends of the spiral arms. Companion galaxies are found to be systematically redshifted with respect to the central galaxy, indicating them to be first-generation descendants. The same pattern extends to groupings of galaxies in clusters and of clusters in superclusters.
Our own Milky Way galaxy is a member of the Local Group, centered on the giant Sb spiral M31, known as the "Andromeda" galaxy, which is the most massive of the group. All members of the group, including our galaxy, are redshifted with respect to M31, indicating it to be the source from which the rest were ejected as young, high-energy objects at some time. So, when gazing at the immense disk of M31, now about a million light-years away, familiar from almost every astronomy book, we're looking back at our "parent" galaxyand indeed, we see M31 as having a slight negative redshift, or "blueshift," indicating it to be older.
The next nearest major group to us is the M81 group, again centered on the same kind of massive Sb spiral galaxy as M31. Once more, every major companion to M81 is redshifted with respect to it. In fact there are many clusters like the M31 and M81 groups, which together form the Local Supercluster. At its center one finds the Virgo Cluster, which consists of the full range of morphological galaxy types, the smaller ones showing a systematic redshift with respect to the giant spirals. Apart from M31, only six other major galaxies show a negative redshift. All six are in the Virgo Cluster and consist of giant spiral types of galaxy, marking them as the older and originally dominant members. It's quite possible, therefore, that these are the origin of M31 and our entire Local Group. So with Virgo we are looking back at our "grandparent."
On a final note, all the way down, this hierarchy has exhibited the pattern of new objects being produced in pairs. The Virgo Supercluster itself, viewed in terms of the configuration of its dominant originating galaxies and the clusters of groups they have spawned, turns out to be a virtual twin of the Fornax Supercluster, seen from the Southern Hemisphere.
If redshift isn't a measure of a recessional velocity at all, and hence not of distance either, what does this do to the scale of distances that has been constructed, mapping structures out to 10 billion or more light-years away? Although the observational evidence has been there for twenty years, conventional astronomy has never really accepted that the redshifts are quantized, and has tried strenuously to find arguments to show that there is no quantization. Quantized means that the values are not continuous through the range like heights of points on a hill from bottom to top, but occur in a series of jumps like a staircase. Since, in general, an object can be moving in any direction relative to us, the radial components of the velocities, i.e., the part of the motion that is directly toward or directly away (which is what the Doppler effect measures) should, if redshift indicates velocity, come in all values. Hence, the conventional theory can't allow it not to.
If redshift correlates with galaxy ages, then what quantization would imply is that the ejections of new generations of proto-galaxies in the form of quasars occur episodically in bursts, separated by periods of quiescencerather like the generations of cell division in a biological culture. This fits with the kind of way we'd imagine a cascade model of the kind we've sketched would work. It also has the interesting implication that interpreting the redshift as distance instead of age would give the appearance of galaxies occurring in sheets separated by empty voids, which of course is what the conventional picture shows.
So what happens to the immense distances? it appears that they largely go away. Arp's studies indicate that on an age interpretation basis, the Local Supercluster becomes a far more crowded place than is commonly supposed, with all of the quasars and other objects that we feel we know much about existing within it, and not very much at all beyond. So suddenly the universe shrinks back to something in the order of the size it was before Hubble (or, more correctly, the Hubble advocates who grabbed his constant and ran with it) detonated it. No wonder the Establishment puts Arp in the same league as the medieval Church did Giordano Bruno.
Through the last several pages we've been talking about a hierarchy in which redshift correlates inversely with the ages of galaxies and other cosmological objectsi.e., as redshift increases, they become younger. Is it possible, then, to say what, exactly, redshift is indicating? In short, what causes it?
Isaac Newton performed an experiment in which he suspended a pail containing water on a twisted rope. When the pail is released it spins, and the centrifugal force causes the water to pile up toward the sides, changing the shape of the surface from flat to curved. The question is, in an otherwise empty universe, how would the water "know" whether to assume a flat surface or a curved one? In other words, what determines rotationor for that matter, accelerations in general? Ernst Mach, an Austrian physicist who lived around the turn of the twentieth century, argued that the only sense in which the term has meaning is with respect to the "fixed," or distant stars. So the property an object exhibits when it resists changes of motionits "inertial mass"arises from its interacting with the total mass of the universe. It "senses" that the rest of the universe is out there. Einstein believed that Mach was correct and set out with the intention of developing GRT on a fully Machian basis, but somewhere along the way it turned into a "local" theory.
Jayant Narlikar is director of the Inter University Center for Astronomy and Astrophysics in Pune, India, and has collaborated with Fred Hoyle and others in looking deeply at some of the fundamental issues confronting physics. In 1977 he rewrote the equations of GRT in a more general form, yielding solutions in which mass is not a constant but can take the form of a quantity that increases with time. 62 Now, the way mathematics is taught is that the proper way to solve an equation is to derive the general form first, and then make any simplifications or approximations that might be appropriate to a particular problem. The approximations that Aleksandr Friedmann used in 1922 in solving the GRT equations to produce the expanding universe solution were made in such a way as to force any changes in mass to be expressed in the geometry of the situation instead. This is what leads to the models involving the curved spacetime that helps give relativity its reputation for incomprehensibility, and which science-fiction writers have so much fun with. But with the full range of dynamical expressions that permit mass to vary, curved spacetime isn't needed.
According to Narlikar's version, a newly created particle, new to the universe, begins its existence with zero mass. That's because it doesn't "know" yet of the existence of any other mass out there, which is necessary for it to begin exhibiting the properties of mass. Its "awareness" grows as an ever-widening sphere of interaction with other masses, and as it does so the particle's own mass proceeds to increase accordingly, rapidly at first and leveling off exponentially. Note, this isn't the same process as pair production in an accelerator, which is matter conversion from already existing (and hence "aged") energy. It represents the introduction of new mass-energy into the universe, induced in the vicinity of concentrations of existing matterin the form of short-lived "Planck particles," which according to quantum mechanical dynamics rapidly decay into the more familiar forms.
This, then, is what's going on in the nuclei of energetic galaxies like Seyferts. New matter is coming into existence and being ejected at high velocities because of its low initial mass. As the mass increases it slows to conserve momentum, forming the sequence of quasars, BL Lac Objects (highly variable radio and X-Ray sources transitional between quasars and more regular galaxies), BSOs, and the like, eventually evolving into the galaxy clusters that we see. The universe thus grows as a pattern of new generations appearing and maturing before giving rise to the next, unfolding from within itself. This is certainly no more bizarre than a Big Bang that has all the matter in the universe being created at once in a pinpoint. Furthermore, its fundamental process is one of continual production and ejection of material, which is what's seen everywhere we look, unlike exotic mechanisms built around black holes whose function is just the opposite. And to survive as a theory it doesn't have to depend on the burying and suppression of observational data.
But here's the really interesting thing. Consider an electron in some remote part of the universe (in the Local Supercluster if that's all there is to it), that's still relatively new and therefore of low mass. If it has joined with a nucleus to become part of an atom, and if it makes a transition from one energy state to another, the energy of the transition will be less than that of the same transition measured in a laboratory here on Earth, because the mass involved is less. Thus the emitted or absorbed photon will be lower in energy, which means longer in wavelength, i.e., redder. So the correlation between the age hierarchy and the redshift hierarchy is explained. The reason why young objects like quasars have high redshifts is that high redshifts mean exactly that: recently created matter. Redshifts don't measure velocities; they measure youth, decreasing as matter ages. And for objects that are even older than the massive, luminous spiral that we inhabit, such as its parent, Andromeda, or the dominant galaxies in Virgo that are of the generation before that, it becomes a blueshift.