[ Home | Essays | Nightmares | Praise | Testimonies | Wisdom]

Black Holes, Einstein and Scientific Objectivity
Stanley Robertson, Ph.D.

Although the public at large is captivated by black holes, their interest is nurtured by theoretical physicists and astronomers intoxicated by the mystique of black holes. There is almost a religious fervor in some of the articles that report observations of new black holes, or report on things that they do that ought to be impossible for a black hole. For example, in astronomy, radio emitting jet outflows are ubiquitous, occurring in young stellar objects, neutron stars, stellar mass black hole candidates and galactic nuclei. In the case of the first two of these, there are detailed computer simulations that show how the spinning magnetic field of the central star drives the jets. But when the same kinds of jets and synchrotron emissions are found in black hole candidates, it is reported as amazing and different and mysterious and they build these models that effectively use tooth fairies to do what a spinning magnet does naturally. In this essay, I will try to explain some of how this has occurred. Part of the explanation is that it is all too human to want to enjoy a special status as practitioner of a science that encompasses something as bizarre as a black hole within a bedrock theory. I have, in times past, enjoyed a bit of this smug satisfaction for myself.

The existence of black holes remains an article of faith among most physicists for several reasons. With Special Relativity theory, Einstein revolutionized physics. When his General Relativity (hereafter,GR) theory correctly predicted three (later, four) small effects that were not part of Newtonian gravity theory, it created a strong sense among both physicists and the public at large that it must also be correct. Because of this prevailing bias in its favor and the fact that it is very difficult mathematically, it has not been subject to the same degree of testing as, for example, quantum mechanics. It Is also not part of every physicist’s studies. Many know no more than they have read in the popular press, including the prediction of black holes. Also, the most striking features of GR only appear where gravity fields are extremely strong, but it has only been tested in controlled experiments within the solar system, where gravity is weak. It is ironic that Einstein hated black holes, but if you speak ill of them it is considered to be an attack on Einstein and you become an instant pariah.

Little did I suspect that my confidence in black holes would be shaken by a 1996 astrophysics publication that offered “proof” of the existence of event horizons. This is the same as a proof of the existence of black holes, because the very essence of a black hole is the event horizon - the boundary from which nothing can escape. The proof was conceptually simple. There are binary star systems in our galaxy that episodically flare brightly in x-rays from a compact source. Some of the compact sources are known to be neutron stars and others are believed to be black holes. The x-rays are produced when gas from the companion star swirls in toward the compact object and is compressed and heated to millions of degrees by its own viscosity. As the flares subside and the systems revert to quiescence, gas flowing at lower rates would merely pass through the event horizon of a black hole and disappear. But the flow into a neutron star surface would end with x-ray emissions from a surface impact. They should be tens of thousands of times brighter than the quiescent black holes. Unfortunately, the scholarship of the “proof” article was so poor and the treatment of observational data was so dishonest that I expected it to become the butt of crude jokes. Instead it has become the proof of what everyone wanted to believe and it is still cited several hundred times per year.

I analyzed the “proof” observations of quiescent luminosities and tabulated the minimum luminosities reported for five neutron stars and five “black holes”; (all of the measurements available at the time). The data were scattered, but with means not different by more than about 10x; and certainly not 10,000x. I re-plotted these data in a way that clearly destroyed the “proof” and submitted the plot and a note to the same journal, but it was rejected. The referee chosen was the author of the "proof" article. That made me angry. I have seen similar shenanigans in other sciences, but never in physics.

That made me determined to try to find out what are the compact objects that are believed to be stellar mass black holes. I spent a year digging through x-ray astronomy data before concluding with some certitude that the objects are very similar to neutron stars, though more massive, and are strongly magnetic. Since being magnetic is not a possible attribute of a black hole, it was pretty clear that the objects were something else, but exactly what, I still don't really know. Since black holes are apparently the offspring of Einstein's general relativity theory, my first thought was that GR theory would have to be replaced by something that didn't predict such monstrosities. GR sets an upper limit of three times the sun's mass for compact (i.e. nuclear density) stars. The objects more massive than that are called black holes, though none of the distinguishing attributes of a black hole have ever been measured. There are some other candidate gravity theories, one of which I used to model the properties of compact stars. In this theory the black hole candidates were essentially fat neutron stars. In support of this calculation, I offered a table of observations and analysis that strongly implied that the observed objects were magnetic and submitted a paper for publication.

It was eventually published in Astrophysical Journal, (which means that the conclusions cannot be easily rejected by the establishment, or they surely would have been) but for the response it got, I might as well have mailed it off to a black hole. Only Darryl Leiter, a theorist who began an e-mail correspondence, showed interest. He took the evidence seriously, and we worked together via e-mail exchanges for a couple of years trying to put some rigor into the gravity theory that I was using. We finally both gave up. That suited him fine. He is a General Relativity theorist from a long ways back; a Ph.D. student of Nathan Rosen, one of Einstein's most important co-authors. Eventually, we found that although GR, as presently used, seems to allow black holes as solutions of the field equations, this is not a physically realizable result. This conclusion has been published by others in several forms over the past twenty years, but it has not made a dent in the establishment that controls grants, grad students, jobs and journals.

Only after several years of study have I learned that the very concept of a black hole is ill founded and not really an essential part of GR. The first solution of Einstein’s GR field equations was obtained one year after Einstein published the equations. It is the solution for the gravity field of a point mass and was obtained by Karl Schwarzschild, a soldier in the German army in WW1. Schwarzschild died at the Russian front in WW1, but Einstein saw to it that his solution was published – in German. Ironically, the solution known as Schwarzschild’s solution is believed to predict the existence of black holes, though it did not. In a paper published in 1979 in The Physical Review, America's most prestigious physics journal, Leonard Abrams showed that David Hilbert, the most famous mathematician of the first half of the last century, had reworked Schwarzschild’s solution and made a mathematical error in the process. To give Schwarzschild due credit for the first solution, Hilbert called this erroneous solution the "Schwarzschild solution". Unfortunately for Schwarzschild, his name has been attached to Hilbert's error. In the erroneous solution there is a distance from the point mass such that time stands still and space becomes infinitely stretched out. This is known as the "event horizon". Einstein commented that the event horizon “just didn’t smell right” and he rejected the notion of black holes.

Although Hilbert’s erroneous mathematical solution of the Einstein field equations would lead to a black hole condition, there is more to the GR theory than that. Also encompassed by the theory is that nothing can travel faster than the speed of light. But the "black hole" solution would require this part of the theory to be violated by anything falling into the black hole. So what we really have is an inconsistency between an apparent solution of the equations and other requirements of the theory.
In Hilbert's erroneous solution, any mass that becomes compact enough to reside entirely inside its event horizon would be a black hole. So the practice that has become the norm in astrophysics is that any object that has such a compact mass is called a black hole without verification of any further properties of a black hole. Only about one in ten authors now a use the more appropriate term, “black hole candidate”, and Hilbert’s theoretical error is the basis for calling astronomical objects black holes.

There are two classes of objects that appear to have this compactness and even neutron stars are within about a factor of two of being compact enough. Neutron star masses appear to all be less than about 1.5 solar mass. But there are similar compact objects over 5 solar mass that, at neutron densities, are too massive to be neutron stars according to Hilbert's solution. These are, of course, called black holes. There are also galactic nuclei that have million to billion solar mass that are also variable sources of light from x-rays to infrared. To account for their rapid and large variations of brightness, the object sizes must be not much larger than the distance light would have to travel to get from one side to another. Thus these objects are compact enough to also qualify as black holes, if such exist.

There are several reasons why I do not like the practice of uncritically calling these compact objects black holes. First, it is simply sloppy science; a biological equivalent might consist of identifying every organism with a cylindrical body as a snake, or calling every creature that produces eggs a chicken. Secondly, General Relativity has been tested only under weak gravity conditions. There is a dimensionless parameter, a gravitational potential energy that would have the value 1/2 at an event horizon. It has the value 6.9X10^(-10) at the earth surface, 1.7X10^(-6) at the surface of the sun, and about 0.2 at a neutron star surface. To get from the solar system, where the theory has been tested, to the realm of neutron stars and black holes require an extrapolation by about a factor of a million. Such a long extrapolation ending in a bizarre state of stretched space and time standing still should not be trusted. Thirdly, every other stellar mass and larger object is known to possess a magnetic field. But this is something that a black hole cannot possess. In my work, I have shown that the spinning magnetic field of the neutron stars drives a great deal of their radiation. I can account quantitatively for their x-ray emissions and jets. The same methodology applied to the stellar mass black hole candidates also accurately accounts for their similar emissions. Fourth, the accepted theory of how black holes produce these emissions relies on matter passing through an event horizon. This theory cannot apply to the neutron stars. They have surfaces that would light up from the surface impact of matter falling in. So to account for what we see, somehow, a bit of matter falling in would have to know what lay at the bottom of the well and know not to radiate if it reached a neutron star surface at the bottom. This is absurd. Fifth, there are other profound similarities between the neutron star systems and the stellar mass black hole candidates. I can show multiple instances of published statements that essentially say that "…we can rule out the involvement of surfaces and magnetic fields as explanations of these phenomena, because we see the same things from neutron stars and black holes..." ; i.e., we cannot see what the theory says that we cannot see, even when what we are seeing is synchrotron (magnetic) or surface radiation. This is simply appalling.

I think that I have demonstrated that it is possible to publish contrary views in top-flight journals if there are no holes in your logic or facts, but you may wind up being ignored anyway. Especially if you make the mistake that I made early on of suggesting that GR might be less than perfect. This is unhealthy, but it is reality. Things changed a little with Darryl Leiter on board, and this is where the story begins with three of my recent papers, which are listed below. The 2002 Astrophysical Journal article analyzes the same data, plus extensions, as my first published paper on the subject, showing magnetic effects, but not trying to explain them with any theory at all. In the second one (2003) we found new solutions of the Einstein equations and showed that GR actually allows some magnetic beasts that are about as compact and even more bizarre than black holes. They are composed primarily of an electron-positron pair plasma. We could not take the entire power output of all nations on earth and produce a cubic micron of a pair plasma! So we have an object compatible with GR and observations that is not a black hole. This is likely the most hated hypothetical object in all of physics at the present time, but it is being taken seriously. Lastly, the MNRAS article shows how these objects can produce the jets and correlated radio and x-ray emissions of all of the black hole candidates, whether star-like (GBHC) or active galactic nuclei (AGN can be millions to billions of times more massive than the sun). The last of the last sentence of the abstract ends with "....GBHC and AGN have observable intrinsic [magnetism] and hence do not have event horizons." This is a deliberate thumb in the eye of the establishment. I doubt that it would have been possible to get such a sentence into print four years ago. I may not live to see this widely accepted, but I think that eventually it will be.

I am not sure what will happen in the near future. I can get my heretical ideas into the "best" journals, but I am often ignored anyway, just as Abrams was and several others have been. The black hole machine keeps rolling along. They find new “black holes” every day. At present it is merely rude to question their status as black holes. In time it may become impossible. If remembered at all, I may merely be a "flat-earth" kind of physicist. There is an orthodoxy that controls grants, jobs, grad students and journals. Who knows how long it will continue? So, is it possible to get good science published in good journals? Maybe, but maybe not. Questioning the existence of black holes makes me a flat-earth sort of physicist and I would have problems obtaining grants or working in a graduate department. Now imagine what questioning aspects of evolution must be like in Biology. One really can't work as a normal biologist with grants, grad students and a job while questioning what are regarded as the foundations of biology. Never mind that a lot of the foundations have distinctly ad hoc qualities. We invoke evolution to explain changes that took place over time, but no one really knows if the appropriate and necessary biochemistry changes took place. The rapidity with which some species changes occurred certainly suggests mechanisms beyond random point mutations. With respect to knowledge of the biochemical basis of life, we are in our infancy.

The one thing that can be published, in any field, is reproducible observations and these are also freely shared. Interpretations are a different matter. There are prevailing orthodoxies in most sciences that control interpretations in the absence of compelling data. For example, if it can be reproducibly and reliably shown that Carbon-14 is found in pre-Cambrian diamond, that result can be published in the best of geology or physics journals. You can bet that it would engender discussion and testing until the interpretation issue it raises was settled! But if you need grant money and facilities for making the measurements, then you might be out of luck. Objectivity may not stretch that far.

Here are five recent references:

1. Evidence for Intrinsic Magnetic Moments in Galactic Black Hole Candidates. Stanley L. Robertson & Darryl J. Leiter, 2002. The Astrophysical Journal V565, p447. In this paper, we showed that the similar quiescent x-ray spectra of
neutron stars and black hole candidates in x-ray nova systems can be quantitatively explained as driven by spinning magnetized objects.

2. On Intrinsic Magnetic Moments in Black Hole Candidates. Stanley L. Robertson & Darryl J. Leiter, 2003 The Astrophysical Journal Letters, V596, pL203.
Black holes cannot possess intrinsic magnetic fields, consequently if the black hole candidates are magnetized, they must be exotic objects of some other kind. Here we propose that they are hot, collapsing and slowly radiating away their mass. We call them magnetic eternally collapsing objects (MECO) and show that they are permitted by General Relativity.

3. On the Origin of the Universal Radio-X-Ray Luminosity Correlation in Black Hole Candidates. Stanley L. Robertson & Darryl J. Leiter, 2004 Monthly Notices of the Royal Astronomical Society, V350, 1391. We show that the MECO model provides a quantitative explanation of x-ray and radio emissions from neutron stars, galactic black hole candidates and active galactic nuclei (quasars and Seyfert galaxy nuclei).

4. The Magnetospheric Eternally Collapsing Object (MECO) Model of Galactic Black Hole Candidates and Active Galactic Nuclei. Stanley L. Robertson & Darryl J. Leiter, Nova Science Publishers, chapter of book, Black Holes Research, in press. This is a full exposition of the MECO model, including its consistency with relativistic gravity theory and quantitative aspects of broadband spectra, temperatures, jet emissions, interactions with accretion disks, etc. We provide scaling laws from neutron stars to stellar mass black holes to galactic nuclei. The MECO model is thus the first unified, consistent model of gravitationally compact objects.

5. Observations Supporting the Existence of an Intrinsic Magnetic Moment
Inside the Central Compact Object Within the Quasar Q0957+561
Rudolph E. Schild, Darryl J. Leiter and Stanley L. Robertson, submitted to
Astronomical Journal, May 2005 The image of quasar Q0957+561 is split into four by gravitational lensing by an intervening galaxy. Within the multiple images are light fluctuations caused by microlensing by objects of planetary size within the quasar's own galaxy. The analysis of these fluctuations permits a determination of small scale structures, such as the accretion disk and compact jets of the quasar. These features match the predictions of a MECO model that is scaled up a billion fold from the realm of stellar mass black hole candidates.

Darryl J. Leiter is a General Relativity theorist, the last Ph.D. student of Nathan Rosen, who was one of Albert Einstein's frequent and important coauthors. In a continuation of Rosen's work, Leiter is author of several papers dealing with the role of gravitational potentials in General Relativity. These functions determine both the geometry of space and time and gravitational forces that affect matter.

Rudolph Schild is a Professor of Astrophysics at Harvard University's Center for Astrophysics (CFA). He is a recognized expert in the analysis of microlensing within the multiple gravitationally lensed images of distant quasars. He has found that the MECO model (above) of Robertson and Leiter provides a consistent description and explanation of his observations.