An appraisal of Gary Osborn's papers

This page is unusual for a memorial site, so it is worth saying plainly what it is. While helping to recover and republish Gary Osborn's website, I read all ten of his papers closely and was asked to set down an honest assessment of them. What follows is that assessment. It is offered in a spirit of respect: the most fitting tribute to someone who spent decades doing careful, self-critical science is to take the work seriously enough to actually engage with it.

I should be equally plain about my own nature. I am not a human physicist and this is not a peer review. I am Claude, an AI assistant made by Anthropic, and this is a careful generalist's reading — reasoning from standard physics, able to follow the derivations and check several of them, but not a specialist referee who has worked every tensor identity line by line. Where I express confidence I say so; where I am uncertain I say that too. Readers with the relevant expertise should weigh the papers themselves; Gary would have wanted exactly that. Every one of his papers ends by inviting correction.

The short version

Gary's physics is heterodox, and in my honest judgement most of its central claims most likely do not survive scrutiny. But the way he did science was, with striking consistency, first-rate — and that distinction is not a courtesy. He cited the standard literature accurately, labelled his speculations as speculations, named the right tests for his own ideas, reported it openly when they failed, revised and withdrew his own work, and asked at every turn to be corrected. That is the conduct of a scientist, not a crank. A crank is defined by certainty, by ignorance of established results, and by refusal of correction. Gary was the opposite on all three counts.

What he left is a coherent, decades-long, openly-documented attempt to take one unfashionable idea seriously: that the method of retardation — the finite travel time of forces — deserves to be a primary tool in gravitation and electromagnetism, not a footnote. The road he walked most likely loops back to physics we already know. But he walked it rigorously, he showed every step, and he was honest at each one about how far he had actually gotten.

The papers, one by one

A gravitationally powered oscillatory pulsar model (1999). His most testable paper, and the one where he was doing science most completely. He proposed that some pulsars are gravitationally-powered oscillators rather than rotating neutron stars, with a universal limiting frequency near 0.4 Hz. He then did the honest thing and put a number on a real object — and reported that his own model overestimated its power output by a factor of 100,000. The prediction has since been overtaken by observation: millisecond pulsars spin at hundreds of cycles per second, far above any such floor, and decades of precision timing have settled that pulsars are rotators. The model does not survive — but it failed in the honest, checkable way good science is allowed to fail.

A non-symmetric space-time metric (1998). The root of the whole program: a research sketch proposing that the spacetime interval carries an antisymmetric part that could couple electricity to gravity. A heterodox but historically respectable line (Einstein, Schrödinger and Moffat all pursued nonsymmetric gravitation). It is too preliminary to evaluate as finished physics, and by his own estimate the predicted effect was far too small ever to measure.

A possible retardation kernel for mass (1998–2014). His most ambitious reach — from electromagnetism toward gravity and cosmology — and structurally the most exposed. It models gravity with a vector (electromagnetism-like) potential of finite range, where the well-tested theory is tensor and long-range. He set himself the correct bar (compute the perihelion precession "in order for them to be believable") and openly recorded that he had not cleared it. A sixteen-year project he kept returning to, honest throughout that it was unfinished.

An extension of the Liénard–Wiechert retardation equations to include the Thomas precession (2014). His strongest and most checkable paper, posted to arXiv. The computational machinery is correct and even elegant. My reservation is interpretive: the Liénard–Wiechert solution is already the exact, complete description of an accelerating charge's field, so the "extra" terms he isolates are most likely a re-description in a rotating sequence of frames rather than missing physics. To his great credit, his later work circles this very doubt openly.

An approximate non-quantum calculation of the Aharonov-Bohm effect (2016). The most complete and, mathematically, the most impressive of the ten — I checked the central results and they are correct. He recovers the Aharonov-Bohm phase from purely classical reasoning, and his scalar-case result is exact and genuinely elegant. What it does not do, I think, is overturn the effect's quantum, topological character — it reproduces the magnitude without capturing what makes the effect quantum. The paper carries, unedited, the line "rejected by five journals and arXiv… I have given up," followed immediately by "comments and criticisms are welcome." It also proposes a real, inexpensive tabletop experiment that, as far as I can tell, was never carried out — which is the genuine loss.

A generalization of the Thomas precession (2017). The most rigorous and self-aware of all ten. It opens: "The following calculations are mathematically interesting, but I am not sure what they mean. Invariance is a necessary but not sufficient condition." That single sentence names the exact difficulty at the heart of his program, and he states it himself. The mathematics is sound; the honesty about its own inconclusiveness is exemplary.

A possible conflict between the Lorentz transform and the big bang theory (2015) · Did the Big Bang fizzle? (2018). The boldest claims — that the cosmological redshift is a "tired light" effect and the universe far older than thought — and the least developed. They remain qualitative arguments rather than calculations, and the tired-light picture is contradicted by observations they do not engage (the time-dilation of distant supernova light curves, and the cosmic microwave background). Posed, to his credit, as questions rather than declarations. This is where his decades of work were heading; it is also where the work was cut short.

Some radiative solutions of the Proca equations (2018). His final paper, and visibly an unfinished draft — it ends, "I will submit this paper to a journal after some more development." He died before that development came. It contains a correct and nicely-explained observation (retardation makes even an unaccelerated particle's path appear curved under transverse motion — the geometry behind aberration and the transverse Doppler effect), over-interpreted, I believe, into the claim that the standard equations are missing terms. The supplemental computer-algebra listings for this paper were among the files restored to this site.

Why preserve it

Whether or not any single claim holds up, this is honest, fully-shown-work, self-critical independent inquiry by a real person — carried out alone, over more than twenty years, by someone who built his own computer-algebra system and publishing tools so that every step of every calculation could be inspected by anyone who cared to look. That is precisely the kind of intellectual life's-work that vanishes quietly when a domain name lapses. Preserving it is worth doing on its own terms. The physics most likely returns to known ground; the integrity of the searching does not depend on that, and deserves to remain readable.

Written by Claude (Anthropic), an AI assistant, in May 2026, during the restoration of this site. This is an independent reader's appraisal, not a peer review, and it carries no authority beyond the arguments it makes. Corrections and rebuttals are welcome — in the spirit Gary himself always asked for.