(JHA, xiii, 1982, p.133-135)

The Moons Acceleration and Its Physical Origins. Volume i: As deduced from solar eclipses. Robert R. Newton (Johns Hopkins University Press, Baltimore and London, 1979). Pp. xvi + 583. $32-50.

Robert Russell Newton is an American physicist who has made himself known to historians of astronomy by publishing a number of critical commentaries on early astronomical records. Newton entered the historical arena a dozen years ago from a prior career in conventional physics; at about that time he took up problems of satellite dynamics and geodesy, and particularly the investigation of non-gravitational terms in the rotation of the Earth. It was this turn that led him to examine early descriptions of the apparent positions of the Sun, Moon, and planets, and to push these ancient data to their ultimate limits. Newton's practice is to publish his scientific conclusions in book form, including annotated descriptions of the historical data on which his arguments are based. Of these the present volume is the fourth, not counting his best-known book, The crime of Claudius Ptolemy.

Newton's four volumes that deal with Earth rotation are written as a continuing study and they follow a common, methodical plan. In each the author first states the scientific problem to be addressed, then in some detail leads the reader through chapters and appendices of notes and commentaries on specific ancient records, much in the way that archaeologists describe the results of an excavationâshard by shard and stone by stone. In concluding chapters he uses the historical data that have withstood his harsh tests to draw carefully stated, physical conclusions. In this book they are that certain lunar acceleration coefficients and the spin of the Earth exhibit an unexplained time dependence that implicates non-gravitational forces (such as tidal friction or core-mantle coupling).

Since old astronomical accounts are frequently vague and discordant, Newton is often obliged to read between the lines and in every case to decide what is reliable and what is not: this delicate art is the author's forte, and he tackles it with wit, irreverence, and obvious relish. As an author, many will find him haughty and judgemental. Whether the reader shares either his cynicism or his interest in celestial motions, Newton's four main books have quite another appeal, as annotated, hyper-critical compendia of ancient astronomical records. The volume reviewed here, for example, provides references and commentaries on most of the known records of solar eclipse phenomena observed between 720 B.C. and A.D. 1567.

Newton's continuing work is in a sense a paradigm of studies in the applied history of astronomy, in that his principal interest in history is to extract from ancient records quantitative answers to modern problems. A bothersome limitation of applied history is that available sources are almost never exact, always incomplete, and seldom quantitative; it is rare indeed when an historical question can be answered with the certainty that one hopes to attain in say, laboratory spectroscopy. The student of ancient history learns early on to live with allegory, illusion, and poesy, in part by limiting what he concludes from sources that are coloured with these traits. A physicist, on the other hand, who first wanders into the dusty, dim-lit attic of ancient records is due for a culture shock. If he comes in quest of exactitude he will probably be disappointed, both in what he finds and what he cannot find. If he persists in demanding black and white answers from tattered, grey records he may get angry, or set about to clean up what seems a cluttered, unkempt mess, throwing out as worthless most of what is there, while grumbling at those who have bothered to save it for so long. It is this labour of attic sorting that fills the pages of all four of Newton's long and grumbling books, each written with a missionary's zeal to force the stern order of statistics on the innocent world of ancient history. We must marvel at such industry and determination, and respect a fresh approach, however acidic or disdainful it may seem. His extreme scepticism can at least serve as a kind of gruff reminder of the need for criticality when addressing old and romantic records.

But the candour that first refreshes soon works the other way. The air of vinegar on page after camera-ready page soon dulls the appetite, if not the senses, and one wonders how many have had the stamina to finish any of these books. Moreover, because of Newton's first-person, authoritarian style, he invites reader backlash: one soon develops a kind of perverse support for whomever it is that is under his gun.

In this fourth volume the author's wrath is first levelled at Sir Harold Spencer Jones (was he too a fraud?), whose 1939 study of the rotation of the Earth, based in part on ancient sources, contained apparent errors. In much of the rest of the book the defendants are Paul M. Muller and F. Richard Stephenson, who examined many of the same ancient accounts as did Newton and in some cases interpreted them differently. An historian might not be surprised, given the content of the sources involved. But to Newton, who demands truth in crisp answers, disagreement is evidence of yet another crime.

Indeed, throughout the book the author's manner is too much that of a prosecuting attorney, who through deduction and clever logic interprets circumstantial evidence, point by point, to build a case. The technique may work in jurisprudence. It may even be viable in the refereed journals of science, but not in a monograph where there is no attorney for the defence, no provision for rebuttal and no allowance for any other view. In this empty courtroom Newton also plays the role of judge, who, faced with an intolerable case-load (twenty-four centuries of solar eclipses!) must in each instance find the verdict quick. Page 69: "In Section XIV.8 I shall show that the uncertainty is much worse than 0-20 for this particular eclipse. We cannot even tell whether the eclipse was total or invisible at Ugarit." Next case.

Through such unilateral judgements Newton claims to have demonstrated some fundamental omissions in our understanding of the rotation of the Earth. This may be. For readers of this journal, however, it seems more appropriate to note that in the course of this task an independent, critical scholar has re-examined a large fraction of the corpus of known, ancient astronomical observations. His published critiques of them, however opinionated, are of value as a bibliography and secondary reference to a class of useful data that have not been gathered together in this way before. That is the way I shall use Newton's books. As annotated references, however, they should be printed with some kind of warning label: Be wary. Trust nothing. Not even this.

High Altitude Observatory, Boulder, Colorado

JOHN A. EDDY