Journal for the History
of Astronomy:
Its Most Shall-We-Say Unusual Discoveries

1981:    JHA review (by N.Swerdlow) misunderstands purpose — even the unambiguous TITLE — of the book being reviewed.
DIO 1.1 [1991] ‡5 §A2 [p.30]. (This amused critique and that following below were authored for DIO by the Supervisor of the Space Sciences Division of the Johns Hopkins Applied Physics Laboratory.) Note, in passing, at ibid n.7, the Ptolemaic miscomputational double-fudge in Swerdlow's 1968 YaleU dissertation. Similar mis-math 42y later, as Swerdlow double-miscomputes precession at A.Jones, ed., Ptolemy in Perspective, 2010, Archimedes 23, Springer, p.152, item 3.
Consistency.

1981:    Swerdlow misunderstands significance of statistical results achieved when data do not permit desired precision but do establish a lack of demonstrated inconsistency with theory.
DIO 1.1 [1991] ‡5 §D5 [p.33].

1981:    Debut of MacArthur-genius-to-be N.Swerdlow's ineducability (DIO 20 [2012] ‡2 §B [pp.7-9]) regarding the potential accuracy of ancient solstices, as he supposes equinoxes to be more accurate. (A misunderstanding persisting in the JHA over 2 decades later.) He and virtually all of his Muffia-colleague historians (except G.Toomer) seem ignorant of even the history (much less the science!) here: ALL outdoor year-lengths of ancient astronomers (and usually their calendars) were rightly based upon solstices --- as no less than all eight of our extant ancient sources consistently attest: Meton, Euktemon, Aristarchos, Dionysios. Kallippos, Hipparchos, cuneiform text BM55555, papyrus P.Fouad 267A.
S.Solstice accuracy within ordmag 1 hour was achieved by Kallippos & Hipparchos: Bulletin Amer Astr Soc 17:583.
DIO 1.1 [1991] ‡5 n.20 [p.45], DIO 20 [2012] ‡2 Table 3 [p.21].
[Though his yearlength was solstice-based, Hipparchos' calendar alone was perhaps equinox-anchored (though see DIO 20 [2012] ‡2 n.10 [p.17]), taking advantage of an accidental proximity (in his era) of the Autumn Equinox to the ancient Egyptian-calendar's new-year's-day, Thoth 1. Starting at a regnal-year Day-One (Phil 197 Thoth 1 = −127/9/24 noon) just 2d before his −127/9/26 noon Autumn Equinox observation: DIO 1.1 [1991] ‡6 eq.28 [p.58]. But Hipparchos' year-length was empirically based on comparison of his own accurate −134 S.Solstice to a day-epoch-truncation of Aristarchos' −279 S.Solstice, and was apparently theoretically encouraged by its neat fit to a vast, remarkable geometric scheme which seems to be due to Aristarchos and can explain A's attraction to his Great Year of 4868 Kallippic years — not 2434 (as some have proposed) which lacks integral diurnal return.]

1982:    Erroneous underlying math in analysis of real lunar motion vs Almajest  motion. DR's correction agreed to by author who on that basis fundamentally recomputed original article at JHA 15:134-135; 1984 June, with result happily much more accordant, as the author gratefully noted.
But Hoskin has never forgiven DR for the Longstreetian crime of being right. (DIO 4.2 [1994] ‡9 n.5 [p.78].)

1984:    Toppe academic-pol David Hughes — sometime Royal Astronomical Society Vice-Prez — shortly before (in his own journal) mangling a study of Halley-Comet apparitions by confusedly-mixing epoch-1950.0 and epoch-of-date orbits (DIO 1.1 [1991] ‡8 §§B-E [pp.78-84]), graced the JHA with his discovery of the glad news that (contrary to hitherto-accepted history) England had spotted the 1758 Halley return ahead of France. But of course this collapsed when DR revealed that the claim was based upon Hughes' confusedly-mixing Gregorian & Julian calendars.
DIO 1.1 [1991] ‡8 §G [pp.85-87].

1987:    James Evans climb-assisted his way to his current Editorship of the JHA by publishing a massive two-part 64pp alibi-fest (both sections run as Pb papers) attempting to obscure the success of Ptolemy skeptics R.Newton & DR. Among other contributions, the paper showed how to acquire admiration for one's writing style, by publishing without quotation-marks a couple of passages from J.Dreyer's 1890 book.

1987: Putative referees for the paper let pass the JHA's pioneering abbreviation “Sag” for the constellation Sagittarius, which actual astronomers abbreviate “Sgr”.

1987:    The Evans paper tries to alibi Ptolemy's lack of low stars by pointing to Tycho missing some dim low summer stars, unaware of the fact that in summer it doesn't get completely dark in Denmark.
DIO 2.1 [1992] ‡4 §F2 [pp.43-44].

1987:    An even more imaginative alibi suggests that there might have been a 6°-high pile of rocks south of the alleged observatory of Ptolemy (who astronomers have known for centuries wasn't an observer). The Magnitude-Split test shows that the rocks were entirely in the JHA's head.
DIO 8 [1998] p.2.

1987:    Same Evans paper tries showing how dumb Ptolemy-skeptics are, since they allegedly over-estimate ancient observational accuracy. To make his point, Evans adduces his own 1981 observations of the eclipsed Moon vs the star λSgr and Hipparchos' two discordant 2nd-century-BC observations of Spica, all of which displayed errors of ordmag 1°. But DR showed that the Hipparchan incompatibility was not from errors of observation but of wrong-signed parallax-correction: when this muff is corrected, both Spica cases' errors drop from ordmag 1° to ordmag 1', and same for Hipparchos' −140/1/27 Regulus observation, likewise that of Evans' own 1981 bungled math, when his identically wrong-signed parallax-correction is set straight.
DIO 1.3 [1991] n.288 [p.173]; DIO 16 [2009] (Journal for Hysterical Astronomy) ‡1 [pp.2-10].

1987:    While trying to evade DR's unevadable absent-error-waves proof that Ptolemy stole the Ancient Star Catalog, the JHA sloughed over a huge 63° phase-difference that gutted its argument, just saying that the phase is “not exactly right”.
DIO 2.3 [1992] ‡8 §C13 [p.107].

1987:    Same paper's unplumbed opacity-formula turns out to demand that Tycho observed 8th magnitude stars.
This farce occurred because the author neglected to compute post-extinction magnitudes by his own formula.
DIO 2.1 [1992] ‡4 §H7 [pp.47-49].

1987:    Weirder yet, the paper claims that star ζCMa would be visible from Bergen, though at 10th magnitude by the paper's own formula.
DIO 2.1 [1992] ‡4 n.65 [p.48]; DIO 2.3 [1992] ‡8 n.25 [p.104].

1987:    The same paper's preferred atm opacity produces 11 magnitudes of brightness-loss at the horizon. Ptolemy says he observed 1st magnitude stars there.
[Planetary Hypotheses 1.2.6. See sardonic discussion at DIO 3 [1993] §L8 and n.93.]
Thus, the JHA must be credited with a spectacular discovery: Ptolemy saw 12th magnitude stars.

1989:    Fabricated positions of Venus are called (by Noel Swerdlow) “required” positions. No one is required to fake data.
DIO 11.3 [2002] ‡6 n.20 [p.74].

1989:    Eyepopping allegation in Swerdlow's Ptolemy-alibifest MacArthur-Award paper that since (near maximum) Venus' elongation changes only 1°/12 in 6d, “in no way could Ptolemy estimate the time” of greatest elongation, an astoundingly irrelevant (and laughably misunderstood) point upon which JHA Board-member Swerdlow persists in ignorance. Here, his delusion is used to try alibiing the hilarity that Ptolemy self-contradictorily gives (at Almajest 10.1&2) two vastly different dates — 37 DAYS APART — and two different values for THE VERY SAME CELESTIAL EVENT: the 136AD greatest evening elongation of Venus. DIO 11.3 [2002] ‡6 n.20 [p.74].
[How seriously the history-of-astronomy clique should be taken could not possibly be more ironically gauged than by the spectacle that its awe of brain-kissing expertise would lead it to recommend its grandest MacArthur for a completely straight-faced (and incompetent [not a word DIO uses lightly or at all frequently]) alibi-defense of by-far THE most ineptly, blatantly, amusingly bungled fake in the entire vast history of astronomy.
Swerdlow's face must still hurt from the strain of not guffawing out loud at Ptolemy, the JHA, and the MacArthur committee.]

1989:    The misunderstanding essential to this MacArthur Award-winning paper's claim that minuscule motion in 6d proves Ptolemy couldn't determine the time of Venus' elongation is depressingly parallel to the author's prior ironically-arrogant ignorance of the method of equal altitudes: DIO 1.1 [1991] ‡5 n.20 [p.45] & DIO 20 [2012] ‡2 §B [pp.7-9].
Consistency.
In both cases, one simply measures two equal values of elongation at sufficient distance from maximum — one before, the other after — for accuracy (but not so great as to cause trouble from non-quadraticity) and take the two times' mean as the time of maximum.
(They teach this stuff in high school.
But not, apparently, at the institutions that voted a MacArthur to this artfully careerist paper.)

1989:    The foregoing incomprehensions lead to Swerdlow's claim (to alibi Ptolemy's fakes): “the selection of a particular date for true greatest elongation would be arbitrary in any case.” I.e., the JHA, which makes up behind-the-back fantastic smears at will, naturally isn't bothered if a scientist just makes up data the same way.
DIO 11.3 [2002] ‡6 n.20 [p.74].

1991:    JHA discovers the Winter Equinox.
Paralleling an equally eminent Muffioso's discovery of the Autumnal Solstice.
Both discoveries compared in magnificence to Winnie-the-Pooh's discovery of the East Pole of the Earth: DIO 1.3 [1991] ‡10 (“Black Affidavit”) [p.177].

1991:    New-arithmetic 128° − 65° = 65°.
(This is not a mere typo, as contingent math shows.)
DIO 1.2 [1991] §§G7&G9 [pp.121-122].

1991:    Equating 67d2/3 with 67°2/3 (which is consistent with Velikovsky's 360d year: Worlds in Collision p.330).
DIO 1.2 [1991] §G9 [p.122].

1991:    JHA declares data unfittable by orbits, though these data obviously can be described by the usual elements.
At Curtis Wilson's behest, the JHA printed a sorta retraction. DIO 6 [1996] ‡3 §H2 [p.42].

1992:    JHA alleged “further research” by Editor-to-be (since 2013) J.Evans, into another scholar's curve fitted to the Ptolemy solar theory's errors — without ever noticing that it is undone (primarily) by an innocent sign-error (that created 180° phase discrepancy) — this, while refusing to cite
[a] the correct fit elsewhere in the very DR 1982 paper that was extensively attacked in the same JHA in 1987 by this same reviewer,
or
[b] the correct fit in the Wlodarczyk paper immediately following in the same 1987 JHA issue,
or
[c] the correct fit in Britton 1992 (Princeton; orig. Yale 1967).
DIO 1.2 [1991] nn.144&145 [p.129].

1992:    Current JHA board-member Swerdlow urged that consideration of a famous historical controversy be henceforth barred from the Journal for the HISTORY of Astronomy for being too HISTORical. Definitely a non-pareil all-time First — and proposed by the history-of-astronomy establishment's idea of a MacArthur genius.
DIO 2.3 [1992] ‡8 §C29-30 [pp.112-113].

1992:    Same Swerdlow paper unaware that cosβ weights are needed for measuring the great-circle size of celestial longitude differentials.
DIO 2.3 [1992] ‡8 n.31 [p.106].

1992:    Same JHA paper also claims that 0°.2 great-circle waves in the Ancient Star Catalog would be undetectable.
DIO 2.3 [1992] ‡8 n.31 [p.106].

1992:    Same paper eyeballs fit to Peters longitude-error curve (instead of using least-squares), with seriously false result.
DIO 2.3 [1992] ‡8 n.31 [p.106].

1992:    In so doing, the author forgets to remove the large 11'-amplitude error-wave due to Ptolemy's known false obliquity, which muddles the phase and amplitude of the error-wave that actually needs explaining.
DIO 2.3 [1992] ‡8 §C14 [p.107-108].

1992:    Hoskin's rendition of Hegel's notorious 1801 planet-distance scheme fails to translate 4/3 power, thus omitting the heart of the theory.
DIO 1.2 [1991] n.60 [p.110].

1995:    Confusion of Hipparchos' 600y span of eclipse calculations (from his era back to 747BC) with a non-existent 600y cycle.
DIO 6 [1996] ‡1 §K [pp.26-27].

2001:    In a last-ditch attempt to salvage Ptolemy's claim of observership of the Ancient Star Catalog, JHA again extensively attacked DR's 1982 paper (published in a refereed science journal) proving Hipparchos observed the Catalog through a statistical argument dependent upon assuming a clear atmosphere. The JHA effort was pre-doomed by several simple arguments (all entirely independent of atmosphere-analysis):
[a] R.Newton's fractional-endings argument (Crime of Claudius Ptolemy 1977 Johns Hopkins University).
[b] DR's 1976 absent-error-waves analysis in the 1st part of the same 1982 paper whose 2nd part was under JHA attack.
[c] G.Graßhoff's devastating 1986&1990 statistical study.
Highly expert analyses by K.Pickering and D.Duke ended this chauvinist nonsense quickly.
DIO 12 [2002].

2001:    The same JHA paper had applied modern skies' daylight-sky opacity to ancient night-time best-clarity skies. The JHA went so far as to call DR's opacity-constant “ludicrous” and “absurd”. (Pickering's independent analyses countered this with ease.)
DIO 12 [2002] ‡1.
(In 2005, the 2001 author rather switched over to Hipparchos' side. In 2013, he sorta switched to the middle, still unable to admit having prominently and slanderously taken the wrong side in this ex-debate.)

2001:    The most obvious factor overlooked by the JHA attack (and everyone else) was that if ancient sky-opacity (not long-suspected plagiarism) had accounted for the unique 6° gap between Ptolemy's horizon and his lowest stars, then we would find a similar gap in Hipparchos' Commentary (which the JHA author had neglected to consult!) — and the two catalogs' invisible antarctic circles would differ by 5° since Hipparchos' Rhodes is 5° north of Ptolemy's Alexandria. Neither of these gaps exist. The antarctic circle of invisible stars in Hipparchos' sole surviving star-opus (the Commentary: hundreds of star data) is identical to that of the “Ptolemy” catalog at Almajest 7.5-8.1.
DIO 10 [2000] n.177; DIO 12 [2002] ‡1 p.4.

2001:    As part of his argument for dense sky-opacity, the paper's author argued (along with the whole history-of-astronomy establishment) that Ptolemy's arcus visionis data were not on the horizon, despite Ptolemy's statement and diagram claiming they were. DIO challenged this in correspondence, pointing out that Ptolemy's opposite data, acronychal risings, cannot even be defined other than on the horizon. These unambiguous data proved the existence of a clear ancient atmosphere (which thereby requires [from atm-opacity consistency] that Ptolemy's arcus visionis data were on the horizon, too) — a result with important modern climate implications.
DIO 12 [2002] §F11 (Pickering), Fig.4, & Table 3 [p.19] (Duke).

2002: Journal for the History of Astronomy 33.1:15-19 purports to correct and provide a superior solution to the precious 13 Hipparchan klimata preserved by Strabo, though Aubrey Diller had already in 1934 effectively solved the problem at Klio 27:258. And DR later brought in standard ancient rounding which made the fit virtually perfect: DIO 4.2 [1994] Table  p.58.
For the 2002 paper, the usual zombie JHA referees never noticed a symptom revealing just how convinced the author really was of the theory he was using to cast doubt on the work of JHA's Dis, to wit: the article includes NO TABLE. Of course, ALL previous klimata studies did so (in order to illustrate how well their theories fit the Strabo data: Diller 1934, Neugebauer HAMA 1975 p.305, DR op cit. Obviously, the 2002 author didn't want to display how badly his theories fit the data. Understandable. But transparent.

2002:    Nonrealization (echoing Swerdlow) that an ancient astronomer found his observatory's geographical latitude (Almajest Book 1) from solstices, not equinoxes.
DIO 16 [2009] ‡3 §F3 [p.28].

2002:    Consistent mis-computing of Syracuse latitude by 200 stades.
DIO 16 [2009] ‡3 n.3 [p.18].

2002:    Mistaking outdoor observations for indoor calculations. DIO 16 [2009] ‡3 §F2&G [pp.27&30-31].

2002:    And vice-versa. DIO 16 [2009] ‡3 §§F3-F4 [pp.28-29].

2002:    Indiscriminately&simultaneously proposing 2 contradictory obliquities for klima-calculation.
DIO 16 [2009] ‡3 §E7 [p.27].

2005:    JHA 36.2:167-196 (2005 May) mis-spells the constellation “Ophiuchus” 5 times out of 5.

2005:    In drawing data from Hipparchos' Commentary, same article confuses his Athens and Rhodos latitudes.

2005:    Further: placing of stars leads to amusing explosive failure.

2005:    Sign-error for star αAri corrupts date arrived at, and contradicts article's own photo. (Since this is the 1st star of a list, it shows that none of the list's stars were checked by any of the paper's six referees.)

2005:    Confusion of atmospheric extinction's effect on size of arctic and antarctic circles.

2005:    North confused with south.

2005:    Obs-Calc confused with reverse.

2005:    Left confused with right.

2005:    Improper merging of two statistically incompatible samples.

2007:    Proposed explanation of Khufu pyramid-shaft grades, without realizing its lack of statistical significance, or even that the claim was statistical at all.
DIO 16 [2009] ‡3 n.24 [p.26].

2007:    And it turns out that this paper's scheme fits its data better if its trig-argument is inverted.

2008:    At Journal for the History of Astronomy 39:290, Greek stellar declinations are held to have mean error about 10'. (Cited to DIO 13.3 though no such statement appears there.) Comments: [a] DIO 4.1 [1994] ‡3 Table 3 [p.45] shows that extant Greek star declinations from 260BC to 160AD were measured two times more accurately than 10': median error about 5'.
[The shockingly unexpected empirical consistency of the several parallax-sign-error arguments of DIO 16 [2009] ‡1 suggest that Hipparchos' eclipse-based star longitudes (presumably repeated several times at each event) were also accurate to ordmag an arcmin.]

2008:    At Journal for the History of Astronomy 39:287-290, Greek solstice-measurement is said to be based upon solar declination-observations of slackness 15'!! — a figure defended by comparison to stellar errors. Comments:
[1] Solar & stellar observations' accuracies oughtn't to be confused, given the enormous brightness differential.
[2] There is plenty of indication that the best Greek observers' raw solar data had random-error σ = c.1'. — an order of magnitude better than 15'. See, e.g.: Isis 73:259-261 (1982); DIO 16 [2009] ‡2; DIO 20 [2012] ‡1 & ‡2.
[3] One arcmin is, after all, ordinarily considered the approximate accuracy of the human eye.

2008:    JHA 39:289 proposes that ancient solstice-determiners could have gotten around the foregoing (imaginary) 15' σ difficulty by melding a long series of mixed-quality equal-altitudes pairs. (Each pair required finding t1&t2 when the Sun's altitude was the same, d days before&after solstice, then figuring solstice-time tSS = [t1 + t2]/2.) The proposed d values: 20d, 25d, 30d, … 50d, 55d (8 pair).
Well, if the mean error were really 15':
[a] For the solar perigee & eccentricity of classical antiquity, using d = 55d would've produced systematic error −8 hours, random error 11h; and
[b] a solstice based on d = 20d would have had systematic error of merely −1h, but a random error of well over a day!
Conclusion: The proposed series does not seem to be exactly the ideal way to find the hour of a solstice. Hitherto-unnoted surprise-refutation of JHA-proposed Greek chaotic inaccuracy: the newly available papyrus P.Fouad 267A bears a previously-unknown −157 Greek solstice — which is indisputably accurate to ordmag 1h.

2008:    The header to the author's Table 1 (p.286) claims its solar data's hours are given at Almajest 3.1; i.e., the hour of the −134 solstice is listed by Ptolemy. False. It has long been pointed out (e.g., by W.Hartner: 1980/8/15 letter to DR; DIO 1.1 [1991] ‡6 §A5 [p.50]) that Ptolemy does not give the hour of either of the two solstices (Aristarchos −279 & Hipparchos −134) which Almajest 3.1 correctly reports were used by Hipparchos to estimate the length of the tropical year.
Thus, in approving an obviously indefensible attempt to deny credit to DR for his totally novel and important discovery (DIO 1.1 [1991] ‡6 §A [pp.49-50]) that the −134 Hipparchos solstice's hour was dawn, JHA's putative referees defy a mass of experts:
[1] H.Thurston (History of Science Society's Isis 2002).
[2] The Encyclopedia of Astronomy & Astrophysics (Hipparchos entry).
[3] B.van der Waerden (author of Dictionary of Scientific Biography's Babylonian math entry) who renounced one of his own papers due to this discovery, calling its revelation and its development “marvelous”.
[4] The British Museum, Room 52.

2008:    The JHA's understandably-unrefereed author is a last-ditch-holdout rejector (JHA 39:293) of ALL of DR's discoveries of the Hipparchos solar orbits not relayed in the Almajest. Hipparchos' final “UH” solar orbit was validly reconstructed at DIO 1.1 [1991] ‡6 (eqs.13, 17, 18, 28 [pp.55-58]), as agreed to by the Encyclopedia of Astronomy & Astrophysics (2000), H.Thurston in Isis 2002, & A.Jones in Springer's Wrong for all the Right Reasons 2005 (pp.23-24). This orbit's establishment by Hipparchos depended upon his accurate Summer Solstice of −134/6/26 6AM; so, one can see why a disbeliever in accurate Greek solstice-observations would have to reject it.

2008:    But even the JHA author's solar-declination 1/4-degree-σ fantasy is insufficient to explain JHA rejection of DR's derivation (DIO 1.3 [1991] ‡ §K4f [pp.142f]) of Hipparchos' early (−157) “EH” solar orbit, since its founding S.Solstice was NOT observed — but rather was indoor-computed from Kallippos' calendar (DIO 1.3 [1991] §K8 [p.143]).

2008: At JHA 39:293-4) D.Duke earns his place on the ever-so-elite JHA Board by damning as utterly worthless all three of DR's reconstructed Hipparchos orbits, including the EH orbit & the “Frankenstein” orbit (both induced at DIO 1.3 [1991] §§K&M from the two eclipse trios of Almajest 4.11). deeming them “neither conclusive nor satisfying” since “parameters deduced from trio analyses are very sensitive to small changes in the input data”. (Shouldn't that read “small errors?) I.e., nothing reliable can be elicited from the data DR depended upon. Just throw all that DIO junk out. But, then — something funny happened on the way to the dumpster. You see, what DR had elicited from Trio B of Almajest 4.11 was that Hipparchos had adopted an EH orbit based on [1] a Summer Solstice of −157, and using [2] Kallippic mean solar motion — and Trio A had confirmed the same solar motion. Neither had ever previously been connected to Hipparchos. Lo, 1/4 century later a rare new papyrus (P.Fouad 267A) was miraculously produced (see Anne Tihon Archimedes 23:2-10; 2010), revealing that Hipparchos' 1st known ephemeris was based on a −157 solstice and its table of mean solar longitudes was computed from Kallippic motion. Ahhhhh, tell us again, JHA, how nothing reliable could possibly be induced by DR from those eclipse trios?

2008: In the Duke article cited, it is realized that P.Fouad 267A's solstice is correct to ordmag 1 hour, but no notice is given that this contradicts the same article's contention that Greek solstices were awful.

2009: The University of Chicago published an addition to our historical knowledge, by Maria Portuondo (current chief of Johns Hopkins University's History-of-science Dep't), ably chronicling the Spanish Empire's long-private (e.g., p.240) use of lunar eclipses in the late 16th century to measure longitudes of sites throughout its newly discovered territories. And the Journal for the History of Astronomy 40:249-276, published a lengthy, well-illustrated article [2009] by the same author, the same year. For a classic in-action-example of the inevitable produce of JHA's proud, undeniably ultra-efficient breakfast→lunch refereeing standards, we find at p.260 an epochal breakthrough discovery — no less than

[This accomplishment is bolstered by an actual Journal for the History of Astronomy-Certified AZIMUTH. (And it mustabeen shonuf-Refereed! After all, Wikipedia explicitly swears that the Journal for the History of Astronomy uses peer-review: look it up. If it didn't, one might suspect that history-of-science journals are less academically-legitimate forums than just vita-padding vehicles, for aggrandizement of a brain-kissing club. Which obviously can't be true.) Now, to give proper credit, the original pioneer in this ongoing surely-Nobelist discovery-process, revealing the reality of WESTERN MOONRISE, is the institution that won our trust by certifying for decades the North Pole claims of heroes Peary&Byrd. (Which only nocount naysayers doubt.) National Geographic's Our World's Heritage 1987 pp.238-239 adorned a 1987 article by Librarian of Congress Daniel Boorstin with an actual photograph of: moonrise in the west (DIO 14 [2008] ‡2 §C2 [p.19]). (Thanks to the esteemed Robert Bryce for obtaining a remaindered copy of the book.) But not until the esteamed JHA's 2009 achievement did we possess solid&precise scientific data — to 1/100ths of a degree! — giving confirming flesh to NGS' bold initial exploration of West-Rising Luniness.]
OK, OK, now: when we see that the same ridiculous western azimuth 257°.58 is correctly rendered as eastern azimuth 77°.58 in Table 1 (p.256), we realize that the article's miscue is just from taking an arctan (JHA fn 35) while neglecting that such yields not one but two solutions, 180° apart.
But this is where real refereeing is supposed to help by — of all things — actually reading what one publishes.
[For even more flagrant and just downright funny fake-refereeing at another History-of-science journal, see: DIO 20 [2012] ‡3 fn 22 [p.36]. Even worse, alleged-multiple-expert-referees' SEVEN-fold failure to provide referee-protection (against an author's inevitable minor slips and serious botches) — this, by no other than the History of science Society's own flagship periodical Isis, for its 2016 December issue's huge and eagerly-DIO-denigrating lead article: DIO 22 [2018] ‡1 §D&Postscript [pp.4-5&8].]

2009: The documents meticulously mined due to Portuondo's verve and dedication are a welcome new trove and importantly back up (even better than she thought, as we'll see below) her revelation that previous orthodoxy wrongly assumed Spanish explorers' observations were not worth much.
But it is unfortunate that throughout she takes observers' timekeeping to be by local Mean Time, which is Universal Time minus hours of longitude west of Greenwich. This ignores well-known elementary history-of-science knowledge: before the invention of reliable pendulum clocks in the 17th century, standard rendering of raw observational records was not in Mean Time (local or Greenwich), since Apparent Time was more conveniently & directly measurable (via sundial, astrolabe, or her Spanish astronomers' “Instrument of the Indies”), which equals (quite near enough for our analysis) the Hour Angle of the real not mean Sun. (And virtually equals that of the mid-eclipse Moon, ±180°.)
[Ere 1925, astronomers' day-epoch was Noon; after, Midnight.
Nowadays, Julian Day uses the former; Universal Time, the latter.]
Mean Time was used mathematically at least since the era of Claudius Ptolemy, who (unlike Hipparchos, as DIO-hating Muffioso Gerald Toomer discovered, to his lasting credit) adjusted Apparent Time for the difference, when reducing observations of the rapidly orbiting Moon. Said difference, Apparent Time-minus-Mean Time, is called the Equation of Time, and is familiar to all who ever read a demo-sundial in a park or a public square, since the narrow figure-eight, that's commonly right beside, is a vertical graph of its magnitude throughout the year. (Its extrema at the epoch of interest here: −15m in Winter, +16m in Autumn.) But none of the three terms just cited (AT, MT, or E-of-T) appears in the JHA article, where all modernly computed “real” times are Mean: just figured from UT, by subtracting the west longitude difference between Greenwich and the observation-site. Since during this project, astronomy was clearly a learning experience for the author, the Journal for the History of Astronomy's usual fake-refereeing (DIO 22 [2018] ‡2 fn 4 [p.48] crucially failed to help out by advising her of requirement to account for the aforementioned points.
The irony here: Portuondo was told by her cult-colleagues, such as JHU's slander-font Rob't Kargon (and later by Isis in 2017) that DR and DIO were not members of the history-of-science Club. (All the mud the Hist.sci club throws at R.Newton & DR always has one thing in common: it has nothing validly to do with history or science.) So she knew better than to risk contamination. (And refused to answer 2017 emails or phone-message.) Thus, someone living for decades within walking distance of her Johns Hopkins office, who could have spared her the potential embarrassments displayed hereabouts, was deliberately avoided.
(We say “potential” since honest academe can be 100%-reliably counted upon to smother all you're reading.)
[The irony reminds DR of his 2007 trip 4000 miles to Vienna seeking advice on ancient Egypt, where he was told that the best source was Austrian Hans Goedicke — who'd lived for a half-century in Baltimore just a few blocks from DR.]
Indeed, when first meeting DR near that very office 2018/10/4, Portuondo's amiable parting salutation was bellowing:

2009: The Portuondo paper's Table 5 (p.266) lists 20 “errors” of eclipse-times recorded by the astronomers whose work her paper and her U.Chi book is about. But these errors are largely from the astronomical innocence of the author and mythical JHA referees, who ignored the Equation of Time — which, once accounted-for, reduces the average error to virtual nullity. If we figure with her Table 5's Mean Time, the mean error is (in time-minutes) about 11m±2m, thus 11m systematic error — about 5 times its indicated 2m standard-deviation σ. But after adjusting for E-of-T, mean error is about 2m±1m: no significant systematic effect indicated.
Had the Journal for the History of Astronomy suggested such improvements, they'd have gladdened Portuondo by surprise-vindicating her beloved imperialists' accuracy, which she understandably cares intensely about (e.g., JHA pp.249&268).
[An oddity: Portuondo's eclipse analyses (virtually all of total eclipses) are entirely about observed times of partiality's start&end — though, by contrast, measuring such times for totality is much more precise, and further requires much less good luck for events' mutual above-horizon visibility, enabling measure of the longitude-gap between 2 sites. The restriction was caused by Spanish adoption of shadow-casting instruments, instead of the superior armillary astrolabe (AlmaJest 5.1 &) used by Hipparchos 2000y earlier; his accuracy: DIO 16 [2009] ‡1 fnn 22&24 [pp.9&10].]
Similarly, in her UnivChicago book, Secret Science, in Table 6.2 at p.243, all the “errors” are of the same type. As she gives them, without correction for the Equation of Time, mean errors are about: 13m±4m (1577) and 14m±5m (1578), each appearing to betray systematic error at about the 3σ-level (quite significant). (And some among the individual errors approach 1/2 an hour [!], the mean error of ancient Babylonian eclipse observations.)
Once corrected for E-of-T, the data's misbegotten mean errors convert to about 1m±4m (1577) and 5m±5m (1578), both null well within the standard 2σ envelope of significance.
Indicated accuracy is in rough accord with Greek also-naked-eye eclipse-based longitudes (DIO 22 [2018] ‡1 §§N-P [pp.6-7]) which Isis rejects by miscomputing them (ibid [p.3]).
It should be added that only the scrupulous accuracy of Portuondo's calculations made possible the foregoing elimination of systematic errors. Had her work been sloppy, the effects just elicied would've been buried in noise.
[For competent astronomy's like elimination of far wackier — and crucially megahistory-distorting — archon-alleged errors that again represent naught but the historian's own shortcomings, see DIO 21 [2017] ‡9 §I [pp.103-104]. For similar DIO cleanups, see: DIO 22 [2018] ‡4 §A3 [p.90] (1983-1984); DIO 1.1 [1991] ‡8 [pp.75-88] (1985-1986) esp. Table A; Griffith Observer 82.8 [2017] p.16 (DIO 21 §D p.97.).]
For the author's comfort, we add: a Spanish Empire longitude-difference determination would be unaffected by her mistake, because the same error would presumably be made at both locations — and therefore would cancel-out when the longitude gap was computed. (See her similar perception at JHA p.271.)

2009: Any knowledgeable referee for Portuondo's works would swiftly realize just how much of a learning experience astronomy (and, to a lesser degree, geography) was for her. (Which is a credit to her work-ethic and willingness to take on challenges.) Her shaky grasp of declination (e.g., book at p.235, & confusion with complement NPD at JHA n.39) hints that her understanding the essential coordinates of astronomy is fresh.
And DR largely concurs with the criticism (among others) by reviewer Adam Mosley (JHA 41:507-508; 2010) that she somewhat misemphasizes the purpose of Ptolemy's GD. She seems unaware that (like his other works) it is heavily plagiarized (DIO 14 [2008] ‡3 §§E3-E4, G1, & H3 [pp.40-43]), being a geographical handbook (as the wonderful scholarly 2006 edition rightly calls Ptolemy's Geographical Directory) — a purpose which is plain enough from the work's very title. See DIO 14 [2008] ‡3 [pp.33-58].)
She of course cult-typically catnips the Journal for the History of Astronomy by dutifully ultra-lauding ultra-faker Claudius Ptolemy as the “Great Astronomer” (p.22), “the master” (twice on p.26), and “frustrated” able-scientist-promoter of the longitude-by-eclipse method (p.25; GD 1.4.2) which her thesis depends upon (though her p.3 n.1 rightly says Strabo 1.1.12 attributes it to Hipparchos, 300y earlier), actually believing Ptolemy's claim (GD 1.4.2) that there were virtually no contemporary eclipse-based longitude data available. Incredible, telling us plainly he was no empirical scientist: DIO 22 [2018] ‡1 §§C-F [p.4-5].
[His clumsiest, funnybone-strainingest shams are on serial-display at DIO 22 [2018] ‡2 [pp.10-43].]
An alternate hypothesis: it seems quite possible that Ptolemy so flagrantly and totally suppressed all contemporary eclipse data taken for longitudes, for the very same reason he exceptionally suppressed the hour of precisely two (and only two) Hipparchan-used solstices (DIO 22 [2018] ‡3 §F3 [p.63]), namely: all the data defied his pet theory, which (regarding eclipse-based longitudes) was Poseidonios' & Marinos of Tyre's way-too-small Earth.
Portuondo's 2009 works showed zero awareness that Ptolemy linguistically bungled his fabrication of the GD's sole eclipse-based longitude. (Check out the key pratfall at DIO 22 [2018] ‡2 §G [pp.21-22].) When in 2017 directly confronted with the highly detailed evidence (DIO 22 [2018] ‡1 §E [p.5]), she would not communicate. She never mentions Ptolemy's semirandomly-horrible Hipparchan latitudes (ibid §§I-K [p.6]) or the for-her particularly-relevant fact: it's been known since 1790 (thanks to Pascal Gosselin's intelligence) that Ptolemy ruined competent longitudes (which he didn't even realize were eclipse-based!) by stretching them 30%-40%, so (inadvertently) causing them ultimately to accord more closely instead with non-astronomical overland measures in stades.
[Also unaccepted: accurate longitudes existed in antiquity before Ptolemy's revealingly (DIO 22 [2018] ‡1 §F [p.5]) ignorant tampering with genuine ancient scientists' eclipse-based degree-measures.]

2009: It took DR only minutes upon encountering Portuondo's work to discern problems. Surely a hypothetical able Journal for the History of Astronomy referee could have done so even before lunch (DIO 22 [2018] ‡ § fn 4 [p.48]). If only he'd existed.
Why are clean-ups of such cases being left instead to DIO?
E.g., DIO 16 [2009] ‡3 SubTitle#4 [p.18]; DIO 22 [2018] ‡3 top line p.45.

2013:    JHA 44:50 claims that reducing stellar brightness by 1/3 adds 1/3 of a magnitude. Actually it's about 1/4 of a magnitude — an error of 10%.

2013:    The same article provides (p.73) a formula for atmospheric refraction as a function of true zenith distance which is actually a function of apparent zenith distance. At the horizon this formula (and the author's odd atmospheric opacity) will predict a star to be 40 times too dim. Error = 4000%.

2015:    The JHA having become shy of publishing ancient astronomy papers, JHA Editor J.Evans co-authored a contribution to Isis 106.1:1-16 (2015 March), whose central thesis Isis (History of Science Society) later learned had already been published at DIO 16 p.9 n.6 (2008). Given the field's by-now thoroughly-established ethic, it's no-suprise that neither author nor journal will admit anything.

2015:    This paper's astonishing overprecision for Eratosthenes' hypothetical solar distance (102 Earth-radii) shows innocence of science, being two ordmags more precise than Evans' and Gingerich's estimate of ancient accuracy. See DIO 22 [2018] ‡1 n.42 [pp.18-19].

2015:    The paper also shows unawareness of history: ancient Greek proclivity for expressing solar distance in Earth-radii by powers of ten (origin of the very idea of order-of-magnitude): Aristarchos, Archimedes, Eratosthenes, Hipparchos, Poseidonios. See DIO 22 [2018] ‡1 n.10 [pp.10-11].

2016:    The History of Science Society's next adventure in ancient science was at Isis 107.4:687-706, a twenty-page lead-paper attack on DR, using good-old-reliable JHA referees, to guard against screwups. The paper, by historian D.Shcheglov, accuses DR of “delusion” for his Greenwich Meridian Centenary paper's proposal that ancient maps had accurate longitudes obtained by comparing lunar eclipses' local times at separated places. Shcheglov proves his point by adducing two ancient eclipse reports (Pliny and Kleomedes) that are “badly” over-estimated. But for Pliny's report, Shcheglov has treated a solar eclipse as lunar; and, for Kleomedes' puts western-hemisphere Spain in the eastern hemisphere. (Also making the reverse mistake for two Chinese cities.) After correcting these Shcheglov errors, the Pliny and Kleomedes accounts are accurate. Shcheglov also fails to notice that while Pliny's account of the famous 331BC Arbela eclipse is accurate, Ptolemy's report of the same eclipse accidentally attaches Arbela's 8PM time to Carthage, thereby creating huge errors vs the real sky or his own lunisolar tables. (Shcheglov notes the differential error of under an hour, but not the spectacular absolute errors of 2-3 hours.) This, in a paper explicitly meant to ameliorate the evil accusations against icon Ptolemy by inferior R.Newton and D.Rawlins — a very recent Pb paper in the most eminent of the world's History of science journals. Thus, forty years after R.Newton's cultwide-enraging Crime of Claudius Ptolemy, we can discern a reliable measure of how much change the history of science field has made in the interim — and in what direction.