Kepler as Historian of Science: Precursors of Copernican Heliocentrism according to "De

Revolutionibus," I, 10
Author(s): Bruce Stansfield Eastwood
Source: Proceedings of the American Philosophical Society, Vol. 126, No. 5 (Oct. 21, 1982), pp.
367-394
Published by: American Philosophical Society
Stable URL: http://www.jstor.org/stable/986213
Accessed: 19-03-2015 17:20 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact [email protected].

American Philosophical Society is collaborating with JSTOR to digitize, preserve and extend access to Proceedings of the
American Philosophical Society.

http://www.jstor.org

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLER AS HISTORIAN OF SCIENCE:
PRECURSORS OF COPERNICAN HELIOCENTRISM ACCORDING
TO DE REVOLUTIONIBUS, I, 10.*

BRUCESTANSFIELDEASTWOOD

Department of History, University of Kentucky

I. PROLOGUE against him as well. Unfortunately, Kepler

had written in 1595 a laudatory letter to
Johannes Kepler (1571-1630), while con-
Ursus, whom he called at that time the lead-
cerned with Copernican astronomy
from early in his career,1 developed his ing mathematician of the age, and Ursus
had quietly included this letter of praise in
sense of the distinctive nature of the Coper- his own published polemic against Tycho.
nican and later astronomical hypotheses To still Tycho's suspicions Kepler agreed to
only gradually.2 One stage in the develop- put together a rejoinder to Ursus, an "Apo-
ment appeared in a work undertaken reluc-
logia Tychonis contra Ursum," in which the
tantly, which seems to have sharpened Ke- nature of astronomical hypothesis is ex-
pler's views about the issues involved in the amined and the claims of Ursus are shown
making of astronomical hypotheses. At the to be incorrect.3 This tract "Contra Ursum"
bidding of Tycho Brahe, Kepler agreed to was written in late 1600-1601 but remained
compose a riposte against Nicolaus Ray- unfinished at Tycho's death (24 October
marus Ursus, whom Tycho had accused in
1601; Ursus had died in August 1600) and
1596 of plagiarism. In 1597 Ursus claimed
unpublished until the first modern edition
in print that the Tychonic system was not of Kepler's works.4 Although Kepler's ap-
Tycho's and attacked Tycho viciously. Ty- proach to the issues in "Contra Ursum" was
cho responded with the charge that Ursus rather factual and straightforward, he surely
was a plagiarist and brought a court action had little patience with the claims of Ursus.
When writing to Michael Maistlin (29 Au-
* The research for this
study was conducted during gust 1599), Kepler spoke of the absurdity
a year of support as NEH Fellow and member of the in Ursus' claim that Tycho's way to his hy-
Institute for Advanced Study (Princeton), 1979-80. I
am extremely grateful for this support. An early draft pothesis was paved by Copernicus' De revo-
of this study was read and criticized by Professors lutionibus V, 35 (heliostatic explanation of
Owen Gingerich and G. J. Toomer. While I benefited
greatly from their criticisms, neither critic should in any
way be held responsible for the form or the content
presented here. 3 Max
Caspar, Johannes Kepler (Stuttgart: Kohlham-
1 Owen Gingerich, "Kepler, Johannes," Dictionary of mer, 1948), pp. 97-8, 123-4, 136, provides background
Scientific Biography, edd. C. C. Gillispie et al. (New to the writing of Kepler's "Contra Ursum." See also the
York: Scribners, 1973), 7: 289-312, provides a useful brief review of the facts in Nicholas Jardine, "The Forg-
introduction. ing of Modem Realism: Clavius and Kepler Against the
2 Robert S. Westman, "Kepler's Theory of Hypoth-
Sceptics," Studies in History and Philosophy of Science,
esis and the 'Realist Dilemma'," Studies in History and 10 (1979): 143.
Philosophy of Science, 3 (1972-1973): 233-264, surveys 4 Johannes
Kepler, Opera omnia, 8 vols., ed. Christian
the various elements and the stages in Kepler's thought Frisch (Frankfurt a.M.: Heyder and Zimmer, 1858-
about astronomical hypothesis. 1871), "Contra Ursum" in 1: 215-287.
PROCEEDINGS OF THE AMERICAN PHILOSOPHICAL SOCIETY, VOL. 126, NO. 5, 1982

367

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
368 BRUCE STANSFIELD EASTWOOD

the stations and retrogradations of Mercury claimed as a source for Tycho's hypotheses,
and Venus); and when Ursus was so ridic- but Ursus had not even paid attention to the
ulous as to find Tycho's hypotheses in De simplest of writers in which heliocentric
revolutionibus III, 25 (computing the appar- ideas could be discovered.
ent position of the sun), Kepler wrote of The core of Kepler's reference in this re-
"asinus ille, non jam ursus." In short, said view of the past was Copernicus' De revo-
Kepler, Ursus cannot have read Ptolemy nor lutionibus I, 10, in which we read, after a
have understood Copernicus.5 On the other rehearsal of the evidence from the orbital
hand, after having thought out and written periods of the planets in favor of a reasoned
on the matter (and after the deaths of the heliocentric order, the following suggestion.
two antagonists), Kepler could show himself
ConsequentlyI think we should certainly not
in a more moderate mood. In a letter to
despise the argumentwhich was well known to
David Fabricius (2 December 1602), Kepler MartianusCapella,who wrote the Encyclopedia,
noted that he wished to study Proclus and and certainother Latinwriters.For they believe
Averroes on the question of hypothesis be- that Venus and Mercuryrevolve round the Sun
fore publishing "Contra Ursum," and in any which is in the middle of them, and they think
case that he would like to see the air clear thatis the reasonwhy they do not divergefurther
of the dust of polemics before having the fromit than the curvatureof theirspheresallows,
treatise printed.6 In between these two let- becausethey do not go round the Earth,like the
ters, while writing the defense of Tycho, rest, but have converse apsides (absidesconver-
Kepler organized his arguments under four sae). What else, then, do they mean, than that
the centre of their orbits is in the region of
chapter headings: (1) what is astronomical the Sun?9
hypothesis, (2) a history of astronomical
hypotheses, (3) a consideration of Apollo- The possible roots of this passage, both im-
nius' lost astronomical hypotheses, (4) the mediate and more distant, led Kepler into
originality of Tycho's hypotheses.7 Within an intriguing historical essay. Let us turn to
the last of these chapters can be found a the details of Kepler's account in chapter 4
fascinating review by Kepler of the possible of "Contra Ursum" and consider the points
predecessors for a geo-heliocentric plane- he makes and the sources from which he
tary system, among whom he counted not draws.
only Martianus Capella but also Macrobius, Ursus, Kepler remarks, noted the doctrine
Pliny, Vitruvius, and even Plato.8 This ex- of heliocentrism attributed to Martianus by
cursus on origins of the geo-heliocentric Copernicus and so thought Martianus to be
idea provides an insight into Kepler's un- the source. But the doctrine of heliocentric
derstanding of the forerunners of Coper- Mercury and Venus can be found, according
nicanism and his reconstruction of the
meanings of the texts involved. It is almost 9 Nicholaus
as an aside that these sources were men- Copernicus, On the Revolution of the
Heavenly Spheres, transl. Alistair M. Duncan (New
tioned. Kepler deliberately posed them as York: Barnes and Noble, 1976), p. 48. The original text
an affront to Ursus. Not only had Ursus reads, "Quapropter minime contemnendum arbitror,
failed to understand properly the writings quod Martianus Capella, qui Encyclopaediam scripsit,
et quidam alii Latinorumpercalluerunt.Existimant
of Apollonius, which Ursus erroneously enim, quod Venus et Mercurius circumcurrant Solem
in medio existentem, et earn ob causam ab illo non
ulterius digredi putant, quam suorum convexitas or-
5 Ibid., p. 235. bium patiatur, quoniam terram non ambiunt ut caeteri,
6 Ibid.
7 The text of
sed absidas conversas (my emphasis) habent. Quid ergo
the treatise itself, excluding Frisch's in- aliud volunt significare, quam circa Solem esse centrum
troduction and notes, appears in ibid., pp. 236-276. illorum orbium." De revolutionibus orbium coelestium
8
Ch. 4 appears at ibid., pp. 270-276. (Nurnmberg: Petreius, 1543), f. 8v.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 369

to Kepler, not only in Martianus but also in Martianus and Pliny from this group and
Macrobius, Pliny, and Vitruvius.?1 While that he (Kepler) will clarify the Platonic in-
Vitruvius only mentions the idea, Macrobius tention further by exposing not only the
and an early commentary (vetus commen- Platonic roots of Martianus and Pliny but
tarius) on Bede attribute the hypothesis to also the utility of others, preeminently Ma-
Plato, but it has been subject to corruption crobius, for laying bare the full meaning of
by Platonists and many others. Macrobius Plato regarding the orbits of Mercury and
even relates the system to the Egyptians and Venus. We shall expand below these out-
so has connected it with the early roots of lines of Kepler's explorations into prisca as-
astronomy. All this, however, has escaped tronomia.
Ursus. Surely he might have asked who
were those "others among the Latins" who, II. a. MARTIANUS CAPELLA
according to Copernicus, were well-versed
in the doctrine. It would also have been Kepler begins with the unambiguous
statements of Martianus,13 quoting a few
useful for Ursus to have sought the author
lines for exactitude. Martianus, it turns out,
of the words, absides conversae,1 which
is the only Latin writer of late Antiquity who
Copernicus took from Pliny.12
The phrase, "converse apsides," becomes clearly offers a heliocentric pattern for Mer-
cury and Venus.14 Kepler, in turn, rather
Kepler's clue to a host of ancient authorities
for heliocentrism. The line of ideological
development which Kepler claims to have 13
Martianus Capella, reputedly a Carthaginian,
identified is a set of direct connections from composed his encyclopedic work, The Marriageof Phil-
ology and Mercury, sometime just before the Vandal
Plato:to Vitruvius, to Pliny, and to Macro- conquest of North Africa in A.D. 426. Among recent
bius. Martianus Capella is said to have taken appreciations of the science in this book is William H.
information from Pliny and to have under- Stahl, "The Quadrivium of Martianus Capella. Its Place
in the Intellectual History of Western Europe," Arts
stood it in a Platonic sense. Finally, Macro- liberaux et philosophie au moyen dge (Montreal: Institut
bius appears as the crucial interpreter of d'Etudes Medievales, 1969), pp. 959-967. Stahl is also
Plato's meaning, and the Macrobian inter- author of a very serviceable translation-the only one
into English-Martianus Capella and the Seven Liberal
pretation is confirmed by a medieval com- Arts, vol. 2: The Marriage of Philology and Mercury,
mentary on Bede. We have as a result a transll. W. H. Stahl and Richard Johnson (New York:
group of Platonic witnesses, according to Columbia U.P., 1977). Of the nine books, Book VIII
covers astronomy.
Kepler. He claims that Copernicus used 14 Kepler seems to have culled carefully the text of
Martianus for the best citations. Comparison with the
standard modem edition shows almost letter-perfect
10
Kepler, Opera, ed. Frisch, 1: 271. Here and below identity; see Martianus Capella, De nuptiis philologiae
I paraphrase and condense Kepler's account. et mercurii, ed. Adolph Dick, addenda Jean Preaux
n This
phrase, translated by Duncan as "their spheres (Stuttgart: Teubner, 1969), a reprinting with typo-
turned the other way" (above, n. 9), is only one among graphic corrections and bibliographical updating from
a series of historical points made by Kepler. It is my the original edition of 1925. Kepler's quotes are: "Tria
focal point in another essay (to be published) for in- item ex his cum Sole Lunaque orbem Terrae circu-
terpreting the understanding of Copernicus. meunt, Venus vero et Mercurius non ambiunt Terram
12 This
peculiar use of the phrase from Pliny was .... nam licet ortus occasusque quotidianos osten-
recognized by Kepler, previously referred to by Rhe- dant, tamen eorum circuli Terras omnino non ambiunt,
ticus, and has been so labeled by commentators ever sed circa Solem laxiore ambitu circulantur. Denique
since. On the other hand, the meaning of the phrase circulorum suorum centron in Sole constituunt, ita ut
and the reason for Copernicus' choice of the phrase to supra ipsum aliquando, infra plerumque propinquiores
describe a heliocentric pattern have never been ex- Terris ferantur." (Kepler, Opera, ed. Frisch, 1: 271.
plained satisfactorily. In fact, Kepler's historical spec- Martianus, De nuptiis, ed. Dick, p. 449, 26-450, 2; 450,
ulation is the only serious attempt at explanation that 19-451, 1.) Venus and Mercury "circa Solem peragra-
I am aware of. See, for example, Owen Gingerich's tione mundana volvuntur .... [horum] circulos epi-
review of Duncan's translation of De revolutionibus in cyclos esse, superioribus (ait)memorasse, id est non in-
Isis, 69 (1978): 293-5, esp. 294. tra ambitum proprium rotunditatem Telluris includere,

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
370 BRUCE STANSFIELD EASTWOOD

neatly avoids parts of Martianus' descrip- Vienna, 1516, contains only Books I-II.17
tion which introduce difficulties of interpre- For the remaining editions two features
tation. For Kepler's purposes the main theme common to all should be emphasized. First,
of heliocentrism seems sufficient. His quo- no edition prior to 1600 has astronomical
tations run as follows: diagrams of any sort. Second, all editions
Threeof these planets,togetherwith the sun and before 1600 have the astronomy divided by
the moon, have their orbitsabout the earth,but chapter headings which are identical to
Venus and Mercurydo not go about the earth. those found in the manuscripts of an Italian
. . . Now Venus and Mercury,although they group, dating back to the eleventh century,
have daily risings and settings, do not travel and not found in any other manuscripts I
aboutthe earthat all;ratherthey encirclethe sun have seen.18 This suggests a wide influence
in widerrevolutions.The center(centron)of their for that set of manuscripts. When we turn
orbitsis set in the sun. As a resultthey are some- to the printed texts which concern the paths
timesabove the sun;moreoften they arebeneath of Mercury and Venus, there appear two
it in a closer proximityto the earth. . . . Venus
and Mercuryorbit around the sun as it makes groupings in the early editions. Later edi-
its celestialrevolution. . . ; their circles. . . as tions, those of 1577 (Basel) and 1599 (Lei-
I have said above, are epicycles;that is, they do den), have one reading, while the others
not encompassthe globe of the earthwithin their have a different version of the section de-
orbits,but describean orbitto one side, in some scribing the heliocentric orbits of Mercury
way. . . . Located on its own epicycle (circulo), and Venus. The earlier printings, in the
Venus goes about the sun, varying its course; chapter headed "That the Earth is Not the
sometimesit passes ahead of the sun, sometimes Center for all the Planets," uniformly follow
it follows after it, and does not catch it; again at a reading which requires intersecting orbits
times it is borne above the sun and at times be- for the inner planets, in a pattern ascribed
neath it. 15
by early manuscripts to Martianus, based
The printed versions of Martianus Ca- on placing Mercury closer to the sun when
pella's encyclopedia begin with the Vicenza above it, and Venus closer to the sun when
edition of 1499 and number eight by 1599 below it.19 In contrast to this reading we
when the annotated edition of Hugo Grotius
appears.16 One of the editions, that of 17 De
nuptiis mercuriiet philologiae cum adnotationibus
Ioannis Dubravii [Vienna; H. Vietor, 1516]. Dubravius'
commentary to Books I-II contains much medieval lore
sed de latere Terrae quodammodo circumduci." (Ke- relevant to their contents.
pler, 1: 271. Martianus, 464, 1-2; 464, 4-7). Kepler 18 These MSS.are Florence San Marco 190 (S. XI) and
further notes that, however the two planets move in its copies in Vat. Urbin. 329 (A.D. 1474-82), Venice
that epicycle, Martianus later says of Venus, "in suo XIV. 35 (4054) (ca. 1485), and Florence Plut. 51. 13
posita circulo eum varia diversitate circumdat, quia (A.D. 1490). I have completed one study involving these
aliquando eum transcurrit, aliquando subsequitur nec MSS.in an article, "The Chaster Path of Venus (orbis
comprehendit, aliquando superfertur, nonnunquam Veneris castior) in the Astronomy of Martianus Ca-
subjacet." (Kepler, 1: 271; Martianus, 446, 3-6). pella," Archives internationales d'histoire des sciences
15These are Kepler's selections, given in Latin in
(forthcoming); further study on these and related MSS.
n. 14 above. The translation is from Stahl, Martianus will appear in another essay, dealing with medieval
Capella, 2: 332 (S. 854), 333 (S. 857), 341 (S. 879), 342 theories of circumsolar planetary motion.
(S. 882). 19 The first four editions (probably the sixth as well,
16 I have been able to examine all
eight of these print- assuming it to be a reprinting of the fourth; see above,
ings except for that of Lyon, 1592, of which a copy n. 16) all read alike here. These are the editions of
does not seem to exist in North America; the British Vicenza: Henricus de Sancto Urso, 1499; Modena:
Library has a copy. It appears that the Lyon 1592 Dionysius Berthocus, 1500; Basel: H. Petrus, 1532;
edition is essentially a reprinting of the Lyon 1539 Lyon: S. Vincentius, 1539. The text concerned reads,
edition, which I have seen; both are octavo, both have "Nam Venus Mercuriusque, licet ortus occasusque quo-
397 pages, and both are from the publishers Vincentius tidianos ostendant, tamen eorum circuli terras omnino
(the earlier by S. Vincent, the later by B. Vincent). non ambiunt. Sed circa solem laxiore ambitu circulan-

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 371

have that of the editions usually considered 857) on the heliocentric inner planets comes
the first two critical editions.20 Vulcanius' a word which caused much medieval com-
edition (1577) relates the positions of the ment. Of all the Renaissance editions not
inner planets to the earth rather than to the one questions, emends, or glosses the word.
sun with the result that the orbits are best It passes without comment until Kopp's
understood as concentric circles around the edition of 1836. The word in question, cast-
sun.21 Grotius' edition repeats Vulcanius' iore, appears as part of a phrase, "a path
reading and notes two revisions, one of [of Venus] more chaste (castiore) and more
which is the preference of intra to infra. In diffuse" than that of Mercury. No medieval
addition, Grotius notes that there are manu- manuscript is known to justify the replace-
script readings which allot to "Mercurius" ment of castiore by vastiore, adopted in the
the one and one-half sign elongation from 1925 edition as a preferable reading. All
the sun, even though it should obviously be medieval glosses on the word suggest a syn-
"Venus," which Grotius adds in the text of onym like strictiore or breviore, which fits
his edition.22 At the close of the passage (S. well with the pattern of intersecting orbits
for Mercury and Venus.23
The sum of our review of the heliocentric
tur. Denique circulorum suorum centron in sole con- passage in Martianus is a difference of two
stituunt. Ita ut supra ipsum aliquando, infra (my em-
words between the earlier editions and
phasis) plerumque propinquiores terris ferantur, a quo
quidem signo uno et parte dimidia Venus disparatur. those of Vulcanius and Grotius. One change
Sed cum supra solem sunt, propinquior est ei (my em- involves a significant shift in the positions
phasis) Mercurius cum intra solem Venus, utpote orbe of the orbits of Venus and Mercury relative
castiore diffusioreque curvetur." The folios of the 1499
and 1500 editions are not numbered; in the 1532 ed., to each other.24 The other change is unim-
this reading appears on p. 191; in the 1539 ed., on portant for meaning but a useful tool for
p. 338. The key word, changing the spatial configu-
ration, is the emphasized ei. tracing quotations.
20 These are the edition of Bonaventura Vulcanius One way to obtain a perspective on the
of Bruges, published together with and following Isi- meaning of the heliocentric passage in Mar-
dore of Seville, Originumlibri viginti ex antiquitate eruti. tianus is to look for interpretations by other
et Martiani Capellae, De nuptiis Philologiae et Mercurii
Librinovem (Basel: Petrus Pema, 1577); and the edition readers. It is clear that Copernicus under-
of Hugo Grotius, Satyricon, in quo De nuptiis Philogiae stood (perhaps, chose to understand) the
et Mercurii libri duo, et De septem artibus liberalibus libri
Martianus passage to involve circles con-
singulares (Leiden: Christophorus Raphelengius, 1599).
Just how critical these editions are is debatable. See, centric to the sun.25 On the other hand, im-
for example, the discussions in Kopp's 1836 edition,
p. vi, and in Dick's 1925 edition, p. 4.
21 Vulcanius' 23
edition (Basel, 1577), col. 192, reads the A detailed study of this text in Martianus' work,
same as that given in the earlier printed editions (above, the medieval MSS.and their glosses, the appended dia-
n. 19), except for a terris in place of the ei and an intra grams, and the plausible meaning of the text in its orig-
in place of the infra. This shifts the relative motion of inal form, will be found in my article, "The Chaster
Mercury and Venus from the sun to the earth in de- Path of Venus . . . ." (above n. 18).
scribing distances and thereby changes the relative 24 The later sections of Martianus' De
nuptiis on
placements of their orbits. The replacement of infra by Venus and Mercury (VIII, 879-883) do not offer infor-
intra represents the large majority among the manu- mation on the exact shape and relative position of each
scripts. However, the infra appears in Avranches 226 orbit. However, it can be noted that there is a unifor-
and in every one of the Italian group of MSS. mity of reading for these sections in every edition from
22
Grotius ed. (Leiden, 1599), p. 289, has the relevant 1499 to 1599, including certain omissions recognized
text, with glosses indicating the alternative readings by later editors.
infra and Mercurius. Curiously, he does not refer to the 25 If
Copernicus used the MS. reading of Martianus
change from ei to terris. His reference to the reading according to the uniform text of the Italian group of
Mercurius may stem from various MSS.,but the most MSS.,he would be supported in seeing concentric orbits
likely is Leiden B.P.L. 88, which carries the variant as for Mercury and Venus, for those MSS.have the terris,
well as a crossed-out venus. This MS.had come to Lei- relating the closeness of the planets to the earth, rather
den just prior to Grotius' editing of De. nuptiis. than the ei, relating the closeness of the planets to the

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
372 BRUCE STANSFIELD EASTWOOD

ages of an unambiguously "Capellan" sys- Brahe's library28 and would, in turn, be

tem, all of which show the concentric orbits, available to Kepler.29 Kepler obviously
are not common before the seventeenth knows what he wants to find in Martianus.
century, when one may find it, for example, In his quotations there is a notable collection
in Edward Sherbume's verse translation of of those statements and phrases which di-
the first book of Manilius' astrology (1675) rectly describe the heliocentric paths of
and in Riccioli's astronomy (1651).26 From Mercury and Venus without including the
the sixteenth century the only published sentence with the troublesome word, cast-
diagram I have encountered for the system iore.30How he understands the castiore we
of Martianus appears in a less than ubiq- do not know. Perhaps he sees through it
uitous elementary textbook on astronomy, without a second thought. Yet there remains
by Valentin Naibod, published at Venice in a difficulty in the text which Kepler must
1573.27 It shows up as an item in Tycho

Brahe, and Michael Maestlin," The CopernicanAchieve-

sun. As already explained, the ei requires intersecting ment, ed. Robert S. Westman (Berkeley: U. California
orbits while the terris allows concentric orbits. All Press, 1976), pp. 285-345, esp. 322-324, where the
printed texts of De nuptiis in Copernicus' lifetime have diagram is also reproduced. Westman, p. 324, calls to
the reading ei. mind the suggestiveness of Naibod's diagram for Ty-
26 Edward Sherbume, The
Sphere of M. Manilius cho. May we not wonder whether this led to any in-
(London, 1675), p. 132; G. B. Riccioli, Almagestum no- quiry by Tycho into manuscripts of Martianus, espe-
vum (Bologna, 1651), 1: frontispiece. Both of these il- cially the Italian group, during his travels of 1575-76?
lustrations are conveniently located in Simeon K. Hen- 28 Ibid., p. 323
inger, Jr., The Cosmographical Glass. Renaissance Dia- 29
Kepler may have had little need for such reassur-
grams of the Universe (San Marino, Cal.: Huntington ance. But if he had, he would probably not have found
Library, 1977), p. 72 fig. 50, p. 69 fig. 48c, respectively. it elsewhere in print. Even in the seventeenth century
Each diagram has a central earth and concentric, he- the name of Martianus was not consistently applied to
liocentric orbits for Mercury and Venus. Sherburne the system. Despite such widely available attributions
attributes the scheme to "Vitruvius, Martianus Capella, as those provided by Riccioli and Sherburne, the system
Macrobius, Beda, and Argol"; this finds a home for the of Martianus (always as concentric orbits for Mercury
presumably prescient system in every authority whose and Venus) was either unrecognized, as in Andreas
words might bear the interpretation. Notably, Plato Cellarius, Harmonia macrocosmica,seu Atlas universalis
and Heraclides are absent. "Argol" is Andrea Argoli et novus (Amsterdam, editions of 1660, 1661, 1708),
(1570-1657), professor of mathematics at Padua (1632- which shows many planetary schemata but none quite
57), among whose astronomical and astrological pub- like that of Martianus, or else the system was named
lications appears Pandosion Sphaericum(Padua: Paulus for others and not Martianus, as in Johann G. Dop-
Frambottus, 1644 [2nd edition, 1653]), in which the pelmaier, Atlas Novus Coelestis (Niimberg, editions of
system of Martianus is depicted on p. 14, with Mercury 1716, 1742), which represents many cosmographies of
and Venus concentrically heliocentric, and labeled as less note (by modem standards) than Martianus. The
"Systema nostrum." On p. 10 Argoli mentions that this system of Martianus is included in this volume but is
system was previously outlined by Vitruvius and Mar- labeled simply as the Egyptian system, a label derived
tianus Capella but without the eccentrics and epicycles most likely from Macrobius, though, unlike Macrobius,
necessary for calculating positions. He claims agree- Doppelmaier does not consider the Egyptian system to
ment between his calculations and those of Tycho. be Plato's, for a separate diagram appears with Plato's
Riccioli, like Sherburne later, calls the scheme the name, following the Timaeus version of a strictly geo-
"Egyptian system"; Riccioli attributes it to Vitruvius, centric system. The continually changing sense of con-
Martianus, Macrobius, Bede, "etc.," which suggests tent as well as advance in knowledge in astronomy
that Riccioli considers the system widely known (cer- during the later sixteenth and seventeenth centuries is
tainly to Jesuits). evoked by Michael Hoskin and Christine Jones, "Prob-
27 Valentinus Naibod, Primarum de coelo et terra in- lems in Late Renaissance Astronomy," Le soleil d la
stitutionum quotidianarumquemundi revolutionum libri Renaissance. Sciences et mythes, Colloque international
tres (Venice, 1573), f. 41r, presents the diagram of (avril 1963) (Brussels: Presses Universitaires de Brux-
Martianus' system; it is attributed to Martianus alone. elles, 1963), pp. 21-31, including notice (p. 27) of Jesuit
Heninger, CosmographicalGlass, p. 58 fig. 43b, repro- uses of the system of Martianus in the first half of the
duces the figure. On Naibod, a professor at the uni- seventeenth century.
versity at Koln who called himself "Physicus et as- 30See above at n. 15 for the complete translation of
tronomus," see Robert S. Westman, "Three Responses Kepler's excerpts from Martianus; the original in Ke-
to the Copernican Theory: Johannes Praetorius, Tycho pler, Opera ed. Frisch, 1: 271.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 373

confront, if he is to be considered a serious (with ei), we learn quickly how he chooses

student of the history of heliocentrism. He to deal with it. Kepler simply evades the
omits the final sentence of Martianus, S. sentence before him in the text, quoting
857, which contains not only the castiore but those passages appropriate to his already
also the passage describing the relative formed conception of the proper interpre-
paths of Mercury and Venus as they circle tation of Martianus Capella's heliocentrism.
above and below the sun. We are in a rea-
sonably good position to see why the omis- II. b. PLINY
sion occurs. Kepler almost certainly has be-
All the statements Kepler takes from Mar-
fore him one of the five editions relating the
tianus are considered derivations from ear-
paths to the sun rather than to the earth.
lier authors, preeminently Pliny. Pliny is
That is, first, Kepler's quotations from Mar-
tianus include an infra,31 which does not distinctly ambiguous and obscure in the
eyes of Kepler. Martianus himself seems not
appear in the text given by Vulcanius and to have fathomed Pliny's mind, but Kepler
Grotius; second, the editions prior to these
believes that the task is not impossible and
uniformly read in the final sentence, "When
proceeds to it. Surely, he says, we must rec-
they are above the sun, Mercury is closer
to it (ei) [the sun], and when inside the sun ognize that Pliny set the paths of Venus and
Venus [is closer], inasmuch as it orbits in a Mercury below the sun.33 Kepler then refers
to three recent commentaries on a vexing
path which is more chaste and more dif-
fuse."32We are not now concerned with the text, where Pliny says,
final phrase, "more chaste (castiore) and First therefore let it be stated why these two
more diffuse," but with the much clearer [planets,Venus and Mercury,]never recede far-
and more disturbing description of the rel- ther [than 46 degrees and 20 degrees respec-
ative positions of the inner planets above tively] from the sun, often turningback toward
and then below the sun, which involves the the sun [before reaching these extremes].Both
have converse apsides (conversasabsidas),being
model of intersecting orbits. Having iden-
set inferiorto the sun;and as muchof theircircles
tified Kepler's source as one of the editions, is below [the sun] as there is of the aforesaid
perhaps that of 1499, with this vexing text [circles]above [the sun], and so they cannot ex-
tend farther,since the peripheryof theirapsides
31
In the phrase "infraplerumque propinquiores Ter-
has no greaterlength.34
ris"; see above, n. 14, for Kepler's complete quotation;
cf. De nuptiis, ed. Dick 450, 24, where the reading is
"intra plerumque ...." 33Kepler, 1: 271, quotes from Pliny: "infra Solem
32 See above, n. 19, for the Latin text, which is there ambit ingens sidus appellatum Veneris" (II, 36); "prox-
drawn from the editio princeps of 1499. It might be imum Veneri sidus Mercurii inferiore circulo fertur"
argued that Kepler's edition was that of 1499 and no (II, 39). C. Plinius Secundus, Naturalis Historiae libri
other on a minute basis. His quotation uses the spelling XXXVII, edd. Ludwig von Jan and Karl Mayhoff (Le-
"plerumque" (my emphasis) rather than "plerunque" ipzig: Teubner, 1906), 1: 138, lines 1, 20-21.
(my emphasis) in the phrase "infra plerumque propin- 34
My translationfollows Kepler'stext, adding con-
quiores Terris" (see above, n. 31). Only the 1499 edition tent from the Jan-Mayhoff edition as needed; Kepler
uses this spelling; all subsequent editions through 1599 has abbreviated it. As Kepler comments, Pliny is indeed
use "plerunque." While Kepler shows the bias of ellip- difficult at this point. See the passage in Pliny, II, 72
sis in his selection of quotations, he does not in "Contra (ed. Jan-Mayhoff, pp. 149-150). The English translation
Ursum" show the carelessness of unwitting error. Even in the Loeb version (Pliny, Natural History, transl. H.
so, I would rather not argue on such small grounds. Rackham, London, Heinemann, 1938, 1: 219) is both
In any case, he is using an edition with a very prob- inadequate and misleading at this point, partly because
lematical reading, the ei before Mercurius. Only this the Latin text established by Detlefsen is preferred to
last point is dependable now, for the MS. folio con- the Jan-Mayhoff text. The text reported by Kepler reads,
taining the Martianus quotation has been lost since "Primum igitur dicatur, cur hi duo nunquam longius
Frisch used it; I have this information from Nicholas a Sole abscedant, saepe ad Solem reciprocent? Con-
Jardine (letter of 8 September 1981). versas habent utraeque absidas ut infra Solem sitae,

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
374 BRUCE STANSFIELD EASTWOOD

Each of the commentators used by Kepler tator on Pliny, Collimitius, or Georg Tann-
found this passage obscure. Jakob Ziegler stetter von Thannau (1482-1535), also
(1480-1549) of Landau, a professor of mentioned by Kepler, gave a somewhat dif-
mathematics at Uppsala, published a com- ferent explanation. Kepler seems to ignore
mentary on Book 2 of Pliny's Natural History the difference in interpretation,36 perhaps
and proposed two successive interpreta- because Collimitius like Ziegler was giving
tions. At first he suggested that the conversae an account in terms of eccentrics alone, not
absides of Venus and Mercury were simply epicycles. Collimitius understood conversae
apsides (or orbits) with "differences in posi- absides to refer to the apogees of the excen-
tion" from those beyond the sun. Assuming ters, not to the complete orbits as seen by
the orbits of sun and inferior planets to be Ziegler. While Pliny himself had used
successively surrounding each other as ec- absides at different places to mean either
centric orbits in the descending order Sun- eccentric orbit or apogee of excenter, the
Venus-Mercury Ziegler explained that Pliny difference between the commentators rep-
finds the smaller paths of the inferior plan- resents a difference in the manuscript tra-
ets to be limited, or bound, to the vicinity ditions, leading the commentators to choose
of the sun, unlike the superior planets which separate meanings of absis.37 Thinking of
are unlimited. On the other hand, this lim- apsidal points, not apsidal paths, Collimitius
iting in longitude of the inferior planets is considered Pliny to have argued for apogees
balanced by the greater freedom in latitude for the superior planets always at or near
of the inferior planets.35 Another commen- conjunction with the sun and apogees of the
inferior planets always in opposition to the
tantumque circulis earum subter est quantum supeme
praedictarum, et ideo non possunt abesse amplius,
quoniam curvatura absidum non habet ibi longitudi- A
nem majorem." (Kepler, 1: 271-272).
35 Jakob
Ziegler, In C. Plinii de naturali historia librum
secundum commentarius,quo difficultatesPlinianae, prae-
sertim astronomicaeomnes tolluntur .... Item, G. Col-
limitii, et Joachimi Vadiani, in eundem secundum Plinii
scholia (Basel: H. Petrus, 1531), pp. 193-194. The com-
mentary to Pliny's sentence beginning Primum igitur
(ed. Jan-Mayhoff 149, 25-150, 2), asking why the
bounded elongation of Venus and Mercury: "Respon-
det, Conversas, id est, differentes positione apsidas ha-
bent, ut quae infra solem sitae sunt tantis spaciis, quan-
tis apsides trium superiorum sunt supeme sive supra
solem sitae, proptereaque quanto sunt breviore ambitu
orbium, tanto reperiuntur distancia a sole minore. Con-
cipit enim Plinius . . . orbes stellarum alios in eade
superficie ratis spaciis distare, ecentros ad se et mundi
centrum sicut in subiecta formula, sit A Solis, B Veneris, (fig., p. 194)
C Mercurii, et Apsidum B et C curvaturae sint. Hic BD, 36
Kepler1: 272.
BE, illhic CF et CG, signum medium orbium sit H et 37
Interestingly, this difference in manuscript read-
diameter AH, quae secet inferiores apsides communiter ings is preserved in two leading critical editions. De-
in K. Ad hanc designationem aptemus verba Plinii. Non tlefsen's edition, used for the Loeb translation, reads,
possunt, inquit, abesse a sole amplius, ideo quoniam "conversas habent utraeque apsidas ut infra solem si-
curvatura DE et FG apsidum B et C non habet longi- tae, tantumquecirculi earum sub terra est (my emphasis)
tudinem a sole per spacia zodiaci et coeli maiorem quantum supeme praedictarum . . ." (p. 218, 11. 4-
quam sit ex H medio signo ad margines apsidum, ergo 6).The Jan-Mayhoff edition, which I prefer, reads, "con-
utraeque stellae statuunt margines suarum apsidum versas habent utraeque apsidas ut infra solem sitae,
prope modum simili ratione, quam diximus, verum tantumquecirculis earumsubter est (my emphasis) quan-
quod minus superioribus abeunt a sole per longitudi- tum superne praedictarum. . ." (p. 150, 11. 2-4). Col-
nem zodiaci et coeli, hoc latius illis excurrunt ab eclip- limitius explains on the basis of the former reading,
tica in austrum vel boream." Ziegler on the latter.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 375

sun. Collimitius concluded that this makes the basis of his photographic copy of the Leningrad
MSS.he informs me (letter of 7 January 1980) that the
no sense, for Mars, etc. have apogees in text does indeed read "Mycillum" (f. 296r) and "My-
places different from the sun, and Venus is cilloque" (f. 296v). Jacob Milich (1501-1559), a pro-
never opposite the sun at apogee.38 fessor of mathematics at Wittenberg, produced his first
edition of and commentary on Pliny's Book II in 1535;
Kepler, we must remember, reports that a subsequent edition (1537) was reprinted in 1538 and
he read both Ziegler and Collimitius on ab- 1543; his final edition appeared in 1553. The edition
sides conversae and that both rejected their of 1537 was essentially the final version, and the edi-
tion of 1553 still carries at p. 143, in commentary on
initial explanations as impossible. Although the motion of Mercury, a position computed "die 13
different in details ignored by Kepler, both Februariipraesentis anni 1537." C Plinii LiberSecundus
used eccentric circles in their accounts. Zie- de mundi historia, cum commentariis Jacobi Milichii
(Frankfurt a.M.: P. Brubachius, 1553), carries Pliny's
gler did not give up with one try, however, conversas absidas on p. 137, and the relevant commen-
and returned to the task with an interpre- tary on pp. 139, 140. Milich says, "Accidunt autem
tation using epicycles, which was acceptable haec omnia, quod centrum Epicycli proportionabiliter
movetur super centro aequantis ipsi Soli, ita ut aequali
to another sixteenth century commentator,
tempore aequales angulos singuli super suis centris
Milichius,39 according to Kepler. The inter- describant, Sol super centro sui deferentis, inferiores
vero super centro sui aequantis, ita ut lineae medii
motus horum trium semper sint in eodem Zodiaci loco
38 Collimitius'
(and Vadianus') comments are pub- secundum longitudinem. Unde potest intelligi causa,
lished together with those of Ziegler (above, n. 35). See cur hi Planetae neque ad oppositionem, neque ad re-
p. 438 for Collimitius' explanation of conversas habent liquos aspectus digredi a Sole possint" (p. 139). Further
absidas: "Conversas absidas intelligere videtur, quod on (p. 140) Milich says, "Atque haec est sententia Plinii,
contrario loco cum absidibus trium superiorum suas cum inquit: Conversas habent utraeque absidas, hoc
habeant. Cunque illi supra terram maiorem ecentrici est, utraque stella habet Epicyclum, qui converso modo
partem habeant, hi eandem infra terram. Ideoque in circumfertur, supra secundum successionem signorum,
minori arcu nobis extante Venere et Mercurio, longius infra vero contra. Sunt autem hi Planetae infra Solem
abscedere non posse. Quod totum ratione caret, cum siti certa positione, ita ut linea a centro deferentis per
tres superiores diversis in locis a Sole suas habent centrum corporis solaris ducta, secet has Absidas seu
auges. Item Venus in Cancro ubi Sol, ergo non in op- Epicyclos in duas fere aequales partes. Existente igitur
posito." Sole, et per consequens etiam centro Epicycli duorum
39Kepler refers to a "Mycillus" at this point. This inferiorum, in horizonte, tum altera ipsius pars est su-
spelling is a variant on the more common "Micyllus," pra, altera vero est infra terram, ut in parte horizontis
a well known sixteenth-century classical scholar. Mi- occidentali pars dextra est infra, sinistra supra." Like
cyllus, or Jakob Moltzer (1503-1558), never edited or Ziegler Milich (p. 137) reads a crucial clause of Pliny
wrote a commentary on Pliny. Johannes Classen, Jacob as "tantumque circulis earum subter est" (see above,
Micyllus, Rectorzu Frankfortund Professorzu Heidelberg n. 37).Milich moves the center of epicycles for Mercury
von 1524 bis 1558, als Schulmann, Dichter und Gelehrter and Venus with the sun and thus explains in Ptolemaic
(Frankfurt a.M.: Verlag fur Kunst und Wissenschaft, fashion Pliny's reference to the bounded elongation of
1859), makes no statement that would clarify Kepler's these two planets. He then proceeds to conversas absidas
reference. A search through Micyllus' works most likely and suggests that it means only that the inferior planets
to say something about Pliny in the context of astron- have epicycles, or absides, rotating so that the planet
omy reveals no clue either. I have gone through his moves positively in the zodiac when supra and nega-
edition of C. Iulii Hygini. . . Arati. . . Procli de sphaera tively in the zodiac when infra. A further gloss on
libellus (Basel: Hervagius, 1549), his Arithmeticae log- Pliny's text is the statement that infra and supra with
isticae libri duo (Basel: Oporinus, 1555), which reference to horizon phenomena of the inferior planets
has an appendix "De temporum supputatione" are to be taken to mean below and above the horizon,
(pp. 269-290) and an appendix "De partibus astron- viz. when the epicycle of either planet is divided by
omicis et earum supputatione" (pp. 201-268), the latter the horizon. Therefore "infra solem sitae" in Pliny is
being published separately as well (Paris: Wechelus, taken to mean "after (or following) the sun." Earlier
1557); and his posthumous Sylvarum libri quinque, ed. printings of the commentary can be found under the
Julius Micyllus (Frankfurt a.M.: Brubacchius, 1564); same title in the edition of Wittenberg: Petrus Bruba-
none of these speaks of Pliny. It seems fruitless to go chius, 1537 (reprinted 1538); and Frankfurt: Petrus
further, through Micyllus' editions of Lucan, Martial, Brubachius, 1543; f. 69r-v. Under the title Commentarii
et al., when we can show that Kepler slipped and re- in librum secundum Historiae mundi C. Plinii conscripti
ferred to Micyllus when he meant Milichius. Consid- a Iacobo Milichio (Hagenau: Petrus Brubacchius, 1535),
ering the possibility of a slip by Frisch, the editor of f. 47r-v, appears his first version, in which he had said:
Kepler's "Contra Ursum," I have consulted Nicholas "... inquit Plinius caussa cur Venus non ultra 46 et
Jardine (Dept. of History and Philosophy of Science, Mercurius non ultra 23 partes discedant a sole, est quod
Cambridge), who is editing the "Contra Ursum"; on utraeque habent conversas apsidas, id est, positione

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
376 BRUCE STANSFIELD EASTWOOD

pretation via epicycles asserted simply that Kepler's investigation of Pliny involves
the upper planets are in conjunction with somewhat different tactics than the subse-
the sun on the outer arcs of epicycles, while quent interpretations of Macrobius, Vitru-
the inferior planets are in conjunction on vius, or Plato himself. For Pliny, Kepler
the inner arcs of epicycles. As a result the finds it useful to introduce other students
morning risings and stations occur on op- of Plinian cosmology. By showing the in-
posite sides of the sun (longitudinally) for adequacy of epicyclic analysis to account for
the two groups of planets. Ziegler proceeded Pliny's phrase, absides conversae, Kepler
further with his epicyclic interpretation of moves to establish a further connection in
Pliny to say that the centers of the epicycles the Platonic chain. Epicyclic analysis-that
of Venus and Mercury are always below the proposed by Milich, Ziegler, and Tannstet-
sun and thus account for Pliny's remarks. ter-is not so much wrong as insufficient.
However, as Ziegler finally noted, the epi- Kepler may well reflect at other points a tra-
cyclic version of absides conversae has no dition witnessed already by Theon of
basis in Pliny's text. With this Ziegler aban- Smyrna (fl. ca. A.D. 100) that Plato seemed
doned the attempt to work out what Pliny to prefer epicycles to eccentrics in astron-
himself meant by the phrase.40 omy, since the former offer a more perspic-
uous uniformity of planetary motion,41 but
inversas, ita videlicet ut superior pars istorum circu-
lorum sit brevior, inferior vero amplior; contrario plane
modo atque accidit in superioribus. Ideo etiam non
possunt abesse a sole amplius quam curvatura apsidum
ipsis concedit. Vocat autem curvaturam apsidum non
circulum deferentem sed alium quendam qui ipsas in
altitudinem tollit, nos ilium epicyclum vo-
camus. .. ."
40 Ziegler offers the following commentary to Pliny:
"conversasapsidas habent. Haec non est causa unde tan-
tum ultra sive citra absint, sed est secundum caput se-
quentis canonicae. Habent itaque absidas conversas, id
est, positione differentes ab his trium superiorum. Hae
enim coniunguntur soli existentes in partibus summis
ep[ic]yclorum ubi E, unde exeunt in apparitionem et
stationem matutinam G lateris. Inferiores duae iun-
guntur soli in partibus imis epicyclorum ubi F, exeu-
ntque in apparitionem et stationem matutinam lateris
H propriorum epicyclorum. Postea enim Plinius plu-
ribus argumentis rerum prodit ut habeant conversas
apsidas."
Ziegler continues, with his comment on ut infra solem
sitae tantumquecirculis nempe absidibus earum stellarum
subter solem est quantum est superne praedictarumtrium:
"Hoc quidem recoeptum est, quod infra solem sunt, ita
ut tres aliae supra solem totis orbibus, sed non inde
causa petitur, cur conversas habeant apsidas, sed latet
haec in consiliis naturae. Sive dicamus, Infra solem si-
tas, id est, mediis centris epicyclorum existere sub loco
solis, unde illis accidit altera ad hunc coniunctio. Ex
qua alia sequuntur has diversa a tribus [superioribus].
Ad hanc sententiam infra solem sitas esse tam est
verum, quam certe a Plinio ita intellectum non fuit ut
tota eius oratio convincit." Continuing with et ideo non-
possunt abesse longius etc.: "Quia enim ut diximus, cen- (fig. p. 197)
tra epicyclorum sunt perpetuo sub linea motus solis,
non possunt stellae abesse longius, quam fert compre- 41
Theon of Smyrna, Exposition des connaissances
hensio epicycli." Ziegler, Plinii de naturali historia, p. mathimatiques utiles pour la lecture de Platon, ed. and
198. transl. J. Dupuis (Paris: Hachette, 1892), p. 304; here,

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 BRUCE STANSFIELD EASTWOOD 377

when investigating Pliny, Kepler records Copernicus' thoughts. As far as we know,

and passes beyond the epicyclic account. He Kepler can have had no solid evidence for
wanders purposefully through commentar- his elaborate fabrication, and yet he pro-
ies close to his own time and contemporary ceeds as follows.
with Copernicus. These commentaries on ButCopernicus,takenby both myself and Ursus
Pliny fail to offer a satisfactory interpreta- as witness in this matter,neglectingchaptereight
tion of the Plinian phrase conversae absides [of Pliny's BookII]and emendingfor coherence,
in terms of either eccentrics or epicycles. which is repeatedlyneeded for Pliny, made it so
Insofar as this search for the meaning of that Martianuscould be heard in the words of
Pliny is also part of a discourse on the nature Pliny. Thus in the opinion of CopernicusPliny
of astronomical hypothesis, the path fol- surely wrote, "conversashabent utraequeabsi-
lowed by Kepler here seems to parallel one das ut circum [replacing"infra"]Solem sitae;
of his larger concerns both in "Contra Ur- tantumquede circulisearumsubterest quantum
sum" and in the previously published Mys- ["praedictarum"omitted] superne." The word
"infra"was an adulterationof the text, and so
terium cosmographicum (1596). Kepler is
was the word "praedictarum"accordingto Co-
searching for an astronomical account of pernicus.Whatmoved Copernicusto this, so that
physical things, not simply for a geometrical he believed Pliny to have spoken thus, was that,
device.42 So Pliny, who seems to have been for VitruviusbeforePliny and Martianusas para-
in touch with a sense of physical things, phraserof Pliny, having the same teaching on
cannot be adequately explained by the ge- the planets and with the same sentences, errors,
ometry of eccentrics or epicycles when con- and often using the same orderof words,45each
ceived only as calculating tools. borrowedfrom his predecessor.46At this point,
As construed by Kepler, Copernicus' in- whereit dealtwith the holdingtogetherof Venus
tent was to find coherence in Pliny. There- and Mercurywith the sun, they surelyexpressed
fore, says Kepler, Copernicus ignored an this intent which Copernicusrestoredto Pliny.47
earlier section of Pliny in which the spheres Kepler, the determined Platonizer, man-
of Venus and Mercury are given the tradi- ages not only to find the heliocentric notion
tional, intra-solar positions,43 and proceeded in Pliny via emendation rather than direct
to read Pliny so that the words of Martianus evidence, but also to give to Copernicus a
Capella could be fit to Pliny's text.44 What similar view of Pliny and of the pedigree of
is remarkable here is the length to which the heliocentric idea. If we give Kepler his
Kepler goes in his imaginative recreation of due, we must remember that Copernicus did
not explain the linkage of Martianus Capella
with the Plinian phrase; the compressed and
in ch. 34, Theon says Plato seemed to prefer epicycles elliptical reference by Copernicus would not
to eccentrics, since he spoke of planets carried by circles in itself tell Kepler how the conjunction of
not spheres. Such an interpretation already shows the
speculative search for Platonic roots operating in the
Martianus and absides conversaecame about.
first century A.D. Relevant to Platonic realism is Theon's
remark (ch. 18, ed. Dupuis p. 240) that Plato considered
45
retrograde motion a reality, while Adrastus (contem- Kepler here conveniently forgets that Pliny never
porary of Theon) understood it to be an appearance. said, clearly and precisely as did Martianus, that Mer-
42 See Westman, "Kepler's Theory of Hypothesis," cury and Venus do not have the earth as center and
pp. 238-241, 246; more recently, Nicholas Jardine, that they do have the sun as center. The close similar-
"The Forging of Modem Realism: Clavius and Kepler ities between Pliny and Martianus are mostly in ob-
Against the Sceptics," Studies in History and Philosophy servational accounts and the transfer of data.
of Science, 10 (1979): 141-173, esp. 165-171. 46 To make a claim for Pliny regarding Vitruvius is
43Kepler, "Contra Ursum," ed. Frisch, I, 271; the far more speculative than the exaggerated claim already
relevant sections are Pliny, N.H., II, 36 and 39 (ed. Jan- made for Martianus regarding Pliny. Kepler seems to
Mayhoff, p. 138). indulge in wishing more than theorizing at this point.
44 Kepler, ed. Frisch, 1: 272. 47
Kepler, ed. Frisch, 1: 272.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
378 BRUCE STANSFIELD EASTWOOD

Left in the dark, Kepler chooses to bring his cury and Venus circle the sun and its rays
own light to bear on the confusing material as a center, according to Kepler.50Vitruvius'
before him rather than leave an apparent language and the inexact character of this
chaos of statements unordered. When Ke- presentation allow for ambiguity. What is
pler finishes, Pliny is believably (?) Platonic certain is that he understands the inferior
and heliocentrist,48based on the speculative planets to move "about" the sun as center
clarification of his relationships to Marti- of their bounded elongations, which need
anus later and Vitruvius earlier. not be circular paths around the sun. We

II. c. VITRUVIUS
back into Latin, I have checked the only German edi-
Kepler bolsters his recreation of Pliny's tion, Vitruvius teutsch, transl. Walther Ryff (Niimberg:
Johannes Petreius, 1548), f. cclxxiir,21-26; the text here
intent by using Vitruvius and Martianus, clearly translates into the Latin of the 15c.-16c. edi-
who were, he says, respectively master and tions, not into the abbreviated version of Kepler, viz.:
student to Pliny. Each can properly be used "haben doch der Mon, der Stem Mercurii, Veneris, die
Sohn selber, der stem Martis, Jupiters unnd Satumi,
to interpret Pliny, who did not, says Kepler, iren eygnen lauff disem entgegen das sie in den auffs-
seem to perceive adequately the reasonings teigenden graden iren lauff der massen halten, das einer
of the astronomers. Pliny read through his mer oder minder in grossem oder kleinen bezirck von
Occident genen Orient ein yrrigen lauff hat in der hi-
sources quickly, mixed them all together, mmlischen Spher." To solve the problem of Kepler's
understood things only in part, and de- apparent divergence from all readings at the time, I
ceived both himself and his readers with the propose the simplest possible solution. Noting that he
uses the same verbiage in his ordered list of the planets
obscurity of his words. Thus Kepler on as the readings in the early editions and that he also
Pliny. Looking at Vitruvius, Kepler explains uses the same verbiage as those editions again in his
that his ordering of the planets in the tra- second quotation from Vitruvius (below, n. 50), I as-
sume that either Kepler himself or his modem editor,
ditional way (Moon-Mercury-Venus-Sun- Frisch, mistakenly extended the quotation beyond the
Mars-Jupiter-Saturn) may have mislead word "Saturni," which is the last word that should be
Pliny.49Vitruvius went on to state that Mer- included. Unfortunately, the original manuscript for
ed., Frisch 272, 31-40, containing the Vitruvius quo-
tation, is now lost. I suggest that we read the latter part
of the statement, "contrariam fixis viam sub zodiaci
48
This is clearly Kepler's reasoning and intention; signis peragrent," as Kepler's or Frisch's paraphrase of
see Kepler, 1: 272. Nor was Kepler wrong in supposing Vitruvius rather than as a partial quotation from a
that Copernicus believed Pliny to be this sort of helio- mysterious and unidentifiable source. Since the mean-
centrist, but Copernicus' reasons for the belief require ing of the paraphrase is fully in accord with the text
further investigation, which I provide in a separate it replaces, the assumption of an incorrect placing of
study. the end of the quotation seems plausible.
49There are twelve 50 Kepler's reading of Vitruvius (cf. ed. Krohn, 203,
printings of the Latin Vitruvius p.
prior to 1600. Among these not one gives the same 3-7) here is, "Mercurii autem et Veneris stellae circum
phrasing of the first short quotation in Kepler's account, Solis radios, Solem ipsum uti centrum itineribus co-
standing for and equivalent in meaning to the text given ronantes, regressus retrorsum et retardationes faciunt"
in De architectura libri decem, ed. F. Krohn (Leipzig: (ed. Frisch, 1: 272). This reading agrees exactly with
Teubner, 1912), p. 202, 20-23, which reads, "luna, six of the twelve Latin editions to Kepler's time. These
stella Mercurii, Veneris, ipse sol itemque Martis et lovis are the editions of: Florence, Philippus de Giunta, 1513,
et Saturni ut per graduum ascensionem percurrentes f. 147v; Florence, haeredes Philippi Juntae, 1522, f.
alius alia circumitionis magnitudine ab occidente ad 352v; Lyons, G. Huyon, 1523, f. 161r; Strassbourg,
orientem in mundo pervagantur." Kepler represents Gregorius Machaeropioeus, 1543, p. 212; Lyons, Ioannes
this statement with the words, "Luna, stella Mercurii, Tomaesius, 1552, pp. 370-371; Venice, F. Franciscius
Veneris, ipse Sol, itemque Martis, Jovis et Satumi con- and J. Crugher, 1567, p. 287. The last of these contains
trariam fixis viam sub zodiaci signis peragrent" (Kepler, the commentary of Daniele Barbaro, who points out
Opera, ed. Frisch, 1: 272). What is remarkable is the the extensive borrowing made by Pliny from Vitruvius
uniformity of the early editions regarding this text. With and Pliny's disagreement with the mechanism of ret-
the exception of the substitution of "circuitionis" for rograde motion advanced in the same chapter by Vi-
Krohn's reading of "circumitionis," the twelve early truvius for the inferior planets. The problem, says Bar-
editions all agree with Krohn's reading to the letter. baro, is that Vitruvius did not use eccentrics, epicycles,
With the thought that Kepler might have translated and so forth (ed. Venice, 1567, p. 287, 17 if.).

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 379

have already seen Kepler's reference to planets so as to explain the phenomena ob-
Martianus,51 and the placing of Pliny in a served. Vitruvius offers the following order
line of thought from the clarified Vitruvius of information. (1) The diurnal appearances
to Martianus suggests little choice other come from the rotation of the heavens on
than a heliocentric reading for Pliny's the- an axis passing through the center of the
ory of the inferior planets. earth.54(2) The globe is divided into north-
Kepler's use of the Vitruvius passage ern and southern hemispheres by an equa-
brings focus on a very circumscribed state- tor, and the zodiacal belt is inclined to the
ment, for which the conceptual framework equator.55(3) The twelve signs of the zodiac
is brought from Kepler's wholly external are equal divisions of the belt, and they have
discussion. The framework offered in Vitru- representational images. No more than six
vius' book finds no place in Kepler's dis- signs will ever be visible from the earth, half
cussion. Vitruvius' mention of the paths of of the zodiac always being below the earth
Mercury and Venus is so concise as to be (between setting and rising). One and the
mysterious, if we assume that he has more same force (vis) accounts both for the dis-
to say than is apparent. But, unlike Kepler, appearance and the reappearance of these
we shall not assume the unspoken. stars.56 (4) The seven planets, given in the
In his Ten Books on Architecture Vitruvius traditional order (Moon-Mercury-Venus-
begins Book 9 with praise for the legendary Sun-Mars-Jupiter-Saturn), move from west
achievements of Plato, Pythagoras, Ar- to east through the zodiac in opposition to
chimedes, and others concerned with math- the diurnal motion. Ascending outwards
ematical science.52 Having dealt in the first from the earth, each planet encountered has
seven books with buildings, in the eighth an orbital period whose length is increas-
with water and waterways, he is prepared ingly greater.57(5) The moon completes its
to discuss the sundial and its rationale, in- orbit in something more than twenty-eight
cluding the effects of solar rays.53 Chapter days, during which the sun moves through
1 then introduces the general architecture only one sign. The sun, then, requires
of the cosmos, oriented around the lumi- twelve months to pass through twelve signs
naries whose positions and effects should of the zodiac.58 (6) Mercury and Venus re-
be understood in order properly to under- main close to the sun, oscillating to either
stand the sundial. While the sun and the side of it, and their anomalies are caused by
zodiac are preeminent in Book 9, the initial this continuing proximity to the sun
chapter summarizes the whole celestial throughout the zodiac. This section of
framework. The reader is presented with the Chapter 1 is, of course, the passage always
turnings and orbits of the fixed stars and cited (often incompletely) in support of the
notion that heliocentrism appears in the
51Above at n. 15. Architecture. Vitruvius' exact statement is:
52
For a useful introduction to Vitruvius and relevant On the other hand, the planets Mercury and
bibliography see John Ward Perkins, Derek de Solla
Price, and G. J. Toomer, "Vitruvius Pollio," Dictionary Venus, in the vicinity of (circa)the rays of the
of Scientific Biography(New York: Scribners, 1978), 15 sun, as if crowning it by travels (itineribus)
(Suppl. 1): 514-521. Vitruvius, De architectura libri de-
cem, ed. F. Krohn (Leipzig: Teubner, 1912), pp. 195-
200; Krohn considers this to be preliminary material 54Ibid., p. 201, 16-27.
and begins the numbering of chapters in Book IX with 55 Ibid., pp. 201, 27-202, 3.
the following material. For convenience, I use Krohn's 56
Ibid., p. 202, 3-19.
numbering of chapters, but every Latin edition of Vi- 57
Ibid., p. 202, 20-23.Note that the word circumitio
truvius before 1600 numbered the chapter on planetary (line 22) means the full path through the zodiac.
orbits as chapter 4, not chapter 1. 58 Ibid., pp. 202, 24-203, 3. The language remains
53
Ibid., p. 200, 30-201, 16. general, and there is no exact measure of the year.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
380 BRUCE STANSFIELD EASTWOOD

through the center (percentrum)they make re- Vitruvius. When we move ahead for his
turns backwardsand retardations,and likewise discussion of Mars, Jupiter, and Saturn, we
at stations because of this circuit(circinationem) shall see that he says nothing about centers
they hold back in the intervalsof the [zodiacal] of their orbits there, and he says nothing
signs."59 about centers for the orbits of Mercury and
This statement as translated requires certain Venus here.
clarifications. First, it must be emphasized Leaving behind the supposedly crucial
that circa (or circum) does not require trans- passage in Chapter 1, we find that Vitruvius
lation in the sense of travel along a geo- has more to say about the two unusual plan-
metrical circle. Justification for such a non- ets-unusual in that their orbits have the
geometrical sense comes from the rest of the same period as the sun. (7) He says that we
passage and from the larger framework. The can learn best about what has just been
paths of the planets are described as itinera, stated by observing Venus; it follows the
hardly a precise geometrical or astronomical sun and precedes the sun, and it stays longer
term. These itinera are said to move through in some signs than in others.61 But not a
the center (per centrum), not around it. The word about a heliocentric orbit! (8) Mercury
word circinatio means observed circuit, or takes 360 days to complete the zodiac, pass-
path, through the twelve signs of the zodiac ing in this period again "to that sign from
as will appear later, equivalent to circumitio which in the previous circuit (circumlatione)
used previously in Chapter 1 as well as it began to make its passage."62 (9) When
later.60In other words, both iter and circina- Venus is "freed from the hindrance of the
tio refer to the observed path of the planets rays of the sun" (liberata ab inpeditione ra-
through the zodiac. Both prior to this pas- diorum solis), the planet covers a sign in 30
sage on Mercury and Venus and following days, but it can take much longer as well.63
it the discussion focuses on the time of each Here we learn that the rays of the sun cause
planet's path through the zodiac. As intro- the variations in the planet's path. Similarly
duction to more detailed treatment of Venus above, we found that the retrogradations
and Mercury, this passage is a statement of occurred somehow in relation to the sun's
the time the two planets take to course
through the zodiac, that is, the same time
as the sun. Hence the word circa means 61
Ibid., p. 203, 7-19. Iter (line 15) and circumitio(line
nothing about the more theoretical shape 19) are equivalent.
of the orbit of either Mercury or Venus, but 62
Ibid., p. 203, 21-22, for the quoted text; lines 19-
only that the appearancesof these two plan- 24 for the complete description of the planet's zodiacal
path. The words iter (lines 19, 23) and circumlatio (line
ets are such that they remain close to the 22) have the same meaning, which is zodiacal path.
sun, for nowhere else in the whole chapter The annotations of Gulielmus Philander to Vitruvius
does this essential information appear. appear in the editions of Strassbourg, 1550; Lyon, 1552;
and Lyon 1586. Philander says of Sun-Mercury-Venus
Careful reading of the text reveals that the that they have the same zodiacal period, and he cites
whole account is observational. The ap- Pliny for the ordering of Venus and Mercury succes-
pearances of the planets as they pass through sively in orbits within the sun's orbit; in the three
editions see respectively pp. 407-408, p. 377, pp. 372-
the zodiac are the only points of concern for 373. A separate printing of Philander's Annotationes,
without Vitruvius' text, contains the same comments;
see the edition of Rome, loannes Andreas Dossena,
59
Ibid., p. 203, 3-7: "Mercurii autem et Veneris stel- 1544, pp. 316-317. The edition of Lyon, 1523, illus-
lae circa solis radios uti per centrum eum itineribus trates the planetary spheres in strictly geocentric order
coronantes regressus retrorsus et retardationes faciunt, (f. 160v). The German translation by Ryff (Niimberg,
etiam stationibus propter eam circinationem morantur 1548) gives a planetary diagram in Ptolemaic order
in spatiis signorum." (f. cclxxvV).
60 63Ibid., pp. 203, 24-204, 1.
See above, n. 57.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 381

rays (circa solis radios . . . eum . . . coron- diac, whether iter, circinatio, circumitio, or
antes). (10) Each of the superior planets has circumlatio.7'None refers to a geometrically
its own proper zodiacal period as well. It defined path, either eccentric or epicyclic-
seems noteworthy that Vitruvius does not not even a geocentric circle.72 Finally, the
mention any sort of center here for Mars64 commonplace image of a wheel is set forth.
or Jupiter65or Saturn.66Instead he continues The turning of the wheel is analogous to the
to provide an observational account re- diurnal appearances, and each of the seven
marking that the superior planets are espe- planets has its position in an ordered inter-
cially affected by the sun at trine.67(11) The val from the center outwards. The closer to
effects of the sun are discussed in some de- the center, the faster the motion of the
tail,68and the sun's rays, as if coherent pow- planet contrary to the diurnal motion in
ers, are the focus of the discussion. At one order to explain the observed orbital move-
point Vitruvius speaks of the "violent im- ment of each planet. This last image con-
petus of the sun from its rays at trine."69 firms again our impression that there is
This theme of the sun's rays as the activating nothing heliocentric about Vitruvius' de-
force in the planetary anomalies appears as scription of planetary motion.
a leitmotif in the first chapter of Book 9. Kepler's interpretation of Vitruvius evi-
Solar force is, in fact, the only theoretical dently suffers not only from presupposition
notion used in discussing the phenomena but also from a correlative hyperselectivity.
of the heavens-apart from the general as- When only the briefest text is cited, only the
sumption that circular motion is the pattern briefest understanding is possible. Kepler is
for celestial motions. In this regard Kepler's able to achieve his goal, the establishment
insistence that Pliny and Vitruvius had the of a Platonic pedigree, by highly restricted
same theory of the planets has an ironical quotations as well as by unduly biased in-
correctness. Both men said nothing about terpretations.
heliocentrism, and both said a good deal
about the force of the sun as an agent in
II. d. PLATO
planetary anomalies.70
To complete our survey of Vitruvius on Plato's words on the inner planets are
the paths of the planets we should recall also obscure, Kepler remarks, but there are
that his various terms for planetary path all clues which show his real meaning as Ma-
refer to the observed path through the zo- crobius later explained.73 The case is made
for Plato by using words from Plato, Mar-
tianus, and especially Macrobius. Kepler
64
Ibid., p. 204, 2-6. begins with a quotation from Timaeus
65
Ibid., p. 204, 6-9. 38 C-D.
66Ibid., p. 204, 9-15.
67 Ibid., p. 204, 15-19. When God had made bodies of the several[wan-
68
Ibid., pp. 204, 19-205, 23. derers],he set them into revolutionswhich the
69
Ibid., p. 205, 3-4.
70 It is beyond the purpose of this study to pursue
Kepler's interest in Pliny, but it is worth noting that
71
both Pliny and Vitruvius, especially Pliny, spoke of the Among an even longer list of locations, the fol-
force of the sun's rays as a motive power for the planets. lowing pages and lines show the common use and
Pliny, followed by Martianus, elaborated on this and meaning of these terms: 202, 22; 203, 6; 203, 15-16;
accounted for the stations and retrogradations of the 203, 19; 203, 22; 203, 28; 204, 14; 204, 15; 205, 24;
planets through the force of solar rays. Kepler seems 205, 27-206, 1.
to have this in mind when he discusses the contrariam 72 The earth is
only mentioned as center when Vi-
vim in Plato's Timaeus. Finally, the resonance between truvius describes the turning of the sphere of the fixed
the suggestions of Pliny et al. and Kepler's own con- stars.
73
ception of the force of solar rays is difficult to avoid. Kepler, I. 272.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
382 BRUCE STANSFIELD EASTWOOD

circuit of each74 would require, seven [bodies] in In explication of this text Kepler points out
seven [orbits], the Moon in the first next to the that the planets Mercury and Venus are in-
earth, the Sun in the second above the earth, cluded in the same orb as the sun, but the
then the Evening Star [Venus] and the one they orb is then divided into three to account for
call sacred to Mercury in such [circles] as by rea- their variations. He would have us under-
son of their speeds run equal to the orbit of the
stand this as equivalent to the sun on an
Sun, yet have received a power contrary (con-
eccentric with Venus and Mercury each
trariamvim) to it, whence the Sun, Mercury, and
Venus include (comprehendunt)and are included having an epicycle around the sun. Thus,
by each other.75 he says, Plato can be seen as if using the
schema of Martianus and revising the plan-
etary order to Sun-Mercury-Venus, not Sun-
74
Kepler's text seems to be his own translation. At Venus-Mercury, for Mercury in this reason-
this point Plato's text refers to "The Other"/"The Dif- ing travels closer to the sun. Also the use
ferent" (OarTpov) while Kepler translates as "cujusque"
("each one"). See n. 75 below.
of comprehenderefor antecederebrings Plato
75
Regarding the interpretation of this text see Francis together with Martianus. Furthermore, says
Macdonald Comford, Plato's Cosmology. The Timaeus Kepler, Pliny seems to have deduced his
of Plato (London: Routledge and Kegan Paul, 1937), p.
105, esp. n. 2; compare A. E. Taylor, A Commentaryon
conversae absides from Plato's contrariam
Plato's Timaeus (Oxford: Clarendon Press, 1928), 196- vim, which allows us to connect both Pliny
202, comm. on 38 D, 3 and 38 D, 4. Kepler's version and Martianus to Plato at this point.76
reads, "Cum effecisset Deus corpora singulorum, ea
It is noteworthy that Kepler uses only the
posuit in circumlationes, quas cujusque (my emphasis)
circuitus requireret, septena in septenas, Lunam qui- Timaeus passage and not the corroborative
dem in primum circa Terram, Solem vero in secundum passage in the Republic. His reasons for this
supra Terram, Vesperum vero et quam Mercurio sacram
dicunt in tales, qui celeritatis quidem ratione aeque- are not clear. He may have chosen to use
currentem Soli orbem eunt, contrariam vero ei vim sor- only the work known during the Latin Mid-
titi sunt; quare et comprehendunt et comprehenduntur dle Ages. More likely he chose because the
a se mutuo et secundum eadem Solem, Mercurium et
Venerem." This text is different from the text provided Timaeus is so thoroughly cosmological in its
by Calcidius, Timaeus a Calcidio translatus commentar- concerns and it, not the Republic, would
ioque instructus, ed. J. H. Waszink (London: Warburg
Institute, 1962), pp. 30, 22-31, 4, which differs at only
one point (reordering to "circunfertur ab eo", p. 31, I have found in all the printings of Ficino's translation
line 2) from identity with the same passage in Timaeus is the infrequent substitution of "sydus" (e.g. Lyon,
Platonis, sive de Universitate, Interpretibus, M. Tullio 1590) for the emphasized "orbem." While Ficino's
Cicerone, Chalcidio, una cum eius docta explanatione. translation is in some ways closer to Kepler's than is
(Paris: Guil. Morelius, 1563), p. 79. However, the am- Calcidius', it appears that one of two later translations
biguous verb comprehendereappears in each. Calcidius' is Kepler's basis for his own reading of Plato. The trans-
translation was edited in 1520 (at Paris) and again in lation of loannes Serranus (Geneva: Henricus Ste-
1563. By far the most commonLatintranslationof the phanus, 1578) is thoroughly unlike Kepler's reading
period before 1600 was that of Marsiglio Ficino, whose (see transl. Serranus, III, p. 38 D). But the version by
version appeared at least twenty-three times from ca. lanus Comarius (Basel: Frobenius, 1561) is quite close
1484 (at Florence) to the edition of Geneva, 1592. The to Kepler's and is as follows: "Corpora vero ipsorum
text of Ficino's translation of Timaeus 38 D remained singulorum quum fecisset deus, posuit ipsa quae sep-
constant throughout this period and is represented as tem essent, in circumvolutiones septem numero, quibus
well by any one as any other example. I choose at alterius circuitus progrediebatur. Lunam quidem in pri-
random the separate printing of the Timaeus (Paris: mum circum terram circulum. Solem vero in secundum
Franciscus Zampinus, 1527), [f. 14v, 11-21: ] "Cum super terram. Luciferum vero, et sacram Mercurii dic-
vero stellarum huiusmodi corpora septem fecisset deus, tam stellam, in eos orbes qui velocitate quidem, ae-
septem orbibus qui alterius ipsius diversaeque naturae qualem cursu cum Sole circulum progrediuntur, con-
circuiti volvuntur, adhibuit. Lunam in primo supra ter- trariam autem ipsi vim sortiti sunt. Unde et compre-
ram circuitu posuit. In secundo Solem. Luciferi deinde hendunt et comprehenduntur etiam secundum eadem
globum, et Mercurii sacrum ut dicitur orbem (my em- a seipsis vicissim et Sol et Mercurius et Lucifer" (p. 737,
phasis) circulis affixit, Soli velocitate aequalibus, po- 35-41). Both the "contrariam vim" and the verb com-
tentia vero illi contrariis. Quo fit ut apprehendant se prehendere, preferred to antecedere, are especially re-
invicem, et a se vicissim apprehendantur similiter hae marked by Kepler.
stellae, Sol, Lucifer, et Mercurius." The only variation 76
Kepler,1: 272-273.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 383

seem to be the appropriate source for a cos- are overtaken by each other. Given the
mological theme. Whatever Kepler's rea- homocentric spheres, either theory or ob-
sons, his choice of the Timaeus source alone servational description yields a non-heli-
is to be noted. ocentric interpretation, and the theoretical
A more dispassionate assessment of the interpretation is the more questionable one
passage from Plato requires only the reader's as with Vitruvius above and Macrobius be-
agreement that vocabulary must be taken low. We simply have no authority and little
in context. For an author who says nothing reason for shifting from a broadly based
elsewhere about eccentrics or epicycles, it cosmographical description to a technical or
seems rather a far stretch of the imagination mathematically theoretical account unless
to assume that he has suddenly taken an we find that authority or reason within the
active interest in the mathematical devices text. In the case of Plato as with the others,
of astronomy. Similarly, for an author whose the only grounds for such a shift come from
concern is cosmological it seems curious that external criteria-the demand by a later
a definite departure from the contemporary reader that the best explanation supposedly
assumption of concentric orbits, as in the available to the author be assumed to have
Eudoxan system of homocentric spheres, been used by that author. Such is Kepler's
would not be marked by a very clear de- demand of Plato. It constitutes the final step
scription, either technical or commonplace. in establishing the Platonic pedigree for a
Plato, like Vitruvius, Pliny, and Calcidius, heliocentric idea. Yet the idea is absent from
appears to mean retrograde motion when all of Kepler's named sources prior to Mar-
he speaks of a contrary power (contrariam tianus Capella, and the Platonic conception
vim)77 whereby Mercury and Venus move appears rather different in Kepler's repre-
with the same general motion as the sun but sentation than in Plato's writing. In fact, it
not with the same vis. As a result of the is Copernicus who is the more careful, if less
contrariety Plato has the three fellow trav- inspired, student of Plato, for Copernicus
elers with distinctive powers "comprehend says quite simply that Plato's Timaeus places
and be comprehended" (comprehendunt et Mercury and Venus above the Sun.78
comprehenduntur)by each other. Following
directly on the statement of what we call II. e. MACROBIUS
retrograde motion, it might seem reasonable
In establishing the Platonic pedigree, Ke-
that Plato should introduce heliocentric epi-
pler uses Macrobius' account as centerpiece
cycles to explain retrograde motion for Mer-
for the reconstruction of the hidden ancient
cury and Venus. However, it is not proven
truth. Through a succession of quotations
that Plato had any interest in epicycles at
from Macrobius' Commentaryon the Dream
all. To read the devices of the time of Apol-
lonius (ca. 200 B.c.) back a century and a of Scipio, I, 19, Kepler makes the following
half is not an acceptable practice. The ap- points. (1) Both the supra-solar and the in-
fra-solar traditions concerning the position
propriate astronomical model with which
to parallel Plato is the model of homocentric of Mercury and Venus are correct, and Ma-
crobius understood this.79 The cause of the
spheres. On this pattern the comprehendunt
et comprehenduntur makes perfectly good
78De revolutionibus
sense. On the background of the fixed stars (1543), f. 7v, 19-22; transl. Ro-
the sun, Mercury, and Venus overtake and sen, On the Revolutions, pp. 18, 35-36.
79
Kepler, 1: 273; he begins at Macrobius, Commen-
tarii in CiceronisSomniumScipionis, ed. Ludwig von Jan
(Quedlinburg: G. Bassius, 1848), pp. 102-103 (I, xix,
77Translation of the complete passage above at 2). This edition of Macrobius is superior to the more
n. 75. available version of Eyssenhardt (Teubner, 1893) and

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
384 BRUCE STANSFIELD EASTWOOD

intermittent reversal of Mercury and Venus In Kepler's reading the "it" (hic) which "is
is not far to seek, and Macrobius saw that encircled as an inferior" leaves an ambi-
Plato had an explanation. (2) Plato reasoned guity; we cannot tell whether the reference
that the order of the planets corresponds to is to "sun" or "circle" of the sun. However,
their orbital periods.80 (3) In presenting the Kepler's addition of the epicyclic interpre-
orbits of Venus, Mercury, and Sun, Macro- tation and his further emendation in the
bius spoke of the three as having a single reading of "inferior" make the sense of
coelum in which they orbit in a period of, Macrobius take a quick turn towards certi-
more or less, one year. This Kepler finds tude, and, we should add, a false certitude
significant, for the coelum has a distinctly at that. With Kepler's suggestions, the reader
physical and not just phenomenal quality- of Macrobius can easily imagine the order
as if the three are tied together and do not Sun-Mercury-Venus and Mercury and
simply appear together.81 (4) The planets Venus moving concentrically either around
Mercury and Venus are much more properly the sun itself or around the center of the
associated with the sun than with the moon, sun's epicycle. Plato, as explained by Ma-
for their motions are closer in timing to sun's crobius, who is in turn explained by Kepler,
than to the moon's.82 (5) The whole sense can be understood as a heliocentrist. (6) Fi-
of Plato as interpreted by way of Macrobius nally, Macrobius' remark about relative ease
can be aided by the improvement of a single of observation of Mercury and Venus also
word, says Kepler. The heliocentric pattern makes more sense, says Kepler, in the light
for Mercury and Venus is suggested when of a heliocentric motion for Mercury and
we read a crucial passage of Macrobius in Venus. If they are more easily seen when
the following way:83 lower and more difficult to see when higher,
The circle along which the sun travels (you we can only agree, for a heliocentric pattern
should understandan epicycle on a concentric, makes these observations the most readily
by which we are accustomedto save the eccen- expected.85
tricity;if you do otherwise,the words of Macro- Macrobius86 receives the most extensive
bius will not fit togetherfor you), it is encircled
as an inferior(preferably"interior")by Mercury, by Kepler. Kepler's text of Macrobius reads, "Circulus
and the superiorcircleof Venus also includes it, per quem Sol discurrit . . . hic, inquam, a Mercurii
and so it happens that these two stars, when circulo ut inferior (malim interior) ambitur, illum quo-
traversingthe superiorarcs of their circles, are que superior circulus Veneris includat, atque ita fit, ut
hae duae stellae, cum per superiores circulorum suorum
thought to be above the sun; when they pass vertices currunt, intelligantur supra Solem locatae; cum
throughthe inferior[arcs]of the circles,the sun vero per inferiora commeant circulorum, Sol eis su-
is consideredto be above these [two planets].84 perior aestimetur."
85
Kepler, 1: 274; Macrobius, ed. Jan., pp. 104-5 (I,
xix, 7-8).
is not improved upon by the edition (2nd) of Willis 86
Macrobius' Commentariiin Ciceronis somnium Sci-
(Teubner, 1970); the stricturesusually placed upon pionis was far more widely available than Martianus'
Eyssenhardt'sedition are more appropriatelyreserved De nuptiis. The Commentarii was first published by
for his edition of the Saturnalia, not the Comm. in Som- Nicolaus Jenson at Venice in 1472. It was printed four
nium. more times to 1500 and had thirty-three printings be-
80 Kepler, 1: 273; Macrobius,ed. Jan., p. 103 (I, fore 1600. I have inspected all 33 printings to compare
xix, 3). with Kepler's (and the modem) version of I, xix, 2-8.
81 Kepler, 1: 273; Macrobius,ed. Jan., p. 103 (I, This has only been possible with the help of Nicholas
xix, 4). Jardine's corrections, based on his recent reading of the
82 Kepler, 1: 273; Macrobius, ed. Jan., pp. 103-104 MS.,to Kepler's text in Frisch's edition (conveyed to me
(I, xix, 5). in a letter of 8 September 1981). Kepler's reading is not
83
Kepler's additions to Macrobius' text are enclosed identical to any single printing. The modest divergences
in parentheses in the following quotation. in his quotation from Macrobius, aside from the emen-
84 Kepler, 1: 273-274; Macrobius, ed. Jan.,
p. 104 dations emphasized, are small grammatical and typo-
(I, xix, 6). Here I discuss Macrobius only as interpreted graphical corrections and improvements in style. More

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 385

treatment in Kepler's search. While Marti- constitutes Kepler's key witness in tracing
anus' text is more straightforward (espe- the Platonic connection. Of all Kepler's
cially when carefully excerpted), it does not sources Macrobius is the only author both
refer to Plato in the context of planetary recognizably Platonist and arguably helio-
orbits. Macrobius does. In confronting Ke- centrist. The relevant section of the Com-
pler's reading of Macrobius not only are we mentary has been said to have the dubious
faced with well chosen omissions as with virtue of vagueness, and even obscurity,
the reading of Martianus, but we also en- which might permit a reader to find what
counter questionable emendations to the he wants more easily than would otherwise
text. Macrobius' Commentary on the Dream be the case. Having seen what Kepler dis-
of Scipio inspires Kepler to raise the game covers in Macrobius, we shall begin anew
of interpretation to a higher level of so- by outlining what we can find in Macrobius.
phistication. The passage translated below includes all
In many ways Macrobius' Commentary of Kepler's quotations and adds slightly
at the beginning, more substantially at
the end.87
of these occur with the Macrobius text than other quo-
tations because of its length. Kepler (Opera, ed. Frisch, Next a few things must be said concerningthe
1: 271-274) quotes about two lines from the vetus com-
order of the spheres (sphaerarum),
about which
mentarius, seven from Plato, four from Vitruvius, five
from Pliny, twelve from Martianus, and about thirty- Cicero can be found to disagree with Plato by
six lines from Macrobius. Judging from word choice saying the sphere of the sun is fourth of seven,
first and from word order, spelling, and grammatical that is, located in the middle; Plato places the
endings secondarily, I tentatively identify one group sun next above the moon, that is, holding among
of four printings as most likely sources, another group
of four printings as reasonable sources, and all other the seven the sixth position down from the top.
printings as unlikely sources for Kepler's quotation. The And the doctrine of the Chaldeans unites Ar-
four "reasonable," or second, choices are the editions chimedes to Cicero, while Plato has followed the
of 1485 (Brescia), 1492 (Venice), 1500 (Venice), and
Egyptians, progenitors of all the branches of phi-
1521 (Venice). Better choices are the "most likely"
editions of 1483 (Brescia), 1501 (Brescia), 1513 (Venice) losophy, who considered the sun to be positioned
and 1515 (Paris). Kepler's word choices (and some cor- between the moon and Mercury.88 They more-
rections/spellings) dictating my selection of these are
the following. (I use as a page and line reference text
the most widely distributed modem edition of Macro- 87 The translation which
follows is rendered mostly
bius, that of Eyssenhardt in the Teubner series, 1893, in very literal fashion, sometimes painfully so. My pur-
in which Kepler's quotation runs from 558, 31 to 560, poses in doing this are (1) to preserve where appro-
2.) 558, 35: om. et1; 559, 2: 1. existimant pro aestimant; priate the tone and even innuendo of Macrobius, (2)
559, 10: 1. est tantum pro tanto est; 559, 11: 1. pera- to preserve exact vocabulary of a potentially technical
gendum pro peragrandum; 559, 24: confundit, soler- character, (3) to preserve what seems to me to be vague-
tiam; 559, 29: intelligantur; 559, 31: 1. existimetur pro ness and obscurity in at least one part of the passage
aestimetur; 560, 1: 1. meliorem pro veriorem. Readers (unlike the obscuring of the obscurity in William Harris
will notice that some of Kepler's readings here were Stahl's translation of Macrobius, Commentary on the
not correctly transcribed by Frisch, hence my debt to Dream of Scipio, New York, Columbia, U.P., 1952,
Dr. Jardine. From this preferred group I am willing to pp. 162-164).
speculate that the 1483 (Brescia) edition may be the 88 Many students of Macrobius have observed his
edition actually used by Kepler, because this edition incorrect representation here of Plato's ordering of the
alone, of the eight first and second choice printings, planets. Among others, see Jan's ed. of Macrobius,
does not have 559,8: tamen pro tantum, thereby adding Commentarii,p. 102, note; and W. H. Stahl's transl. of
one more similarity to Kepler's, which also reads "tan- Macrobius, Commentary,p. 162, n. 1, for a useful, brief
tum." Frisch seems to have made some use of printed review. Plato's ordering of the planets was Moon-Sun-
editions of Macrobius in establishing the quotation, and Venus-Mercury-Mars, and so forth. There is a curious
he must have used one of the later editions, since he inversion of appropriate labels in Sherbume's transl.
reads (at 560,1) "veriorem" rather than "meliorem"; of Manilus, The Sphere, p. 131, where the "Systema
"meliorem" is the only choice to be found until the Platonicum" is given with two variants: (1) "Plato"-
edition of Koln, 1526, in which (and in many subse- Earth-Moon-Sun-Mercury-Venus. . . , (2) "Porphy-
quent editions) a marginal note glosses "meliorem" as rius"-Earth-Moon-Sun-Venus-Mercury. ... The same
"alias, veriorem." inversion and labels appear in William Leyboum, Cur-

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
386 BRUCE STANSFIELD EASTWOOD

over grasped and made known to reason why seen this during that course (cursu) of the planets
the sun was believed by so many to be above which sometimes, as we said, seems (videtur) to
both Mercury and Venus. Nor have those who be inferior; and this appearance is really more
so opine (aestimant/existimant) strayed far from noticeable, for it can be seen more clearly (liberius
a semblance of what is true (a specie veri procul apparet) then. When they occupy the superiors
aberrant).Indeed, the following sort of thinking they are hidden more [quickly] by the sun's rays
led to the belief in such a variant (permutationis). (Nam cum superiora tenent, magis radiis occulun-
From Saturn's sphere, which is first of the seven, tur). And so this latter conviction (persuasio) has
all the way to the sphere of Jupiter, second from gained strength, and the corresponding order of
the highest, there is so great a distance of inter- the planets has been accepted almost universally.
vening space that the higher planet completes a Nevertheless, more perspicacious observation
course of the zodiac in thirty years while the discerns a truer order, which, apart from the ex-
lower does so in twelve. Further, the sphere of amination of the appearances, the following rea-
Mars stands so far from Jupiter as to run the same soning also recommends: that the moon, which
course in two years. Likewise Venus is so far lacks a light of its own and borrows from the
below the realm of Mars that a year suffices it sun, is necessarily subordinated (subjectam)to the
for coursing the zodiac. Now in fact the planet source of its light. That the moon has no light
Mercury is so near Venus and the sun so near of its own, that all the other planets are seen by
Mercury that these three circle through their their own light, the following explanation ac-
heavenly realm (caelum) in the same interval of counts for: those planets above the sun are lo-
time, that is, in a year more or less. And so Cicero cated in that purest aether, in which everything
called these two courses the companions of the whatever is naturally and intrinsically light (lux),
sun, because in the same interval they never which, complete with its fire, inclines toward the
withdraw far from each other. The moon fell so sphere of the sun, so that the zones of heaven
far below these that it covered in twenty-eight distant from the sun are perpetually burdened
days what they did in a year. Therefore among by cold, as will be shown later on. Indeed, the
the ancients there has been no disagreement over moon, since it alone is beneath the sun and next
the order of the three higher planets, which to the non-luminous realm of transitory things,
shows forth obviously and clearly by virtue of could have no light if not for the sun above it,
immense distance, nor disagreements over the to which it shines back. And so because the low-
realm of the moon. The nearness of those three est part of the whole universe is the earth and
companions, Venus, Mercury, and the Sun, has the moon is the lowest part of the aether, they
confounded their ordering, but only for some call the moon also an earth but an aethereal
authorities. For the explanation has not escaped one.89
the shrewdness of the Egyptians and is thus. The
orbit (circulus) along which the sun travels is sur- In order to interpret fully Macrobius'
rounded as an inferior by the orbit (circulus) of meaning in this passage, only part of which
Mercury;this latter orbit in turn the superior orbit
of Venus contains, and it happens thus that these
89 Macrobius,Commentarii, ed. Jan, pp. 102-105 (I,
two planets, when traveling through the superior
vertices of their orbits are thought (intellegantur) xix, 1-10); cf. ed. Eyssenhardt(Leipzig:Teubner,1893),
pp. 558, 26-560, 17; the only differencebetween these
to be located above the sun, and when they pass two editions over this passage is at ed. Eyssenhardt,
easily (commeant) through the inferiors of their p. 560, line 12, where Janreads ipsafor Eyssenhardt's
orbits, the sun is believed (aestimetur/existimetur) ima. I follow Jan'sreading.Willis,MacrobiiCommen-
to be superior to them. Therefore, those who tarii,2nd ed. (1970), pp. 73, 12-75, 3 suggests a few
more, less significantchanges, none of which strikes
have said that these planetary spheres (sphaeras) me as commendable.Willis,I, pp. vii-viii, lists only six
[of Mercury and Venus] are below the sun have MSS.used for his edition of the Comm. in somnium. The
firstsatisfyingdiscussionof the MSS.sinceJan'sin 1848
is that of BruceBarker-Benfield, "The Manuscriptsof
sus Mathematicus (London,1690), fig. opp. p. 434. See Macrobius'Commentaryon the SomniumScipionis,"D.
illustrationsin Heninger,Cosmographical Glasspp. 73, Phil.thesis,OxfordUniversity,1975;this is a masterful
80. These confusionsseem to derive from Macrobius. and authoritativestudy.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 387

is used by Kepler, we must pay attention tion requires a number of unacceptable sup-
not only to the short section referring to the positions. It requires first of all that Macro-
orbits of the sun and its "companions," bius read Calcidius' Commentaryon Plato's
Mercury and Venus, but also to Macrobius' Timaeus90and considered Plato's system to
vocabulary and tone throughout the pas-
sage and even to other parts of his Com- 90Just what Macrobius took from Plato's Timaeus
mentary which bear on the meaning of the and what he took from Calcidius' Commentaryon the
passage. The tone is set by appellations and Timaeus is not always clear. However, it seems reason-
able to credit Macrobius with a bit more sense than
prejudicial words in comparisons of the two
does his translator, William Harris Stahl, who is so
points of view being considered. While often ready to indict Macrobius for carelessness and/
Plato's predecessors include those (Egyp- or stupidity. Unfortunately, Stahl rarely carries through
with a fair trial for his suspect. For example, Stahl
tians) who were the "progenitors of all the
transl., Commentary,p. 196, n. 20, finds inconsistencies
branches of philosophy", Cicero's authori- between the statement by Macrobius of differentspeeds
ties have no such auspicious label. One side for Sun, Mercury, Venus and the earlier statement that
of the argument is clever enough to have the three travel together. Briefly, Stahl's confusions can
be unraveled as follows. At II, iv, 2-4, Macrobius says
"grasped and made known the reason why" that outer planets have higher tones and move faster;
the other side arrived at its position; only as noted a long time ago (ed. Jan, p. 149, n. 4), this
Plato's Egyptians were shrewd enough for statement refers to the daily rotation of all the spheres.
At I, xxi, 6-7, Macrobius speaks of varying speeds for
this. On the other hand, Cicero and others the planets due only to varying distances from the cen-
with the same point of view can be forgiven, ter; this statement clearly refers to the orbital periods
of revolution, and there is no mention at all here of the
says Macrobius, for supposing (existimant)
sun and its "companions," since they have the same
as they did, for they have not "strayed [too] period. At I, xix, 4, as we have already seen, Sun-Mer-
far from a semblance" of the true point of cury-Venus course together, and this contradicts nei-
view. So he explains patiently how the be- ther of the previously cited statements. The contradic-
tions existed only in the translator's mind. More to the
lief in such a permutatio ever came about. point, the contradictions which can be found are not
The way it happened, he says, is by way of fabrications of Macrobius but are difficulties in the Pla-
the appearances. (And we all know that sav- tonic tradition. While Macrobius may not have the cor-
rect order of Plato, it is clear that Plato's Timaeus alone
ing the appearances is not an exercise in would not provide that knowledge. In Calcidius' trans-
determining the exclusively correct system!) lation of Timaeus 38 D a reader will learn only that
The appearances result in what are to be Mercury and Venus travel "together" with the sun; a
relative order is not given for the two planets (ed.
called estimations, judgments, or opinions, Waszink, p. 31, 1-4). In Calcidius' version of Timaeus
to wit persuasiones. But a more intelligent 36 D (ed. Waszink, p. 28, 21-24), the relative motion
observer will discover a "truer order" than of each planet is not given; one finds only that three
move at the same speed and the other four at propor-
that of Cicero and the Chaldeans, a "truer tional speeds. These translations by Calcidius represent
order" aso reinforced by a proper under- Plato adequately, so that one must look to commen-
standing of the moon's light. In the face of taries on Plato for more definition. Calcidius' com-
mentary on the Timaeus offers no explicit clarification
anything less than direct contradiction, the of the exact order of Sun-Mercury-Venus in elaborating
distinctly biased vocabulary of Macrobius' on 38 D (see Calcidius, Commentarius,cc. 108-112; ed.
account should be sufficient to make it clear Waszink, pp. 156-159), and we learn the detailed order
only from his commentary on 36 D (see Commentarius),
(1) that Plato's system is to be taken as cor- cc. 96-97; ed. Waszink, pp. 148-150), where the pro-
rect, and (2) that the alternative to Plato's portions of planetary motions are given. Calcidius (c.
system has "strayed" and is explicable as 96, Waszink 148) uses the sequence 1, 2, 3, 4, 8, 9, 27
as multiples of the earth-moon unit distance in locating
an opinion based on appearances. the relative positions of the planets. Macrobius follows
Two terms used in the passage and re- a lost work of Porphyry, not Calcidius, in using the
quiring delimitation are circulus and vertex. sequence 1, 2, 3, 4, 9, 8, 27 as successive multiples each
of the immediately preceding planetary distance (II, iii,
Macrobius uses circulus as orbit, not epi- 14-15; ed. Jan, p. 148), so that Mars, for example, is
cycle. To suppose the epicyclic interpreta- 9 times the distance between earth and Mercury. Only

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
388 BRUCE STANSFIELD EASTWOOD

involve the epicyclic explanations found in the assumption that his audience, ostensibly
that Commentary.9'Yet even Calcidius' Com- his young son, Eustachius, and more gen-
mentaryuses circulus commonly for orbit or erally a fairly unsophisticated audience (as-
apparent orbit instead of epicycle.92 Fur- tronomically), will immediately understand
thermore, Calcidius distinctly describes the shift from circulus (orbit) of the sun to
Plato's order of the planets as they are given circulus (epicycle) of the planet, even though
by Plato,93 which happens to differ from Macrobius never mentions nor explains epi-
Macrobius' "Platonic" order of Moon-Mer- cycles anywhere in the Commentary!In fact,
cury-Venus, and so forth. The interpretation it seems a bit unreasonable to assume that
of Macrobius' circulus as epicycle requires Macrobius should introduce such an audi-
ence to epicycles at all, and there is no solid
evidence that he ever does. A careful read-
with this later elaboration of Plato by Platonic com-
ing of Macrobius' explanation will show
mentators, such as Calcidius and Porphyry, is the Ti-
maeus order of the planets given adequate definition, that he is describing what appears to hap-
and all the commentators seem to ignore the difficulty pen. Epicycles do not appear; they are de-
in affirming that the sun with Venus and Mercury have vised. Finally, Macrobius seems to use cir-
similar orbital periods while giving the three remark-
ably different "harmonic" intervals. Here Macrobius
culus and cursus to refer to the observed or
seems to contradict himself, and he is in good Platonic the traced path of the planet and sphaera
company. This last contradiction may be eliminated, to mean a more bluntly physical being of
if we use Taylor's explanation of Timaeus 35 D-36 D
(A. E. Taylor, Commentary, pp. 154-174), where he the planet's orbit. Both are circular, but cir-
shows that the "speeds" of the planets are understood culus suggests a calculation or observation,
as the times of revolution; three planets may thus have while sphaera suggests an entity as such.
the same "speed" and different distances from the com-
mon center of revolution while revolving together. But Each planet, including Mercury and Venus,
Taylor (p. 173, n. 1) also finds further contradiction in has a distinct sphaera, and there is no sug-
Macrobius at II, iv, 9. It is worth remarking that we
have no knowledge of the early history of the illustra-
gestion that one intersects or depends upon
tions in Calcidius' translation and commentary. Ma- another.94
crobius may or may not have had them. Also, Calcidius In his description of the orbits (circuli) of
comes tantalizingly close to giving the Sun-Venus-Mer-
Mercury and Venus, Macrobius accounts for
cury order in cc. 72-73 (Waszink 119-121) but does
not do so. the appearances in a way much closer to
91 For example, Calcidius, Commentarius,c. 109 (ed. that of Pliny than to that of the mathematici,
Waszink, p. 156). and the use by Macrobius of the term vertex
92
Ibid., c. 110 (p. 157); as explained generally by
A. E. Taylor, Commentaryon Plato's Timaeus, pp. 200- seems to fit especially well into a Plinian
201; more extensively by Godfrey Evans, "The As- sort of understanding here. Initially we may
tronomy of Heracleides Ponticus," Classical Quarterly, be unsure of his meaning in the use of the
64 (1970): 102-111; with a useful technical addendum
by Otto Neugebauer, "On the Allegedly Heliocentric word. It may, after all, be of rather vague
Theory of Venus by Heraclides Ponticus," American meaning. A bit later (I, xxii, 7)95Macrobius
Journal of Philology, 93 (1972); 600-601. H. B. Gotts- uses vertex in the phrase vertex sphaerae to
chalk, Heraclides of Pontus (Oxford: Clarendon Press,
1980), pp. 69-81, reviews the whole question of the mean outermost part of the sphere (viz. sur-
Calcidius text and decides against Evans and Neuge- face). While vertex would commonly be
bauer. Gottschalk's uses of Vitruvius and Macrobius translated to mean extremity or high point-
taint his argument. His review of arguments building
on some of the Calcidius MS. diagrams is a step in the in uses by Cicero, Vergil, Aratus-I believe
right direction. However, I am preparing an argument Macrobius has in mind a more precise
based on a rereading of the text and a review of all the
MS.diagrams to the effect that the Calcidius text, which
is the sole remaining bastion for the Heraclidean helio- 94
In I, xx, 14-16 (Jan 113-114), Macrobius uses cir-
centric Venus (and Mercury?), does not support a he- culus and sphaera as apparently interchangeable terms.
liocentric interpretation. Even here, I believe, the distinction I have made will
93 Calcidius,
Commentarius, c. 96 (ed. Waszink hold.
p. 148). 95Jan ed., p. 128.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 389

meaning (at I, xix, 6), derived from Pliny's After a review of the tone and vocabulary
Natural History. In Book II Pliny uses ex- of Macrobius' passage on the order of the
tremitascirculi96 in his observational account planets, a straightforward discussion of the
of the planetary appearances. With explicit doctrines explicitly espoused seems almost
reference to Mercury and Venus, the phrase superfluous. Nonetheless, let us consider
extremitas circuli is made to explain the them succinctly. Macrobius says that there
greater difficulty in seeing these planets are two fixed orders of the planets, the
where they are judged to be "superior." Egyptian and the Chaldean, and nowhere
Macrobius' vertex circuli (I, xix, 6) seems to does he introduce any variable order for the
have the same sense as Pliny's extremitas planets Mercury and Venus. Remembering
circuli. Pliny's account is, in fact, an account that the references to positions above and
of the operation of his conversae absides, below the sun designate, first, one or an-
and Macrobius seems to mean that the su- other fixed order and, second, the appear-
perior vertices are the absides altissimae of ances to be contrasted to reality, one can see
the planets. He does not use the term vertex that no variability in the true order is ever
with inferior but only with superior, which suggested by Macrobius. To add weight to
is just what we would expect for apogees his explanation of the appearances, Macro-
(absides altissimae). According to Pliny's ex- bius introduces a "reasoning" to further ad-
planation, the extremitascirculi is the fastest vance his position on the "truer order" of
part of the orbit for the inner planets, just the planets. This reasoning, an interpreta-
the opposite of the outer planets, and so the tion of the moon's light, concludes the dis-
phenomena of the inner planets are less cussion of planetary order. According to
apparent when Mercury or Venus is in that Macrobius the moon alone requires sunlight
(superior) part of its orbit. This is exactly in order to be illuminated. All the other
what Macrobius wishes to convey when he planets, from the sun outwards, have their
introduces the superiores vertices. The por- own lights, and this reaffirms the supra-so-
tion of the translated passage beginning lar positions of Mercury and Venus. Like the
with the vertices and ending with the phrase other planets, Venus and Mercury have
liberius apparet is clearly an observational their own lights. Thus Venus and Mercury,
account. Macrobius' recasting at this point being quite unlike the moon in quality, are
of the information found in Pliny tells us also quite unlike the moon in location. Only
that Mercury and Venus observed only near the moon is positioned below the sun. If the
perigee are incorrectly interpreted by some foregoing statements and arguments are not
(Chaldeans) to be in orbits actually below enough, we have, finally, Macrobius' simple
the sun, since these planets are more easily and unambiguous description (at I, xxi, 27)
visible at perigee.97 of what he calls the "order which Plato as-
signed to the spheres." It places the moon
first, then the sun, next Mercury, fourth
96
Pliny, N. H., II, 74 (Jan-Mayhoff 150, 21). Venus, Mars beyond, then Jupiter, finally
97
It is ironical that Macrobius is able to use Pliny
to discredit a planetary order supported by Pliny. With
Saturn.98While this is not the correct Pla-
reference to the same part of Macrobius' argument, tonic order-it stems from Eratosthenes99--
Stahl (transl.), Commentary,p. 250, says, "Macrobius it is explicit and it explicitly locates Mercury
is undoubtedly alluding to the Herclidean theory in this
statement, but either he himself does not understand
the theory, or he is purposely vague in referring to it."
Stahl's approach to the passage is not defensible, and be credited with recognizing that Macrobius endorsed
it might be said that his understanding of Macrobius a fixed order.
at this point is as confused or vague as he claims Ma- 98Jan ed., p. 124.
crobius' account to be. On the other hand, Stahl should 99See above, n. 88.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
390 BRUCE STANSFIELD EASTWOOD

and Venus beyond the sun, which Plato did doubt it. On the basis of Kepler's quotations
also. Turning to Book II of Macrobius' Com- from Macrobius, the text of the Leiden 1597
mentary, one will find the supra-solar place edition seems not to have been one of Ke-
of Mercury and Venus once more, and this pler's sources, at least when he sat down to
time in the correct "Platonic" order of copy the text of Macrobius. Other printings,
Moon-Sun-Venus-Mercury, and so forth.1?? without the annotation on Martianus, are
All in all, a careful reader of Macrobius will closer to the text given in Kepler's "Contra
find certain confusions but we can now see Ursum."103What is quite certain is that the
that there is unequivocally no allusion to a edition(s) used by Kepler present(s) no sig-
heliocentric arrangement for the planets nificant variants, that is, no variants in
Mercury and Venus. meaning, when compared to the passage as
How does Kepler find a heliocentric pat- given above. Equally certain is that every
tern in Macrobius' passage? Prima facie it edition down to 1600 presents a variant in
seems that he knows in advance what he the final section, dealing with the reflected
wishes or expects to find. Collateral evi- light of the moon, which Kepler does not
dence helps frame his opportunities. On the quote.104 This variant, while not fully ob-
one hand, none of the earlier references in scuring the argument, changes the last sen-
the sixteenth century to the system of Mar- tence, which must, according to the six-
tianus makes any connective reference to teenth-century editions, be read as follows:
Macrobius, nor do seventeenth-century ref- "And so because one part of the whole uni-
erences such as Riccioli, Argoli, Sherburne, verse is the earth, and the moon is one part
and Cellarius.?10 On the other hand, the of the aether, they call the moon also an
editions of Macrobius begin to refer to Mar- earth but an aethereal one."105The reading
tianus Capella in Kepler's time. Of thirty- makes little sense; its function in its more
three separate printings before 1600 of Ma- correct form is as a final comment and a
crobius' Commentaryon the Dream of Scipio transition to the next thought in Macrobius'
there is one, that of Leiden, 1597, which
adds a note that Macrobius' statement con- 103
See the discussion above in n. 86.
cerning the orbits (circuli) of Sun- Mercury- 104
Uniformly the editions to 1600 all read "una" for
Venus is equivalent to Copernicus' under- "ima" at two points, making the argument slightly less
standing of Martianus Capella.102It would clear. The sixteenth-century variant of the passage at
this place, beginning where Kepler ends, reads as fol-
be convenient to add that Kepler saw this lows: "quem praeter indaginem visus haec quoque ratio
annotation, but there is some reason to commendat, quod lunam, quae luce propria caret et de
sole mutuatur, necesse est fonti luminis sui esse su-
biectam. Haec enim ratio facit lunam non habere lumen
proprium, ceteras omnes stellas lucere suo, quod illae
100
II, iii, 14 (ed. Jan, p. 148); see also, n. 90 above. supra solem locatae in ipso purissimo aether sunt, in
101The first three are
given above, n. 26. The fourth quo omne quicquid est, lux naturalis et sua est, quae
is Andreas Cellarius, whose Harmoniamacrocosmicaseu tota cum igne suo ita sphaerae solis incumbit, ut coeli
Atlas universalis et novus (Amsterdam: loannes lanson- zonae, quae procul a sole sunt, perpetuo frigore op-
ius, 1661), contains 29 fascinating plates with descrip- pressae sint, sicut infra ostendetur, luna vero, quia sola
tion, plus an even more intriguing review of cosmology ipsa sub sole est et caducorum iam regioni luce sua
and a history of the "progress" of mathematics and carenti proxima, lucem nisi desuper posito sole, cui res-
astronomy since their beginnings (preface of 124 pp.). plendet, habere non potuit, denique quia totius mundi
His preface p. 65 refers to Argoli and the system of una pars terra est, aetheris autem una pars luna est,
Martianus. lunam quoque terram sed aetheream vocaverunt." The
102 The printing of interest here is Macrobius, Opera, two emphasized words pose the only noteworthy vari-
et Saturnaliorumlibros, ed. Johannes Isacius Pontanus, ants. Inter alia, this reading can be found in the pre-
annot. I. Meursius (Leiden: Fr. Raphelengius, 1597), in viously cited edition of Leiden, 1597, pp. 73-74.
which Isacius' note to I, xix, appears on p. 680; the 105 The full sixteenth-century text is above in n. 104.

relevant text (as translated above in this study) appears See the translation, which is changed by this variant,
on pp. 72-74, the exact section on p. 73. above at n. 89 in my text.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 391

Commentary,of a very different nature than Mercury's epicycle. Redirected in this way
the preceding. by Kepler, a reader is likely to say, "Of
How Kepler initially came to the idea that course!" and see that Macrobius has, after
Macrobius offers a heliocentric schema106is all, an awareness of the right idea! More
not clear, but we can see that he ignores important to Kepler is the connection Ma-
Macrobius' argument for the moon as the crobius provides with Plato. If Macrobius
only subsolar planet. We have already seen can be made to reveal a heliocentric idea,
that his interpretation of the heart of the there is then more strength in the specula-
Macrobius passage, on the appearances of tion that Plato's intention-it was certainly
Mercury and Venus, involves three distinct not explicit in Plato-was to have the sun's
modifications.107 He insists on an epicyclic companions understood as solar satellites.
interpretation, first. This is, strategically, his With his conclusions from Macrobius, Ke-
cardinal error-or, perhaps, device. Once pler seems to believe there is just such a case
the statement of Macrobius about inferior for a Platonic pedigree.
and superior locations of the two planets is
read theoretically rather than observation- II. f. VETUS COMMENTARIUS
ally, Kepler's path is the only reasonable one
Kepler's last source for enlightenment on
to take. We need not attribute to him a de-
Plato is the writing identified as the vetus
ceitful intent but only a biased direction.
commentarius on Bede. The citation makes
Already looking for heliocentric patterns, he it clear that this vetus commentariususes the
is inclined to such a theoretical rather than
commentary of Calcidius on Plato's Ti-
observational interpretation of Macrobius'
maeus, but Kepler does not recognize and
words. Thus his other two modifications,
identify the work that way.'09 He develops
interior replacing inferiorand a usefully am-
biguous pronoun (hic) inserted,108 assist the 109
Kepler, 1: 274, refers to an edition of Koln, 1537.
direction of Kepler's thought. Interior is a This identifies the book he uses as Bede, Opuscula cum-
less ambiguous and more theoretical term, plura de temporum ratione diligenter castigata, atque il-
not lending itself as well as inferior to an lustrata veteribus quibusdam annotationibus una cum
scholiis in obscuriores aliquot locos, aucthore Johanne
observational account. The ambiguity of the Noviomago (Koln: J. Prael, 1537). The source identified
pronoun is useful, for it suggests to the here as vetus commentariusis not used in commenting
knowledgeable reader that either the sun on Bede's De temporumratione but is used for two chap-
ters of Bede's De natura rerum (cc. 13, 14); the com-
itself or the epicycle of the sun (given Ke- mentary appears at if. 6v-7r, 8r-10r, and a series of
pler's epicyclic reinterpretation) is within diagrams with the commentary are clearly derivable
from cc. 80, 81, 85, 88, 90, 92, 111, 112 of Calcidius'
commentary on Plato's Timaeus, though often misla-
106
The illustrations in the fifteenth- and sixteenth- beled and/or misconstrued in the 1537 printing. No-
century editions are equally unhelpful. In relation to viomagus' edition is one of three used in the edition
Macrobius, Commentary,I, xxi, 24-27 (ed. Jan, p. 123- of J. P. Migne, Patrologia latina, vol. 90, esp. coll. 215-
124), which notes the placement at the birth of the 229. In the Migne edition the figures included with the
world of each planet in a zodiacal sign and the order vetus commentarius are Calcidian (coll. 219-226) and
in which the planets were originally set, the various Plinian (coll. 227-229). The Calcidian diagrams, with
printings give either the Ptolemaic order (Earth-Moon- some modification, appear in Calcidius' commentary,
Mercury-Venus-Sun-Mars-Jupiter-Satum) or no illus- ed. Waszink, pp. 129-162. The three Plinian diagrams,
tration at all. These illustrations appear despite the which are not to be found in the 1537 vetus commen-
uniform statement in the text of Macrobius' version of tarius, appear respectively in the following manuscript
Plato's system: Earth-Moon-Sun-Mercury-Venus-Mars- locations: (1) Erfurt Ampl. 4?. 351, f. lv (this is an
Jupiter-Saturn. Of the sixteenth-century editions, only approximate diagram; I have not yet found an identical
that of Leiden 1597 does not carry a diagram for the copy); (2) Madrid 3307, f. 65v; (3) Madrid 3307, f. 66r.
order of the planets. If one reads the last lines of Calcidius' commentary,
107 Translation of
Kepler's version given above at c. 116, and compares the lines of the vetus commentarius
n. 84. (PL 90, coll. 227-229), it will become apparent how the
108
Kepler, Opera, ed. Frisch, 1: 273-274. three Pliny diagrams have been interpolated into the

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
392 BRUCE STANSFIELD EASTWOOD

a somewhat rhetorical flourish here, prelim- represent exactly what Martianus intended,
inary to his retaking the field directly against that is, the epicycle of Venus rotating with
Ursus. Kepler argues that within the vetus the sun at the center of the epicycle, al-
commentarius one can find justification for though this diagram has been corrupted and
locating the epicycle of Venus in any one requires correction, which Kepler obligingly
of three positions. If we take the meaning provides."12 Here, says Kepler, "nearer to
of one limited part of this commentary, viz. the sun" means visually closer, and "nearer
"nearer to the sun it [Venus] seems higher, to the earth" refers to the distance between
nearer to the earth it seems lower,"110 we centers. Finally, if we look at the left-hand
can understand the commentator to have diagram of the same pair, we see that, while
placed the epicycle of Venus completely it may be in error, it appears to represent
below the sun. Again, if we look at the right- the understanding of those (unnamed) who
hand diagram of two appearing in the vetus place the epicycle of Venus entirely above
commentarius,111we can understand it to the sun.1"3And so in reading the vetus com-
mentarius to Bede it is possible to discover
text; the first diagram (incomplete) gives zodiacal lati- the sun above, within, or below the epicycle
tudes for Venus, Mercury, Sun, Moon; the second gives of Venus. It is the version according to Mar-
apsides, which are supposed to locate zodiacally the
planetary retrogradations; the third gives again zodiacal
tianus, thereby in accord with Plato, that we
latitudes, for all planets except Mercury and Saturn,
whose names appear but without paths drawn in.
Charles W. Jones, Bedae Pseudepigrapha:Scientific Writ- The text for the two diagrams, taken from Calcidius'
ing Falsely Attributed to Bede (Ithaca: Cornell U.P., commentary (cc. 111-112), reads as follows in the 1537
1939), p. 10, discusses the sources of the vetus com- edition, f. 9v. "Ad solis et Veneris demonstrationem,
mentarius but does not note the extensive extract from erit una linea directa ex terra medietate solem demon-
Calcidius. strans a litera x. duae vero aliae dextera levaque nih-
l0oKepler, 1: 274. Kepler cites the phrases, "proxi-
ilominus directae lineae, a sole quidem distantes L.
mum Soli videri excelsiorem, proximum Terrae humi- momentis a se autem invicem C. dextera quidem a parte
liorem," which appear in the 1537 Koln edition, f. 9v; orientis per x et A. leva vero ab occidente per x et r.
along with a preceding quotation, this comes from zodiacus quoque circulus sit ABr. quae singulae distant
Calcidius' commentary (c. 112) on Timaeus; see ed. a se momentis, L. et cum sol sit in B, et per XB. lineam,
Waszink, p. 159, 2-4. Kepler is also making tacit use sit punctus solis in litera K. haec, id est XF linea prius
here of Macrobius' Comm. in Somnium, I, xix, 6-7. occidit et prius oritur quia sol, illa vero alia xA. posterius
1ll Noviomagus ed. (K6oln,1537), f. lOr, reproduces
occidit et posterius oritur. Necesse est igitur ut litera A.
erroneously two diagrams from Calcidius' commen- demonstret hesperum post solis occasum, r. vero lu-
tary. See ed. Waszink, p. 158 for the diagrams, and ciferum ante solis ortum. At vero Plato et alii, aliquanto
pp. 157-159 (cc. 111-112) for the relevant texts; ed. quam solis est, elatiorem Luciferi globum astruunt, qui
Noviomagus, f. 9v for the same texts. The diagram limitatur, AEZH. contingens, A. quidem lineam per E.
supposed by Kepler to represent Martianus is as follows litteram KF. vero per H. Cum ergo fuerit in E. Lucifer
in the 1537 edition (see also PL 90, col. 225, lower videtur esse in A. et cum in H. putabitur esse in E[r].
figure): cum vero penes A. dubium non est proximum soli videri
B excelsiorem, et cum in Z. proximum terrae humiliorem.
lam illud observandum quod sive ad orientem, sive ad
occidentem Lucifer secesserit diebus fere D. LXXXIIII.
ad id in quo fuerat remeare pridem, et HAE. quidem
peragrat diebus CCCCXLV. ut maiorem ambitum, mi-
norem vero depressioremque reliquis diebus CXXXIII."
112 Kepler's corrected version, fig. 7 in Kepler, I, 274,
has the sun fixed at a point k in the center of Venus'
epicycle. This diagram (Kepler's corrected version) can
be found with different letters at f. 8r of the 1537
edition as part of Noviomagus' scholia, not as part of
the vetus commentarius;it derives from Calcidius' com-
mentary (c. 85), ed. Waszink p. 136. See also PL 90,
col. 220, lower figure.
113
Noviomagus ed. (Koln, 1537), f. lOr, reproduced
here; the version in PL 90, col. 226, upper figure, has
been corrected. The original is in Calcidius, commen-
tary, ed. Waszink, p. 158.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
 KEPLERAS HISTORIAN OF SCIENCE 393

are to understand as the most useful reading Kepler's insistence on uncovering distinct
here. All these versions, not of computa- and definable world-systems in each of the
tional systems but of formae mundi, forms authors from Copernicus backwards in the
of the physical world system, are different Platonic line of thought. While the case of
from the Tychonic hypotheses. But Ursus, the Bede commentary is hopelessly distorted
says Kepler, has paid no attention to all this by Kepler's willingness to use the Calcidius
in charging Tycho with having lifted his material as evidence for world-systems as
hypotheses from one or another earlier au- well as for calculating systems,114 the cases
thor. of Pliny and Macrobius show rather directly
what Kepler is about. A severe and careful
III. CONCLUDING REMARKS
reading of these two sources will reveal that
neither has a clear world-system to offer. In
What is the character of Kepler's histor- fact, both start with phenomena and then
ical scholarship in his account of the frame- give hints (Macrobius many, Pliny few) that
work for understanding Copernicus' refer- some kind of physical system lies behind the
ence to precursors? Kepler casts his discus- phenomena. But neither can be made to
sion in the mold of an interpretive essay on reveal the exact nature of that system by
a Platonic view of planetary motion. Ex- anything less than historical misrepresen-
plicitly the interpretation involves a com- tation. Kepler makes historical misrepresen-
mitment to a certain relationship of the tations in order to find a train of thought-
planets Venus-Mercury-Sun in their motion a Platonic line of thought-in his sources.
around the earth. This relationship is said He appears to identify a series of more or
to be understood in some sense by a line of less right-minded thinkers, who could pre-
later authors: Vitruvius, Pliny, Martianus, sumably be set against those with less ac-
Macrobius, a commentator on Bede, and, of ceptable astronomical hypotheses. It might
course, Copernicus. Implicitly Kepler's in- be added that Kepler can hardly give a full
terpretation follows the conception that coup de grace to Ursus' claims of plagiarism,
each of these authors back to Plato-the unless it can be shown just what the earlier
Platonic line of thought-is essentially fo- hypotheses were in order to point out that
cused on the form of the world, so that geo-
metrical equivalents to such forms do not
adequately explain what they are about. 114
The text of Bede is essentially phenomenist and
With this in mind we can find reason for based on Pliny at this point. While Pliny does indeed
have a sort of world-system in mind, this chapter in
Bede and its source in Pliny present a calculational
device for locating apsides of the planets individually,
with no relationship between planets either in terms
of order, distance, or orbit. See Bede, DNR, c. 14; Pliny,
ed. Jan.-Mayhoff, p. 147, 7-17; also Wilhelm Kroll [and
Heinrich Vogt], Die Kosmologie des Plinius (Breslau:
Marcus, 1930), pp. 67-73. The freedom of represen-
tation of the theme is remarkable in the medieval
manuscripts extracting this and other sections of Pliny's
planetary doctrines. While the traditions for other mat-
ters, such as the order of the planets or the zodiacal
latitudes of the planets, find fairly consistent diagrams
to represent them, the tradition of diagrams to repre-
sent the apsides is highly divergent, going so far as to
distort or reorder the zodiac in some cases. In other
words, the issue under discussion, in the Bede chapter
where the vetus commentariusappears, does not relate
directly to any physical hypothesis; it is a thoroughly
calculational matter under discussion. Kepler bends the
discussion to import physical meaning.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions
394 BRUCE STANSFIELD EASTWOOD

Tycho's hypotheses were not inspired by For Kepler, the question is not, "Where
them. Insofar as this motive contributed to did Copernicus get the idea?," but rather,
Kepler's interpretations, we can consider his "Where did the idea originate?" Delving for
historical work to be essentially polemical, the roots of the correct theory, Kepler finds
even though rather well informed.115 it in diverse sources without claiming that
Copernicus knew all of them. For Kepler it
is sufficient that the new cosmography
115 After
completing this study I had the opportunity
to see the new edition, translation, and thorough study draws somehow on the long-known truth.
of Kepler's Apologia Tychonis contra Ursum (forthcom- If Copernicus knew Martianus and Pliny
ing publication) by Nicholas Jardine (University of ("and others among the Latins"), this is suf-
Cambridge). Dr. Jardine shows most convincingly that
Kepler's Apologia is especially concerned to refute the
ficient for Kepler, who obligingly completes
claim that Apollonius was the source of Tycho's hy- the pedigree of the idea.
potheses; the pedigree of Copernicus' reference in De
revolutionibus I, 10, is a subordinate concern. Jardine's
study illuminates nicely the new depths of historio-
graphical and theoretical awareness in Kepler's Apolo- far beyond the present study, which is explicitly limited
gia as compared to contemporary and earlier works on to Kepler's survey of pre-Copemican mentionings of
and in astronomy. The issues addressed by Jardine go the idea of circumsolar Mercury and Venus.

This content downloaded from 148.210.115.53 on Thu, 19 Mar 2015 17:20:03 UTC
All use subject to JSTOR Terms and Conditions