JHA, xlii (2011) PETRUS APIANUS DRAWS UP A CALENDAR ANTHONY GRAFTON, Princeton University The dullest page in Petrus Apianus’s magnificently illustrated Astronomicum Caesareum is folio K I verso (Figure 1). Under the title “Comparison of the Greek and Latin months” appears a partial calendar in two columns. The column on the left begins with 26 July and includes the rest of July, August, September and 1 October, in the form assigned them by the Julian ecclesiastical calendar, the standard calendar of the medieval and Renaissance west until the Gregorian Calendar reform of the calendar decreed in February 1582. The column on the right begins with the first day of Hecatombaion, the first month of the calendar of ancient Athens. It continues through Metageitnion, the second month, and the beginning of Boedromion, the third. Apianus treats these months, oddly, as if they too were Roman, as a comparison with the left-hand column shows. He makes the first day of each Athenian month the Kalends. Then he counts down to the Ides, which he sets on the fifth day; to the Nones, which he sets on the 13th of Hecatombaion and the 15th of Metageitnion; and finally, after the Nones, to the Kalends of the next month. Though Apianus supplies no detailed explanation, he does offer supplementary materials on the right: a list of the Athenian months in order, a list of the Athenian names for the days of the lunar month (which was divided quite differently from the Roman month, into three sections of almost equal length), and a couple of historical references to Nicias — the leader of the Athenian expedition to Syracuse during the Peloponnesian War, whose defeat in 413 B.C. Thucydides narrates in the magnificent, climactic books of his History.1 Both subject matter and presentation seem dry enough to eliminate the need for caucus races: dry enough, indeed, to induce slumber in the normal reader. The table contrasts sharply with many of the splendid and original visual effects achieved by Apianus elsewhere in his book. The Astronomicum, after all, included gorgeous diagrams of planetary models, splendid images of eclipses, and ingenious efforts to record the changing appearance of Halley’s comet and four others, day by day. The comet images gave Apianus’s work lasting value as a source of data for astronomers — including Halley — long after Kepler denounced it as a fanciful rechauffé of Ptolemy.2 Apianus, after all, was not only a humanist, an antiquary, and a printer of astronomical textbooks and tables, but also a man with a keen visual sensibility. A skilled observer of the heavens, he instructed readers on how to look safely at eclipses through two thick pieces of glass of different colours, with a thin sheet of paper glued between then and perforated at one point, and accurately noted that comets’ tails are normally directed away from the Sun. No wonder, then, that his book seemed a marvel of design even in the decade that saw the appearance of Fuchs’s Herbal and Vesalius’s Fabrica.3 In the brilliant company of Apianus’s volvelles and eclipse diagrams, carefully coloured in his shop before the book was distributed to buyers, the calendar table seems a verbal wallflower. Apianus himself, however, saw it as 0021-8286/11/4201-0055/$10.00 © 2011 Science History Publications Ltd 56 Anthony Grafton FIG. 1. Petrus Apianus, Athenian and Roman calendars for the end of the Sicilian Expedition and the defeat of Nicias, Apianus, Astronomicum Caesareum (Ingolstadt: Petrus Apianus, 1540), KI verso. Copyright the Master and Fellows of Peterhouse, Cambridge. Petrus Apianus 57 a particularly instructive visual tool: “With the aid of this journal”, he told readers, “your own eyes can teach you everything about the months and the beginning of the Attic year, more clearly than light itself”.4 What then did the Comparison teach, and how, and why did Apianus think it mattered? And what does it tell us about the uses of illustrations in astronomical and other literature of this period? The Roman part of the diagram — the left-hand column — is a standard, if rather partial, treatment of the Julian ecclesiastical calendar, as that took shape in the fifth and sixth centuries and continued to develop, in its sacred components, over time. Apianus lays out the data in parallel columns. They include: the names of the months, a repetitive series of letters from A to G, and the Roman dates for the days in question. The letters are the so-called Dominical or weekday letters, which appear beside the days in the Julian ecclesiastical calendar. Each year is denominated by the letter on which its first Sunday falls. Because the year is 365 days long and 365/7 = 52 rem 1, the calendar should move smoothly through a cycle of seven years in which the Dominical letter of any given year is one later than that of its predecessor. But the intercalary day that extends every fourth year interrupts the cycle. Each leap year has two Dominical letters, one for the period before the end of February and one for the period after, and the whole cycle takes 7 × 4 or 28 years to complete itself. To find the Dominical letter of a given year in the Christian era, one adds 9 and divides by 28 — or looks it up in a calendar. Once in possession of a year’s Dominical letter, one can work out on which day of the week any given date in the Julian ecclesiastical calendar will fall for the year in question. It is possible, in other words, to tie the abstract year of the calendar to a particular year, in the future, the present, or the past, and literally see the year pass, week by week and saint’s day by saint’s day. All of these facts had been well understood since the early Middle Ages, and all of these data appeared in any standard calendar designed for use in the present (for example, that of Regiomontanus). Like many astronomers, Apianus built up his reputation and, perhaps, his astrological clientele by issuing wall calendars that included all of this information for a year at a time — along with saints’ days, new and full moons and other relevant information, sacred and astronomical alike. Though only a few fragments and one incomplete calendar by Apianus survive, the evidence shows that he and his brother and son produced these from the 1520s until the mid-1550s — the same period in which he and Gemma Frisius devised the series of cosmographical works that bore his name to an immense readership. These books, too, contained calendrical material. No wonder, then, that Apianus found it easy to apply this form in the Astronomicum. The tradition of the printed calendar, as founded by Regiomontanus, included astronomical matter and illustrations as well as standard calendrical tables, as we will see in due course. This precedent no doubt helped Apianus to make the graphical transition that he carried out in the Astronomicum, where he used the conventions of one semiastronomical form to enlarge and enrich the subject matter of another. And that helps us to begin to place Apianus’s efforts at the visual presentation of data into a context. In a classic article on the uses of images in astronomy and anatomy, Martin Kemp 58 Anthony Grafton maintained that “for astronomers in the Renaissance, the fundamental processes of representation do not seem to have been essentially different from those of Ptolemy or his Islamic successors”.5 Where Vesalius and other anatomists transformed their subject by their ability to marshal visual evidence, astronomers went on drawing up tables and diagrams in a form that Ptolemy would have found immediately familiar, at least until Copernicus forced them to think in more holistic, aesthetic terms. More recently, however, Isabelle Pantin has sketched a more complex history. She concedes that the tables and diagrams that played the central visual role in formal presentations of planetary theory retained their traditional forms even after De revolutionibus appeared. But she also notes that in other areas of the tradition — such as the lively, hybrid world of textbooks — considerable visual experimentation took place. Three-dimensional images of the planetary spheres, for example, appeared alongside the old two-dimensional ones.6 Apianus produced plenty of textbooks, and that may help to explain his willingness to meld traditions. In the Astronomicum, though, he drew up a hypothetical calendar to enrich a massive, ambitious work on astronomy — a project on a grander scale than those identified by Pantin as the primary loci of visual innovation. Apianus, moreover, laid out the Julian calendar for a particular period long before the Julian reform itself: the part of the year in which Nicias was defeated and the Athenian expedition to Syracuse came to its bitter end (in our terms, though not in his, the late summer of 413 B.C.). He did so, moreover — or so at least his comparative table claimed — with absolute precision, day by day. The reader could now look up the exact Julian equivalents, down to the days of the week, for the Greek dates, transmitted by the ancient sources, on which Nicias’s eclipse and then his defeat and capture took place: in this case, the night of 8–9 September, and 20 September. Apianus clearly believed that in accomplishing this he was achieving something remarkable: hence his praise for his own table. To achieve this novel level of precision about the past, Apianus had to attain two subsidiary goals. The first of these was the one most obviously germane to his current enterprise: he had to fix a Julian date for the defeat of Nicias. And the way he did this was typical of his method. Both Plutarch and Thucydides noted that Nicias, after suffering defeat in the harbour at Syracuse, decided to leave Sicily. Just then a lunar eclipse took place. Nicias and his followers, who did not see lunar eclipses as regular phenomena like solar ones, thought this one an ill omen. As Plutarch explained, But just as everything was prepared for this and none of the enemy were on the watch, since they did not expect the move at all, there came an eclipse of the Moon by night. This was a great terror to Nicias and all those who were ignorant or superstitious enough to quake at such a sight. The obscuration of the Sun towards the end of the month was already understood, even by the common folk as caused somehow or other by the Moon; but what it was that the Moon encountered, and how, being at the full, she should on a sudden lose her light and emit all sorts of colours, this was no easy thing to comprehend (Nicias 23.1–2, transl. B. Perrin).7 Petrus Apianus 59 So the Athenians stayed where they were and suffered total defeat. In an interesting digression, Plutarch noted that Anaxagoras had already explained the mechanism of lunar eclipses, though his mechanistic explanations had alienated many Greeks. But this sort of knowledge — and the naturalistic explanations of apparent omens that it made possible — remained confined to a few philosophers.8 Plutarch’s history of astronomy was not quite accurate, but his larger point had merit. Though the 19-year luni-solar cycle and the correlation between full moons and lunar eclipses had been well known in Babylon for some centuries, it seems that Meton of Athens was the first to describe the cycle in Greek, exactly in this period, and that the mechanism of lunar eclipses came to be generally understood in Greece in the late fifth and early fourth centuries B.C.9 For Apianus — and here he seems to have been a pioneer — this account and similar ones seemed charged with an extraordinary potential. The Astronomicum Caesareum included a long section on eclipses.10 Apianus provided volvelles with which the reader could reckon the Julian dates of solar and lunar eclipses, before and after the birth of Jesus. He offered precise dates and times for a number of eclipses that had taken place in the last few decades, above all those associated with the lives of Charles V and Ferdinand I. And he made his accounts of these eclipses vivid by drawing up illustrations that laid out, in schematic form, the course, duration and degree of fullness of the eclipses in question. In doing so he tied his work — and his own experience as an observer of eclipses — to the tradition established by Peurbach at Vienna and carried on after him by Regiomontanus, Schöner and others. These men observed eclipses and recorded what they saw in great detail. They also adapted the traditional iconography of eclipses — an iconography that went back through the western Middle Ages into the Islamic world before that — to new ends. A combination of two colours, one light and one dark, had long served in eclipse diagrams to make clear exactly how the Earth’s shadow cone could swallow the Moon at full moon and how the Moon could come between the Earth and the Sun at new moon.11 In his calendar, Regiomontanus provided such images for eclipses of the Sun and Moon that would take place from 1474 to 1530 — along with their dates and timing. Schöner emulated him. These images fascinated readers, who copied them and reused them: thus Sebastian Münster inserted Schöner’s images of solar eclipses that would take place from 1526 to 1583 into his own treatment of the Hebrew calendar, which appeared in 1527.12 The pullulating imagery of eclipses enriched textbooks and almanacs with a new set of images of the celestial dramas that would take place during the lifetime of each almanac’s owner and after. When Apianus included them in the Astronomicum, he underlined what Pantin might call the hybrid nature of his work. In this case, as in that of the partial calendar for 414 B.C., he fused the contents and conventions of traditional models of planetary theory with materials more normally found in almanacs. Yet Apianus’s ambitions went far beyond the traditional goals of almanac making and, more generally, of astronomy. For he defined his work against another technical genre of great elegance in design — one located entirely outside the world of 60 Anthony Grafton astronomy. Since late Antiquity, when Eusebius and others had created the tradition of Christian chronology, historians interested in establishing the dates of events in Antiquity had relied on comparative tables of the sort Eusebius seems to have invented. Tabular chronologies lined up important events in the different ancient states. They synchronized the lives of rulers in dynasties that had ruled at the same time and dated events in intervals of years from the time when those dynasties began (dates in Assyrian history, for example, were reckoned in intervals of whole years from the accession of Ninus, the first ruler of Assyria, not long after the Flood). The Chronicle of Eusebius, in the Latin edition by Jerome, tabulated the histories of nineteen different peoples. But it also showed the hand of Providence at work in the past. Kingdom after kingdom, city after city, rose and fell. At the end, however, all the other kingdoms funnelled down and disappeared. Only Rome and Israel remained, as universal monarchy made the world accessible to the Jewish-born Messiah of Christianity: and soon, after the fall of Jerusalem in 70 A.D., there was only Rome. The result was both a meticulous display of complex information and a dazzling statement of its meaning — one that contemporary and later readers appreciated, as nineteenth- and twentieth-century readers appreciated Charles Minard’s famous graphical representation of Napoleon’s invasion of Russia and its impact on his army.13 As the Roman aristocrat and scholar Cassiodorus commented, in the sixth century, “Eusebius composed, in Greek, a Chronicle: that is, an image of history and a condensed record of the past”.14 The Latin Chronicle of Eusebius and Jerome served as a desk reference through the Middle Ages and beyond, its afterlife extended by the varied continuators who were still adding supplements to bring it up to date as late as the fifteenth century. It remained a central source for humanists from Petrarch, who annotated his copy heavily, to Polydore Vergil, who pillaged it for the information that he deployed in a different form in his book on inventors. Printers seized on it fairly early, overcame the difficulties involved in reproducing it, and added to it not only further continuations, but also indexes that made it much easier to consult. Renaissance world chroniclers from Werner Rolewinck to Paulus Constantinus Phrygio borrowed its contents and emulated its form. Printed editions of the Latin Chronicle encased it in a still larger system: they synchronized the dates given by Eusebius himself, which he reckoned only from the Flood, with a system of dates from the Creation. It all looked complete, perfect, a closed system: an image, as Cassiodorus had said, of history itself.15 From the standpoint of an astronomer — someone used to working with intervals dated in precise sums of Egyptian years, months, and days, from equally precise eras — a serious problem remained. All the dates Eusebius gave, as Apianus complained, were approximate. The chronologers reckoned only in whole years, not in months or days. Worse still, they dated every event from a previous event, the date of which itself could be established only in approximate terms. If any mooring point proved loose or any interval approximate, all dates in a given chronicle could be called into question. These problems were apparent not only to Apianus, but to any astronomer who took an interest in connecting astronomical data with historical events — a Petrus Apianus 61 subject of serious technical interest to Copernicus, but also to a great many less original astrologers.16 From the beginning of the Astronomicum, Apianus insisted that his work offered the basis for a radical transformation of historical scholarship and its practices. If historians attended to the dateable eclipses mentioned in their sources, he argued, they could attain a completely novel level of precision — one that scholarship had never reached before. For his work offered the only way to escape the imprecision that dogged all traditional chronologies, based as they were on lists that skimped on vital facts: To show you the extraordinary knowledge, excellence and utility that eclipses possess, I found myself compelled to offer some examples from [the time] before Christ. These will reveal the utility of eclipses more certainly than later examples. Everyone knows that historians go wildly wrong when they state periods of years from the creation and the founding of Rome. The learned can judge how much damage the obscurity, ignorance, doubt and confusion introduced by this mistake have wrought to all efforts to assess and understand history. For the result is that if they inform the Christian reader about some event that either preceded or fell not long after the founding of Rome, they cannot give a consistent account. Only knowledge of eclipses can correct this great evil and improve matters. For eclipses can make it possible to fix all events to particular years, before Christ just as much as after him. Once the historian has obtained the unquestionable date for some event, he can infer the other intervals, both those before and those after it.17 Apianus boasted more than once of the importance of this methodological innovation: Not the least important application of my work, I think, is that fact that I have shown how the errors in histories can be corrected — something that no one before me has ever said or perceived. Anyone who considers how important orderly history is in human affairs will see how much this matters.18 To support this argument he included in his work detailed treatments not only of present and future eclipses, but of past ones. Apianus started (Figure 2) with a strategically chosen event: Alexander’s defeat of Darius at Gaugamela. All good Christians knew Nebuchadnezzar’s vision of a great statue, as the prophet Daniel related it to the king: Thou, O king, sawest, and behold a great image. This great image, whose brightness [was] excellent, stood before thee; and the form thereof [was] terrible. This image’s head [was] of fine gold, his breast and his arms of silver, his belly and his thighs of brass, his legs of iron, his feet part of iron and part of clay. Thou sawest till that a stone was cut out without hands, which smote the image upon his feet [that were] of iron and clay, and brake them to pieces. Then was the iron, the clay, the brass, the silver, and the gold, broken to pieces together, and became 62 Anthony Grafton FIG. 2. Petrus Apianus, representation of the lunar eclipse of Gaugamela, Apianus, Astronomicum Caesareum (Ingolstadt: Petrus Apianus, 1540), I, iii, verso. Copyright the Master and Fellows of Peterhouse, Cambridge. Petrus Apianus 63 like the chaff of the summer threshingfloors; and the wind carried them away, that no place was found for them: and the stone that smote the image became a great mountain, and filled the whole earth (Daniel 2:31–35, KJV). Daniel explained to the king that the statue represented the past and the future, down to the time, in the future, when the Lord would create a final kingdom, never to be destroyed. Standard exegesis of the passage held that four, and only four, empires would exist between the Flood and the coming of the Messiah: the Assyrian, the Medo-Persian, the Greek (or Macedonian), and the Roman (the last of which, all subjects of Charles V knew, still existed). Treatments of world history from textbooks to Albrecht Altdorfer’s magnificent painting of Alexander’s victory over Darius embodied this ordering of time. They dramatized the defeat of Persia by Macedon as the hinge of history: halftime in the great narrative that stretched from the Flood to the Millennium. But even this crucial event, as Apianus complained, they dated qualitatively, in terms of an interval of years from another approximate date. Eusebius, for example, set it in the year 4871 after the Creation, which took place in 5200 B.C. — or 328 B.C. Plutarch, by contrast, tied the earthly event to a celestial one. He noted with happy precision that Alexander defeated the Persians 11 days after a total lunar eclipse, which took place in the month of Boedromion, around the beginning of the Mysteries at Athens (Alexander 31.8). Boedromion, the third month of the Athenian year, fell at some point in the warm part of the year. Apianus examined the possible full moons and discovered that the only one that could have experienced the eclipse in question actually fell on 28 June 326 B.C. Suddenly, he could fix the change of empires not only to a year, but to a month and day. History took on the precision of astronomy. Apianus was not the first astronomer to reckon the date of a historical eclipse. In the first century B.C., the antiquary Varro asked Lucius Tarrutius of Firmum, an Etruscan diviner and astrologer, to use astrological evidence to fix the true date of the founding of Rome. Tarrutius worked out that Romulus had been conceived during a total solar eclipse that took place on 23 Choiak of Olympiad 2, 1 — or 24 June 772 B.C. He went on to determine the date of Romulus’s birth and that of the foundation of the city as well by a combination of astrological inference and astronomical computation.19 Legions of later astrologers did their best to correlate the births of princes and the founding of states to conjunctions and eclipses.20 It seems likely that Apianus shared with his ancient and medieval predecessors the view that the eclipses that accompanied such great events, though predictable, were still ominous. True, Apianus argued that eclipses mattered to historians for a different reason: the greater precision they could afford. That was the spirit in which he decided to challenge Eusebius, who dated the defeat of Nicias to the year of the world 4782 or 417 B.C.21 On this road too, though, Apianus was not the first traveller.22 Paul of Middelburg, the Dutch-born bishop of Fossombrone in the Italian Marche, had included a strikingly similar set of arguments in his Paulina, a long work on chronology and calendars written in preparation for the Fifth Lateran Council. He set out to move the 64 Anthony Grafton reign of Augustus himself, basing his argument on what he took to be a firmly dated eclipse associated with the emperor — though misdated — by a reliable historian. The chronicle of Eusebius and Jerome notes at year 56 of Augustus’s reign that “a solar eclipse took place and Augustus died at age 76 at Atella, in Campania, and he was buried in the campus Martius at Rome”.23 If Eusebius and Jerome were correct, the eclipse would have taken place in the last year of Augustus’s reign, 13–14 A.D. But as Paul noted, If we wish to work backwards according to the usual computation of years of our Lord to these past times, we will find that no eclipse of the Sun took place at Rome in the year where the aforesaid chronologer mentioned that Augustus died, nor we will find that any eclipse took place in the following year. Nor was there any remarkable eclipse in the preceding years. Only in the third year after this one, that is, in the year of our Lord, following the normal computation, 17, will we find that a really massive eclipse of the Sun took place, in which the Sun was visibly obscured at Rome and in many parts of Italy, on 15 February, around noon.24 Paul believed that eclipse dating had a special certainty. Eclipse computation, he thought, usually yielded fairly clear results. And the identification of the eclipses mentioned by historians presented no great difficulties. “It is clear”, he explained, that Eusebius and the historians wrote only about this eclipse, since it was full, so much so that when the Sun approached the meridian it became dark, and in the years around it no noteworthy eclipse took place that deserved mention by the chronographers. For generally chronographers and historians record in their annals only those eclipses, in which the Sun is entirely obscured and hidden from us, so that in some cases the stars of the starry heaven can be seen and counted. These happen rather seldom.25 Paul’s dating provoked debate — rightly so, as his thesis, like the original notice in Eusebius’s Chronicle, was wrong.26 The solar eclipse that actually took place on 15 February 17 A.D. had nothing whatever to do with the death of Augustus, which it followed after an interval of almost three years. An examination of Suetonius’s life of Augustus — a source composed in the early second century, by a writer from the Roman equestrian order, and thus in a much better position than Eusebius or Jerome to know the facts — would have shown that the emperor’s death, though preceded by “evidentissimis ostentis” (97.1), was not presaged by a total eclipse of the Sun. Even in its weaknesses, though, as we will see, Paul’s project clearly had much in common with that of Apianus. Yet Apianus cast his historical net more widely than Paul had. He was, after all, an innovative student of the human past as well as of the sky. In 1534 he and a colleague at Ingolstadt, the historian Bartholomaeus Amantius, published the first sylloge, or collection, of ancient inscriptions to appear in northern Europe. The Inscriptiones sacrosanctae vetustatis appeared only 13 years after the first great printed collection, Petrus Apianus 65 the Epigrammata urbis, which Jacopo Mazzocchi had published in Rome. And in many ways it was more ambitious than Mazzocchi’s collection. The Italian had focused on Rome: the Germans offered other inscriptions (a fair number of them, unfortunately, forged), collected everywhere from Portugal to Africa. The Italian had concentrated on inscriptions and buildings: the Germans, emulating makers of manuscript collections, included ancient works of art as well, from the collection of their patron Raymund Fugger. Apianus’s technical interests manifested themselves more than once — as when he gave dimensions, as Mazzocchi had not, for the Vatican obelisk (he also denied that it had served as the resting-place for Caesar’s ashes).27 And he and Amantius clearly saw — as they noted a propos the last image in their collection, a depiction of a tattooed barbarian whom they could not identify — that modern readers liked to see the past depicted in the most vivid possible way, on the basis of ancient monuments, and that providing such images had a value of its own even when the iconography and context of the work of art in question remained elusive.28 The Astronomicum represented a complementary form of antiquarian scholarship, no longer based on material remains but still vividly visual in character. It enabled the reader to visualize not the costumes and customs of the ancients, but the sometimes-frightening night sky, as they themselves had seen it. Apianus’s enterprise really was a hybrid. Apianus’s second subsidiary goal lay in a related but not identical realm. He had to work out how the ancient Athenian and other calendars mentioned by his sources actually functioned. In one sense, of course, conversion between calendars formed an organic part of the astronomical tradition. From Ptolemy on, astronomers had generally dated celestial events from the accession of Nabonassar to the throne of Babylon on 26 February 747 B.C., or from the later era of Philip. They had dated them, moreover, in Egyptian years of exactly 365 days, which made computation simpler than Julian years, with their different lengths and intercalary days. And they had known how to work between this astronomical calendar and others. Any fifteenth- or sixteenth-century astronomer knew how to convert dates given in the Julian ecclesiastical calendar, the Egyptian calendar, or the Muslim lunar calendar.29 Problems arose, however, as soon as one tried identify and to work with the numerous other calendars mentioned in or relevant to other sources. Which calendar did the Hebrew text of Genesis 7:11 refer to when it stated that “In the six hundredth year of Noah’s life, in the second month, the seventeenth day of the month, the same day were all the fountains of the great deep broken up, and the windows of heaven were opened”? How did the calendar of ancient Athens — dates which appeared with some frequency in favourite texts like Plutarch’s Lives and Moralia and Aristotle’s History of animals — actually function? In the second century A.D., scholars still knew, or thought they knew, how the ancient Athenian calendar operated. The members of a cult or club organized by Herodes Atticus in honour of Dionysus met on the 8th of the Athenian month Elaphebolion, as a famous inscription records. Evidently the Athenian calendar still ran, along with the imperial one of Rome, which had long ruled Greece.30 Ptolemy or his sources cited astronomical observations by Athenian 66 Anthony Grafton month and day. Apparently some astronomers could convert Athenian dates into astronomical ones reckoned from Nabonassar. But the data they provided were and remain too skimpy to reveal the principles they followed.31 Astronomers, for all their proficiency in converting dates between the calendars they understood and used, did not, and could not, provide tables or rules for use with the Athenian calendar, as they could for the Egyptian calendar that they used in practice.32 How then could one know which Athenian month corresponded to which Julian one? The question mattered. The late Byzantine scholar George Gemisthos Pletho, who set out to revive the ancient gods of Greece, also set out to create an updated version of the calendar that regulated their worship.33 Western scholars who wrote letters to their friends in Greek — as many did — needed a way to date them.34 Marsilio Ficino, who wanted to celebrate the birthday of Plato, knew from Diogenes Laertius (3.2) that his hero was born on the 7th of Thargelion, the 11th month of the Athenian year. But when was Thargelion? Ancient sources gave some help. Many sources made clear that the Athenian calendar was lunar. Greek-Latin glossaries listed the Athenian months in order. A vital passage in Plato’s Laws (767C) showed — or so most scholars agreed — that the Athenian year normally began, in Hecatombaion, around the summer solstice, presumably late in June. But though a number of writers referred to an 8-year cycle, no ancient writer explained, in detail, the system of intercalation that the Athenians had actually used to regulate the connection between their lunar year and the tropical one. The possibilities for error were immense. Ficino, for example, simply assumed that the 12 Attic months corresponded to the 12 Julian ones, took Hecatombaion as January, and in 1468 he and his friends celebrated Plato’s birthday in November, the 11th Julian month.35 A superb Greek scholar, in this instance Ficino showed himself oblivious to both well-known texts (like the one from Plato) and more obscure ones (Aristotle notes that bears have sex in Elaphebolion, the 9th month, and Pliny that this takes place early in the winter (“hiemis initio”); accordingly, Thargelion, the 11th month, can hardly correspond to November).36 Translators of the works of Plutarch and Aristotle wrestled with these same problems, which often bested them as well. By reckoning the date of the eclipse of Nicias — so it seemed to Apianus — he had solved this difficult and tormenting problem. The Athenian calendar was lunar. Plutarch set the eclipse in the month of Metageitnion, the second of the Athenian year. An eclipse of the Moon could take place only at full moon, in the middle of the month. Accordingly, Apianus could state, with absolute certainty, that 15 Metageitnion, the night of the eclipse, corresponded to the night of 8 September 414 B.C. By running the Athenian calendar backwards from Metageitnion through Hecatombaion, Apianus could reveal to his startled contemporaries how the calendar had actually worked at a time when Socrates, Plato and Thucydides lived by it. To his readers — more of whom were probably expert users of calendars than were expert astronomers — it should have been clear that he had brought off a triumphant wedding of astronomy and philology. But to ensure that no one missed the point, Apianus made it the climactic moment of his discussion of eclipses, and praised his own table, as we Petrus Apianus 67 have seen, for presenting the facts about the Athenian calendar in a manner “clearer than light”. Apianus thus arrived at what he saw as an image of the past that satisfied his passion for precision — not the eclipse images themselves, which represented the sky at historically important moments, but a tabular, day-by-day presentation of the events that unrolled below the sky, on the infinitely more obscure and elusive surface of the Earth. In fact, as Paulus Crusius and Joseph Scaliger would soon show, Apianus had all the details wrong. The eclipse of Nicias, a spectacular one visible throughout the ancient Mediterranean, actually took place on 27 August 413 B.C. The eclipse of 8 September 414 B.C., a partial one, was not visible in the western Mediterranean, and therefore cannot have been the event that terrified Nicias. The eclipse he connected with Alexander’s defeat of Darius, which took place on 28 June 326 B.C., was total and widely visible. But both calendrical information and interlocking historical accounts that Apianus did not take into account show that the actual eclipse of Gaugamela took place happened five years earlier, on 20 September 331 B.C. (So does a Babylonian lunar diary to which he and his contemporaries had no access.37) Eusebius’s dating was actually more accurate than Apianus’s. So much for the plan to bring in astronomy to correct “historicorum errores”. Apianus, moreover, knew perfectly well that he had not solved all of the problems that his sources posed him. He thought that the Alfonsine Tables seemed to support yet another, much later date for Alexander’s victory — one that he could reconcile neither with his own eclipse date nor with Eusebius.38 “But”, he plaintively asked the reader, “I ask you, which of them should one believe? Alfonso? But he nowhere provides a clear argument in support his dating. Eusebius? He wove his histories together in so precise a sequence that it seems unbelievable that he can have gone wrong by as many as 12 years. Yet, probably as a result of carelessness, Eusebius missed the three years that I detected, as is clearly correct”.39 In the end, Apianus confessed, he could not provide rigorous new dates, though he remained convinced that his method would eventually make it possible to do so: However the matter may turn out, I have long since ceased trying to offer a final verdict. For it is not part of my current project to correct the historians, but only to show one of the paths that will lead to doing that, and to pave it. In the future, if God grants me a long enough life, I will offer a full new treatment of history and will support it in detail with computations of eclipses. There I will repeat this and other computations.40 Apianus’s sense of the difficulties involved in applying his seemingly rigorous method was prescient, as anyone acquainted with recent controversies about historical chronology will know. But he was right on the central point. In the period from 1569 to 1583, Gerardus Mercator, Paulus Crusius, Joseph Scaliger and Heinrich Bünting would prove that identifying the eclipses mentioned in historical accounts could enable historians to erect a new and solid framework for substantial parts of the record — at least for the period after era Nabonassar.41 68 Anthony Grafton In another, period sense as well, Apianus triumphantly anticipated the ways in which historical chronology would develop over the rest of the sixteenth century. Many scholars would emulate his effort to work out the relations between the Julian calendar on the one hand and the calendars of ancient Greece (and Israel) on the other. Just ten years after the Astronomicum Caesareum appeared, the Wittenberg astronomer Paulus Eber produced a diary that laid out the year, a day at a time, in the Roman, Egyptian, Macedonian, Athenian and Jewish calendars.42 Below the dates appeared printed accounts of great events, biblical and classical, that had taken place on the day in question. Plenty of room remained for the user to enter engagements for the future or records of the immediate past, as he or she pleased (most seem to have done the latter). Eber made his task easier for himself by assuming that all of these systems corresponded to the Julian ecclesiastical calendar, so that their months could simply be lined up with one another — a working principle that ensured that all segments of his information that came from the arbitrary Egyptian or the lunar Athenian calendar would be wrong. But his book went through several editions, and the annotated copies that appear in libraries around the world make clear that readers appreciated the information he provided. After his time every chronologer worth his salt — Erasmus Oswald Schreckenfuchs, Joseph Scaliger, Seth Calvisius, Denys Petau, Johannes Kepler — offered his own rules for converting dates given in the Athenian and other ancient calendars to Julian ones. Central to this effort — and to the chronologies of Mercator and Scaliger as a whole — was the realization that one could run the Julian calendar backwards for any given period. Before Apianus wrote, chronologers apparently assumed that the Julian year was a historical creation that began with Julius Caesar’s reform of the calendar in 45 B.C. and attained its definitive form when Augustus repaired the intercalation cycle thirty-six years later.43 Apianus, by contrast, showed that one could imagine the Julian calendar as functioning long before its historical date of creation. Reinhold incorporated this way of thinking into his Prutenic Tables, as he explained to his students: “the Julian year should be understood as in use back in time before Christ and Caesar, so that we may coordinate the histories of all periods with it. For of all the civil calendars it seems the one best fitted for this task.”44 By accepting this view, Mercator devised his own, much more approximate way of reckoning eclipse dates. More important, on the same basis Scaliger worked out his central chronological device: the Julian period, a period of 7980 Julian years, beginning on 1 January 4713 B.C., which could be used to fix celestial and terrestrial events to a single, rigorous scale (and avoided the problem that there was no year 0). The Julian day numbers that astronomers still use to date and time celestial phenomena are the distant descendants of Apianus’s modest effort to fix two Greek calendar dates unequivocally to a scale that he could measure. In period terms, however, we can express the meaning of Apianus’s collation of calendars rather differently. In a world in which debates about the true calendar could lead to public disturbances and even to civil war; in which calendars designed to be posted on walls or hung from belts gave ordinary people such purchase as they could claim over time; and in Petrus Apianus 69 which the spread of Lutheran and Calvinist calendars betokened, as Max Engammare has argued, something like the invention of punctuality itself, an image of the past that took the form of a calendrical table seemed incomparably solid and rigorous.45 To sum up: Apianus’s wall flower of a table turns out to be the unprepossessing remnant of a powerful and, as we now know, colourful enterprise: the first of the many efforts that have been made, from the sixteenth century to the twenty-first, to empower historians by putting a rigorous, quantitative method in their service. Every time that a historian or scientist tells us — as many have in recent years — that demography or development economics, ecology or evolutionary psychology, genetics or, indeed, astronomy is the tool that will at last make history a science, he or she is Apianus’s heir.46 No wonder that he was so proud of the apparently modest comparative table he had wrought. REFERENCES 1. For basic information on the Greek and Roman calendrical terms mentioned in this article, see F. K. Ginzel, Handbuch der mathematischen und technischen Chronologie: Das Zeitrechnungswesen der Völker (3 vols, Leipzig, 1906–14); E. J. Bickerman, Chronology of the ancient world, 2nd edn (Ithaca, NY, 1980); Alan Samuel, Greek and Roman chronology: Calendars and years in Classical Antiquity (Handbuch der Altertumswissenschaft, 1. Abt., 7. T.; Munich, 1972); Robert Hannah, Greek and Roman calendars: Constructions of time in the Classical world (London, 2005). 2. See in general Siegmund Günther, “Peter und Philipp Apian, zwei deutsche Mathematiker u. Kartographen: Ein Beitrag zur Gelehrten-Geschichte des XVI. Jahrhunderts”, Abhandlungen der Königl. Böhm. Gesellschaft der Wissenschaften, VI Folge, xi (Mathematischenaturwissenschaftliche Klasse, Nr. 4) (Prague, 1882); Peter Apian: Astronomie, Kosmographie und Mathematik am Beginn der Neuzeit, mit Ausstellungskatalog, ed. by Karl Röttel (Buxheim, 1995). 3. See e.g. Sachiko Kusukawa, “Leonhart Fuchs on the importance of pictures”, Journal of the history of ideas, lviii (1997), 403–27; Andrea Carlino, Books of the body: Anatomical ritual and Renaissance learning, transl. by John Tedeschi and Anne Tedeschi (Chicago and London, 1999). 4. Petrus Apianus, Astronomicum Caesareum (Ingolstadt, 1540), K I verso: “Quicquid autem hucusque super mensibus annique Atthici principio dictum fuit, hoc totum insequenti diario Luce clarius oculi te docere possunt.” 5. Martin Kemp, “Temples of the body and temples of the cosmos: Vision and visualisation in the Vesalian and Copernican revolutions”, in Picturing knowledge: Historical and philosophical problems concerning the use of art in science, ed. by Brian Baigrie (Toronto, 1996), 40–85, p. 82, quoted and discussed by Sven Dupré, “Visualization in Renaissance optics: The function of geometrical diagrams and pictures in the transmission of practical knowledge”, in Transmitting knowledge: Words, images and instruments in early modern Europe, ed. by Sachiko Kusukawa and Ian Maclean (Oxford, 2006), 11–39. 6. See Isabelle Pantin, “L’illustration des livres d’astronomie à la Renaissance: L’évolution d’une discipline à travers ses images”, in Immagini per conoscere: Dal Rinascimento alla Rivoluzione Scientifica, ed. by Fabrizio Meroi and Claudio Pogliano (Florence, 2001), 3–41; and “Kepler’s Epitome: New images for an innovative book”, in Transmitting knowledge, ed. by Kusukawa and Maclean (ref. 5), 217–37. On the uses of diagrams see also The power of images in early modern science, ed. by Wolfgang Lefèvre, Jürgen Renn and Urs Schoepflin (Basel, Boston and Berlin, 2003); Adam Mosley, “Objects of knowledge: Mathematics and models in sixteenth-century cosmology and astronomy”, in Transmitting knowledge, ed. by Kusukawa and Maclean (ref. 5), 193–216; and Christoph Lüthy and Alexis Smets, “Words, lines, diagrams, images: Towards a history of scientific imagery”, Early science and medicine, xiv (2009), 398–439. 7. Thucydides also rebukes Nicias for being too given to superstition (7.50.4). 70 Anthony Grafton 8. Plutarch, Nicias 23.2–4. 9. Cf. Anthony Grafton and Noel Swerdlow, “Greek chronography in Roman epic: The calendrical date of the fall of Troy in the Aeneid”, Classical quarterly, n.s., xxxvi (1986), 212–18. 10. On the importance of being able to predict eclipses, see Owen Gingerich, “Astronomy in the age of Columbus”, Scientific American, no. 267 (November 1992), 100–5. 11. A rich selection of eclipse illustrations appears in the classic work of John Murdoch, Album of science: Antiquity and the Middle Ages (New York, 1984). 12. Sebastian Mnster, Kalendarium hebraicum (Basel, 1527). 13. See in general Anthony Grafton and Megan Williams, Christianity and the transformation of the book (Cambridge, MA, 2006), chaps. 3–4; and Rosamund McKitterick, Perceptions of the past in the early Middle Ages (Notre Dame, IN, 2006). 14. Cassiodorus, Institutiones 1.17.1: “Chronica vero, quae sunt imagines historiarum brevissimaeque commemorationes temporum, scripsit graece Eusebius.” See Brian Croke, “The originality of Eusebius’s Chronicle”, American journal of philology, ciii (1982), 195–200. 15. See in general Daniel Rosenberg and Anthony Grafton, Cartographies of time (New York, 2010). 16. Noel Swerdlow and Otto Neugebauer, Mathematical astronomy in Copernicus’s De revolutionibus (2 vols, New York, Berlin, Heidelberg and Tokyo, 1984), i, 183–8. 17. Apianus, Astronomicum (ref. 4), J III verso: “Quo videas penitus quidnam ecleipsium cognitio virtutis, praestantiae, utilitatisque in se complectatur, aliquot ante Christum exempla, unde certius quam ex prioribus commodum conspici quaeat, adiungere coactus sum. Nemo ignorat, rerum historiarumque traditores in enumerandis tum mundi, tum urbis conditae annis, tantum non, ut dicitur, coelo ipso excidere. Qui quidem error quantum tenebrarum, ignorantiae, dubii, confusionisque in omni alia historia iudicanda intelligendaque importet, doctorum esto iudicium. Inde enim fit, ut si cuiuspiam rei, memorabilis saltem, lectorem christianum commonefaciant, quae urbem conditam aut praecesserit aut non ita multo post, subsequuta sit, ipsis constare nullo modo possint. Quod quidem tam grande malum, sola ecleipsium cognitio emendare et in melius vertere potest. Per ecleipses enim omnia certos in annos reduci possunt, Christum praecedentes, non minus, quam sequentes. Historicus autem certum semel tempus rei gestae nactus, reliqua per se, tam ante quam retro elapsa gestorum spacia, colligere valet. Quae omnia exempla uberrime patebunt.” 18. Apianus, Astronomicum (ref. 4), A ii verso: “Inter quae omnia, non ultimum opere precium reor, monstratum a nobis, quo pacto historiarum mendae corrigi debeant, prius a nullo unquam dictum, neque animadversum, quae res quantum momenti habeat, ille demum videbit, qui quid historiarum ordo in rebus humanis queat, perpendet.” 19. Anthony Grafton and Noel Swerdlow, “Technical chronology and astrological history in Varro, Censorinus and others”, Classical quarterly, n.s., xxxv (1985), 454–65. 20. John North, Horoscopes and history (London, 1986). 21. The notice in the Chronicle of Eusebius/Jerome (dated to AM 4782 = Ol. 90, 2 in the Paris editions of 1512 and 1518, 67 verso) reads: “Clades quae Atheniensibus accidit in Sicilia.” 22. Roger Bacon had called for a reform of chronology on the basis of astronomical data, which will be discussed by Philipp Nothaft in a forthcoming study. In a larger sense, the many astrologers who drew correlations between celestial portents, including great conjunctions of Jupiter and Saturn as well as eclipses, and events on earth also needed to be able to connect heavenly and earthly chronologies. 23. Eusebius and Jerome, Chronici canones, ed. by John Knight Fotheringham (London, 1923), 253: “Defectio solis facta et Augustus LXXVI aetatis suae anno Atellae in Campania moritur sepeliturque Romae in campo Martio.” 24. Paul of Middleburg, Paulina de recta paschae celebratione: et de die passionis domini nostri Jesu Christi (Fossombrone, 1513), T ii verso: “sed si iuxta usitatam annorum domini computationem ad transacta tempora retrocedere velimus, nullam solis eclipsim Romae factam inveniemus eo anno quo chronographus praefatus Augustum vita functum meminit, neque sequenti anno Petrus Apianus 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 71 eclipsim aliquam factam comperiemus, neque praecedentibus annis eclipsis aliqua insignis facta fuit, sed tertio anno sequente videlicet anno domini salvatoris iuxta consuetam supputationem. 17. maximam solis eclipsim factam inveniemus, in qua totus sol Romae et in plurimis Italiae regionibus visus fuit obscurari die Februarii quintadecima, hora fere meridiana.” Paul of Middelburg, Paulina (ref. 24), T ii verso: “constat certe Eusebium et historiographos de hac duntaxat eclipsi scripsisse, cum maxima extiterit, adeo quod sole prope meridianum existente tenebrae obortae fuerint, neque proximis annis eclipsis aliqua insignis facta fuit, quae a chronographis memoriae commendari meruerit. mos enim historiographis atque chronographis esse solet illas duntaxat eclipses solares in annales referre et annotare, in quibus totus sol a visu nostro occultatur et absconditur, adeo ut interdiu astra caeli stellati videri et numerari possint, quales raro admodum contingunt.” For a refutation of an earlier version see Petrus de Rivo, Tractatus III de anno die et feria dominice passionis et resurrectionis (Louvain, 1492), III.8, e 3 recto – e 4 recto. See Inscriptiones sacrosanctae vetustatis, ed. by P. Apianus and B. Amantius (Ingolstadt,1534), p. ccxiv: “Obeliscus in Vaticano iuxta Basilicam B. Petri non Iulia vocitatur, sed ex Aegypto Caii Principis iussu abductus, ut scribit Ioan. Tor. in altitudine brachia habet 45. latitudine in fundo cuiuslibet quadri brach. 4. cum dimidio, in summitate vero brach. 2. cum dimidio.” Cf. Epigrammata antiquae urbis, ed. by Jacopo Mazzocchi (Rome, 1521), X verso: “In Obelisco qui nunc dicitur Iulia. S. Petri in Vaticano.” For Apianus’s and Amantius’s antiquarian work and its context see Christopher Wood, Forgery, replica, fiction: Temporalities of German Renaissance art (Chicago, 2008). Inscriptiones, ed. by Apianus and Amantius (ref. 27), [cccccxviii], commenting on a statue of a bearded and tattooed barbarian wearing a shawl over his shoulders, and a loincloth: “Cuius haec sit effigies, ne divinare quidem licuit, et quamvis diu ad hanc rem cogitaverimus, quapropter tuo iudicio o Lector subiicimus, etsi ignoremus tamen apposuisse nihil nocet, quod gratum sit videre et intelligere morem vetustae vestis et indumenti germanorum, quibuscunque hominibus, qui saltem veterum monumentis oblectari solent.” Erasmus Reinhold, Prutenicae tabulae coelestium motuum (Tübingen, 1551), I, includes a discussion of the Egyptian, Julian and Muslim calendars, 20 verso – 22 verso; of how to convert any given day in a Julian year, B.C. or A.D., to a day in an Egyptian year, 22 verso – 23 verso; how to convert any given day in an Egyptian year to a Julian date, 23 verso – 24 recto; and how to identify the weekday of any given day in a Julian year, 24 recto – 25 verso. Part II includes tables for converting Julian into Egyptian dates and vice versa. The British Library copy (C.113.c.7) contains notes taken at Reinhold’s lectures, beginning, according to a note on the title page, on 19 October 1551. Robert Wilken, The Christians as the Romans saw them, 2nd edn (New Haven, 2003), 41–4. See Anthony Grafton and Noel Swerdlow, “Calendar dates and ominous days in ancient historiography”, Journal of the Warburg and Courtauld Institutes, li (1988), 14–42. Reinhold, Tabulae (ref. 29), 22 recto, writes: “Olympiadum anni sunt Lunares ad Solis tamen cursum utcunque accommodati, quorum initium Attico more semper a meridie primi diei Hecatombaeonis et conversione Solis aestiva proficiscitur.” A marginal note in British Library C.113.c.7, probably recording a comment from his lectures, explains that “Olympiadum anni” are “Graeci vel Atheniensium”. Beyond that he does not seem to have gone. Milton Anastos, “Pletho’s calendar and liturgy”, Dumbarton Oaks papers, iv (1948), 183–305. Paul Botley, “Renaissance scholarship and the Athenian calendar”, Greek, Roman and Byzantine studies, xlvi (2006), 395–431. Grafton and Swerdlow, “Calendar days” (ref. 31), 38. Aristotle, Historia animalium 579a25; Pliny, Naturalis historia 8.54.126. Lorenzo Verderame, V. F. Polcaro and G. B. Valsecchi, “The Gaugamela battle eclipse: An archaeoastronomical analysis”, Mediterranean archaeology, viii/2 (2008), 55–64. Reinhold, Tabulae (ref. 29), II, 142 recto, offers a “Catalogus in quo insignium et memorabilium rerum gestarum intervalla, Iulianis annis descripta, proponuntur”. The eras of the Alfonsine 72 39. 40. 41. 42. 43. 44. 45. 46. Anthony Grafton tables follow on 142 verso – 143 recto, but Reinhold warns, on 142 recto: “Sequitur Canon aerarum Alphonsi ex ipsius sententia correctus, etsi a vera historia multis in partibus plurimum discrepat” — a warning that, had it been made just over a decade earlier, could have saved Apianus considerable time and effort. Apianus, Astronomicum (ref. 4), [J IV recto]: “Sed quaeso cuinam horum potius fidas? Alfonso ne? verum hic suae computationis rationem nullibi manifestam producit. Eusebio ne? qui historias tam accurata sequela connectit, ut incredibile sit eum posse tot annis, nimirum 12 falli, tribus tamen annis idem Eusebius quoque excidit, quos per incuriam omissos, ut veri simile est, ego dubio procul omni recte cognovi.” Apianus, Astronomicum (ref. 4), [J IV recto]: “Verum rem utut se habeat, iudicare longius desino, cum instituti praesentis non sit, historias ad plenum emendare, sed viam tantummodo aliquam ad illud conficiendum indicare et praestruere. Olim, si Deus vitam concesserit, historias ex integro retractaturi, ecleipsiumque rationibus vere firmaturi sumus, ubi et hanc et alias iterabimus.” See Anthony Grafton, Joseph Scaliger (2 vols, Oxford 1983–93), ii. Paul Eber, Calendarium historicum conscriptum a Paulo Ebero Kitthingensi (Wittenberg, 1550). For an example see Giovanni Maria Tolosani, Opusculum de emendationibus temporum (Venice, 1537), 112 recto – verso. Tolosani wrote under the pseudonym Johannes Lucidus Samotheus, which was exposed by Edward Rosen, “Was Copernicus’s Revolutions approved by the Pope?”, Journal of the history of ideas, xxxvi (1975), 531–42. Reinhold, Tabulae (ref. 29), 23 recto: “Supra autem dixi retro ante Christum et C. Caesarem cogitandum esse usum anni Iuliani, ita ut omnium temporum historias ad eum referamus, propterea quod inter civiles annos videtur huic negocio potissimum convenire.” A comment ad loc. in the British Library copy, C.113.c.7, underlines his statement: “Vsus anni Iuliani retro etiam ante Christum et Caesarem cogitandus est.” See Bernd Roeck, Eine Stadt in Krieg und Frieden: Studien zur Geschichte der Reichsstadt Augsburg zwischen Kalenderstreit und Parität (Göttingen, 1989); David Cressy, Bonfires and bells: National memory and the Protestant calendar in Elizabethan and Stuart England (London, 1989); and Max Engammare, L’ordre du temps: L’invention de la ponctualité au XVIe siècle (Geneva, 2004). E.g. Daniel Smail, On deep history and the brain (Berkeley, 2008).