On the crowded mantelpiece of contributors to the debate on the anachronism of Dido and Aeneas explored in our first Interlude, the scholar who stands out in the history of chronology is Joseph Scaliger (1540–1609), a French Protestant based in Leiden in the later stages of his career. Scaliger has been called ‘the founder of modern chronological science’,1 and even (in the words of the Victorian scholar Mark Pattison) ‘perhaps the most extraordinary man who has ever devoted his life to letters’.2 In this chapter we will use Scaliger’s scholarship as a means of exploring both ancient attitudes to anachronism as chronological error – the sense that the word acquired in Scaliger’s lifetime – and the place of chronology in the history of temporalities ancient and modern. The upshot of these discussions is that ancient chronological method was informed by a strong sense of historical process. Far from being an isolated, purely technical discipline, ancient chronography parallels its modern successors in its reflection of ideological concerns.
We start with ancient calendars and their irregularities, the topic to which Scaliger devoted the first books of his first major chronological work, De emendatione temporum (On the Improvement of Times, 1583). We then analyse the concern with anachronism revealed by the dating systems used to record the past as well as the ancient interest in detecting isolated chronological mistakes. We will close by exploring how a sensitivity to anachronism was fostered by the conflicting temporalities – and the divergent religious practices – found in different cultures. First, however, we discuss in more detail a topic at which we glanced in Chapter 1 – Scaliger’s own place in the history of chronology and anachronism.
The myth of Scaliger has been unsettled by Anthony Grafton’s remarkable intellectual biography and by other scholars following in Grafton’s steps. The claims in Scaliger’s favour made by admirers such as Pattison were that he opened up the chronology of the ancient world to critical study by comparing classical and Near Eastern sources (some previously unknown) and by using astronomy to provide secure dates for eclipses, and that he secularized time by treating years as mere numbers rather than as clues to the patterns of history. Grafton and others have shown that the comparative approach to chronology was widespread, that the use of astronomy for dating can be traced back to the thirteenth century, and that Scaliger’s arguments about the date of the Passion were entangled in contemporary religious disputes as well as strongly typological (that is, an Old Testament event – in this case Abraham’s sacrifice of a lamb – prefigures one in the New Testament – here Christ’s death). The immensely learned but fallible Scaliger who emerges from recent accounts is not quite a modern, but nonetheless still a pivotal figure in the history of chronology.3
A topic not treated by Grafton is Scaliger’s contribution to the language of anachronism. As we saw in Chapter 1, Scaliger has often been credited with the spread of the Greek word to Western Europe. In fact the adoption of the term in Latin and Italian pre-dates him by some decades, and his own earliest uses show no consciousness that the word is a new coinage: he did not gloss it in a letter he wrote in French in 1591 in which he argued that ‘anachronismes’ in the Old Testament book of Judith show that it is a Jewish fiction,4 nor when he first used it in print (in its Greek form) in the second edition of De emendatione (1598). While not a neologism, anachronismos did lend powerful support to his advocacy of the value of Near Eastern chronological traditions. Responding to a Catholic historian, Cesare Baronio, who had criticized him for rejecting the date for the destruction of the temple in Jerusalem offered by the Christian chronicler Eusebius of Caesarea (c. 260–339), Scaliger suggested that Eusebius had been led by his contempt for the Jewish historian Josephus into ‘those anachronismoi by which his whole interweaving of the times is disturbed’. Again, in his notes on a fragment of the Babylonian historian Berossus he suggested that ‘portentous anachronismoi’ were born of chronologers’ ‘barbaric contempt’ for local chronological traditions.5
The idea of Scaliger as innovator in the study of anachronism does nonetheless have some justification. He elaborated in his 1606 Thesaurus temporum (Treasury of Times) a distinction between anachronismus, defined as ‘perversion of a heading of time, or dislocation of an epoch’, and two specific types of error: prochronismus (‘displacement of an epoch through removal, when less is said’) and metachronismus (‘displacement of an epoch through addition, when more is said’).6 These definitions were the prop for the long commentary on Eusebius that Scaliger included in Thesaurus temporum. Eusebius’ Chronicle consisted of two parts, a Chronography (a collection of king-lists and other chronological data which survives in excerpts and in an Armenian translation that was unknown in the sixteenth century) and Chronikoi Kanōnes (Temporal Tables); the second part, which was widely known through Jerome’s translation into Latin, arranged the information from the Chronography in parallel columns with a summary of events alongside a continuous numbering of years from the birth of Abraham (2016 BC in our terms) as well as other dating systems (regnal years, Olympiads) where appropriate (Fig. 5).7 In his commentary, Scaliger scrutinized Eusebius for internal chronological inconsistencies as well as for contradictions with other evidence. He was able, for instance, to detect a ‘prochronism of six years’ in the entry for the year of Abraham 1598, ‘the disaster which happened to the Athenians in Sicily’, because of ‘the lunar eclipse which intervened in that disaster’ and because historians such as Thucydides and Diodorus preserved a year-date relative to the start of the Peloponnesian War which could be aligned with Olympiad years.8 The originality of the commentary lay, however, in scope rather than in method. The eclipse in Sicily had been used to disprove Eusebius’ date in Petrus Apianus’ 1540 Astronomicum Caesareum,9 and even the language of ‘removal’ and ‘addition’ in Scaliger’s definitions of prochronism and metachronism was drawn from the third-century Christian chronographer Julius Africanus (F34 Wallraff).10 But no one before Scaliger had devoted so much time and learning to elucidating the chronological errors in a single work.
Scaliger’s keen eye for anachronism contrasts with the lax attitude to chronology that is often attributed to antiquity. As we noted in Chapter 2, some modern scholars attribute to the Greeks and Romans a sense of time as episodic, relative and multiple, with no conception of the past as an entity in itself and ‘no cognitive awareness of absolute chronology’.11 Such views might be supported by the approach to chronological exactitude that we noted in our Prelude. Plutarch, we saw, was prepared to retell the meeting of Solon and Croesus even though its historicity was strongly questioned by ‘the so-called temporal tables (chronikoi kanōnes) which countless people are to this day correcting without being able to bring the contradictions into any general agreement’ (Solon 27.1). And he took an even stronger stand in his biography of the Roman king Numa, berating as childish those writers who prided themselves on solving intractable problems such as the tradition that Numa was educated by the Greek philosopher Pythagoras (Numa 8.21).12 But Plutarch also notes that part of the problem lay in the ‘late’ date at which the Greeks turned to chronology (Numa 1.6, alluding to the Olympiad list constructed by Hippias of Elis at the end of the fifth century BC). And his polemic reveals at the same time that many people in antiquity did take such questions seriously. One of Horace’s poems is set at a party where a guest keeps going on about the number of years that separated Inachus, first king of Argos, from the Athenian king Codrus; Horace would rather he got on with the drinking (Odes 3.19.1–3). Dry-as-dust pedantry inspired such dislike partly because it had so many adherents.
Fig. 5 A page from a manuscript of Jerome’s version of Eusebius’ Chronicle, showing the date for the Athenian defeat in Sicily during the 90th Olympiad. Bodleian Library, Oxford. MS. Auct. T. 2. 26, fol. 87r. Photo Bodleian Libraries.
Scaliger saw himself as heir to the scholarly strand of classical chronography.13 At the start of De emendatione, while admitting (like Plutarch) that the Greeks and Romans were ‘later than they should have been’ in applying themselves to the reckoning of times, he laments that much of what they wrote has been lost; in particular, of the Greek writings ‘which concerned chronology’, ‘nothing is left us except a longing for them’.14 He then singles Eusebius’ Chronicle out from those works that do survive (in some form at least), but he does so critically, as one of those books which ‘today in the lack of better ones are valuable to us’ – but which would be ignored ‘if the tables (canones) of Thallus, Castor, Phlegon, and Eratosthenes were extant’.15 Looking beyond Christian chronography, Scaliger here traces out a pagan chronological tradition that started in Hellenistic Alexandria, where Eratosthenes was based, and continued under the Roman empire with writers such as Castor of Rhodes (first century BC) and Phlegon of Tralles (second century AD). Scaliger knew from their scanty remains that these authors had specified the intervals in years between epochs such as the sack of Troy and the military expeditions of Xerxes and Alexander – the same task that he would himself undertake, in much amplified form, in Books Five and Six of De emendatione.
Scaliger linked the task of the chronologer with another Renaissance science, archaeology. What chronologers attempt is, in his words, ‘to pull back fugitive antiquity’ and ‘each day with heroic courage to dig up (eruere) chronology from the darkness and silence of oblivion, dead and buried as it is through the neglect of earlier generations’.16 Among his own contributions to this scholarly pursuit was the editing of ‘fragments’ – a term that itself evokes a sense of the past as broken like the pots and statues dug from the ground. He also edited and commented on many previously unpublished chronological works. More than that, he himself included in Thesaurus temporum an Olympiad chronography, Anagraphē Olympiadōn, that he had worked over time into a form that seemed so authentic that careless readers mistook it for an ancient text – and that even in the nineteenth century received the tribute of a commentary of its own that corrected its occasional mistakes with the help of the Scaligerian panoply of metachronisms and prochronisms.17
‘Chronology’ was itself a recent coinage in Latin when Scaliger defined its remit. First attested in Greek in the Byzantine chronographer George Syncellus, the word entered Latin in 1531 in an edition of Livy with a chronological appendix as well as a dedicatory letter by Erasmus which praises chronologia as the ‘thread’ without which ‘inextricable error envelops even the erudite reader in the labyrinths of historical events’.18 Scaliger now becomes our thread as we analyse ancient attitudes to chronological anachronism as well as chronology’s role in the intersections of ancient and modern temporal consciousness.
One of the pieces of evidence about the Greek calendar with which Scaliger grappled was the episode we discussed in our Prelude: Herodotus’ account of Solon’s lecture to Croesus on the fragility of human life. Solon there calculates that the normal human life of seventy years has 26,250 days – on any of which misfortune may occur. His calculation is based on a year of twelve thirty-day months with another thirty-day month added every other year so that ‘the seasons tally as necessary’ (1.32.3). Solon’s inclusion of this intercalary month boosts his rhetoric (an extra 1,050 days for disaster to strike). But his method would signally fail to keep the seasons in sync: as the Greeks were aware (by the sixth or fifth century BC), a 360-day year would need to add twenty-one days every four years rather than thirty every two to do that.
Scaliger’s response to this puzzle was to include a section ‘On Herodotus’ false year’ at the end of the first book of De emendatione. As the title suggests, he put the blame for Solon’s ridiculous system firmly on Herodotus. The problem was not so much the 360-day year in itself. Scaliger in fact argued for a new interpretation of the Athenian calendar based on twelve thirty-day months, thereby abandoning the consensus that Athenian months were lunar; elsewhere in De emendatione, moreover, he allows that a 360-day year was widely used ‘in ancient times, especially in Egypt’, as ‘the most ancient and simple form of the year – and the one most appropriate to popular time-keeping … because it divides up so evenly’. But he assumes that Herodotus did not understand the principles of intercalation: by Herodotus’ time, in Scaliger’s view, the Athenians were using a complex system based on a seventy-six year period comprising nineteen four-year cycles.19
In the second edition of De emendatione, Scaliger again devoted a section to Herodotus’ ‘false year’. He had in the meantime developed a radically different view of the fifth-century Athenian calendar. He now posited a year of 354 days formed of six ‘hollow’ (twenty-nine-day) and six ‘full’ (thirty-day) months, with three thirty-day months intercalated every eight years. Herodotus’ mistake, he suggested accordingly, was based on a misinterpretation of the hollow/full alternation (days were numbered up to thirty even in hollow months, with a day earlier in the month being subtracted). Besides adjusting his view of how Herodotus slipped up, Scaliger attempted to cast new light on Herodotus with the help of a Greek text that first appeared in print in 1590: Introduction to the Phenomena by Geminus, a first-century BC astronomer and mathematician. Included in Geminus’ treatise was a discussion of the history of the months (8.26): ‘The ancients had thirty-day months, with intercalary months every other year, but, the truth being swiftly tested in the phenomena, because the days and months were not in harmony with the moon and the years were not aligned with the sun, they sought a period that would be in harmony with the Sun in respect of the years and with the moon in respect of the months and days.’ According to Scaliger, this new evidence from a ‘most accurate writer’ confirmed that the system of intercalation applied by Herodotus’ Solon had been used ‘in ancient times’.20 He did not consider that Herodotus himself might have been Geminus’ source and that Geminus’ goal might be pedagogical rather than historical: modern editors suggest that ‘he means to show his reader how one might begin to solve the problem of finding a good lunisolar cycle with a simple first step’.21
Scaliger’s approach to the ancient Greek calendar is throughout marked by a strong sense of anachronism. By equating the ‘most ancient’ and the ‘most simple’, he points both to his own cognitive distance from antiquity and to historical changes within antiquity itself. In positing that the ancient calendar had gradually diverged from an ancient simplicity, he was adopting a model of uneven development that was commonly found in ancient accounts of the evolution of human society (as we shall see in Chapter 5) and had been applied to the determination of months and years by many earlier writers from antiquity to the Byzantine period.
An intermediate source for the developmental model was a work On Months by Theodore Gaza, a fifteenth-century scholar from Thessalonica who played an important part in introducing the study of Greek in Italy.22 Gaza’s work presents the classical order of the months as developing from disorder: the ‘most ancient’ peoples determined the months ‘without calculation or order’ (On Months 7 (Patrologia Graeca 19.1188): alogistōs kai ataktōs). Gaza here follows the common ancient pattern of defining early human existence in terms of the absence of features that emerged later.23 His inspiration for applying this model to the calendar was Plutarch, who spoke in the same terms of the Roman calendar under Romulus (Numa 18.1: alogōs … kai ataktōs); both Gaza and Plutarch were drawing in turn on the famous cosmogony of Plato’s Timaeus (43b1–2 ataktōs … kai alogōs; cf. alogōs at 43e3, 53a8 and ataktōs at 30a5, 69b3). A similar image of early calendrical disorder is found in a sixth-century AD treatise On Months by John the Lydian, where the Sicyonians and Arcadians, the first inhabitants of Greece after the flood, are said to have been called proselēnoi (pre-moons) ‘because they are said to have arisen before the delineation of the months’ (3.1).24
The accounts of the calendar that survive from antiquity offered Scaliger schematic outlines of the process by which order had been achieved. Geminus, following his initial account of the ‘two-year’ cycle, describes the successive creation of cycles of eight, nineteen and seventy years, each one offering a superior fit between the rhythms of the moon and the sun (8.26–59). A more detailed but equally schematic account is offered by the Roman grammarian Censorinus in On the Birthday, an unusual work offered as a present to his patron in AD 238 which includes chapters on numerous different aspects of time. The ‘great year’ (a term for the cycles Geminus describes) was in the ‘ancient states of Greece’ two years, with one intercalated month (the system Herodotus’ Solon uses, but without specification of days25); Censorinus adds the explanation that twelve or thirteen new moons were observed in each solar year. Then, when ‘error was recognized’, the ‘great year’ was doubled to four years, and then again to eight; and in due course more complicated cycles of nineteen, fifty-nine and seventy-six years were created (18). Similarly, in relation to months Censorinus writes that the Egyptians initially had one, then four, then thirteen (emended by Scaliger to twelve) with an extra five days (19). This pattern of steady numerical increase is found in the histories of many other domains, including ship design (Pliny, Natural History 7.207–8), the colours used in painting (Philostratus, Life of Apollonius 2.22.4), actors in tragedy (Aristotle, Poetics 1449a15–18) and holes or strings on musical instruments (Horace, Art of Poetry 202–19; Athenaeus 4.183c). Scaliger followed this evolutionary model in criticizing Censorinus’ suggestion that the eight-year cycle might have been devised by the fourth-century BC mathematician Eudoxus of Cnidus, even though the Athenian Meton (active in the fifth century) had already created a nineteen-year cycle.26
Scaliger’s approach to the Roman calendar was shaped by similar assumptions – and again he was following a pattern established in antiquity. According to a common tradition, picked up in the third-century AD miscellany-writer Solinus (1.35–9), the year set up by the founder of Rome, Romulus, had 304 days divided into ten months (starting with March);27 the second king, Numa, raised the number of days to 355 by adding two extra months, January and February. Scaliger ridiculed the assumption that a ‘shepherd’ such as Romulus could have set up a calendar system, and attributed the first (inept) reforms to ‘Numa or some other bumpkin or rustic’.28 Similarly the fifth-century writer Macrobius (Saturnalia 1.13.1) describes the initial Roman calendar as the product of an ‘uncouth and still unpolished age’ (saeculo rudi et adhuc impolito), while Ovid in his Fasti (a month-by-month account of the Roman year) attributes Romulus’ ‘error’ to the fact that he ‘knew arms better than stars’ (1.29). Ovid further suggests that Romulus named March after the god of war and made it the first month in part as tribute to the Romans’ warlike character (3.79–80).29
Scaliger also followed ancient sources in his approach to the calendar introduced by Julius Caesar (a year of 365 days with one day inserted every four years). Ovid writes that ‘the times were wandering even now (errabant etiam nunc tempora)’ – until Caesar took control (Fasti 3.155–6); the phrase ‘even now’ (often used to express the surprising survival of a practice (p. 126)) here suggests that the Romans were surprisingly late to introduce changes. A similar account is found in Censorinus: Caesar ‘corrected’ the calendar after observing that ‘the months did not correspond as they should to the moon, nor the year to the sun’ (22.8). Other writers add that the leap-year rules were misapplied after Caesar’s death until the error was corrected by Augustus (Suetonius, Augustus 31.2; Solinus 1.45–7; Macrobius, Saturnalia 1.14.13–15).30 Heir to these developmental assumptions, Scaliger was even more vigorous in his language: all of Caesar’s military triumphs, he wrote, are ‘mere trifles set against this arrangement of the year’.31
The idea of progress paraded by ancient accounts of calendar reform gave rise to the possibility that different temporal strata could be conceived as co-existing.32 Plutarch was one of a number of ancient authors to note that the last six months of the Roman year were named after the ordinals ‘fifth’ to ‘tenth’ (or the last four after Quintilis and Sextilis had been renamed after Julius Caesar and Augustus). The names of these months supported reconstructions according to which the original Roman year, whether formed of ten or twelve months, started in March; on either view, the months kept their names when the calendar was reformed even though they had lost their etymological significance (Plutarch, Roman Questions 268a–b). A similar conservatism was observed in the festival calendar: a speaker in Macrobius notes that with Caesar’s calendar (which added two days to December) the Saturnalia were now celebrated on the sixteenth rather than the fourteenth day before the Kalends of January, but that some of the common people kept up their celebrations ‘according to the old custom’ (vetere more); the result of this confusion was that the festival expanded from one day to three (Saturnalia 1.10.2).
The progressivist understanding of ancient calendar reform has been criticized by some recent scholars. In his important study Caesar’s Calendar, Denis Feeney suggests that the sort of attitude reflected in Scaliger’s celebration of Caesar’s triumph is anachronistic: ‘as inhabitants of Caesar’s grid we take it for granted that a calendar is there precisely to measure time, to create an ideal synthesis of natural and socially or humanly organized time and in the process to capture a “time” that is out there, waiting to be measured’.33 That is, prior to Caesar’s reforms calendars were geared towards the civic time of festivals and other human institutions. It was the creation of a system with identical years (or at least years differing by one day every four years) that gave rise to the idea of a regular succession of measurable time independent of political manipulation. Feeney’s argument is supported by the fact that many cities kept their traditional systems of intercalation even when more precise systems were available; in addition, the standardization of calendars in the first millennium BC can be seen as part of the administrative centralization carried out by large imperial powers such as the Seleucids and Romans – even if the Romans also tolerated local exceptions such as the prestigious Athenian calendar.34
Feeney offers a brilliant picture of the changes wrought by Caesar’s calendar. He points, for instance, to changes in accounts of the festival and agricultural year before and after the reforms as well as to the new importance of anniversaries as pathetic markers of transience.35 Nonetheless, the notion that a calendar captures a time that is there to be measured should not be dismissed as the product of the post-Caesarian mindset. A strong awareness of the continuity of ‘time’ as a separate entity is caught in images of chronos as ‘best saviour for just men’ (Pindar fr. 159), ‘ancient father of days’ (Euripides, Suppliant Women 787–8) and ‘unwearying’ and ‘begetting itself in its ever-flowing stream’ (Critias, DK 88 B18).36 The calendar year, moreover, was used for measurement not just in rhetorical demonstrations such as Solon’s diatribe on the fragility of human existence, but also in the realms of politics and diplomacy: when the Athenians ostracized politicians for ten years (from 487 BC), or made a treaty with the Spartans for thirty, the years were defined by the magistrate associated with each calendar year (archons at Athens, ephors at Sparta); at Athens, decrees began to be inscribed with the archon’s name from the 430s BC, and a list of annual archons was inscribed in the 420s.
The practice of intercalation in ancient calendars is itself a further reason for thinking that progressivist approaches are not totally anachronistic. The practice implies that cities were keen to keep some form of correspondence between months and the solar year – even if pre-Caesarian Rome tolerated calendars that were at times two or three months out of joint with the seasons. A desire for close correspondence is, moreover, at times made explicit in justifications of intercalation: Herodotus, as we have noted, speaks of ensuring that the seasons ‘tally (sumbainōsi) as necessary’ (1.32.3; cf. 2.4.1), and a reform to the 365-day Egyptian calendar proposed in the third-century BC (though only applied by Augustus) aimed to ensure that winter and summer festivals were held in the proper season (OGIS 56.40–6: the Canopus Decree, 238 BC). Something of the Scaligerian mind set likewise underlies the Moon’s complaints (as reported by the chorus of Aristophanes’ Clouds (607–26)) that the festival calendar has gone awry.
We have seen, then, that ancient accounts of calendars show a sense of anachronism in the developments they posit from simple to complex, and that calendrical systems reveal a desire to match the different rhythms of the moon and the sun. As we shall see in the rest of this chapter, detection of historical anachronisms is founded on a similar concern to match different temporal cycles.
One of Scaliger’s contributions to chronology was the invention of the Julian Period, a period of 7,980 years based on a 28-year solar cycle, a 19-year lunar cycle and a 15-year indiction cycle (a system used for administrative purposes from late antiquity to the early modern period), with a starting date equivalent to 4713 BC (year 1 in each of the three cycles). The advantage of the Julian Period was that it provided a continuous timeline well adapted for astronomical calculations (it is still used for this purpose). Scaliger also, however, displayed his historicizing instincts by looking at the evidence for other dating systems and even, as we have noted, creating his own Olympiad-based chronography of ancient Greek history.
What seems to be an odd slip appears in Scaliger’s chronography in the entry for the second year of the eighty-seventh Olympiad. He offers the name of the Athenian archon, Euthydemus, and then a single historical notice: ‘Beginning of the Peloponnesian War, which Thucydides son of Olorus wrote up.’37 As the nineteenth-century German scholar Ewald Scheibel noted in his commentary on the chronography, Scaliger was following a view found in Diodorus (12.38) and implied in Aulus Gellius’ dating of the start of the war to the 323rd year from Rome’s foundation (17.21.16).38 Thucydides, on the other hand, offers a different and, in Scheibel’s view, more accurate date at the start of his second book. After announcing that he will give an account of the war by successive summers and winters, he fixes the war’s start with the Theban invasion of Plataea in ‘the fifteenth year’ of ‘the Thirty-Year Treaty made after the capture of Euboea’, ‘when Chrysis had been priestess at Argos for forty-eight years, when Aenesias was ephor at Sparta and Pythodorus [Euthydemus’ predecessor] still had two months to serve as archon at Athens, in the sixth month after the battle at Potidaea and at the start of spring’ (2.2.1). If this date is accepted, Scaliger seems to have committed a metachronism of one year.
Scaliger’s decision not to follow Thucydides was not due to a lack of knowledge. In De emendatione he uses this passage as evidence for the start of the Athenian year, and he alludes to it again, albeit sloppily, in his commentary on Eusebius’ Chronicle, which was published alongside Anagraphē Olympiadōn, to explain why the start of the war is there placed the year before Euthydemus’ archonship: ‘It is true the Archidamian or Peloponnesian War began when there were still a few days left of Pythodorus’ magistracy.’39 But he at once adds that ‘writers of this war take its beginning from the following magistrate Euthydemus’. That is, Scaliger consciously overrides Thucydides even though by his usual principles he would have favoured him as contemporary with the events he describes.40
The anachronism committed by Diodorus and other ‘writers of this war’ results from a clash of different dating systems. Diodorus dates each year by Athenian archons and Roman consuls while noting Olympiads every four years. Since each of these systems had a different starting point within the calendar year, it was possible to date an event correctly in one system but incorrectly in the others: thus in our example Diodorus dates the start of the Peloponnesian War correctly in Roman terms but to the wrong archon and Olympiad year. Scaliger himself had no such excuse since his chronography included only Olympiad and archon years. He may have felt that his dating was justified instead on historicist grounds: he was producing what a (reasonably accurate) Olympiad list might have looked like.
Thucydides’ own dating system reflects his concern to avoid precisely the sort of anachronism found (four centuries later) in Diodorus. His synchronism of different systems at the start of Book 2 implicitly underlines their disjunction both with each other and with the seasons which control the rhythm of military campaigns by land and sea; the passage is a concealed polemic against his near-contemporary Hellanicus of Lesbos, who composed a Panhellenic chronicle entitled, after its dating system, Priestesses of Hera at Argos as well as an Atthis (history of Athens) which used archon years. The polemic becomes explicit later in the work when Thucydides defends his claim that the first part of the war lasted for almost exactly ten years: only dating by seasons, he suggests, can bring out its length with the required accuracy (5.20).41
Further light is shed on Thucydides’ hostility to local dating systems by his account of the closing years of the first part of the war. In the ninth year, a year-long armistice was agreed between Athens and Sparta; two years later, a short-lived fifty-year peace was formalized. Thucydides cites the terms of both treaties, including their start dates in each city’s calendar: for the armistice, 14 Elaphebolion at Athens, 12 Geraestius at Sparta (4.118.12); for the peace, 25 Elaphebolion and 27 Artemesius (5.19.1). The difference in the days of the month in itself points to the lack of alignment between the calendars;42 the second set of dates shows, moreover, that in two years the alignment of both months and days in the two cities’ calendars has changed. This variability of local calendars is one reason why Thucydides was unwilling to use them for dating events within his war seasons; other, equally deep (and related) reasons are their parochialism and their absence from his main literary model, Homer.
The example of the armistice does nonetheless reveal the role that local calendars could play in detecting chronological errors. Thucydides reports how news of the armistice reached the Spartan general Brasidas as he was in the process of welcoming Scione, a city defecting from Athenian control. A dispute then arose as to whether this revolt happened before the truce, as Brasidas claimed, or afterwards, as the Athenian envoy asserted ‘by reckoning up the days’ (4.122.3). It is only at the end of his account of the episode that Thucydides reveals that the Scionians revolted ‘two days too late’ (4.122.6) – evidently on 16 Elaphebolion in the Athenian calendar. Rather than revealing that date at the outset, Thucydides leaves it to be inferred at the end. He thereby preserves narrative tension while showing how the meaning of events was controlled by the perceptions of participants.
The propensity of annalistic dating to give rise to anachronism was discussed by other historians too. Both Polybius in his history of Rome’s rise to power and Tacitus in his account of the Roman emperors offer year-by-year narratives, but they note on occasion that they have grouped together in a single year events that took place in more than one year, either because the events would seem insignificant if narrated separately or (in Tacitus’ case) to offer some respite from a relentless narrative of internal discord (Polybius, Histories 14.12.4–5, 32.11.2–7; Tacitus, Annals 6.38.1, 12.40.4). Polybius further explains that one consequence of his geographical arrangement within each year’s narrative (outlined at 5.30.8–33) is that later events in one area are narrated before earlier events in another area even when they are causally connected (15.24a, 28.16.5–11).
Another problem in annalistic dating was alleged by the late antique historian Eunapius in his continuation of the Historical Chronicle composed by the third-century AD Athenian Dexippus. Dexippus’ work evidently used Athenian archons and Roman consuls as its chronological frame, presumably with a king-list for earlier periods. Eunapius (F1 Blockley) in his proem suggests that he has learnt from his predecessor that writers of annalistic history are forced to admit that their narratives are chronologically imprecise: Dexippus himself, he claims, described his chronicle as ‘wandering and full of contradictions, like an unchaired assembly’. Eunapius proceeds to dismiss excessive precision about hours and days as fitting only for accountants and astrologers: ‘What contribution do dates make to Socrates’ wisdom?’ He even appeals in support of his position to Thucydides’ account of the dispute over Scione, which he distorts entirely by claiming that the disagreement was unsolvable. While this sort of distortion of historiographical predecessors was not uncommon, Eunapius was probably unusual in his willingness to profess his neglect of chronological accuracy for the relatively recent past. The problematic dates to which Dexippus himself alluded presumably sprang from the large scope of his work: as Livy too noted (From the Foundation of the City 2.21.4), it could be hard when treating distant times to know to which year events belonged.43
Besides facilitating anachronism as chronological error, ancient dating systems could reflect a sense of anachronism as historical difference. Part of the difference of the distant past for ancient historians was the difficulty of discovering reliable information about it: many ancient writers evoked this with the imagery of darkness,44 while Plutarch compared undertaking his Theseus with travel to the uncertain spaces on the edges of maps (1.1). The point at which knowledge became more secure did, however, vary from author to author and region to region. Some historians (Thucydides among them) were prepared to offer year-dates after the Trojan War; others (for instance Diodorus) started from the return of the Heraclids (placed by Thucydides eighty years after the Sack of Troy); others again rejected reliable records before the first Olympiad (e.g. Julius Africanus F34 Wallraff). Castor of Rhodes, by contrast, provided year-dates before the Trojan War using the king-list of Sicyon; his method was developed by Christian chronographers such as Eusebius, who synchronized regnal years at Sicyon, Argos, Mycenae and Athens prior to the Trojan War. Similar variety can be found in Roman historians: Livy, for instance, offers for the regal period reign-lengths without much internal precision; it is only with the start of the Republic that he adopts an annalistic structure. A sense of when reliable chronology becomes possible can also be revealed by patterns of incidental dates within works that lack a strong chronological frame: Phlegon of Tralles, for instance, author of a book of marvels in addition to a chronology, offers a narrative of sex changes in which he presents no dates for Tiresias and Caeneus, whose bodily transformations resulted from divine intervention, while offering archon- and consul-dates for some more recent sex changes accomplished without divine help (evidently cases where intersex children were re-classified at puberty) (FGrH 257 F4.6–10).
The difficulty of dating by years in early periods is reflected in the continuing use of generations as chronological signposts. Dating by generations, found to a limited extent in the Homeric poems, structured the heroic age in the Hesiodic Catalogue of Women and again in the more systematic chronologies found in early prose works such as Hecataeus’ Genealogies (early fifth century BC); it also played an important role in Herodotus’ structuring of the more recent past. Generations continued to be useful chronological tools, moreover, even after some chronologers had offered year-dates for the heroic age: thus Pausanias (10.17.4) rejects as ‘nonsense’ a story that Daedalus took part in a colonizing expedition with Aristaeus since Daedalus lived ‘at the time of Oedipus’ while Aristaeus married the daughter of Cadmus (the unstated part of the argument being that Cadmus was Oedipus’ great-great-grandfather);45 similarly he rejects a local Laconian tradition that Achilles was one of the suitors of Helen (3.24.11) by appealing to explicit indications of generations in the Homeric poems.
The changing nature of the historical record was itself a basis for periodization. Particularly notable is a passage in Censorinus (On the Birthday 21) to which Scaliger was repeatedly drawn. Censorinus records three ‘distinctions of times’ (discrimina temporum) defined by Varro, who was himself probably drawing on a Hellenistic chronographer such as Eratosthenes: the first interval, ‘from the beginning of humans to the earlier flood’, was ‘called “unclear” (adēlon) because of our ignorance’, and was of uncertain length (perhaps even infinite); the second, ‘from the earlier flood to the first Olympiad’, was ‘called “mythical” (muthikon) because many fabulous things are reported in it’, and was believed to be around 1,600 years; the third, ‘from the first Olympiad to us’, was ‘called “historical” (historikon), because the events done in it are contained in true histories’. The interest of this passage for Scaliger lay initially in the fact that the first Olympiad was the basis for the reckoning of times;46 subsequently he suggested that ‘heroic’ was a better label than ‘mythical’ for Varro’s middle period (a small tweak, as Varro equally allowed for a substantial non-fabulous element in this period) and, more daringly, equated the first two periods with the new concept of ‘proleptic’ time that he developed in response to Babylonian and Egyptian king-lists that stretched beyond the Biblical creation date.47
Variations in dating systems expressed a sense of political development as well as the changing contours of historical knowledge. Eusebius tracks political change at Athens by successively dating years through kings, life-archons and ten-year archons; this schematic and probably unhistorical progression culminates in the establishment of the annual archonship, at which point he drops Athenian dating from his parallel columns. The Athenian annual archonship nonetheless retained its use for historians such as Diodorus and Dionysius of Halicarnassus, partly because it was used by the Athenian Apollodorus (second century BC) in his very influential verse chronology, partly as homage to Athens’ past cultural superiority; it remained so familiar that the geographer Pausanias felt the need to explain the changes in the archon system when he synchronized an early Olympiad dating with the fifth year of a (ten-year) Athenian archon’s office (4.5.10).
The potential of chronology itself to mark difference is well illustrated by the Parian Marble (FGrH 239) (Figs. 6 and 7), a chronological list inscribed on stone on the island of Paros in the third century BC. This inscription uses two different systems: it dates events back by years from the archonship of Diognetus in Athens and his equivalent in Paros (264/3 BC), using a system of acrophonic numerals (i.e. with thousands, hundreds, fifties, tens, fives and ones marked by the first letter of the Greek word for the number);48 and it names the Athenian king or archon in that year (using the term ‘king’ for life-archons too, while omitting the ten-year archons and marking the start of annual archons as an event in itself). Only three times does it mark the regnal year: for the start and end of the Trojan War and the foundation of Syracuse (the final event listed before the annual archonship). The status of the Trojan War as a chronological marker is further highlighted by the unique inclusion of a calendar date (with Athenian months) for its fall and possibly by an empty space on the inscription (the relevant part has been lost since its transcription in the seventeenth century).49
Fig. 6 Middle section of chronographic inscription from Paros, known as the Parian Marble, c. 266/5 BC. ANC Chandler.2.23. © Ashmolean Museum, University of Oxford.
Fig. 7 Detail of text from the Parian Marble, showing the use of both acrophonic numerals (e.g. Δ = 10, H = 100) and archon names for dates. © Ashmolean Museum, University of Oxford.
External political change, too, could be reflected in changes of dating systems. Eusebius’ synchronistic tables tracked the rise and fall of kingdoms: the parallel columns become fewer and fewer until the Roman conquest of Judaea, at which point only Rome is left.50 The rise of Rome similarly underlies a change in the system used by Polybius. For most of his work he adopts an annalistic method, dating by Olympiad year, but at the very start of his main narrative he treats in single units events in different regions in the four years of the 140th Olympiad. This variation is linked to his key thesis that the rise of Rome led to an unprecedented ‘interconnection’ (sumplokē) of events throughout the inhabited world in the third year of the 140th Olympiad; previously, he explains, he introduced synchronisms purely to clarify the chronological link between events (4.28).51
Accuracy in synchronism is a key concern of the various dating systems we have explored in this section: Thucydides synchronizes local dates for the start of the Peloponnesian War; Polybius offers explicit synchronisms of (initially unconnected, later connected) events occurring in different parts of the Mediterranean; Eusebius combines dating by various regnal systems with continuing use of the era of Abraham’s birth. As we shall see in our next section, this interest in synchronism went hand-in-hand with concern for its flipside, anachronism.
In the preface to the posthumous edition of De emendatione, Joseph Scaliger wrote of his ambition to save chronology from the ‘ravings of the ignorant and the tyranny of sophists’ by preparing the ground for the completion of a ‘perfect Chronicle’ (Chronicon absolutissimum). A key step towards that goal was the commentary on Eusebius’ Chronicle that he provided in Thesaurus temporum – a series of detailed notes on those entries (and there were many of them) where he detected chronological problems. The commentary was throughout embellished with his new typology of anachronism. At times, as we have seen with Athens’ Sicilian disaster, he identifies prochronisms (entries placed too early). Entries placed too late include those for years of Abraham 1701, where Plato’s nephew Speusippus is listed as a distinguished philosopher, and 1859, where Terence’s death is announced (metachronisms of twenty-four years and one year respectively).52 At times, too, Scaliger uses ‘anachronism’ without any indication of years: the report ‘Theseus seized Helen’, placed by Eusebius in the thirteenth year of Theseus’ reign, is a ‘ridiculous anachronism’, since ‘Theseus seized Helen during the reign of his father Aegeus’.53 Often, too, he departs from his typology by indicating chronological mistakes through periphrastic expressions: the time allotted to Phoenix and Cadmus ‘is not congruent’ with that assigned to Busiris; ‘the epoch of Capua’s foundation … is thirty-four years before the time assigned by Eusebius or rather by Diodorus’.54 Again, there are entries, such as that for Hannibal’s death, where he simply notes the existence of conflicting sources.55 And occasionally he even explains that Eusebius got the date (more or less) right, as with the placing of the poet Terpander twenty-nine years after his victory at the Spartan Carneia.56
Scaliger’s commentary on Eusebius offers a snapshot of the scholarly laboratory in which the chronology of the ancient world was distilled from a vast range of famous and obscure authorities. It is not that he was the first to attempt the task of chronological reconstruction in early modern times: the correct BC date of one key epoch, the first Olympiad, had been established slightly earlier in the sixteenth century by Paul Crusius.57 In many cases, indeed, Scaliger was picking up ancient controversies over dates – while also, as we shall see, suppressing in the cause of chronological precision some of the distinctive ideological disputes that lay behind those controversies.
Let us start with the date of Hannibal’s death. Scaliger notes that three possible consul-years are preserved for this date; for two of these he names sources (the late antique historians Eutropius and Orosius). He omits to mention that the first-century BC biographer and chronologer Cornelius Nepos (Hannibal 13.1) had offered the same three consul-years, with a source for each possibility: the Annals of Cicero’s friend Atticus (FRHist 33 F8), Polybius (23.13) and Sulpicius Blitho (FRHist 55 F1), a writer known only from this citation. For a sense of what was at stake in the dispute, however, we need to turn to Polybius (the one author named by Nepos whose account survives independently) and Livy (who wrote after Nepos). Polybius places Hannibal’s death in the same year as the deaths of Scipio, Hannibal’s great adversary in the Second Punic War, and of his own political hero Philopoemen, who had attempted to defend Greek independence in the face of Roman expansion; he further stresses the synchronism by writing parallel obituaries for the three men (23.12–14). While Polybius was evidently trying to boost the standing of his fellow Achaean Philopoemen, Livy rejects the synchronism with detailed arguments for the date of Scipio’s death (directed against Polybius and the historian Valerius Antias (FRHist 25 F56)) based on the dates of his replacement as princeps senatus and of a court case in which he was involved.58 Livy then offers a different basis for comparison: ‘the deaths of these three men, each the most famous among his own people, seem comparable not so much because of the coincidence of their times, as because no one of them met an end worthy of the brilliance of his life’ (39.52.1–3). At stake in the debate over Hannibal’s death, then, was the use of synchronism as a means both of pattern-finding in history and of ethnic self-definition.
Disputes over Helen’s dates, by contrast, reveal normative assumptions about gender. One important chronological problem was addressed by Eusebius himself in an entry where he places the Argonautic expedition under the year 756. Noting that Castor and Pollux were included among the Argonauts, Eusebius queries how Helen could be their sister, ‘as she was seized as a virgin by Theseus many years later’ (he puts her rapes by Theseus and Paris in years 795 and 826). The obvious solution (which Scaliger adopted59) was to place the Argonautic expedition much closer to the Trojan War. But even with that chronology, fixing the date of Theseus’ rape of Helen posed difficulties: while Scaliger, as we have seen, placed this event before Theseus became king, many ancient writers followed the version of Hellanicus (FGrH 4 F168a, from Plutarch, Theseus 31.1), according to which Theseus was fifty years old at the time of the rape – and all the more reprehensible on that account: thus Plutarch (Syncrisis of Theseus and Romulus 6.1) blames Theseus for seizing Helen when ‘she was an unripe child, while he was already of an age too great for even lawful wedlock’.
The chronology that Eusebius questioned could also be put to moralizing ends. Lucian’s dialogue Cock starts with a poor man who dreams of wealth complaining of being woken up by a cock. The cock, it turns out, is none other than the latest incarnation of Pythagoras – who had famously claimed to be the Trojan warrior Euphorbus in a still earlier incarnation. As part of a typically Cynic diatribe against wealth, the cock rejects the version of the Trojan War told by Homer (who was a camel in Bactria at the time) – including the reputed beauty of Helen: ‘she was very old, almost the same age as Hecuba, inasmuch as Theseus, who lived in Heracles’ time, seized her first and held her at Aphidnae, and Heracles captured Troy on the former occasion in our fathers’ time, our then fathers that is’ (22.17). Lucian’s engagement with chronology here is strictly ad cock: elsewhere he plays on the contrast between her beauty at the time of the Trojan War and her sorry state in the Underworld, exposing the futile sacrifice of so many lives for her sake (77.5).
Different issues are involved in ancient debates on the foundation of the Campanian capital Capua. Scaliger’s view that Eusebius dated the foundation of this city thirty-four years late applies to a notice for the year 1580 that refers more generally to the unification of the Campanians.60 The date Scaliger calculates is based on evidence from Cato the Elder (FRHist 5 F52) that is preserved by the first-century AD historian Velleius Paterculus. Applying his usual method of dating back from the consulship of Marcus Vinicius (AD 30), Velleius counts 240 years to the sack of Capua by the Romans while also citing Cato’s claim that the city’s sack happened 260 years after its foundation. What Scaliger fails to note is that Velleius alludes to the foundation as an event ‘in which there has been much error and in which the views of the authorities show great discrepancy’. Velleius himself agrees with the ‘vastly different’ view that Capua was founded 830 (rather than 500) years before his own time: ‘I can hardly believe that such a great city grew, flourished, fell, and rose again so quickly’ (1.7.2–4). Velleius’ reasoning here is based on an a priori assumption about the rhythms of history that is close to the model of anachronism developed by Vico in the eighteenth century (p. 24): just as Vico detected an anachronism in putting events from the ‘age of gods’ in the 200 years of the ‘age of heroes’, so Velleius suggests that Cato’s date for Capua’s foundation puts too many events into 500 years. Vico himself listed the Capuans as one of three peoples conquered by Rome who had resisted the natural course of human civil institutions – in their case because they ‘developed too rapidly because of their mild climate and the abundance of fertile Campania’.61
Scaliger’s notes on Eusebius frequently reflect one of the key areas of chronological interest in antiquity, literary history. From the fifth century BC onwards, there was a persistent interest in the dates of Homer and Hesiod;62 from the Hellenistic period onwards there were biographies of them, as of early Greek poets, generally with little firm evidence to go on. Divergent views on Homer’s dates were grist to the mill of Christian apologists such as Tatian (To the Greeks 31) and Clement (Miscellanies 1.117), while their near-contemporary Pausanias was so alarmed by the ‘censoriousness’ of present-day professors of poetry that he refused to publish his own research on the issue (9.30.3). One place where Scaliger pays attention to this Homeric question is an entry for the historian Herodotus (prochronistically placed in year 1548, i.e. 468 BC). Scaliger digresses to note that Herodotus claimed in his Histories that Homer and Hesiod lived ‘not more than 400 years’ before his own time (2.53), but that his biography of Homer gave an interval of 622 years between his life and Xerxes’ crossing from Asia to Europe (in 480 BC). While modern scholars reject the attribution of the biography to Herodotus, Scaliger’s solution was to emend the text to 422.63
The complexity of literary chronology can be seen equally well in the case of the poet Terpander. As we noted, Scaliger accepted Eusebius’ dating for Terpander as approximately correct owing to its proximity to the first Carneia. A passage in Athenaeus – a small diversion in a long discussion of the disputed chronology of certain musical instruments – reveals how comparison of different scholarly sources on the history of poetry could throw up very different datings (14.635e–f):
Terpander was the first victor at the Carneia, as Hellanicus reports in both his metrical and prose Carneian Victors (FGrH 4 F85a). The Carneia was established in the twenty-sixth Olympiad, as Sosibius states in his On Dates (FGrH 595 F3). But Hieronymus in On Citharodes – the fifth part of On Poets – says that Terpander lived at the time of Lycurgus the lawgiver, who is reported by all without disagreement to have organized the first numbered Olympics together with Iphitus of Elis (fr. 33 Wehrli).
Athenaeus’ discussion shows that Terpander’s link with the Carneia clashed with a different synchronism, with the lawgiver Lycurgus, who flourished over a hundred years earlier. Other sources muddy the picture further by showing that Lycurgus’ date was much more contested than Athenaeus implies (Plutarch begins his Lycurgus by stating that ‘there is nothing to say that is undisputed … least of all the dates when the man lived’).64
Even though biographical traditions about early Latin poets were recorded much sooner after their deaths, they too gave rise to chronological disputes. Cicero’s dialogue on the history of Roman oratory, Brutus, attests early and late chronologies (by Varro and Accius respectively) for one of the first figures of Latin literature, Livius Andronicus (72–3); accepting the former, Scaliger blames a metachronism of fifty-four years in Jerome’s entry on his ‘haste, the mother of hallucination’.65 For the death of Terence, on the other hand, Scaliger notes that Jerome follows Suetonius’ Lives of the Poets, albeit with a metachronism of one year. But he omits to note three scandalous stories which Suetonius mentions – but rejects on chronological grounds. One of these stories was that Terence was born a slave at Carthage and came to Rome as a prisoner of war; Suetonius dismisses this story on the grounds that Terence was born before the Second and died before the Third Punic War, and that the Romans did not at that time trade with Carthage’s other enemies (Terence 1). Terence’s date of birth was also cited against claims that he had been the younger lover of the Roman aristocrats Scipio Africanus and Laelius and that he had received help from them in writing his plays (Terence 1, 4, citing the historian Fenestella and grammarian Santra); the basis for these rumours seems to have been in the one case a scurrilous poem cited by Suetonius and in the other some oblique lines in the preface of one of Terence’s own plays (Brothers 15–21). These anachronisms follow a common pattern whereby famous figures are attached to other significant figures or events: there were similarly stories that the comic playwright Eupolis died at sea during the Sicilian expedition and that the Sicilian philosopher Empedocles fought in that war – stories rejected on the grounds that Eupolis had written plays dated later (Eratosthenes FGrH 241 F19) and that Empedocles was either dead or too old to fight at that time (Apollodorus FGrH 244 F32).66 At the same time, the disputes over Livius Andronicus and Terence reflect a pervasive Roman interest in mapping the (belated) development of their own culture against Greek literary history.
The propensity for anachronism in the biographical tradition about philosophers was equally strong. Scaliger notes that Speusippus, who succeeded Plato as head of the Academy, is mentioned as having died twenty-four years prior to his appearance in Eusebius as a ‘distinguished’ philosopher in year 1701; indeed, he is mentioned as ‘distinguished’ already in year 1622. Scaliger thought that such doublets resulted from Eusebius’ use of different sources; the variation among the sources may itself have reflected either attempts to attach dates to free-floating anecdotes or the influence of deliberate comic fictions. The particular interest in the chronology of philosophers lay in the importance of teacher–pupil relationships in ancient biographical traditions (as reflected, for instance, in Diogenes Laertius’ Lives of the Philosophers, which is structured by the various philosophical schools that claimed descent from Socrates). The attempt to map chains of pedagogical influence gave rise to many suspect ties: claims that Themistocles and Thucydides were taught by the philosopher Anaxagoras were rejected on chronological grounds by Plutarch (Themistocles 2.5) and the Byzantine scholar John Tzetzes (scholia on Thucydides 8.109, responding to Marcellinus, Life of Thucydides 22) respectively.67
The ideological implications of pedagogical anachronism are highlighted by a short treatise by Isocrates, Busiris. The work is in the form of a letter to a sophist Polycrates criticizing his paradoxical Defence of Busiris, a king of Egypt who was said to have sacrificed foreigners until, in one version, he was killed by Heracles. Isocrates says that Polycrates made the Egyptian even worse than traditional accounts; he counters by presenting Busiris as the founder of a political order that bears some resemblance to the imaginary city Callipolis described in Plato’s Republic. The charge of anachronism is raised against Polycrates twice. First Isocrates rejects Polycrates’ claim that Busiris emulated Aeolus and Orpheus, whose deeds were the opposite of his and whose fathers were not yet born in Busiris’ time (7–8). Later he argues that the claims made about Busiris’ misdeeds fail on chronological grounds because they are linked with the story that he was killed by Heracles, ‘but it is agreed by all writers that Heracles was four generations younger than Perseus, son of Zeus and Danae, and that Busiris was older by more than 200 years’ (36–7). While, as often, claims about the universal agreement of chronographers deserve some scepticism (Eusebius offered up, to Scaliger’s bafflement, a ‘first Hercules’ even earlier than Busiris68), Isocrates’ rejection of the claim that Busiris was influenced by non-Egyptian figures and his dismissal of the Heracles story together prepare for the hints that Busiris’ own political measures have been copied by others.69
The stakes become higher still when chronologies have to deal with different cultures’ perceptions of the divine. Herodotus tells how the historian Hecataeus, during a visit to a temple in Egyptian Thebes, boasted of being descended from a god in the sixteenth generation; the priests responded by showing statues of the holders of a priesthood that had been handed down from father to son for more than 300 generations, with no gods in sight (2.143). Herodotus in addition dismisses as far too late the dates assigned by the Greeks to Dionysus and Pan (1,000 and 800 years before his own time), suggesting instead that those were the dates when they first received knowledge of these gods from Egypt (2.145). And when he hears of a Heracles who was deified some 17,000 years earlier, he infers that this Heracles is separate from ‘the other Heracles, whom the Greeks know’, that is, the son of Amphitryon whom they worship as a hero (2.43).70 Herodotus, that is, undermines Greek pretensions by exposing the anachronisms in their views of the gods and their relative belatedness by comparison with the Egyptians: it was, he writes, ‘yesterday or the day before (prōēn te kai chthes), so to speak,’ that they learnt about the gods from Homer and Hesiod (2.53).71
With these Herodotean musings on the chronologies of the gods we return once more to concerns central to Scaliger’s chronological researches. On the title-page of the first edition of De emendatione, Scaliger quoted a dictum of the second-century Christian theologian Tatian on the impossibility of historical truth when ‘the recording of dates is incoherent’ (To the Greeks 31). In context, that dictum was directed against the contradictory chronological records of the Greeks by contrast with the secure antiquity of Moses and Jewish law – a common refrain of the early Christian chronographers. While Scaliger himself avoided the polemical anti-paganism of writers such as Tertullian, who gave the earlier Hebrew prophets credit for Solon’s exhortation to Croesus to look to the end (Apologeticus 19), he did in his later commentary on Eusebius praise Eusebius’ success in proving the priority of Jewish traditions even while he showed up anachronisms within Christian chronography. Enemies of Christianity, he noted, used the age of their own gods as a weapon against Christians, suggesting that ‘Christianity appeared yesterday and the day before (chthes kai prōēn)’72 – a pointed echo of Herodotus’ comment on the recentness of Greek knowledge of the gods. In response, Christian writers such as Tatian and Clement dated Moses to the time of Inachus, first king of Argos, but (and now Scaliger shifts to first-person possessive adjectives) Eusebius had not thought it enough that ‘gentiles be overwhelmed by their own testimonies, unless they could be refuted by ours too’ (that is, by the ‘sacred page’). Eusebius’ achievement, Scaliger concluded, had been to draw on a wider range of evidence in placing Moses 400 years later – but still to show the greater antiquity of the Jewish traditions.73
Our analysis in this chapter has suggested that a sense of historical process was at stake in many of the pre-Christian tussles over time-reckoning. We have encountered many authors in both the Greek and Roman worlds who presented calendrical systems as developing from simple to complex or whose approach to chronology modelled their understanding of political change. Even seemingly pedantic disputes over isolated chronological difficulties often had ideological and ethical sub-texts, not least when they involved clashes between different religious and cultural systems. In particular, Egypt was in some respects for Greece what Greece was later to be for both pagan and Christian chronologers under the Roman empire – a site of chronological comparison that exposed the uneven rhythms of historical development.74 As we shall see in our next chapter, this perception of uneven rhythms was accompanied by a heightened awareness of the possibility of cultural outdatedness and obsolescence – concepts that over the past 200 years have come to be seen as the definitive characteristics of anachronism itself.