Anachronism in Oxford: the case of the Marmor Parium

Oxford’s Ashmolean Museum is home to one of the more intriguing objects to have survived from the ancient world, the central fragment of the Marmor Parium, a historical chronicle inscribed on stone. The Marmor Parium, regarded by the museum as one of its ‘greatest treasures’, has long been a focus for explorations of ancient historiography and questions of anachronism in ancient texts, from the time its text was first published in the 17th century. Set up on the island of Paros during the third century BCE, the inscription lists events from Greek myth and history, starting with the accession of King Cecrops, the half-man, half-serpent first king of Athens (in our 1581 BCE), and finishing with the conflict between Demetrius and Cassander, in a series of largely formulaic entries referred to as ‘epochs’; here is the entry for the end of the Trojan war, in Rotstein’s translation.

24. From the time Troy was conquered, 945 years (= 1209/8 BCE), when [Menesthe]us was king of Athens, in his ⟨twenty⟩ second year, in the month of Th[argeli]on, in the seventh day, (counting) from the end of the month.

The last readable entry is for our 299/8 BCE, but most of its text is damaged or missing, and it’s possible that the end of the inscription is likewise lost. The Oxford section includes the entries for the dates 895-355 BCE.

Jacoby’s drawing of the Oxford section of the Marmor Parium (IG XII.5 444).

The Marmor Parium offers some intriguing insights into ancient thinking about the past, as well as raising many questions (for example, quite why an Athenocentric history should have been inscribed and displayed in Paros). With its long chronological span stretching deep into the past, it has been an invaluable document for thinking about problems of ancient chronography, despite the brevity of its entries and its focus on literary rather than political history. It combines two dating systems, one with years expressed in numbers counted backwards from a fixed point, the time of composition, and one with years identified by king or archon. Genealogy and chronology run in parallel, although the former changes gear in line with political changes, and the latter has many peculiarities.

The inscription has long been studied in Oxford, where Marmora Oxoniensa, edited by Joseph Chandler (1737-1810), was published in 1763, containing an improved version of its text. This edition was used by an early commentator, Joseph Robertson (1726–1802), who had concerns about the authenticity of the inscription. Some of these arose from its chronological errors: these included ‘prochronisms’ where events were placed too early (such as the birth of Euripides, in epoch 51), and ‘parachronisms’ in the Sicilian events of epochs 53 and 55 where the temporal confusion is broader (Robertson 1788: 166-7). Robertson is more broadly concerned with authenticity beyond this particular text; he also discusses the poems of Thomas Chatterton (1752-70), which he feels display ‘some apparent anachronisms’ (Robertson 1788: 204), which had recently been revealed to be Chatterton’s work rather than those of a mediaeval bard. It is intriguing to think that our project echoes the interests of these early scholars.

robertson parachronism
Robertson on parachronism, from The Parian Chronicle, 1788, p. 167.

Despite the gaps and losses, the Marmor Parium’s text shows how an ancient chronicle can combine past times and spaces that we would regard as quite distinct in kind into a single narrative structure – the spatium mythicum, a world in which the king of Athens can have serpent form and the spatium historicum, a world in which the city is a trophy for the warring successors of Alexander the Great’s disintegrating empire. In this sense, the structures modern historiographers attempt to impose on Greek accounts of the past, and the distinction between historiography and mythography, look as if they might themselves be anachronistic retrojections alien to the ancient sources.

While the Marmor Parium’s apparently unbroken chronology suggests continuity, recent research has pointed to subtleties within its narrative and language that mark some changes in the style of its account. Veit Rosenberger finds evidence in the chronicle’s entries of the ‘floating gap’ between the mythical and historical past; following the details of various events we treat as mythical, the period between 1202/1 BCE and 604/3 BCE has very few entries, but then more is recorded for subsequent years. Rosenberger argues that the second of these shifts in the frequency of recorded events marks the starting point of Greek literary history, possibly in the work of the historian and mythographer Hecataeus. The stone therefore encodes a frozen ‘floating gap’ that marks the start of Greek written historical accounts. The second section of the stone, the Paros fragment, covers dates that fall within the 80 years before the chronicle’s composition, and thus within the scope of oral history at the time of composition. But a physical gap of text, covering a critical 19-year period, lost between the Oxford and Paros sections makes it impossible to identify the exact date at which this increased level of detail begins.

The afterlife of the Marmor Parium is as intriguing as its origin. The first section was acquired by Lord Arundel’s agents in Smyrna (and so divorced from its archaeological context) and arrived in England in 1627, and drawings and transcriptions were made and published soon afterwards in Marmora Arundelliana, along with the rest of the Arundel collection of classical inscriptions, by John Selden. Selden’s publication is the only record for this section, which was lost between 1627 and the donation of the Arundel collection to Oxford in 1667, most likely during the turmoil of the Civil War; it was possibly used as building material to repair Arundel House. Editors ever since have been striving to improve Selden’s text, occasionally with enthusiastic supplements and emendations. More recent editors wish, anachronistically, that Selden, and indeed Felix Jacoby (1876-1959) in his two editions, had been able to use Leiden convention markings for doubtful characters and spaces (Rotstein 2016: 17-20).

A further section of the chronicle, covering the dates 336 BCE to 299/8 BCE, was discovered on Paros in 1897, sparking a further flurry of editions and commentaries. That section is now on display in Paros; as Rotstein observes, the history of the marble (variously identified as Arundellian, Oxonian, and Parian) is itself a microcosm of the history of the ‘early European appropriation of antiquities’ (Rotstein 2016: 5). Museum visitors, whether in Paros or Oxford, may wonder how much more legible or accessible the stone, with its tiny lettering, was to its original readers in Hellenistic Paros.


Counting backwards: genealogy and anachronism

The shifting boundary between the near and the distant past is blurred by ancient Greek writers when they establish temporal frameworks by counting backwards in years and, once a more distant and less well-known period is reached, generations. With multiple lists in operation – one for every city and Panhellenic temple, victors from the Olympics – and with the genealogies of royal dynasties stretching back to incorporate divine ancestors, there was plenty of opportunity for the manipulation of lists, for error and invention, and for debates about accuracy. Generating synchronisms, placing the same event on points in different lists, was a particular challenge for historians, and so became a site of historiographical criticism. Failed synchronisms and arguments about them result in a type of anachronism that is characteristic of Greek historiographic debate and spills over into other genres whenever the past is debated, as their use by both Thucydides and Isocrates shows.

For writers of contemporary history such as Thucydides, the use of officer lists based on the records of cities is transparent and supported by documentary evidence. While Thucydides organises his account of the Peloponnesian War by seasons, he uses the officer-list system to establish its start date (2.1-2.2.1), and his Athenian readers could have referred to an inscribed version of the list (IG I 3 1031) that had been set up in Athens during the later part of the fifth century, possibly as late as 410 BCE:

My account sets out the events in chronological order, by summers and winters. The Thirty Years Treaty agreed after the conquest of Euboea lasted for fourteen years. In the fifteenth year, when Chrysis was in her forty-eighth year as priestess at Argos, Aenesias was ephor in Sparta, and Pythodorus had two more months of his archonship in Athens, in the sixth month after the battle at Potidaea, and at the beginning of spring, in the first watch of the night an armed force of slightly over three hundred Thebans entered Plataea, a city in Boeotia allied to Athens. (Thucydides 2.1-2.2.1, translation Hammond)

inscription fragment
Athenian Archon List (527/6-522/1 BC?) IG I3 1031, fragment c. Agora Museum, Athens (I 4120); squeeze from Centre for the Study of Ancient Documents, Oxford.

Thucydides found the practice of using office-holder lists to establish dates imprecise (5.20.2-3), and suggested his own improvements for greater precision in narrative and analysis. But not all historians were, like Thucydides, concerned with the very recent past. The further back in time the Greeks went, the less precise their dating systems became. In classical Athens, for example, before the list of annual archons began, there were lists of officials with longer terms in office, and kings stretching back to Kekrops, the half-snake figure who emerged from the ground to found and rule the city. The kind of information in the lists changes as they go back in time, a change perhaps marked by a shift from years to generations as the unit of counting, and the separation of these distinct lists in the Athenian tradition; although the origins and development of the lists are unclear, later writers transmitted complete versions of them.

Making temporal connections within the distant past posed a challenge. Genealogies, the typical form of lists from the distant past, could be used to establish the kind of synchronism that Thucydides uses at the start of his histories. This process permitted synchronism between the foundation myths of different poleis and the characters of different mythical cycles, but here there was less possibility of consulting records and more reliance on the conventional form of mythical narratives.

Arguing for and against the accuracy of synchronisms between myths became an important mode of criticism as such stories were used as examples in political contexts to establish relationships between cities. There were many possible causes of error; the corruption of ancestry lists, the need to establish synchronism with significant events such as the return of the Heraclids from the Trojan War to the Peloponnese, and the urge to assert priority for one’s patriotic account of civic origins. Myths involving culture heroes such as Heracles and Theseus generate problematic synchronisms as writers try to fit them into coherent narrative frameworks, or to establish a claim to temporal priority. Xenophon, for example, asserts that Lycurgus established the Spartan constitution at the same time as the return of the Heraclids (Xen. Lac. Pol. 10.6), a claim that is contrary to other developmental accounts of Spartan history that place Lycurgus after the early (mythical) history of the Dorian League founded by the returning Heraclids; Plato in his Laws suggests that the Lycurgan constitution resolved the problems of this earlier period, but again myth and history intertwine in a complex way (Pl. Leg. 3.683c-693c).

Arguing with myth in this way provided opportunity for parody and the comic repurposing of mythical material for rhetorical argument. Reading Isocrates’ Busiris, a complicated and paradoxical text that still puzzles commentators, shows how the critique of a claimed synchronism can be used as the starting point for broader criticisms. Isocrates is ostensibly writing to Polycrates the sophist, to point out errors in his defence speech in praise of Busiris, the infamous Egyptian king of heroic times. Busiris was said to sacrifice his guests, and was eventually killed by Heracles, as the Greek culture hero avoided this grisly fate, a scene frequently depicted on Greek vases.

vase painting - Heracles and Busiris
Heracles killing Busiris. Attic red-figure kalpis (hydria), ca 480 BCE. From Vulci. Staatliche Antikensammlungen 2428.

Isocrates aims to show that earlier writers on Busiris have got their genealogical calculations wrong, and that the encounter between Heracles and Busiris could not have happened. Polycrates could have used this simple method in his defence of Busiris:

Furthermore, it could be easily proved on chronological grounds (tois chronois) also that the statements of the detractors of Busiris are false. For the same writers who accuse Busiris of slaying strangers also assert that he died at the hands of Heracles; but all chroniclers agree that Heracles was later by four generations than Perseus, son of Zeus and Danaë, and that Busiris lived more than two hundred years earlier than Perseus. (Busiris 36-37, translation Van Hook)

The evidence that the criticism of Busiris is misplaced in time is a piece of evidence (pistis) that is clear (enargê). Of course, the idea of establishing an accurate genealogy of characters from the far-distant past that operates in a similar way to the chronology of a contemporary historian like Thucydides is itself rather paradoxical and a long way from being ‘clear’; Although there were standard exchange rates between years and generations, Isocrates’ use of both emphasises the paradoxes involved in chronological calculations of mythical narratives. Isocrates’ subsequent comments expand the related problem of the difficulty of knowing the different past, given the lurid slanders (blasphêmiais) written by poets that attribute all kinds of bad behaviour to the gods (38-40).

Isocrates’ real target in the Busiris is probably not the obscure sophist Polycrates but his rival educator Plato. The criticism of poets for slandering the gods is reminiscent of Socrates’ arguments against poetry in the Republic. Using Heracles and Egypt to think about the possibility of knowledge of the past further links both writers, via Herodotus’ challenges to the genealogical reckonings Greeks used to date Heracles (2.142-6); Plato echoes this passage in his Timaeus-Critias, with Solon replacing Hecataeus as the Greek visitor to Egypt.

With the traditional criticism of Busiris destroyed by chronology, Isocrates aims to show instead that Busiris should be praised for establishing the Egyptian constitution. However, the constitution that Isocrates describes bears a detailed resemblance to that of Plato’s Kallipolis in the Republic. Plato, Isocrates seems to be hinting, has not revealed a timeless ideal of how we should live, but has borrowed from a historical model itself established by a notorious barbarian and in existence at a time that has been identified with precision.

Isocrates’ manipulation of history and myth in the pursuit of political argument is a feature of his work that deserves more exploration, along with the construction and argumentative use of genealogies and temporal frameworks by other Greek writers. I will continue to investigate the political activation of anachronism in imaginary time by Greek historians and political theorists as our project continues.