The ancient sources of information for this tangled, confused succession of events are very numerous indeed, and one of the historian’s principal problems is to detach what is important from a mass of detail. And yet all these sources are too fragmentary to be adequate. ‘The period is like a dark tunnel, illumined from either end, and by rare and exiguous light wells in the interval.’7 Those words were written about the central half-century of the epoch (235–84), but even the illumination from either end, though perhaps bright in contrast with the obscurity in the middle, is only fitful and flickering.8 Knowledge of what happened has to be built up by a process of jigsaw puzzle and inference far more difficult than reconstructions of the earlier world of Rome or the later empire of Constantinople.
The most serious deficiency is the absence of any comprehensive account by an ancient historian of distinction. The earlier years are to some extent covered by surviving portions of the account written in Greek by Dio Cassius of Nicaea (Iznik) in Asia Minor. A senator and holder of high office, he wrote a history of Rome from its beginnings to AD 229, in eighty books. But for the last hundred and eighty-three years (except for incomplete versions of the two final books) we possess his history only in the form of an eleventh-century compendium,9 and that too is lacking at times. Dio Cassius was a contemporary and eyewitness of events after 18010; he is anecdotal and reminiscent and undetached, and that is both his value and his weakness. Herodian, a Hellenized Syrian, wrote an eight-book history of the years 180–238. Repetitive, pompous, superficial and inaccurate, he has preserved a certain amount of street-corner gossip and gives us a third-century view of the passing of the Antonine Golden Age. There are also other Greek historians of later times who incorporate a good many isolated facts and events that would otherwise be lost.
The Latin Historia Augusta, ostensibly the work of six authors supporting the senatorial aristocratic tradition, is a collection of lives of emperors from Hadrian to the accession of Diocletian (117–284), though the sections from AD 244 to 253 are missing. Several of the biographies purport to be dedicated to Diocletian and Constantine, but internal evidence suggests that the work, at least in the form in which it has come down to us, is later than their age by half a century, or a century, or even two centuries. Although this indication that the ostensible authorship is false makes the whole work highly unreliable as a historical source, the major second-century lives in the Historia Augusta include in haphazard form much material, not found elsewhere, which may tentatively be regarded as factual. Probably this goes back in large part to Marius Maximus, the author of a sequel to Suetonius’ Twelve Caesars, who has been identified as one of Septimius Severus’ generals. The Historia contains a disproportionately long biography of Severus Alexander (222–35) which may be a remodelled account of the life of Julian the Apostate (361–3); and most of the other third-century lives, especially those of the more transient figures, seem from what contemporary evidence we can collect to be stuffed with fiction, including alleged documents that are manifestly forged.
The last great doctor, Galen, speaks for himself, at great length. He writes in Greek; and so, except for the marvellous Latin of Apuleius, do the novelists. The outstanding philosophers Marcus Aurelius and Plotinus—the latter in his Enneads, published by a pupil Porphyry— likewise use the Greek language. On religion, pagan and Christian alike, there are innumerable authors in a variety of languages, the outstanding historical work on the Christian side being the Ecclesiastical History written in Greek by Eusebius (260–340).11 An immense amount of historical information can be disinterred from the Digest of Justinian, of which a large proportion was derived, with or without modification, from jurists of Severan times. The thirteenth book of the so-called ‘Sibylline Oracles’ throws light on eastern events in the third century.
Papyri, revolutionising our comprehension of the history of the ancient novel and illustrating almost innumerable aspects of public and private life, have been found in Egypt in their thousands.12 Archaeology and art provide their usual abundant aids, and epigraphic evidence of bewildering quantity and variety includes the military calendar of Dura Europos (Feriale Duranum), more than sixty fragments of Diocletian’s Edict of Prices found in thirty-two different places, and the Sassanian inscription of Shapur I in the Kaaba of Zoroaster (Naksh-i-Rustam), which counterbalances equally propagandist Roman accounts of Romano-Persian hostilities and, in particular, the inscriptions on imperial Roman coins. For this is the epoch, above all others, in which the designs and circulation and mintage of coins (augmented by commemorative medallions) have a far-reaching significance which spans the fields of politics, publicity, war, economics, religion and art. Occasionally coins even depict and name self-styled Augusti and the relatives of Augusti whose very names are otherwise totally unknown.13 ‘It is beyond doubt that in the coins lies a treasure, partly won, partly awaiting further study as a condition of its full exploitation; a treasure which, failing new discoveries of inscriptions or manuscripts, offers almost our only chance of penetrating the thick darkness that still envelops so much of the history of the third century.’14