By Kyle Orton (@KyleWOrton) on 12 June 2021

A painting representing the Byzantines use of “Greek fire” to repel the Arab siege of Constantinople, 674-78

The use of the word “Byzantium” for the Eastern Roman Empire is inescapable, though it is best thought of as a term of convenience rather than definition. For one thing, the inhabitants never called themselves “Byzantines”: they always thought of themselves as “Romans”, though using the Greek work “Romaioi”. The easiest way to see the problem is to ask: When did the Byzantine Empire begin?

Two helpful sources drawn on for this post are Judith Herrin’s classic 1987 book, The Formation of Christendom (pp. 137-39), and Cyril Mango’s 2002 book, The Oxford History of Byzantium (pp. 2-5).

One suggested date is 324 AD, when the Emperor Constantine (r. 306-37) re-founded the city of Byzantium and renamed it Constantinople, or the ceremonial inauguration of the city in 330 as the second capital, the “Second Rome”.

The official Christianization of the Empire in 380 under Theodosius (r. 379-95) is another possible date, given that this process did radically reorganise society, and occurred in parallel with an equally radical transformation of the state into a much more powerful structure.

The formal division of the Empire into east and west in 395 is another possibility—as, by this definition, are 286 or 293, when the Emperor Diocletian (r. 284-305) initiated the first efforts to make the Latin West and Greek East autonomous.

The collapse of the western half of the Empire in 476 has been a popular candidate, and, indeed, the emergence of the term “Byzantium” in English was somewhat premised on this idea. “Byzantium” began to be used in English-language historiography with any frequency probably towards the end of the eighteenth century and was established by the nineteenth century, having been in use slightly earlier in French, German, and Russian. (One author argues it was used much earlier in French.) The reasoning of Victorian historians was that an Empire no longer ruling the city of Rome could not be “Roman”. There is a logic to this, but it is complicated by the fact the Justinian expedition recaptured Rome for “Byzantium” in 536 and held it until 568.

Ultimately, all these attempts to identify an origin point for Byzantium collapse in on themselves because the continuity with Rome is too strong. Any attempt to find a distinction can be argued, in the final analysis, to be artificial.

That said, the best case for a distinction between Rome and Byzantium looks to the processes of societal recalibration—born out of a series of crises—that took place in the sixth and seventh centuries. This is sometimes expressed by periodization, with the Roman Empire as ancient and Byzantium as medieval. Broadly, the transition from ancient to medieval can be said to begin under Justinian I (r. 527-65) and reach completion under Heraclius (r. 610-41). Some older histories argued for the ancient-modern faultline being under Leo III (r. 717-41), but this is generally agreed to be too late because what historians are getting at with these dates is the change, on par with the earlier reformation as Christianity triumphed, that led to the death of the old Roman urban way of life, leaving a more rural, somewhat more egalitarian, and more heavily militarized society.

An important inflection point is the 530s, when lots of the physical assets—material and human—extant from the Roman past were destroyed. The Nika Revolt in Constantinople in 532 burned down half the city, and when Justinian rebuilt it was a much more visibly Christian and Greek city than it had been Roman and Latin. Later in the 530s, Justinian’s very effort to restore Rome to its former greatness—devoutly Christian, he was nonetheless something of a throwback to the great martial Emperors of pagan Rome—would literally destroy Rome, the city: having recaptured it relatively easily, the fighting when the Ostrogoths counter-attacked devastated the infrastructure (above all the aqueducts), displaced most of the population, and tore to shreds the effectively Roman pattern of socio-political life that had endured under Barbarian rule since 476. The third great city of Byzantium, Antioch, was levelled by the Persians in 540.

The Justinian Plague that began in 541 was a near-apocalyptic event: it may have killed half the Empire’s population and it left the state a shell of its former self, removing the base for taxes and the legions. Barbarians like the Avars began to make inroads in the Balkans in the late sixth century, but for the time-being the Empire’s structures and boundaries largely held.

Then came the titanic clash with the superpower rival in Persia, lasting a generation (602-28). Though ending in a technical Byzantine victory, the Empire’s foundations had been shattered and, in the 630s and 640s, Byzantine authority in provinces recaptured from the Persians collapsed, giving way to new Arab rulers, who in time fashioned the Islamic creed. The armies of Islam would be Byzantium’s undoing 800 years later, but it was a close-run thing in the first round with the Arabs. The Arabs reached the walls of Constantinople in the 670s, and the effort needed to throw back this challenge indelibly changed Byzantium.

Already without Rome, the other great cities—Alexandria, Antioch, Jerusalem, Damascus—were now gone. The previous urban culture could not be sustained with just one city, not even one as splendid as Constantinople, and not in these new conditions of relative scarcity and danger. Truncated in territory, stripped of its richest provinces by conquest, and reduced in population by plague, the diminution in Roman power and wealth at the end of the seventh century when compared with the Antonine age—or even the Theodosian era—was stark. The struggle with the Arabs as they erupted out of the Near East had been a near-death experience for Byzantium, which emerged on the other side as a garrison state whose people had little memory of the time when their predecessors had been masters of the known world. If there is a rupture point, this is it.