The question of what fell when Rome fell has been answered three different ways in the last three centuries, and the disagreement is mostly about which thing is being asked after.
The previous lesson supplied a mechanism: the western empire lost the provinces that paid for its army, and an army that cannot be paid becomes somebody else's. That explains a political outcome. It does not say whether anything larger ended, and this lesson tests that against evidence that is independent of politics, using the same method as the earlier lesson on the imperial economy: several proxies, each stated with its defect.
Three answers
Edward Gibbon, publishing across twelve years from 1776, wrote a narrative of decline running to 1453 and summarised its western half as the triumph of barbarism and religion. He had grounds for both terms and a personal stake in the second: the empire's adoption of Christianity diverted resources and attention, produced quarrels that consumed emperors, and, in his reading, drained the civic virtue that had made Rome formidable.
Henri Pirenne, writing in the 1930s, moved the ending by two centuries. The Germanic kings, he pointed out, preserved Roman institutions, Roman law and the Mediterranean trade that had always defined the Roman world; what broke that world was the Arab conquests of the seventh century, which turned the Mediterranean from a Roman lake into a frontier and pushed the centre of western Europe northwards. His formulation was that without Muhammad, Charlemagne would be inconceivable.
The third answer, dominant in the later twentieth century, is that the question is wrong. On this view the period is a transformation rather than a fall: barbarians were accommodated rather than invaders, elites adapted, the church carried Roman culture forward, and late antiquity was a creative age in its own right rather than the wreckage of a better one.
Weighing them
Example. Assess Gibbon's two words as causes.
Barbarism does not survive the previous lesson. The peoples involved were few, largely Romanised, mostly admitted by treaty and often already in Roman service, and the same peoples faced the eastern empire without destroying it.
Religion is more interesting and still fails as a primary cause, for a reason that is decisive and simple: the eastern empire was more Christian, earlier, more thoroughly, and with far more disruptive theological quarrels, and it survived for another thousand years. Any factor shared by both halves cannot explain a difference between them. The church did absorb resources and did compete for the loyalty of able men, and Gibbon is right that both happened, but the comparison shows it cannot have been what tipped the balance.
What deserves credit is Gibbon's method rather than his conclusions. He looked for structural causes running across centuries instead of blaming individuals, and he framed the problem as needing an explanation at all, where his predecessors had treated Rome's fall as the natural death of any great thing. His answers are wrong in ways his own approach eventually exposed.
Now you. How would you test Pirenne's thesis?
Answer
By dating the interruption of Mediterranean exchange directly, which is exactly what the archaeology of the last fifty years has done, and the result goes against him. Pirenne worked mainly from written sources and from a small set of imported goods such as papyrus and spices. The pottery evidence, which is far more abundant and much better dated, shows long-distance distribution in the western Mediterranean falling sharply from the fifth century, well before any Arab army existed, with the sharpest breaks tracking the loss of the African provinces and the collapse of the fiscal system that moved goods across the sea.
The verdict is that Pirenne asked the right kind of question and had the wrong date. His insight, that the Roman world should be defined by Mediterranean exchange rather than by who wore the purple, is now standard. His causal claim was tested by evidence he did not have and did not survive it. That sequence is worth noticing, because it is how the subject has advanced: a thesis stated clearly enough to be checked is more useful than a cautious one, even when it turns out to be wrong.
The test in pottery
Pottery is the historian's luckiest material. It is cheap, it breaks constantly, the pieces are indestructible, and the shapes and fabrics change often enough to be dated closely and traced to their kilns. That combination makes it the best surviving indicator of how complex an economy was.
Example. In Roman Britain, wheel-thrown pottery was made in industrial quantities, distributed across the province and used by ordinary farming households; so were mass-produced roof tiles. By the middle of the fifth century, wheel-thrown pottery has effectively ceased to be made in Britain, tile production stops, and coins go out of circulation. What does that show, and what is the honest limit?
It shows the disappearance of specialisation and distribution, not merely of wealth. Mass-produced pottery requires potters who do nothing else, kilns run continuously, a means of moving the output tens or hundreds of miles, and customers with something to trade for it. When it vanishes, every one of those conditions has failed together, and the replacement is handmade, locally fired ware of a quality below what pre-Roman Iron Age Britain had produced. That is the striking part: the material culture of the fifth century falls not to the level before the Romans arrived but below it.
Two honest limits. First, Britain is the extreme case, the province that had been most dependent on military spending and lost it most abruptly, and Italy, Africa and the eastern Mediterranean show far gentler curves; a claim about Britain is not a claim about the empire. Second, pottery measures complexity, not happiness. It tells you the economy simplified. It does not tell you that any individual was worse off, which the next question takes up.
Now you. Some skeletal evidence suggests that people in post-Roman northern Europe were, if anything, slightly taller than under Roman rule. Does complexity equal welfare?
Answer
No, and an earlier lesson made the same point in the other direction, when the imperial economy was expanding and adult stature was flat or falling. Height reflects childhood nutrition and disease load, and both improve when population falls relative to land, when people leave crowded cities, and when the tax that took a share of the harvest is no longer collected. A peasant who keeps more of his crop and lives at lower density may well be better fed in a world with no coins and no roof tiles.
So two things are true at once. The economy of the post-Roman west was drastically simpler, poorer in goods, and less connected, and that is not seriously disputed by anyone who has looked at the material. And a substantial number of individual people may have experienced the change as a relief. Holding both is uncomfortable, and it is the correct position. What the fall of the western empire destroyed was a system, and systems are not the same as the welfare of the people inside them. The reason to care about the destruction anyway is that the complexity was not recovered for many centuries, and what was lost with it includes literacy, building in stone, and the ability of a society to do anything requiring more resources than one district can raise.
The other proxies agree
The pottery does not stand alone, and the corroboration comes from the same independent directions used earlier in this course.
The Greenland ice, which recorded lead at four times natural background around the beginning of our era, falls back to near-background levels in the post-Roman centuries and does not return to Roman levels for well over a thousand years. Coinage tells the same story in a different register: the disappearance of small-denomination bronze is more informative than any change in gold, because low-value coin is what an economy uses when ordinary people buy ordinary things, and its absence means everyday exchange has stopped being monetary. Cattle bones from Britain shrink after the Roman period, reverting towards pre-Roman size, which points to the loss of managed breeding kept up by a market for meat and hides.
And the city of Rome itself, which had held something near a million people when the empire was at its largest, is usually reckoned at around 100,000 by 500 and perhaps 30,000 in the centuries after the Gothic wars of Justinian's reconquest: a fall of more than thirty to one. No western European city reached a million again until London around 1800.
The empire that did not end
Meanwhile the empire continued. The eastern half, governed from Constantinople, ran without a break in legitimacy from Constantine to 29 May 1453, when the walls were breached by Mehmed II, which is 977 years after the deposition in Italy and 1,480 years after Augustus's settlement.
Its inhabitants called themselves Romans throughout, in Greek, and it is worth registering how odd our habit of calling them Byzantines is: the word was coined by a scholar in the sixteenth century and no citizen of that state ever used it. In the sixth century Justinian retook Africa and Italy, and had the whole of Roman law codified into the compilation that carries it to the present day. The consulship, an office created when the kings were expelled a thousand years earlier, was last held in the west in 541.
Example. List what survived the western collapse, and then say what all the survivals have in common.
The law survived, reissued by Germanic kings for their Roman subjects and codified in the east, and revived in the Italian universities from the eleventh century as the foundation of continental civil law. The church survived with its provinces mapped onto the old imperial ones, its language Latin, and its bishops still doing the civic work the previous lessons described. The eastern state survived. The name survived: the Ottoman sultan took the title Caesar of Rome, the western kingdom crowned in 800 called itself the Roman empire until 1806, and Kaiser and Tsar are both the word Caesar.
What they have in common is that none of them required a tax system. A legal code is a text and can be copied. A church is funded by donation, land and its own authority. A title is a claim. Every one of these could be maintained by a society far poorer and simpler than the empire that produced it, which is precisely why they lasted.
Now you. So state, in one sentence, what actually ended in the west, and then say what the sentence is doing.
Answer
What ended was a state that taxed a large territory in order to pay a professional standing army, and with it the complex, specialised, long-distance economy that only that kind of state could hold together.
What the sentence is doing is picking out the load-bearing element, and its warrant is that it explains all the other facts rather than being one of them. It explains why the survivals are the things that need no fiscal machinery, and why the losses are the things that do: mass-produced goods, everyday coinage, quarried stone and tile, and the movement of ordinary bulk cargo across a sea. It explains why the east, which kept its revenue, kept everything else. It explains the timing, since the decisive years are the ones in which provinces stopped paying rather than the ones with famous battles. And it survives the comparative test that defeated Gibbon, because it names something the two halves did not share.
It is also falsifiable, which is what makes it worth stating. If the western economy had simplified before the tax system failed, or if the east had simplified in the same way while keeping its revenues, the account would be wrong. Hold onto that structure rather than the conclusion. This subject has no experiments, so its claims are tested by whether they predict evidence gathered for other reasons, and an explanation that cannot fail is not an explanation.
Ending
The empire ran, in the west, for five centuries after Augustus, and it ran on an arrangement that the first lessons of this course laid out and the last ones watched come apart. A monarchy with no name and no rule of succession, so that the army decided. An army that was the largest expense any pre-industrial state had taken on. A tax of about five per cent, light enough to be tolerable and too light to leave any margin. Government delegated to a couple of thousand cities whose elites wanted the job. A citizenship extended until everyone had it. A frontier that stopped moving, so nothing came in from outside. Every one of those is visible in the fifth century, doing the opposite of what it had done in the first.
What is worth keeping from this is not the date. It is the habit of asking, about any large and apparently permanent structure, what it is actually made of and what is paying for it. The Roman empire looked to its inhabitants like a fact about the world, in the way such things always do, and it turned out to be a machine with a specific input. When the input stopped, the parts that needed it stopped too, and the parts that did not are still here: the law in the courts, the church in the towns, the calendar, the alphabet, the roads under the roads, and the word Caesar in three languages.