Chapter 1The Subura

Foreword

The Gauls left no text. They left ground.

That sentence is the reason for this book and it explains almost everything about how it is built.

Julius Caesar fought a war in Gaul for nine years, killed and enslaved at a scale that troubled Romans who were not squeamish, and then wrote the only surviving account of it — in the third person, in prose so clean that Cicero praised it while its author was still alive, and which European schools then set as the beginner's Latin text for the next four hundred years.

There is no Gallic version. Not a document, not a fragment, not a name in a language that was not Latin or Greek.

So the war and the book called the Gallic War are one artifact, and this book reads them as one. Eleven of its twenty-two chapters sit in Gaul, and Chapter 17 stops the narrative entirely to anatomize the text, because the argument does not work unless the reader watches the campaign and its cover story being manufactured in the same nine years.

What is documented, and every item of it is documented because Caesar reported it himself:

Fifty-three thousand Aduatuci sold at auction in a single transaction, after terms had been agreed and weapons surrendered. The Usipetes and Tencteri attacked while their chiefs were being detained in the Roman camp, and their non-combatants cut down at the water. The ruling council of the Veneti executed and the population sold, because Roman officers sent to requisition grain had been detained. The Eburones hunted through the Ardennes across two campaigning seasons — with neighboring peoples invited in to do the plundering, so that, as he explains, Gauls rather than legionaries would take the risk in the woods and the stock and the name of the people would be wiped out — until they do not appear in history again. The Mandubii, expelled from Alesia so the garrison could eat, refused passage through the Roman lines and left to starve in the ground between the siegeworks. And at Uxellodunum, survivors released with their hands cut off, which Hirtius explains had to be visible because Caesar's clemency was famous and might otherwise be mistaken for weakness.

He wrote all of that down himself, without embarrassment, in the same calm administrative register he used for the joinery of a bridge.

This book does not settle whether the word genocide fits, and Chapter 16 explains why it cannot: the pre-conquest population of Gaul is unknown, every casualty figure in the Commentarii comes from the man who caused the casualties, and David Henige has demonstrated that not one of the ancient totals is usable.

The honest verdict is that we do not know the death toll and never will, and that this is not the same as saying it was small.

One number needs correcting at the front, because the tradition has got it wrong for two thousand years.

At Bellum Gallicum 4.15 Caesar gives 430,000 as the count of the Usipetes and Tencteri who had left home — the size of a host, not a body count — and sets it against a Roman return of all safe to a man, very few even wounded. He reports the number where a death toll would go and lets the reader do the arithmetic.

The handbooks, the textbooks and the internet have laundered that into 430,000 dead. This book quotes the construction instead.

The counter-case to all of this is carried inside the chapters and not conceded in a note, because my own instincts run the other way and a reader is entitled to know it.

Adrian Goldsworthy holds that this was ordinary Roman brutality at extraordinary scale rather than something different in kind. Erich Gruen holds that the Republic was structurally sound and was killed rather than dying of its own contradictions. Robert Morstein-Marx holds that Caesar answered throughout to a real popular constituency and was not a lone man driving at a crown. Zvi Yavetz holds that the divine honors of 44 BC were improvised and retro-fitted by the winners of the next war.

Those four are named in Chapters 18 and 22, where they belong, arguing.

What is new here is a counter-archive that the standard biographies were written too early to use.

At Kessel in North Brabant, skeletal material including women and children dredged from the Maas, identified in 2015 with the destruction of the Usipetes and Tencteri — a strong identification, and a contested one, and this book presents it as both. A Roman camp at Hermeskeil. Franco-German excavation at Alise-Sainte-Reine in the 1990s that put the double siege line of Alesia on the ground, where Napoleon III had dug in the 1860s for reasons of his own. And a defensive ditch enclosing more than twenty hectares at Ebbsfleet in Pegwell Bay, now a kilometer inland through silting, which is the leading candidate for the landing beach of 54 BC.

Candidate is the word.

Two smaller disciplines the reader should know about.

Every speech in Sallust, Plutarch and Suetonius is a composition, written decades or centuries later according to a well-understood convention. The Catiline debate of 5 December 63 BC, which Sallust preserves in full and which is one of the great scenes in Roman history, was written twenty years after it happened by a man who was not in the room. This book says so where it uses it.

And what Caesar said at the Rubicon is not recoverable. Suetonius gives one sentence in Latin; Plutarch gives a different one in Greek, quoting a line from Menander. They are not the same remark. What he wrote afterwards, in his own account of the civil war, is dignitas — that his standing had been insulted.

The last thing.

Sixty-odd men killed him on 15 March 44 BC in a hall attached to Pompey's theater. Most of them were men he had pardoned. There were twenty-three wounds, of which the physician Antistius judged one to have been fatal.

Suetonius reports his last words in two versions and then says that most sources say he spoke none.

This book closes on the ledger rather than on the funeral.

Chapter 1 · The Subura

The Subura was a quarter of Rome, and the address is very nearly all that survives. There is no building to visit and no street to walk. No lease, no deed, no inventory of the furniture, no name cut over a door, no wall with a scratch on it that can be tied to this family. What the record holds is the name of the district, the fact that the family lived in an apartment block in it, and the three words that describe what living there was like: loud, poor, flammable. Into that block, in July of 100 BC, a patrician family put a son.

Each of those words is a statement about money. A quarter is loud because people are stacked on top of one another, which is what an apartment block is for. It is poor because stacking is how a city houses the people who cannot pay not to be stacked. It is flammable because everything that cooked a meal or lit a room after dark was an open fire, and a building of that kind was full of them. Rome had ground for families with money and ground for families without it, and the Julii were housed among the second sort. This is not an atmospheric detail. It is the first hard fact in the life, and it is a fact about the family's finances rather than about its rank.

It should be said at once that even this fact has soft edges. The record attaches the family to the quarter. It does not establish how long they had been there, whether they owned the block or rented in it, whether the boy was born in that building or only raised in it, or when, if ever, they got out. What survives is an association between a name and a district, of the sort that later writers repeat because it is memorable and because it makes a point about a great man's beginnings. The point it makes happens to be true. That does not make the sources any more precise than they are.

The rank was not in question. The gens Julia was patrician — not merely ancient, and not merely once wealthy, but a member of the closed hereditary caste of the oldest Roman families, which a man entered by being born into it and could not otherwise enter at all. Patrician status could not be bought, voted, earned in the field, or married into by the man who wanted it. It was the one thing in Roman public life that was purely inherited, and the Julii had it, and it did not put food on a table.

They claimed a good deal more than that. The family's descent, as the family told it, ran from the goddess Venus, through Iulus, to them. It is worth being exact about what a claim of that kind was and was not. It was not evidence, and nobody in Rome, the Julii included, mistook it for evidence. It was a standing assertion about who the family was, maintained across generations in a society where a house's account of its own origins was part of its political equipment, repeated at funerals and understood as a form of property. It was also the cheapest thing the family owned, and it outlived every other asset in the estate, because nobody had to pay anything to keep a goddess on a family tree.

Against the goddess, the ledger. The gens Julia had produced almost nothing politically for a century. In a Republic that measured a family's standing by the offices its ancestors had held, and that expected a young man to be pushed up the ladder by the accumulated weight of his relatives, a hundred years of silence was not a neutral condition. It compounded. Each generation that failed to reach high office made it harder for the next to be taken seriously as a candidate for it, and the remedy — canvassing, entertaining, giving the public something to remember at the moment of the vote — cost money the family did not have. The Julii held the one qualification that could not be purchased and lacked the ordinary means by which men without it got elected anyway. Why the house had gone quiet, the record does not explain. It records only that it was quiet, and then it records that a boy was born to it in a rented block in the Subura.

The birth is dated to the twelfth or the thirteenth of July, and the tradition does not choose between the two days. The year is given as 100 BC, which is standard and is not secure: 102 BC has been argued, by Mommsen among others, because 100 sits awkwardly against the Republican rules governing the minimum ages at which a man could hold each office in turn. The problem is genuine and it is not going to be settled here. What follows uses the standard year, and the reader should hold it loosely, because a life that later turns on precisely dated days — a river crossed on a particular night, a man killed on a particular morning — opens with a date that cannot be fixed to a year, and within the year cannot be fixed to a day.

Ask where any of this comes from and the answer is uncomfortable. Nothing written at the time about this household survives. The later historians who supply the childhood — Plutarch, Suetonius, Appian, Dio — were working from Caesar's own books plus intermediaries that are now lost, at a remove of 100 to 250 years. Suetonius wrote Divus Julius, the deified Julius, early in the second century AD. The title is worth a moment, because it tells you where the tradition had arrived before it tells you anything about where the man began: by the time the ancient book that supplies most of the famous anecdotes was compiled, its subject was a state god, and the compiler knew every single thing that was going to happen to him. Nothing in the Subura chapter of that book was written by anyone who did not already know how it ended.

Suetonius is also a particular kind of witness. His book is thematic rather than chronological, which means it does not walk through a life so much as sort it into categories — appearance here, vices there, omens somewhere else. He had archival access and no consistent standard of evidence. He will set down rival versions of a story and decline to decide between them, which is an honest habit and is also the reason he cannot be used the way a document can be used. Plutarch's Life of Caesar is moral biography, written to arrive somewhere. Neither man was trying to establish what happened inside a rented apartment in the Subura, and neither was in any position to. The wider difficulty is structural: the surviving record of the Roman Republic is overwhelmingly a record of public acts — elections, laws, trials, commands, prosecutions. A household performs no public acts. A boy performs none until he is old enough to stand for something, and so, in the only kind of evidence the period generates in quantity, he does not yet exist.

There is one great contemporary archive for this period and it belongs to Cicero. Roughly 900 letters survive, many of them never intended for publication, and across the later chapters of this book they are the single most valuable non-Caesarian source there is: real time, unguarded, written by a man who did not know how the story ended. They do not begin to help until the mid-60s. For the years in the Subura they are silent, because in those years Cicero was not yet a man whose letters anybody kept, and the boy in the apartment block was not yet a man anybody would write one about.

So the childhood is not thin. It is absent. We do not know what he was taught or by whom. We do not know what he read, what languages were used in the room, who else lived in it, what the family's income was or where it came from, or what any of them made of the ancestor who was a goddess. Everything beyond a small handful of items is inference from what patrician boys in general did, and it ought to be called inference, because that is what it is. A biography that fills this space with a plausible Roman boyhood is not recovering the childhood of Julius Caesar. It is describing a type and attaching his name to it.

We do not even know whether he was a healthy child. The ancient sources report collapses in the later life, described in Greek terms usually rendered as the falling sickness, and modern medicine has proposed epilepsy, a series of small strokes, and other conditions besides. None of those proposals reaches back into the Subura, and none of them is a finding. Retrospective diagnosis at two millennia is not evidence, and the honest position is to record that the collapses were reported, that the reports come from writers who were not physicians and were not present, and that the cause is not recoverable — rather than to select a modern label and let it explain a boyhood nobody described.

The silence is not only a silence about the boy. It is a silence in his own voice, and it never entirely lifts. His letters do not survive as a collection. What we have of them survives because other people quoted them, chiefly inside Cicero's correspondence, which means that even his private sentences reach us only where somebody else found them worth reproducing. Of the writing he did that was not a report on a war, almost everything is lost. He is one of the very few figures of this period whose private voice is essentially inaccessible — and the one voice of his that does survive in bulk, and survives entire, is the one he composed for publication about his own campaigns.

That absence sets the method for everything that follows, so it is worth stating now. A modern historian of a war sets one army's war diary against the other army's unit record, and sets both against a soldier's letter home, and the truth is assembled out of the disagreements. For most of this life, that option does not exist. There is going to be one narrative source for the central years, and its author is the defendant; there are Cicero's letters as an outside line where they reach; there is a later tradition that read the defendant's book and repeated it; and there is the ground — bone, ditch, buried coin, burn layer, dug in the last few decades and still being argued about, which is the only archive the losing side left, and which this book treats as an archive, with the same seriousness and the same skepticism it applies to the winner's prose. For the Subura years even that is unavailable. There is no text to cross-examine and no ground to dig. There is a district, a date with two days in it, a name, and one relative.

His aunt Julia was married to Gaius Marius, seven times consul.

That sentence has to be allowed to land, because in a family that had produced almost nothing politically for a century it is the only political fact on offer. The consulship was the summit of the Roman ladder. Marius had stood on it seven times. Through his aunt's marriage, the son of an obscure branch of an ancient house was related by law to the holder of seven consulships — which, by the Republic's own way of measuring a man, made Marius the largest political figure of the preceding generation. The gap between the family's rank and the family's circumstances suddenly had something standing in it.

In Rome this was not sentiment. Connection was the mechanism by which public life actually moved: how a young man was introduced and to whom, whose clients turned out for him on a voting day, who lent him money against a future he had not yet earned, whose obligations transferred to him and whose enemies became his before he had done anything at all to earn them. A name opened a door. The Julii had the name and nothing behind the door. The aunt's marriage put something behind it.

It also put the wrong thing behind it, or at least a thing that could turn — and this is the part the arrangement of the facts makes plain without any source stating it outright. A great connection in Rome was not a possession. It was a side. A man did not hold it so much as stand on it, and its value depended entirely on how things afterward went for the man at the other end. Nor was it chosen. A boy does not select his aunt's husband. He is born into the alliance exactly as he is born into the caste, and he carries it whether or not he ever exchanges a word with the man. Within this boy's lifetime it would become illegal in Rome to display the images of Marius, which is as clean a measure as the record offers of what became of the family's one asset, and of how completely a single fact could reverse its sign. The most valuable thing the Julii owned turned into a thing that could not lawfully be shown in public.

Of his father, the tradition kept a death.

He was at Pisae, and not at Rome, and in 85 BC he dropped dead there while putting on his shoes.

That is the whole of it. The record does not say why he was at Pisae or how long he had been there, what he was doing that morning or what he intended to do with the rest of the day, what killed him, who was in the room, or what was done afterward with the body. It preserves a town, a year and a gesture — a man bending down to his own feet and not getting up.

The item is worth grading rather than simply using, because this book will have to grade a great many more consequential things later on, and the standard should be visible early. A sudden death mid-gesture is precisely the kind of detail an anecdotalist keeps. It is short, it is vivid, it requires no explanation, and it can be told in a single line, long afterward, by somebody who was not there. Details of that shape survive because they are repeatable, not because anyone checked them. Nothing written at the time confirms this one and nothing contradicts it. In its favor, mildly, is that it carries no moral, flatters nobody and serves no argument: men do die suddenly in the morning, and a story that has no work to do is a story with less reason to have been improved. That is a reason to keep it, not a reason to trust it. What can be said without qualification is that the family lost the head of its household in that year, and that everything afterward in the son's life happened to a young man who no longer had a father to arrange it for him.

Between the July in the Subura and the morning at Pisae there are fifteen years, and they are the least documented stretch of a life that is otherwise among the best attested in the ancient world. That mismatch is not incidental to this book. It is the book's whole problem, appearing here in its simplest form.

Set out the estate. He inherited a name that could not be taken from him and could not be spent. He inherited a goddess in the genealogy, worth exactly what an audience was prepared to grant it and not a coin more. He inherited no money that the record can find: no property is listed, no sum is preserved, and no public inventory of this family's holdings exists for any point in these years. And he inherited a connection to a man who had been consul seven times, which was the only item in the estate with real political value and the only item in it that could get him killed.

That is a hard beginning, but it is a beginning with a shape, and the shape is worth naming before the narrative moves. An ancient name, no money and a dangerous inheritance is not a description of poverty. It is a description of a man who owns one asset, cannot sell it, cannot decline it, and cannot control what it is about to be worth. Everything he did afterward was done by someone who had learned early that his position was inherited, valuable, and not in his hands.

The last thing to say about these years is why they are so dark, because the answer explains the rest of the book. The record of Julius Caesar does not thicken gradually, the way a life normally comes into focus as its subject becomes worth noticing. It thickens because, from 58 BC onward, he wrote it. And it thickens strangely: the stretch of the life with more narrative detail than any other is also the stretch with the least corroboration, and those are the same years for the same reason. For the nine years of the Gallic War there is one narrative source, and its author is the defendant. He composed it in the third person, so that the man issuing the orders appears on the page as a character whom other people obey — a figure observed from outside his own body, moving through the flat administrative register of an official report, recording what he had done and never once explaining why he had done it. It is superb Latin. It was believed for a very long time. The reasons it was believed are the subject of this book.

The Subura, then, is what the record of this man looks like when he is not holding the pen: a district and no address, a birthday given as two days, a year that is still argued over, a goddess in the family tree, an empty ledger, an aunt's marriage, and a father dying over his shoes in a town nobody can explain his being in. Everything after this gets brighter. It is worth remembering, when it does, who arranged the lighting.