Caesar's Gallic Wars: What His Own Commentaries Say, and How Far His Numbers Can Be Trusted

Start hereCaesar says his men came back from the attack of 55 BC "all safe to a man". He also gives the enemy's number as 430,000. So how many people died that day?

Between 58 and 50 BC Caesar conquered the rest of Gaul up to the left bank of the Rhine. Most of what we know about it comes from one book: Caesar’s own account. Here is that book, the war year by year and the death tolls source by source. It ends with a way to read any commander’s own story.

A note before you start: parts of this page describe the killing of civilians, as Caesar and later writers report it.

011. What Caesar’s Gallic War is

Suetonius says Caesar left memoirs of the Gallic war, and that Aulus Hirtius “supplied the final book of the Gallic War, which Caesar left unwritten”. Suetonius seems to have written mostly under the emperors Trajan and Hadrian (AD 98–138), and was for a time Hadrian’s secretary. Caesar’s seven books end with 52 BC; the eighth covers 51–50 BC. Hirtius had served with Caesar in Gaul as a legate.

The book opens with a famous line: Gallia est omnis divisa in partes tres, “All Gaul is divided into three parts”. It refers to its author as “Caesar”, as when lists “were brought to Caesar”. In a 1998 study, David Henige notes that it has long been regarded as one of the great war reports, “refreshingly free of the intrusive ‘I’”.

Why he wrote it

The ancient answer is modest. Cicero, quoted by Suetonius, praised the memoirs and said Caesar’s purpose “was to supply material to others” who wished to write history. Hirtius wrote that they “were published for the use of historians”.

Modern readers see more in it. The 1911 Encyclopaedia Britannica judged that their publication “was doubtless timed to impress on the mind of the Roman people the great services rendered by Caesar to Rome”. The archaeologist Nico Roymans notes recent critical readings that treat the book as “an instrument of political communication and manipulation”.

A 2018 study by the scholar Sophie Hulot describes a founding work by M. Rambaud that stresses Caesar’s skill at distorting history, and his propaganda. In her reading, Caesar’s account of the 55 BC massacre carries at least a version of events built to protect him from political attacks. She adds that it is not certain the book was written as a reply to Cato’s attack on Caesar over that massacre.

When he wrote it

This part is still argued. The 1911 Britannica said the seven books “appear to have been written” in 51 BC.

Hulot writes that two views still compete: that it was written year by year, or all at once, after the victory over Vercingetorix. A third view has it written in three stages. She notes that the single, final writing is now more and more widely accepted, citing C. S. Kraus (2009).

How far to trust it

Doubts began in Roman times. Suetonius reports that Asinius Pollio thought the memoirs were put together “somewhat carelessly and without strict regard for truth”. Pollio thought Caesar too ready to believe other men’s reports, and that he gave “a perverted account of his own, either designedly or perhaps from forgetfulness”. Pollio also thought Caesar meant to rewrite them.

Henige points out that the book has been “unchallenged by rival accounts”. He adds that until fairly recently its testimony “was seldom seriously challenged”. In 1911 the Britannica could still say “the accuracy of its narrative has never been seriously shaken”.

Our reading: we have the winner’s version, written by the winner, about his own war. That does not make it false. It does mean every number in it needs a second look.

022. The war, year by year

The table gives the main campaigns of each year, with one detail from the book itself.

Year	Main campaigns	One detail from the account
58 BC	The Helvetii; the German king Ariovistus	Ariovistus escaped across the Rhine in a small boat
57 BC	The Belgae, the Nervii and the Aduatuci	The Aduatuci were sold: 53,000, by the buyers’ count
56 BC	The Veneti of Brittany, at sea	Their senate was put to death and the rest sold
55 BC	The Usipetes and Tencteri; a bridge over the Rhine; Britain	The bridge was finished within ten days
54 BC	Britain again; a revolt in the north	A legion and five cohorts were destroyed by the Eburones
53 BC	The Eburones; a second Rhine crossing	Their leader Ambiorix escaped
52 BC	Vercingetorix: Avaricum, Gergovia, Alesia	Gergovia cost Caesar 46 centurions
51–50 BC	Last resistance, ending at Uxellodunum	Those who had fought had their hands cut off

58–56 BC: the first campaigns

The war began with a people on the move. In 58 BC the Helvetii, a Celtic people of what is now Switzerland, set out for the shores of the Atlantic, and Caesar defeated them at Bibracte. Then came the Germans under Ariovistus. Their king escaped across the Rhine in a small boat; both his wives died in the flight.

In 57 BC he marched north against the Belgae and the Nervii. Caesar says the battle left “the nation and name of the Nervii” almost “reduced to annihilation”. Their envoys spoke of 60,000 men cut to scarcely 500 who could bear arms. He spared the survivors, he writes, “that he might appear to use compassion”.

When the town of the Aduatuci fell, he sold everyone in it as booty: the buyers reported 53,000 people.

The Veneti had held two Roman officers as hostages. After defeating them at sea, Caesar “put to death all their senate” and sold the rest as slaves. He gave a reason: so that in future the rights of ambassadors would be more carefully respected by barbarians.

55–53 BC: the Rhine, Britain and a revolt

He bridged the Rhine within ten days of starting to gather the timber. He crossed to Britain because, he says, the Britons had sent help to his enemies in Gaul. After the second crossing, in 54 BC, Caesar set a yearly tribute, which the 1911 Britannica says was never paid.

Late in 54 BC the Gauls struck back. At Aduatuca a newly raised legion and five cohorts were destroyed by the Eburones under Ambiorix.

52–50 BC: Vercingetorix

Early in 52 BC the Arverni revolted under Vercingetorix, and other peoples quickly joined them.

At Avaricum, Caesar says, his soldiers “spared neither those worn out with years, women, or children”: of about 40,000, scarcely 800 escaped. At Gergovia he was driven back, losing 46 centurions and a little under seven hundred soldiers.

He shut Vercingetorix up in Alesia and surrounded the town with siege lines. The townspeople, the Mandubii, were sent out with their wives and children. They begged to be taken as slaves and fed. Caesar “forbade them to be admitted”. Caesar’s book does not mention them again. Cassius Dio was born between AD 155 and 164, and his history ends in AD 229. He says the people driven out “perished most miserably between the city and the camp, because neither party would receive them”.

The surrender is told in different ways. In Caesar’s version, Vercingetorix first offers himself to his council, to be put to death or handed over alive. Then the Gallic chiefs surrender him and lay down their arms. In Plutarch, Vercingetorix rides out in his finest armour, circles Caesar and sits motionless at his feet.

In Dio, he comes unannounced, says nothing and falls to his knees. Dio says Caesar kept him in chains, showed him in his triumph and then had him put to death.

At Uxellodunum, Hirtius writes, Caesar “cut off the hands of those who had borne arms against him”.

033. The numbers, source by source

Caesar’s book gives figures for single events. The totals come from later writers, and their figures do not agree.

Event	Caesar’s own figure	Plutarch’s figure
Helvetii, 58 BC	368,000 set out; 110,000 counted going home	300,000, of whom 190,000 fighting men
Ariovistus’s Germans, 58 BC	No count: his cavalry “slew all the rest”	Dead “said to have been eighty thousand”
Nervii, 57 BC	Their envoys: 60,000 men down to scarcely 500 who could bear arms	Of 60,000, only 500 “said to have come off alive”
Usipetes and Tencteri, 55 BC	430,000, “the number of the enemy”	400,000 cut to pieces (Caesar); 300,000 slain, “it was believed” (Cato)
Alesia, 52 BC	80,000 inside; 8,000 cavalry and about 240,000 infantry coming to relieve it	At least 170,000 inside; 300,000 outside

• Velleius Paterculus: more than 400,000 of the enemy killed by Caesar, and a greater number taken prisoner. He wrote in AD 30, for the consulship of his friend Marcus Vinicius.
• Pliny the Elder: 1,192,000 killed in all Caesar’s battles outside the civil wars, not only in Gaul. He dedicated his Natural History to Titus in AD 77.
• Plutarch: in not quite ten years in Gaul, three million fought, one million killed and as many taken prisoner. He lived about AD 45 to 120.

Pliny’s Latin also says Caesar fought fifty-two pitched battles (bis et quinquagiens), more than M. Marcellus’s thirty-nine. The 1855 English translation prints “fifty”.

Why historians doubt the numbers

Hulot notes that such figures always call for caution in ancient history, and that modern historians have repeatedly stressed that Caesar’s 430,000 is exaggerated.

She also points out that Caesar does not say 430,000 were killed, only that the two peoples numbered that many. His Latin says the number of the enemy “had been” 430,000 heads: hostium numerus capitum CCCCXXX milium fuisset.

Some counts were second-hand even for Caesar: the Nervii figures came from their own envoys. He also called the Nervii almost annihilated in 57 BC.

Yet in 54 BC Caesar has “the Eburones, the Nervii, and the Aduatuci” attacking a Roman camp. Our reading: “almost reduced to annihilation” cannot be taken literally.

Our reading: the ancient totals for Gaul run from more than 400,000 killed (Velleius) to one million (Plutarch). All were written long after the war, and none shows how it was counted. Very many people were killed or enslaved. No source lets us say how many.

044. The massacre of 55 BC, and the word “genocide”

What Caesar says happened

Caesar tells it in book 4. In 55 BC the Usipetes and Tencteri crossed the Rhine with their families, after years of attacks by the Suevi. Caesar held talks with their envoys. On a day their envoys had asked to be kept as a truce, Caesar says, no more than 800 of their horsemen attacked Caesar’s 5,000 and routed them.

Next morning their princes and elders came to his camp; Caesar says they came to deceive him again. They said they had come to excuse the skirmish of the day before. Caesar, “rejoicing that they had fallen into his power, ordered them to be detained”.

He then attacked their camp. Those who could grab weapons fought among the carts. The rest, Caesar says, were “boys and women”; they fled, and he sent the cavalry after them.

The men in the camp, Caesar writes, heard the noise behind them and “saw that their families were being slain”. They fled to the meeting of the Meuse and the Rhine; after “a great number” of them had been killed, the survivors threw themselves into the river and drowned. Caesar gives no count of the dead. Caesar says his own men came back “all safe to a man”.

What Romans said about it

Plutarch, citing a writer named Tanusius, says that when the senate voted thanksgivings, Cato proposed handing Caesar over to the barbarians for breaking the truce. In Plutarch’s life of Cato, Cato urged Rome to hand Caesar over to the people he had wronged. After a second clash in the senate, Plutarch says, “nothing was done”, beyond saying that Caesar should be given a successor.

Suetonius does not name the massacre. He says that once, when the senate sent a commission to inquire into the Gallic provinces, some even recommended handing Caesar over to the enemy. Pliny the Elder refused to count the dead of Caesar’s battles outside the civil wars to his glory. In the 1855 English translation he calls it “an outrage committed upon mankind”, though Caesar may have been acting under necessity.

The finds at Kessel

Archaeology entered the story much later. In December 2015 the archaeologist Nico Roymans announced skeletal remains, swords, spearheads and a helmet from Kessel, in the Netherlands, linked to this massacre. The remains are of men, women and children, with spear and sword wounds. Tests on the teeth of three people showed they were not from the area.

Hulot notes that the finds were dredged from the Meuse between 1975 and 1995. The real novelty, she says, was a radiocarbon date that could match the middle of the first century BC in part. Doubts remain, she writes, about the real link with the event Caesar describes, and Roymans sets them out himself.

Is “genocide” the right word?

This is a question about how to describe the past, so we lay out the views rather than give a verdict.

Hulot sums up the change: the massacre “has long been seen as a war crime. More recently, it has increasingly been described as a genocide”.

She names M. Grant as, to her knowledge, the first to use the word for this massacre, followed by E. Horst. The word spread more widely around the turn of the millennium, she says. L. Canfora, M. Wyke, M. Á. Novillo López and Roymans use it too, and B. Kiernan speaks of a genocidal massacre.

She adds that Roymans mostly says “massacre” or “mass violence”, and uses “genocide” only to discuss it. Roymans himself wrote in 2019 about “the use of extreme mass violence against resistant groups” in Rome’s conquests.

Hulot’s own view is that the label “seems inappropriate”, because the ancient accusations “are of a different kind”. The ancient charge, she argues, was that Caesar attacked during a truce and laid hands on envoys whose persons were sacred.

She warns that calling ancient massacres genocide risks diluting the word’s meaning. She also argues that, in military terms, the massacre was neither a turning point nor of a proven scale.

The legal definition comes from Article II of the 1948 United Nations Genocide Convention. It covers acts “committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such”.

Some who use the word point to Caesar’s own text. In a 2023 chapter, the historian Tristan Taylor writes that Caesar “clearly states what can be regarded as a genocidal desire” about the Eburones. In 53 BC Caesar invited neighbouring peoples to plunder them, so that Gauls rather than his legionaries would risk the forests. He adds a second aim: that “the race and name of that state may be annihilated for such a crime”.

Taylor adds that mass enslavement, “although not readily caught by modern legal definitions of genocide”, would have the same effect, ending a people’s existence as a distinct group. He also notes that Caesar’s text presents such acts as justified, as punishment for wrongs done to Rome.

In 2021 the historian Kurt Raaflaub wrote of “the atrocities his army committed in slaughtering or enslaving entire nations”. He asked whether it is fair to condemn Caesar for a crime “recognized as such only recently”. Without condoning Caesar’s acts, he placed them within Roman and ancient customs of imperial warfare.

As Hulot notes, the voice of the victims is hard to reach in Roman history.

055. What the war meant

For Caesar it meant money and an army. Suetonius says he made Gaul a province and set a yearly tribute of 40,000,000 sesterces. He adds that Caesar “pillaged shrines and temples” there and “oftener sacked towns for the sake of plunder than for any fault”.

Plutarch says he used money taken from the enemy to win over citizens at Rome. Suetonius also says he “did not let slip any pretext for war, however unjust and dangerous it might be”.

After Julia died in 54 BC and Crassus was killed in 53, Pompey drifted apart from Caesar and became the champion of the senate. In January 49 BC Caesar crossed the Rubicon, the river that separated Cisalpine Gaul from the rest of Italy, with a single legion, and began a civil war against Pompey.

Our page on crossing the Rubicon tells what happened at the river. Our summary of Caesar’s life follows the story to the Ides of March.

For the Gauls it meant conquest, death and enslavement. After Alesia, Caesar gave “one of the remaining captives to each soldier, throughout the entire army, as plunder”. The 1911 Britannica says he then dealt “liberally with the conquered tribes, whose cantons were not broken up”.

066. How to read a commander’s own account

Caesar’s book is a good place to practise. The same habits work on any memoir of a war, a company or a campaign.

1. Ask who is speaking, and what he needed. Caesar had enemies in Rome, and some scholars read parts of his book as written with them in mind.
2. Separate what he saw from what he was told. The Nervii count was their envoys’ claim.
3. Ask what a number counts. In Hulot’s reading, the 430,000 counts two whole peoples, not the dead.
4. Compare a second source. At Alesia, Caesar and Plutarch give different armies and different surrenders.
5. Notice who never speaks. The Mandubii and the Usipetes have no voice here.
6. Watch the words that excuse. Hirtius opens the story of the cut-off hands by saying Caesar was “convinced that his lenity was known to all men”.

More tools: numbers that mislead and the SIFT method for checking a claim.

07Quick answers

08Use this

09Quick quiz

How we checked this · Last checked October 2026

• We read Caesar's book in the McDevitte and Bohn translation and checked key lines in the Latin text.
• Suetonius, Plutarch, Velleius, Pliny and Dio were read in their Loeb or older translations, and Pliny's battle count in his Latin.
• We read Sophie Hulot's 2018 article in full. David Henige's 1998 article could be read only on its first page, because the full text sits behind a bot check.
• Nico Roymans's 2018 chapter on Kessel could not be opened, and his 2019 article only on its first page. For the scholars Hulot lists as using the word genocide, we report only what she says.
• Appian's and Orosius's figures are left out, because we read them only in Hulot's summary.
• Tristan Taylor's 2023 chapter and Kurt Raaflaub's 2021 article were read in their published abstracts only.
• C. S. Kraus's 2009 chapter, which Hulot cites on when the book was written, could not be opened. We report only that she cites it.
• The 1948 legal definition is quoted from the United Nations page on genocide prevention.
• Dates for Dio, Velleius, Plutarch and Suetonius follow the Loeb introductions by E. Cary and F. W. Shipley, the Stanford Encyclopedia of Philosophy, the 1911 Britannica and the Historia Augusta.

How we check every article →

10Sources