History · Archaeology & historical evidence

How historians read a source

History is built from sources made by people with their own aims. Learn to ask who made a source, when, for whom and why, and what historians do when sources disagree, fall silent or get rewritten.

4 short lessons · about 17 minutes · 7 sources · updated November 2026

What you will learn

  • Use who, when, for whom and why to judge what a source can be trusted to tell you
  • Explain how a slanted source can still give trustworthy evidence by accident
  • Explain how historians use independent sources to check each other
  • Explain why silence and lost records limit what surviving sources can prove

Lesson 1 of 4 · 4 min

Four questions to ask any source

Before you readJulius Caesar wrote the main surviving account of his own war in Gaul. What do historians do with it?

Correct.Not quite. The answer is Use it, asking what Caesar wanted people to think.

Historians use it, with care. A source made by someone with an aim is still evidence, as long as you ask what the aim was.

In 58 BCE, Julius Caesar says he found a census in an enemy camp. Written in Greek on tablets, it counted 263,000 Helvetii and 105,000 allies, and exactly a quarter were fighters. Should you believe him?

Caesar's account is a primary source, something made at the time you're studying. Historians start with four questions. Who made it? Caesar, the commander, writing about himself in the third person. When? During or soon after the war, published by 46 BCE.

For whom? Roman readers back home. Why? To make his war look justified and his wins look huge. The historian David Henige notes that Caesar was the aggressor, though he claimed to be defending Rome.

So the census looks shaky. Why would Gauls keep records in Greek, or haul piles of tablets on a migration? And a neat quarter is suspiciously convenient. The questions don't throw the source out. They tell you which parts lean toward what its author needed.

Check yourselfA general's account of his own campaign says the enemy army was huge and his losses were tiny. Which question helps a historian most in judging those numbers?

Correct.Not quite. The answer is Why did the general write it, and for whom?.

The general had a reason to make the enemy big and his losses small, so purpose and audience come first. Being there gave him knowledge, but it didn't remove his motive.

A slanted source still tells you things its author never meant to say. The historian Louis Gottschalk gave an example. A Roman road inscription says a certain governor built the road while Augustus ruled.

You might doubt the governor did the building, since officials liked credit. It's much harder to doubt that the road existed under Augustus. Nobody had a reason to fake that part. Historians call this kind of detail incidental evidence.

Gottschalk's second example was an advert boasting about a brand of coffee at an unusual price. Doubt the boast if you like. The advert still proves the brand existed. So you trust each statement separately, by asking whether the author gained anything from it.

Try to recallA Roman inscription credits a governor with building a road during Augustus's reign. Why might a historian doubt the credit but trust the date?

Officials gained prestige from claiming public works, so the governor's credit may be inflated. Nobody gained anything by placing the road in Augustus's reign, so that incidental detail is much harder to doubt.

Ask who made a source, when, for whom and why, and you'll see which claims lean toward the author's aims. Then look for the details nobody had a reason to invent, because those are often the most trustworthy.

Lessons in this topic

  1. Four questions to ask any source

    See how historians judge a source by who made it, when, for whom and why, and how even a slanted source gives away facts by accident.

    Essentials · 4 min

  2. When sources disagree, or stay silent

    Learn how historians test one source against another, why copies don't count as extra witnesses, and what to make of silence and lost records.

    Essentials · 4 min

  3. Reading a king's own account of himself

    Read Darius's cliff inscription and Augustus's list of his deeds as boasts with a purpose, and see what comparing them with other sources reveals.

    Deeper · 5 min

  4. Why history gets rewritten, and when it's denial

    See how new finds and new questions change what historians say, and what separates honest revision from denial that bends the evidence.

    Deeper · 4 min

Three quick stories from this topic

A boast on a cliff

Darius had his life story carved about 100 meters up a cliff above a royal road

The Behistun inscription is about 15 meters tall and written in three languages. Darius sent the text out in copies too, and one in Aramaic turned up in Egypt.

Written for the future

The text tells future readers to believe it and not to hold it to be lies. It also curses anyone who destroys the carving.

Why historians doubt it

Darius says he killed an impostor posing as the old king's brother. Most modern historians think the man he killed was the real prince.

The pharaoh's disputed victory

A Hittite king complained in writing about the pharaoh's claim to have won at Kadesh

Ramesses II carved his victory at Kadesh on temple after temple in Egypt. Hittite records say he left in defeat, and a letter from Ramesses answers the Hittite king Hattusili III's complaint about his boasts.

Two sides, two endings

Egypt's carvings show the pharaoh charging the enemy and saving the day. Records from the Hittite capital, Hattusa, say he went home beaten.

What both agree on

Ramesses didn't take Kadesh, and the land near it was Hittite again within a year. Most historians call the battle a practical draw.

A Roman birthday invitation

A birthday invitation from about AD 100 may be the oldest Latin writing by a woman

Claudia Severa invited her friend Sulpicia Lepidina to a party at a fort near Hadrian's Wall. A scribe wrote most of the wooden tablet, and Severa added the closing greeting in her own hand.

Mistaken for shavings

The Vindolanda tablets were first taken for wood shavings in 1973. The ink faded in the air, so infrared photographs were needed to read it.

Why it matters

Letters like this put wives, friends and parties into the story of a frontier known mostly through soldiers. New evidence like this is one reason history gets rewritten.

Sources

  • Pompeii's body casts and what they can prove

    See how Pompeii's plaster figures were made, what bombing, repair and copying did to them, and why DNA overturned stories long told about the victims.

    4 lessons · 17 min

  • How a dig knows what came first

    Nothing at an archaeological site comes with a date attached. Two methods do almost all the work. One puts events in order, and the other pins a piece of wood to a calendar year.

    4 lessons · 23 min

  • Calendars, dates and historical periods

    Year numbers and labels like "Middle Ages" are tools people invented. See how BCE and CE work, why other calendars count differently, where period names came from and how historians date events no calendar recorded.

    4 lessons · 16 min

  • Alexander the Great

    A Macedonian king who never lost a battle and fought in India before he turned 30. See how he took the Persian Empire, how he tried to rule it, and what was left when he died at 32.

    4 lessons · 16 min

  • Ancient Greece

    Hundreds of small city-states shared a language and gods but never a government. See how Athens and Sparta worked, how Greeks spread around the sea, and why their culture outlasted their freedom.

    4 lessons · 16 min

  • How the First World War began

    Trace the decisions that turned an assassination into a wider war in 1914. Along the way, you'll separate the trigger from long-running tensions and from choices made under pressure.

    4 lessons · 17 min

All history topics · Take the quiz