How do we know what happened? Sources, bias, and historiography

The Problem

You open a history book and read: "The French Revolution began in 1789." That sounds like a fact. It has a year. It is in a book. It must be true.

But someone wrote that sentence. Someone chose to say "began" rather than "erupted" or "was triggered" or "had been building for decades." Someone decided that 1789 was the start date rather than 1787 (when the Assembly of Notables met) or 1792 (when the monarchy was abolished). Someone chose to call it a revolution rather than a civil war, an uprising, a betrayal, a liberation, or a catastrophe.

None of these choices are wrong. But each one is a choice, and every choice carries an angle. This essay is about those angles -- where they come from, how to spot them, and what to do about them.

What Counts as a Source

Historians work with evidence. The evidence falls into rough categories:

Primary sources are materials produced during the period being studied, by people who were there: letters, diaries, government records, newspapers, photographs, physical objects, buildings. A soldier's letter home from the trenches of 1916 is a primary source. So is the treaty that ended the war.

Secondary sources are works produced later, by people who were not there, analysing and interpreting the primary sources. The history book you opened is a secondary source. So is this essay.

Tertiary sources are compilations: encyclopaedias, textbooks, reference works. They summarise secondary sources for a general audience.

The further you get from the primary source, the more interpretation has been layered on. This is not a defect -- interpretation is how history becomes intelligible. But it means the reader must keep track of how many layers stand between them and the original evidence.

The gaps

Most of human history left no written records at all. For the vast majority of people who have ever lived -- farmers, labourers, enslaved people, women in most pre-modern societies, children, the poor -- there are no letters, no diaries, no government records. Their experiences exist only indirectly: in tax records that list them as property, in court documents that prosecute them, in archaeological evidence of how they lived. The historian must reconstruct their lives from traces that were created by people who did not consider them important enough to write about directly.

This means that "history" as it has been written is disproportionately the history of the literate, the powerful, and the victorious. The people who could write (or who employed scribes) are the ones whose perspective survives. The people who could not write are, in many periods, simply absent from the record.

Oral traditions are one corrective. Many cultures preserved their histories through spoken narrative rather than written text. These traditions can be remarkably durable -- some Aboriginal Australian oral histories describe geographic features that existed over 10,000 years ago, before sea levels rose at the end of the last ice age (Nunn & Reid 2016). But oral histories were often dismissed by European historians as unreliable, a bias that itself reflects the privileging of written evidence over other forms of knowledge.

Bias Is Not a Flaw -- It Is a Starting Condition

Every source has a perspective. A king's proclamation is biased toward the king's legitimacy. A rebel's manifesto is biased toward the rebel's cause. A newspaper report is shaped by the publication's audience, ownership, and political context. A diary is shaped by what the diarist chose to record and what they chose to omit.

The word "bias" is often used as an accusation, as if a source could be contaminated by it. But bias is not contamination -- it is the natural condition of any account produced by a human being with interests, loyalties, fears, and a specific position in the world. The question is not whether a source is biased (it always is) but how it is biased and what that bias reveals.

A practical approach:

1. Who wrote it? What was their position in society? What did they have to gain or lose?
2. For whom was it written? A private diary, a public speech, and a classified government memo tell different versions of the same events because they address different audiences.
3. When was it written? An account written during an event has immediacy but limited perspective. An account written decades later has perspective but may be coloured by subsequent events and the author's changing views.
4. What is left out? Absence is evidence. What a source does not say can be as revealing as what it does.

Historiography: The History of History

The study of how history has been written is called historiography. It asks not "what happened?" but "how have different historians understood what happened, and why did they understand it differently?"

Historiography reveals that historical interpretation changes over time, not because the facts change (the dates and events are mostly stable) but because the questions historians ask change.

Consider the fall of the Western Roman Empire. Edward Gibbon (The Decline and Fall of the Roman Empire, 1776-1789) attributed it primarily to internal moral decay and the rise of Christianity. Late 19th-century historians emphasised economic and administrative failure. Mid-20th-century historians like A. H. M. Jones stressed structural overextension. Peter Brown (1967) reframed the entire period as "Late Antiquity" rather than "decline," arguing that what Gibbon saw as fall was actually transformation. Contemporary historians like Bryan Ward-Perkins (2005) have pushed back, arguing that the collapse was genuinely catastrophic for living standards.

These are not merely different opinions about the same facts. They reflect fundamentally different frameworks for understanding what matters in history: moral character, economic structure, institutional design, cultural continuity, or material welfare. Each framework illuminates something the others miss. None captures the whole picture.

Major historiographical traditions

• Rankean history (19th century): Leopold von Ranke argued that the historian's task is to show "what actually happened" (wie es eigentlich gewesen) through rigorous use of primary sources. This aspiration to objectivity shaped academic history for a century, though later historians have pointed out that Ranke's own work was shaped by his Protestant, Prussian, liberal-nationalist perspective.
• Marxist historiography: history is driven by class struggle and the material conditions of production. Economic relations are primary; ideas, culture, and politics are superstructure. This tradition produced powerful analyses of social movements, labour history, and colonialism, but has been criticised for reducing complex events to economic determinism.
• The Annales School (20th century, France): shifted focus from political events and great individuals to long-term social, economic, and environmental structures (la longue duree). A Braudellian history of the Mediterranean is less about battles and treaties and more about geography, climate, trade routes, and grain prices.
• Postcolonial history: challenges Eurocentric narratives and foregrounds the perspectives of colonised peoples. Edward Said's Orientalism (1978) argued that Western scholarship about the "East" was fundamentally shaped by imperial power relations. Subaltern studies (Guha, Spivak, Chakrabarty) attempt to recover the histories of people who left few written records.
• Gender history: examines how gender roles, expectations, and power relations have shaped historical events and been shaped by them. Joan Scott's influential 1986 article argued that gender is a "primary way of signifying relationships of power" -- not merely a topic added to existing history but a category that reorganises how history is understood.

No single tradition has a monopoly on truth. Each asks questions the others neglect. A student who reads only one tradition gets a partial picture. A student who reads several traditions gets something closer to the complexity of what actually happened.

The Honest Position

This essay does not conclude with a method for eliminating bias. There is no such method. Every historian has a perspective, every source has a slant, and every framework highlights some facts while obscuring others. The honest position is not pretended neutrality but transparent positioning: state your sources, name your framework, acknowledge what you are not covering, and let the reader evaluate.

The history essays in this section follow this principle. Each one presents an event or question from multiple perspectives, names the historians and traditions behind each perspective, and leaves the reader to decide what they think. There is no single narrative being advanced. There are many narratives, and the skill being taught is the ability to hold several of them in mind at once.

That skill -- the ability to understand that the same event looks different depending on where you stand -- is not just a historical skill. It is a thinking skill. It is what this section is for.