Historical Method and Document Analysis: Sources, Causality, Continuity and Argument
The method section is tested inside document-based questions: you must identify a document, place it in context, judge its value and limits, and use it as evidence in a structured argument. The exam rewards explicit reasoning, not recall alone.
Reading a document
- Nature: primary (produced in the period - law, petition, cartoon, census, letter, speech) or secondary (a historian's interpretation, textbook, article).
- Origin: author, date, place, position (governor, bishop, union leader, journalist).
- Purpose and audience: persuade, inform, justify, record.
- Content: what is said, and what is omitted or assumed.
- Context: which event, which debate, which period.
Evaluating evidence
Ask: could this author know? What interest do they have? Is the claim fact or opinion? Is it corroborated by another source? A law proves intent, not results; a census proves a count, not a cause; a cartoon proves an argument, not an event.
Causality and change
- Distinguish long-term structures (seigneurial tenure, clerical control of education, staple exports) from short-term triggers (a strike, an election, a crash).
- Distinguish causes, consequences and interpretations.
- Continuity and change: name what persisted (French civil law after 1763, confessional schools until 1998) as well as what changed.
- Chronology: know the hinges - 1763, 1774, 1791, 1840, 1867, 1940, 1960, 1982.
Building an argument
Thesis → three evidence-based points → documents used as proof → a counter-argument addressed → conclusion. Cite documents by nature and date, never "the document says".
Worked example
Two 1917 editorials: Bourassa's Le Devoir opposing conscription and a Montreal English daily supporting it. Reasoning: both are primary and persuasive, and both select evidence. Bourassa argues francophones have no imperial obligation; the English paper argues duty and sacrifice. Neither proves what "most Canadians" thought. To weigh them, add enlistment statistics, the 1917 election results by riding and Borden's speeches. Conclusion: the split is regional and linguistic, visible in returns, not only in rhetoric.
Traps
- Treating a secondary source as if it were the event.
- Ignoring date and author and reading the present into the past (anachronism).
- Confusing correlation with causation in statistics.
- Describing documents instead of using them as evidence.
- Narrating instead of answering the question asked.