Did you know? Two historical documents can disagree about a date without either one being a deliberate lie. One writer may record the first day something was used; another may remember the public ceremony; a third may write down the day the news reached them. An impressive-looking date on an old page is therefore a starting point for investigation, not automatically the last word.
Resolving a historical date disagreement means identifying the precise event each source describes, checking when and why the source was created, comparing independent evidence, and reporting the most defensible conclusion with its remaining uncertainty. Do not choose the older-looking document or the tidier date by instinct. If the evidence does not establish one date, a good historical answer can clearly say so.
This guide gives students, parents and teachers a complete fictional school-history exercise, a worked evidence table, a practical source-checking method, questions for diagnosing misunderstandings and guided-to-independent practice. Its purpose is not to make children distrust every book. It is to teach them how responsible historians move from documents to claims.
First ask a deceptively difficult question: ‘What exactly happened?’
The sentence ‘The bridge opened in September’ hides several possible events: construction finished, the first crossing was allowed, the ribbon was cut, or an official announcement was published. These milestones may occur on different days. If two accounts are using ‘opened’ for different milestones, simply writing them side by side can create a false contradiction.
The very first step is to define the claim narrowly. Instead of ‘What was the opening date?’ ask ‘What date was ordinary public use first permitted?’ or ‘On what date was the formal ceremony held?’ Make a separate line for each. Only then can you decide whether sources genuinely disagree about the same event. In history, accuracy begins with a carefully framed question.
The Library of Congress guide to primary-source inquiry recommends observing sources, questioning their context and comparing evidence. The US National Archives document-analysis materials similarly ask students to inspect a document before using it as proof. These are general research practices; the bridge and people in our example below are deliberately invented.
The worked example: two ‘opening’ dates on a fictional school-history poster
Imagine a student researching the fictional North Quay Footbridge. The names and records in this exercise have been invented solely to make the method visible. No claim is being made about a real bridge, town or historical event. In the exercise, the student receives two documents and is asked to write one date on a display.
- Source A: a clerk’s dated operations entry, written on 8 September 1924, says, ‘Pedestrian passage first permitted from 6 September 1924.’
- Source B: a newspaper article published on 8 September 1924 reports, ‘The bridge was officially opened yesterday,’ giving a ceremony date of 7 September 1924.
- Initial student answer: ‘One source says 6 September and another says 7 September, so one is wrong.’
Not so fast. The documents make different specific claims. Source A discusses first permission for passage; Source B discusses the formal opening ceremony. Both could be accurate. The student’s task is not to average the dates, pick the most recent publication or force agreement. It is to show the difference in meaning.
Here is a defensible poster sentence under the fictional assumptions that both records are authentic and accurately transcribed: ‘The operations entry records public passage beginning on 6 September 1924, while the newspaper reports an official opening ceremony on 7 September 1924.’ Notice that the sentence keeps the source attached to each claim. It does not turn ‘the newspaper reported’ into an unqualified eyewitness fact.
Source date, event date and publication date are three separate things
In Source B, the newspaper was published on 8 September, but its word ‘yesterday’ refers to 7 September. The article’s publication date and the date of the reported ceremony are not the same. Source A was also written on 8 September, but records permission beginning on 6 September. The same source date can therefore document an earlier event.
The arithmetic here is simple but worth checking: in this invented example, the day before 8 September is 7 September. No time-zone shift or midnight timestamp has been provided. If the newspaper date were uncertain, or if the text said ‘last Sunday’, we would need further context before assigning a calendar day. The role of the number is to identify a claim, not to create unwarranted confidence.
For longer projects, draw three columns headed date written, date claimed and event described. This tiny change prevents many false arguments. If a student puts 8 September into the ceremony column merely because the paper is dated 8 September, the error is not in historical opinion. It is a reading and evidence-label error that can be corrected immediately.
Now make the example genuinely contradictory
Suppose a third fictional record, Source C, is a handwritten ceremony programme bearing the line ‘Formal bridge opening: 8 September 1924’. Source B indicates a ceremony on the seventh, and C indicates the eighth. Once we define the event as the formal ceremony, the sources appear to contradict each other. Source C might be a draft programme, contain a printing mistake or record a postponed ceremony. Source B might have used a mistaken date. None of these explanations is established merely by proposing it.
A strong response is: ‘Sources B and C give different ceremony dates. I need to check whether the programme was final, whether the newspaper report described an event that actually occurred and whether independent evidence confirms either day.’ This is not indecision. It identifies exactly what is missing from the case.
How to build a small evidence table
- Claim: First pedestrian use began on 6 September. Support: Source A operations entry; remaining check: whether the entry is authentic and contemporaneous.
- Claim: Official ceremony occurred on 7 September. Support: Source B newspaper retrospective ‘yesterday’; remaining check: the reporter’s information and an independent corroborating source.
- Claim: Official ceremony occurred on 8 September. Support: Source C programme; remaining check: whether the programme is a final record or a preliminary schedule.
- Safe interim conclusion: the earliest-use date and ceremony date should not be merged; the ceremony date remains disputed between B and C.
Students sometimes believe an evidence table must end with a single bold answer. It does not. The table should let a reader see what each source actually establishes. If a source only tells us what somebody planned, it may not establish what occurred. A dated invitation, for example, is usually strong evidence of an announced event date, but not by itself proof that the event went ahead that day.
Five questions that make historical evidence easier to evaluate
- Who produced it? Was it a participant, clerk, witness, editor or later compiler?
- When was it produced? During the event, soon afterwards, or decades later?
- What did it mean to record? An action that occurred, a planned action, a rumour or a commemoration?
- How directly could the creator know? Does the source report personal observation, an official register or somebody else’s account?
- Can we independently corroborate it? Are several sources truly separate, or are they repeating one original report?
Source credibility is not a contest in which a photograph always beats a newspaper or an official stamp always wins. A photograph may lack a trustworthy date. A newspaper may have good access but rely on a single reporter. A register may be excellent for transactions while saying nothing about the public celebration. Choose evidence that is well matched to the specific question.
Singapore’s MOE educator resource Reading multiple historical texts, updated 8 September 2026, discusses sourcing, corroboration and contextualisation as skills for comparing differing texts. It supports teaching students to investigate inconsistencies rather than treating all texts as identical lists of facts. That resource does not certify the invented North Quay scenario; the worked example here is an original teaching illustration.
Why three websites are not always three witnesses
Imagine that Sources D, E and F all reproduce Source B’s newspaper sentence, including the same punctuation and a distinctive error. Counting them as three votes for 7 September would overstate the evidence if they all copied one article. Verify their reference notes and original publication trail. The related Core Aim lesson, how to check whether history sources are independent or repeat one report, gives this problem its own focused treatment.
A useful rule is to compare independent chains of evidence, not just the number of pages found by a search engine. A later historian may provide strong reasoning based on several archived records, but the student should examine the historian’s citations rather than assuming a confident narrative sentence must be correct.
A diagnostic exercise for a student who says ‘I found two dates’
Ask them to underline the verb that each source uses. Does one say ‘announced’, one ‘planned’, one ‘began’, and another ‘celebrated’? Then ask for the name of the precise event. A student who can separate announcement, occurrence and remembrance has identified why the dates may differ. A student who says only ‘old sources can be unreliable’ needs a more specific explanation.
Next, ask: ‘What new source would best help distinguish a planned ceremony from the ceremony that occurred?’ Possibilities include a dated photograph with reliable provenance, a contemporary diary describing attendance, a subsequent official minute or an independent newspaper report. Each has limits. A single source may support the claim without proving it beyond all doubt.
Guided practice one: relative dates in a document
An entirely fictitious letter is dated 12 March 1931 and says, ‘The exhibition opened yesterday.’ Under the straightforward calendar assumption, the writer claims an opening on 11 March 1931. But we should still label it as the letter writer’s claim, not an independently established event. Have the pupil write: date written = 12 March; date claimed = 11 March; wording = ‘yesterday’.
Guided practice two: plan versus occurrence
A fictional invitation printed on 1 August announces a debate for 5 August. A classroom newspaper report dated 7 August says the event happened on 6 August after a weather delay. There is no necessary contradiction if the event was postponed. The invitation supports the original plan; the report supports the asserted occurrence. Ask the student to write two sentences rather than silently changing one document to match the other.
Independent practice: make a qualified historical statement
A fictional school records a foundation-stone ceremony in an invitation dated 20 June, while an undated photograph label says ‘opening day, 21 June’. You do not know whether the label refers to the same ceremony, a separate public opening or a later archivist’s mistake. Write a two-sentence caption that names both pieces of evidence and states a useful next check. A valid response should avoid asserting that either document is definitely false.
One possible answer: ‘A surviving invitation schedules the foundation-stone ceremony for 20 June, while a photograph label gives 21 June as an opening day. The label needs independent dating and clarification of which event it names before these dates can be treated as contradictory.’ Your own answer may use different words if it keeps the evidence and uncertainty honest.
How parents and teachers can see progress
For a short four-source task, score progress by whether a learner can: identify each document’s creator and date; distinguish the document date from the event date; define the precise event; explain why sources may disagree; and write a qualified conclusion. A child who gives a tentative answer supported by sources may show more historical understanding than a child who quickly memorises a date without knowing where it came from.
A teacher can provide a fictional pair first, then introduce real archived sources with reliable catalogue descriptions. A parent can ask, ‘What does this source actually say?’ and ‘What else could we check?’ without turning the discussion into a test of obscure historical facts. When the student is ready, invite them to show another person the evidence table and explain the differences independently.
Common mistakes and what to do instead
- Choosing the newest publication: a later article is not automatically more accurate; inspect its evidence.
- Choosing the oldest-looking paper: age is not a substitute for authenticity, relevance or context.
- Taking a schedule as proof: a planned ceremony can be moved or cancelled.
- Combining events with different names: first use and formal inauguration may occur separately.
- Counting copied articles as independent corroboration: follow the source trail.
- Writing one date despite unresolved contradiction: state what is known and what would settle it.
Frequently asked questions
Is a primary source always correct?
No. A primary source may give valuable firsthand or contemporary evidence while containing omissions, errors or bias. Ask what the creator was in a position to know.
Can a school history answer include two dates?
Yes, when the sources describe distinct events or when a genuine disagreement remains. The wording should explain why both dates appear; it should not simply list alternatives without analysis.
What if the only source is a later textbook?
Use the textbook carefully and cite it. If a high-stakes claim matters, check whether it identifies earlier evidence and whether those sources are accessible. An unverified secondary account can be a starting point, not an automatic final authority.
How should I place an uncertain date on a timeline?
Label the uncertainty or show the range and the source. For a separate practical treatment, read how to build a timeline when a source gives an approximate date. A disputed exact date and a genuinely approximate date are related but not identical problems.
What this teaches beyond the history classroom
A learner who can separate ‘reported’, ‘planned’, ‘observed’ and ‘verified’ is also better prepared to read science reports, examine internet claims and write balanced explanations. This evidence-first process reflects the diagnosis, explanation and independent checking approach of eduKateSG’s Clementi Mathematics reference, adapted to historical inquiry. More reading pathways are available through the Education Hub and the Core Aim of Education hub.
Remember: when historical dates disagree, first ask whether the sources are describing the same event. If they are, examine the provenance and evidence; if they are not, explain the different milestones. A clear statement of what remains unknown is a mark of careful thinking, not a failure.