Historical Document Set Curator
Design a document set for a document-based lesson — selecting and sequencing sources for analytical tension around a central question. Use when assembling sources for a new lesson or when an existing set produces flat responses.
What it does
Designs a document set for a document-based history lesson — recommending which sources to include, how to sequence them, and why each document earns its place in the set. The output includes the set design with rationale, the analytical tensions built into the set, a mapping of which documents support which historical thinking skills, a sequencing rationale, and a representation audit identifying whose perspectives are present and absent.
A document set is not a collection of sources on a topic. It is a carefully designed evidentiary environment that creates productive analytical tension around a central question. The difference matters: a collection of five documents about World War I gives students information to read; a curated set of five documents that contradict, complement, and complicate each other gives students a problem to solve. The quality of the document set determines whether students engage in genuine corroboration (weighing accounts against each other) or sequential summarisation (reading one source, then the next, without connecting them).
The Reading Like a Historian curriculum (Reisman, 2012; Wineburg, Martin & Monte-Sano, 2011) used 2–5 documents per lesson, selected for specific analytical properties: contradictions between accounts, differences in perspective or purpose, tensions between primary evidence and received narratives. The Pocahontas lesson paired Smith's two contradictory self-accounts with four historians who reached different conclusions from the same evidence — creating layered analytical tension that makes corroboration not just possible but necessary.
This skill also addresses the representation question that Caswell (2014) raised: whose history is documented, whose is absent, and what the implications of those absences are for what students can learn. A document set on early American colonisation that includes only English-authored sources doesn't just limit perspectives — it structurally prevents students from practising certain analytical moves (they cannot corroborate across genuinely different viewpoints if all viewpoints share the same cultural assumptions). The skill includes a representation audit that names these gaps honestly.
The evidence behind it
Reisman (2012) used document sets of 2–5 primary sources per lesson in the Reading Like a Historian curriculum. Each lesson's documents were selected to create a central tension requiring investigation. The significant treatment effects for sourcing and close reading (but not corroboration) suggest that document set design matters: the sources must create opportunities for the target skills. Reisman attributed the null corroboration finding partly to the near-absence of whole-class discussion, but the document set design also plays a role — corroboration requires sources that genuinely tension with each other, not just sources that say different things.
Wineburg, Martin, and Monte-Sano (2011) provided detailed rationales for their document set designs. The Pocahontas chapter uses six sources in three rounds: Round 1 (Smith's two accounts) creates the foundational contradiction; Round 2 (Adams and Lemay) shows historians disagreeing about what the contradiction means; Round 3 (Lewis and Barbour) adds further interpretive complexity. Each round deepens the investigation without overwhelming students with too many documents at once. This progressive architecture — contradiction first, then competing interpretations, then further complication — is a design principle, not an accident.
Wineburg and Martin (2004) demonstrated that document selection shapes what students can discover. The Pocahontas unit begins with Disney's version to activate prior knowledge, then dismantles it through primary source comparison. The documents are sequenced to create cognitive dissonance: what students think they know (from Disney) is destabilised by what the sources reveal. This sequencing principle — start with the familiar, then complicate — is central to effective document set design.
Caswell (2014) argued that archival absence constitutes symbolic annihilation — when a community's history is not preserved, that community is rendered invisible in the historical record. For document set design, this means that the available pool of sources is not neutral. A document set on South Asian American history requires the kind of community-archiving work that SAADA does; without it, the curriculum has nothing to work with. Teachers should be aware that some topics will have richer source pools than others, and that absence from the archive is itself historically significant.
Theimer (2012) emphasised that how sources are collected, arranged, and presented shapes what historical thinking is possible. A source removed from its archival context — stripped of information about who collected it, how it was preserved, and what other materials surrounded it — loses the contextual information that sourcing and contextualisation depend on. Document sets should preserve provenance information wherever possible.
Sources
- Reisman (2012) — Reading like a historian: a document-based history curriculum intervention in urban high schools
- Wineburg, Martin & Monte-Sano (2011) — Reading like a historian: teaching literacy in middle and high school history classrooms
- Wineburg & Martin (2004) — Reading and rewriting history
- Wineburg (2007) — Unnatural and essential: the nature of historical thinking
- Caswell (2014) — Seeing yourself in history: community archives and the fight against symbolic annihilation
- Theimer (2012) — Archives in context and as context
How to use it in your lesson
For the best results with EvidenceLesson, give it:
- central_question — The central historical question the document set must support — the inquiry that students will investigate using these sources
- student_level — Age/year group, reading level, and experience with document-based inquiry
- target_skills — Which historical thinking skills the document set should foreground — e.g. sourcing and corroboration, close reading, all four
- existing_sources (optional) — Sources the teacher already has or plans to use — the skill will evaluate these and suggest additions, removals, or resequencing
- set_size (optional) — Target number of documents — typically 2–5 for a single lesson, more for a multi-day unit
- source_constraints (optional) — Practical constraints — e.g. sources must be freely available online, must be in English, must be shorter than 200 words each, must include visual sources
- representation_priorities (optional) — Whose perspectives should be represented in the set — e.g. ensure Indigenous voices are included, include perspectives from multiple social classes, include women's perspectives
- historical_topic (optional) — The broader topic or unit the lesson sits within
- curriculum_framework (optional) — From context engine: relevant curriculum standards or historical thinking framework in use
Known limitations
- This skill designs document sets but cannot supply the documents themselves. It recommends specific sources by author, date, and type, but the teacher must locate, obtain, and (where necessary) adapt them. For freely available sources, the SHEG/DIG curriculum at sheg.stanford.edu provides adapted primary sources for many US history topics. For other topics, teachers will need to source documents from archives, textbooks, or digital collections.
- The representation audit identifies gaps but cannot always fill them. For some historical topics, the voices of marginalised groups were not documented — not because they didn't exist but because the power structures that determine what gets preserved excluded them (Caswell, 2014; Theimer, 2012). The skill names these silences honestly rather than pretending they can be easily resolved. In some cases, the absence itself becomes the most important thing to teach.
- Document set design is inseparable from the central question. A set designed for "Did Pocahontas rescue John Smith?" will not work for "What was the nature of Powhatan-English relations?" The skill evaluates the set against the specific question provided. If the question changes, the set should be re-evaluated.