How we research history
A documentary about real people is a set of claims about what happened to them. This page explains how we decide which claims to make, and how you can check them.
Every claim has a source, and every source has a label
Each episode and Knowledge Base entry lists its sources in full. Every source carries two labels: what kind of evidence it is, and whether we have checked it against the original.
- Primary sources
- Material created by people who took part in or witnessed the event, at or near the time: letters, diaries, orders, logs, telegrams.
- Contemporary records
- Documents produced at the time by institutions rather than individuals: hospital registers, casualty returns, inquiry reports, newspapers.
- Eyewitness testimony
- What someone who was there said about it. Valuable and fallible. We note when it was given and to whom.
- Archival documents
- Records held by an archive, cited by collection and item number so anyone can find the original.
- Scholarly secondary sources
- Peer-reviewed research and books by historians who have worked through the primary material.
- Later recollections
- Memoirs and interviews given years or decades afterwards. Memory changes. We frame these as memory: "Speaking in 1989…"
- Disputed claims
- Where sources disagree, we give each version and who holds it. We do not pick a winner to make the story cleaner.
- Historical interpretation
- Arguments historians make about why something happened. We attribute them, and we don't present them as fact.
Checked, not yet checked, disputed
- Checked against the source
- We have read the original document, or a reliable scan of it, and confirmed the wording or figure.
- Not yet checked
- The citation comes from our research files and points to a real source, but we have not yet confirmed the specific passage. Pages with unchecked claims are marked as drafts and list what still needs checking.
- Accounts conflict
- Two or more sources disagree. We show each version.
When the sources disagree
They often do. How many people were aboard the Wilhelm Gustloff: three counts in our sources, about 3,000 apart. Whether Cholula was planning an ambush in 1519: the Spanish and Tlaxcalan accounts say yes, Mexica tradition remembers it differently. The death toll of the Halifax Explosion: sources range from 1,950 to over 2,000.
We do not resolve these by choosing the number that sounds best. Every episode page has a section called What is still disputed, which sets out each disagreement and who holds each position.
Whose words are these?
A quotation has to be what someone actually said or wrote. When an account was written down by someone else, translated, or retold decades later, we say so on the quotation itself. Two examples from our own research:
- The account of Dr. Clement Ligoure's work after the Halifax Explosion survives as a third-person record made by the archivist Archibald MacMechan. Some articles online rewrite it as Ligoure's own words. We quote it as MacMechan's.
- Hans Rittner's account of the Gustloff sinking was told to and written by A. A. Hoehling. We label it as mediated and do not use its figures.
Composite characters and reconstruction
Some episodes follow a composite: a figure assembled from the evidence to stand for people the sources describe without naming. Composites are always labelled as composites, on screen and on this site. They are never given a real person's name or experiences.
Our documentaries are illustrated, not filmed. Illustration declares itself: nobody should mistake a drawing for a photograph. Every illustrated scene is a reconstruction based on the sources listed with that episode. Where generative tools are used in production, every frame goes through an accuracy check against the research before it is used, and we follow YouTube's disclosure rules.
What we will not show
- We show aftermath, not the act. The field after, the hands after, the room after.
- No animated gore, and no deaths staged for effect.
- Medical procedures are shown the way a medic would describe them: hands, instruments, the patient's face.
- Real archival photographs are never animated or redrawn. They stay still, sourced and labelled.
- We never put a dead or dying identifiable person on a thumbnail.
What a modern medic's eye adds, and what it doesn't
The project's creator served as a Navy corpsman with the Marines. That shapes the questions we ask: what the wound was, who treated it, how long evacuation took. It does not make modern experience an authority on Roman surgery or 1917 ophthalmology. Historical claims stand on historical evidence. Where we explain an injury in modern medical terms, we say that the explanation is modern.
Corrections
History gets revised when better evidence turns up, and so will this site. If you find an error, or know a better source, email corrections@humancostofhistory.com with the page, the claim and your source. Every correction we make is logged on the corrections page with the date and what changed.