Methodology & limitations
mementomorti documents publicly reported homicides and attempted homicides in Europe and follows each case through the justice system. This page explains how the archive is built and — just as importantly — what it cannot tell you.
What we record
Each entry is tied to a place, a date, a crime category, a judicial status, and one or more original sources. We record:
- where and when the incident was reported,
- the crime category (homicide, attempted homicide),
- the case's judicial status and how it changed over time, each step linked to a source,
- links to the original reporting, with archived copies where available.
What we deliberately do not record
- No names. We do not publish the names of victims or of suspects, accused, or convicted persons. Entries are identified by place, date, and source links only. This protects against defamation, respects the presumption of innocence, and limits the processing of personal data.
- No characterization. Summaries are factual and neutral. We do not editorialize.
- No ethnicity, nationality, or immigration data. These are not part of the archive.
Sources
We accept only sources from an approved whitelist: national newspapers, public broadcasters, reputable local outlets, and official police or court communications. Submissions citing social media, anonymous reports, or unverified claims are rejected. The whitelist is published in full for scrutiny.
Judicial status
The unattended publication workflow researches sources and applies automated scope, consistency, duplicate, and anonymization checks. These checks do not constitute independent human editorial review or guarantee that a source's account is correct. Unresolved required facts block publication.
A single case moves through stages — alleged → charged → on trial → convicted / acquitted / dismissed. News coverage almost always stops at the arrest. We track the outcome, because an arrest is not a conviction and a charge is not guilt. Every status change cites its source.
Limitations — please read
This archive is a documented floor, not a complete count.
- It undercounts. We capture only what is publicly reported and what we find. The true number of incidents is higher and unknown. Any total shown here is a lower bound, never a complete figure.
- It is not a representative sample. Media coverage is uneven; some cases are reported widely and others not at all. The set of cases here therefore cannot support population-level conclusions or comparisons between groups, regions, or time periods.
- For authoritative totals, consult official statistics such as Eurostat and national statistical offices. We link to them rather than competing with them.
- Information changes. Cases are reviewed and updated; corrections are logged publicly on each entry.
Archive publication dates
Monthly archive briefings record additions, material updates, and corrections from the publication ledger introduced at rollout. Earlier records form a baseline without invented publication dates. Publication timestamps mark preparation of validated records for the publication commit; deployment follows that commit. They are distinct from incident dates and judicial-event dates. Review-only timestamp changes do not count as material updates. Withdrawn records are excluded from public briefings and feeds.
Corrections & removal
Every entry shows a "report" option and a public corrections log. See Corrections & takedown for how to flag an error or request review.