Great Purge
Enable an AI agent to identify, contextualise and assess evidence about the Great Purge while distinguishing its campaigns, victims, institutions and contested historical boundaries.
Research draft, second pass
A second pass drafted this model: the structure a model of this thing needs, and what is known about it in the world. The line under this one says how the second half was obtained - researched against sources, or recalled without web access, in which case nothing here was read anywhere and every claim is a lead to verify. Unreviewed either way.
recalled by Codex without web access - no source was read
Researched by: Codex
Purpose and description
Enable an AI agent to identify, contextualise and assess evidence about the Great Purge while distinguishing its campaigns, victims, institutions and contested historical boundaries.
The Great Purge was a campaign of political repression in the Soviet Union under Joseph Stalin, conventionally dated to 1936-1938, involving party and military purges, staged trials, mass arrests, executions, and imprisonment, with mass terror peaking in 1937-1938.
It can be Determine whether a documented episode belongs within a stated definition of the Great Purge.; Trace connections among directives, implementing bodies, accusations and outcomes.; Compare population estimates after aligning their scope and counting rules.; Identify unsupported inferences from coerced testimony or official targeting labels.; Build evidence-linked timelines of persecution and subsequent case review.; Flag unresolved attribution, contradictory records and missing documentation for further research..
Distinguishing features
An included episode must have an evidenced connection to the specified Soviet purge campaigns; occurrence under Stalin alone is insufficient.
The model distinguishes party expulsion from arrest, imprisonment and execution rather than treating all uses of purge as equivalent.
Public political trials and less visible mass operations are represented separately, even when a historical account groups them under one label.
Campaign membership requires a documented institutional, procedural or targeting connection rather than merely a date within 1936-1938.
Great Purge and Great Terror are treated as source-dependent scope labels whose equivalence must be established for each account.
Scope
+ Competing periodisations and definitions of the Great Purge and Great Terror
+ Relationships among party purges, public trials, military repression and mass operations
+ Authorisation, implementation and revision of repressive campaigns
+ Targeting, accusations, coercive procedures and victim outcomes
+ Evidence, population estimates, rehabilitation and historical interpretation
- Soviet repression across its entire history
- Collectivisation, famine and deportation campaigns except where directly connected to the defined purge episode
- Complete biographies of Stalin, officials, perpetrators or victims
- The entire organisational history of the Communist Party, NKVD or Gulag
- Political purges as a general class of action
- Later commemorative institutions as independently modelled organisations
Characteristics
- Historical designation
- Great Purge; Great Terror; source-specific alternative Terminology can change which campaigns and populations an account includes.
- Temporal boundary
- Start and end dates, date precision and supporting periodisation Totals and causal arguments depend on the interval selected.
- Campaign affiliation
- Links to named trials, operations, directives and institutional purges Connects individual episodes to the broader event without collapsing their differences.
- Targeting basis
- Political position; attributed social background; nationality or foreign connection; personal association; other documented designation Separates authorities' classifications from verified attributes of the people targeted.
- Decision mechanism
- Party disciplinary process; court or military tribunal; extrajudicial body; other documented mechanism; unresolved Identifies how accusations became sanctions and which institutions participated.
- Recorded outcome
- Expulsion; dismissal; arrest; release; imprisonment; execution; death in custody; other; unknown; multiple sequential outcomes Prevents distinct harms and successive events from becoming interchangeable counts.
- Affected population estimate
- Persons or cases, with outcome, interval, territory, inclusion rule, uncertainty and deduplication method Makes quantitative claims comparable and exposes incompatible denominators.
- Evidence assessment
- Unassessed; corroborated for a specified claim; disputed; insufficient; inaccessible A document may establish an official accusation without establishing its truth.
- Subsequent case disposition
- Not established; reviewed; sentence revised; conviction annulled; rehabilitated; other documented disposition Separates later legal or administrative decisions from original outcomes and historical interpretation.
Also called
Where this came from
wikidata · CC0 1.0
Drafted structure
Bundle to layer to finding to question, as the second pass will find it: 5 bundles · 9 layers · 15 findings · 25 questions.
Event boundaries and campaigns Represent what a particular account includes under Great Purge and how its component campaigns relate.
The event cannot be recognised consistently if terminology, dates and campaign membership are silently assumed.
Terminology and periodisation
Record attributed definitions and the evidence used to establish temporal boundaries.
Attributed event definition
Capture each source's use of Great Purge or Great Terror, including whether it foregrounds elite repression, mass operations or both.
- What does this source include under Great Purge, and does it distinguish that label from Great Terror? definition
- Which developments justify its starting and ending dates, and which connected episodes fall outside them? boundary
Component campaign membership
Differentiate public trials, party and military purges, and mass operations while recording their connections.
Episode inclusion test
Require an explicit reason for connecting a trial, arrest wave or institutional purge to the modelled event.
- Which directive, institutional process or historical argument connects this episode to the Great Purge? provenance
- Is this episode a component campaign, a precursor, an aftermath or a separate repressive action? boundary
Authorisation and implementation Trace how political decisions became operational instructions and local practices.
Assessment of responsibility requires distinctions among authorisation, implementation, escalation and retrospective explanation.
Directives and command
Connect leadership decisions, NKVD orders and implementing authorities using documented relationships.
Documented authority chain
Record who issued, approved, transmitted and executed an instruction without inferring personal involvement solely from rank.
- Which records establish the roles of Stalin, party leadership, NKVD leadership and regional authorities in this decision? provenance
- Does the evidence establish authorisation, knowledge, implementation or a later attribution of responsibility? definition
Operational limits and local practice
Represent campaign-specific categories, numerical limits, revision requests and implementation differences.
Operation-specific controls
Examine mechanisms such as those associated with NKVD Order No. 00447 separately from procedures used in other operations.
- What categories, numerical limits and approval procedures did the applicable order prescribe, and how were they revised? measurement
- Which records distinguish central instruction from regional requests, local initiative or departures from prescribed procedure? provenance
Targeting, procedure and harm Follow how people were classified, accused, processed and affected.
The model must preserve the difference between perpetrators' claims, coercive case production and documented human outcomes.
Target selection and accusation
Record official targeting categories and accusations as attributed claims.
Classification versus established conduct
Distinguish labels such as enemy, conspirator or spy from independently supported conduct, including targeting through association or attributed background.
- Was the person selected through office, attributed class background, nationality, foreign ties, denunciation or association with another suspect? provenance
- What independent evidence, if any, supports the alleged conduct beyond an accusation or confession? provenance
Case processing and outcomes
Represent interrogation, adjudication and sanctions as a sequence with identifiable evidentiary gaps.
Coercion and disposition
Distinguish public trial narratives, tribunal decisions and extrajudicial processing while tracing outcomes for individuals and affected relatives.
- Which body processed the case, and what evidence establishes coercion, access to defence or the absence of a meaningful hearing? provenance
- Which outcomes are documented for the person and affected relatives, and which dates or dispositions remain unknown? measurement
Evidence, reckoning and interpretation Evaluate documentary claims, numerical estimates and later reassessments.
Records produced by repression, later investigations and historical research answer different questions and require distinct treatment.
Record criticism and counting
Assess source provenance and make population claims comparable.
Claim-specific source assessment
Evaluate what orders, case files, execution records, testimony and later compilations can establish for a particular claim.
- Who created this record, for what purpose, and is the consulted version an original, transcription, translation or later compilation? provenance
- What corroboration is required before an agent treats its contents as evidence of an event rather than evidence of an official allegation? action
Comparable victim estimates
Separate arrests, sentences, executions and deaths in custody, retaining scope and overlap information for every estimate.
- Does this figure count people, cases or decisions, and what dates, territories, campaigns and outcomes does it cover? measurement
- How are repeat arrests, revised sentences, overlapping categories and missing records handled before estimates are compared or combined? measurement
Rehabilitation and historical explanation
Track later case review and competing explanations without conflating legal findings with historical conclusions.
Later reassessment
Record the authority, date and scope of rehabilitation or other review decisions and their implications for earlier claims.
- Which authority reviewed the case, what decision did it make, and which original allegations or sanctions did that decision address? provenance
- Which claims should an agent revise after this decision, and which require independent historical evidence? action
Competing causal accounts
Represent attributed explanations involving leadership choices, institutional incentives, security fears and local dynamics as arguments to evaluate.
- Which historian or research tradition advances this explanation, and what evidence supports its proposed mechanism? provenance
- Does the explanation account for elite trials, mass operations and regional variation equally well, or only for a defined subset? boundary
Evidence and external alignment What the world already says about this thing, gathered so the model can be checked against it.
A model that cannot be lined up against existing standards, identifiers and practice cannot be adopted by anyone who already uses them.
Reported evidence
Findings from the breadth pass, kept separate from the structural claims.
Check these first
Recalled without web access and unsourced; every item is a lead to verify.
- This describes the Soviet historical campaign, not political purges in general.
- Period boundaries and the relationship between the labels Great Purge and Great Terror vary among historians.
- Any victim totals require checking the dates, territories, archival categories, and distinction between executions, imprisonment, and deaths in custody; none are supplied from recall.
- Which of these check these first hold for the sense of Great Purge this model covers, and on what evidence? provenance
Real-world use
Recalled without web access and unsourced; every item is a lead to verify.
- A historical case for studying dictatorship, state violence, coerced confessions, and the destruction of political and institutional independence.
- Archival investigation, rehabilitation proceedings, and commemoration of victims of Soviet repression.
- Which of these real-world use hold for the sense of Great Purge this model covers, and on what evidence? provenance
Failure modes and hazards
Recalled without web access and unsourced; every item is a lead to verify.
- Mass execution, imprisonment in the Gulag, torture, and coercion of confessions.
- Collective targeting through social classifications, political allegations, and perceived foreign or national affiliations.
- Family separation, punishment of relatives, and enduring stigma.
- Loss of experienced personnel and disruption of military, administrative, scientific, and cultural institutions.
- Historical accounts can mislead by treating elite show trials as the whole campaign or conflating its victims with those of all Stalin-era repression.
- Which of these failure modes and hazards hold for the sense of Great Purge this model covers, and on what evidence? provenance
Regional variation
Recalled without web access and unsourced; every item is a lead to verify.
- Repression extended across the Soviet Union, but its intensity, targeted populations, and implementation varied between republics and localities.
- NKVD national operations targeted categories associated with particular nationalities or foreign connections, including Poles and Germans.
- Regional authorities participated in proposing and implementing arrest and execution limits, producing locally different patterns of repression.
- Which of these regional variation hold for the sense of Great Purge this model covers, and on what evidence? provenance
Neighbouring kinds and how to tell them apart
Recalled without web access and unsourced; every item is a lead to verify.
- Great Terror - Often used synonymously with Great Purge, but can refer more specifically to the mass repression of 1937-1938; usage must be checked in each account.
- Stalinist repression - The broader category includes coercive campaigns and persecution before and after the Great Purge.
- Moscow Trials - The prominent staged trials of former Soviet leaders were components of the campaign, not its full extent.
- Gulag - The Soviet forced-labor camp system received many victims but existed before and after the campaign.
- Party purge - An organizational purge can mean expulsion or screening of party members; the Great Purge encompassed mass violence against people far beyond the party.
- Which of these neighbouring kinds and how to tell them apart hold for the sense of Great Purge this model covers, and on what evidence? provenance
What the second pass must settle
- Which researched definition and periodisation should anchor this registry entry, and how should alternative uses of Great Terror be mapped to it?
- Which component campaigns and peripheral episodes satisfy an explicit inclusion rule rather than merely overlapping chronologically?
- Which accessible primary records and critical editions can support operation-level claims about authorisation, numerical limits and regional implementation?
- Which victim estimates become comparable after aligning populations, outcomes, territory and dates, and what uncertainty remains from missing or overlapping records?
- Where do rehabilitation decisions, contemporary records and later scholarship conflict, and which disagreements must remain explicit?