Islamic State administrative apparatus in Iraq and Syria

Administration and violence were one apparatus

The organization calling itself Islamic State did not administer empty territory. It seized public facilities, oil and other assets, professional routines, workers, records, and service systems that Iraqi and Syrian communities and governments had already built. Its central bodies and provincial offices joined military, security, finance, religious enforcement, courts, and service administration. UNITAD's financial investigation identifies the Delegated Committee and diwans for treasury, alms, soldiery, and resources as parts of a top-down apparatus that linked policy and money to military operations and international crimes.¹

That structure does not make the organization a lawful state or make residents' survival under its control political acceptance. An office could distribute food, record a payment, or keep a utility operating while the larger system confiscated property, punished dissent, financed warfare, and classified people for persecution. Administrative capacity and atrocity therefore cannot be balanced as separate sides of a ledger.

Hierarchy depended on captured capability and unequal kinds of membership

Daniel Milton's study of captured personnel files reconstructs a central hierarchy, provinces, and governance offices from two large payroll sheets and related memoranda. In a late-2016 snapshot weighted toward Iraq, he identified more than 60,000 unique identification numbers and roughly 166,000 listed dependents. Of records that could be assigned a function, about 18 percent were in governance rather than military roles. The figures describe the captured files, not a census of everyone who ever worked, fought, died, defected, or received support across Iraq and Syria.²

The source boundary matters. Milton could not disclose every provenance detail; the data contain anomalies, unclear geographic coverage, possible miscoding, and people whose status had changed. A propaganda video naming fourteen ministries shows the chart the organization projected, not which offices existed everywhere or performed as advertised. Cross-checking captured files against other collections improves confidence in particular records, but the files remain situated primary evidence: they establish what an apparent record prescribed or recorded, not independent truth, public consent, or lawful authority.³

Civilian employees were not one category. Matthew Bamber's interviewees distinguished pledged members from non-pledged civilian supporters and workers. Pledged members generally held more authority, pay, and protection; non-pledged employees could be recruited for inherited technical skill, required to attend ideological courses, pressured to pledge, or threatened when trying to leave. Specialist workers sometimes retained more bargaining room than replaceable workers. Coercion constrained agency without erasing every individual choice, which is why a job title alone cannot settle postwar culpability.⁴

This dependence limits claims that the apparatus created a complete administration. Doctors, teachers, engineers, clerks, utility workers, and other employees carried knowledge embedded in earlier institutions. Supervisors, security bodies, curricula, and payroll rules redirected that capability. The organization could therefore claim an inherited service as its achievement while the people performing it faced sharply unequal choices.

Taxation made people and property legible to both rule and war

Mara Revkin's study maps seven revenue-extracting policies in nineteen Syrian districts from 2013 through 2017 and uses al-Mayadin as an oil-rich case. It finds that tax-like policies were as prevalent in resource-rich as in resource-poor districts and argues that ideology and the costs of warfare help explain the pattern. The evidence shows that oil wealth did not displace administrative extraction; it does not make the payments voluntary, lawful, or representative.⁵

UNITAD separately traces treasury, alms, military, resource, and Delegated Committee structures through witnesses, Iraqi judicial material, transaction data, financial experts, and case files. Its public factsheet reports that treasury revenues included oil exploitation, confiscation, taxation, and pillage, and that central authorities used the apparatus to finance military operations and persecution. The factsheet summarizes investigations and case assessments shared with Iraqi courts; it is not the underlying transaction ledger or a public criminal judgment.¹

These records make several distributions visible. Members and dependents received organizational pay and support. Merchants, farms, transporters, money exchanges, and property holders became objects of fees, surveillance, confiscation, or control. Residents could receive an output from an office while the same revenue system sustained the armed conditions under which refusal and exit were dangerous.

Genocide was classified, transported, registered, and distributed

UNITAD's public Sinjar assessment applies a reasonable-grounds standard to survivor, witness, expert, and alleged-perpetrator testimony; documentary, forensic, audiovisual, telecommunications, and open-source material; and Iraqi judicial evidence. It redacts names and details for safety and privacy. The Independent International Commission of Inquiry on Syria reached a compatible genocide conclusion through forty-five interviews and corroborating documentary material, while expressly limiting its mandate to violations in Syria and using Iraqi events to establish context. These are official investigative and legal findings, not convictions of every alleged participant.⁶

For Kocho on August 15, 2014, UNITAD estimates that ISIL forces executed about 422 men and older boys and that eighteen survived. About 1,250 civilians had remained in the village before the separation. Fighters gathered residents at a school, separated men and older boys from women and younger children, took valuables, and moved the groups to different sites. Those numbers are the assessment's reconstruction from available evidence, not a complete village census or final identification of every victim.⁷

The apparatus continued after capture. UNITAD documents registration, photography, transfer, selection, lotteries, formal and informal slave markets, sales and gifts, and court-issued property contracts or proofs of ownership for Yazidi women and girls. Administrative form did not reduce the violence; it made captivity, sexual enslavement, and reassignment repeatable across places and personnel.⁸

Peer-reviewed research by Revkin and Elisabeth Wood helps separate policy from practice. Drawing on organizational documents, social media, interviews, and secondary sources, they find policies authorizing sexual slavery and child marriage, alongside forms of sexual violence that violated stated policy but were tolerated by commanders. Their interview evidence is purposive and snowball-sampled, and their internal documents record rules, receipts, and claims. The study supports an institutional explanation of patterned violence without implying that written policy perfectly predicted every act.⁹

A final German Federal Court of Justice judgment gives a narrower kind of evidence. In one defendant's case, the court upheld findings concerning her membership and her treatment, with her husband, of two Yazidi captives as domestic slaves, including forced prayer, renaming, and abuse. The judgment can establish adjudicated facts and liability for that defendant; it should not be expanded into a judgment against every employee or member.¹⁰

Children were not incidental to the design. UNITAD records forced Arabic language, prayer, renaming, separation into purported orphanages, conversion, enslavement, and conscription or military training. The Commission of Inquiry also documents forcible transfer of Yazidi women and children into Syria and the training of boys. Neither report measures every child's later health, education, family reunification, or recovery.¹¹

Use of a court or service did not establish free consent

People under armed territorial control still needed water, health care, bread, schooling, documents, and dispute resolution. Mara Revkin's short peer-reviewed account draws on a household survey of 1,458 Mosul residents, more than 200 interviews, and over 1,500 primary documents. It argues that legal institutions helped the organization initially attract cooperation in some settings, while procedural injustice, corruption, discrimination, extraction, and arbitrary violence contributed to failure. The proceedings article is a compact synthesis of a larger research program, not a full publication of every instrument or subgroup estimate.¹²

The distinction is essential. Using the only available court, accepting food, or continuing work needed to keep a city alive can reflect necessity, fear, calculation, partial approval, or some combination. No single observed act proves allegiance. Conversely, coercion does not mean every worker had identical agency or that every decision is exempt from individual examination. Evidence must remain role-specific, time-specific, and connected to conduct.

Internal records also reveal purpose without proving performance. Gina Vale's analysis of twenty-four internal documents shows prescriptions for behavioral codes, education, and forced conversion and interprets them as a strategy of civilian control. Those documents are authoritative for what the organization ordered or presented as piety; they are not independent observations of compliance, acceptance, or service quality.¹³

Contraction produced adaptation, not an administrative moral conversion

Matthew Bamber-Zryd reconstructs repeated governance cycles from more than 1,000 provincial documents and 117 interviews with members, civilian workers, and residents. He identifies phases of insurgency, territorial acquisition, institution-building, and territorial loss and interprets captured lessons-learned material and later changes as organizational reflection and adaptation. The fieldwork spans Iraq, Syria, Lebanon, and Turkey but is not a representative survey; the cyclical model is the author's historical interpretation.¹⁴

That evidence supports modest profile claims about doctrinal revision, after-action review, central reconfiguration, local iteration, and crisis mobilization. It does not show reliable learning toward public welfare. As territory and resources contracted, offices and priorities changed inside an armed organization whose security, extraction, and ideological commitments remained constitutive.

Environmental damage and public rebuilding exceed the office chart

UN Environment's rapid scoping note for areas retaken around Mosul identifies oil wells deliberately set on fire near Qayyarah and describes pollution from conflict, deliberate sabotage, and looting of industrial and civil infrastructure as a health, livelihood, reconstruction, and peacebuilding burden that can persist for decades. It is an official rapid assessment, not a complete ecological inventory or a causal allocation of every damaged site among ISIL, Iraqi forces, coalition forces, and other armed participants.¹⁵

The ecosystem record is therefore a documented burden with an attribution boundary. The located sources do not provide a comparable audit of animals killed or displaced, livestock and agricultural use, wildlife exposure, or animal suffering during occupation, siege, flight, bombardment, and cleanup. Nonhuman-life direction remains unclear because the research is missing, not because the impact was absent.

Records support accountability only when custody and privacy remain visible

Payrolls, tax records, court instruments, property documents, captive lists, and personnel files can corroborate crimes, locate responsibility, support property claims, or help families search for missing people. They can also expose survivors, coerced workers, relatives, and witnesses. UNITAD's use of redaction, corroboration, and a stated evidentiary threshold shows one public investigative approach; it does not eliminate the need for survivor access, Iraqi legal authority, chain of custody, data protection, and contestable interpretation.

The public ISIS Files repository makes the custody dispute concrete. George Washington University says it released a selection from roughly 15,000 digitized pages collected in Iraq by a New York Times team, returned the originals to the Iraqi government after digitization, consulted widely, and redacted personally identifying information. The Society of American Archivists had objected to the documents' removal and planned dissemination, citing Iraqi documentary heritage, privacy, human rights, and the risks of dispersal. The later return and redaction are relevant responses; they do not make the earlier disagreement disappear.¹⁶

A 2024 statement organized by Yazda and forty-seven other Yazidi-led civil society organizations locates authority differently. It calls government, international, and local actors to commemorate in Sinjar, attend to insecurity, and listen to a unified Yazidi voice amid continued displacement and barriers to return. It is authoritative for the coalition's expressed priorities, not an independent prevalence survey of all Yazidis or a substitute for individual survivors' differing views.¹⁷

Comparisons identify mechanisms, not equivalence or ancestry

The Rwanda genocide local administrative machinery is an institution comparison about classification, local reach, and administrative participation in genocide. Democratic Kampuchea is an organizational-case comparison about ideological command, forced labor, and lethal classification. The Great Leap Forward reporting system is an organizational-case comparison about distorted information, coercive hierarchy, and the human cost of suppressed correction. None asserts equivalent victims, law, ideology, scale, sovereignty, or historical origin.

The Prussian-German General Staff is an institution comparison limited to hierarchy, specialist staff work, records, and adaptation under military pressure. It does not assert moral equivalence, influence, shared lineage, or transferable performance. Authority, legitimacy, and acceptance and governance, stewardship, and accountability are idea relations used to keep coercive capacity separate from accepted authority and to ask who can inspect, challenge, repair, and judge administrative action. Historical participants did not use these Atlas categories.¹⁸

The comparative fingerprint makes the emphasis inspectable

Score 3 marks authority, legitimacy, and acceptance, coordination, communication, and common understanding, structure, hierarchy, and scale, strategy, competition, and adaptation, and governance, stewardship, and accountability, together with executive attention, information, and organizational sensing. These are defining because the apparatus joined coercive authority, hierarchy, provincial coordination, strategic contraction, and vertical information to governance without public accountability.

Score 2 marks purpose, mission, and institutional legitimacy, delegation, decentralization, and responsibility, decision-making, judgment, and bounded rationality, measurement, accounting, and control, cooperation, incentives, and organizational equilibrium, work design, productivity, and automation. These lenses explain ideological purpose, provincial discretion, judgment, payroll and registration, differentiated membership, and specialized work without replacing the entry's principal authority-and-governance comparison.

Score 1 marks knowledge, expertise, and professional autonomy, learning, quality, and reliability, culture, informal organization, trust, and voice. Expertise and adaptation are documented, but professional autonomy, reliable learning, and protected voice were sharply bounded by ideology and coercion.

Score 0 leaves innovation, entrepreneurship, and renewal outside the organizing comparison, together with organizational ignorance. Zero does not mean that no local change or information failure occurred. It means the located evidence is better explained by seizure, recombination, coercion, learning, adaptation, and direct suppression than by innovation or organizational ignorance as defining lenses.¹⁸

The profile is an editorial synthesis. Authority combines military-security, religion-theology, and state-bureaucracy; decisions span central-executive, divisional, rule-bound, and frontline loci. The historical-polity ownership label marks a comparative territorial form and does not recognize sovereignty or lawful title. Hierarchy, rule and ritual, doctrine, training, and metrics coordinate work; knowledge moves top-down, bottom-up, through specialist staff, and through embedded practice. Financial, operational, and behavioral measurement fit the documented records.

Doctrinal revision and after-action review describe the bounded evidence of learning; central reconfiguration, local iteration, and crisis mobilization describe adaptation. Members are the sole named beneficiary group because the payroll and worker evidence supports an internal distribution to members and dependents. That label is descriptive, not approval. Capture, leader-dependence, suppressed voice, financial extraction, externalized harm, and fragility summarize recurring risks visible in the cited record, not quantitative frequencies.¹⁸

Benefits and burdens cannot be collapsed into service delivery

Workers received income in some roles and sometimes exercised specialist leverage, while coercion, unequal membership, ideological supervision, threats, and postwar guilt-by-employment created burdens. Customers and users received or used necessary services under armed rule; that use cannot establish free consent. Suppliers and purported partners were subject to revenue and control systems whose sector-by-sector effects remain unmeasured.⁴¹²⁵

For owners and investors, the direction is burden because the relevant subject is the people, communities, firms, and public bodies dispossessed—not the armed organization's asserted title. For members, the record supports a bounded benefit: pay, dependent support, office, and relative privilege. It does not measure every member's net welfare or excuse individual conduct.¹²

For communities, the documented direction is burden, centrally through genocide against Yazidis. Public institutions carried captured work and later rebuilding and contamination costs. Mission-beneficiaries is used here as a repair lens for survivors, missing families, displaced people, and communities seeking safe return and justice, not as the armed organization's category.⁶¹⁷¹⁵

For ecosystems, deliberate oil-well fires and the wider conflict left documented pollution and reconstruction burdens, with incomplete causal allocation. For nonhuman life, direction remains unclear because no systematic audit was located. For future generations, forced transfer, indoctrination, recruitment, family separation, damaged infrastructure, and contamination are documented mechanisms of burden, but their full duration and distribution have not been measured.¹¹¹⁵¹⁹

Paths into deeper study

• Build survivor- and community-governed access rules for captive, property, detention, personnel, and missing-person records, with privacy, correction, chain-of-custody, repatriation, and remedy procedures.
• Compare payroll, worker testimony, public-institution records, and individual conduct to distinguish coercion, survival work, material support, command, and direct participation without guilt by job category.
• Publish disaggregated evidence on courts, utilities, health, education, food, agriculture, and local administration that separates inherited capacity, worker contribution, organizational prescription, actual performance, and civilian consent.
• Extend accountability research beyond the Yazidi case without flattening distinct campaigns, legal findings, affected groups, survivor priorities, and evidentiary standards.
• Conduct contaminated-site, ecosystem, livestock, wildlife, and animal-welfare assessments that distinguish occupation, scorched-earth tactics, siege, bombardment, looting, cleanup, and reconstruction.
• Trace intergenerational effects through family reunification, missing-person resolution, education, health, land and property restoration, cultural continuity, environmental exposure, and institutional trust.