Intelligence collection is organized by source type into distinct disciplines, each with its own methods, capabilities, limitations, and institutional culture. These are sometimes called “the INTs” — a shorthand for the suffix shared by each discipline’s acronym.
The disciplines
- HUMINT (Human Intelligence) — information from human sources. The oldest discipline. High value for accessing intent and context; slow, difficult to verify, vulnerable to deception and manipulation.
- SIGINT (Signals Intelligence) — information from intercepted communications and electronic emissions. Scalable and automatable; dependent on the adversary communicating electronically and on the collector’s ability to decrypt or interpret intercepts.
- IMINT (Imagery Intelligence) — information from visual and multispectral imagery. Provides physical evidence of ground conditions; requires expert interpretation and is vulnerable to camouflage, concealment, and deception.
- OSINT (Open-Source Intelligence) — information from publicly available sources. Cheap and abundant; the challenge is filtering signal from volume and evaluating reliability without institutional controls.
- MASINT (Measurement and Signature Intelligence) — information from physical signatures. Highly specific and difficult to fabricate; narrow in scope and technically demanding.
- GEOINT (Geospatial Intelligence) — integration of imagery with geospatial data. Provides the spatial framework within which other collection is situated.
Multi-INT fusion
No single collection discipline provides a complete picture. Each discipline reveals one dimension of the adversary’s activity while remaining blind to others. SIGINT intercepts a communication but cannot confirm whether the message was deception. IMINT photographs a military formation but cannot explain its purpose. HUMINT reports intent but cannot verify capability. Effective intelligence work fuses collection from multiple disciplines to cross-validate findings, fill gaps, and resist adversary denial and deception.
Multi-INT fusion is an analytic practice, not an automatic process. It requires analysts who understand the strengths and limitations of each discipline, can weight evidence accordingly, and can recognize when disciplines produce contradictory results — contradiction that may indicate either analytic error or adversary deception.
Institutional dynamics
Each collection discipline has developed its own institutional culture, budgets, training pipelines, and organizational identity. HUMINT officers are trained differently from SIGINT analysts; IMINT interpreters work in different organizations from MASINT technicians. These institutional divisions create both specialization and stovepiping — the tendency for disciplines to collect and analyze within their own channels without adequate integration.
The history of intelligence failures is in part a history of stovepiping failures: information available in one discipline that was not shared with analysts in another, connections that were visible across disciplines but invisible within any single one. The intelligence cycle’s analytic phase is where multi-INT fusion is supposed to occur, but institutional and classification barriers often prevent it.
Relation to the Stasi model
The Stasi’s analogue surveillance apparatus, documented in Pattern Before Person, operated without the formal INTs framework but practiced a functional equivalent: postal interception (analogous to SIGINT on communications), physical surveillance (analogous to IMINT), informant networks (analogous to HUMINT), forensic analysis (analogous to MASINT), and cross-departmental correlation (analogous to multi-INT fusion). The Stasi’s blob-first epistemology — constructing patterns from heterogeneous fragments before identifying individuals — represents a model of multi-source fusion that modern computational surveillance systems have formalized and scaled.