Morgenstern’s unfinished RAND memorandum is a methodological clearing of the ground for a future quantitative theory of organization. Its central claim is that economics, administration, and sociology cannot yet formalize organization because they lack an adequate descriptive language for it. The firm, the army, the university, and the logistics system cannot be understood as mere aggregates of inputs, since their capacities depend on arrangements of authority, information, timing, discretion, and control.
Work in the field of the theory of organization will for a long time have to be essentially phenomenological.
This phenomenological starting point is not anti-mathematical. Morgenstern wants concepts precise enough to become mathematical later, but he resists premature axiomatization. Game theory offers a partial model, especially for relations among organizations, yet it does not explain the internal order of a centrally directed body. Economics treats the firm largely as a black box; Morgenstern makes its internal organization part of what determines costs, feasible strategies, and outcomes.
The memorandum’s constructive vocabulary begins with centrally directed social organizations. Morgenstern distinguishes “inner” activity, by which the organization arranges persons, materials, machines, and offices, from “outer” activity, by which it acts upon nature or other organizations. The basic unit is the operation: not a static task but a realized sequence selected from possible activities.
an operation consists of acts, which are choices.
Organization therefore becomes a distribution of choice. Instead of organic metaphors, Morgenstern introduces “competences”: units authorized to choose values of assigned variables within specified domains. These competences originate in delegation from the highest authority, the Source, but delegation is not reducible to a simple hierarchy. A subordinate may choose because the superior lacks time, information, skill, or position; in certain sequences, a guard or pilot may exercise temporary authority over nominal superiors. Thus the organization chart is inadequate unless supplemented by the rules and signals that determine when competences actually act.
Every organization relies on at least one signaling system; but there can be many.
The discussions of signaling, information, and memory give the memorandum its proto-cybernetic character. Signals activate competences and may be internal or external, intentional or accidental, clear or noisy. Since signals connect variables to decision points, the same formal hierarchy can produce different operations under different communication systems. Memory is likewise constitutive: routines, drills, files, procedures, and “protocol memory” make continuity and learning possible. Learning, however, requires some way to identify error, and error depends on aim. Morgenstern therefore distinguishes tolerable deviations from structural mistakes: a commercial stockout may be acceptable, while a missing irreplaceable naval supply may be disastrous.
Control appears as the organizational answer to disorder. Passive control compares duty and performance after the fact; active control narrows the possible range of a variable or converts it into a constant. Morgenstern’s point is not simply managerial. Organization is fragile because it must continually preserve form against entropy, drift, ignorance, and miscommunication.
Organization in itself, wherever it occurs, whether in social or natural life, is something exceptional and extraordinary.
The later sections return this framework to familiar economic categories and show their insufficiency. Input-output tables, cost-profit analysis, and ordinary accounts cannot capture the strategic and structural differences between organizations using similar resources. The difficulty is greater for universities, hospitals, churches, prisons, and armies, whose aims are hard to define and harder to quantify. Size and growth are equally resistant to simple treatment: expansion may intensify an existing structure, but it may also require mutation into another organizational type.
The memorandum anticipates information theory, operations research, bounded rationality, cybernetics, and organizational design. Morgenstern treats organization neither as a production function nor as a chart of reporting lines, but as a system of delegated variables, constrained choices, signals, memories, controls, errors, and adaptive transformations.
This work was divided into 18 sections when it entered the library's research corpus—an apparatus for search and citation, not necessarily the author's own table of contents. Each title opens its summary.
Put a question to this work; the Librarian answers from its 18 sections and cites the passage.
Ask the Librarian