This short theoretical article, originally published in 1973 and reprinted in 1976, proposes a research programme for game theory by reconsidering the relation between strategies and resources. Its two sections move from the material conditions of strategic choice to the interconnected games through which individuals conduct economic life. Morgenstern’s central move is to treat available strategies as dependent on resource allocation, rather than as a fixed repertoire within which optimization takes place.
Section I distinguishes “neutral” games from “power” games:
The neutral games are those where the wealth or the resources of the participants have no significance in regard to the number and kind of strategies available to each player.
Chess illustrates this neutrality: the physical means of executing permissible moves are already supplied. Stakes can make wealth relevant, especially through repeated play and possible ruin, but do not themselves define the repertoire of moves. In economic and political power games, by contrast, resources determine which strategies can be made available. Morgenstern initially assumes homogeneous resources to isolate the problem of allocating them among strategic possibilities.
Possessing a “strategy” means that certain resources are used in execution and playing of the game.
This material interpretation turns the payoff matrix into a guide to investment. In a strictly determined game, players need equip themselves only for the strategies guaranteeing the saddle-point payoff; with probabilistic choice, investment concerns the strategies actually eligible for selection. Construction may follow the choice, but the capacity to implement the selected strategy must already exist. Strategic availability thus becomes a question of feasible commitment as well as formal permission.
The distinction also changes the analysis of repeated play. A payoff retained as resources can enlarge the winner’s strategy space and alter the next game’s payoff matrix. Repetition then transforms the conditions of play instead of merely reproducing them.
The repetition of this process cannot be treated by the current theory of the supergame; a far more complicated problem emerges.
Coalitions introduce a parallel difficulty. Describing a coalition as controlling its members’ combined strategies is formally correct, yet may fail to capture its effective power. Strategies can reinforce one another, become redundant, or conflict. Their aggregation therefore requires an account of their practical compatibility and complementarity. Morgenstern presents these as problems awaiting precise formulation, not as results of a completed theory.
Section II expands the allocation problem from strategies within one game to an individual’s simultaneous participation in multiple games—buying, earning, investing, and interacting under different market structures.
We call the entirety of these games the “integral game” for the reason that they all are power games and have to be viewed as one game.
The unity of the integral game comes from the resources allocated across its constituent activities. Their payoff structures depend on the individual’s allocations and on those of other players, which may be unknown. Even the identities and relationships of participants cross game boundaries: two people can be antagonists in a two-person zero-sum game while cooperating in another game. Morgenstern consequently recommends restrictive initial assumptions, including complete information where possible and simplified relations among participants, as steps toward a tractable account.
Different durations add a temporal dimension. When one game ends, resources may become available for another. Such reallocations suggest a conception of social dynamics grounded in strategic interdependence rather than simply inherited from mechanics. The article’s relevance lies in making resource commitments, overlapping participation, and changing strategic capacities central to the modelling of society.
The conclusion gives this programme its sharpest economic implication:
The optimal use of resources is thus never independent from other players.
Morgenstern argues that marginal economics cannot adequately approach allocation even within a single power game, much less across an integral game. He confines marginal analysis and optimization to situations without interference from other agents. The article thus challenges the sufficiency of isolated optimization while outlining why a strategic alternative remains both necessary and exceptionally difficult.
This work was divided into 3 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 3 sections and cites the passage.
Ask the Librarian