¹ My attention was drawn to the comparison of money and power by a paper of Parsons, Talcott, “On the Concept of Influence,” Public Opinion Quarterly, 27 (Spring 1963), 37–62 CrossRefGoogle Scholar. See also my detailed critique of this paper, and Parsons' reply, Coleman, James S., “Comment on ‘On the Concept of Influence’”, Public Opinion Quarterly, 27 (Spring 1963), 63–82 CrossRefGoogle Scholar, and Parsons, ibid., pp. 87–92. I am greatly indebted to Parsons for this initial stimulation.

² This is not to say that chronologically this development occurred late in the history of economic institutions, nor that in every society it followed the development of commodity money. In barter systems, the absence of a monetary unit has often led to a system of widespread obligations or debts, these obligations serving in place of a system of money, to allow nonsimultaneous exchanges to take place. There seem to be no good examples in primitive monetary systems of such obligations being systematically transferred and used as legal tender. (See, for example, Einzig, Paul, Primitive Money, 2nd edition, London, 1966, pp. 365–366 Google Scholar.) However, Hawtrey describes the possibility of such a system, where in a barter economy, a medium of exchange is made unnecessary by a balancing off of debts and credits by the parties in the market. (See Hawtrey, R. G., Currency and Credit, 3rd edition, London, 1928.Google Scholar) While this is not so extensive a system of transfer of the obligation from a creditor to a third party as the institution of legal tender, it does represent a transfer of the obligation to a third party, if only for balancing accounts.

³ There is one important historical example in which such chains of obligation did constitute a form of money. This was in the early industrial activity of Lancashire in England. (For an account of the use of bills of exchange in Lancashire, see Ashton, T. S., “The Bill of Exchange and Private Banks in Lancashire, 1790–1830,” Economic History Review, Vol. 15, Nos. 1, 2, 1945 CrossRefGoogle Scholar, reprinted in Papers in English Monetary History, edited by Ashton, T. S. and Sayers, R. S., Oxford University Press, London, 1953.Google Scholar) While in most areas of England in the late 18th century, the notes of merchant bankers or goldsmiths became accepted currency, in the Lancashire industrial area, these bank notes were less freely accepted (and then only with a discount) than bills of exchange. Bills of exchange were debts incurred by one manufacturer upon another. The system that developed was one in which the creditor himself would endorse the bill and use it to pay one of his creditors. The use of bills to settle debts had been (and still is) widespread in large commercial transactions, but generally only in order to raise cash in a time of need, and generally at a discount. In Lancashire they were used widely by manufacturers for small transactions, apparently for nearly all transactions other than payment of workers (which was done in coin), and with little or no discount, though they were dated for payment some months in the future. Neither the Bank of England notes nor the notes of country banks or merchants were accepted as freely as the bills of exchange. A bill of exchange would pass through many hands, being endorsed and guaranteed by each holder in turn. The confidence of manufacturers and merchants in industrially dense Lancashire was apparently greater in one another collectively than in any one merchant who undertook to specialize as a banker by issuing notes of his own.

⁴ See Unwin, G., Samuel Oldknow and the Arkwrights (Manchester, 1924), Chap. 12Google Scholar.

⁵ Einzig, op. cit., p. 38.

⁶ See Adcock, F. E., Roman Political Ideas and Practice (Ann Arbor: University of Michigan, Ann Arbor Paperbacks, 1964)Google Scholar.

⁷ Political power in modern society is not, however, associated with a particular person. This separation of power from the particular person and blood lines is a major achievement in the rationalization of society, and is one of the crucial characteristics of bureaucratic authority.

⁸ When an occupant of a position threatens to resign if he fails to get his way on a given issue, he could do so only a few times at most. This is use of personal power, rather than power associated with the role. Only if a man has filled an office in such a way that his personal occupancy of the position is of value to others does this threat have any power.

⁹ This does not exactly correspond to Samuelson's definition of a public good, which involves both the jointness of supply and the non-additivity of consumption. However, the aim here is not to say that all political power concerns control of the supply of public goods, but rather to give a rough description of the kind of actions which political power controls.

¹⁰ It is Bentley's, Arthur E. Process of Government (1908)Google Scholar that was most pioneering in this direction, and remains the most theoretically elegant, although it was formulated very loosely. Bentley had a conception of society as composed of activities, and of interest groups forming to implement these activities, with the outcomes of governmental actions a resultant of the pressures of these interest groups that were founded on activities. Bentley was never specific about what he included as activities, but it was clear that he intended them to include, but not be limited to, economic activities. In one example he uses, church-going activity is opposed to saloon-keeping activity (with other activities participating as well) over the issue of saloon operation on Sunday. Bentley, however, did not take the next step and examine the possibility that these activities could be the recipient of citizens' votes, and thus represent them in a legislative body. He pays little attention altogether to the form of government, making the implicit assumption that the pressure of the activities through their interest groups would be independent of the governmental form.

¹¹ A Colonel Blotto game is a game in which two opposing players, each with a set of forces, must deploy these forces among several sites. The player with the greatest size of force at a given site wins on that site, and the overall payoff to each player is determined by the sum of his gains at each victorious site. This is obviously a game of pure conflict. See Luce, R. D. and Raiffa, Howard, Games and Decisions (New York: John Wiley and Sons, Inc., 1957)Google Scholar.