¹ Almond, Gabriel A., “Comparative Political Systems,” Journal of Politics, Vol. 18 (1956), p. 392.CrossRefGoogle Scholar

² E.g.: Y has power over X's doing x; with respect to Y, X is free to do x. Here I shall be concerned only with freedom in the interpersonal sense, which is closely connected with the key concept of power. Interpersonal freedom, must be distinguished from freedom of choice, the feeling of freedom, free actions, and of course from the many normative meanings this word tends to assume.

³ Some examples of other types of comparative statements in this area: Y₁ has more power over X's doing x than has Y₂. (E.g., parents have greater power over their children's attending Sunday School than the Church, and children have greater freedom to stay away from Sunday School with respect to the latter than with respect to the former.) Y has greater power over X₁'s doing x than over X₂'s performing the same kind of action. With respect to Y, X is more free to do x ₂ than to do x ₂.

⁴ Some recent attempts to explicate the concept of degree of influence or power are: Simon, Herbert A., “Notes on the Observation and Measurement of Political Power,” Journal of Politics, Vol. 15 (1953), pp. 500–516 CrossRefGoogle Scholar; March, James G., “An Introduction to the Theory and Measurement of Influence,” this Review, Vol. 59 (1955), pp. 431–451 Google Scholar and “Measurement Concepts in the Theory of Influence,” Journal of Politics, Vol. 19 (1957), pp. 202–226; Dahl, Robert A., “The Concept of Power,” Behavioral Science, Vol. 2 (1957), pp. 201–215.CrossRefGoogle Scholar However, these authors do not con sider all of the components which seem to me important. Furthermore, they are concerned only with the degree of influence or power, and I shall extend the analysis to the concepts of degree of interpersonal unfreedom and freedom.

⁵ Influence (e.g., persuasion, with or without threat of sanction) is another instance of power, but not of unfreedom. If Y dissuades X from doing x, Y has power over X's not doing x, but X remains, with respect to Y, free to do x. A complete definition of “Y has power over X's doing x” would be as follows: Y either influences X to do x or makes it necessary for him to do so or punishable to act otherwise. Concerning the relationships between influence and power (or control), cf. Oppenheim, Felix E., “An Analysis of Political Control: Actual and Potential,” Journal of Politics, Vol. 20 (1958), pp. 515–534.CrossRefGoogle Scholar

⁶ Dahl holds that this can be done only if the actors are “power comparable,” loc. cit., p. 208. However, we need not make this restriction, once we have made the other one, namely that we consider only the component of probability of being prevented or penalized.

⁷ Actually, interpersonal freedom is not the opposite of interpersonal unfreedom. With respect to Y, X is free to do x if and only if X is, with respect to Y, neither unfree to do x nor unfree to abstain from doing x. However, the second condition is often trivially fulfilled, and we may therefore disregard this clause in the present context.

⁸ Cf. Dahl, loc. cit. p. 203. Goldhamer, Herbert and Shils, Edward E. claim that “the amount of power exercised by an individual may be measured ... by the ratio of successful power acts to all his attempted power acts.” “Types of Power and Status,” American Journal of Sociology, Vol. 45 (1939), p. 176.Google Scholar By this test, someone who attempts little but is always successful would be more powerful than someone who achieves just as much but tries for more. Power is by definition successful; if Y makes an unsuccessful attempt at preventing X from doing x (or constraining him to do so), no power relation obtains between Y, X and the latter's action x. However, Y's unsuccessful endeavor may generate future power relations between himself and X.

⁹ However, the following two problems in utility theory are still controversial: (1) Is it possible to attach a rigorous meaning to interpersonal comparisons of utility (e.g., to the assertion that a miser is more deprived by a parking rule than a spendthrift)? Cf. Robbins, Lionel, An Essay on the Nature and Significance of Economic Science, 2d ed. (London, 1940), ch. 6Google Scholar; Little, I. M. D., A Critique of Welfare Economics, 2d ed. (Oxford, 1957), ch. 4.Google Scholar (2) Is it possible to attach numerical values to utilities, and to add or subtract them? Cf. Hicks, J. R., Value and Capital (Oxford, 1939), ch. 1Google Scholar; Neumann, J. von and Morgenstern, O., Theory of Games and Economic Behavior, 3d ed. (Princeton, 1953), ch. 1, pt. 3Google Scholar; Robertson, D. H., “Utility and All What?,” The Economic Journal, Vol. 64 (1954), pp. 665–678.CrossRefGoogle Scholar

¹⁰ Merton, Robert K., Social Theory and Social Structure, rev. ed. (Glencoe, Ill., 1957), p. 414.Google Scholar

¹¹ DeGré, Gerard, “Freedom and Social Structure,” American Sociological Review, Vol. 11 (1946), p. 529 CrossRefGoogle Scholar; cf. also Fuller, Lon L., “Freedom—A Suggested Analysis,” Harvard Law Review, Vol. 68 (1955), p. 1317 CrossRefGoogle Scholar: “… the widest range of possible choice—that is freedom.”

¹² Cf. supra, note 6.

¹³ This term is used by Lasswell, H. D. and Kaplan, A., Power and Society, (New Haven, 1950), p. 73.Google Scholar

¹⁴ Cf. Dahl, loc. cit. p. 207.

¹⁵ Even if “significance” could be denned operationally, we would still be confronted with problems of the following kind: Suppose that Y₁ is just a little more powerful than Y₂ in the most significant component, but a great deal less powerful in all other components, would we nevertheless consider Y₁ to have more total power than Y₂?

¹⁶ Indeed, the power of the Governor extends only to part of that range of the citizen's activities over which the Federal Government has no jurisdiction. As judged by the probability of sanctions and by the degree of deprivation, there is no significant difference between the power of the two officials.

¹⁷ In cases in which probability of prevention or punishment and degree of deprivation are the significant components, it might be possible to combine these two variables into a single one which describes the degree of an actor's total power or freedom in different circumstances. Indeed, economists, following Von Neumann and Morgenstern, have shown that for persons whose psychological make-up can be described by a standard set of axioms, utility and probability numbers can be combined into a single figure which predicts their preferences correctly. If deprivation itself could be assigned a utility number, this device might be useful in reducing the two variables of probability and deprivation down to one. However, the relevance of the Von Neumann-Morgenstern utility axioms to the political concepts and problems under discussion remains to be explored.

¹⁸ Cf. Truman, David B., The Governmental Process (New York: Alfred A. Knopf, 1955), p. 191 Google Scholar: “… leadership can be measured in quantitative terms by a ratio between the frequency with which an individual initiates actions in a group situation and the frequency with which he responds to the initiative of others.”

¹⁹ In the case of McCarthy this class consisted at one time of all other Senators, including of course Senator Tydings.

²⁰ Neumann, Franz L., The Democratic and the Authoritarian State (Glencoe, Ill., 1957), p. 76.Google Scholar

²¹ It is therefore meaningless to claim that “any interference with individual freedom … even by the most ideally democratic government, is to be justified only on the ground that it increases total freedom, immediately or in the long run.” Knight, Frank, Freedom and Reform (New York, 1947), p. 192.Google Scholar

²² If it were possible to count them, we would perhaps come to the opposite conclusion, at least in certain areas. There are probably “more” income tax and traffic regulations in the United States than in Soviet Russia.