¹ Almond, and Coleman, James S., eds. (Princeton 1960).Google Scholar

² P. viii: “… The functional theory presented in the first part of the book has not been fully applied in the individual area chapters.”

³ It may be that Almond is taking for granted the existence of a body of structuralfunctional theory widely accepted by students of the social sciences. But as a matter of fact such a theoretical corpus does not exist; see Merton, Robert K., Social Theory and Social Structure, rev. ed. (Glencoe 1957), 22–24Google Scholar; and Martindale, Don, The Nature and Types of Sociological Theory (New York 1960, London 1961), 443–44.Google Scholar

⁴ P. 17. These are the functional requisites of the political system, but it appears (p. 16) that they were derived from Western political systems. Whether or not they are held to be necessary to any actual political system is not clear, but presumably this would be a matter of investigation. The concept of systemic needs, however, is a very difficult one. See Rex, John, Key Problems of Sociological Theory (London 1961), 76.CrossRefGoogle Scholar

⁵ This point has also been noticed by Binder, Leonard, in Iran (Berkeley 1962), 7.Google Scholar

⁶ See Braithwaite, R. B., Scientific Explanation (New York 1960), 12–21.Google Scholar This is very similar to Mill's “inverse deduction.”

⁷ This will be shown in the body of this article, but in anticipation it may be remarked that (1) since there is no deductive system nothing is proved as opposed to asserted, (2) nothing can strictly be deduced unless terms are clear, and (3) tautologies are not amenable to disproof.

⁸ That is, Kingsley Davis seems mistaken when he suggests that the very diversity of functional analysis means we are all functionalists. Possibly the lack of analytical vigor is a better explanation; see Davis, , “The Myth of Functional Analysis,” American Sociological Review, XXIX (December 1959), 757–72.CrossRefGoogle Scholar

⁹ Emmet, Dorothy, Function, Purpose and Powers (London and New York 1958), 47.Google Scholar

¹⁰ P. 7. Almond assumes that agreement about the use of the word “system” is widespread and that the term is unambiguous. Neither assumption is correct. For a useful account of the logical difficulties involved in the use of “system,” see Ruck, R. C., “On the Logic of General Behavior Systems Theory,” Minnesota Studies in the Philosophy of Science, 1 (1956), 223–38.Google Scholar

¹¹ Runciman, W. G., Social Science and Political Theory (Cambridge 1963), 41.Google Scholar

¹² For the demonstration of this, see Popper, Karl R., The Poverty of Historicism (London 1957)Google Scholar, passim.

¹³ This may not always be rigorously possible: e.g., “Why has John a cold?” is not completely explained by saying that he was playing with Fred who has a cold, since not everybody who played with Fred caught a cold. We are not yet in a position to make “if—then” law statements about colds as we are with falling bodies or rusting iron.

¹⁴ Hempel, Carl G., “The Logic of Functional Analysis,” in Gross, Llewellyn, Symposium on Sociological Theory (Evanston 1959), 295–96.Google Scholar

¹⁵ There is a sort of cargo cult implied here. Fill in the appropriate categories and all will be delivered in the end. But meanwhile the prophet has put off the cargo-day by drawing up a few more categories to be filled in.

¹⁶ P. 7 (my italics).

¹⁷ P. 8. A difficulty here is that although the explanation may be true of an analytic statement—it can be attributed—it may, as a matter of fact, be false for an actual political “system.” It is possible that there can be more, less, or no functional interrelationship, and that the latter may be more stabilizing than the former. For example, if interdependence is complete, then “negative” feedback (a vicious circle) could destroy a system, whereas with a low degree of systemic interdependence the feedback would be limited. An example of this would probably be Tinker's “brokenbacked” states—Burma and Indonesia—where central dislocation is hardly felt peripherally; see Tinker, Hugh, “Broken-Backed States,” New Society, III (January 30, 1964).Google Scholar

¹⁸ Notice that this leaves untouched the extremely important question of what can be said to constitute change in a system; similarly it neglects the question of how much change is needed before a system becomes a different system.

¹⁹ “Reciprocity and Autonomy,” in Gross, Symposium, 249.

²⁰ See my paper to the 1965 Annual Conference of the British Sociological Association, “Ghana: One-Party or Totalitarian?” British Journal of Sociology (forthcoming).

²¹ P. 8.

²² P. 9 (my italics).

²³ This difficulty is also noticed by Professor Morris-Jones, W. H. in a suggestive review, British Journal of Sociology, XIII (June 1964), 169–73.Google Scholar

²⁴ HempeI, 295. Similarly, Merton, 52: “Embedded in every functional analysis is some conception, tacit or expressed, of functional requirements of the system under observation. … This remains one of the cloudiest and empirically most debatable concepts of the functional theory.”

²⁵ Pp. 35–36.

²⁶ The lack of any serious attempt to integrate either China or the USSR into the suggested framework seems to be a serious taxonomic weakness. On the implicit liberal bias of a great deal of functionalism, see Horowitz, I. L., “Sociology and Politics: The Myth of Functionalism Revisited,” Journal of Politics, XXV (May 1963), 248–64.CrossRefGoogle Scholar

²⁷ It is difficult to know whether the word “boundary,” as used in the introduction, is purely analytic, arising from the idea of a limited system, or whether it has a geopolitical reference. If it is the former, then the use of the term is open to crippling difficulties, and if the latter, there are also difficulties. For a coherent, although not fully satisfactory, examination of the way in which the term can be used, see Easton, David, A Framework for Political Analysis (Englewood Cliffs 1965)Google Scholar, chap. 5.

²⁸ P. 542.

²⁹ Evans-Pritchard, E. E., The Nuer (London 1940).Google Scholar Presumably intrasystemic feuding or war can be eufunctional for the system if the right mix is hit upon or dysfunctional if the right one is not discovered. It might prove as difficult to specify in advance the proportions of peace- and feud-providing optimum social solidarity as it would be to estimate the number of convicted criminals necessary for the same purpose.

³⁰ Evans-Pritchard states quite specifically that “adjacent tribes, and the Dinka who face them, form political systems, since the internal organisation of each tribe can only be fully understood in terms of their mutual opposition.” And further, “Opposition to their neighbours gives them consciousness of kind and a strong sense of exclusion” (p. 123).

³¹ Something like the same problem would occur in a function-oriented analysis of ideologies, incorporating supersubtle foreigners plotting takeovers, i.e., Stalinism, Nazism, and so on.

³² Since there are at least seven permissible uses of the word “legitimate,” it is not possible to discover the significance of its appearing in Almond's definition. These possible and actual uses are (1) in accord with a “liberal” constitution, (2) in accord with the majority, (3) in accord with national well-being, (4) acceptable without revolution, (5) done by an elected government, (6) having extraterritorial recognition, and (7) in accord with due process of law.

³³ Pp. 8, 58–64.

³⁴ It is necessary if a hypothesis is to be disproved that it should be in a nontautological form and that the forecast should be a rigorous one. In systemic forecasting this is clearly impossible unless the stated variables are given some sort of quantitative expression and the system itself is clearly defined. Neither of these conditions is met in Almond's introduction.

³⁵ There is something odd about the bald dismissal of logical analysts as joyous barkers (p. 4) by one who claims scientific status for a theory. A scientific theory, if it ignores elementary logical needs, is neither scientific nor theory.

³⁶ Almond, , in “A Developmental Approach to Political System,” World Politics, XVII (January 1965), 183–214CrossRefGoogle Scholar, apparently fails to grasp the implications of some of the logical points made by such barkers as Hempel and Gouldner. He suggests several dozen new categories, but does make a serious attempt to relate them to each other, if not to a functional theory. His defense of functionalism is that it provides “insights"; this may be true, but is not true enough since indigestion and ideology, science fiction and Marxism—to name but a few—can all rightly claim to provide insight. The point here is an extremely simple one: False theories may yield correct results—alchemy, astrology, and demonology resulted in accurate calendars and primitive psychiatry—but the theories are no less false because the results are subsequently confirmed as correct.

⁸⁷ P. 52.

³⁸ See, for example, LaPalombara, Joseph G., Bureaucracy and Political Development (Princeton 1963)CrossRefGoogle Scholar, passim.

³⁹ But the familiar is redescribed and thus made unfamiliar, e.g., pp. 36–38. Homans, G. C., “Bringing Men Back,” in American Sociological Review, XXIX (December 1964), 809–18CrossRefGoogle Scholar, claims of functionalists that “much of the work they called theoretical consisted in showing how the words in other languages, including that of everyday life, could be translated into theirs.”

⁴⁰ For a much more satisfactory and rigorous account of functionalism than that of Almond, see Merton, Social Theory, and Bredemeier, Harry C., “The Methodology of Functionalism,” American Sociological Review, XX (April 1955), 173–80.CrossRefGoogle Scholar

⁴¹ Flanigan, William and Fogelman, Edwin, “Functionalism in Political Science,” in Martindale, Don, ed., Functionalism in the Social Sciences, Monograph 5, American Academy of Political and Social Science (Philadelphia 1965), 111–26.Google Scholar

⁴² Nagel, Ernest, Logic Without Metaphysics (Glencoe 1956), 251–52.Google Scholar

⁴³ The point about a homeostatic device is that it can cope with “stresses” of a limited intensity and duration; beyond that it is likely to be ineffective. It is also only homeostatic in terms of limited purpose, e.g., guiding aircraft or maintaining a temperature range.

⁴⁴ Merton.

⁴⁵ Bottomore, T. B., Sociology (London 1962), 52.Google Scholar

⁴⁶ P. 40. Authoritarian is subdivided into authoritarian and totalitarian.

⁴⁷ P. 43.

⁴⁸ This is not, of course, to suggest that precisely the contrary has not also been suggested, i.e., that in the absence of class structure and social differentiation, only one party is necessary.

⁴⁹ There is a difficulty here familiar to functional theory: It is sometimes logically false to argue from consequences to causes; but it can easily be the case that those to whom one is attributing agency actually expressed the desire to bring about what occurred. This is true of both Lenin and Nkrumah.

⁵⁰ In his extremely interesting paper “The Development of Social Policy in England: 1800–1914,” Transactions of the Fifth World Congress of Sociology, IV (1962)Google Scholar, I. H. Goldthorpe develops this point in some detail.

⁵¹ Huaco, George, “A Logical Analysis of the Davis-Moore Theory of Stratification,” American Sociological Review, XXVIII (October 1963).Google Scholar

⁵² This is implicitly recognized in all the area studies in the volume and, indeed, in practically every other study of nation-building, all of which have strong historical leanings. It should be added that the point I am making can be proved formafly; see I. C. Jarvie, “The Limits of Functionalism in Anthropology,” in Martindale, 18–34.

⁵³ Compare, for example, Houphouet-Boigny with Nkrumah and Touré or, as Crick has suggested, the leadership of Ghana and Israel, in In Defence of Politics (London 1964), 90.Google Scholar

⁵⁴ Rex, 75–76, is getting at this point in a more generalized way.

⁵⁵ Most of the area studies in fact do focus attention on political ideology. For an excellent, stimulating example of comparative political analysis at this level see Brzezinski, Zbigniew and Huntington, Samuel P., Political Power: USA/USSR (London and New York 1964).Google Scholar