Misinterpreting Dahl on power

Abstract

Dahl’s critics have often called attention to important aspects of power relations – e.g. the suppression of issues and consciousness control. The critics, however, have not always portrayed Dahl’s views as accurately as one might wish. Distinguishing between an abstract concept of power and operational definitions adopted for purposes of specific research projects is fundamental for Dahl. Furthermore, Dahl’s concept of power implies nothing about the preferences of B, is not zero-sum, does not necessitate compulsion, may or may not be subtle or visible, is not confined to material resources, and may be either direct and immediate or indirect and long term.

Keywords:

• power
• faces of power
• zero-sum
• interests
• preferences

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

This article draws on the author’s forthcoming book entitled Power and International Relations.

1. See, for example, Gruber (2000), Hayward (2000), Morriss (2002), Barnett and Duvall (2005), Berenskoetter and Williams (2007), Gallarotti (2010), Finnemore and Goldstein (2013), Guzzini (2013).

2. Following Dahl’s usage in 1957, the terms power and influence will be used interchangeably here.

3. Describing the community power debate as one between Dahl and his critics can be misleading, since Dahl is more accurately described as a target for his critics than as an active participant in the debate. Shortly before his retirement, he observed that he had not wanted to spend his time ‘answering critics’ and mused that he may have ‘done less of that than [he was] properly obliged to do’ (Baer et al. 1991, p. 176). Dahl’s reluctance to engage his critics may account for some of the later misinterpretations of his concept of power.

4. Most of Dahl’s critics make no reference to the ‘conclusion’ of this article. Stewart R. Clegg is an exception. He portrays the conceptualizer (‘a nit-picking sort of character’) as the antagonist and the operationalist (with whom ‘we are clearly led to sympathize’) as the protagonist (1989, p. 54).The title of the article, the thrust of the argument, and the fact that Dahl gives the first and last word to the conceptualizer would seem to argue against Clegg’s interpretation.

5. For example, even though the ability to ‘initiate’ successful policy proposals plays an important part in the analysis, the concept of initiation is ignored. For a discussion of the concept of initiation, see Baldwin (1966).

6. The terms operational definition and operational measure are often used interchangeably.

7. An influential collection of articles on power published between 1950 and 1968 contained 27 entries, of which only 5 focused on the ‘community power’ debate (Bell et al. 1969).

8. For an overview of the ‘community power debate’, see Ricci (1980). This article hardly mentions abstract concepts of power, but focuses instead on methodology and democracy. See also Polsby (1980).

9. Whether Lukes was familiar with Dahl’s 1968 article on ‘Power’ is problematic. He neither cites it nor includes it in the bibliography of the 1974 edition. Nor does he consider either of the first two editions of Modern Political Analysis.

10. A Google Scholar search reveals more than twice as many citations to Lukes’ book as to Dahl’s 1957 article – about the same number as citations for Dahl’s article and Who Governs? combined.

11. While no single author subscribes to all of the elements of this dominant narrative, the following provide examples of various combinations of its elements: Clegg (1989), Strange (1994), Gruber (2000), Hayward (2000), Barnett and Duvall (2005), Berenskoetter (2007), Grewal (2008), Nye (2011), Finnemore and Goldstein (2013).

12. In his annotated bibliography Lukes dismisses Dahl’s ‘Concept of Power’ as a 'first, rather crude effort to define and operationalise “power”’. He praises Who Governs? as ‘a finer, subtler work than its critics and defenders might suggest, partly because it contains the evidential basis for criticizing its conclusions’. (1974, p. 60).

13. On the question of whether power should be considered as ‘asymmetric’, see Baldwin (1978).

14. ‘Domination’ is perhaps the most abused and ill-defined term in the lexicon of power.

15. Any parent of a two-year-old or a rebellious teenager, of course, could have identified this defect in Lukes’ 1974 definition of power. Of course, there are also times when ‘disempowering’ a two-year-old by decreasing his resources is in his interest – e.g. ‘you can’t play with that gun, hammer, or knife’. Likewise, it may be in the interest of a teenager to ‘disempower’ him or her by taking away the car keys.

16. In the second edition, Lukes does mention Boulding’s (1989) book as a ‘thoughtful’ book.

17. On the importance of drawing a ‘clear distinction between measurement issues and disputes about concepts’, see Adcock and Collier (2001).

18. The divergence between operational concepts and the abstract concept they are intended to operationalize is not unique to Dahl’s concept of power. King et al. (1994) make a similar point. Barnett and Duvall (2005, p. 40) contend that ‘there is a widely accepted conceptualization [of power] that is viewed as the only way to understand power: how one state uses its material resources to compel another state to do something it does not want to do’. They add that ‘attempts by scholars to operationalize power follow from this definition’. Actually, a wide variety of operational definitions could ‘follow from’ this definition, depending on the specific research project at hand.

19. Examples abound, but see Clegg (1989), Hayward (2000), Berenskoetter (2007), Grewal (2008), and Nye (2011). A particularly influential – and misleading – example is Gaventa (1980, 2007), who quotes Dahl’s 1957 ‘intuitive idea of power’ and immediately follows it with a quote from Polsby (1980) to the effect that community power may be studied by examining ‘who participates, who gains and who loses, and who prevails in decision-making’. Gaventa neglects to point out that Polsby is identifying only one of many different ways to operationalize Dahl’s abstract concept of power, thus leaving the reader with the impression that there is some necessary connection between the two. Contrary to this impression, Dahl’s abstract concept of power implies nothing about participation, who gains, who loses, or who prevails in decision-making.

20. The Marshall Plan example is solely for illustrative purposes. I should like to sidestep the question of whether European countries really preferred integration.

21. As noted above, Lukes himself is careful to distinguish between Dahl’s ‘concept’ of power and what Lukes labels Dahl’s ‘view’ of power. His readers, however, often overlook this distinction. Dahl’s concept of power is depicted as narrow and one-dimensional by Nye (2004, 2011), Gruber (2000), Hayward (2000), Barnett and Duvall (2005), Grewal (2008), Gaventa (1980, 2007), Berenskoetter (2007), and countless others. Oddly, Berenskoetter attributes Dahl’s definition of ‘A getting B to do something B would otherwise not do’ to Who Governs? and to Dahl’s article in the International Encyclopedia of the Social Sciences even though it appears in neither.

22. Any given instance of A’s influence with respect to B may have both favorable and unfavorable effects on the interests of either actor. Thus, winning, losing, and drawing are here defined in terms of their net impact on the interests of each actor.

23. Dahl also described the process for calculating the 14,000 ‘faces’ in a paper delivered at the 1964 annual meeting of the American Political Association and reprinted in Dahl (1997, pp. 295–296).

24. Parsons also attributed a zero-sum concept of power to C. Wright Mills, V. O. Key, and to Lasswell and Kaplan. For a discussion of whether this applied to Lasswell and Kaplan, see Baldwin (1989). Others who have designated Dahl’s concept of power as zero-sum include Scott (2001), Barnett and Duvall (2005), Berenskoetter (2007), and Dowding (2012). The attribution of a zero-sum concept of power to Dahl is usually asserted without explanation of the basis for the assertion, as if it were self-evident.

25. There are only three possible outcomes to a zero-sum game (win, lose, draw). As Table 1 shows, however, Dahl’s concept of power allows for nine possible outcomes.

26. Early on, Dahl often used the terms power and influence interchangeably. In later editions of Modern Political Analysis, he adopts the language of Lasswell and Kaplan in treating ‘power’ as a subtype of ‘influence’. This is more of a terminological change than a conceptual one. This change of terminology, however, does not explain the comments above, since they rely on the 1957 article and on Who Governs?

27. This assertion about ‘most texts’ is not supported by references to specific texts.

28. Hayward (2000, p. 36) goes further than Barnett and Duvall in describing Dahl’s comment about ‘no action at a distance’ as ‘explicit and uncompromising’. When terms are explicitly left undefined, however, phrases like ‘implicit and wishy-washy’ would seem to be more apt descriptions.