For one trained in the literary criticism of high modernist theater (Williams wrote Drama from Ibsen to Brecht), this "defining characteristic" was new, since theater, like literature, was traditionally experienced in highly bounded performances of single works.

     Williams coined the term "planned flow," but he did not originate the concept. It was propounded by another British literary critic, Terence Hawkes, in a 1967 radio talk published in The Listener (June 8, 1967), which was Williams's own vehicle for TV criticism (Hawkes, pp. 229-41; O'Connor, 1989). Hawkes proposed that the "television experience" included "plays, ... news bulletins, comedy shows, music, and other diverse activities, in the same unit." Reception defined television: "the basic and irreducible constituent of the medium is not the ... individual pro- gramme ... but the much larger unit ... that emerges from the receiving set"'-as shaped by other shows and by the home and family context of reception (p. 234). He argued that "detailed analysis of a text" was not appropriate to television, and that "television's ephemerality in fact forms part of its nature, as an element in its grammar that relates directly to the structure of its units" (pp. 237-38).

     Planned flow had occurred before television. Popu­lar cinema had evolved a programming repertoire in which a session contained comedy cartoons, cliffhanger serials, newsreels, travelogs, B and A features-the "cultural form" that Williams ascribed to television. Radio broadcasting had been defining itself for more than half a century, and audiences were habituated to planned flow in the home. Even TV's theatrical antecedents, such as music hall, were hybrid forms with internal segmented flow, which transferred directly to television (e.g., the long-running Sunday Night at the London Palladium and Royal Variety Shows).

     Thus, flow was not a new or newly noticed phenomenon when Williams leant his weight to it. His intervention was important because it marked a change in theoretical perspective. Television simply defeated high modernist textual empiricism (Hawkes, pp. 235- 37), not least because some of its most important textual content was the "television that wasn't there," such as advertisements, trailers, station Ids, and other gaps between the programs that produced the flow (Hartley, ch. 11). Analysis of single shows could not lead to an understanding of television. Rather, it was necessary to relate the organization of production and distribution (i.e., the "planned" aspect of planned flow) to the family home context of viewing, the experience of consumption, and the identity or subjectivity of audiences. The flow that they were thought to experience cleared the necessary theoretical ground for a tum away from textual analysis to the subsequent flood of audience ethnographies in TV studies.

     Television criticism, as an attempt to educate audiences in civic virtues or aesthetic values, was abandoned. Instead, television studies arose as an attempt to specify the relation between producers and audiences in terms of power. Audiences were not seen as having the freedom to form their own opinions and conclusions regarding what they saw on television. On the contrary, the notion of planned flow allowed the consumer experience to be thought of in the most general and abstract terms, as subject to the plans of broadcasters-namely, to be an ideological practice. Analyzing actual flows as experienced by individual viewers was rarely undertaken, because Williams had installed planned flow as a defining characteristic of television, not as a hypothesis to be tested against evidence.

     Television was watched in households and nations, both highly fraught and ideological institutions where gender, class, race, and other aspects of identity were constantly in contention and pervaded by power, whether power was understood in Marxist terms (as struggle) or in Foucault's (as the administration of life). Individuals' experience of flow did not interest investigators (unless it was their own), because they already knew what it meant as an instance of abstract power relations. In this respect flow was not unlike other "fabulous powers" (Ian Connell's phrase) that people have attributed to electronic media throughout modernity. Jeffrey Sconce made the connection: "fan­tastic conceptions of media presence" (grounded in a "metaphysics of electricity") "have often evoked a series of interrelated metaphors of 'flow,' suggesting analogies between electricity, consciousness and information" (p. 7).

     Flow became prominent via its uptake in cinema studies just as that field was going through a highly abstract theory-oriented phase. But even as they paid their respects to Williams, film theorists could not bring themselves to agree with his concept of flow. Thus John Ellis in the United Kingdom and Jane Feuer in the United States, among many others, refined and redefined the concept. Ellis determined that the "small­ est signifying unit" of television was the segment, which led to "segmented flow." Feuer pointed out that there was no such thing as pure flow, only a dialectic between segmentation and flow. Previously, Hawkes had drawn attention to the fact that watching television could result in "disconcerting juxtaposition" (the opposite of flow) when shows or segments were in jarring contradiction to one another. For example, he mentioned having seen The Black and White Minstrel Show followed immediately by a news program marked by a main story of race riots in the United States (Hawkes, p. 238; see also Comer, chs. 5-6). John Caldwell mounted an especially spirited critique of the concept of flow in his book Televisuality (Caldwell, pp. 158-64, 264).

     People do not remember flows; they only remember shows. But once installed as a founding concept in television theory, flow remained available for application to new media to which it was even less suited, that is, interactive and computer-based media. Ellen Seiter, for instance, suggested that in the hands of commercial and advertising "programmers" the Web resembled television; here. planned flow consisted of the attempt to "guide the user through a pre-planned sequence of screens and links." But at best the new interactive media could only claim "flow, interrupted," since the "cultural form" of interactivity meant that users could not be carried along uncontrollably (though they could follow a planned sequence if they chose to). Therefore, the concept has limited application to new media and serves mostly as a reminder of the period when ideology theory required users who were passive and uncritical, at the mercy of the persuasive blandishments of the marketing communication that wants them to stay tuned, during prime time, at any price.