summarize and analysis these 3 articles.

University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Sociology.

http://www.jstor.org

"All Hits Are Flukes": Institutionalized Decision Making and the Rhetoric of Network Prime-Time Program Development Author(s): William T. Bielby and Denise D. Bielby Source: American Journal of Sociology, Vol. 99, No. 5 (Mar., 1994), pp. 1287-1313 Published by: University of Chicago Press Stable URL: http://www.jstor.org/stable/2781150 Accessed: 05-11-2015 22:00 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

"All Hits Are Flukes": Institutionalized Decision Making and the Rhetoric of Network Prime-Time Program Development'

William T. Bielby and Denise D. Bielby University of California, Santa Barbara

Drawing upon institutionalist theory this article analyzes how the introduction of new cultural objects produced for a mass audience is managed through an organized discourse. Data come from an- nouncements of prime-time television series in development for the 1991-92 season by the four U.S. television networks. Maximum- likelihood logit analyses support the conclusion that network pro- grammers working in a highly institutionalized context use reputa- tion, imitation, and genre as rhetorical strategies to rationalize and legitimize their actions. This study contributes to institutionalist theory and the sociology of culture by explaining the content and consequences of business discourse in a culture industry.

In this article we examine how the rhetoric of network television pro- grammers shapes the introduction and evaluation of new prime-time se- ries. Held accountable for decisions made under conditions of ambiguity and uncertainty, network programmers employ a carefully organized dis- course to frame how their actions are appraised. Our analysis shows how programmers' rhetorical claims are organized to manage the reception of their decisions by important constituencies.

1 An earlier version of this paper was presented at the 1991 annual meetings of the American Sociological Association, held in Cincinnati, Ohio. We thank Robert Faulk- ner, Ryken Grattet, Valerie Jenness, John Meyer, Harvey Molotch, Richard Peterson, John Sutton, Joseph Turow, the AJS reviewers, and the members of the Comparative Institutions Seminar in the University of California, Santa Barbara, Department of Sociology for their comments and suggestions. We are grateful to Betsy Frank and to the employees of ABC/Capital Cities, CBS Inc., Fox Broadcasting Co., and NBC Entertainment who shared with us network program development promotional mate- rials. This research was supported by the National Science Foundation (SES 89-10039) and by the Academic Senate of the University of California. Address correspondence to William T. Bielby, Department of Sociology, University of California, Santa Bar- bara, California 93106.

? 1994 by The University of Chicago. All rights reserved. 0002-9602/94/9905-0005$01 .50

AJS Volume 99 Number 5 (March 1994): 1287-1313 1287

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

We apply institutionalist theories of organizations (Powell and DiMag- gio 1991) to understand how decision makers operating under conditions of ambiguity and uncertainty cope with contradictory commercial and aesthetic assessment criteria. We develop hypotheses regarding the con- tent and consequences of network programmers' rhetorical strategies for introducing new series, and we test our hypotheses using data from the networks' program development announcements for the 1991-92 prime- time season.

Our study contributes to an understanding of how the introduction of new cultural objects is actively managed in highly institutionalized and centralized culture industries. We analyze how industry context shapes the organization and content of decision makers' discourse and demon- strate how that discourse is used to manage the conflicts and contradic- tions inherent in the industry's structure. In the concluding section we discuss the implications of our work for the study of culture industries more generally and suggest avenues for future research.

INTERDEPENDENCE AMONG SUPPLIERS, NETWORKS, AND MARKETS

Each year, the four networks evaluate thousands of concepts for new series and purchase approximately 600 pilot scripts. From these, the networks select about 20% to be produced as pilots at the networks' expense (CBS Inc. 1990). About one-third of the pilots eventually appear on the prime-time schedule. For example, of the 112 pilots commissioned by ABC, CBS, NBC, and Fox for 1991, 23 debuted in the fall and another 15 were scheduled as midseason replacements.

Until very recently, the FCC's Financial Interest and Syndication Rules (the "Fin-Syn" rules) have placed strict limits on the amount of prime-time programming that can be produced by the networks them- selves, so most prime-time pilots and series are supplied by independent writer-producers working for outside production companies. Program suppliers include small independent companies and the television subsidi- aries of the major film studios. Either as sole producers or in joint ven- tures with smaller companies, the major studios account for a majority of the series produced for network prime time. Program suppliers retain ownership of the series and license networks to one first-run and one rerun broadcast of each episode (Bielby and Bielby 1990; Cantor and Cantor 1992).

Since (with a few exceptions) the networks do not own the shows they broadcast, advertising revenues from network broadcasts are their primary source of profit from prime-time series. At the same time, the

1288

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

licensing fees received by the program suppliers typically do not cover the cost of production. A typical sitcom costs about $600,000 per episode to produce and is licensed to the network for about $500,000. An episode of a typical one-hour drama costs about $1.2-$1.3 million to produce and is licensed to the network for $250,000-$300,000 less than the cost of production (Channels 1990). Program suppliers incur these short-term losses in anticipation of substantial profits from eventual syndication of successful series.

Thus, program suppliers and network programmers are mutually de- pendent. On the one hand, a writer-producer who creates a new series seeks access to a network's prime-time schedule. The supplier is hoping for a network run of at least three or four seasons, so that enough episodes will be produced to make the series profitable in subsequent syndication. On the other hand, the network programmer is dependent upon program suppliers for new series that will attract audiences that advertisers want to reach. A series that fails to deliver a sizable audience with a desirable demographic composition will be unprofitable for the network and its affiliates. However, participants in network program development are not simply interacting in a product market. They are also embedded in an institutional context that introduces a symbolic dimension that they manage and enact through their decisions.

DECISION MAKING AMID AMBIGUITY AND UNCERTAINTY

"Institutions" are routinely reproduced, taken-for-granted social prac- tices that have "rule-like status" in thought and action (Meyer and Rowan 1977; Jepperson 1991). Network prime-time program develop- ment is an institution in this sense, and decisions about introducing new series are made in an "institutionalized" context. In such contexts, deci- sion makers cope with ambiguity and uncertainty by substituting imita- tion, routines, and rules of thumb for rational calculation as decision criteria (March and Olsen 1976). To maintain legitimacy, they are likely to engage in activities that have "ritual significance" to outsiders, pro- viding "prudent, rational, and legitimate accounts" of their actions (Meyer and Rowan 1977).

Highly institutionalized decision contexts exhibit two distinctive fea- tures: (1) technologies are poorly understood; and (2) actions are evaluated according to multiple, ambiguous, and contradictory criteria by external constituencies (Meyer and Rowan 1977). The context for decisions about new prime-time series is highly institutionalized in both respects. First, network programmers are making decisions about productions for which there are no agreed-upon standards of competence. An experienced pro-

1289

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

grammer can probably distinguish well-crafted from mediocre scripts and make informed judgments about the quality of acting, editing, and direction of a pilot. Nevertheless, the programmer has no reliable basis for predicting whether audiences, advertisers, and critics will accept the series. In the words of Jeff Sagansky, president of CBS Entertainment, "All hits are flukes" (Frank 1991, p. 1; also see Cantor and Cantor 1992, p. 70). Gitlin (1983) found network executives consistently expressing views like these and concluded that the "problem of knowing" is a key feature of program development decisions.

Second, network programmers are making decisions that are subject to multiple, ambiguous, and often conflicting assessment criteria. Least ambiguous are criteria of commercial success. A commercially successful series is one that delivers a large audience with a demographic compo- sition valued by advertisers. However, feedback from ratings comes months after the decision is made to schedule a series, and even that feedback can be misleading for a series that slowly builds an audience over the course of one or two seasons.

Criteria for critical acclaim are more varied, and critics often reserve their most favorable assessments for so-called quality series that the net- works have dropped from their prime-time schedules. Critics' "obituar- ies" for their favorite canceled series often reveal the conflicting criteria that underlie judgments about aesthetic versus commercial success. For example, at the end of the 1990-91 season, Wall Street Journal critic Robert Goldberg (1991) celebrated the qualities of "subtlety in depicting relationships," "solid writing," and "innovative ideas" among canceled shows on his "Best of 1990-91" list, while Los Angeles Times critic Rick Du Brow (1991a) recently lamented the loss of "humanistic ensem- ble dramas," "well-crafted TV fiction," and "work of major social import. "

Dozens of special interest groups also monitor network executives' programming decisions. They evaluate network series on issues as diverse as representations of race, age, gender, drug use, violence, birth control, and "traditional" family values (Montgomery 1989). In short, as a cul- tural object, a television series has a range of meanings attributed to it (Griswold 1987). A decision to develop a series for prime time is simulta- neously a choice about a commercial commodity, an aesthetic endeavor, and a social institution. As a result, those making programming decisions will be variously evaluated according to perceptions of their business judgment, their aesthetic tastes, and the values they impart.

Because the industry is so highly centralized, programmers are more directly accountable to commercial interests than to other constituencies. DiMaggio (1977) has described such an industry context as a "centralized

1290

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

brokerage system." Network programmers2 mediate the business rela- tionship between creative personnel employed by independent production companies and advertisers who bear the costs of distribution. The pro- grammer "brokers" the inherent conflict between the creative interests of the writer-producers who create and supply programs and the commer- cial interests of advertisers. Programmers perform the monitoring func- tion that would otherwise reside in the authority relations of a bureau- cratic structure or in the professional standards and contractually mandated product specifications of a craft structure.

In mediating the relationship between writer-producers and the net- works, programmers do not give equal weight to creative and commercial concerns. Writer-producers have almost no alternative to the four net- works as distribution channels for their creations, despite the growth of cable and first-run syndication in recent years. Most are also dependent upon one of the seven major studios for financing, and, of course, series creators have their own interests in commercial success. Thus, as DiMag- gio (1977) has theorized, in a mass-culture industry with a high level of market concentration such as network television, brokers are more di- rectly accountable to commercial interests than to creative interests.

RHETORICAL STRATEGIES FOR INTRODUCING NEW SERIES

"Framing" Prime-Time Program Development

A "frame" is a central organizing idea for making sense of events (Gam- son 1988; Gamson and Modigliani 1989). Gamson (1988) and Snow and Benford (1988; Benford 1993) have applied the concept to represent how social movement elites develop rhetorical strategies to mobilize constitu- encies and shape understandings. According to Gamson and Modigliani, meaning in media discourse is managed by assembling frames into "inter- pretive packages." A package "offers a number of different condensing symbols that suggest the core frame and positions in shorthand, making it possible to display the package as a whole with a deft metaphor, catchphrase, or other symbolic device" (Gamson and Modigliani 1989, p. 3).

We apply the frame concept to the vocabularies articulated by decision makers to justify their actions in highly institutionalized contexts. In

2 The term "network programmer" refers to the top executives who make decisions about the prime time schedule and work with independent producers to develop new series. At the top of the management hierarchy is the president of the network's entertainment division. Reporting to the president are vice presidents responsible for research, marketing, program development, and current prime-time programs.

1291

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

developing new prime-time series, network programmers are compelled to provide legitimate accounts for their decisions (Meyer and Rowan 1977). We argue that they do so by organizing discourse around widely accepted frames. Specifically, in describing series in development, net- work executives invoke the framing devices of genre, reputation, and imitation.

Genre, Reputation, and Imitation

Genre.-Television genres are conventions regarding the content of television series-formulas that prescribe format, themes, premises, characterizations, etc. In contemporary television, consensus among writers, producers, programmers, advertisers, and audiences over the boundaries of genres is probably greater than in any other area of popular culture. Ideas for new series are "pitched" in terms of widely recognized genres, while network scheduling decisions, advertisers' purchasing deci- sions, and audience viewing patterns are all based, to some extent, upon shared understandings about program categories.

Since the late 1950s the television industry has recognized two basic genres for prime-time network series: the half-hour situation comedy ("sitcom") and the one-hour drama. These genres are recognized and reproduced in the industry's organizational structures. Since the late 1970s the entertainment divisions of ABC, CBS, and NBC have con- tained separate units for comedy development and drama development, and corresponding units exist in the television divisions of the major studios. Pilots for new sitcoms and dramas are easily recognized and labeled as such.

Over the past several years the "reality" series has become recognized as a third basic prime-time genre. The reality genre label is typically applied to inexpensively produced half-hour nonfiction series other than programs produced by the network news divisions. The reality genre is less established than the sitcom and drama categories. Nevertheless, the reality series' existence as a distinct genre is recognized as such by pro- grammers, producers, creative personnel, advertisers, critics, the indus- try press, and, increasingly, audiences.

As industry conventions, genres can be viewed as forms of social orga- nization that facilitate the coordination of production. Work outside of established genres disrupts shared understandings and requires more ef- fort to coordinate and promote (Becker 1982). Accordingly, a writer- producer seeking to sell a pilot that defies the conventions of established genres can expect resistance from network programmers. Moreover, a programmer seeking to place such a series on the network schedule can

1292

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

expect advertisers and local affiliates (who sell local advertising time) to demand reassurance that despite-or because of-its innovative ele- ments, the new series will be accessible to audiences and commercially viable.

Genres are more than just shared understandings that economize on the costs of doing business. They are taken-for-granted categories that guide thought and action, serving to rationalize and legitimate decisions made in a context that is characterized by a high degree of ambiguity and uncertainty. As such, they are readily available as a framing device. Thus, locating a series pilot with respect to an established genre provides an imme- diate frame of reference for the new and unknown cultural product. Even before the pilot is produced, the potential new series is linked to a category that is widely perceived as familiar, understandable, and appropriate.

Reputation. -In culture industries the success of new products cannot be known a priori. As a result, those who propose new products are likely to be evaluated on the basis of reputations built upon prior suc- cesses (DiMaggio 1977, p. 442). The importance of reputation has been documented in studies of studio musicians (Faulkner 1983), filmmakers (Baker and Faulkner 1991), and television writers (Bielby and Bielby 1993; Gitlin 1983). Accordingly, we expect reputation to figure promi- nently in programmers' rhetoric about series in development. Linking new series to producers' prior hits reassures commercial constituencies that well-crafted episodes will be produced in an orderly and timely manner and will contain elements proven successful with audiences in the past. Similarly, linking a new series to a well-known celebrity estab- lishes an association with a familiar and successful commodity.

Imitation.-When decisions are made under conditions of ambiguity and uncertainty, decision makers often attempt to establish legitimacy by imitating the successful efforts of others (March and Olsen 1976; Di- Maggio and Powell 1983). Accordingly, we expect network programmers to rely upon imitation as a rhetorical strategy in their discourse about series in development. A new production can be described as similar to another hit series, even if it is supplied by a different production team. In addition, network executives can emphasize a series' similarity to or origins in a product from another popular medium, such as a hit film, play, comic book, or novel.

In sum, in a context where formulas for producing successful products do not exist and "all hits are flukes," decision makers rely on rhetorical strategies to demonstrate that their actions are rational and appropriate. Network executives construct succinct interpretive packages by linking new series to established genres, to reputable producers, and to other popular and successful cultural commodities. If the series is successful,

1293

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

external constituencies will view the executives' actions as understand- able and familiar (Hirsch 1986).

The Sequential Organization of Programmers' Rhetorical Strategies

Network executives' discourse about program development is articulated in four stages. First, in March executives from each network make pre- sentations to advertisers' representatives in which they announce the pilots in development for the new season. Second, at the affiliates' meet- ings in May network executives brief the press, advertisers, and execu- tives from their affiliated stations about their fall schedules. Third, the annual press tour is held in late July. Critics from around the country come to Los Angeles to meet with network executives about the new season. Finally, in late summer the all-out marketing of the new season begins, including on-air promotions, advertising in the print media, and other publicity efforts. Throughout these four stages, the network execu- tives are able to shape the reception of new series before critics, advertis- ers, and audiences have had the opportunity to view them. Each stage is covered in detail by the industry press, so network executives are speaking indirectly to the larger business and creative community as well as to those with whom they interact directly.

While informed observers of the industry are unlikely to take network piogrammers' claims at face value, until the series air observers have little independent information with which to judge those claims. More- over, by introducing the new season in stages, network programmers can manage the decoupling of commercial and aesthetic evaluation. They can attend primarily to the bottom-line financial concerns of producers, advertisers, and the networks and speak to the critical community in their own language later as the new season approaches.

In their meetings with advertisers in March, network programmers outline their overall programming strategies and describe each of the series pilots they have commissioned for the new season. At this stage, the pilots exist only as scripts, and the programmers themselves have yet to see the product they are describing.3 Their discourse about the pilots is carefully constructed around the frames of genre, reputation, and imi- tation. The "interpretive packages" they present are, literally, packages. Each of the 300 or so advertising executives attending the meeting re- ceives a press kit containing brief synopses of the network's pilots. We rely on these written descriptions from March 1991 for network program- mers' claims about potential new series. Claims about overall develop-

3There are a few exceptions, such as pilots that are holdovers from the previous season and spin-off pilots that are episodes of existing series.

1294

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

HOME IMPROVEMENT priIG[ (working title) _iuw a COMEDY/HALF HOUR DEVELOPMENT

It's not easy being a man these days. But Tim Allen, the star of this new domestic comedy from the creator of "Roseanne" and "Carole & Co.," is determined to re-establish the role of husband and father as provider and protector. He's a Mid-western everyman who is determined to re-establish the natural order of things -- with men in charge. A celebrty in his hometown, Tim epitomizes 'Mr. Fix It" as the host of a local home improvement show. Unfortunately, his "personal home audience" is comprised of a no-nonsense wife and three skeptical boys who challenge Tim's attitudes and often question his solutions. But he pushes ahead, confident in the belief that most problems can be solved by simply increasing the voltage.

Production Company: Disney Executive Producers/Writers: Matt Williams

Carmen Finestra David McFadzean

Director: TBA

CAST

Tim Allen

FIG. 1.-An example of a program development announcement for the 1991- 92 prime-time season.

ment strategies are ascertained from industry press accounts of state- ments by network programmers at the March meetings. Based on our understanding of the institutional context, we develop hypotheses about the content of the claims about potential new series and about the rela- tionship between the claims and the probability of a pilot making it onto the network prime-time schedule.

DATA AND MEASURES

Linkage Claims: Reputation and Imitation

Synopses of potential new series are intended to generate interest in and support for the network's new offerings. Each description attempts to convey the content and potential of a new series in just a few sentences. The example in figure 1, from the ABC press kit, is typical.

In this example of ABC's Home Improvement, the series is linked to a genre (half-hour comedy), subgenre (domestic comedy), a celebrity (Tim

1295

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

Allen), and a creator with a proven track record in situation comedy (Matt Williams, of Roseanne and Carole and Company).

We coded binary variables to represent linkages such as those in figure 1 for each of the 112 synopses in the press kits for the four networks. Two of the linkages represent claims regarding reputation. A producer link denotes that a series is identified with an executive producer associ- ated with a prior successful series. A celebrity link indicates that the name of at least one cast member appears in the series' description.

Two linkages represent claims of imitation. First, some pilots are de- scribed in terms of prior successful series, even when the pilot's producer is not associated with the earlier show. For example, the Fox network's description of The 50-Minute Man begins as follows: "Magnum . . . Remington . . . Hunter . . . Forget 'em. We bring you Henry Fellows, private eye." We code descriptions such as these as linked to prior series. Second, some series are described in terms of the relationship to success- ful novels, films, or plays. For example, NBC's description of Eerie, Indiana makes reference to the work of author Stephen King, and CBS's Big Girls Don't Cry is described as a spin-off of the film Mermaids. We code claims such as these as linkages to other media.

The meetings in March and the descriptions of potential new series are covered in detail by the industry trade papers, the Hollywood Re- porter and Daily Variety.4 Overall, there is a high degree of agreement between the linkages claimed in the network synopses and those reported in the industry press. For example, the Hollywood Reporter covered Home Improvement as follows: "Roseanne creator Matt Williams has created a domestic comedy called Home Improvement, starring Tim Allen as a Midwestern everyman who is determined to re-establish the natural order of things with men in charge. Williams, Carmen Finestra and David McFadzean executive produce for Walt Disney Television" (de Moraes 1991a, p. 6). Overall, nearly 90% of the linkages appearing in the network press kits were reproduced in the coverage by Daily Variety and the Hollywood Reporter, according to independent codings of the network and press descriptions. This consensus suggests that the net- works are effectively using the framing mechanisms of reputation and imitation to shape media discourse about program development. That is, the industry trade papers are attending to and reproducing the claims contained in the structured discourse of the network programming execu- tives. As Gamson and Modigliani (1989, p. 9) found in their study of media discourse on nuclear power, journalists "make official packages the starting point for discussing an issue."

4 Coverage of network executives' presentations of their development slates for the 1991-92 season appeared in Daily Variety and the Hollywood Reporter on March 20 andMarch 21, 1991.

1296

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

Genre Claims

In addition to reputation and imitation, network programmers use genre as a framing device for discourse about new series. Without exception, network synopses locate the series with respect to the genres of comedy, drama, and reality, so we coded binary variables to reflect those classifi- cations. In some instances, pilots were described in ways that mixed or even transcended the three established genres. For example, ABC de- scribes The Chameleon as a "new form comedy series" combining live action with old film clips, while Moe's World, a half-hour drama is de- scribed as combining fantasy, dreamscape, and animation. Overall, we classified 24 of the synopses as mixing or transcending established genres. Our coding is based on the networks' claims, not on whether the series actually departs significantly from institutionalized conventions. Our in- terest is in how programmers use genre as a framing device, not in the actual format and content of the series.

Again, the success of network programmers' rhetorical strategies is reflected in part by the extent to which their claims are reproduced in the industry trade papers. When the network snyopsis contains language suggesting mixing or transcending genres, unconventional labels are often used in the trade papers as well. For example, NBC's animated Fish Police is described in the industry press as an "animated 'fish noir"' and "comedy spoof," ABC's Saturdays is described as an "hour long anthology comedy," and Fox's Chameleon is described as "action com- edy, animated/live. "

Outcomes: Selection for Network Prime-Time Schedule

Our quantitative analyses examine the relationship between network claims regarding their series in development and the likelihood that a pilot will be "picked up" for the network prime-time schedule. The four networks announce their schedules at meetings in May with representa- tives from their local affiliated broadcast stations. Based on those an- nouncements, we code whether a pilot has been selected to premiere in the fall and whether it has been selected as a possible midseason replacement.5

5Between March and May, programming executives view the pilot episodes, conduct their marketing research, and make their decisions about the upcoming season. Deci- sions about the fall schedule are often made days and sometimes just hours before the meetings with affiliates and are accompanied by another round of claims about the series that have been selected for the upcoming season. In this article, we do not analyze the claims surrounding the schedule announcements. Nor do we analyze schedule changes made following the initial announcements.

1297

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

Finally, our analyses include several control variables. Three binary variables control for differences among the four networks in the series development process. A fourth binary variable denotes whether or not a pilot has originated from the in-house production division of a network. Since networks profit directly from the eventual syndication of the few series they are allowed to own, in-house pilots may receive unique treat- ment in the process of making decisions about the prime-time schedule.

HYPOTHESES

Content of Programmers' Claims

We have argued that network programmers actively manage the recep- tion of new series through an organized discourse of program develop- ment. That discourse is structured around the framing devices of reputa- tion, imitation, and genre. Our hypotheses address both the content and consequences of the interpretive packages articulated by network pro- grammers. We expect network programmers' discourse will be designed to (1) provide a legitimate account of decisions made in a context of ambiguity and uncertainty and (2) manage the multiple and often con- flicting assessment criteria of commercial and creative constituencies.

Since the mid-1960s the evaluation of network prime-time program- ming on the basis of aesthetic criteria has been decoupled from assess- ment based on commercial success. Television criticism that was taken seriously by industry elites disappeared in the 1950s, and today television critics speak to and for viewers, not to the art world of writers, producers, and programmers (Lang 1958; Boddy 1990). The decoupling of assess- ment based on business and aesthetic criteria allows programmers to articulate seemingly contradictory claims about the creativity and com- mercial viability of new series. However, because they primarily repre- sent management interests in brokering the relationship between series' creators and the networks, programmers' claims will emphasize the com- mercial viability of the new series. Specifically, programmers will de- scribe new series in ways that draw attention to the reputations of the series' producers. Accordingly, we offer the following hypothesis:

HYPOTHESIS 1. -In describing new series, claims emphasizing the rep- utations of series' creators are more frequent than any other kind of claim.

A linkage to an established writer-producer is a powerful symbol for reassuring advertisers, affiliates, and network top excutives that a new series is commercially viable. Establishing such a claim reduces the need to use other reputational or imitative rhetorical strategies to describe a new series. It also reduces the need to make claims about departing from widely shared conventions. Thus, we hypothesize:

1298

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

HYPOTHESIS 2.-A linkage to an established writer-producer reduces the probability that a series will be described in terms of other reputa- tional or imitative linkages.

HYPOTHESIS 3.-A linkage to an established writer-producer reduces the probability that a series is described as mixing or transcending tradi- tional genres.

Since the Fox network's inception in 1987, executives there have framed their programming strategies as innovative and unconventional (Block 1990; Du Brow 1991b). Accordingly, we expect descriptions of new series by Fox to contain fewer reputational and imitative claims and more claims about unconventional approaches. Specifically, we hypoth- esize:

HYPOTHESIS 4.-Compared to the other three networks, the Fox net- work is less likely to frame new series in terms of linkages to established producers, stars, prior series, and products from other media. The Fox network is more likely than the others to describe new series as mixing or transcending established genres.

Consequences of Programmers' Claims

Network programmers frame their new series in terms of the reputations of their producers and stars, their similarity to prior series and successful products from other media, and their conformity to or departure from established genres. We expect the actual scheduling decisions made by network programmers to be more highly associated with reputational claims about series' creators than with any of the other framing devices. More than any other group, network programmers seek to reassure those with commercial interests in the new productions. We expect that empha- sis on the reputation of the series' creator will be the most effective rhetorical strategy for providing that reassurance. In addition, claims about series' creators reflect structural relationships between program- mers as brokers and the writer-producers who supply new series. Accord- ingly, we expect reputational claims about series' creators to be more strongly associated with scheduling decisions than are other framing de- vices. As a result, we hypothesize:

HYPOTHESIS 5.-Series described as originating from successful writer-producers are more likely to be selectedfor the network prime-time schedule than series not linked to creators with established reputations. Other reputational and imitative linkages have little or no effect on sched- uling decisions.

Claims about creativity and innovativeness reassure those judging net- work decisions on aesthetic grounds. However, series described as de- parting from established conventions are likely to be perceived as com-

1299

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

mercially risky by advertisers, affiliates, and top network executives. Accordingly, we hypothesize:

HYPOTHESIS 6.-Series described as mixing or transcending estab- lished genres are less likely to be selected for the prime-time schedule than series described as conforming to those genres.

Again, we expect the Fox network to depart from the overall pattern. Because they rely on a less conventional programming strategy, we ex- pect claims about the reputation of a series' creator to have a smaller impact on scheduling decisions at Fox than at the other three networks.

Finally, the institutional context of network programming makes it almost impossible to predict which series will succeed or fail before they appear on the network schedule. Even though programmers are likely to rely on the reputations of program suppliers in making scheduling decisions, prior success does not guarantee audiences will embrace a new series. Accordingly, we hypothesize:

HYPOTHESIS 7.-Claims about the reputations of a new series' pro- ducer and cast, claims about the series' similarity to other shows or products from other media, and claims about its departures from estab- lished genres do not predict the series' subsequent commercial success.

FINDINGS

Content of Programmers' Claims

Network programmers' statements about their overall programming strategies are often at odds with the language they use to describe specific series in development. The overall strategy is typically framed as a quest for fresh ideas and new talent, while established reputations are empha- sized more heavily in descriptions of specific productions. For example, in describing program development at ABC, Stuart Bloomberg, vice president for development at ABC said, "We keep trying to branch out and bring new people to the network, like David Lynch and those kind of folks" (Bernstein 1991, p. F2). The article in the Los Angeles Times continues, "The plan, according to executives involved in the projects, is to bring new, less-expensive blood to the networks, while giving expo- sure to the cable producers." Some features of ABC's development slate were consistent with this claim. With 40 pilots in development, the net- work was looking at more potential new series than any other network. Furthermore, ABC had indeed commissioned a pilot from David Lynch as well as pilots from three cable networks (HBO, MTV, and Nickel- odeon).

But as table 1 reveals, nearly half of ABC's synopses framed new series in terms of the reputations of their writer-producers. For example, the description of ABC's Good and Evil begins, "In the spirit of Soap,

1300

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

TABLE 1

TRAITS OF SERIES PILOTS, 1991-92

Number %

Distribution of series pilots by network and genre: Networks:

ABC .............. ................................. 40 36 CBS ............. .................................. 27 24 NBC 22 20 Fox ............................................... 23 21

Total ................. .............................. 112 101* Genres:

Comedy .................. ............................. 52 46 Drama ............................................... 47 42 Reality ............................................... 13 12

Total ............................................... 112 100 Attributes of series pilots:

Linkages (not mutually exclusive): To successful producer ........................................... 52 46 To celebrity .......................... ..................... 26 23 To prior series . ............................................... 7 6 To other media ............................................... 17 15 No linkages .......................... ..................... 26 23

Producer links by network: ABC .............. ................................. 19 48 CBS ............. .................................. 11 41 NBC .............. ................................. 15 68 Fox ............................................... 7 30

Mixed or new genre ............................................... 24 21 In-house ............................................... 15 13 On fall schedule ................................. .............. 23 21 Midseason replacement ............................................. 15 13

NOTE.-N = 112. * Total exceeds 100% due to rounding error.

the acclaimed producing team of Paul Junger Witt, Tony Thomas, and Susan Harris have created a domestic comedy awash with intrigue and misalliances." An untitled pilot from Tom Miller and Bob Boyett is described as "in the spirit of Full House and Family Matters," two successful series from the same producers.6 Another untitled pilot for

6 While ABC described this pilot as a show that "springs from a common phenomenon in American society-the blended family," Daily Variety reframed this claim, calling the show a "'Brady Bunch'-type blended family comedy" (Lowry 1991b). In this case, the interpretive package supplied by the network was repackaged by the industry press by invoking the framing devices of subgenre and imitation.

1301

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

ABC is described as produced by "Jim Brooks, the creative force behind The Mary Tyler Moore Show." ABC's drama pilots included series iden- tified with such successful producers as Stephen Bochco and Bill Fin- kelstein (L.A. Law), David Jacobs (Dallas & Knots Landing), and Stephen Cannell, "in the timeless tradition of Rockford, Baretta, Columbo, and Kojak. "

The disparity between an overall programming strategy framed in terms of "new blood" and an emphasis upon producers' reputations in descriptions of specific series was even more pronounced at NBC. Bran- don Tartikoff, according to Daily Variety's coverage of NBC's develop- ment slate (Lowry 1991a, p. 1), told his audience of advertisers' represen- tatives to expect "a whole new NBC" in the fall. Tartikoff claimed that "the network would place less emphasis on 'pedigree' than promise, seeking to work more with unproven producers who may generate the new hits of the '90s as well as known commodities." In fact, more than two-thirds of NBC's series were described in terms of production teams associated with prior hit series (table 1). NBC's pilots were framed as coming from such "known commodities" as the producers of Married with Children, WKRP in Cincinnati, Barney Miller, St. Elsewhere, Eight Is Enough, and Quantum Leap.

Like his counterparts at ABC and NBC, Fox Entertainment Group President Peter Chernin emphasized innovative approaches, promising to bring back "excitement" to the one-hour form and to move away from "living room-based sitcoms" (Brennan 1991, p. 7). In contrast to the other networks, his network's framing of its pilots was more consis- tent with its claims about overall programming strategies. Less than one-third of the potential new series for Fox were linked to established writer-producers. As stated in hypothesis 4, Fox's propensity to make such reputational links was the lowest among the four networks (table 1).

Unlike executives from the other three networks, CBS's Entertainment President Jeff Sagansky refrained from emphasizing new and innovative approaches in outlining his network's overall programming strategy. In- stead, Sagansky stressed a strategy geared toward "franchise" program- ming with broad audience appeal (de Moraes 1991b). His claims at the March meeting echoed his earlier comments about a strategy to attract "brand-name" producers back to his network and borrow "pretested" concepts (Lippman 1990). Yet despite Sagansky's willingness to articu- late a conventional overall programming strategy, his network's descrip- tions of its pilots were actually less likely to include reputational links to established producers than were those of ABC and NBC (table 1).

Overall, table 1 provides strong support for our first hypothesis. Over three-fourths of the potential new series were framed in terms of at least one reputational linkage. Almost half were linked to an established

1302

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

TABLE 2

DETERMINANTS OF LINKAGE OF SERIES PILOT TO ESTABLISHED PRODUCER

Links to Links to Links to Mixed or New

Independent Variable Producer Celebrity Other Media Genre

Networks: ABC ............................... .73 2.13*** 2.16*** -.47

CBS ............................... .41 1.02 -2.44*** -2.02**

NBC ............................... 1.60** .82 -.53 -.77

Genres: Drama .............................. -.58* -.37 .05 .08

Reality .............................. -.76 -.07 -2.05*a -.49

Links to producer .................. . . . -.58 - 2.48*** -.69*

Constant .............................. -.49 - 2.11 .39 -.31

Log likelihood ....................... -72.56 -54.53 -35.42 -53.74

Test of network effects: X2 (3df) 6.86* 11.14** 11.90** 6.74*

NOTE.-N = 112. a Nondirectional test.

* P < .10. ** P < .05.

*** P < .01.

writer-producer, much more than any other kind of linkage. In contrast, less than one-fourth were linked to an established celebrity, just 15% were linked to products from other media, and only 6% were linked to other series. Moreover, only 21% were described as mixing or tran- scending established genres. Thus, in framing their pilots to advertisers, network executives placed more emphasis on their ability to "broker" relationships with the established members of the creative community than they did on any other kind of framing device.

Estimates from a series of logistic regressions, reported in table 2, address hypotheses 2, 3, and 4 regarding the content of programmers' claims. For each of the 112 network synopses, the probability of invoking a specific framing device is specified to be a function of the network and genre. Since we hypothesize that the ability to link a new series to an established producer reduces the need to make other kinds of claims, the variable capturing producer links is included as a predictor of the other framing devices.

Overall, the results in table 2 are supportive of our hypotheses regard- ing the primacy of the producer's reputation as a rhetorical strategy. Series linked to established producers are less likely than others to be

7Because only seven series were linked to prior series (i.e., earlier shows other than those produced by the series' creator), a separate logistic regression for this framing device was not estimated.

1303

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

framed in terms of successful properties from other media or to be de- scribed as mixing or transcending established genres. Although not statis- tically significant, the effect of producer links on the propensity to link a series to an established celebrity is in the predicted direction as well.

The findings in table 2 also support our hypothesis about the distinc- tiveness of the Fox network's strategy for describing new series. Since Fox is the reference category for the network variables, the results show that the Fox network is less likely than the other three networks to frame series in terms of the reputations of established writer-producers and stars. Also, as hypothesized, the Fox network is more likely than others to describe new series as mixing or transcending established genres. Each of the point estimates for the network coefficients is in the predicted direction, and for each outcome the hypothesis of no network differences is rejected with a three degree-of-freedom chi-square test.

Although we offered no hypothesis about Fox programmers' propen- sity to frame new series in terms of properties from other media, the results from table 2 show that they were much more likely than their counterparts from the other networks to use this rhetorical strategy. Thus, even as the Fox programmers describe their strategy as defying the conventions of prime-time television, they seek to reassure commercial constituencies by making associations between their new series and suc- cessful films, plays, and novels.

Consequences of Programmers' Claims

To examine the consequences of programmers' claims, we evaluate whether or not a pilot is selected for the prime-time network schedule as a function of network, linkages, and genre. We examine two outcomes: (1) whether or not a series is selected for the network schedule (either for fall or as a midseason replacement); and (2) among those selected, whether a series is scheduled for fall or held in reserve as a midseason replacement. Hypothesis 5 is addressed by the relative impact of linkages to producers, starts, prior series, and other media on the likelihood of selection. Hypothesis 6 is addressed by the impact of framing with respect to mixing or transcending genres on scheduling decisions.

We control for network since the probability of selection will be influ- enced by differences across networks in the number of pilots commis- sioned relative to the amount of time to be filled with new series. An interaction term is included to test whether the impact of framing with respect to established producers is less consequential for scheduling deci- sions at the Fox network. This expectation is based on two factors. First, as described above, the Fox network has framed its programming strat- egy as departing from conventional network practice. Second, established

1304

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

producers are less likely to have produced previous series for the Fox network, and therefore the ties between programming executives and the community of established producers are probably weaker at Fox than at the other networks.8

Finally, we include a binary variable for in-house producton, denoting whether a pilot is produced by the network's production division (e.g., a pilot from ABC Productions for ABC). Relaxation of the Financial Interest and Syndication Rules allowed the networks to produce more of their own programming in 1991-92 than had been the case during the previous two decades. Networks may have an economic incentive to favor their own pilots over those from independent producers, since fu- ture earnings from syndicating in-house series would return to the network.9

The dependent variable for results reported in table 3 is whether a pilot is selected for either the fall schedule or as a midseason replacement versus rejected altogether."0 Results reported in table 3 support our hy- pothesis regarding the primacy of producer linkages for scheduling deci- sions (hypothesis 5). The single best predictor of whether a pilot is se- lected for the prime-time schedule is whether or not it has been described as originating with an established producer of a prior successful series." The negative interaction term is not statistically significant, suggesting that despite their claims regarding programming strategy, the Fox net- work is about as likely as the other networks to schedule series that have been described as originating from successful producers.

Claims made in March about innovative pilots may have reassured critics and members of the creative community. However, as expected (hypothesis 6), pilots described as deviating from established genres were less likely than other series to be selected for the network prime-time

8 Our original specifications also included an interaction term for the "mixed or new genre" variable and the Fox network to test the hypothesis that the Fox network was more likely than the other networks to select unconventional programming. However, there were too few cases to obtain a stable estimate of this interaction effect. 9 When program development and scheduling decisions were made for 1991-92, the ultimate outcome of proposed changes in government regulation was (and still is) largely uncertain. As a result, the effect of the "in-house" variable in our models provides only weak evidence on the issue of a network bias toward in-house produc- tions. 10 Slightly stronger relationships are detected when the dependent variable is defined as whether or not a series is selected for the fall schedule. " The logistic coefficient of .89 implies that series framed in terms of linkages to established producers are substantially more likely to be selected by the networks than series not associated with successful producers. Evaluated at the mean, the probability of selection increases from .33 to .55 when a series has been described as originating from a writer-producer with a proven track record in network prime-time television.

1305

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

TABLE 3

DETERMINANTS OF LIKELIHOOD THAT A SERIES PILOT IS SCHEDULED TO APPEAR

ON THE 1991-92 NETWORK PRIME-TIME SCHEDULE

Independent Variable Logistic Coefficient

Networks: ABC .-.28 CBS .-.31 NBC .-.56

Linkages: Producer. .89** Fox x producer. -.28 Prior series ..11 Celebrity .................................. .33 Other ..22

Genres: Drama ..03 Reality ..32

Mixed or new genre. -.75* In-house ..21 Constant .- .86 Log likelihood .......... -68.56

NOTE.-N = 112. * P < .10.

** P < .05.

schedules. Indeed, according to the logistic coefficients in table 3, the disadvantage of departing from established genres (- .75) almost exactly offsets the advantage of linkage to a successful writer-producer (.89).

Results reported in table 4 reveal factors that differentiate between series that are held in reserve and those that premiere in the fall, among the 38 that were selected by the networks for their prime-time schedules. There are several reasons why networks are more likely to schedule series that pose the greatest commercial risks as midseason replacements. First, from the beginning of the season, networks attempt to build momentum among viewers for each evenings' prime-time offerings. A single ratings failure can adversely affect viewership of the series that follow. Accord- ingly, networks are more likely to begin their season with new series they perceive viewers will readily accept. Second, by scheduling an unusual or unconventional series in midseason, the network has more time to use its promotional efforts to situate viewers to the "preferred reading" of the new show. With so many new series debuting in the fall, viewers might be less willing to sample those that are not easily situated within established genres. Finally, an order for a midseason replacement typi-

1306

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

TABLE 4

DETERMINANTS OF LIKELIHOOD THAT A SERIES FOR 1991-92 IS SCHEDULED

AS A MIDSEASON REPLACEMENT

Independent Variable Logistic Coefficient

Fox network ................................................................... 2.20

Linkages: Producer ................................................................... -2.76** Celebrity ................................................................... .89 Other .................................................................... -2.04

Genres: Drama ................................................................... 3.30**a

Reality ....................................................................-2.. 50

Mixed or new genre ........................................................... 3.72**

In-house ................................................................ ... -2.14*

Constant ................................................................... -1.04

Log likelihood .................................................................. - 15.94

NOTE.-N = 38. a Nondirectional test.

* P < .10. ** P < .05.

cally requires less of a financial commitment from the network. The network can take a chance on more unconventional programming by ordering just six episodes rather than the usual commitment of at least 13. Moreover, if the new series fails to find an audience quickly, it can be moved to a slot on the prime-time schedule that the network has already conceded to the competition.

Results reported in table 4 provide modest support for our expecta- tions. The logistic coefficients are quite large, but with so few cases they have relatively large standard errors.12 Series that are framed as de- parting from established genres are more likely to debut in midseason, as are those that have not been linked to experienced producers. Thus, network programmers apparently believed that series they had described as innovative or unconventional or originating from unproven producers would have a more difficult time building an audience.

In short, claims made in March about links to established producers had real consequences for the scheduling decisions announced in May,

12 To conserve degrees of freedom, the three binary variables for differences among the four networks have been replaced with a single variable differentiating Fox from the others. There were not enough cases in the sample to compute a reliable estimate of the effect of linkages to prior series, so that variable has been omitted from the model.

1307

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

but genre and imitative claims by programming executives were largely irrelevant to those decisions. Despite claims by three of the four networks that their programming strategies would place a premium on innovative fare, pilots described as departing from conventional genres were actually less likely than others to be selected for the network prime-time sched- ules. If series described as unconventional were selected, they were typi- cally held in reserve as midseason replacements. Following the rather spectacular ratings failure of such unconventional fare as Twin Peaks and Cop Rock during the previous season, it is perhaps not surprising that network programmers favored series that could be framed in terms of producers with proven track records over those that could be described as challenging the conventions of established genres.

Our final hypothesis, that programmers' claims would have no measur- able effect on the commercial viability of new series, was tested in two ways. First, for series that were broadcast during the 1991-92 season, we examined whether our measures of how a series is framed predicted its average ratings for the season. Second, we examined whether the same measures predicted whether or not the series was selected to return for a second season in 1992-93.

Both tests supported our hypothesis; that is, we found no significant relationship between how a series is framed and measures of commercial success. In t-tests of an individual coefficient and in joint tests that all coefficients are zero, the null hypothesis of no effect could not be rejected (we have not compiled tables to report these nonsignificant results). In contrast, the association between ratings and the likelihood that a series will return for a second season is significant and substantial.

In sum, working in highly institutionalized contexts, programmers use the framing devices of reputation, imitation, and genre to construct inter- pretive packages that rationalize and legitimize their decisions. Their discourse emphasizes reputation over innovation and creativity, typically calling attention to the reputations of writer-producers with proven track records. The successful decoupling of commercial and aesthetic assess- ment is demonstrated by the opposite effects of claims about producers and claims about departures from established genres on scheduling deci- sions. Specifically, series described as originating from established pro- ducers are more likely to be selected for the prime-time schedule, and those described as defying conventions are less likely to be selected. Though network programmers appear to rely more heavily on established producers than on new talent, there is no evidence that the series originat- ing from the more reputable producers have greater commercial viability. Thus, it does indeed appear to be true that in the highly institutionalized context of network television all hits are flukes.

1308

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

DISCUSSION Overall, what is most striking about our findings is the importance of claims about linkages to established writer-producers in the program development process. For the 1991-92 season, the framing device of linking a potential new series to established writer-producers is invoked much more frequently than any other rhetorical strategy, and it is the only one positively associated with the likelihood of a series being selected for a network's prime-time schedule. Other framing devices appear to be strictly symbolic, designed to shape perceptions of constituencies with conflicting assessment criteria in a context characterized by uncertainty and ambiguity. In an industry where all hits are flukes, decision makers use rhetorical strategies to reassure others that their decisions are ratio- nal, appropriate, and legitimate.

If anything, our statistical findings understate the importance of actual links (as opposed to claims about links) between networks and established writer-producers. Not reflected in our statistical analysis are several new series from veteran producers selected for the network schedules without going through the normal development process of pilot production and evaluation. For example, Flesh 'n' Blood, from the producers of Cheers, was scheduled by NBC based on a six-minute presentation to network executives (de Moraes 1991c), while Gary David Goldberg's Brooklyn Bridge was sold to CBS on the basis of a script alone (Lowry 1991b).

The commitments to these shows reflect a recent change in the network development process. The networks are increasingly making multiyear, multiseries commitments to the most sought-after producers. For exam- ple, the network may commit to ordering a specific number of episodes of a series before seeing a pilot, or even a script, or it may agree to pay a financial penalty to the producer if the network fails to order a fixed number of episodes (CBS Inc. 1990, p. 26). At least six of the series in our data, all linked to established producers, had such commitments from the networks at the time the development slates were announced in March.

As the level of risk and uncertainty facing the industry increases (Marich 1991), these kinds of arrangements are likely to proliferate. Reli- ance on established writer-producers is likely to intensify, which should be reflected in the discourse framing the program development process in the future. In terms of the defining characteristics of a centralized brokerage system, (1) it is becoming more difficult to predict the success of a new series a priori, (2) pressures for commercial success are increas- ing, and (3) network programmers are becoming more dependent on contributions of the creative personnel who supply series. As a result, actively managing the reception of new series through carefully organized

1309

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

rhetorical strategies will become increasingly important, but doing so successfully will become more difficult.

CONCLUSIONS

Our study provides an explanation for the organization and content of discourse that introduces new cultural objects marketed to a mass audi- ence. Our research extends Griswold's (1987) methodological framework for the sociology of culture by emphasizing how social agents actively manage the reception of new cultural objects through rhetorical strate- gies. Our analysis of prime-time television suggests that three factors shape the content of rhetorical strategies for introducing new cultural objects: (1) the degree of centralization in social arrangements that medi- ate the relationship among creators, audiences, critics, and commercial constituencies; (2) the degree to which the commercial viability of a new product can be unambiguously evaluated based on measurable features of the clinical object; and (3) the degree to which commercial and critical assessment of a new cultural object are successfully decoupled from one another. Our case study of prime-time television applies to a context in which the brokered relationship between creators and business interests is highly centralized, commercial and critical success cannot be predicted in advance, and critical and commercial assessment are decoupled from one another. We found that in such circumstances, decision-making bro- kers use linguistic framing devices of reputation, imitation, and genre to reassure commercial and creative constituencies that their actions are appropriate, legitimate, and rational. Because of the primacy of commer- cial viability in evaluating success, claims regarding the reputation of creators are emphasized over other rhetorical strategies.

Variation along these three dimensions should have consequences for the discourse used to introduce new cultural objects. For example, when the brokerage between creators and business interests is less centralized and success is more predictable, less emphasis should be placed on repu- tational claims about creators. In such circumstances, claims about cre- ators' reputations should also be less strongly associated with access to channels of distribution. Furthermore, when critical and commercial as- sessments are more tightly coupled, linguistic claims about aesthetic qual- ities of the product should receive greater emphasis relative to claims about reputation and imitation.

The validity of our explanatory scheme for understanding the manage- ment of the introduction of new cultural objects can be assessed empiri- cally in two ways. One is through historical analysis of the discourse in a specific culture industry. For example, in the early years of network prime-time television, critical acceptance should have been more impor-

1310

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

tant for establishing the legitimacy of the new medium as a profit-making enterprise (Boddy 1990). Accordingly, we should find claims about aes- thetic quality to be more common in programmers' discourse about new prime-time series in the late 1940s and early 1950s. Similarly, if attacks on the values imparted by contemporary television begin to seriously threaten the legitimacy of the industry, we can expect programmers' claims to increasingly emphasize the positive social values of the series they broadcast.

A second way to assess our approach to the management of the intro- duction of new cultural objects is through comparative analysis across culture industries. Prime-time television, feature film, popular fiction, and popular music differ in the organization of production, in levels of risk, uncertainty, and ambiguity, and in the relative importance of criti- cal reception for commerical success. They also differ in features of the product life cycle such as the length of time between creation, production, and distribution and the duration of popularity with audiences. These industry traits should be systematically related to the kinds of rhetorical strategies invoked by brokers and their effectiveness for shaping the re- ception of new cultural objects by business interests, critics, and audi- ences.

In sum, we have analyzed how language is used to manage uncertainty and ambiguity in a highly centralized and institutionalized culture indus- try. By specifying how institutional context and industry structure shape interaction among those- engaged in the production, distribution, and consumption of popular culture, we have demonstrated the importance of three phenomena. First, we show that when confronted with high levels of amibiguity and uncertainty and conflicting assessment criteria, it is incumbent upon decision makers to develop rhetorical strategies that provide legitimate accounts of their actions. Second, we demonstrate that the content of that discourse will be organized around widely shared categories and symbols that have meaning to varied constituencies. Third, we show that in centralized culture industries, brokers' rhetorical strategies have consequences for creators' access to product markets and audiences. Others have shown how the structure of industries and mar- kets constrain innovation in culture industries (Peterson and Berger 1975; DiMaggio 1977). Our analysis suggests that the strategic use of language and categories to manage commercial and creative interests can also have a causal impact on the sources and diversity of cultural objects reaching the marketplace.

REFERENCES

Baker, Wayne E., and Robert R. Faulkner. 1991. "Role as Resource in the Hollywood Film Industry." American Journal of Sociology 97:279-309.

1311

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

American Journal of Sociology

Becker, Howard S. 1982. Art Worlds. Berkeley and Los Angeles: University of Cali- fornia Press.

Benford, Robert D. 1993. "Frame Disputes within the Nuclear Disarmament Move- ment." Social Forces 71:677-701.

Bernstein, Sharon. 1991. "Strange Bedfellows." Los Angeles Times (May 4), pp. Fl, F8.

Bielby, Denise D., and William T. Bielby. 1993. "The Hollywood 'Graylist'? Audi- ence Demographics and Age Stratification among Television Writers." Pp. 141-70 in Creators of Culture, edited by Muriel G. Cantor and Cheryl Zollars. Vol. 8 of Current Research on Occupations and Professions. Greenwich, Conn.: JAI Press.

Bielby, William T., and Denise D. Bielby. 1990. "Producing Prime Time: Organiza- tional Concentration and the Diversity of Network Television Programming." Pa- per presented at the annual meeting of the American Sociological Association, Washington, D.C., August.

Block, Ben Alex. 1990. "Twenty-first Century Fox." Channels 10 (January): 36-40. Boddy, William. 1990. Fifties Television: The Industry and Its Critics. Urbana: Uni-

versity of Illinois Press. Brennan, Steve. 1991. "FBC Strengthens Program Reserves with 23-Pilot Slate."

Hollywood Reporter (March 21), pp. 1, 7. Cantor, Muriel G., and Joel M. Cantor. 1992. Prime-Time Television: Content and

Control, 2d ed. Newbury Park, Calif.: Sage. CBS Inc. 1990. "Comments of CBS Inc." In the Matter of Evaluation of the Syndica-

tion and Financial Interest Rules. Federal Communications Commission, Docket no. 90-162. June 14.

Channels. 1990. "Primetime's Price Tag." Channels 10 (September 10): 50-51. de Moraes, Lisa. 1991a. "ABC Pilot Slate Full of Big Names and Cable Concepts."

Hollywood Reporter (March 20), pp. 1, 6. . 1991b. "New CBS Pilots Long on Machismo, Women's Humor." Hollywood

Reporter (March 20), pp. 1, 5. . 1991c. "NBC to Bow Six Comedies, Two Dramas in Fall Shake-up." Holly-

wood Reporter (May 21), pp. 1, 78. DiMaggio, Paul J. 1977. "Market Structure, the Creative Process, and Popular Cul-

ture: Toward an Organizational Reinterpretation of Mass Culture Theory. " Journal of Popular Culture 11:433-5 1.

DiMaggio, Paul J., and Walter W. Powell. 1983. "The Iron Cage Revisited: Institu- tional Isomorphism and Collective Rationality in Organizational Fields." American Sociological Review 48:147-60.

Du Brow, Rick. 1991a. "Humanity Loses to Reality." Los Angeles Times (March 30), pp. Fl, F12.

. 1991b. "Fox Network Goes Out on a Limb Again." Los Angeles Times (May 22), pp. Fl, F7.

Faulkner, Robert R. 1983. Music on Demand: Composers and Careers in the Holly- wood Film Industry. New Brunswick, N.J.: Transaction.

Frank, Betsy. 1991. On Air: Primetime Program Development, 1991-92. New York: Saatchi & Saatchi Advertising.

Gamson, William A. 1988. "Political Discourse and Collective Action." Pp. 219-44 in From Structure to Action: Social Movement Participation across Cultures, edited by Bert Klandermans, Hanspeter Kriesi, and Sidney Tarrow. Greenwich, Conn.: JAI Press.

Gamson, William A., and Andre Modigliani. 1989. "Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach." American Journal of Sociology 95:1-37.

Gitlin, Todd. 1983. Inside Prime Time. New York: Pantheon.

1312

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions

Prime-Time Programs

Goldberg, Robert. 1991. "TV: Winners That Were Losers." Wall Street Journal (June 24), p. A9.

Griswold, Wendy. 1987. "A Methodological Framework for the Sociology of Cul- ture." Pp. 1-35 in Sociological Methodology 1987, edited by Clifford C. Clogg. Washington, D.C.: American Sociological Association.

Hirsch, Paul M. 1986. "From Ambushes to Golden Parachutes: Corporate Takeovers as an Instance of Cultural Framing and Institutional Integration." American Jour- nal of Sociology 91:800-37.

Jepperson, Ronald L. 1991. "Institutions, Institutional Effects, and Institutionalism." Pp. 143-63 in The New Institutionalism in Organizational Analysis, edited by Walter W. Powell and Paul J. DiMaggio. Chicago: University of Chicago Press.

Lang, Kurt. 1958. "Mass, Class, and the Reviewer." Social Problems 6:11-21. Lippman, John. 1990. "Running CBS' New Game Plan." Los Angeles Times Calen-

dar (December 30), pp. 5, 77-79. Lowry, Brian. 1991a. "NBC, FBC Pilots Fall into Place." Daily Variety 231 (March

21): 1, 18-19. . 1991b. "Pilots Lighted for Fall." Daily Variety 231 (March 20): 1, 14, 18.

March, James G., and Johan P. Olsen. 1976. Ambiguity and Choice in Organizations. Bergen: Universitetsforlaget.

Marich, Robert. 1991. "Networks Suffer 61% Profit Drop in 5-Year Period." Holly- wood Reporter (May 24), pp. 1, 8.

Meyer, John W., and Brian Rowan. 1977. "Institutionalized Organizations: Formal Structure as Myth and Ceremony." American Journal of Sociology 83:340-63.

Montgomery, Kathryn C. 1989. Target Prime Time: Advocacy Groups and the Struggle over Entertainment Television. New York: Oxford University Press.

Peterson, Richard A., and David G. Berger. 1975. "Cycles of Symbol Production: The Case of Popular Music." American Sociological Review 40:158-73.

Powell, Walter W., and Paul J. DiMaggio. 1991. The New Institutionalism in Organi- zational Analysis. Chicago: University of Chicago Press.

Snow, David A., and Robert D. Benford. 1988. "Ideology, Frame Resonance, and Participant Mobilization." Pp. 197-217 in From Structure to Action: Social Move- ment Participation across Cultures, edited by Bert Klandermans, Hanspeter Kriesi, and Sidney Tarrow. Greenwich, Conn.: JAI Press.

1313

This content downloaded from 131.179.44.100 on Thu, 05 Nov 2015 22:00:22 UTC All use subject to JSTOR Terms and Conditions