![]()
FOUR
Fatty Arbuckle and the Creation of Public Attention
Nothing catches public attention like a “juicy” scandal. At first glance, scandal seems to exemplify a strong constructionist perspective on social problems. A scandal appears suddenly, from the ether—a chance occurrence that moral entrepreneurs seize for their own ends. Yet, it is precisely this apparently random selection process that should give pause. Is the creation of scandal and social problems as “idiosyncratic” as it often appears?
Ironically, much constructionist theory takes social conditions into account: examining conditions that produce successful moral entrepreneurs and effective claimsmaking, while problematizing or dismissing the conditions that produce the problems themselves: looking forward from the first claim, not looking back (Best 1989; Miller and Holstein 1989; Schneider 1985). I argue that a constructionist model can be enriched when a concern with the existence of problematic conditions is added to the understanding of the dynamics of claims. Such a view is congruent with current approaches to the relationship between structure and agency.
Scandals1 are analytically distinct from social problems, although they often connect to emerging social problems and, when they capture public attention, may influence (magnifying, concretizing, transforming, or minimizing) the development of the social problem they are said to represent. The depiction of the scandal comes to symbolize the problem for the public, and, thus, the response to the scandal shapes the response to the social problem.
In speaking of scandal, I refer to public and media perceptions of specific behaviors and events that deserve moral opprobrium. This might suggest that every felony is a scandal. Such is clearly not the case. Scandal also must be situated within an institutional structure (e.g., politics, business, the media) and, more significantly, must be said to characterize that structure. Put simply, scandal represents a breach in institutional trust: the particular actions come to characterize an institution. Watergate was a scandal, not because individuals committed moral infractions, but because it exemplified an immoral order, and, in so doing, created a center of public attention and discourse (Lang and Lang 1981; Schudson 1992). Even scandals linked to private doings of politicians or ministers (e.g., Bill Clinton or Jimmy Swaggart) gain significance because of what they say about trust in institutions (Jenkins 1996).
The power of scandal to focus public attention by questioning institutional legitimacy justifies reconsidering the dismissal of societal conditions and institutional strain in the development of social problems (Alexander 1988). The agency of scandalous actors and claimsmakers is set within obdurate structural conditions: agency and structure define each other (Giddens 1984; Sewell 1992). While claimsmakers play a heavy role in the creation of social problems, scholars in social problems theory and collective behavior routinely ignore structural conditions of a social order, only referring to these conditions obliquely. I reject the strong or strict version of constructionism (Ibarra and Kitsuse 1993; Woolgar and Pawluch 1985) that denies any structural reality. I intend to expand the weak or contextual version (Best 1989, 1993; Goode 1994) by arguing that the structural conditions of a social order contribute to how social agents perceive and respond to social problems (Eitzen 1984:10; Jones, McFalls, and Gallagher 1989). Context, while processual, is also grounded in structural relationships. Social structures create the specific contexts in which interaction and discourse occur. Examining scandal allows us to analyze social problems through a “cautious naturalism” (Gubrium 1993:100) grounded on recent theoretical attempts to link structure and action.
While much important recent theoretical work examines relations between agency and structure (Fine 1992), I draw upon a classic, if neglected, analysis that provides a similar model in which structure and action continually recreate each other in dynamic tension. Specifically, I incorporate Neil Smelser’s (1962) value-added model into an approach based on cautious naturalism. Smelser’s model, once a highly influential approach to collective behavior,2 has been routinely dismissed (Smelser 1969) by subjectivists who reject the possibility of specifying “objective” structural conditions and who suggest that “reality” is up for grabs: an exercise in ontological hopelessness. Because of its seemingly objectivist and functional view of structure and strain, many sociologists neglect other elements of the model that are fully consistent with constructionism. Smelser’s perspective is a particularly useful addition to a weak constructionist view because it incorporates structure (both cultural and institutional) and agency into the same model—a model whose temporal sequence recognizes that the establishment of collective action is necessarily processual. Unfortunately, Smelser never presented a detailed empirical case study that would have justified his model, was ambivalent as to whether social forces were objective, and never provided techniques for operationalizing structure.3
Consistent with Smelser’s value-added approach (and with cautious naturalism) we must confront the legitimacy of consensually recognized (“objective”) social conditions, while claiming that these conditions gain power through collective interpretations. Institutional and economic orders are powerful, even if their effects are mediated by collective sense-making. The recognition that we give to effects of claimsmakers should also be given to effects of social institutions. The creation and development of a social problem depend on the interests of claimsmakers (moral entrepreneurs), operating within a body of discursive traditions (resonant rhetoric [Schudson 1989]), constrained by an obdurate social reality.
Given its eclipse, I briefly summarize Smelser’s value-added approach. Six determinants explain collective behavior; each builds on previous ones in this temporal sequence: structural conduciveness, structural strain, generalized belief, precipitating events, mobilization, and social control (what I term institutional response). First, structural conduciveness explains how the organization of social institutions permits (or dampens) collective action (Smelser 1962:384; Galliher and Cross 1983:21). Some organizational forms are structurally more conducive to collective action than others. In conditions of structural conduciveness, structural strain (the second determinant) may be present. Strain refers to the routine impairment of relations among institutional actors and the consequent tensions among them. In interactional terms, strain refers to the friction or breakdown in negotiations among institutional actors (Farberman 1975); strain is reflected in alternative models of how systems should operate.4 Without some measure of structural strain it is unlikely that a problem’s claim will have sufficient rhetorical resonance for moral entrepreneurs to generate public attention. Third, a generalized belief must crystallize and be publicized (Smelser 1962:384). As emergent norm theory asserts, people facing ambiguity assess their situation by answering the question, “What is happening here?” (Turner and Killian 1957; Shibutani 1966). Fourth, a precipitating event or triggering mechanism (Fine 1979) focuses public attention. Fifth, a dramatic and relevant belief does not produce action by itself, but individuals must be mobilized for action (see McPhail 1991). Moral entrepreneurs transform attention into a demand for action. The public must be organized by those with interests in doing so. Finally, rival groups put forth alternative explanatory claims in a struggle for control of the meaning of an event. Facing attack, institutions respond, and when backed by incentives or disincentives for action, their response constitutes social control, in that one group’s vision determines the actions of others.
The first three components specify the antecedent conditions necessary for collective action. The final three reflect the construction of a social problem. Together these six components enrich social problems theory by connecting social conditions to the claimsmakers’ activities. The challenge for constructionists is to take seriously the reality of the structure and organization of a society in which meaning creation occurs, while not ignoring the role of social agents. The power of social conditions, strain, and generalized beliefs denies that social problems are created from “thin air.” In organizing the paper around Smelser’s six components, I do not embrace the underlying functional (Parsonian) theory, but argue that this model permits constructionists to incorporate social conditions in the analysis of claimsmaking.
To examine a constructionist model of public attention that takes structure seriously, I focus on the trial of comedian Roscoe “Fatty” Arbuckle (1887–1933), who was, along with Charlie Chaplin, one of the most popular comedians of his day (see Lewis 1990). A young woman’s death after a party that Arbuckle hosted in 1921 led to his being accused of murder and subsequently tried on a charge of manslaughter. After two hung juries, Arbuckle was acquitted by a third jury. I use the intense public interest in this case to explore how structural conditions, coupled with the actions of a set of moral entrepreneurs and agents of social control, shaped how the case was given meaning.5
THE CASE
On Labor Day Weekend of 1921, Roscoe Arbuckle and two friends (director Fred Fischbach and actor Lowell Sherman) motored to San Francisco.6 Arbuckle, a top star at Paramount, the largest American film company, had just completed shooting three feature films and had six feature films playing in Los Angeles and New York that week. On Monday, September 5, the men decided to hold a small party in their luxury suite at the St. Francis Hotel. Word of the party spread and several people connected with the motion picture industry arrived, including a young actress and model, Virginia Rappe. Like many parties of the period, this one was generously lubricated with bootleg whiskey and gin. Prohibition had only recently taken effect, and both Hollywood and San Francisco were “wet.”
During the party Rappe became ill, started to scream, and tore off her clothing. Attempts to calm her (by Arbuckle and others) proved in vain, and eventually the hotel manager was called. Rappe was moved to a separate room, and a doctor sent for. Rappe’s condition did not improve, and eventually she was transferred to a sanitarium where, on Friday, September 9, 1921, she died from complications of a ruptured bladder caused by peritonitis. No physical evidence suggested that a murder or rape had occurred.
Shortly after Rappe fell ill, a friend and fellow partygoer, Maude Bambina Delmont, accused Arbuckle of having “jumped on Virginia and crushed her bladder.”7 This statement, coupled with some ambiguous remarks that Rappe made while semi-conscious, led San Francisco District Attorney Matthew Brady to charge Arbuckle with murder.
Despite Brady’s attempt to bring a murder charge, Judge Lazarus of the San Francisco Police Court reduced the charge to manslaughter on September 28.8 The first trial, which began on November 18, ended on December 4 with a hung jury voting 10 to 2 for acquittal. Perhaps Arbuckle’s attorneys became overconfident: the hung jury in the second trial (January 11–February 4, 1922) voted 10 to 2 for conviction. In the third trial (March 13–April 12, 1922) Arbuckle’s lawyers defended him vigorously, and after minutes the jury returned with an acquittal, remarking:
Acquittal is not enough for Roscoe Arbuckle. We feel that a great injustice had been done him. We feel also that it was only our plain duty to give him this exoneration, under the evidence, that there was not the slightest proof adduced to connect him in any way with the comission of a crime. (Young 1994:71)
Less than a week later Will Hays, the head of the Motion Picture Producers and Distributors Association, banned Arbuckle from appearing in films. It was not until 1932 that Arbuckle again appeared on the screen.
It is easy to suggest that the Arbuckle case represented a clear instance of social construction, in that various figures felt that it was in their interest to become moral entrepreneurs. This interpretation, while accurate, ignores social conditions that made the Arbuckle case so resonant. While the party at which Rappe fell ill was a “random perturbation,” it occurred in the context of social strain.
STRUCTURAL CONDUCIVENESS
Every society is constituted through institutional arrangements, linking politics, culture, and economics in recursive relations (Giddens 1984). Not even the most ardent constructionist argues that all behavior is locally produced. Institutional arrangements matter, even if they are not fully causal, separate from agents’ mediation and choices (Sewell 1992). The problem is how to account for these structural arrangements: to recognize their reality while seeing their interpretation “in play.” To describe a complex society is possible, as long as one accepts historical “plausibility,” and does not separate structural arrangements from their interpretation, a concern in all historical analyses. The choices of which social conditions to include, and how to define them, is itself a matter of construction, and like all analysts, I draw on those conditions I consider most relevant to the issues at hand, based on my reading of the texts of the period. Admittedly all empirical evidence is constructed in some measure, but where there is historical consensus, these constructions can be treated (cautiously) as having an isomorphic relationship with institutional realities. For this analysis I focus on the organization of the communications...
