In American sociology, interest in the media exploded during the 1920s, remained as a visible presence in the flagship journals until the end of the 1950s, and then seemed to disappear during the 1960s. There was a renewed interest in the media during the 1970s and early 1980s, but by 1990 this movement had collapsed.Footnote 1 What happened? How could an institution that is touted in all introductory textbooks as one of the most influential in American society simply disappear, on two separate occasions, as an object of analysis in the major sociology journals? This is the central question that this article seeks to answer.

My general argument is that media sociology is most vibrant when it is attempting to understand the dominant cultural structures that shape the public sphere. This kind of attempt emerged first and most notably during the 1920s and 1930s, with Robert Park’s work on newspapers and the cultural power of the press. It continued, albeit in a slightly different form, with a series of 1950s studies of propaganda and the press. Sociological interest in the media declined in the 1960s, for reasons that Elihu Katz discusses in an article in this issue, but also because of a turn away from the earlier focus on culture and the public sphere. There was a resurgence of interest during the late-1970s and early 1980s, with the newsroom ethnographies and organizational studies of the press.

While the 1970s and 1980s newsroom studies were interesting, they continued the general analytical shift away from culture and the public sphere, in ways that ultimately demonstrated that there was nothing particularly distinctive or unusual about media discourse. In effect, these studies argued that media outcomes were caused by other social factors – social networks, organizational routines, etc. – and that the media was therefore just a platform where other social processes operated. From this analytical vantage point, it is perhaps not surprising that sociological interest shifted away from the media, and back toward those social structures and processes that presumably had more of an independent causal impact: demography, organizational ecology, resource mobilization, and elite reproduction. In order to bring media sociology back to its previously prominent position, I want to suggest a return to research that investigates the relationship between culture, media, and the public sphere. Indeed, as Rod Benson’s article in this issue suggests, there is some evidence that such a return may already be underway. In order to maximize the disciplinary impact of these new developments, however, I want to suggest that a clearer connection be made to the earlier work of the Chicago School, and to Robert Park in particular.

Robert Park, the Chicago School, and the Golden Age of Media Sociology

While Robert Park is probably best known as a scholar of race, urban ecology, and collective behavior, he also maintained a serious and sustained interest in the mass media. This is not surprising, perhaps, given that Park spent ten years as a newspaper journalist before entering graduate school in 1898. Park’s scholarly writing on the press began shortly after he joined the faculty of the University of Chicago’s sociology department in 1914, and continued even after his formal retirement from the university. All told, Park published seven articles about the press in AJS between 1923–1941, in addition to an important book about the immigrant press (Park 1922).

Park’s media sociology covered two related themes. Initially, the focus was on news and the power of the press. While the conceptual vocabulary was different, this work was clearly engaged with questions that today would invoke the language of media and public sphere. In “The Natural History of the Newspaper”, Park argued that the modern newspaper was initially made possible by the growth of cities, which expanded the size of the reading public and provided “an impersonal account of manners and life” (Park 1923: 277). Park argued that one of the most significant and telling differences between urban and rural life was the prevalence of daily newspaper reading in the cities; even within metropolitan regions, Park was able to show, there was a pattern of declining newspaper reading as one moved outward in concentric circles away from the urban core (Park 1929: 62).

By the time he published “News and the Power of the Press” in 1941, Park had a fully-developed theory of media and the public sphere. In essence, Park distinguished between an elite public sphere and a popular public sphere, and linked each to a specific part of the newspaper. The elite public sphere consisted of politicians, party leaders, and intellectuals, all of whom argued about how the events of the day supported or challenged specific policy proposals. For these elites, the editorial page was the most important part of the paper; having its origins in the letter to the editor – “in which men interested in political matters sought to express their opinion in regard to debatable measures proposed or undertaken by the government” (Park 1941a: 8) – the editorial was designed to interpret the news through a consistent and coherent political philosophy. During periods of political stability, when there was more likely to be party discipline and when it was much easier for editorial columnists “to maintain contact with events as recorded in the news” (Park 1941a: 9), then Park thought that the editorial page and the elite public sphere could maintain significant influence and control over political society.

On the other hand, during periods of change and social transformation, Park’s theory of press and public sphere pointed to the importance of the news column and the way that it organized political discussion among ordinary citizens. Park argued that the news editor was not constrained by the demands of philosophical consistency, but rather by the goal of expanding the circle of readers. By publishing news items that would be of interest to the widest range of readers, the news editor helped to increase the number of people who were aware of what was going on in their society. And this allowed for the development of public opinion formation.

Park continued to develop this argument in more detail in “News as a Form of Knowledge” (Park 1940). Park argued that news was different from other forms of knowledge, in that news was defined by its transient and ephemeral quality (Park 1940: 676). The key to news’s liveliness, as compared to history, was that it was reported as a series of isolated incidents, rather than a teleological sequence.Footnote 2 The reason for this was that news was a public commodity; it did not become news through any individual reading, but rather through the fact that it was part of a public communication. Because news treated events as if they were isolated, this required individuals to talk about them; from those conversations public opinion emerged:

“The first typical reaction of an individual to the news is likely to be a desire to repeat it to someone. This makes conversation, arouses further comment, and perhaps starts a discussion….. The clash of opinions and sentiments which discussion invariably evokes usually terminates in some sort of consensus or collective opinion – what we call public opinion. It is upon the interpretation of present events, i.e., news, that public opinion rests.” (Park 1940: 677)

This description of news and public opinion rejected the image of the isolated individual, who reads the paper alone, contemplates the meaning of current events, and develops an individual opinion about those events that is carried into public encounters. Instead, Park’s media sociology provided a thoroughly social and public account of the news. If an item of news is interesting enough for an individual to reflect upon it, then this reflection will usually be accompanied by – indeed, often preceded by – the desire to discuss the news item with another individual. And so the opinions that individuals develop from news items are almost always already public opinions, in the sense that they emerge from discussions with others. As Park (1941a: 6) argued,

“There can be no public opinion in regard to any political action unless the people who constitute the public know, in a general way at least, what is going on….. For public opinion, as it ordinarily functions in a free society, is the product of discussion. In turn, discussion arises from differing interpretations which different individuals, different political parties and groups give to events. But this again assumes the existence in any public of a general understanding and a community of interest among all parties to make discussion possible.” (emphasis added)

This last point was crucial for Park, because it pointed to the important role that the newspaper played in providing a common horizon of attention, around which discussion could take place about matters of common concern. And this common horizon was a cultural matter. Indeed, in a gentle critique of Walter Lippmann, Park suggested that, while it was certainly important for the news to be free from propaganda, it was equally true that the power of the press could only be realized if it was supported by a shared interpretive community, “in which there is a body of tradition and common understanding in terms of which events are ordinarily interpreted” (Park 1941a: 11). In other words, the structure of news discourse was an important matter, because it shaped the ability of the newspaper to encourage public opinion formation.

This focus on the different cultural forms of newspaper discourse was the second component of Park’s media sociology. As early as his 1923 “Natural History of the Newspaper”, Park argued that newspapers of the time were developing their own distinctive literary forms, which borrowed from fiction and other dramatic forms, in a way that made the cultural structure of the news story and the fiction story very similar (Park 1923: 283–4). In fact, Park (1938: 204) argued that a good deal of what was printed in the newspaper – most notably, the human-interest story – was read by its audience as if it was literature, in the sense that it was read to stir the imagination rather than to focus public discussion or public action. The same was true of many breaking news stories, particularly when they were reported as a series of stories and updates:

“As a story it becomes more enthralling just because it is published in installments which give opportunity for readers to reflect, speculate, or brood over the significance of each successive installment. Under the circumstances readers of the news interpret these instances and all the details in terms of memories and of similar tragic episodes with which they are familiar. In this way the news ceases to be mere news and acquires the significance of literature, but of realistic literature like the ‘true stories’ of the popular magazines and of the earlier ballads that preceded them in the history of the newspaper.” (Park 1941b: 374–375)

But Park did not only focus his attention on the ways in which news resembled literature. He was also interested in the ways that literature was following the news, in its choice of subject matter as well as its use of specific poetic techniques. Thus, as Park (1940: 686) commented, “Emile Zola’s novels were essentially reports upon contemporary manners in France just as Steinbeck’s The Grapes of Wrath has been described as an epoch-making report on the share-cropper in the United States.”

While his comments on the mutual influence and intertextual relationship between news and fictional formats foreshadowed one of the key arguments of today’s cultural sociology of the media (e.g., Alexander 2006; Alexander and Jacobs 1998; Jacobs 2005, 2007), Park never developed a complete accounting of the different cultural structures that shaped news discourse. Nor did he consider the way that fictional and literary texts created their own spaces of discussion, resulting in a “literary public sphere” (Habermas 1989) or “aesthetic public sphere” (Jacobs 2007; Jones 2007). Instead, Park’s analysis emphasized the mutual constitution of news discourse and interpretive communities. On the one hand, news writers tailored their texts to fit the cultural sensibilities of their audiences. On the other hand, those news texts reacted back on those cultural sensibilities, by providing a common horizon of attention that made public discussion possible. In the end, for Park, the issue of most importance was the degree to which the news encouraged people to talk about current events, in a way that would bring about public opinion formation. It was the public character of the newspaper that was Park’s central sociological concern. And this concern was made more urgent by one of the central issues of the day: namely, the relationship between media and propaganda.

Studies of Media and Propaganda

Sociological interest in media and the public sphere continued throughout the 1940s and 1950s, with a series of studies of propaganda and mass society. Initially, the concern focused on the different techniques of propaganda and mass persuasion, and the question of whether media served to amplify or disperse propaganda. Later, the focus would shift to the isolated, atomized individual of mass society, whose lack of social connections made him more susceptible to all forms of mass-mediated influence, including propaganda. By the mid-1950s, with the publication of Katz and Lazarsfeld’s (1955) Personal Influence, both of these questions would be unmasked as overly simplistic. Until this happened, however, most media sociology continued to share Park’s concern with the cultural power of the press.

From Park’s perspective, news and propaganda were opposed to one another. On the one hand, propaganda served to link events together around a common and strongly defined notion of a singular truth. In this sense, propaganda was close to the kind of knowledge produced on the editorial page, with the main difference being that propaganda was expressed with a greater sense of force and obligation (Park 1941b: 364). On the other hand news, as Park understood it, tended to undercut the power of propaganda: first, because it resisted incorporation into a singular truth; and second, because it was accompanied by discussion and public opinion formation. News – particularly when it was reported as a series of disconnected events – tended to disperse attention and intensify differences, in a way that weakened the desired effects of propagandists (Park 1941b: 373–375). This was why efforts at propaganda were so often accompanied by corresponding attempts at press censorship.

There was a certain amount of inconsistency in Park’s theory of propaganda. On the one hand, the connection he drew between propaganda and censorship treated propaganda as a specific and pernicious form of communication. And yet, on the other hand, Park also noted similarities between propaganda and the editorial page of the newspaper, which suggested that propaganda was a more general form of culture and communication that was characteristic of public life in mass society. Lee (1945) picked up on this more general understanding, arguing that propaganda was simply a means by which certain individuals tried to convey ideas rapidly to a large number of people. Propaganda could be rational or emotional, true or false, or somewhere in the middle of these two extremes. An efficient method of attracting sales as well as votes in a modern society, propaganda “facilitates and may even be said to be one of the elements making it possible for us to develop – for better or for worse – modern mass-political and mass-business action” (Lee 1945: 126). Rejecting a dualism that opposed a putatively evil propaganda with its sacred other, Lee (1945: 127) defined propaganda as broadly as possible, as “the use of words, symbols, ideas, events, and personalities with the intention of forwarding or attacking an interest, cause, project, institution, or person in the eyes and minds of a public.”Footnote 3 In the end, Lee (1945: 135) suggested the following set of questions be used in the analysis of propaganda techniques:

  1. 1.

    For what organization, and toward what interest or objective is the propagandist working?

  2. 2.

    What is the social and institutional context in which the propagandist is trying to circulate his message?

  3. 3.

    What rhetorical techniques are being used in the propaganda?

While Lee developed this programmatic research statement using the language of propaganda, interest, and influence, it effectively served to continue the focus on media, culture, and the public sphere. Lee’s propaganda research aimed to examine how individuals use the media to try to introduce a topic of common concern into public debate, and how they tried to represent the issue in such a way that the debate would proceed in a manner that was likely to favor their desired position. This concern is clear in many of the specific techniques that Lee pointed to as typical: (1) selecting the issue, (2) case-making, (3) simplification, and (4) “hot potato” (“the technique through which a propagandist springs an event, a trap, a situation upon his opponent that will be interpreted by most people to the discredit of the opponent” (Lee 1945: 134). Other techniques were more clearly linked to the symbolic and aesthetic dimensions of communication in the public sphere: techniques such as (1) name calling, (2) glittering generality (“short-cutting discussion by associating an idea with a ‘virtue word’” (Lee 1945: 134), (3) symbolic transfer, and (4) establishing a symbolic identity between propagandist and audience, through the rhetoric of “plain folks”. In focusing on specific rhetorical techniques used in propaganda, Lee advanced the search to identify different cultural structures that shaped media discourse (even if he did so in overly strategic terms).

If Lee’s approach emphasized the strategic nature of communication in the public sphere, Wirth (1948) provided a cultural correction to this perspective in his 1947 Presidential Address to the American Sociological Association. Returning to some of the ideas Park had developed in his writings about news, Wirth emphasized how strategic communication could only be effective if there was already a common cultural framework shared by their audience:

“The extent to which force and authority, law, religious sanction and leadership, propaganda and education, and the apparatus of symbols can be used effectively depends in large part upon this substratum of a common basis of knowledge, belief, and standards molded by tradition and reinforced by the ongoing social life that embodies that tradition.” (Wirth 1948: 6)

While Wirth emphasized the fact that mass communication could only be effective if it was aligned with an already existing common culture, he also emphasized some important new features of mass society, which complicated the relationship between tradition, culture, and media discourse. First, the increasing circulation of people, goods, and culture had led to an erosion of local community cohesion and unitary, self-enclosed doctrines; an increasing skepticism toward dogmas and ideologies; and a growing sophistication in the individual’s attitude toward attempts at influence or persuasion. These developments were responsible for the growing recognition that public consent was something that needed to be “engineered”, and to the development of much more sophisticated techniques of propaganda that could more effectively produce this engineering.

The second important social change Wirth noted was the bifurcation of most societies into two types of groups. On one side were those individuals, who were affiliated with a variety of different organizations and interest groups, and who imagined that their various interests were represented by the leaders of these different groups. On the other side was a large group of individuals who were “unattached to any stable group and in that sense constitute unorganized masses and thereby leave the decision-making to those who are organized and can exercise their corporate power” (Wirth 1948: 8). Taken together, the existence of these two groups made it difficult to think about consensus as a process of agreement on all issues by all members of society. For those who were organized into interest groups, they were likely to become more inflexible in their willingness to compromise on their interests, and less interested in issues that were not connected to the agenda of the different interest groups to which they began. For the unorganized masses, who lacked effective representation, there was a general movement toward apathy and disengagement. The solution to this dilemma involved a rethinking of the idea of consensus itself, which Wirth (1948: 9) thought should focus on the goal of establishing the habit of intercommunication, discussion, debate, and tolerance of a certain amount of disagreement and indifference. This type of intercommunication could only be organized through mass media, and for this reason Wirth (1948: 15) concluded his address by arguing that learning about the effective use of media was the most important task for sociology.

As sociologists continued to think about media and propaganda in mass society, the focus shifted away from the techniques of propaganda and toward the characteristics and behavior of the mass audience. Much of this research centered on the question of how to disseminate information and/or propaganda to the mass. Thus, Defleur and Rainboth (1952) published the results of a study that examined the most effective way to distribute airborne leaflets to a mass population, finding that the best techniques involved distributing a high ration of leaflets per person, and that message diffusion could be increased further by dropping leaflets on three successive days. They also argued that there was a small amount of “purely oral message diffusion” (Defleur and Rainboth 1952: 737), suggesting that the individuals receiving the leaflets were unorganized and socially disconnected. Graham (1954) held a similar assumption in his study of the factors relating to the adoption of television; beginning with the assumption that television requires “a penchant for passive recreation of the spectator kind” (Graham 1954: 167), Graham went on to argue that television only really appeals to the “mass audience”, which he saw as less educated, less active, and less socially connected. Similar assumptions under girded Adorno’s research on mass society and fascism – in particular, the study of the authoritarian personality (Adorno et. al. 1950), which attempted to identify which combination of personality traits were connected to the greatest susceptibility to fascist influence, and which did so without reference or concern for how the behavior of individuals is also related to the ways they are connected in specific social networks and local community relationships.Footnote 4

On the other hand, there was always throughout the 1950s an alternative argument about media and mass society, which continued the research traditions at Chicago. Freidson (1953), for example, argued that it was incorrect simply to assume the existence of the masses; what was needed was a concrete, sociological, and empirically specifiable definition of the masses, which would enable the social analyst to determine whether this was an accurate or useful concept. Taking his cue from Blumer’s (1946, 1948) writings about collective behavior and public opinion, Freidson (1953: 313) identified four key characteristics of the mass: (a) heterogenous, (b) composed of individuals who do not know each other, (c) are spatially separated, and (d) have no definite leadership or discernible organization. If the audience is in fact a mass, then their media choices and practices should be able to be explained without reference to their local group experience, and thus should be reducible to factors such as age, sex, education, socio-economic status, and basic personality characteristics (Freidson 1953: 313). In the end, Freidson insisted that the concept of the mass could not be used in a useful way to describe the media audience. Instead, Freidson (1953: 317) suggested research that begins from a focus on “the local audience itself as a social group composed of individuals who have absorbed mass communications into their relatively settled ways of behaving and who….behave towards mass communication in an organized, social manner.”

Empirical studies by Bogart (1955) and Larsen and Hill (1954) supported Freidson’s claim that the mass was not an empirically accurate way to talk about the media audience. In a study of the diffusion of a news event, Larsen and Hill examined how people came to learn about the death of Senator Taft in 1953. What they found was that most people integrated the information they got from media with the information they got from their social networks. For all groups, radio was the first place they heard about the event, followed by television for the working class population and by interpersonal communication for the faculty population in the study. Importantly, the central place of media as the source of information did not mean that these individuals were socially disconnected. Most people talked to at least three others about the event, and the working class population – despite their greater involvement with mass media – actually talked to more people about the event than the faculty population (Larsen and Hill 1955: 431). Furthermore, the overwhelming preponderance of this talk took place between people who already knew about the event, reinforcing Park’s observations about news and public opinion formation. And this observation could be extended to less “serious” media, as well, as Bogart’s (1955) study of adult talk about newspaper comics demonstrated. Borrowing from Simmel’s arguments about talk and sociability – in which the continuation of conversation is more important than the content of that conversation – Bogart (1955: 26) argued that newspapers, television, popular magazines, and movies “all provide grist for the mill of conversation in America today.”

By the end of the 1950s, the model of the socially connected media audience had largely come to replace the model of social isolation in mass society. The more that sociologists studied real audiences, the more likely it was that they would come to the same conclusion that Katz and Lazarsfeld (1955) had come to in their pathbreaking study, Personal Influence: specifically, that the effects of mass culture are always mediated through (1) the different uses they make of the media text, and (2) the specific networks to which they belong, as well as the dynamics of influence that operate within those networks. In many ways, Katz’s (1960) “Communication Research and the Image of Society” signaled the end of mass society theories, and the victory of the Chicago/Columbia model of social networks, interpersonal communication, and public opinion formation. Summarizing the state of knowledge about media and its effects – in sociology as well as the discipline of communication studies, Katz (1960: 436) wrote that

“What research on mass communication has learned in its three decades is that the mass media are far less potent than had been expected…. The shared values in groups of family, friends, and co-workers and the networks of communication which are their structure, the decision of their influential members to accept or resist a new idea – all these are interpersonal processes which ‘intervene’ between the campaign in the mass media and the individual who is the ultimate target. These recent discoveries, of course, upset the traditional image of the individuated audience upon which the discipline has been based.”

Katz’s model of media influence emphasized the interaction between media and social networks. During earlier, more informational periods, media have a good deal of influence, particularly among the early adopters and informational elite who constitute the “opinion leaders” of many social networks. During the later periods of deliberation and decision making, though, it tends to be the interpersonal relationships that are most influential.Footnote 5 The key point was that mass media and social networks were complementary rather than competitive, an insight that was difficult to arrive at from a mass society theory that assumed the existence of isolated and disconnected individuals.

If Katz’s 1960 AJS article signaled the final victory of sociology over mass society theory in media research, then the disappearance of media sociology from the flagship journals during the remainder of the 1960s is quite surprising. Katz gives his own explanation for this disappearance in the article he has contributed to this journal. First, the “limited effects paradigm” developed at Columbia showed that the media was not as powerful as people might have thought – a finding that discouraged sociologists from continuing their research on media in general. Relatedly, the new focus on audiences and social networks required time to develop and settle on a specific set of conceptual and methodological tools necessary for the accumulation of knowledge, and sociologists were too impatient to wait. Finally, mainstream sociology seemed to lump media research into the general category of “collective behavior”, a paradigm which was losing influence in the 1950s and 1960s.

I want to suggest a slightly different reason for why media sociology declined during the 1960s. My account is connected to the explanation Katz offers, but I want to place more emphasis on the withdrawal of analytical focus away from the public character of media. What Park’s research shared with the studies of propaganda was a concern for how the press influenced public opinion formation, by helping to organize the public sphere in a particular way. Wirth’s attempt to rethink the idea of consensus in a mediated mass society continued this focus. With Freidson, though, and continuing with Katz’s own work on media influence, the focus shifted toward a concern with the way that social networks intervene between media messages and individual behaviors. This finding could easily have been connected to Park’s concern with public opinion formation, but instead it was used to reject the premises of mass society theory, and its corresponding erroneous belief that there was a direct relationship between media content and individual beliefs and actions. To be sure, this was an important move to make, because it brought media research back into a thoroughly sociological framework. Unfortunately, what was lost in this analytical move was the sense that media provided a distinctive type of social output – whether that was influence, solidarity, public sociability, or public opinion formation. As a result, it became easier for subsequent researchers to develop a more reductionist media sociology, which was concerned only with the social factors that intervened between media and society. And this was clearly the approach that was taken with the organizational studies that led to a brief revival of media sociology in the early 1970s.

Organizational Studies of the Newsroom and the Continued Disappearance of the Public Sphere from Media Sociology

In its January 1972 issue, American Journal of Sociology published three articles about the media, as well as a comment by Herbert Gans which was titled “The Famine in American Mass Communications Research”. In his comment, Gans (1972: 697) commented on the vitality that had previously existed in media sociology, particularly during the 1930s and 1940s, and he went on to give his own reasons for why this vitality had disappeared during the 1960s: (1) the high cost of doing good media research, (2) the decision by the large media companies to set up their own research divisions, and (3) the disdain that many sociologists felt toward the media and its product. At the same time, Gans (1972: 701) was hopeful that there would be a revival of interest in media sociology; not only was there a growing public concern about the quality of media products, but there was also an increasing number of public attacks on the news media, which seemed to be part of the political strategy of the Nixon administration. As media came to be perceived as more of a social problem, Gans thought that sociologists would turn their attention back to the study of newspapers, television, and other mass media organizations. Indeed, media sociology did enjoy a brief revival during the 1970s. But it was a media sociology that tended to focus on organizations, and to spend less time considering the public nature of mass media.

While the three articles that informed Gans’s comment included an organizational analysis of the cultural industry (Hirsch 1972) and a content analysis of popular fiction representations of aggression (Gecas 1972), it was Tuchman’s (1972) newsroom ethnography that clearly engaged him the most. This was due in part, no doubt, to the fact that Gans himself was involved in his own newsroom ethnography, which would ultimately be published as Deciding What’s News (Gans 1979). But Gans’s interest in the organization of the news room was clearly also part of a larger trend in sociological interest; five different articles about the organization of news work were published in the flagship journals from 1972–1975.

Taken as a group, the general argument of the newsroom studies was that news, as a social accomplishment, resulted from the attempt by news workers to solve the organizational problems that emerged from their work. Tuchman (1972), for example, argued that news workers defined “objective facts” in a way that allowed them to protect themselves from internal and external critics. Most elements of news objectivity were designed to separate journalists from the risky demand of having to make truth claims: for example, presenting conflicting interpretations of a “fact”; presenting supplementary evidence to support a fact; using quotation marks to signal that the reporter is not making a truth claim, but simply reporting the “fact” of a truth-claim; and carefully separating “fact” and “analysis” with clear labels attached to the latter. In a similar way, news workers relied upon typifications in order to force “unexpected events” into normal news routines, and thereby to control the variability and contingency of news work. Importantly, Tuchman (1973: 117–118) insisted that the typifications news workers relied upon reflected the practical tasks associated with their work:

“newsmen’s distinctions between hard news and soft news reflect questions of scheduling; the newsman’s distinction between spot news and developing news pertain to the allocation of resources….. and the typification ‘continuing news’ is based upon problems in predicting the course of events-as-news.”

The argument that news practices reflect the practical challenges facing news organizations was also central to Sigelman’s (1973) study of media bias. Relying on ethnographic data as well as interviews, Sigelman argued that biased news coverage was caused primarily by the organization of the news room, and by the shared desire by reporters as well as editors to avoid conflict. A top-down model of news control by executives over reporters was inaccurate, because such a model would violate the “institutional mythology of objective reporting” in a way that would actually increase conflict. “The key to understanding bias”, Sigelman (1973: 149) argued, “lies not in conspiracies, not in conflict, but in cooperation and shared satisfactions.”

Molotch and Lester (1974, 1975) also argued that media ideology could be explained as a reflection of organizational processes. Studying news coverage of an oil spill, for example, they found that federal officials and business spokesman had better access to news workers than conservationists or local officials. Like Tuchman, however, Molotch and Lester wanted to move beyond an explanation of this access as the result of conspiracy or coercion, and wanted to explain the finding as the result of routinized news gathering practices. Because federal executives and large corporations are routinely part of the news gathering process, it is easier and more natural to journalists to turn to these sources, even during non-routine news. This explains why non-routine news stories increase the privilege given to these sources as they continue to develop in time; as the story gets typified into a particular category of news story, the normal routines of news work come to exert an ever-greater influence over the news-making process (Molotch and Lester 1975: 255–257).

For the most part, the organizational studies of the newsroom were unconcerned with public opinion formation, the public sphere, or other public aspects of media. The analytical focus was on news workers and news promoters, who were viewed as active participants in media production, participating in purposive behavior that was enmeshed within complex social networks. The people who consumed the news were rarely discussed, and when they were an object of consideration they were described as undiscriminating individuals who accepted the “reality” of the news without recognizing its socially constructed nature. This view comes out most clearly in Molotch and Lester’s “News as Purposive Behavior”, which argues for a media sociology devoted to “examining media for the event needs and the methods through which those with access come to determine the experience of publics” (Molotch and Lester 1974: 111; emphasis added).

Thus, while the newsroom studies made important contributions toward an understanding of the production of culture, the assumptions they made about the audience made it difficult to connect them to the tradition of media sociology that had developed during the previous fifty years. Essentially, and unfortunately, the 1970s media studies reproduced the assumptions about the passive and socially isolated audience that had been so thoroughly rejected by Park, Wirth, Freidson, and Katz. This critique is perhaps an overstatement, given that most of the newsroom studies ignored the audience altogether. Furthermore, as was the case with the earlier research of Freidson and Katz, one could imagine linking these organizational studies with a consideration of public opinion formation, or with the ways that media products circulate as public commodities, whose effects are connected in important ways to the specific processes by which they circulate in public. But these connections were not made. And the result was that media sociology became disconnected from its past in a way that made it less central to the institutional history of American sociology.

By the end of the 1970s, the organizational studies of the newsroom had run their course, and the brief return of media sociology to the flagship journals was over. Almost without exception, articles published about the media during the 1980s concerned the relationship between media and imitative suicide or violence (e.g., Philips 1982, 1983; Mazur 1982; Baron and Reiss 1985; Stack 1987). But these articles were completely disconnected from the historical tradition of media sociology. In fact, the primary reference point of the 1980s studies was the psychological literature on media, which took as its starting point the image of the isolated, atomized individual who consumed the products of the media in a manner that was completely disconnected from social networks or local conversations. That these articles could be published in the flagship sociology journals, despite their thoroughly un-sociological understanding of media and society was testament to the decline of American media sociology.

Culture, the Public Sphere, and the Rebirth of Media Sociology

Beginning around 1990, one could begin to see signs of a different media sociology, which was concerned with the public character of media, and which was interested in the relationship between media, culture, and the public sphere. These developments can be traced to three broad intellectual movements within sociology: (1) the focus on framing processes in social movement research, (2) the introduction of the concept of the public sphere into the lexicon of mainstream sociology, and (3) the rise of cultural sociology.

In social movement research, the 1980s saw a shift away from the resource mobilization paradigm, and toward a focus on political opportunity structures and cultural framing processes. Both of these analytical shifts emphasized that the public’s openness to a movement’s goals was variable, and depended on how the movement was able to respond to the public agenda and insert its issue effectively into public debate. The idea of a political opportunity structure referred to the fact that social movements needed to pay attention to political developments and changes in public opinion, and to be ready to mobilize for collective action when the public/political environment maximized their chances for success (McAdam et al. 1996; McAdam 1996). Furthermore, when these opportunities did arise, the movements needed to pay attention to the cultural environment in which they made their claims. Social movement scholars tended to conceptualize this cultural work as a process of frame alignment, in which movement actors put forth a public definition of a social problem that relies upon a central organizing idea – a frame – that defines the issue in a way that resonates with public sentiment (Snow and Benford 1988).Footnote 6

These paradigms of social movement research placed a renewed emphasis on mass media, which helped movements to monitor the political environment at the same time that it provided the arena where they would try to engage in successful frame alignment when the time was right. Importantly, as Gamson and Modigliani (1989) argued, media discourse had a cultural logic of its own, which did not merely reflect political coalitions or public opinion. In the first place, most public issues were defined through multiple and competing interpretive packages, or frames. These media packages did not produce public opinion on their own, but they were available for individuals to use (together with their own experiences and interpersonal discussions) in constructing their own understanding of the issue. As specific media packages rose and fell in prominence, they became correspondingly more and less readily available as cultural resources for those individuals; as a result, they became more or less useful as potential frames to be mobilized by social movements. Furthermore, the combination of specific media packages had an influence on the amount of public ambivalence surrounding an issue, which movements had to take into account when assessing the structure of political opportunity. This idea, that media organize a public arena of public issues and public problems, was also echoed in an important article about social problems by Hilgartner and Bosk (1988).

As Benson has forcefully argued in his article in this issue, media sociology also received a significant boost by the 1989 English translation of Habermas’s Structural Transformation of the Public Sphere, and the rapid diffusion of the public sphere concept throughout American sociology. As Benson notes, Habermas’s work encouraged normative thinking about the proper role of media in democracy, at the same time that it stimulated empirical research into the structural features and communicative processes that shape “actually existing public spheres”.Footnote 7

Ironically, as Benson notes, empirical research on media and the public sphere has tended to emphasize the inadequacies of Habermas’s theory, and yet it has done so without weakening any of the enthusiasm for the concept. Benson points to Bourdieuian field theory and new institutionalism as theoretical resources that can help to ensure that the public sphere remains an empirically viable concept for media sociology. While I agree with him that these are useful and important, I want to suggest an additional theoretical resource, which has become increasingly visible and influential during the last fifteen years: namely, the rise of cultural sociology.

One of the key insights of cultural sociology has been the insistence on cultural autonomy.Footnote 8 This is to say that the meaning of a given text or utterance cannot be read off of some external dimension, such as organizational ecology, the practical demands of cultural production, or the social-structural location of the producer or receiver of the message. Rather, a cultural sociology insists that communication is shaped by larger meaning structures – such as binary codes, narratives, or scripts – and that these meaning structures have their own logic. And this has real consequences for how one thinks about the relationship between media and the public sphere.

For cultural sociologists who are interested in the media, the problem with Habermas’s theory of the public sphere is that it fails to recognize that media discourse – whether in its empirical or in its idealized state – is organized through structures of meaning. Elsewhere (Jacobs 2007), I have suggested that Habermas’s failure to appreciate the full cultural significance of his public sphere theory is related to his pragmatic theory of communication, which is insufficiently cultural in its orientation. For Habermas, communicative action is driven by the pragmatic goal of reaching mutual understanding. His hope is that the mutual quest for common understanding will lead to an increase in solidarity and inclusion, a willingness to be persuaded by the force of the better argument, and, ultimately, a release of heightened levels of critical rationality. The problem is that the content of Habermas’s communicative action is completely devoid of any cultural structures. Instead of texts and performances, there are speech acts. Instead of cultural environments, there is the principle of generalizable validity. Instead of biographical or collective narratives, there is rational debate and persuasion. Unfortunately, this theory of public communication completely misunderstands the nature of public opinion formation. Habermas believes that people can be convinced to accept the more rational argument, when in fact public opinion and deliberation tends to move in the direction of the more persuasive narrative and the more convincing performance.

This goal of developing a more cultural model of media and the public sphere was the central task of my own 1996 article about the Rodney King beating, which compared different media narratives of racial crisis (Jacobs 1996). In a challenge to Habermas, this article found that empirical publics are only rarely arenas of reason or consensus; more typically, they operate as agonistic arenas for aesthetic politics and symbolic contestation. Comparing African-American and “mainstream” newspaper coverage of the 1991 Rodney King beating, this article found important differences between the two types of papers. Specifically, the Los Angeles Times reported about the crisis primarily as a story of police brutality and political divisiveness, while the Los Angeles Sentinel wrote stories about police brutality, white insincerity, and the need for African-American empowerment. The Los Angeles Times reported about the beating as the first event in a new narrative of crisis, which ended with the Christopher Commission report in a story of political unification and effectiveness. In contrast, the Los Angeles Sentinel reported about the Rodney King beating as part of an ongoing historical narrative about civil rights and police brutality, which did not end with the release of the commission report but merely continued the troubling pattern of police brutality and white insincerity. As I argued several years later, in a study on the 1992 Los Angeles uprisings (Jacobs 2000), these different narratives had profound consequences for the city of Los Angeles, helping to explain the disbelief of white citizens and the rage of African-Americans in the face of the not-guilty verdicts against the police officers charged in the beating of Rodney King. Trying to force these representational dynamics into a normative and un-cultural model of deliberation or procedural rationality is neither illuminating nor helpful.

Conclusion

It remains an open question as to whether these recent developments will lead to a more regular presence of media sociology within the flagship sociology journals. To be sure, the current situation does not seem terribly encouraging, when compared to the golden age of media sociology during the 1930s and 1940s. Of course, one could mount a convincing argument that it is less important than it used to be for the flagship journals to publish work on media. After all, media research is much more interdisciplinary than it used to be, and media sociologists have plenty of good places to publish their work, such as Media Culture and Society, Political Communication, or Theory and Society. Nevertheless, I think that most media sociologists would prefer to see their work get more visibility in the flagship journals.

If media sociologists want to increase their visibility within the mainstream general-interest journals such as American Journal of Sociology, I want to suggest that they do more work to connect their scholarship to the Chicago School tradition, and in particular to the work of Robert Park. As I have shown in this article, Park’s media sociology emphasized three points: (1) that media are a particular and distinctive form of culture or knowledge; (2) that media are a public commodity, generating discussion and public opinion formation; and (3) that media have an intertextual relationship with other types of cultural products. Based on my own specific research interests, I believe that the Park tradition of media sociology can most effectively be reclaimed through a research agenda that combines cultural sociology and public sphere studies. Yet, any media research that takes the time to connect itself to Park would make progress toward institutionalizing a recognizable disciplinary tradition of media sociology. Time and time again, this opportunity to build a coherent disciplinary tradition has not been taken: not by the Columbia tradition of limited and indirect effects, not by the organizational studies of the 1970s, and not by the more cultural studies of the late 1980s and 1990s.

As Alexander (1987: 28) has argued, every intellectual movement or research tradition in the social sciences needs the legitimation that comes by attaching itself to one of the classical founders of the discipline. Attachment to a classical founder or a classical text provides symbolic condensation, which provides a common reference point and a minimal baseline of understanding between specialists and non-specialists. Attachment to a classical founder also provides an easy answer to a question that is so often raised by the flagship journals: namely, what kind of general sociological contribution is being made by this manuscript? Media sociology has not attached itself to a classical founder, and so it has a difficult time answering this question. A sociological project that examines media and the public sphere should have an easy time answering such a question, though, if it is willing to claim Robert Park as its classical founder.