Note: the following is an unedited seminar paper I wrote in my first summer of grad school exported to HTML. At the time I was exploring writing a dissertation about the history of the academic communications research. I'm putting it up not as an academic article, but I occasionally encounter people working on similar topics with the Rockefeller Foundation archives and I'm extremely unlikely to do anything further with the materials here. (There is a bit more about the Lazarsfeld-Stanton Program Analyzer in my dissertation, and I have a little additional research.)

Radio Reception and Theories of Society:

The Rise and Fall of Communications Research, 1936–1943

Benjamin MacDonald Schmidt

December 19, 2006


Among the many effects of the rise of new electronic media on American life was a newly rigorous focus on the concept of communication itself. A new type of mass media and mass culture, embodied above in the swift establishment of radio as a central element in the day-to-day life of most Americans and Europeans, drove social scientists to conceptualize “communications” as, for the first time, a distinct field of study in the human sciences. But though it was clear that ‘communications’ represented an important aspect of social life, no one knew exactly how to explore it. As a result, communications studies became, in its early years, an incubator of diverse perspectives from throughout the social sciences and humanities. For some it seemed that a fully articulated field of communications would encompass a new theory of society as whole.

How this field formed, and how it functioned in its early years, is an important and largely overlooked facet in the evolution of the American social sciences in the period before the Second World War. Never definitively enshrined as a peer of older fields like sociology, economics, or anthropology in the universities, communications plays only a passing role in studies of the social sciences during the interwar period as a whole. Histories of the field written by modern communications researchers, on the other hand, largely treat the early history of the field in relation to present conflicts within the discipline, ignoring other important elements of its early development. The opening statement in one recent history that communications “crystallized into a distinct discipline within sociology—complete with colleges, curricula, the authority to grant doctorates, and so forth—between about 1950 and 1955”1 is only an extreme example of the disregard most writers have for the flourishing field of communications in the late 1930s and 1940s. As a result, both sides in the major division in the history of the field—between advocates of a “received history” advocating a linear narrative of scientific progress, and critics of the “dominant paradigm” who seek to integrate voices from the past outside the academic discipline who anticipated a more critical perspective towards society and the media (generally, a perspective shared by the writers themselves)—ignore the prewar research that did not directly feed into later lines of research in communications. 2

This disciplinary perspective is particularly stifling because early communications research is characterized, above all, by its interdisciplinary. From the mid-1930s into the war, communications functioned as an important site for the exchange of ideas, techniques, and individuals not only among the social sciences, but with the government and the broadcasting and advertising industries as well. The novelty of the research problems brought together individuals who would later scatter into the security of more isolated, homogenous enterprises in academics and business. For a brief period, advertisers felt that that they needed psychologists to understand how to market their products; broadcasters felt that they needed academic statisticians to help them analyze their listenership patterns; and researchers felt that they needed corporate and government funding and data to produce studies of the scope that the new medium demanded. Until the increasingly utilitarian needs of the war effort came to restrict the scope of research, the flexibility of approaches in the field meant that communications played an important part in the incorporation of new ideas across disciplinary and professional boundaries.

As many writers have noted, the federal government played a critical role in the early evolution of communications research, particularly in the important subfield of propaganda.3 But particularly in the period before the war, two other sets of institutions played an even greater role in spurring new academic research. The first were corporations involved in publishing and broadcasting, who looked to academic research to help identify their audience and increase their profits, above all through a better understanding of advertising. Second, nonprofit foundations and universities predated the government in their support for research into the effects of new mass communications agencies, and so ensured that substantial resources were available even for studies that did not strictly align with corporate interests.

The result of these forces was a fresh field of study, with low barriers to entry, a socially relevant subject matter, and no single methodological orientation powerful enough to exclude other approaches. As a result, the study of communications briefly attracted leading scholars and rising stars from across the social sciences. Among the most prominent were psychologists like Kurt Lewin and Carl Hovland, sociologists including Paul Lazarsfeld and Robert Merton, and political scientists such as Harold Lasswell and Carl Joachim Friedrich. Equally impressive is the quality of scholars who worked in the field but were temperamentally and ideologically leagues away from the functionalist, quantitative model that came to dominate communications research. Attracted by the possibility of analyzing the massive changes in research brought about by the new media, soon-to-be major social critics including C. Wright Mills, Theodor Adorno, and David Riesman all worked in the field in its early days. And early communications researchers achieved great success in the business world, as well.

Although, as critics of the field have argued, the eventual orientation of communications was behavioristic and quantitative, this early history of the discipline is not a straightforward emergence out of the more empirical branches of the existing social sciences. Nor is its later history simple an account of the academic field of communications through the decades. Indeed, the communications research of the late 1930s and early 1940s should not be viewed exclusively as antecedent to the modern academic field with the same name. Historians of science sometimes like to look at the roads not taken in the evolution of a science; but in the case of communications research, many of them were in fact taken, but by corporate researchers or sociologists or others outside the confines of the present-day field. During the war, many of the leading figures in communications left the field, and the postwar field lacked some of the brilliant diversity that characterized its first methodological flowering. The greatest importance of early communications studies lies in its role as a site of exchange among corporations, government officials, and the various social sciences; this is the network of relations that this paper explores.

The origins of this community lie in a small group of organizations that together constituted the core of the first group to identify themselves as communications researchers. Most were centered around New York City, which was home to the major radio networks, the primary funder of research in the Rockefeller foundation, and the most important research project, at Columbia University. The time frame is relatively narrow as well; this story primarily covers the period from 1936, when the Rockefeller Foundation began to seriously consider the need for a new type of research on the mass media, until 1943, when the increasing subordination of new research to the war aims of the government ended the early period of diverse methodological approaches.

  1. The Major Institutions: Communications Research 1936–1943.

    1. Planning an academic discipline

The christening of communications did not take place until 1939. John Marshall, the assistant director of the Division of Humanities at the Rockefeller foundation, wrote that summer to researchers in “a field which for a lack of a better name I have come to call mass communications.”4 By the time he felt the need to put the field on a new footing, research into the “mass media of communication” (the first two words of the phrase were not yet familiar enough to stand on their own) had been going on for some time. But only in the 1930s did that research come to collectively take place far enough from the media corporations to be seen as a single unitary field. Marshall and the Rockefeller foundation were critical in effecting that change; and so, the genesis of communications research lies in large part in their efforts to unify the various inquiries into media and society already underway.

For the first half of the decade, substantive research about the radio had been primarily driven by the needs of the rapidly expanding ranks of broadcasters. Although commercial broadcasts began only in 1920, in 1930 radios were already more common household items than either telephones or newspaper subscriptions. By 1940, a radio receiver sat in four out of every five American homes.5 As a commercial structure evolved that relied on advertisements as the sole source of revenues, two questions in particular demanded new research techniques. The first was the question of the radio audience: since radio executives had no figure comparable to newspaper circulation to show their advertisers, to convince advertisers to sponsor programs they needed to find a different way of estimating the number who listened to them. A second question, less unique to radio, lay in measuring the efficacy of advertising and seeing what sorts of advertisements had the greatest impact.

Though the former sort of question was most easily pursued via survey research and the latter through laboratory research in applied psychology, in practice the same investigators often answered both of them. The career of Frank Stanton, who emerged out NBC’s research department to head the network in the mid-1940s, is indicative in this regard. As a graduate student in psychology at Ohio State University, Stanton initially investigated the effectiveness of advertising in various media, through a series of tests testing how well people remembered key phrases and brand names from advertising copy that they either saw printed on the page or heard read aloud.6 For his doctoral dissertation, however, he turned to the problem of audience research. Unconvinced of the accuracy of existing survey techniques in measuring the radio audience, he built a number of toaster-sized devices that registered if a radio was turned on or off. After installing them in a number of homes, he compared the self-reported listening habits of the subjects gathered through traditional survey techniques with the data logged by the machines.7 By Stanton’s account, the ideology underlying the device was his “Ohio State behaviorism”: since radio listening, not memory of listening, was the relevant variable, he set out to design a device which would measure the physical fact of the receiver’s activity and nothing else. Fittingly enough, the immediate technical inspiration for his test of listener honesty was the stylus-paper arrangement that records physiological reactions in a polygraph machine.8

While the psychologists were carried on this and similar investigations into the nature of radio, educators were also growing increasingly interested in the medium as they sought to create educational programming, build university radio stations, and schedule regular broadcasts of lectures. Even more than the networks, though, the educational institutions felt the need for specific information about their audience and the effects of their broadcasts. While network broadcasters were generally content to know simply their raw number of listeners, educational and cultural broadcasters were more particularly concerned to learn who was listening to their programs. Did the lectures that many university radio stations sent over the airwaves create a new audience for their subjects, or were their listeners the same people who already frequented the lecture halls? What (if anything) did listeners take away after listening to broadcasts by educational shortwave stations such as W1XAL in Boston? And how could listeners be induced to tune in these sorts of broadcasts instead of the light music and serial dramas that dominated the airwaves?

Not only the educational broadcasters wanted the answers to these questions; their underwriters, too, wanted to see tangible support for the high-flying hopes many educators placed in radio for the dissemination of culture and knowledge. On the other hand, networks, with a comfortable rate structure in place for advertisements, had no need to delve further into the data. As one contemporary put it, “Business is so ‘good’ in broadcasting that the industry is not inclined to spend money for research until they are forced to do so, not only because of the embarrassment such data might cause, but because business does not demand it.”9

As funders like Marshall at the Rockefeller Foundation realized, a substantial investment in research to answer these questions would not only help increase an abstract understanding of the workings of the radio; it would also be invaluable in active efforts to increase the quality of broadcast programming. What kind of research, though, would it take? Studies of the audience could build off of existing survey techniques like those used in the Crossley ratings of radio broadcasts (the predecessor to the modern Nielsen television ratings); but more detailed demographic data would require a large, independent research infrastructure, presumably using the sampling techniques being developed for the new polling companies. Likewise, psychological studies of advertising effects could provide only a starting point for questions about the effects of programming in the long run, and indeed of radio listening in general. Thus, even as they strove to answer new empirical questions, educators and funders contemplating further research found themselves confronting new methodological questions as well. Precise answers to these methodological questions would only emerge once research had actually begun, as projects began to spin off new questions of their own and as researchers from different fields began to discover both what questions they shared and what important subjects none of them studied. It was in the course of this research that what had been thought of simply as “radio research” began to frame its questions and approaches in terms of a larger problematic: mass communications.

Lazarsfeld’s Bureau

From its founding in 1936, the Princeton Radio Research Project would be the most important site where these questions were developed. Marshall received the initial proposal from Princeton psychologist Hadley Cantril, whose The Psychology of Radio struck him “as an almost heaven-sent indication of an interest in work of this kind” and Frank Stanton, who had by then parlayed his Ph.D. in psychology into a job in NBC’s tiny research office.10 They laid out an ambitious plan not just to conduct studies of the social impact of radio, but also, through the development of new methods, to determine “what constitutes the field of radio research” at all.11 As the occupations of its directors suggested, it would build on existing research in both the academy and the industry; but in trying to describe the “social significance of radio,” they would move far beyond either the individual-oriented psychological studies or the raw commercial statistics collated by the networks.

No sooner had Marshall authorized a substantial Rockefeller grant in support of the project—$67,000 over three years from September 1937—than both Cantril and Stanton begged out of the immediate responsibility of running the project. Cantril was already struggling under his workload as a newly tenured professor at Princeton and an editor at the Public Opinion Quarterly, and CBS enticed Stanton, who had been supposed to direct the day-to-day activities of the project, to remain full time at the network with a substantial raise and a promotion to director of research. Over the summer of 1937 they frantically consulted a series of advisors, including the pollster George Gallup and the sociologist Robert Lynd, possible replacements to shape the project. All the leading players came to agree that Paul Lazarsfeld, an Austrian social scientist who was trying to establish himself in the United States, would be the best choice.

Although Lazarsfeld made his name in America as a professor of sociology, those who knew him in his first years in the United States thought of him more as a market researcher or social psychologist. By training, he was actually a mathematician, but Lazarsfeld had been able to switch to work in psychology—at the influential Viennese institute of Karl and Charlotte Bühler—largely because his mathematical expertise let him the lead in social psychological investigations involving the analysis of data. In Vienna, he had founded the Wirtschaftspsychologische Forschungsstelle [economic-psychologicall research center], which organized researchers to develop advanced techniques to engage in contract research for anyone who bid for their services. Lazarsfeld, who had earlier been active in the Viennese socialist movement, happily engaged in studies of topics such as the psychology of shoe purchasing and the factors influencing consumption of tea. The center was, a contemporary later recalled, “a midpoint between an institute for market research… and a university institute for social research.”12 Among these myriad research was one of particular note to the Princeton directors: the first quantitative study ever made of radio listening habits from a social-psychological perspective, which Lazarsfeld completed in 1932.13

Lazarsfeld’s myriad, and nearly contradictory roles in Austria—market researcher, quantitative sociologist, socialist activist, and Viennese psychologist—would all have their influence on the Radio Project, and all represent important strands of early communications research.14 The two former roles, however, were the most important in the project’s early years. Thanks to Rockefeller funding and its Princeton affiliation, the Radio Project was independent of the networks insofar as it could pursue research without having to worry either about its profitability or, for the most part, whether its findings would be embarrassing to the industry. But since the most extensive data on radio listening was held by the networks, and since the radio researchers didn’t want to forswear the help of experienced researchers and broadcasters like Stanton himself, they retained strong connections to corporate needs in their research. As a result, even as ‘radio researchers’ sought to invent techniques that would help educational broadcasters or that might criticize the existing broadcasting system, they had a strong incentive to work at the same time on projects that would pique the industry’s interest. Thus, the Radio Project often acted as a testing ground for advanced techniques that the networks were unwilling to engage in themselves.

This catholic approach to research problems grew out of Lazarsfeld’s own research style. He saw himself as, first and foremost, an expert in the techniques of research; the topic to be studied was of secondary importance. For this reason, he was willing to take on topics that had little intrinsic interest as social science, as evinced most of all in the marketing research he undertook both in Vienna and in New York.15 Thus, even the Office of Radio Research itself was often engaged in studies only tangentially, at best, related to the radio. For example, in his eagerness to sharpen his office’s facility testing and administering questionnaires, Lazarsfeld helped develop an early version of the Cosmopolitan magazine quiz: the ORR produced a report for the magazine testing the reliability of a survey to be administered by a friend on the topic “Are you an overspender?” But the immensity of the corporate operations also presented tremendous opportunity for Lazarsfeld to obtain new data sets to work with, and to fund new investigations: as he put it, “This field of radio research presents three important opportunities to psychologists: challenging problems, unusual data, and more jobs.”16

Some of the project’s most lasting creations shows the ways that ‘radio research’ itself came to serve as a locus for the transition of research techniques from applied psychology to corporate use, much as Frank Stanton had done with his work in graduate school on listening behavior. The Lazarsfeld-Stanton program analyzer (institutional loyalty led to it being called the Stanton-Lazarsfeld program analyzer at CBS) used the same sort of mechanical arrangement as Stanton’s dissertation project—a marker inscribing results over time on a moving strip of paper—but with a technique less heavily grounded in behavioristic ideas. Rather than registering whether a radio receiver was turned on or off, the program analyzer registered if the listener was pushing one of two buttons. A green button indicated approval of what he or she was hearing; a red one, disapproval. Researchers would use this record to guide an interview of the subject to investigate what properties of the program had caused the reaction.17 The researchers rapidly applied the program analyzer to a number of studies involving the effect of broadcasts on listeners, and proclaimed its virtues in Radio Research 1942–1943. But its use as a tool of rigorous research was limited, as well as its suitability for meaningful evaluations of educational broadcasts. A later attempt to use a slightly more complicated version of the same idea as a tool for classroom teaching shows the difficulty. The researchers went through several evaluative pairs to correspond to the red and green buttons (including understand/don’t understand; interesting/uninteresting; important/unimportant) before finding that the only couplet that produced consistent results was “I am getting what I want/I am not getting what I want.”18 The result neatly sums up the successes and failures of the device. It produces feedback that cannot easily guide improvement in the tester’s goals; it only enables him or her to react to the immediate wants of the person being tested.

This reliance on the irreducible feelings of the experimental subject, although it crippled the program analyzer’s usefulness in academics, proved to be perfectly suited to the networks’ needs. By allowing the quantification of preferences on programs, the program analyzer filled an analogous need to that served by ratings—executives were finally able, without the expense of broadcasting and Crossleys, to predict which shows would be successful on the air. Stanton immediately began using it in his research department at CBS, and the advertising firm McCann-Erickson paid a regular royalty from 1943 for the “exclusive” use of the machine (although it was not patentable, and others would use it).19 Like much communications research, this corporate tool also found itself swept up in the war effort—specifically, in work done as part of Samuel Stouffer’s massive project to study the attitudes of the American military. Stouffer (himself an alumnus of the radio project) assigned Yale psychologist Carl Hovland to help test the efficacy of Frank Capra’s propaganda films “Why We Fight.” Hovland contracted Lazarsfeld’s ORR to run assessments of Capra’s film, and one of the most recent additions to Columbia’s faculty, Robert K. Merton, directed the study. The accuracy of the device allowed Merton to confront the filmmakers decisions: “Even though Frank Capra thought he was reaching hoi polloi, these kids didn’t know what he was talking about,” Merton later recalled.20 More recently, the usefulness of the device in helping to shape popular entertainment and propaganda has helped win it a position as an essential element in the arsenal of the modern campaign manager and pollster.

Other studies began as outright collaborations, rather than gradually finding their way into corporate and government applications as did the program analyzer. For example, researchers realized that although the CAB rating data included information on socioeconomic status of the respondents,21 both the surveyors and the networks used the rating data only to create raw listenership numbers. The project’s directors contracted NBC research director H.M. Beville to investigate the possible further applications of this data, with the stated aim “to acquaint a larger public, presumably one interested in educational broadcasting, with the vast fund of material which commercial agencies have been collecting for years and which is quite astounding as to amount and regularity of canvass.”22 Lazarsfeld’s Austrian experience clearly imprinted itself on the study’s conception. Above all, the devoted attention to social class reflects the lingering political ambition in Lazarsfeld’s early work to show the distinctive features of the “proletarian buying public.” (Indeed, some have found this attention to class the only noticeable income of his early leftism on Lazarsfeld’s later research.)23

But while Beville’s study did manage to squeeze a great deal more information out of the existing Crossley numbers than had previously been attempted, the benefits again redounded chiefly to the commercial community. The data that Beville analyzed were available only to the advertisers, agencies and networks which subscribed to the Crossley reports. While the report’s conclusions were not without interest for the educational community (they included, for instance, data about cultural programming like the General Motors symphony hour), they were severely limited by the fact that Crossley numbers included only commercial programs; neither smaller educational broadcasters, nor the networks own unsponsored “sustaining” broadcasts were included. Thus, as with the program analyzer, the research efforts of the Radio Project brought new techniques to bear on problems that the networks themselves had regarded as unimportant, but more to the benefit of the networks than to the research community. Lazarsfeld proudly reported to the Rockefeller foundation that Beville had told him “NBC really ought to have paid him for doing this job because by doing it, he learned how much more this contained than he or any of his colleagues had ever before suspected.”24 Advertisers, too, stood to gain from such research. When the time came to market the second collection of project essays, the circular promoting the book read “Every advertiser, every agencyman knows the names Paul F. Lazarsfled and Frank Stanton. Lazarsfeld is Director, Office of Radio Research, Columbia University and Stanton is v. p., CBS. Together they have compiled a book which will be of great value to advertisers. The book is: Radio Research 1942–43.25

While these commercial applications represent a crucial part of the early work in radio research, they nonetheless formed an essential part of a research model that helped establish communications as an independent field. As the use of socioeconomic analysis in the Beville study suggests, not all these advances could be traced back to origins using the sort of psychological equipment that was epitomized in the program analyzer. This insistence on integrating a wider social perspective in its work is one of the central developments that made the new communications research fundamentally different from the sorts of studies that had come before. The dynamics of this shift were already evident even before Lazarsfeld arrived: by asking Robert Lynd for advice on the new director of their project, Cantril and Stanton were implicitly admitting that the psychological techniques they used could only go so far in illuminating “the social situation of radio.” But it was in the persons of Lazarsfeld and Lynd that the research began to actually engage with these new problems.

Just how closely communications research looked to sociology as a model is evident in the field’s relationship with the Columbia sociologist Robert Lynd. Lynd’s immensely popular study Middletown, co-written with his wife Helen, served as a dominant model for the sort of work early communications researchers hoped to undertake. They hoped to expand on his exploration of the ways changing American values, particularly increasing capitalist acquisitiveness, had disrupted traditional patterns of American life. Likewise, they wanted to be able to look at the impact of new communications technologies on typical American communities, not just individuals, and to be able to offer holistic sociological descriptions of those communities in the same way that Lynd had transformed Muncie, Indiana into an archetypal “Middletown.” Moreover, Middletown itself presented a model investigation of modern mass society in a way few other studies could. The Lynds did not fully explore the role of media and communications in Muncie’s civic life; but it was easy for communications researchers to present their work as adding that new dimension to complete the picture of mass society that Lynd had begun. Accordingly, Leo Rosten introduced his Motion Picture Research Project to the readers of the Public Opinion Quarterly as “A ‘Middletown’ Study of Hollywood,”26 and Paul Lazarsfeld’s study of voting patterns in Sandusky, Ohio seems to have evolved out of an initial plan to select “a kind of radio ‘Middletown’ to which we can refer whenever needed.”27

Moreover, although Lynd did not himself direct any studies of mass communications, his post-Middletown renown as a professor at Columbia, allowed him to serve as a senior statesman lending credibility and guidance to the nascent field. Lynd initially sponsored Lazarsfeld’s first trips to the United States under a Rockefeller grant, suggested Lazarsfeld as a likely head of the Princeton Radio Research Project, and later saw that his Office of Radio Research found a home at Columbia; Lynd sat on the advisory board to Rosten’s study of Hollywood and joined John Marshall’s seminar on Communications at the Rockefeller Foundation. Even so, he remained something of an outsider—as he wrote John Marshall at the Rockefeller Foundation, “I am interested in the field of communication and regard it as very important. But the doggone thing just isn’t something in which I have either central interest or special competence.”28

While Lynd was willing to remain an outsider, albeit a very influential one, others saw in radio a new way to make their careers the same way Lynd had made his with Middletown. Thus, the Radio Project pursued studies that rather than looking at radio’s effect on the individual level, attempted to see its effects on a community as a whole, using the tools of social psychology, anthropology, and sociology in similar ways to Lynd. Such community-oriented studies became increasingly important as the political crisis in Europe grew. Especially since the world of research was full of European émigrés, expectations were high that America too would quickly become involved. As a result, the public hysteria that broke out after Orson Welles’ Mercury Radio Theater broadcast their famous “War of the Worlds” adaptation appeared an indispensable opportunity to explore the ways that the new medium functioned in disseminating information and panic in a crisis situation. Hadley Cantril and Herta Herzog (who, in Princeton, was just a few miles from the site of the alien landing and therefore ideally located to study its effects) directed what Lazarsfeld called a “Firehouse study” to quickly interview dozens of individuals who had heard the broadcast, and to speculate about both the psychological and sociological causes that made certain individuals and communities more susceptible to panic.

A more widely anticipated event, the 1940 presidential election, afforded Lazarsfeld an opportunity to conduct a “Middletown” study of his own. Although he had initially hoped to conduct a panel study of radio listening, the final work (still done under the auspices of the ORR) studied the forces influencing voter behavior. The published study, The People’s Choice, was widely influential in political science as the first study to focus in depth on a single group of voters, and thus to be able to determine when and why they changed their minds. In communications research it played an important role in structuring the way that researchers thought of information as diffusing through society, introducing the idea of the “two-step flow” which has since become a staple of the communications curriculum. According to the theory, media events do not have an instant and predictable effect on each individual listener in isolation (called the “hypodermic needle” model); instead, opinions and information tend to diffuse from the media into a few key members of the community (“opinion leaders”) who, through personal interactions, tend to guide others in the community in forming their own opinions.

This theory, once accepted, cemented an element of the sociological perspective as a new orthodoxy in the field. If media effects diffused through a community rather than acting on an individual, psychological experiments or even focus group panels would be inadequate to capture their effects. Furthermore, the idea of the two-step flow was crucial in consolidating communications research as a field, by seeing bringing together electronic and print media with interpersonal conservations as a single object of study. Similar ORR studies relying on these insights found wide diffusion in the print media, in academic organs like the Journal of Applied Psychology, and eventually the ORR’s own Radio Research and Communications Research publications. As a result, by 1940, the Office of Radio Research had already brought the mainstream of the field into an understanding of the role of communications in society that went far beyond its origins in corporate research, into a theory of the new dynamics of society the pervasive electronic media was creating.

This theory was by no means intrinsically radical. Indeed, certain features of Lazarsfeld’s approach, like its top-down model of dissemination and its focus on discrete, measurable changes in opinion rather than the more pervasive ways the media might shape society, have led some more recent media theorists to consider the ORR research politically regressive. But even in its origins from Lynd’s Middletown studies—which contained, beneath their insistence on scientific objectivity, a committed progressive stance29—communications research contained a good deal of heavily critical potential. Cantril, Stanton, and Marshall’s choice of a socialist Austrian political refugee in Lazarsfeld to head the project further shows the field’s willingness, at the time of its birth, to criticize fundamentally the government and corporate arrangements that controlled key channels of communication. Those Lazarsfeld hired for the project often did so. Many were, like Lazarsfeld, European émigrés whose position left them eager for work. And because of the political climate in Europe, the scholars most eager to leave often held positions far more radical than Lazarsfeld’s own. Despite these views, however, and thanks to the extraordinary methodological fluidity of the early work, communications research proved an important vector for their incorporation into American society. Their incorporation, in turn, helped communications become of the most important routes through which European currents in psychology and social thought, particularly Marxist and Freudian theories, entered into American social science.

The most famous of these encounters was the tenure of Theodor Adorno as the director of musical research at the Radio Project. Adorno’s experience in New York is often summarily dismissed as a failed experiment, with scant publications and no immediate impact.30 But although few of works found immediate publication, primarily because John Marshall cut Rockefeller funding to his projects before they could be completed, Lazarsfeld’s relationship with Adorno and with the other members of the Institut fur Sozialforschung in exile in New York was actually quite productive. Lazarsfeld himself, despite an occasionally strained personal relationship, was a passionate advocate of the value of Adorno’s work to the Rockefeller Foundation and others, going so far as to invite a number of guests from the foundation and from the networks to attend Adorno’s staff presentation of his work at the Office of Radio Research, an unprecedented move. The eventual removal of Adorno from the Office’s payroll came only over his repeated objection. Moreover, he continued to work closely with other members of the Frankfurt School: Leo Lowenthal drew on a few of the lessons of content analysis for an article on biographies in popular magazines in the 1942-43 issue of radio research, and the official organ of the Institute, Studies in Philosophy and Social Science, handed its reins to Lazarsfeld for an entire issue devoted to the social criticism of radio research.

In many respects, Lazarsfeld’s program was not really so far from that of the Frankfurt School. Although the sociologist’s Marxism seems to have nearly disappeared after he arrived in the United States (aside from his continuing attention to the importance of class distinctions), he persisted in questioning the role radio played in society. A letter to Robert Lynd after Adorno left the project reveals the depths of his suspicions about the new medium:

However, this new idea of two-way communication now clouds one very important aspect of communication research, and that is the study of what radio (chosen as an example) does to people unintentionally. Here really the whole Adorno aspect would have to come in. In ‘Radio and the Printed page,’ for instance, at the end of the sixth chapter I have outlined why radio would always reinforces conservative, middle-class mores. It always shies away from any discussion of really controversial matters, that is to say all controversies except those between the Republicans and Democratics [sic]. It tremendously increases authoritarian patterns by continually playing up the biggest band leaders, the best soap, the greatest song hit, the most famous actor, and so on. It also impoverishes the little bit of variety we have in our leisure-time activities by forcing us to take what we get instead of getting what we want. The tremendous amount of popular music provides steady sources of escape.31

Like Adorno, Lazarsfeld saw authoritarian tendencies reinforced by the new entertainment industries; and like them, he sought an explanation in the psychic needs of the individual. The chief difference was that Lazarsfeld saw this perspective as only one useful front in the larger struggle to define a field of radio research; the critical theorists could not warm to tasks such as using the program analyzer to determine the quality of new shows, because they saw the individual tastes as being too warped by current radio practice to be of much use. But while he was unwilling to curtail his studies to the extent Adorno might have liked, Lazarsfeld remained enthusiastic about the contribution that a “European perspective” of critical Marxism could have in the understanding of radio.

The anti-corporate elements of the Marxist critiques fell on predictably unreceptive ears from many of the industry-associated individuals who worked with the project. One wrote to the Rockefeller Foundation condemning Adorno’s talk as “hardly worthy of serious consideration, except possibly as propaganda. In short, it seems to have an axiom to grind.”32 But it was not conservative resistance that ultimately drove the critical theorists from the field; rather, it was the ameliorative goals of those who thought radio research should help purge the worst excesses from corporate radio. Thus, in Adorno’s case, the final act took the form of a classic struggle between revolutionaries and reformers. In a memorandum explaining why he had cut funding for Adorno over Lazarsfeld’s objections, John Marshall wrote that Adorno “seems psychologically engaged at the moment by his ability to recognize deficiencies in the broadcasting of music to an extent that makes questionable his own drive to find ways of remedying them.”33 Robert Lynd, the most vocal proponent among the Rockefeller group of taking advantage of the national emergency to consolidate greater governmental control of the radio infrastructure, agreed with Marshall: Adorno was too unable to make allies within the system to be a useful critic from outside it.

Psychoanalysis presented fewer political problems, and Lazarsfeld aimed from the beginning to incorporate it into radio research. One of his first actions as director of the project was to pay a group of psychoanalysts to discuss topics like the nature of the entertainment derived from different sorts of radio broadcasts. Indeed, Lazarsfeld’s connection to the world of Freudian psychology was so strong that Frank Stanton himself mistakenly remembered a number of years later that Lazarsfeld had been trained by Freud. At the most basic level, the Freudian influences that pervaded the project served as a corrective to the ascendant behavioralist trends in American psychology. Frank Stanton, as the product of that school, often found himself raging against the constant intrusions into the project’s research of the Freudian idea of the unconscious. In this sense, the Freudianism of the Radio Project can even be thought of as further evidence of its sociological turn, for in the late 1930s Freud’s influence was stronger in that field than in the behaviorist psychological circles that fed into much radio research.34 The most important transformation of Freudian techniques at the radio project, however, involved their application to principles of marketing and propaganda. Much of the Office of Radio Research’s contract work for companies was designed to give advice to advertising copywriters on their particular products: thus, for example, a study on the psychology of refrigerator purchasers framed its central question: “Do people have more subtle desires to which we could link the use of a refrigerator?”35

The Rockefeller Communications Seminar: Nationalizing Research

Lazarsfeld’s project stood at the forefront of media research, due to its size, early founding date, and the number of its innovations. But even as his project was pushing the field in new directions, John Marshall at the Rockefeller foundation sought to bring scholars of other media into conversation with each other in a single field of “mass communications.” The single most important exercise in this attempt was a seminar held at the Rockefeller Foundation in 1939–40 that tried to set a coherent set of goals for researchers in the field. The seminar’s members included researchers who were at the forefront of empirical studies of the media, such as Lazarsfeld and the political scientist Harold Lasswell, whose study of American propaganda after the first world war served as a paradigm for research into media effects in the early field; and comparative outsiders like Lynd and the British scholar I.A. Richards, one of the founders of the New Criticism in literary studies.36

If the seminar was the moment of communications’ birth, the historical moment led the committee to midwife a creature considerably different than had been gestating at Lazarsfeld’s offices and elsewhere across the country. Marshall sent out invitations to the seminar in August 1939; by the first meeting in September, the German invasion of Poland made a university-based, social scientific approach to communications seem untenable. Lazarsfeld’s strategy of allying research to corporate interests paled, now, in comparison to the tremendous funding potential of the New Deal government gearing up to engage in a massive propaganda war. The suspicion of hysteria or manipulation promoted by the media was now dominated by the question of mobilization. “A purely theoretical and academic view of what mass communication is doing in society,” Marshall wrote, was no longer possible: instead, the seminar quickly turned its thoughts to changing and aiding government policy during the national emergency and after.37

This did not prevent the seminar from trying to create a coherent framework for all communications research. Harold Lasswell quickly proposed that communications research should attempt to answer the interrelated questions of “who says what to whom with what effect,” a phrase that came to define a generation of communications research. In explicitly calling for attention to the production of information with the first two questions, they underlined an approach that had been applied only in rare cases in the first stages of Lazarsfeld’s audience-oriented studies. Content analysis, when attempted at all by the office of radio research, had largely been the province of humanistic critics like Adorno’s study of radio music or Leo Lowenthal’s article on the contents of popular biographies in magazines. But Lasswell was developing a new approach which promised to operationalize the question, at least with regards to written or spoken texts. It bore fruit in the fall of 1940 with Rockefeller funding for a project at the Library of Congress under Lasswell to count the appearance of key words, terms, or ideas in the foreign press (for example, changes in the mention of Russia in the German press preceding the Molotov–Ribbentrop pact).

As notable as the additional questions is the question that is not asked: why the communication is happening at all. In a few of the committee’s minutes, the additional phrase actually is included in the formulation (Lasswell, in one draft of a joint memorandum, was to write about both “who communicates” and “for what purpose”).38 Its excision reveals an important change over the course of the seminar. As late as May, nine months into the project, the participants still did not agree on who the audience for communications research ought to be. Gorer harkened back to the original rationale for establishing the Princeton Radio Project by suggesting research should be targeted at educators; Lazarsfeld made the widest claims for their work with the suggestion that it be aimed at everyone who might use a medium. Lasswell, however, tried to situate the research more clearly in the context of the international situation by suggesting policy makers formed the primary audience for the medium.39 Within the black and white world of national policy during wartime, this necessarily meant that study of why the government chose to communicate would be pushed out of the mainstream.

The government-oriented view of communications research won out in the committee’s ultimate self-presentation, a memorandum outlining their proposed direction for further research on mass communications. The essay is cast in terms of government necessity: since “events are obliging our central government to take on wider and wider responsibility for the welfare of the people,” the authors argue that improved communication will be essential in ensuring that “the exercise of that responsibility is to be democratic.”40 As a result, rather than holding the theoretical aim of understanding “the social significance of radio,” as did Lazarsfeld’s project, the seminar opted instead for the instrumental goal of “making communication more effective.” At the same time, the memo expands on Lazarsfeld and Stanton’s close alliance with pollsters by insistently linking communication research and opinion polling as two aspects of a single flow of communication, between the government and the people. (Other actors—such as the networks and advertisers—have dropped out of the picture). The government, they write, generates “a stream of proposal, explanation, and decision:” the people, in turn, produce “an answering stream of counter-proposal, explanation, and consent.” Communication research can aid the government by helping it to more clearly and effectively get its proposals and explanations heard; and it can help the people by making their responses accessible to government, through the use of opinion polling. Indeed, opinion polling, panel studies will be among the most important contributions communications research can make: research will primarily study how effectively the government has got its message across, not aid it in crafting the message. Despite the confident assertion that “the same research could, to be sure, be used to turn communication into powerful propaganda, if those responsible chose to do so,” the memo’s authors maintain a firm confidence that the research insights from their various disciplines and from marketing will bolster the war effort significantly.41

The memo closes with a description of the institutional resources that will enable communications research on the government’s scale to proceed. Gallup and Roper’s polling agencies are mentioned by name, as are Princeton’s Public Opinion Research Project, Columbia’s Office of Radio Research, and the Graduate Library School at Chicago. The members of the seminar offer to immediately begin deploying the existing research in service of government–popular communications, “if they are able to secure the help they need – cooperation from those concerned, and funds to meet the costs.” They end by calling for “an institute of research in communication… similar to the national institutes for research in economics, which would, in assigning responsibility, ensure the comparability of findings, and their pooling in some central formulation and reporting.”42 Through the 1940s, this realignment of existing institutions with government agencies would dramatically change the landscape of the field.

The Harvard Radiobroadcasting Research Project: Research stops at Politics

In tying the future of communications research to the government even more fundamentally than Lazarsfeld had tied it to corporate interests, the changes described by the Communications Seminar continued a process of exclusion that would ultimately supplant the methodological diversity that hallmarked earlier communications research. A good example of how the new focus on government strangled ongoing research can be seen in the history of one of the first organizations to receive the communications seminar’s memorandum: the Harvard Radiobroadcasting Research project, which began in 1940. The Harvard project was directed by Carl Joachim Friedrich, like Lazarsfeld an émigré who had arrived from the German speaking world before political events made it entirely necessary. Friedrich’s office was much smaller than Lazarsfeld’s, did not mount any large-scale statistical studies, and only published six article-length reports. In this, more limited scope, it was typical of the sort of university research in mass communications that was taking place not only at Harvard and Columbia, but around the country—in early 1939, Lazarsfeld identified radio research programs underway at Ohio State, Kansas State College, and the University of Wisconsin, among others. A substantial proportion of this research was done to improve the educational effectiveness of radio, or to use it for community outreach; but other studies at Wayne University and the University of Mississippi were engaged directly in the sort of attitude studies of listeners that Lazarsfeld and the networks were interested in.43 Friedrich’s study, like most of these others, has slipped below the notice of historians of communications, although its reports are still occasionally cited by scholars of early radio.

Friedrich’s study is important not only because it was one instance in a proliferation of research, however: it also represents an attempt to expand communications research by advancing producer-side studies in general, and particularly the relationship between the government and the networks. Friedrich was a political scientist (his best known work was a collaborative study of totalitarianism with his student Zbigniew Brzezinski44), and the project’s studies chiefly looked at various aspects of the interaction among the courts, Congress, the Federal Commmunications Commission, and radio broadcasters. But while, it represented a new path in research, Friedrich maintained strong ties to the burgeoning communications infrastructure. Like Cantril and Stanton’s initial proposal, the Harvard project was initially funded by John Marshall at the Rockefeller Foundation; and its only full-time employee was Jeanette Sayre, whose chief qualification for the job was a year of work with Lazarsfeld’s Office of Radio Research. Friedrich himself continued to teach in the government department at Harvard, including leading a seminar in communications whose visiting guests included David Riesman and Frank Stanton.

The industry was happy to see political studies of radio being done, only provided they supported the broadcasters interests: one NBC executive wrote to ask Friedrich’s support against some of the wartime controls since “if the Fly orders stand unmodified, network broadcasting, as we have known it, is at an end.”45 The researchers could not help being taken for political actors themselves—although the final publication maintains a studious neutrality, Sayre reported to Friedrich that one highly partisan new-dealer she met greeted her by asking if she was from “that lousy Republican outfit in Harvard that had sent out the questionnaire on Federal Agencies.”46 But ultimately, their investment in political issues scared off other communications researchers. Lazarsfeld declined to publish the project’s study of radiobroadcasting control, claiming at the time that it was too “anecdotal” within the context of the ORR’s work. The line between anecdote and a level of political independence that could damage the fledgling field, however, was so close that excluding one might mean excluding the other. Lazarsfeld himself later admitted that he had “a more complex motivation characteristic of someone responsible for a new and struggling institution,” stemming from his desire to put the ORR in a mediating function rather than take positions that might be unpleasant to either the FCC or the networks.47

This is not to say that communications research at in the Harvard school of government failed solely because of insidious political pressures. Lazarsfeld did, later, colloborate with Friedrich and Sayre to publish a study on the less volatile topic of foreign language broadcasting. But ultimately, Friedrich’s project did fall by the wayside, and largely because the field was not interested in accepting political research into the fold. The Rockefeller Foundation declined to continue funding the project after the first grant expired in late 1941. Nor did Harvard, observing how radio research at Columbia and Princeton had proceeded largely without university funding, see any benefits of paying for its own interdisciplinary program in radio research. A university committee convened on the topic concluded that although “study of radio is part of the business of a great education institution,” “there seems to be no reason why they should not proceed within the existing organizations and facilities for research purposes.”48 Thus at once, the Harvard project captures two important trends within the field: the failure of the large private universities to engage systematically in radio or mass communications research, and the disinterest of the field as a whole in studies of control. As the war came more and more to the fore in new work in mass communications, both trends would come to dominate the field.

The war, the industry, and the sequestration of a field

Carl Friedrich’s departure withdrawal from academic communications research was early, but not was only part of broad shift that saw the new field, in its consolidation, begin to shed much of the ideological and methodological pluralism of its early years. Beginning with the war and culminating in the new research institutes of the 1950s, the field began to orient around a new research model, while the personnel and techniques that characterized it in its formative years brought their insights either back to the disciplines from which they came, or sought to apply them working for the government or private sector.

Nationalizing the war effort

The immediate result of the war, and the conscious response to it by the leaders of the field in the Communications Seminar, was an immediate focus on research that would be of immediate use in the war effort. As a result propaganda research, monitoring of radio broadcasts, and domestic opinion research were given the most prominence. Producer-oriented studies of the American broadcasting system fell by the wayside. Methodological innovations became conceived in terms of their utility to the federal government, which by its nature ruled out research which, like Friedrich’s Radio Control study, was neither directly scalable and, moreover, presumed to mediate between government and business rather than serve the former.

For the more quantitative research techniques, this represented an enormous boom, so much so that Harold Lasswell could describe private research in entirely utilitarian terms as an incubator for government work. “The special contribution of private agencies is in the sphere of tool-sharpening, rather than tool-using. Tool-using is a mass duplication job, best done by central direction and firmly knit bureaucracies. Tool-sharpening, on the contrary, calls for new ideas, new devices, new kinds of personnel. Here is where the alert private agency whose resources are mobile can deliver its most telling aid to the national effort.”49 The Rockefeller-funded communications projects recast themselves as just this sort of tool-sharpeners in the service of the war effort. Marshall encouraged Lasswell’s Library of Congress study and Lazarsfeld’s ORR to quickly assemble lists of the staff that they had lost to wartime mobilization, as proof of the benefits to be gained from their research. The numbers were impressive: Lasswell was able to name 39 people trained at his program to the Department of Justice’s special defense unit alone, and when the Office of War Information was established in 1942, it drew many of its staff out of the many Rockefeller projects.

The major new projects in the field also tended to be the same type of tool-sharpening endeavors, even when their founders came out of similar backgrounds to the earlier studies. The Rockefeller foundation authorized a substantial grant for a study of Totalitarian Communications at the New School for Social Research in 1940, shows much of the shift in the field. The new project was led by not one but two German émigrés, Ernst Kris and Hans Speier. They, like Lazarsfeld or Friedrich, brought diverse theoretical backgrounds to the study of communications; Speier as an economist and sociologist who could work with large social issues, and Kris as a psychologist (trained by Freud in psychoanalysis). But, constrained by its political program, the study hardly reflected on the domestic system. Indeed, trust in indigenous American broadcasters came to form a bedrock faith for their work. They once wrote “the foremost result achieved by enemy propagandists in their own countries and elsewhere is the strengthening of distrust of mass communication.” Their job was not to cause suspicion of communications in democratic society; it was to unveil one of the sources of such suspicions. Even in its semantic construction, Speier and Kris argued, totalitarian communication differed fundamentally from democratic communication. Through studies like this, the horizons of mass communication studies shrank even as their size and number proliferated.50

The strand into business

The mobilization of the existing research institutes towards war research, and the loss of much of their staff to the OWI and other wartime offices, marked a major turning point in the relation of academic communications research to the corporate sphere as well. The result was the quick collapse of the uneasy alliance between networks and academics that had characterized the Princeton Radio Project and the earlier interplay between research like Stanton’s doctoral thesis and the research department of the networks. Therefore, even as the wartime propaganda agencies built on the work of the early communications researchers and drew away much of their staff, significant numbers of researchers also departed into the private sector to continue lines of thought with few government applications.

But while commercial communications research was sidelined in the Rockefeller-funded projects and in the universities in general, it had made significant enough strides that it was able to easily ensconce itself back in the networks. Stanton, when hired at the NBC research office in 1936, had been one of two staff members there; by the early forties, the research office was rapidly adding new staff members and Stanton himself had moved from being a peripheral hire in a small department to, at the request of the CEO, the president of the company. Corporate executives like H.M. Beville at NBC no longer needed academics to tell them what their data was good for: the rush of social-scientific interest made them realize its value and shed their earlier complacency about their business model. Academic training would remain a valuable asset in the networks for decades—in 1983 the directors of social research at NBC, CBS, and ABC had all been trained by Lazarsfeld51—but universities had shifted to being a locus of training from a source of actual research findings.

Others followed even farther than had Lazarsfeld in seeing the expansive records of the network as inviting important new research in social science. Matthew Chappell, an academic psychologist who collaborated on a study of the radio audience with a leading pollster, breathlessly concluded in 1943 that “all advertising and media research is psychological research—psychological research of such scope and magnitude that no university could dream of conducting it.” That Chappell identified one of the revolutions in progress in 1943 as “the migration of research from the seclusion of the academy to the field of competitive enterprise”52 might seem to fly in the face of the conclusion that the government gradually took over the functions that foundations had nurtured in university incubators. In fact, though, it simply represents another side in the fruition of the applied techniques developed by the researchers. Just as, by 1943, the techniques of Lasswell’s office were largely well enough developed to allow government agencies to apply them without further exploratory research of their own, so too could Lazarsfeld’s research be applied by radio networks and advertising agencies, who were eager to know about the effects of communication on the individual in large part so they better refine their tools.

Three defections from Lazarsfeld’s team during the war show how succesful the applied techniques of the Office of Radio Research could be in the business world. Two have been mentioned above in passing: the twin hiring, by the leading advertising firm McCann-Erickson, of Herta Herzog and Hans Zeisel in 1943. Herzog was initially hired by Marion Harper to do qualitative research—the kind of follow-up, in depth interviews like those that formed the basis of the Invasion from Mars study (of which Herzog was the co-author) or that explored the choices made on the program analyzer. While Zeisel, another old Austrian associate of Lazarsfeld’s, gravitated back to academia (he took a position as an instructor of statistics at Rutgers at the same time he moved to McCann-Erickson, and in 1953 became a law professor at the University of Chicago), Herzog remained a prominent force in advertising—after rising to vice-president at McCann-Erickson, she became a founding member of the renowned firm Jack Tinker & Partners. Another immigrant hired by the Radio Project, Ernest Dichter, brought Freudian methods even farther into mainstream corporate culture. After moving from the ORR to work as a staff research psychologist at CBS, he became famous in marketing as the founder of “motivation research,” which sought to use Freudian psychological categories to influence consumer behavior.

Thus, it was not only the sophisticated statistical techniques of the early corporate studies that came to dominate (like the program analyzer) from the behaviorist tradition. The core elements of Herzog’s and Dichter’s work lay in Freudian antecedents; they privileged the analyst’s creative powers against amalgamative statistical techniques. But they continued to share with the early psychological studies an individualistic outlook about communications effects that looked at the individual as the subject of a procedure, rather than the participant in a process. In Dichter’s case, it was clear from near the beginning that his research was so instrumentally oriented that it would only fit in as corporate work. But even within the academic discipline, interest was shifting to this sort of instrumental research, and away from the larger aspirations of some of its founders.

Epilogue: Professionalization and Diaspora

In 1959, Bernard Berelson declared in the Public Opinion Quarterly that communication research was “withering away.” The former student of Douglas Waples and collaborator of Paul Lazarsfeld, who had been called the “Great White Hope” of communications research, pronounced, “the innovators have left or are leaving the field, and no ideas of comparable scope and generating power are emerging.”53 He acknowledged the emergence of several new approaches, including a “broad historical approach” led by David Riesman and a “journalistic approach” represented by the new communications schools, but argued that these approaches had either been largely ignored, or were not backed by the same field-shaping intellectual force of the previous generation of communications theorists. Those involved in the expansion of the field into journalism and in the formation of new Ph.D. programs scoffed at Berelson’s claims. Wilbur Schramm wryly suggested death “is a somewhat livelier condition than I had anticipated.” New doctoral students, a continuing synergy across professional and scientific disciplines, and the emergence of a dedicated class of researchers without other disciplinary formations made communications in the late 1950s, he thought, “an extraordinarily vital field.”54

But though Berelson’s claims were incomprehensible for those who saw greater focus in the field, his argument nonetheless points toward a significant shift in the discipline during and following the war. Beginning with Cantril’s defection to public opinion research in 1938–39, almost all of the major figures in early communications research slipped away into other fields. Berelson identified four chief figures in early communications research, and noted that of the three survivors (Kurt Lewin had died in 1947), all had retreated into other disciplines: Lazarsfeld into mathematical applications of social research, Lasswell into political science, and Hovland into increasingly detailed studies of cognitive processes. Indeed, if one looks at the primary contributors to communications publications in the late 1950s and early 1960s, the attrition from the early generation is far greater than can be attributed to death alone. In addition to the five mentioned above, Berelson himself had abandoned most systematic communications research, Friedrich had switched to more conventional political science work even before the war ended, and most of Lazarsfeld’s collaborators at Columbia had moved either into sociology or into the private sector.

What remained was certainly an active field, and one not necessarily hostile to new intellectual influences. Indeed, as researchers of the later history of communications have shown, this was the period when other methodological innovations, such as the information theory of Claude Shannon, began to play an important role in the evolution of the field. But at the same time, it was one that maintained a considerably smaller purview of research. The atomization of society in the television generation was a major concern in works like David Riesman’s The Lonely Crowd. Such concerns, however, lay outside the restricted range of problems in the new field; Riesman himself had been involved in the field in its early days, coming to speak to Friedrich’s communications seminars at Harvard and working on the committee on communications that Berelson led at Chicago. But although he came more slowly than Berelson to the conclusion that communications research was atrophying, he wrote a sort of concurrence to Berelson’s article lamenting lamenting the lack of “small-scale empirical sorties” to answer “large-scale questions” as many of the Columbia and Princeton studies had done.55

Instead, the study of short-term, measurable effects came to lie at the heart of training in communications. Producer-side studies, like Friedrich’s analysis of political control or Adorno’s studies of the music industry, waned. As the field began to train students, it prepared them more and more for the sort of jobs that the earlier mass communications studies had already populated: in broadcasting, in advertising, and in government. The newly ascendant cold war bureaucracy, in particular, encouraged the growth of studies of international communications, which became, even by Paul Lazarsfeld’s estimation, the most dynamic site in an intellectually atrophying field.56 Perhaps the major figure in this new institutional consolidation was Wilbur Schramm, whose 1954 reader The Process and Effects of Mass Communications became one of the definitive texts of this newly restricted field.57 Reviewers hailed it for bringing the unity of a scientific endeavor to a fractured field: one praised it for overcoming the field’s “many false starts, a good deal of esoteric material and an apparent inordinate emphasis on methodology.”58 But the unity it brought had its price. The book began as a reader for the United States Information Agency (the independent federal agency that oversaw many aspects of public diplomacy, including the Voice of America broadcasts), and its main groupings reflect the interests of a broadcaster: it moves through getting the attention of an audience, “getting the meaning understood,” “modifying attitudes and opinions,” and culminates in a discussion of “special problems of achieving an effect with international communications. By comparison, an earlier reader edited by Berelson had included several essays on media control, popular culture, and their standing in a democratic society.59 Such topics did not necessarily become taboo; but rapidly, it became clear to those who were interested in them that communications would not be the arena in which to pursue them.

This change reflected a new ossification that marked the waning of the methodological pluralism of the late 1930s. Communications was a discipline, now, and now longer just a field. And it was a discipline with considerably different structures. The war had been a brush fire that cleared out the old dominant institutions and prepared the ground for a new monoculture. The agencies that formerly lay at the heart of communications research fed the flames of federal research; but interest in communications research in Washington was bound to temporary offices and temporary projects, and when the offices at the OWI and Library of Congress closed down, the ecology of the field was radically different. The leading experts in research had moved on, severing communications’ infant ties sociology and political science: now it operated as a series of research projects in large part funded by outside forces. In some ways, this weaning was necessary for the research to continue; once it was clear that Lazarsfeld, Lasswell, and the rest were not going to devote their careers to the furthering of communications research, their absence helped clear the ground for a new generation of researchers who would. But it was not only individuals who were lost to the field—it was the rich variety of backgrounds that made the early days of research so innovative. Outsiders like Robert Lynd and Theodor Adorno, who were acknowledged to have important contributions they could make (even if they did not always actually contribute), had little role to play in the congealed discipline. As a result of this transformation, it no longer held its earlier promise of offering a theory of society as a whole. Part of the cause was undoubtedly a naturalization of the media—the American social scientists of the 1950s were further removed from the creation of the electronic media in the United States, and many of them could come to seem them as almost natural features of the cultural landscape. It only reinforced their complacency more that they in fact were, much more than their predecessors, American social scientists; with peace in Europe, the influx of Europeans with diverse research interest would not be able to play a formative role in the field.

What was still as known as communications research began to be removed, even spatially, from the foremost writers in its predecessor fields. When Lazarsfeld moved away from research, the new centers of communications arose mostly in the large public universities of the Midwest, particularly at the centers founded by Wilbur Schramm at Iowa and Illinois. To be sure, Stanford did gain a Schramm-founded center of its own: but other private research universities like Harvard, Chicago, and Columbia would not again consider placing radio or communications research at the front of their institutional missions. But while they did not always retain the institutional ties to communications research, the universities, like the broadcasting and advertising corporations, continued to benefit from the significant thinkers who had been drawn to it in its early years. Just as communications itself had taken in the exiles and the marginal researchers from other countries and disciplines, its own diaspora would reverberate far afield.

Works Cited

  1. Christopher Simpson, Science of Coercion: Communication Research and Psychological Warfare, 1945–1960 (New York: Oxford University Press, 1994), 3.↩︎

  2. The term “received history” is from Gertrude J. Robinson and Willard D. Rowland, Jr., "Preface," Communication 10 (1988): 93. The canonical example of this account, due to the preeminence of its primary author, is Wilbur Lang Schramm, Steven H. Chaffee, and Everett M. Rogers, The Beginnings of Communication Study in America : A Personal Memoir (Thousand Oaks: Sage Publications, 1997). By Schramm’s account the “four fathers” are Paul Lazarsfeld, Harold D. Lasswell, Kurt Lewin, and Carl I. Hovland. (This list of founders was first established by Bernard Berelson, "The State of Communication Research," The Public Opinion Quarterly 23, no. 1 (1959).) Younger writers in a similar tradition (most notably, Everett M. Rogers, A History of Communication Study : A Biographical Approach (New York: The Free Press, 1994).) sometimes demote these four to simple forefathers, and consider Schramm himself (who founded three communications research institutes, including the first major program to offer the communications Ph.D., at the University of Illinois in 1947) the true founder of the discipline. Another traditional account is Shearon Lowery and Melvin L. DeFleur, Milestones in Mass Communication Research : Media Effects, 2nd ed. (New York: Longman, 1988).

    The equivalently canonical text for the revisionists is Todd Gitlin, "Media Sociology: The Dominant Paradigm," Theory and Society 6, no. 2 (1978). Some (See, for example, Robinson in Everette E. Dennis and Ellen Wartella, American Communication Research : The Remembered History, Lea's Communication Series (Mahwah, N.J.: Erlbaum, 1996). have suggested that the traditional emphasis on Schramm and the four fathers ignores other important critical antecedents of modern communications research. She and others propose that the history of the field should concentrate on gathering the roles of the various scholars who have influenced modern researchers in the field.↩︎

  3. For instance, Brett Gary, Timothy Richard Glander, Origins of Mass Communications Research During the American Cold War : Educational Effects and Contemporary Implications, Sociocultural, Political, and Historical Studies in Education (Mahwah, N.J.: L. Erlbaum, 2000), Simpson, Science of Coercion: Communication Research and Psychological Warfare, 1945–1960. These studies are notable in that they place the key events in communications’ rise in the 1940s, largely casting aside the story of corporate-academic collaboration in favor of a narrative privileging the history of government intervention.↩︎

  4. John Marshall to I.A. Richards and others, August 16, 1939. Rockefeller Foundation Archives, Box 223, Folder 2672.↩︎

  5. Darrell M. West, The Rise and Fall of the Media Establishment (Boston: Bedford/St. Martin's, 2001), 131.↩︎

  6. It is typical of the close connections between academic research and commercial applications in the field that Stanton later summarized his research, which suggested that aural perception of copy was more memorable than visual, along with a number of other similar psychological findings into a pamphlet designed to convince companies to advertise on NBC programs. Columbia Broadcasting System, Exact Measurements of the Spoken Word (New York City: Columbia Broadcasting System, 1936).↩︎

  7. Frank Nicholas Stanton, "A Critique of Present Methods and a New Plan for Studying Radio Listening Behavior," (The Ohio State University, 1935).↩︎

  8. Frank Nicholas Stanton and Mary Marshall Clark, Frank Stanton Oral History [Transcript of audiorecording] (Columbia University, 1996 [cited June 15 2006]); available from↩︎

  9. Letter from Hadley Cantril to John Marshall, May 11 1937. Rockefeller Foundation Archives Group 1.1, Series 200R Box 271 Folder 3233. All Rockefeller Foundation Archive citations hereafter are in Group 1.1 Series 200R unless otherwise indicated.↩︎

  10. John Marshall to Paul Lazarsfeld, December 29, 1941. Rockefeller Foundation Archives Box 222 Folder 2633. Hadley Cantril and Gordon Willard Allport, The Psychology of Radio (New York and London,: Harper & brothers, 1935).↩︎

  11. “Propositions Regarding the Problems and Findings of the Princeton Radio Research Project, Rockefeller Foundation Archives Box 271 Folder 3237↩︎

  12. “…ein mittelding zwischen einem Institut fur Marktforschung, die damals in oesterreich noch nicht entwickelt war, und einem Universitaetsinstitut fuer Sozialforschung.” Paul Neurath, “Die methodische Bedeutung der RAVAG-Studie von Paul Lazarsfeld: Der Wiener Bericht von 1932 und seine Rolle fur die Entwicklung in America.” Desmond Mark, Paul Lazarsfelds Wiener Ravag-Studie 1932 : Der Beginn Der Modernen Rundfunkforschung (Wien: Guthmann Peterson, 1996).↩︎

  13. See Ibid.↩︎

  14. The Rockefeller funding, and Lazarsfeld’s directorship, would last through a period of name changes in administration that are testament to the sometimes precarious institutional situations in which Lazarsfeld operated. The Princeton Radio Research Project (which, despite its name, was actually located at the University of Newark) became, in 1940, the Office of Radio Research at Columbia University; the Office of Radio Research eventually became Columbia’s Bureau of Applied Social Research. With some notable exceptions, the staff and researchers were retained at each of these junctures. Since the ostensible university affiliation was often not particularly important to the auspices under which research took place, I refer to it primarily as the “Radio Project” in the first period and the “Office of Radio Research” in the second.↩︎

  15. The University of Newark research office picked up where the Vienna surveys on tea and shoes described above had left off, with studies including a 63-page opus “Dislike of Milk among Young People. Development of a Method to Measure and Analyze this Dislike.”↩︎

  16. Glander, Origins of Mass Communications Research During the American Cold War : Educational Effects and Contemporary Implications, 118.↩︎

  17. Firsthand accounts of the Program Analyzer are contained in Lazarsfeld’s and Adorno’s accounts of their American research experience in Donald Fleming and Bernard Bailyn, eds., The Intellectual Migration: Europe and America, 1930–1960 (Cambridge: Harvard University Press, 1969). Stanton takes credit for the bulk of the work in developing the concept in Stanton and Clark, Frank Stanton Oral History. A more recent study of the topic is Mark R. Levy, "The Lazarsfeld-Stanton Program Analyzer: An Historical Note," Journal of Communication 32, no. 4 (1982).↩︎

  18. Delbert C. Miller and Warren W. Philbrick, "The Measurement of Group Learning Process by Use of the Interactional Telemeter," American Sociological Review 18, no. 2 (1953).↩︎

  19. See Levy, "The Lazarsfeld-Stanton Program Analyzer: An Historical Note," 35. Levy gives no reason why Marion Harper, Jr., McCann-Erickson’s president, volunteered to pay Lazarsfeld the money. It is possible the agreement is connected to the fact that around the time McCann-Erickson licensed the Program analyzer, they also hired two collaborators of Lazarsfeld’s dating back to his time in Vienna, Hans Zeisel and Herta Herzog, to work at the agency.↩︎

  20. Elizabeth Kolbert, "Test-Marketing a President," New York Times Magazine, August 30 1992, 20.↩︎

  21. Albeit only in a very crude form; homes were automatically classified as belonging to one of four income groups on the basis of telephone exchange and street address, without any attempt to verify the data.↩︎

  22. H.M. Beville Jr., Social Stratification of the Radio Audience (Typescript, Princeton University Library, 1939).↩︎

  23. See Neurath, “Die methodische Bedeutung der RAVAG-Studie von Paul Lazarsfeld” in Mark, Paul Lazarsfelds Wiener Ravag-Studie 1932 : Der Beginn Der Modernen Rundfunkforschung.↩︎

  24. RF 3242, Lazarsfeld to Marshall, December 1939.↩︎

  25. Carl Joachim Friedrich Papers, Harvard University Archives, series 17.25, Box 3, “Radio” folder↩︎

  26. Leo Calvin Rosten, "A "Middletown" Study of Hollywood," The Public Opinion Quarterly 3, no. 2 (1939).↩︎

  27. Paul Lazarsfeld, “Princeton Radio Research Project Plans and Problems,” Memorandum, January 1 1938. Lazarsfeld Papers, Box 26, Folder “Princeton Radio Research Project—8”. Newspaper coverage also conceived of the work in relation to the Lynds’ earlier study: for instance, "Radio 'Middletown' Being Sought for Princeton Research Project," New York Times, January 16 1938.↩︎

  28. Lynd to Marshall, April 17 1940, Rockefeller Foundation Papers Box 223 folder 2673.↩︎

  29. See Sarah Elizabeth Igo, "America Surveyed: The Making of a Social Scientific Public, 1920–1960," (Princeton University, Doctoral Dissertation, 2001).↩︎

  30. Both Lazarsfeld and Adorno emphasized its failures, particularly compared with Adorno’s later collaborative work in the United States, the Authoritarian Personality, in their autobiographical essays in Fleming and Bailyn, eds., The Intellectual Migration: Europe and America, 1930–1960. The longest treatment is David E. Morrison, "Kultur and Culture: The Case of Theodor W. Adorno and Paul F. Lazarsfeld " Social Research 45, no. 2 (1978).↩︎

  31. Lazarsfeld to Lynd, September 17 1940: RF 2674↩︎

  32. W.G. Preston to John Marshall, December 18 1939: RF 3242↩︎

  33. Memo by John Marshall, January 5 1940 RF 3243.↩︎

  34. Other Rockefeller-funded projects, too, were run or staffed by Freudians, like the “Totalitarian Communications” project at the New School, which was co-directed by Ernst Kris, whom Freud trained as a psychoanalyst in Vienna.↩︎

  35. Bureau of Applied Social Research Publication 0141, Exploratory Study on the Psychology of Refrigerator Purchasers: BASR collection, Columbia University Archives. 1.↩︎

  36. Marshall’s original choices for the committee were Geoffrey Gorer, a British anthropologist who came to advise Lazarsfeld’s Radio Project; Donald Slesinger, Lyman Bryson, Richards, Lasswell and Lynd. Lazarsfeld and Douglas Waples, who studied reading at the University of Chicago library school, began as observers and presenters, while Richards dropped out in April 1940, feeling that he had little to contribute to the social science methodology that dominated the sessions.↩︎

  37. John Marshall, Memo, September 1939, RF 2672.↩︎

  38. Minutes of the Communications Seminar, May 8 1940. RF 2678.↩︎

  39. Ibid.↩︎

  40. Lyman Bryson et al., "Needed Research in Communications," (Harvard University Archives, Carl Joachim Friedrich Papers, 1940). The memo is signed by Lyman Bryson, Lloyd A. Free, Geoffrey Gorer, Harold D. Lasswell, Paul Felix Lazarsfeld, Robert S. Lynd, John Marshall, Charles A. Siepmann, Donald Slesinger, and Douglas Waples.↩︎

  41. Ibid., 3, 1, .↩︎

  42. Ibid., 20–21.↩︎

  43. Isabelle Wagner, "Current Radio Research in Universities," Journal of Applied Psychology XXIII, no. 1 (1939).↩︎

  44. Carl J. Friedrich and Zbigniew K. Brzezinski, Totalitarian Dictatorship and Autocracy (Cambridge,: Harvard University Press, 1956).↩︎

  45. Letter from James Angell June 4, 1941 CJFP Series 17.25 Box 1 Folder A↩︎

  46. Notes from J. Sayre on meeting with Leslie Evan Roberts, CJFP Series 17.6 Box 26.↩︎

  47. Lazarsfeld, “An Episode in the History of Social Research”, in Fleming and Bailyn, eds., The Intellectual Migration: Europe and America, 1930–1960, 314–16. The study was ultimately published in the Political Science Quarterly, and independently by the Harvard project as Carl J. Friedrich and Evelyn Pearl Sternberg, Congress and the Control of Radiobroadcasting, I- Ii (Cambridge, Massachusetts: The Radiobroadcasting Research Project at the Littauer Center of Harvard University, 1944).↩︎

  48. “Report of the Radio Policy Committee of Harvard University,” October 1940. Carl Friedrich papers, Series 17.62, Box 5↩︎

  49. Letter from Harold Lasswell to John Marshall, February 19 1942; RF 2855.↩︎

  50. Report on work of the Totalitarian Communications project, January 29 1942. RF 3102.↩︎

  51. David L. Sills, "Paul F. Lazarsfeld," in Biographical Memoirs, ed. National Academy of Sciences (U.S.) (Washington: National Academy of Sciences, 1987).↩︎

  52. Matthew Napoleon Chappell and Claude Ernest Hooper, Radio Audience Measurement (New York,: S. Daye, 1944), ix–x.↩︎

  53. Berelson, "The State of Communication Research," 1, 4.↩︎

  54. Wilbur Lang Schramm, David Riesman, and Raymond A. Bauer, "The State of Communication Research: Comment," The Public Opinion Quarterly 23, no. 1 (1959): 6, 9.↩︎

  55. Ibid.: 13. On the travails of Berelson and Riesman’s communications committee at Chicago, see Karin Wahl-Jorgensen, "How Not to Found a Field: New Evidence on the Origins of Mass Communication Research," The Journal of Communication 54, no. 3 (2004).↩︎

  56. Paul Felix Lazarsfeld, "The Prognosis for International Communications Research," Public Opinion Quarterly 16, no. 4 (1952–53).. Even Lazarsfeld’s ORR—now rechristened the Bureau of Applied Social Research, and concentrating on far more than just issues of communications—was lured by the appeal of cold war funding, and the expertise gained by the wartime propaganda studies, to issue reports for the government to aid in the dissemination of information. Columbia University Bureau of Applied Social Research and United States Dept. of State International Broadcasting Division, Communications and Public Opinion in Jordan (New York: Bureau of Applied Social Research, 1951). More generally, see Simpson, Science of Coercion: Communication Research and Psychological Warfare, 1945–1960.↩︎

  57. Wilbur Lang Schramm, The Process and Effects of Mass Communication (Urbana,: University of Illinois Press, 1954).↩︎

  58. William F. Swindler, "The Process and Effects of Mass Communications," Political Science Quarterly 70, no. 2 (1955): 290.↩︎

  59. Bernard Berelson and Morris Janowitz, Reader in Public Opinion and Communication (Glencoe, Ill.,: Free Press, 1950).See particularly the articles by John C. Ranney, T. W. Adorno, and David Riesman and Reuel Denney.↩︎