WHAT A GOOD IDEA! IDEOLOGIES AND FRAMESIN SOCIAL MOVEMENT RESEARCH*Pamela E. Oliver and Hank Johnston†Frame theory is often credited with "bringing ideas back in" social movement studies, butframes are not the only useful ideational concepts. The older, more politicized concept ofideology needs to be used in its own right and not recast as a frame. Frame theory is rootedin linguistic studies of interaction, and points to the way shared assumptions and meaningsshape the interpretation of events. Ideology is rooted in politics and the study of politics, andpoints to coherent systems of ideas which provide theories of society coupled with valuecommitments and normative implications for promoting or resisting social change. Ideologiescan function as frames, they can embrace frames, but there is more to ideology than framing.Frame theory offers a relatively shallow conception of the transmission of political ideas asmarketing and resonating, while a recognition of the complexity and depth of ideology pointsto the social construction processes of thinking, reasoning, educating, and socializing. Socialmovements can only be understood by linking social psychological and political sociologyconcepts and traditions, not by trying to rename one group in the language of the other.The study of social movements has always had one foot in social psychology and the otherin political sociology, although at times these two sides have seemed to be at war with eachother. In the 1950s and 1960s, social psychology dominated, and collective behavior theoristssaw social movements as long-lasting panics or crowds. In the 1970s, proponents of resourcemobilization criticized collective behavior theory, and stressed the importance of political andorganizational factors. In the 1980s, social psychologists criticized resource mobilization andpolitical process theories for treating social movements only in organizational and politicalterms, and neglecting the problems of social construction. Snow, Rochford, Worden, andBenford's (1986) article on "frame alignment processes" was central in the socialpsychological turn, and is widely credited with "bringing ideas back in. " t Framing theory hasprovided a way to link ideas and social construction of ideas with organizational and politicalprocess factors. Over a hundred different kinds of frames linked with specific movementshave been identified (Benford 1997).*An earlier draft was presented at 1999 American Sociological Association Annual Conference, Chicago, IL.Thanks to the Project on the Rhetoric of Inquiry at the University of Iowa for the opportunity to present and discussthis paper, and to John Noakes and anonymous peer reviewers for comments on an earlier version. The authorswelcome comments from Mobilization readers.†Pamela E. Oliver is Professor of Sociology at University of Wisconsin, e-mail: [email protected] HankJohnston is editor of Mobilization, Department of San Diego State University, e-mail: [email protected] source was the resurgence of cultural studies and their application to social movement analysis(see Johnston and Klandermans 1995). It is an important trend embracing various perspectives and foci, but willnot be reviewed here as it is tangential to our central argument. Mobilization: An International Jounial, 2000, 4(1): 37-5437

38MobilizationNot surprisingly, frame theory has itself been criticized. Benford's "insider'scritique" (1997) lists several shortcomings in the way the concept is applied in researchstudies, and asserts that the term has become a cliche (p. 415). "Framing" is often inserteduncritically wherever there is a movement-related idea being defined or debated. It has beenpointed out that the concept of frame does not do justice to the ideational complexity of asocial movement (Munson 1999); and that it tends to reduce the richness of culture torecruitment strategies (Jasper 1997: 76). Steinberg (1998) criticizes frame theory as too staticand stresses the contextual and recursive qualities of frames.None of these critiques has identified what we consider to be two central problemsin frame theory: its failure to address the relation between frames and the much older, morepolitical concept of ideology, and the concomitant tendency of many researchers to use"frame" uncritically as a synonym for ideology. Snow and Benford (1988) are often givencredit for insights adopted from the older literature on the functions of and constraints onsocial movement ideologies and renamed as framing tasks and constraints on frames. Theirarticle clearly credits this older literature and specifically says that they are drawing on theolder literature to develop insights about framing processes. In this and their subsequentarticles, they use the terms frame and ideology distinctly and explicitly cite older works.Nevertheless, they neither provide justification for abandoning the term ideology andsubstituting frame in this context nor explain the relation between frames and ideologies.Subsequent scholars have tended to cite the Snow and Benford article and its framing languageas the original work in the area, and to use the terms frame and ideology interchangeably.This leads to muddled frame theory, diverts attention from a serious examination of ideologyand its social construction, and avoids questioning relation between frames and ideologies.Frames and framing processes are powerful concepts. Frame theory's insight intohow movement activists construct their self-presentations so as to draw support from othersis an important process. This line of theorizing has been extraordinarily productive of newresearch and new understandings of social movements. In seeking to back up and revisit aparticular turn in framing theory, we should not be understood as trying to discount the valueof a whole line of work. Nevertheless, the power of frame theory is lost if "frame" is madeto do the work of other concepts. Ideology is of central importance in understanding socialmovements and other political formations, and it is trivialized when it is seen only as a frame.We need both concepts, and we need to understand the relation between them.This is seen most starkly in movements for and against legal abortion. As KristenLuker (1984) argues, these movements are rooted in deeply-held ideologies andunderstandings of the meaning and purpose of a woman's life, as well as in the professionalideologies of physicians. Strong anti-abortion beliefs were in the 1960s rooted in Catholicdoctrine which links sexuality to procreation, condemns artificial birth control, and condemnskilling a fetus even to save the life of the mother (two deaths are morally superior to onemurder); people who live according to these doctrines build lives in which pregnancies canbe accommodated. As the abortion struggles evolved, conservative Protestants also adoptedanti-abortion ideologies which do not necessarily contain all the elements of the coherentCatholic world-view, but strong anti-abortion sentiment remains deeply rooted in religioustraditions and religious world-views. Those with strong anti-abortion ideologies reject abortioneven for the "strict constructionist" reason of saving the mother from the i mmediate risk ofdeath, although laws permitting such abortions do not outrage their moral sense. Initialimpetus for abortion reform was rooted in physicians' desire to clarify the "broadconstructionist" views of the medical necessity for abortion which would include severedeformity of the fetus, and threats to the mother's life and well-being that might includephysical strains of excessive pregnancies or illnesses, psychological distress, and financial

What a Good Idea!39hardship. For physicians, the issue was the right to practice medicine in good conscience,unconstrained by others' religiously motivated intrusions. Physicians were not supporting"abortion on demand," but rather the ideology of themselves as the proper arbiters of medicalnecessity. As the women's movement energized and joined the abortion debate, feministsdeveloped an ideology stressing women's autonomy and need to control their own bodies. AsLuker argues, women who were in the labor force saw pregnancy as capable of disruptinga person's life, valued sex for enjoyment and intimacy, and believed that women shouldchoose to have children when they could devote proper attention to them.Simply renaming these three ideological strands as frames (e.g. religious, medicalnecessity, women's need) would add nothing to the analysis and would, in fact, risk obscuringthe complexity of the belief systems. But this does not mean that frames are unimportant orirrelevant in these debates. Rather, the frame concepts are most powerful precisely if they aresharply distinguished from ideology. The ways in which actors have self-consciouslypositioned the issue over time is very different from what one would think from a simpleextrapolation of the underlying ideologies. Several examples illustrate this. First, Luker arguesthat the 1972 Roe v. Wade Supreme Court decision essentially framed abortion as a churchstate issue: those who filed friend of the court briefs against abortion reform were all religiousorganizations, while those who filed briefs for abortion reform represented a broad spectrumof professional and secular organizations. The decision was constructed in the context of arecent decision that had overturned laws against the sale of contraceptives as representing anunwarranted intrusion of the state and particular religious beliefs into the personal lives ofpeople. Beliefs about abortion were seen (framed) as religious beliefs. Second, the selfnaming of each movement in the politics of the 1970s is a framing turn. From anti-abortionand pro-abortion, the sides renamed themselves as pro-life and pro-choice as the pro-lifemovement sought to position itself in a secular space to reach out to people who did notnecessarily share their religious views, and the pro-choice movement repositioned itself toemphasize its defense of contraception and personal responsibility, with abortion as anecessary backup to failed contraception. Third, both sides have adopted the civil rightsmaster frame. The pro-life movement stresses the right of the fetus to life, while the prochoice movement stresses the right of the woman to control a fundamental aspect of her life.If we think of frames as synonymous with ideologies, we will lack the analytic tools, eventhe very language, for talking about this fascinating instance of the same frame being tied todiametrically opposed ideologies. If we keep the concepts clearly differentiated, we have somevocabulary and tools for talking about how people present their issues in a public space, andwe avoid the danger of simply extrapolating ideologies from their public presentations.We suggest that the turn toward framing theory and away from ideology was largelydue to the legacy of pejorative theories of ideology which laced the social movement writingsin the early 1970s. A second agenda of this paper is to revisit this pejorative legacy and callfor a rehabilitated non-pejorative understanding of ideology in the study of social movements.There is, in fact, a huge literature on ideology to which this paper cannot do justice. Our goalis to revisit the debates that were abandoned in the 1970s, and point to the directions in whichwe think a rehabilitated theory of movement ideology should move.The plan of this paper is as follows. First we summarize the history of the frameconcept and its roots in linguistics and cognitive psychology; then we review the history ofthe ideology concept and its roots in the study of politics. We then discuss the advantages ofkeeping these concepts separate and explore the important issues that are highlighted byconsidering the relations between frames and ideologies. We suggest that frame alignmenttheory correctly captures some of the important particulars of United States political culturein the 1990s, but is misleading for other problems, especially for movements in other timesand other places.

40MobilizationA FRAME IS A FRAME IS A FRAMEThe frame concept is rooted in the study of communicative interaction. Gregory Batesonintroduced the notion of a frame as a metacommunicative device that set parameters for "whatis going on" ([1954] 1972). He showed that interaction always involves interpretativeframeworks by which participants define how others' actions and words should be understood.Twenty years later, frame analysis was introduced to sociological research by ErvingGoffman. In Frame Analysis (1974), and Forms of Talk (1981) Goffman explored types andlevels of framing activities. In Forms of Talk, Goffman discussed the several layers offraming in interaction, and shifted his focus to linguistic analysis of conversationalconventions that mark the application and changes in interpretative frames. Researchersbuilding on Goffman's work have developed an extensive body of empirical knowledge abouthow speech occurs, how cultural knowledge is used, and how these interplay withinteractional intentions and constraints; but this body of knowledge has not been utilized bysocial movement approaches to framing.Within the linguistic tradition, there is divergence between those who treat a frame(or its synonyms, script and. schema) as a relatively fixed template, and those who treat it asmalleable and emergent. Work in anthropological linguistics views frames as fully formedcognitive structures that constitute part of the cultural tool kit of everyday life. Frames arean aspect of cultural knowledge, stored in memory, that permit social actors to move in andout of different experiences as if they were not completely new. Frames are used to explainspeech acts, rituals, and commonly occurring behaviors in other cultures (Hymes 1982, 1974;and Frake 1964). The assumption is that the elements of frames can be elicited throughethnographic interview and reconstituted into a working schema or algorithm. This approachhas also been adopted by researchers in artificial intelligence to explain speech behavior ineveryday situations such joking, gossiping, doing business, lecturing, shooting the bull, etc.(Schank and Ableson 1997; Minsky 1974, cited in Tannen 1993).The other way to view a frame is as an inherently malleable and emergent mentalconstruct. In Bartlett's terms, it is an "active developing structure" (1932) shaped in actionand interaction as additional elements are added to existing structures based on new, incomingdata. Frames are the instruments by which we infer "what is going on," with the caveat thatthey are under constant revision based on new occurrences and unexpected actions by others.Many ethnographic linguists stress the malleability of frames by asserting that the proper unitof analysis is an interactional event or activity. Frake, for example, points out that people are"doing something all the time," and that these activities, not "mental structures," are theproper units of analysis. Gumperz (1982) adds that when people speak, they do things withtheir words within culturally typical speech situations. Frake offers a poignant metaphor forthe fluid and interactive view of frames: Rather than providing a few fixed cognitive mapsto be unrolled and referenced to make sense of situations, culture gives people "a set ofprinciples for mapmaking and navigation, resulting in a whole chart case of rough,improvised, continually revised sketch maps" (1977: 6-7, quoted in Tannen 1993).Imported into the study of social movements, frames have been treated as both fixedand emergent. Early insights into framing focused almost wholly on the interactive level ofanalysis. In Encounters with Unjust Authority, Gamson, Fireman and Rytina (1982) createdartificial focus groups of strangers who gradually were made aware that they were beingmanipulated into giving false statements on camera that could be used deceitfully in a civillawsuit. Gamson and his colleagues focused on the interactive emergence of a frame, of ashared understanding of "what's going on" that they labeled an injustice frame, and the wayin which a public announcement of this frame was essential for rebellion against authority.A few years later, Snow et al. (1986) discussed the improvised and processual quality of

What a Good Idea!41sketch-map frames by elaborating frame alignment processes.Subsequent elaborations of the framing perspective moved to a more fixed conceptionof collective action frames, even though the most influential scholars of framing haveconsistently stressed emergent and processual aspects of framing tasks. This paradoxical effecthas occurred for two reasons. First, the concept of frame resonance (Snow et al. 1986) gaveindividual cognitive schemata an organizational dimension by making their generation astrategic task of the SMO, namely to link the movement's frame to existing beliefs andvalues. By "strategically framing" movement positions in accord with dominant values andfolk beliefs, the SMO elicits greater participation. While strategic framing is a process, theemphasis is on the content. When a collective action frame is recast as something that leadersmust articulate to better "market the movement," the interactive negotiation of "what's goingon here" takes back seat to a one-way, top-down process. The sketch maps are drawn up bythe leaders to be passed on to the grassroots. Simultaneously, the cultural beliefs of the targetsof these efforts are also viewed as relatively fixed, with framers merely putting the right"spin" on their issue to tap into these fixed preconceptions. It would be foolish to deny thei mportance of these processes in the United States in the 1990s, but few scholars with a senseof history would want to say that this is all there is to idea making in social movements.The second source of fixity in framing theory is the growing use of the master frameconcept. Master frames are linked to cycles of protest, and work at the most general level ofanalysis to "turn the heads" of movement participants and movement entrepreneurs to seeissues a certain way. Movement participants draw upon master frames to portray theirperceived injustice in ways that fit the tenor of the times, and thus parallel other movements.Snow and Benford (1992) cite as one example the psychosalvational master frame which TM,est, Scientology, Silva Mind Control, and other groups drew upon in the 1970s. Anotherexample is "rights frame" which was defined by the southern civil rights movement, pickedup by other racial/ethnic movements and the women's movement, and diffused to gay rights,animal rights, abortion rights, fetal rights, and student rights. Master frames are generalassemblages of concepts that are often new and ascendent, but relatively unelaboratedcompared to established ideologies. They are often articulated by early-riser movements andthen used by late-comer movements can draw (Swart 1995; Carroll and Ratner 1996; also seeWilliams 1995, for "rhetorical models" which are utilized rather than master frames).We draw four conclusions regarding frame analysis as it is currently practiced. First,frames are individual cognitive structures, located "within the black box of mental life" thatorient and guide interpretation of individual experience. Frames "enable individuals to locate,perceive, identify and label occurrences" (Snow et al. 1986: 464); and "selectively punctuateand encode objects, situations, events, experiences and sequences of actions within one'spresent and past environment" (Snow and Benford 1992: 137). They are complexinterpretative schemata-not just isolated ideas-which are relevant at different levels ofexperience. Second, frames become important in analyzing collective action insofar as theyare shared by enough individuals to channel individual behaviors into patterned social ones.This presumes an ideal-typical formulation of a f