Notes on : Denzin, N. (1969). Symbolic Interactionism and Ethnomethodology: A Proposed Synthesis. American Sociological Review, 34(6), 922-934.

Dave Harris

There are similarities — problems of social organisation, methodology, socialisation, deviants, social control, face-to-face interaction, and the analysis of science as a social enterprise, basically 'how individuals are shaped by and in turn create elements of social structure' (922), with an emphasis on the subjective side of social life. These two perspectives may not be irreconcilable.

Investigation of both social psychological and sociological problems is required, but approaches have not always been successful. S I and ethno both focus on the individual and subjective aspects. Garfinkel and Cicourel analyse taken for granted expectations by any member of a social order. Scientific is distinguished from everyday activity. Documentary analysis is the preferred strategy. The concern is with how everyday taken for granted meanings are linked with routine patterns of interaction. SI focuses on individual conduct and forms of social organisation, seeing how selves 'emerge out of social structure and social situations. Both suggest a link between persons and social structure with an important role for symbols and common meanings. There are links with structural functionalism, although the unit of analysis is the individual and interaction.

Human beings are seen as capable of analysing their own thoughts and activities thus being able to manipulate symbols and orient their actions towards others, often in a routine way, so that conduct becomes a matter of 'custom, tradition, and ritual'(923). Self-consciousness means that some behaviours are also 'actively constructed in a self-conscious and interpretative fashion'. The empirical problem becomes identifying these different modes of interpretation. There are no perfect differences between interpretative and non-interpretative elements. No objects in our environment carry intrinsic meaning. All social objects are constructs [citing Blumer]. Objects can be anything that can be designated 'in a unitary fashion and around which one can organise action' so the meaning of an object lies in the definitions brought to it, not in itself. 'Meanings typically derive from a group or organised interactional perspective. Human life is group life'. Joint action involves the ability of one human to grasp the direction of the acts of others. Consensual action implies a common community of symbols, but there is always a possibility of negotiation and change with 'certain objects'at least. Selves can produce multiple interpretations and this is what gives 'group life its changing character'. Consensus produces stable patterns of action which implies 'a series of social selves', but humans can undertake 'self – lodging' merging their own identity into those of 'relevant others'. Thus Cooley argues that 'the other exists in "our imaginations of him"' [could be Lacan!]. This produces social bonding and reciprocal relations. Lodging self is not the same as presenting one — 'at some point in the cycle of recurrent interactions, the self moves from the presentational to the lodging phase' [very different indeed from Denzin's later work]. There are variations on the looking glass self which involves 'presentation, identification and subjective interpretation', guided by pleasure or displeasure based on interpretations of the reactions of others, a process of continual evaluation, not an object, especially not 'a calculated and planned object'. We are attracted to those areas where basic features of ourselves have been lodged, and this is 'the basis of motivation'and of patterns of identification. Motivation becomes 'an interactional process' (924) which in turn becomes grounded and ritualised, taking on the form of 'the firm rules of tradition and custom'.

Thus human interaction is not just rational or cognitively developed, but also depends on 'the affective bond between the self and its relevant others'. Empirical 'indicators of self lodging' include personal names, modes of dress or special body movements. Naming, for example shows various levels of intimacy, as does personal variations on speech dress: these are seen as 'an attempt to lodge certain portions of the self in the interaction'. Involvement in conversation in terms of time or topic is another indicator. Frequency of utterances can be used to chart involvement. Interactions can be judged in terms of success at self lodging, and recognition of the attempt.

'Joint actions cannot be resolved solely into individual lines of action'. Observed behaviour emerges in interaction as individual lines of action are fitted together. Such fitting includes taking into account definitions and interpretations of others. Plans of action may be disparate, conflicting or incomplete, so 'interaction may have a variable career', contingent on emerging events, such as the birth of a child. Any rules of conduct might be too vague to prevent conflicting points of view. New points of view in new lines of action, changing creative activity can emerge when old perspectives have 'failed to provide answers' to maintain 'basic forms of thinking and acting common to that world' [classic functionalist notion of moral density?]. Mead was able to use this insight to suggest that social scientists uniquely focused on these challenges — in the dialectical process of confrontation, analysis and resynthesis of perspectives' (925). This anticipated later work on the positive functions of conflict and reintroduced conflict and challenge as an important stage in the formation of group perspectives. For Denzin, interaction actually takes place as a result of a combination of rules — civil legal codes; rules of etiquette like those studied by Goffman; rules which 'display the distinctive nature of enduring social relationships', sometimes derived from the above codes, but sometimes addressing new situations [among them 'white-collar theft']. Sometimes mild deviance from formal rules indicate 'reciprocal lodging' as people relax. Any interaction will display all three categories of rules and will study how people actively construct their own meanings within these standards of conduct. Different collectivities will not always agree on these relational rules. [No preference at this stage for subversive or challenging resistance etc]

Interactionism basically 'endeavours to relate covert symbolic behaviour with overt patterns of interaction', involving an interest in unfolding meaning during interaction. The usual strategy is to start with overt behaviours and then work back to meanings attached to them. Behavioural analysis is not sufficient, nor is just an analysis of meanings or definitions without looking at application in interaction. There is a need to identify different phases of interpretation as the reaction of others become important, and as agreement replaces negotiation. It is the self that becomes the object, meaning that behaviour is to be studied from the perspective of those being studied. This includes occasions of ritual encounter 'where the basic activity lies above the self, or in the interaction process' such as social gains, routine work, participation in religious ceremony. By studying selves, we can soon see what is taken for granted and what is problematic, and this helps us 'escaped the fallacy of objectivism' (926) where the scientist perspective substitutes for the perspective of those actually involved. We take 'the role of the acting other'. This involves us also linking symbols and meanings to social circles and relationships, 'Unless meanings are linked to larger social perspectives, analysis remains largely psychological'. We also have to consider the '"situated aspects" of human conduct', where social situations and their meaning can influence subsequent behaviour. There may be components like whether the interact as part objects, the concrete setting, the meanings brought into the situation and the time taken for the interaction. Variations in behaviour can be produced by different definitions of others and objects '(e.g. furniture, lighting)'. 'The situation as an intrusive variable cannot be ignored' — new people, the failure of mechanical equipment, shifts in levels of involvement can all affect interaction. Both stability and change must be studied, however, observations also become 'subject to personal bias and even ideological preference': concepts are sensitising, only operationalised once situated meanings have been discovered. Standard methods of observation and other methods can be used, in the process of triangulation, since 'each method has restrictions, and if several different methods are combined in the same study, the restrictions of wine are often the strength of another' [naive, and eventually replaced by his idea of the triangle as a diverging crystal]. We also need 'a series of common databases; reliable sampling model… A series of empirical indicators for each database; a series of hypotheses; and a continual reciprocation between data and hypotheses'. Theory is still the common goal, and 'propositions with the greatest universal relevance are sought those ', assuming universal interactional processes. We should end with 'soundly grounded empirical propositions of an all-inclusive universal nature', although interactionism has not yet achieved this goal and is a perspective rather than theory.

Ethnomethodologists ask how social order is possible and for Garfinkel there is both a concern for large collective representations as in Durkheim with an interactionist conception of rules norms and meanings (927). Such rules include an assumption that interaction flows in the temporal sequence, that many topics will only be tacitly recognised, that background and conditions are taken for granted, that definitions hold for the duration of the encounter, that 'any object present in the situation is what it is presented as being', that meanings on one occasion will hold for future occasions, that standard terms, symbols and labels are used to attach meaning, that discrepancies with personal biography and experience are held in abeyance for the duration leading to 'conflict between their public and private definitions'. These rules have to be uncovered by penetrating normal situations, sometimes by disrupting normal social events, in Garfinkel's case through 'quasiexperimental field studies' which appear naturalistic and which challenge existing definitions: students pretend to be boarders in their own home, or pay different amounts for objects purchased, or to violate the usual rules of a simple game. It becomes difficult to challenge these routine rules and sometimes engenders 'distrust, hostility, anger, frustration, and persecution', partially resolved by adopting an experimenter attitude. Garfinkel says that the notion of trust has been infringed, an assumption that everyone shares the same expectations and definitions and everyone agrees to work by them. Without such trust, joint action becomes more difficult. Ethnomethodologists have also studied routine production of persons in social organisations, mental clinics, hospitals, courts, suicide prevention centres and so on. This is to study the special perspectives that members have to deal with comments. So far, studies have suggested that organisations even 'perpetuate themselves through time by generating fictitious records' (928); that comparable organisations assign different meanings to the same events; that the production of organisational records is basically an interactional process and includes many subjective elements [Cicourel]; it is common for members to rely 'on open-ended categories to classify cases' — this is the '"et cetera clause"' where events are fitted into a pattern to confirm ongoing action.

We can see these studies as a contribution to labelling as in Becker, and ethnomethodologists clearly share interactionist's concerns with the social organisation dimensions of deviance. There is another possible convergence where interactionists study the informal structure of organisations, including a moral hierarchy [citing Hughes or Strauss], although interactionists tend to place more emphasis on the formal structure as a factor. Ethnomethodologists suggests that sociologists are concerned with depicting taken for granted affairs of actors, which involves necessary decisions about the relation between sociological concept and observations. Invariably, 'unclassifiable instances appear' to stretch coding schemes or statistical tests, and when sociologist decide whether or not an observation fits that category, they make use of the documentary method of analysis, perhaps unconsciously. So every classification rests on 'assumptions of daily interaction' — some statements are placed in the temporal sequence, some ignored, some common vocabulary is assumed, interactions take place between researchers and actors. The latter is often seen as a good thing when rap or is established, but 'this hypothesis is problematic' and disguises the necessary collaboration by the subject as well as the sociologist — we need to look at 'the routine meanings held by subjects' as well (929).

The assumptions taken from Schutz are important, including the idea of 'an impenetrable barrier between the scientific and the everyday conceptions of reality', and the denial that scientists can ever accurately enter the world of everyday man. The solution is to construct ideal types as rational constructs. Both Garfinkel and Cicourel have constructed models of everyday life based on 'a social game perspective' to grasp the rationality of analysing everyday action.

We need to look at the criticisms of both interactionism and ethno. Theoretically, there are problems with the dramaturgical view, for example, because that overemphasises the project to 'win support for a presented self'. Other conceptions of the self are too vague to ground any empirical observations, and interactionist themselves seem to disagree about the causal status of the self as concept. Generally too few concrete hypotheses have been produced [denied by Hammersley of course]. Interactionism might fail to adequately grasp the larger forms of social organisation, although the sociology of work and organisational settings associated with Hughes might be cited in defence, and collective behaviour can help us grasp 'mass society' [citing among others Blumer 1957 — I don't know it]. Ethno and its 'phenomenological bias'also presents problems for looking at individuals in larger social units. Its demonstration of how taken for granted assumptions actually operate is unclear, and so is its discussion of the documentary method.

Both perspectives failed to 'clearly indicate the source of meanings and definitions' for Denzin [those in politics?]. They offer no way to measure the interaction process. They do offer a way to grasp 'the complex role of interaction in shaping activity'. There are some experimental strategies in Garfinkel, and there are contributions to the sociology of knowledge, as with Garfinkel's analysis of scientific rationalities. Ethnomethodologists's argument that sociological methods are produced in the same way as the meanings of everyday life 'echoes the concern of Mead and others' on the differences between science and everyday activities.

The proposed synthesis is possible on the convergence between the two in terms of the 'meanings given to social objects'(930). Take the 'movement of objects from interpretative to non-interpretative rules'. We can suggest that any event challenging normal interpretation will bring that event into the flow of interaction, perhaps after an initial frustration or groping for meaning. There can be a quick resolution for less important objects, such as when mechanical equipment breaks down at a meeting. However fundamental objects will require further negotiation by being accorded 'explicit interpretative status'. These are as important to study as taken for granted elements. There is no way to classify objects into these two categories a priori. The self may be the most significant object for interpretation, but we need to add 'meanings brought into the situation and perhaps the situation as well', even though concrete situations are 'likely to contain the greatest proportion of non-problematic objects'. Meaning can be treated as 'an element of the covert symbolic act' studied by self-reports and expectations: these might include assumptions about who was going to be present how long it would take, the types of cells to be presented, the degree of knowledge and the types of objects. As interaction develops, overt activities can be linked in to shifts in meaning to study constructions. The extent of interaction can be 'measured by the frequency of joined actions'. Emergent effects can be 'represented by the frequency of disrupted plans of action' and emergent qualities can be studied in terms of the extent to which they involve explicit interpretation.

Thinking of deviance and labelling, we can redefine Garfinkel on trust. If actors assume the other will abide by decisions mutually agreed on, then 'violations of trust become violations of those agreements'. Examples might include not keeping secrets or pretending to interact consensually. We might see the betrayal of trust especially clearly with relations between normals and the mentally ill. We could suggest that continued violations of trust creates strains which in turn 'culminate in attributions of deviants directed towards the trust violator', but this will happen only with certain breakdowns, where more violations of the codes occurred. At the organisational level, members in social control agencies might have to work to validate a deviant label, and again the 'degree of trust violation' (931) will be important, although there is an additional requirement to meet 'collective organisational needs'. We will find more labelling where these needs converge with attributions of deviants — sometimes organisational needs will even dominate, and sometimes labelling will take place without an organisational backing — labelling can become 'a political and ideological issue with members of the social control agencies responding more to political demands'.

There may be conflicting definitions of the relevant objects in the environment, including objects that are taken for granted by some participants but the subject of interaction by others. A sense of dissatisfaction with interactional partners can produce 'reciprocal alienation' and then 'deviance attribution'. We need to study the language of interaction, both 'silent' and 'vocal'. The first turns on things like gestures and bodily controls and often reflect background expectations. The second concerns matters such as 'proper address, naming, tone of voice, and choice of vocabulary' we can take these as an expression of 'the salient features of the silent dialogue' [citing Goffman on behaviour in public places]. interaction involves both languages. We need to see how rules, including implicit ones are taught in socialisation, and how silent languages can be sanctioned. Both perspectives suggest that 'socialisation is never ending… an [sic] ubiquitous feature of all interactions'. The success of socialisation can be measured by 'the ability of persons to fit lines of action together' (932).
Together, both perspectives would help us do organisational analysis. Organisations depend on 'complex spoken and unspoken languages' representing their concerns and prescriptions for action, even where these run counter to formal goals. Each role position might have a special language requiring special socialisation strategies to be tested and reaffirmed within special social orders. These might compete. We need to break organisations down into interactional units of this special kind rather than think of one organisation. An organisation needs to be held together 'if at all' by a few salient symbols [the examples of both universities and mental hospitals] — and the name of the organisation might be the only one agreed.

We can also use both to understand scientific conduct including the development of sociological methodology. Scientists especially are 'self-consciously interpretative' and this produces developments in science. Mead partially anticipated Kuhn on paradigms. Mead and Garfinkel examine scientific rationality and contrasted it to Merton and Parsons. Scientific conduct is 'imbued with elements from everyday life' although this can lead to unfavourable judgement, hence Merton's 'four norms of universalism, communism, disinterestedness and organised scepticism', although other professional organisations also share these. We can now have a better discussion of the 'value free problem'. Science does not avoid values. Scientific norms of rationality, such as official neutrality towards meaning, the disinterest in the real world, and indifference to chronological time, perfect communication, standards of publicity can never be realised fully. Nor can they be in modern sociology. Garfinkel suggests that pursuing these norms leads further away from 'the world of social events… Concrete behavioural analyses of face-to-face interaction'. Sociologists typically overlook irrational elements in their own conduct when they make observations code documents or conduct interviews. But these can never be ignored, and interactionism and ethno can even suggest corrective action. Thus we need multiple observers and multiple methods to overcome individual 'restrictive biases', and this implies that triangulation produces more reliable and valid data. The scientists own unique interpretations must be moderated [but we are not told how different interpretations are reconciled]. Interactions between observers and subjects can affect the data, for example in emergent interaction. Some may be taken for granted. The researcher needs to record interactions carefully 'with a special eye to events they judge to be unique' (933). there might be additional implications [Denzin calls them hypotheses].

Finally, both use documentary method and this can be linked with analytic induction sensitising concepts role taking and 'the strategies of linking individual perspectives with the larger social units'. Garfinkel's quasi-experiments could become a model for more rigourous studies, especially if we acknowledge experimenter effects and subject perceptions.

back to social theory