© The Author(s) 2012
Reprints and permission: sagepub.
reconsidered: The practical
logic of social systems theory
Liu Yu Cheng
National Chengchi University, Taiwan
This article compares ethnomethodology to social systems theory and attempts to
clarify their theoretical concerns, differences and similarities when synthetically
considering the former as a practical logic of the latter. As two poles on a continuum
of research methods, ethnomethodology is a microanalysis of social order and human
agency, whereas systems theory assumes a macro and theoretical abstraction of system
formation and social evolution. When attempts are made to solve the problems of the
micro/macro, part/whole, or individual/society relation, this comparison is conducive
to elucidating how the practical logic is implied on both micro and macro sides, and
to enhance our understanding of the human social world. Three points are suggested.
First, systems theory and ethnomethodology do not contradict each other; they are
complementary in an intricately dialectical process. Second, ethnomethodology can be
observed through indexical rather than objective expressions. Third, due to a higher
level of abstraction of systems theory, this article argues that ethnomethodology can
possibly be utilized to explicate and also be regarded as the practical logic of systems
theory. This comparison may contribute to reformulating systems theory, and also to
some philosophical puzzles.
Autopoiesis, ethnomethodology, indexicality, reflexivity, systems theory
At the heart of the philosophy of social science and sociological theory has been that the
problem of the relation between the individual and society, or between part and whole, is
Liu Yu Cheng, Department of Sociology, National Chengchi University, No. 64, Sec. 2, ZhiNan Road,
Wenshan District, Taipei City, 11605, Taiwan.
426193CSI0010.1177/0011392111426193Yu Cheng2012Current Sociology
582Current Sociology 60(5)
raised and allegedly resolved in one way or another (Callinicos, 2007 ; Coleman,
1990; Elder-Vass, 2007; Giddens, 1984; Parsons, 1937; Sawyer, 2005; Simmel and
Wolff, 1950; Thompson, 1994 ; Weber, 2001 ). Among these authors, it is
agreed either that social structure is imposed on the individual who cannot help but be
influenced by it, or that individuals are the essential parts that constitute the social world,
and also contribute to the only explanatory force of social phenomena. This problem of
the relation between the individual’s action and social structure was recently dealt with
in a way (Thompson, 1994 : 56) that shifted ‘from a static to a dynamic perspective,
from a theory of structure to a theory of structuration’. This shift indicates one thing
important to sociological theory, highlighting thedynamic view of modern societies.
Aside from those supporting either the action or the structure side, this dynamic perspective
advances and deepens our understanding of their relation with regard to social
phenomena. Here we briefly discuss structuration theory and its modification from our
The aim of structuration theory is to put an end to either building an imperialism of
the subject or that of the social subject, since the stance of the two is quite ontological,
according to Anthony Giddens (1984: 2). This attempt succeeds in emphasizing that
social practices ‘are not brought into being by social actors but continually recreated by
them via the very means whereby they express themselvesas actors’ (Giddens, 1984: 2).
The conditions that make these practices possible are reproduced in and through their
activities. Accordingly, Giddens overcomes the traditional view of individuals and structure
as independent of each other, adualism, and proposes instead a duality of structure.
By ‘duality’ he means (Giddens, 1984: 25) ‘the structural properties of social systems are
both medium and outcome of the practices they recursively organize’. Structuration theory
thus focuses more on the idea that structure is continuously sustained and reproduced
in and through the dynamics of action, since for Giddens it is a process or a flow that is
realized repeatedly by social practices and renders structure possible.
However, the attempt fails precisely because of his elaboration of and distinguishing
between structure, structures and structural properties of social systems (Bauman, 1994
: 34–55; Thompson, 1994 : 62–63). The distinction between structure and
system is constitutive of his theory. Structure is comprised of rules and resources,
whereas social systems are regarded as social practices reproduced across time and
space (Giddens, 1984: 17). This distinction is important in that social systems can be
spoken of and formed in a certain structure. Its weakness lies in the fact that structuration
theory overemphasizes the significance of ‘constraint by the social order on the
individual’s action’, or shows how ‘the limitations of individual “presence” are transcended
by the stretching of social relations across time and space’ (Cuff et al., 2006
: 278; Gregory, 1994 : 189). It is confusing to consider, on the one hand,
the ways for understanding where these rules and resources come from, and, on the
other hand, how these rules and resources enter into social practices that reflexively
refer tothe structure and constrain the individual’s action. Neither of these has been
explained by structuration theory itself (Callinicos, 1985; Thompson, 1994 ;
Urry, 1982). Although it is a good attempt to explicate the relation between micro and
macro dimensions, or action and structure, there has to be something else added in order
to bridge the gap between them.
This article draws on ethnomethodology and a systems theoretical perspective to
propose an alternative to elucidating and answering the problem, presupposing that these
two poles of epistemological views are not necessarily contradictory but complementary,
in the sense that ethnomethodology can be regarded as the practical logic of social systems
theory. It is interesting to note that since they have the same heritage as phenomenology,
especially that of Husserl, both ethnomethodology and social systems theory
contribute to the problem elaborated above, and of course to structuration theory.
Theoretical foundations of systems theory and
It has been recognized that there is diversity in systems theoretical approaches.
Luhmann’s theory can be viewed as only one of them; however, its contribution to sociology
and several other fields of research cannot be underestimated (Fuchs, 1997; King,
2001; King and Schutz, 1994; Rasch, 2000). Some of Luhmann’s major works have
been translated into several languages, allowing his thoughts on a general theory of
society to be spread worldwide. Despite the fact that his theory is considered highly
abstract and difficult to apply to social issues and problems, it continues to draw the
attention of many researchers and extends to various fields of study, including religion,
law, art, politics, international relations and media studies (Czerwick, 2008; Green,
1982; Holub, 1994; King and Thornhill, 2006; Seidl and Becker, 2005; Zolo, 1992).
However, his theory is also contentious and debatable in some others’ eyes (Elder-Vass,
2007; Wagner, 1997). It is not our purpose here to introduce or to review those arguments
on whether his theory is convincing or not. When social theory is considered as
an observer observing society and sociological theory as an observer observing sociology,
Luhmann’s theory remains effective as one theory by which it is possible to make
observations on society and sociology.
The features of Luhmann’s theory can be summarized by several concepts that
have been disseminated and utilized, properly or not, in many disciplines: system/
environment, structural coupling,2 autopoiesis,3 distinction, paradox,4 communication,
reflexivity and so on. Among these concepts, the most important may be his distinction
of system and environment by which he substitutes for the distinction between the
whole and its parts (Luhmann, 1995 : 6). His arguments allow us to formulate
long-lasting debates on and to bridge the gap between theory and practice, which had
been discussed by Immanuel Kant, Emile Durkheim, August Comte and the like, up to
the present scientific studies of sociology such as philosophy of science, emergentism,
scientific realism and critical realism. Luhmann’s theory, which he terms ‘a general
theory of social systems’, draws on and is a synthesis of the concept of autopoiesis of
Maturana and Varela (1980 ), the second-order cybernetics of Von Foerster
(1979), the logistics of Spencer-Brown (1969) and the phenomenology of Edmund
Husserl, to name but a few. As a fellow of Talcott Parsons, he also modifies and reconstructs
his action system theory, substituting the role of actor for the concept of ‘communication’
(Luhmann, 1995 ).5
The second theoretical school concerns the ethnomethodology developed by Harold
Garfinkel in the 1960s. Although, most ethnomethodologists would be sceptical of a
584Current Sociology 60(5)
theory per se because of their aversion to macro-theorizing (Kwang-Ki, 2003: xiv).
Whether as a research method or a theoretical perspective, there is something that can be
discerned when Garfinkel tries to explicate what he has observed. Broadly speaking, his
approach derives from the phenomenology of Edmund Husserl and Alfred Schutz, and
also inherits from the linguistic traditions (Leiter, 1980; Mehan and Wood, 1975).
However, his approach is closer overall to Schutz’s concept of a lifeworld than to
Husserl’s transcendental phenomenology. When considering the famous method of ‘conversation
analysis’, the theory’s connection to the phenomenology of the lifeworld is
almost unquestionable (Johnson, 2008). In addition, the relationship between them has
also been claimed by Garfinkel himself, and noted by later scholars.6 So far, as we can
see, the two theorists have something in common with regard to the origins of their theories.
They both inherit from phenomenology and social constructivism in developing and
constituting their views of society and concept of social order, although in opposite ways.
Luhmann’s theory extends several concepts of Husserl’s phenomenology, such as the
concept of horizon and the distinction between cognitive and normative expectations.7
He even claims that the uncovering of the ultimate end of the world, which for him
means the grasping and reduction of complexity, must depend on transcendental phenomenology.
This orientation definitely refers to both Husserl and Schutz (Bednarz,
1984: 55). However, Garfinkel’s theory has affinity with phenomenology; it draws more
on Schutz than Husserl in the sense that Garfinkel (1992 : 76) is more interested
in analysing ‘common sense knowledge of social structures’ than just ‘bracketing’ it. In
other words, his investigations on practices and common sense knowledge, whether of
lay persons or professionals, do not exclude the possible effects of structures, nor do they
indicate that successful ethnomethodological enquiries have to lose or sacrifice the issues
of structure (Garfinkel, 2002: 124). Instead, the practices of social members and their
common sense knowledge of these social structures emerge out of, on the one hand,
their accounting of the settings as reportable and understandable, and, on the other hand,
their accounting practices as a texture of relevances constituting their further accounting
practices. So far as Garfinkel (1992 : 1) ‘treats practical activities, practical circumstances,
and practical sociological reasoning as topics of empirical study’ and regards
those as consisting of ‘an endless, ongoing, contingent accomplishment’, his thoughts on
social yet practical order can be seen as constructivist. It is exactly from this view of
social order as an ongoing and contingent achievement that it is possible to formulate
ethnomethodology as a practical logic of social systems theory.8 This is explained later.
Affinities: Members’ accounting practices and systems’
Following Garfinkel, our understanding of his project starts with the question ‘What is
ethnomethodology?’ This question is also the focus of one of the sections in his book
(Garfinkel, 1992 , 2002). Basically, the term is composed of two words,
ethnomethodand logos; the former designates how social members, lay persons or
professionals, make social phenomena accountable and reportable, whereas the latter
is the knowledge aggregated when studying and attempting to explain the ways that
ethnomethodworks. In Garfinkel’s (1992 : 11) own words, ‘ethnomethodology’
means: ‘the investigation of the rational properties of indexical expressions and other
practical actions as contingent ongoing accomplishments of organized artful practices
of everyday life’. Since it is regarded as a method, there is a boundary set, implicitly or
explicitly, for the study ofethnomethod, limited to ‘the rational properties’ of practical
actions. In other words, the object of ethnomethodology is situated exactly in our daily
life, or in the ‘lifeworld’, that is, everything that happens and that is experienced around
us. Furthermore, what has happened and what has been or will be experienced, since
these are understandable and reportable, assume rational properties. The ‘action’
indicates a certain activity performed by any other social member for some reason and
to achieve some purpose. Following Edmund Husserl, the use of this term by Garfinkel
(2006: 102–104), which compares to ‘conduct’, refers to the ‘meaningful or intentional
The terms ‘meaningful’ or ‘intentional’ indicate that there exists a preconceived condition
as a background or a texture of relevances triggering members’ accounting action.
In the sense that Schutz distinguishesact from action in time series, act can only be
recognized when it has been completed, whereas action is a process that is oriented by
the present, influenced retrospectively and projecting into the future. Therefore the act
consists both of the past and the future. The action is bound to the present. Action is a
process oriented towards projects and thus exists in the form of ‘a continuing achievement’.
The differentiation of act and action suggested by Schutz contributes to the term
‘practice’ used by Garfinkel to demonstrate and understand the ethnomethod as an ongoing
accomplishment of everyday life.9 For the purpose of this research, this distinction
can also be related to the concept of ‘reality’ advocated by Luhmann’s theory in terms of
the temporal dimension underlying the distinction between system and environment.10
Luhmann’s distinction between system and environment consists of his thoughts on
questions such as ‘how is society possible?’ and ‘how can a society observe itself as a
modern one?’ Since his theory is also coloured by phenomenology, Luhmann (1983)
formulates these questions on his own terms. However, this time it is not modified from
the lifeworld perspective but draws on Husserl’s transcendental viewpoint. Hence the
argument is that Luhmann’s theory and ethnomethodology are situated on different levels,
but their underlying mechanisms are quite similar. The mechanism that makes social
order observable and accountable and renders possible a society that observes itself as
modern can be summarized as a dialectical process of, in the former case, settings and
members’ accounts of them, and in the latter case, the system and its environment.
A system can be identified only when simultaneously referring to its environment.
The two sides are inseparable when attempting to understand either one. However, the
condition of this inseparability is based on their distinction from each other. This mutual
reference constitutes a situation of both dependence and interdependence. In addition,
the relationship between system and its environment is asymmetrical since only the
system operates, and through its own operations constitutes its own environment.
Extending this distinction, Luhmann further distinguishes social systems from interaction
systems, and also from society. In this sense, it is not proper to claim that society is
composed of individuals. Rather, individual and society aremutually exclusive. This is
not to say that the argument advocated by Luhmann underemphasizes the role played by
an individual within society. A better way to understand the relation of individual and
586Current Sociology 60(5)
society is to first recognize that these two entities are not identical and that society
cannot be reduced to the individual (Simmel and Wolff, 1950). The element that composes
society is not a person, but something else. This view avoids treating the escalation of
human reason as the ultimate end of human society, and of thus leading to an arrogance
emerging from anyone who is simplyliving in the world.
The distinction between individual and society brings to the fore another differentiation,
between perception and communication. This distinction relates more comprehensively
to the legacy of phenomenology. Despite the fact that this distinction perhaps
parallels the aforementioned act and action, we note here the concept of act and its
replacement with ‘episode’ by Luhmann (1995 : 406). Accordingly, communication,
as well as action, is more social than perception, act, or episode. It is communication
that constitutes social systems, whereas consciousnessqua perception
constitutes psychological systems. Luhmann divides social systems into three types:
interaction, organization and society. ‘Society’ is the most complex social system, and
is also the environment of these interaction and societal systems. The form of ‘system/
environment’ and that of ‘society/interaction’ are mediated by ‘human beings’ as sensors
that contribute to their interpenetration. In addition, communication is also distinct
from action in the sense that action is always attributed communication, but communication
cannot be attributed any action. In other words, action implies communication,
but not vice versa. Hence to analyse the action of social members involves not
only the implicit intentionality that might lead to problems of motivation, but also the
appearance of action, or generally communication, performed by members, including
body gestures, accentuations and so on. This combining of intentionality and appearances
of action is the rationale of ethnomethodology.
Perception, according to Luhmann (1995 : 412), is a ‘less demanding form of
acquiring information than communication’, and it does not ‘depend on the information
being selected and communicated as such’. Luhmann’s transformation from a
psychic-based perception to a social-based communication is made to articulate a
situation ofdouble contingency, i.e. one can perceive that one is perceived. This reflexive
perceiving, compared with explicit communication, demonstrates the features of
ethnomethodology. So far as an observer is concerned, Garfinkel (2006: 102) notices,
‘the rules of procedure by which the manifestations of a person’s “spontaneous life”
are to be realized by the observers; are assigned their significances by the observer in
his task of interpreting an “other”’.
This combination of perception and communication within interaction systems contributes
to the formation of ‘background expectancy’, which Garfinkel also refers to as
‘rational properties’ of indexical expression. To Luhmann, the condition for it lies in
securing the continuation of communications and the fulfilment of expectations, and
also lies in avoiding the disappointments that probably resultfrom within systems’
communicatingabout others. The combination also structures the ways whereby the
objects are able to be perceived. This structuring has different names, such as Luhmann’s
‘institutionalization’, Garfinkel’s (2006: 117) ‘standardization’, or even, loosely
applied, Husserl’s ‘horizons’ or ‘finite province of meaning’. Since perception renders
observation possible without demanding precise information, it must rely on the
‘social dimension of perceptible meaning as a selector’ to maintain its boundary with
communication (Luhmann, 1995). This is how we understand the distinction between
social systems and psychological systems, and how we recognize that this distinction
parallels the distinction between system and environment. Perception and communication
are only analytically separated. Hence perception is necessary to interaction systems,
but a social system can only emerge and continue its operation when perception
is experienced and attributed as communication.
There is a further question concerning the above ‘social dimension’: how can it be
formed and recognized as ‘social’ when perception is necessary to make an observation?
The answer requires considering the concept of autopoiesis and a feature of interaction
systems: ‘presence’. Simply speaking, the concept of autopoiesis means that a system
reproduces the element that constitutes itself and that of presence indicates ‘people’s
being together there guides the selection of perceptions and marks out prospects for
social relevance’ (Luhmann, 1995 : 415). We can see here the affinity with
Garfinkel’s analyses of members’ accounting practices. The operation of a system can
only be realized in the present, and each operation, in the sense of autopoiesis, implies
that reality is always constructed within a system with regard to the present (Luhmann,
2002b: 136–137). The condition of presence suggests, on the one hand, that delimitation
has to be worked on those that can be perceived, and, on the other hand, that the intentionality
and appearances of action are linked to the distinction of communication and
perception. Therefore, it explains how ethnomethodological conversation analysis is
possible. In the course of conversations, perceptionqua presence requires not only
being attributed as communication being communicated by other social systems, but
also waiting for further communication to justify or disqualify the expectation implied
in a system’s earlier operation. As Luhmann (1985 ) elaborates, this is how law is
formed by the distinction of society and interaction.
Ethnomethodology reconsidered and the concept of
When turning to the practical features of action analysed by Garfinkel, the above combination
of perception and communication within members’ accounting practices brings
a practical logic of systems theory to the fore. Social members, in line with Durkheim’s
aphorism, act out their understanding of situations in terms of certain ‘rules’, ‘the objective
reality of social facts’, that are not noticed or even perceived by members using
them (Garfinkel, 2007: 14). These ‘seen but unnoticed’ rules permeate daily life and
consist of each action performed ‘for all practical purposes’. They are taken for granted
when members make the settings observable or expect others’ reactions. This principle,
on which all actions and expectations are founded, contributes to meaningful communication
and interaction among members and others. In other words, something has to
be hidden when a meaningful communication is acting out. Otherwise nothing can be
observed and communicated. The situation of ‘something has to be hidden’ can be interrupted
once social members consciously recognize the rules that they apply through ‘a
breach of the background expectancies of everyday life’ (Garfinkel, 1992 : 54). In
this way theobjective structure taken for granted by social members can be temporarily
sabotaged in order to let something come out of this interruption. It is not impossible
588Current Sociology 60(5)
that this something might be ‘distinction’, as systems theory indicates. It is exactly
this starting point that bridges Luhmann’s theory and ethnomethodology, and then
contributes to micro–macro debates.
The starting point discloses the concept ofreflexivity discussed by both Garfinkel and
Luhmann.11 Although there are differences between them, for both of them the concept
refers to a dynamic, ongoing and self-referential process. It is also a dialectical one that
occurs in ethnomethodology between members’ accounting practices and their accounts
of the settings, whereas in systems theory it is between a system and its environment.
For both, the concept of reflexivity refrains from mere projection or reflection. Without
it, social members or observers cannot see what might be concealed in daily common
knowledge as background expectancies that support and sanction their action or communication.
Hence it is possible to make the operation of reflexivity visible when
observers observe each other and know that this mechanism operates with distinctions
or ‘blind spots’ (Luhmann, 2002b: 190). In this sense, theobjective reality of social facts
is realized by ‘blind spot’, which ‘has to be assumed to be the unity of the system/
environment distinction, when this distinction is employed’ (Luhmann, 2002b: 136).
The concept of reflexivity not only avoids mere projection or reflection, but also
describes members’ accounting practices and accounts of the settings as necessarily
indexical, instead of objective. To some extent, indexical expressions support and constitute
objective expressions. Therefore those expressions that have been recognized as
objective are in fact either backed by indexical ones or they are indexical. Following
ethnomethodological analysis, for example, daily conversations between members are
always context-oriented. These are categorized and can thus be observed by the distinction
between rational and irrational. It is rational because it is recognized socially,
scientifically, or religiously ‘objective’, or irrational because of its social, scientific, or
religious unacceptability. Objective expressions are assumed to be rational in that they
are supported by certain contextual factors. Without context, the distinction between
indexical and objective expressions will not emerge. Hence the term ‘rational’ perhaps
represents only partiality, an ongoing accomplishment of those indexical-based objective
expressions. They are treated by members as rules, possessing powers of ‘agreement’.
In the course of action they turn into background knowledge that (re)frames accounting
practices. The answers to ‘how these agreements work to form background knowledge’
and ‘how social order is possible’ are offered by Garfinkel through extending Schutz’s
assumption of ego–alter relations. Contrary to ethnomethodology, systems theory
borrows from second-order cybernetics and radical yet operational constructivism to
provide another solution.
Consequently, systems are applying respectively expectant assumptions to each
other, and background expectancy is simply one form of these assumptions. In the
analysis of the forming of law, for example, Luhmann (1985 : 30) argues that for
all interactions ‘certainty in the expectation of expectations constitutes the essential
basis’. However, Garfinkel attributes the possibility of ‘rules’ or ‘agreement’ to some
prevalentuncertainties. When he (Garfinkel, 1992 : 72) discusses several ways
of portraying a person as a judgemental ‘dope’ he mentions that those uncertainties in
fact constitute the settings and will emerge when deliberately intervening or inserting
an experiment into it. These uncertainties also suggest that this portrayal is founded on
the assumption that an actor’s agency does exist (Garfinkel, 2006: 107). One thing that
should be emphasized is ‘its reliance upon and its reference to common sense knowledge
of social structures’: when any incongruity occurs, this judgmental work ‘will
force itself upon our attention’ (Garfinkel, 1992 : 71).12 In other words, the emergence
of the actor’s agency can be regarded as the re-entry ofthe distinction, or as the
distinction repeating itself within a system. This dynamic rather than static view of
reflexivity can be found in both ethnomethodology and the theory of social systems.
In this sense, communication starts with common assumptions. Most of the time,
these assumptions are vague and not necessarily recognizable by communicating counterparts.
What they offer is the possibility to select, to draw distinctions, or to reduce
complexities. The extent of it depends on the level of differentiation through which a
system responds to its environment: segmentary, hierarchical, or functional. So far as
the expectable assumptions are realized within communicating systems, two kinds of
expectations are suggested: cognitive and normative. In systems theory they do not
originate in any kind of objectivity. In other words, expectations can always be expected,
and also communications are communicated: ‘Members’ accounts are reflexively and
essentially tied for their rational features to the socially organized occasions of their
use, for they are features of the socially organized occasions of their use’ (Garfinkel,
1992 : 4). This describes the relationship between the settings and members’
accounts of them, indicating not only that the settings will sanction members’ accounts
but also members’ accounts will inversely constitute the settings that they are accounting
for. This is comprehensible when taking into consideration the role played by indexical
expressions. Indexicality confers expectable features on the settings and accounts of
them. The termexpectable means that not only the settings are context dependent, but
also members’ accounts produce their own expectations of expectations. As Luhmann
(2002b: 137) indicates, ‘the foundation for the reality of the system is the simultaneity
of its operation with the conditions of reality that sustain it’. Hence a system’s operation
constitutes simultaneously the reality of the system, and the operation as such is also
conditioned by the reality that supports it. Again, we can recognize here that ethnomethodology
implies a practical logic of systems theory, stating that a member’s accounting
practices can be seen as a system’s operation, and the settings are of a temporal reality
constructed by a system’s operation. Here we come to the formulation of reflexivity
stated by Garfinkel (1992 : 8): members’ accounts are constituent features of the
settings they make observable.
The concept of reflexivity brings together ethnomethodology and the theory of social
systems. They both indicate that, to some extent, purely and clearly causal operations
and agents do not necessarily exist. In the former, members report the settings that have
already been coloured with their using of accounts, which means their accounts apply
themselves to this course of reporting, and hence in turn explain those settings. In the
latter, a system’s operations render the settings accountable or communicable through
the reduction of complexityfrom within. Since the reduction of complexity is characteristic
of the operation of a system, it therefore indicates that this reduction is actually
anchored in the unspecifiable and presuppositionless, i.e. those seen but unnoticed rules.
The seen but unnoticed rules also have their formation in ethnomethodological
enquiries. No matter whether members are situated in the position of first-order or
590Current Sociology 60(5)
second-order observation, considering the experiments of how Garfinkel’s students
tried to explain what happened in a series of conversations, their accounts may always
depend on those seen but unnoticed rules to have expectations carried out in a certain
way. In daily conversations concealed rules have reflexively been used by members to
expect expectations from others, and the expectation of expectations otherwise makes
the rules more concrete than before. When we expect another’s expectations of our
own expectations, we are experiencing a course through which we reach anatural
congruence of mechanisms of generalization(Luhmann, 1985 : 77). The difference
between ethnomethodology and the theory of social systems is that the content of
reflexivity emphasized by the former is characterized by indexicality, whereas in the
latter it refers to a further achievement in that a system constitutes this indexicality
through an asymmetrical operation ofthe distinction between system and environment.
13It also indicates not only asymmetry but also simultaneity in time when
attempting to identify a system through differentiating it from what does not belong to
a system, or through ‘control’ (Luhmann, 2002b: 137; White et al., 2007). Rules in this
way, whether noticed or not, are appropriated reflexively by members simply due to
their necessary indexical characteristics. They are emergingfrom within and can be
The objective of ethnomethodology is not to prove that action is caused by rules, or
to explain how these rules work to make the accounts accountable. Rather, it is to study
how members use these rules to form their mutual ‘understanding’ and ‘reactions’ in the
course of accounting. The concept of social order for ethnomethodologists refers to ‘the
factual character of social reality as a social product of members’ interpretive procedures’
(Leiter, 1980: 191). Theobjective reality occurs only in these procedures, and
hence stands against what some realists describe (for example, Bunge, 2006; Elder-
Vass, 2005; Sawyer, 2005).14 Members’ accounting practices create, complement, or
modify theobjective social orders. Meanwhile, they are produced not only to be valid
for those memberswithin, but also in certain situations to be convincing for those members
without. Members are required to use what they continuously make reportable and
accountable, and social orders emerge exactly from these recursive processes.15 In the
next part the comparison between them will be clearly demonstrated with experiments
conducted by Garfinkel. We come to our conclusion in the last section.
Reviewing experiments through systems theory
In this section we propose to reformulate one of Garfinkel’s experiments through the
systems theory perspective. In order to discover how members use what kind of expectancies
as seen but unnoticed backgrounds of common understandings to recognize what
each other talked about, Garfinkel required his students to treat themselves as boarders
in their ordinary and familiar interactions with their families or friends, and to record the
activities that occurred in these situations for at least 15 minutes. Students were obliged
to assume their positions as strangers and describe their conversations and interactions
with their counterparts. Garfinkel (1992 : 45) also required these students to do
their best to put aside all kinds of social relevance with regard to those people involved.
In the end the students reported the situations they had been through and their thoughts
about this task.
Most students reported difficulties in trying to be a stranger. They were easily
influenced by those originally familiar attributions such as relationships, locations,
motives and so forth. However, while they tried to ‘bracket’ these expectancies, some
unexpected results appeared in their reports. What they viewed was very different from
what they used to be, and what they used to take for granted represented another picture
to them to some extent. This new strangeness in ordinary scenes, according to Garfinkel,
is the function of the seen but unnoticed rules as background expectancies that prevail on
our daily conversations and interactions.
The emergence of strangeness in ordinary settings points out a ‘blind spot’ and
replaces it with another. A distinction has to be used when one is observing. This distinction
constitutes a blind spot for the observer who is using it: ‘an operation that uses
distinctions in order to designate something we will callobservation’, whereby the
observer observes the observed, and the observer is able to recognize what distinction
has been used by the observed (Luhmann, 2002b: 134). In our case, the distinction used
by members performing observations or being observed should be both routine and not
routine. Whether the distinction is used consciously or not, members can develop their
own identity or their accounting of the settings and establish and modify the boundary
with the settings. The distinction appropriated by members will be able to produce and
also to selectfrom within the possibilities generated in order to resonate with the settings
they will account for. Hence the distinction used by members will be fundamental
in forming their expectations and their expectation of expectations.16
The distinction demonstrated above is the source of strangeness. It can be seen but
unnoticed only if it is used by members who are observing. Therefore it is not necessary
that all distinctions have to be the background of our common understanding. Instead,
there are only some, usually most of them, that will form its basis. Thus not only a system’s
operations and its expectations, but also its expectation of expectations constitute
and make ordinary and common talk understandable and reportable. Garfinkel (1992
: 68) notices this when concentrating on observing how members make the
settings accountable and how they are influenced by the unnoticed rules. Ignoring the
property of common understandings portrays a person as a ‘cultural or psychological
dope’: ‘The common feature . . . is the fact that courses of common sense rationalities
of judgment which involve the person’s use of common sense knowledge of social
structures over the temporal “succession” of here and now situations are treated as epiphenomenal.’
The termepiphenomenal refers theoretically to the preferences possessed
by members, lay and professional. It results in the possibility of breaching and intervening
in the procedure, and in the possibility of disclosing ‘the nature of rule governed
actions’. His experiments suggest (Garfinkel, 1992 : 70–71) that the avoided
tests and challenges to the background expectancies constitute the basis of portraying a
person as a judgemental ‘dope’. More deeply,pace Garfinkel, the problem of how
social order is possible starts with the discussion of the property of agreement and
how it is formed and used. Diverging from it, thisepiphenomenon emerges as a parasite
in a system’s constituting and communicatingabout its own environment.
The property of agreement contains not simply objective expression, but indexicalitybased
objectiveexpression since each agreement implies a distinction in order to observe
or make a selection. In the aforementioned experiments students reported that the
592Current Sociology 60(5)
phenomenon they observed was not areal one of their familiar environment, which
means that what they observed in a nearly pure context-free way is contradictory with
what they had already expected, whether consciously or not, of anormal situation. In
fact, the actions or observations of students only present how the background expectancy
affected them rather than pointing out the means used to understand the observed.
Besides, expectancy, overtly or covertly, indicates a selection of distinction. The selection
and application of a distinction also constitutes the foundation of expectations. We
either choose or adapt to what we have already chosen, otherwise we can see nothing in
the daily conversations with others. At the time that the observing system observes, it
makes a distinction that operates as its background expectancy. Even if the observing
system attempts to ‘escape from’, ‘ignore’, ‘disregard, or ‘bracket’ those seen but unnoticed
rules that are used to make sense of the observed, it will ultimately fail due to
‘circularity of self-reference’, which means the observing system uses those rules in the
same way as the observed does. This situation is what the term ‘two black boxes’
describes, two systems confronting each other. As the experiment shows (Garfinkel,
1992 : 46–47), students report that as they process their ‘jobs’ they can hardly
avoid these rules, and their actions are regarded by their counterparts as ‘weird’.
The systems theoretical viewpoint suggests an alternative to observing how a distinction
is used by a system and how it is repeated in the system observing the observed. In
the case of Garfinkel’s experiment, the feedback of the observed, such as asking ‘What’s
wrong with you?’ or ‘Are you out of your mind?’, includes the ways they are using distinctions,
i.e. the extension of previous expectations to the subsequent situations. Hence
each communication is an ongoing accomplishment, and only communication can reproduce
communication, only communication communicates. Each temporary accomplishment
depends on the fulfilment of the expectation of expectations. We communicate
until we get what we expect or not, rather than the factual situation that has been achieved.
In other words, communication is accomplished when the expectation of expectations is
satisfied to some extent or not.
Considering the experiment, each observation, each conversation, each communication
that is held, achieved and worked out by students has its own internal efficacy. One
researcher comments (Fuchs, 1997: 188) that ‘conversations develop their own internal
history, regardless of whatever plans any actors might have’. This ‘internal history’ is
to some extent irrelevant to the actor’s intentions, whether the actor attempts to orient
those conversations towards a certain direction or not. However, it enters into the actor’s
accounting practices of these flowing conversations, or is later exemplified bypassing,
as ‘a texture of relevances manipulated to coming to terms with practical circumstances’
(Garfinkel, 1992 : 137–185). As an observer observes, the observer cannot
understand what has been communicated by the observed. The observer has only partial
information about the observed, and since it is only partial the observed cannot be seen
through. Other references are needed to let the observer decipher, constitute and then
understandwhat the observed expresses. In other words, there are plenty of possibilities
to select; but these possibilities have already been conditioned to a certain range by
situations observed. Hence this internal history indicates that in a somewhat loose way,
or in Garfinkel’s term, there exist ‘seen but unnoticed’ rules that govern and condition
what is and how it will be developed between conversations.
While ethnomethodology and the theory of social systems differ in their disciplinary
orientations, research questions and levels of analysis, they share a common problematic
in the constitution of realities. The term ‘constitution’ refers to the operation of reflexivity
on one side presented by members’ accounting practices of the settings they are
making accountable and reportable, and on the other side operated by the distinction
repeating itself within a system in order to maintain a boundary with its environment,
both of whose redundancy confirms the reality of the identity thus produced. What is
ultimately produced is not the truth of the universe, but the known structures and processes
on which members and social systems premise and account for their settings and
communications. When dealing with knowledge, whether in sociological enquiries, the
professionals, or everyday life, ethnomethodology seeks to show how such knowledge is
possible. It strives to retrace the steps through which descriptions, definitions, concepts,
or formal methods are constituted. The theory of social systems explains constitutional
problems epistemologically, whereas ethnomethodology tackles them practically through
the ethnographic investigation of members’ reporting and accounting practices.
This article compares ethnomethodology with the theory of social systems and concludes
with three implications for theoretical advancement. The first is that the two are
complementary to each other since they share a common ground when dealing with the
mechanism ofhow problems. Second, with the fundamental distinction between indexical
and objective expressions, ethnomethodology can then be observed with a systems
theory perspective. Accordingly, it is practical and indexical because of the distinction
that repeats itself within this same distinction and designating the side of indexical
expression. This can be seen only if,pace ethnomethodology, members appropriate
other distinctions to breach the ones they are using. To this extent, the concept of
‘indexicality’ has its source in the operations of and also between systems communicating
aboutothers. This is an unachievable task due to their non-transparency to each
other. When the extent of differentiation increases, indexicality becomes more important
and more unavailable to each system; effective communications among them will
thus be gradually improbable. In systems terms, in a simpler society the interaction
system is closer to the society, whereas in a more complex one the distance between
them increases and results in the separation of social structure and semantics, thus
requiring other mechanisms to reckon with this situation, such as role systems or more
specializations. Third, due to a higher level of abstraction of systems theory, this article
argues that ethnomethodology may be formulated as a practical logic of systems theory.
While Luhmann’s systems theory has been treated unfairly because of its abstractness
and the difficulties of putting it into practice, it can be complemented and reformulated
with the ethnomethodologicalmethod to display and apply its practical logic to sociological
enquiries, and also to escalate the analytical level when addressing the social
This research received no specific grant from any funding agency in the public, commercial, or
594Current Sociology 60(5)
1. This article was originally presented at the 104th ASA Annual Meeting, 8–11 August 2009,
San Francisco, California.
2. Structural couplings take place between a system and its environment. The most important
aspect of the concept is that ‘it does not indicate a causal relation and certainly not an
instrumental relation, but one of simultaneity’ (Luhmann, 1993: 98). Since a system and its
environment can only exist at the same time, their couplings are difficult to be conceived as
causal or instrumental.
3. Autopoiesis means ‘self-production’, extending from the concept of self-organization. The
system that is autopoietic indicates that ‘everything which is used as a unit by the system is
produced as a unit by the system itself’ (Luhmann, 1986: 174).
4. Paradoxes preserve the form of opposition that always contains two elements that are different
from each other, such as true/false, legal/illegal. Since this form consists of at least two
opposite elements, they are asymmetrical in that only one side will be marked at the same
time. The paradox lies in the fact that the distinction comprised of two sides has to be not only
asymmetrical but also simultaneous (Luhmann, 2005 : 91).
5. Despite the fact that he draws from functionalism and phenomenology, Luhmann was considered
to be not a structural functionalist, or a neofunctionalist, or a postmodernist (Fuchs,
6. Kenneth Leiter (1980) offers an excellent explanation of the relationship between phenomenology
and ethnomethodology. Subtle analyses of this are also to be found in Garfinkel’s
own work (1992 ).
7. It is derived from Husserl’s distinction between open and problematic possibilities. See also
Schutz’s discussion of these ideas (Schutz and Natanson, 1990 : 79–82).
8. Although Garfinkel (2006: 117) confirms this by stating that ‘we decided to reject any notions
regarding how the world “really” is’, hence abandoning any possibility of ontological viewing
of society and members’ actions and then adopts phenomenological concepts from Husserl
and Schutz, he simultaneously leaves aside the former’s concept of ‘transcendental subject’.
9. In the remainder of this article we adopt a loose application of the distinction between act and
action and regard it as a unity of difference.
10. Luhmann (2002a) makes an analogy of time perspectives with Husserl.
11. Schutz (1990 : 48) elaborated this term, transforming it from a human’s mental operations
to social embodied ones. Its phenomenological origin indicates a second-order construct
of indigenous constructs. Garfinkel refers to it to describe members’ accounting and using of
the constructs of the settings as part of their accounts. Tsekeris and Katrivesis (2008: 3) also
clearly point out this inheritance and adoption by ethnomethodology.
12. Bourdieu has already suggested that the term ‘habitus’ means an unlimited source of production
of thoughts, perceptions, expressions and actions (Bourdieu, 1977).
13. Researchers (Tsekeris and Katrivesis, 2008) also claim that this reflexivity with indexicality
will be able to respond and adjust to the environment and also to social phenomena.
14. Similarly, Luhmann describes reality as being purely constructed and worked out within a system.
The reality comes out of a process in which a system not only constitutes its environment
but also thematizes its relation to this imagined environment onlyinside itself.
15. While Heidegger contrasts the concept of ‘ready-to-hand’ with that of ‘presence-at-hand’ and
infers that there are distinctions between modern science and modern technology, he applies
the same strategy later extended by ethnomethodology to describe the relationship between
a human and the world. This reflexivity, for instance, has been demonstrated more clearly
by Ihde (1979: 110–111) as follows: ‘Modern science is embodied technologically.’ He also
states: ‘Technology is the source of science, technology as enframing is the origin of the
scientific view of the world as standing-reserve.’ Drawing on the legacy of phenomenology,
Luhmann offers a similar explanation.
16. It argues (Luhmann, 2002b: 134–135) that the reality of a world is simply imagined by its
members or by their observation of other observing systems, and hence is completely internal
to the system and only emerges in the operation of a system.
Bauman Z (1994 ) Hermeneutics and modern social theory. In: Held D and Thompson J
(eds)Social Theory of Modern Societies: Anthony Giddens and his Critics. Cambridge:
Cambridge University Press, pp. 34–55.
Bednarz J (1984) Complexivity and intersubjectivity: Toward the theory of Niklas Luhmann.
Human Studies7(1): 55–69.
Bourdieu P (1977)Outline of a Theory of Action. Cambridge: Cambridge University Press.
Bunge M (2006)Chasing Reality: Strife for Realism. Toronto, Buffalo and London: University of
Callinicos A (1985) Anthony Giddens: A contemporary critique.Theory and Society 14(2):
Callinicos A (2007 )Making History: Agency, Structure, and Change in Social Theory.
Leiden: Koninklijke Brill NV.
Coleman J (1990)Foundations of Social Theory. Cambridge, MA and London: The Belknap Press
of Harvard University Press.
Cuff EC, Sharrock WW and Francis DW (2006 )Perspectives in Sociology, 5th edn. London
and New York: Routledge.
Czerwick E (2008)Systemtheorie der Demokratie: Begriffe und Strukturen im Werk Luhmanns.
Wiesbaden: VS Verlag für Sozialwissenschaften.
Elder-Vass D (2005) Emergence and the realist account of cause.Journal of Critical Realism 4(2):
Elder-Vass D (2007) Luhmann and emergentism: Competing paradigms for social systems theory?
Philosophy of the Social Sciences37(4): 408–432.
Fuchs S (1997) Review work(s): Social systems.Contemporary Sociology 26(1): 117–118.
Garfinkel H (1992 )Studies in Ethnomethodology. Cambridge: Polity.
Garfinkel H (2002)Ethnomethodology’s Program: Working out Durkheim’s Aphorism. New York
and Oxford: Rowman and Littlefield.
Garfinkel H (2006)Seeing Sociologically: The Routine Grounds of Social Action. Boulder, CO
and London: Paradigm Publishers.
Garfinkel H (2007) Four reasons between literatures of the social scientific movements and their
specific ethnomethodological alternatives. In: Hester S and Francis D (eds)Orders of Ordinary
Action: Respecifying Sociological Knowledge. Farnham: Ashgate, pp. 13–29.
Giddens A (1984)The Constitution of Society: Outline of the Theory of Structuration. Cambridge:
Green G (1982) The sociology of dogmatics: Niklas Luhmann’s challenge to theology.Journal of
the American Academy of Religion50(1): 19–34.
Gregory D (1994 ) Presences and absences: Time–space relations and structuration theory.
In: Held D and Thompson J (eds)Social Theory of Modern Societies: Anthony Giddens and his
Critics. Cambridge: Cambridge University Press, pp. 185–214.
Holub R (1994) Luhmann’s progeny: Systems theory and literary studies in the post-Wall era.New
German Critique61: 143–159.
Ihde D (1979)Technics and Praxis. Dordrecht and Boston, MA: D Reidel.
596Current Sociology 60(5)
Johnson DP (2008)Contemporary Sociological Theory: An Integrated Multi-Level Approach.
New York: Springer Science+Business Media, LLC.
King M (2001) The construction and demolition of the Luhmann heresy.Law and Critique 12(1):
King M and Schutz A (1994) The ambitious modesty of Niklas Luhmann.Journal of Law and
King M and Thornhill C (eds) (2006)Luhmann on Law and Politics. Oxford: Hart Publishing.
Kwang-Ki K (2003)Order and Agency in Modernity: Talcott Parsons, Erving Goffman, and
Harold Garfinkel. Albany, NY: State University of New York Press.
Leiter K (1980)A Primer on Ethnomethodology. New York and Oxford: Oxford University Press.
Luhmann N (1983) Insistence on social systems: Perspectives from Germany – An essay.Social
Luhmann N (1985 )A Sociological Theory of Law, trans. King E and Albrow M. London,
Boston, MA, Melbourne and Henley: Routledge and Kegan Paul.
Luhmann N (1986) The autopoiesis of social systems. In: Geyer F and Van der Zouwen J (eds)
Sociocybernetic Paradoxes. London: Sage, pp. 172–192.
Luhmann N (1993)Risk: A Sociological Theory, trans. Barrett R. Berlin and New York: Walter
Luhmann N (1995 )Social Systems, trans. Bednarz J and Baecker D. Stanford, CA: Stanford
Luhmann N (2002a) The cognitive program of constructivism and the reality that remains
unknown.Theories of Distinction: Redescribing the Descriptions of Modernity, ed.
Rasch W, trans. O’Neil J, Schreiber E, Behnke K et al. Stanford, CA: Stanford University
Press, pp. 128–152.
Luhmann N (2002b)Theories of Distinction: Redescribing the Descriptions of Modernity, ed.
Rasch W, trans. O’Neil J, Schreiber E, Behnke K et al. Stanford, CA: Stanford University
Luhmann N (2005 ) The paradox of decision making. In: Seidl D and Becker KH (eds)
Niklas Luhmann and Organization Studies. Herndon, VA: Liber and Copenhagen Business
School Press, pp. 85–106.
Maturana HR and Varela FJ (1980 )Autopoiesis and Cognition: The Realization of Living,
Vol. 42. Dordrecht: D Reidel.
Mehan H and Wood H (1975)The Reality of Ethnomethodology. New York and London:
Parsons T (1937)The Structure of Social Action. New York: McGraw-Hill.
Rasch W (2000)Niklas Luhmann’s Modernity. Stanford, CA: Stanford University Press.
Sawyer RK (2005)Social Emergence: Societies as Complex Systems. Cambridge and New York:
Cambridge University Press.
Schutz A and Natanson M (eds) (1990 )Collected Papers: The Problem of Social Reality,
Vol. I. Dordrecht, Boston, MA and London: Kluwer Academic.
Seidl D and Becker KH (eds) (2005)Niklas Luhmann and Organization Studies. Herndon, VA:
Liber and Copenhagen Business School Press.
Simmel G and Wolff K (eds) (1950)The Sociology of Georg Simmel. Glencoe, IL: The Free Press.
Spencer-Brown G (1969)Laws of Form. London: George Allen and Unwin.
Thompson J (1994 ) The theory of structuration. In: Held D and Thompson J (eds)Social
Theory of Modern Societies: Anthony Giddens and his Critics. Cambridge: Cambridge
University Press, pp. 56–76.
Tsekeris C and Katrivesis N (2008) Reflexivity in sociological theory and social action.Facta
Urry J (1982) Duality of structure: Some critical issues.Theory, Culture and Society 1(2):
Von Foerster H (1979) Cybernetics of cybernetics. In: Krippendorff K (ed.)Communication and
Control in Society. New York, London and Paris: Gordon and Breach Science, pp. 5–8.
Wagner G (1997) The end of Luhmann’s social systems theory.Philosophy of Social Sciences
Weber M (2001 )The Protestant Ethic and the Spirit of Capitalism. London and New York:
White H, Fuhse J, Thiemann M et al. (2007) Networks and meaning: Styles and switchings.Soziale
Zolo D (1992)Democracy and Complexity: A Realist Approach, trans. McKie D. Cambridge:
Liu Yu Chengis a postdoctoral researcher and college master at National Chengchi University,
Taiwan, teaching sociology of speed, sociological theory and globalization. Current research interests
include systems theory, sociological theory, ethnomethodology, religion and modernity. He
has published on social systems theory and religious studies. Recently he has been working on
systems theory and media studies.
Cet article compare l’ethnométhodologie à la théorie des systèmes sociaux, et envisage
d’éclaircir leurs intérêts théoriques, différences et ressemblances, lorsque l’on considère
de manière synthétique la première comme une logique pratique de la seconde. En tant
que deux pôles dans un continuum de méthodes de recherche, l’ethnométhodologie
est une micro-analyse de l’ordre social et de l’action humaine, tandis que la théorie des
systèmes assume une abstraction à la fois macro et théorique de la formation du système
et de l’évolution sociale. Lorsqu’il est tenté de résoudre les problèmes du micro/macro,
de la part et du tout, ou de la relation individu-société, cette comparaison conduit à
expliquer comment la logique pratique opère aussi bien au niveau macro qu’au niveau
micro, et à renforcer notre compréhension du monde social humain. Trois points sont
ici suggérés. Premièrement, la théorie des systèmes et l’ethnométhodologie n’entrent
pas en contradiction ; elles sont complémentaires dans un processus intrinsèquement
dialectique. L’ethnométhodologie peut, deuxièmement, être appréhendée en
distinguant les expressions indexicales et objectives et suivant ces dernières plutôt que
les premières. Troisièmement, en raison du plus haut niveau d’abstraction de la théorie
des systèmes, ce papier défend l’idée que l’ethnométhodologie peut être utilisée pour
expliquer la théorie des systèmes, tout en pouvant être considérée comme sa logique
pratique. Cette comparaison peut contribuer à une reformulation de la théorie des
systèmes, et aussi de quelques puzzles philosophiques.
ethnométhodologie, théorie des systèmes, indexicalité, réflexivité, auto-poiesis
598Current Sociology 60(5)
Este artículo compara etnometodología y teoría de los sistemas sociales, tratando
de iluminar sus preocupaciones teóricas, diferencias y similitudes; mientras que
sintéticamente considera al primero como una lógica práctica del segundo. Como
dos polos de un continuum de métodos de investigación, la etnometodología es un
microanálisis del orden social y de la agencia humana, mientras que la teoría de los
sistemas presupone una abstracción macroteórica de la formación del sistema y de
la evolución social. Cuando se realizan intentos de resolver los problemas de las
relaciones micro/macro, parte/ todo o individuo/sociedad, esta comparación colabora
para dilucidar cómo la lógica práctica está implicada en ambos lados, micro y macro,
y para aumentar nuestro entendimiento del mundo social humano. Se sugieren tres
puntos. Primero, la teoría de los sistemas y la etnometodología no se contradicen entre
ellas, sino que son complementarias en un intricado proceso dialéctico. Segundo, la
etnometodología puede ser observada con la distinción de expresiones indexicales y
objetivas, entrando en la primera antes que en la segunda. Tercero, debido al alto nivel de
abstracción de la teoría de los sistemas, este trabajo argumenta que la etnometodología
puede posiblemente ser utilizada para explicar y también puede ser considerada como
la lógica práctica de la teoría de los sistemas. Esta comparación puede contribuir para
reformular la teoría de los sistemas, así como para algunos dilemas filosóficos.
Palabras claveetnometodología, teoría de los sistemas, indexicalidad, reflexividad, autopoiesis