See discussions, stats, and author profiles for this publication at: [Link]

net/publication/227918111

Three Problems of Intersubjectivity—And One Solution*

Article in Sociological Theory · March 2010

DOI: 10.1111/j.1467-9558.2009.01364.x

CITATIONS READS

38 2,401

1 author:

Wendelin Reich
Uppsala University
17 PUBLICATIONS 288 CITATIONS

SEE PROFILE

All content following this page was uploaded by Wendelin Reich on 13 April 2018.

The user has requested enhancement of the downloaded file.

 Three Problems of Intersubjectivity—And One Solution∗

WENDELIN REICH
Uppsala University

Social thinkers often use the concept of intersubjectivity to mark out a problem
of theoretical sociology: If people are unable to look into each others’ minds, why
do they often understand each other nonetheless? This issue has been debated ex-
tensively by philosophers and sociologists in three largely disconnected discourses.
The article investigates the three discourses for isolable ideas that can be fitted
into a sociological answer to the problem of intersubjectivity. An interactional so-
lution, fully coherent with key insights from the discourses, is offered at the end of
the article. Its main point is to identify coordinative interactional mechanisms that
compel participants to “make themselves understandable” vis-à-vis their interaction
partners.

The concept of intersubjectivity has fascinated sociologists for more than half a
century. With its implied promise of a connection (“inter”) between seemingly au-
tonomous selves (“subjects”), it appears to express a core aspect of sociality. At
the same time, the mere availability of the concept marks the existence of such a
connection as contingent, as something that cannot always be taken for granted
and that, therefore, stands in need of sociological explanation. Schütz, who took the
concept from Husserl and recast it as a foundational problem of social science, spent
the better part of his professional life trying to solve the “enigma of how man can
understand his fellow man” (Schütz 1962d:179). When Berger and Luckmann intro-
duced both Schütz and his adapted notion of intersubjectivity to the sociological
mainstream, they made clear that this enigma is normally not felt by members of
society, who take the intelligibility of their interaction partners for granted (Berger
and Luckmann 1966:23). Contemporary research in social psychology confirms
this phenomenological intuition, showing that human beings are naturally condi-
tioned to attempt to mind-read others in social situations (Gergely and Csibra 2003;
Papafragou 2002) and to assume that they are continuously being mind-read by
co-present others (Gilovich et al. 1998; Tomasello et al. 2005).
The establishment of a situation where “man can understand his fellow-man” de-
pends normally not only on the skills of the understander (or mind reader), but
also on the cooperation of the person who is to be understood. Far from render-
ing the problem of intersubjectivity obsolete, social-psychological research thus casts
Schütz’s “enigma” as even more mysterious because it fails to answer the question of

∗ Address correspondence to: Wendelin Reich, Department of Sociology, Uppsala University, Box 624,
75126 Uppsala, Sweden (SE). Tel.: +46-73-6482401; E-mail: [Link]@[Link]. I wish to express
my gratitude to Richard Swedberg and Delia Baldassarri, without whom this article would never have
been published. It is based on Chapter 1 of the author’s unpublished Ph.D. thesis in sociology (Uppsala
University, 2003). On a personal note, I would like to add that it is with mixed feelings that I see this
article get to print. “Intersubjectivity” or the understanding of other individuals’ thoughts and feelings
has become an important line of research in recent neuropsychology, developmental cognitive science,
and comparative (primate) psychology. The sociological way of approaching the subject now strikes me
as rather rationalistic, and I can only hope that the article may, nonetheless, have some merit as it stands.

Sociological Theory 28:1 March 2010


C 2010 American Sociological Association. 1430 K Street NW, Washington, DC 20005
INTERSUBJECTIVITY 41

why people’s expectation to understand and to be understood by others is a socially

viable strategy. Having learned from the flaws of Berger and Luckmann’s static,
consensus-oriented conception of intersubjectivity, social thinkers like Habermas
(1988:180), Schegloff (1992b:1298), and Knorr-Cetina (see especially Knorr-Cetina
and Bruegger 2002:922) hold that intersubjectivity must be viewed as an accom-
plishment that is always fragmentary and fallible. However, a point of view is not
a theory, and it is the very lack of a full-fledged theory of intersubjectivity that
leaves current sociology ill equipped to answer questions such as: Under what so-
cial conditions is intersubjectivity likely to be accomplished? How much—or how
little—intersubjectivity can be achieved under optimal conditions? Are agents only
concerned with “defending” intersubjectivity (Schegloff 1992b) or just as much with
simulating or exploiting it for their own purposes? Are there culturally specific insti-
tutions or mechanisms that support the establishment of intersubjectivity? To what
extent do societies—or different fields within them—depend on the continual achieve-
ment of intersubjectivity? When do structural deficiencies in social understanding
threaten the orderliness and projectability of social conduct—and when do they not?
What roles do generalized understanding and misunderstanding play in cultural re-
production, in the legitimation of social inequalities, or in the initiation of social
change?
The aim of this article is to develop a solution to the problem of intersubjectivity—
how members of society come to “understand” each other—which may help theo-
rists and empirical researchers to tackle questions such as these. The path taken
here to develop the solution is to engage the reader in a focused review of exist-
ing attempts to solve—or dispose of—the problem. Several of these attempts are
as original as they are clever, making it necessary to discuss them in a manner
that is more selective than they deserve and to approach them in some cases with
modernized terminology. They can be organized neatly into three philosophical and
sociological discourses—on other minds, intersubjectivity, and double contingency—
and will be discussed in the first three sections of this article. I have chosen to
include the two former, philosophical discourses in this sociological discussion be-
cause both of them have been highly influential in sociology (due to names like
Wittgenstein, Husserl, and Schütz). The final section presents a solution that draws
on an idea that Harvey Sacks, the founder of Conversation Analysis, offered in
two minor, largely disregarded works. As we shall see, Sacks’s idea turns the prob-
lem of intersubjectivity upside down, replacing the question of how person A can
understand B with the question of how A can compel B to make him-/herself
intelligible.
For the sake of generality and comparability, this article will use a simple work-
ing definition of “intersubjectivity” as a situation in which two or more persons
share knowledge reflexively, that is, all know X and know that all others know
this, too. The main objective of this definition is to exclude any normative-emphatic
conceptualization of intersubjectivity as consensus, cultural cohesion, “deep” under-
standing, or soul watching. It also excludes the confused notion of intersubjectivity-
as-process—which appears to be popular among some ethnomethodologists—insofar
as it distinguishes processes of knowledge-sharing from their empirical result (which
may range from full intersubjectivity to utter misunderstanding). An arbitrary—but
useful—convention I shall apply throughout is to use the pronouns “she” and “he”
to distinguish between the formal perspectives of the understanding person (ob-
server, listener) and the person to be understood (observee, speaker), respectively.
Tacit translations from German and French sources are my own.
42 SOCIOLOGICAL THEORY

THE (PHILOSOPHICAL) PROBLEM OF OTHER MINDS

Understanding and misunderstanding are such common and fundamental aspects
of social life that it seems safe to surmise that most societies have developed folk
theories for reflecting on their possibilities and limitations. In view of the fact that
many problems contemporary philosophers deal with can be traced back to medieval
or even antique authors, it is therefore interesting to note that issues of social un-
derstanding and understandability entered the philosophical stage at a comparatively
late moment. Before Descartes’ radicalized subjectivism, it rarely ever seemed rele-
vant or even sensible to assume that access to other people’s thoughts or emotions
could be barred in an absolute manner (Avramides 2001). To be sure, anthropo-
logical conceptions portraying man as witty, strategic, and consequently inscrutable
are much older. With antique predecessors, they reemerge toward the dawn of the
Renaissance in the stories of Boccaccio (1982), and they enter the early European
literature of political consultation in the writings of Guicciardini (1994), Machiavelli
(1988), and Botero (1997). However, all these theories are from the outset supple-
mented by hierarchies of impulses and needs that serve to make man reinterpretable.
In Boccaccio, hypotheses of sexual desire, greed, or overt malevolence allow the ob-
server to bridge the gap between false appearance and true intentions. The political
advisors, on the other hand, rely on the handy distinction between passions and
interests and refer to the latter—especially financial or territorial interests—when
asking what forces underlie the seemingly erratic behavior of princes (Hirschman
1977; Luhmann 1989). Thus, the indeterminacy introduced by an interpretation of
behavior as strategically motivated is domesticated from the outset. Observing what
the other is up to is not yet registered as a fundamental problem.
The situation changes when Descartes’ Discourse on the Method drastically bans
all assumptions about the outer world, including assumptions about other people’s
minds, to the realm of doubt (Descartes 1996). The ordinary idea that people behave
according to a universal, hence intelligible common sense, is unmasked as an inven-
tion of this very common sense. 1 Although Descartes states his axiom of radical
uncertainty in the middle of a treatise on rules for “Rightly Conducting the Reason,
and Seeking Truth in the Sciences” despite this uncertainty, he becomes the reference
point of a long series of philosophical refutations of socioepistemological skepticism.
In the systems of Kant as well as the German idealists, this skepticism is still easily
brushed off by means of a transcendental conception of “reason” or a metaphysical
notion of “spirit” (Hegel 1970; Kant 1974). 2
However, in the 19th century, the growing uneasiness of Anglo-Saxon philosophers
with transcendentalism and metaphysics leads to a more detailed inspection of the
“other-minds problem,” as it comes to be called. John Stuart Mill seems to be one
of the first to put this uneasiness into words when he asks: “By what evidence do I
know, or by what considerations am I led to believe, that there exist other sentient
creatures; that the walking and speaking figures which I see and hear, have sensations
and thoughts, or in other words, possess Minds?” (Mill 1872:41). This question is
open to both epistemological (“do I know”) and psychological/sociological (“am I

1 Descartes gives this observation an ironical twist when he writes that: “Good sense is, of all things
among men, the most equally distributed; for every one thinks himself so abundantly provided with it,
that those even who are the most difficult to satisfy in everything else, do not usually desire a larger
measure of this quality than they already possess” (Descartes 1996:1f).
2 Kant certainly did not hold any naı̈ve beliefs about the understandability of other minds, but his
works do not pursue the problem in detail. For two elaborate reconstructions of Kant’s presumable views
on the accessibility of others’ minds, see Kirk (1986) and Simon (2001).
INTERSUBJECTIVITY 43

led to believe”) answers. Yet it seems to have triggered an almost exclusively episte-
mological discourse (for some highlights, see Aune 1961; Dretske 1973; Glendinning
1998; Hyslop 1995; Malcolm 1958; Putnam 1975; Shoemaker 1965; Sober 2000;
Strawson 1959; Ziff 1965). With the important exception of phenomenology (to be
considered in detail in the next section), the problem that seems to haunt a remark-
able number of 20th-century philosophers is how to justify a belief in the existence
and intelligibility of other minds. For the purposes of this article, it is not necessary
to consider their discussions in any detail. Instead, I would like to single out four
contributors to this discourse: Russell, Wittgenstein, Hollis, and Levin.
Bertrand Russell is presumably the first to suggest that people’s belief in the “ex-
istence of other people’s minds” may be justified by means of analogy from self
to others (Russell 1993:101–05). This idea has had an enormous impact on the
way in which 20th-century analytical philosophy approached the other-minds prob-
lem. It also dovetails with some recent work in cognitive science, which hypothe-
sizes that people reconstruct others’ beliefs by way of “simulation” (Gordon 2004).
Nonetheless, Russell admits that analogical justification can never be more than a
philosophical blessing to psychologically necessary convictions. There is simply no
deductively valid inference from other people’s bodies to their minds—although this
inference seems to do well as a “working hypothesis” (Russell 1993:103) in everyday
life. Russell notes that this “somewhat meager conclusion” (Russell 1993:104) can
only be the starting point of a more thorough analysis of the practical role of such
convictions.
Ludwig Wittgenstein’s views on the problem of other minds are instructive even
though they have to be reconstructed from a set of loosely connected notes (pub-
lished posthumously in the Philosophical Investigations). In a sequence of paragraphs,
he asks how we can judge that someone is “reading a book” when we see him hold-
ing it in front of his face and producing sounds that correspond to the words on the
current page (Wittgenstein 1953 §§156–171). How can we be sure that this person
does not just recite a text he learned by heart? In my understanding, Wittgenstein’s
answer is both radical and fruitful. Although reading versus reciting seems to be
a distinction between mental operations, the predicate “X is reading” must refer
to a difference in observable behavior. For instance, if the person in question has
to read a text in an alphabet that we invented especially for this experiment, and
he appears to look up characters in a table before he pronounces them, we would
certainly say that he is reading and not reciting. In general, such external differences
are all we have in order to make sense of his behavior. As Wittgenstein beautifully
writes, trying to observe the person’s inner experiences directly is similar to mistreat-
ing an artichoke: “In order to find the real artichoke, we divested it of its leaves”
(Wittgenstein 1953 §164). 3 When it comes to observing other minds, the “core,” if
there is such a thing, must be found at the behavioral “surface.”
Martin Hollis makes the important suggestion that knowledge of another mind
depends not on the capacities of the observer, but on the presumed internal organi-
zation of the observee (Hollis 1977:149ff). The other can only become transparent
when he keeps his mental operations consistent. He does not have to follow any

3 This interpretation is in accordance with the standard commentary by Savigny (1988:201–04). A

long-winded argument about the possibility of “feeling someone else’s pain” that appears later in the
Philosophical Investigations (cf. 1988:302) has also caused heated philosophical debates. However, Hyslop
nicely demonstrates that it is actually irrelevant to the problem of justifying beliefs about other minds:
not identity of one’s own state and the other’s state, but only perception of the other’s state would need
to be shown to be possible (Hyslop 1995:9).
44 SOCIOLOGICAL THEORY

specific (e.g., Western, scientific, masculine, etc.) “logic,” but he has to follow some
logic and maintain some internal consistency. The only problem with this precondi-
tion, Hollis explains, is that it is itself not observable. If consistency is a precondition
of observation, it cannot be contingent on observation. Hollis draws the conclusion
that we can only observe the other mind as another mind in virtue of a prior at-
tribution of consistency. 4 What the other mind “actually” does, whether he thinks
consistently or erratically, and whether he “really” is a mind—and not a computer
program or a hallucination—must remain completely irrelevant to this imputation.
The upshot of Hollis’s attributive turn is that the other-minds problem is radically
subjectified. Observing the other mind becomes a matter of adjusting one’s own
belief system.
Finally, Michael Levin, drawing on evolutionary theory, carries Russell’s vision of
a more naturalistic conception of the other-minds problem some important steps
further (Levin 1984). Building implicitly on Hollis’s attributive turn, he asks what
evolutionary advantages an observation of others as other minds could entail. The
key benefit, according to Levin’s answer, is that the behavior of some rather er-
ratically moving (and potentially dangerous) objects in an organism’s environment
becomes more interpretable and predictable. More recent studies in cognitive science
(Gergely and Csibra 2003) and evolutionary anthropology (Tomasello 2006) confirm
that humans—and probably other mammalian species—approach their environment
with a “teleological stance.” According to Levin, the ascription of self-reference and
agenthood that underlies this stance becomes “predictively fruitful” (Levin 1984:350)
by enabling organisms to generate further, more specific assumptions about the inner
organization of another agent. For instance, it allows predators to predict the move-
ments of their prey, and humans to interpret certain types of behavior displayed by
other humans as social actions (cf. Dennett 1987, who makes a similar argument).
The contributions of these four authors anticipate certain aspects of the solution
to the problem(s) of intersubjectivity proposed later in this article. They turn ob-
servation and understanding of other minds from a justificational into a practical
problem. The observer’s task is to generate transparency in the face of opacity. At
no point in time is she able to draw on information about the other’s mental state.
Thus, her only chance is to relate behavioral events directly by assuming (perhaps
counterfactually) that the other mind connects them in internally consistent ways.
However, neither of these theories separately, nor all of them puzzled together,
provide an idea of the full picture. How can an ascription of self-reference and con-
sistency make another mind more predictable? If the essence of observation of other
minds is that it works independently of the existence and the internal organization
of other minds, how can it produce information that is not independent of their ex-
istence and their internal organization? In case the observee is an animal, schematic
supplementary hypotheses about the meaning of its gestures, facial expressions, or
body postures may provide sufficiently reliable information about inner states and
expectable conduct. In the case of humans, however, the individual’s ability to dissim-
ulate intentions on the basis of inferences about her alter ego’s beliefs and intentions
can be expected to be matched by her observee’s ability to do just the same (Schelling
1980:60). Theories of social systems, which I will discuss below, have baptized this

4 His emphasis on categorical attributions as a precondition of cognition places Hollis’s statement

firmly within the context of Kantian epistemology. The idea that a “radical translation” between two
unrelated languages requires reconstruction of internal consistencies goes back to Quine (1960:29ff).
Similar conclusions are arrived at by Davidson (1980:237).
INTERSUBJECTIVITY 45

predicament a “situation of double contingency” and have argued forcefully that its
solution requires support by some form of social order. Thus, the subjective turn
made possible by philosophical insights needs to be complemented by findings about
some form of objective social organization that renders self-determined observations
adequate. Further below, a number of such findings will be extracted from some of
the early works of Harvey Sacks.

THE (PHENOMENOLOGICAL) PROBLEM OF INTERSUBJECTIVITY

While the discourse on “other minds” is largely situated within the analytical, Anglo-
Saxon tradition of philosophical theorizing, the notion of intersubjectivity can be
seen as a prototypical brainchild of the Continental philosophy of the subject. Ed-
mund Husserl is the first to make intersubjectivity a core notion of a complex
philosophical edifice—his phenomenology of the “transcendental ego.” It is chiefly
due to his and his students’ contributions that this notion made an extraordinary
philosophical and sociological career. Naturally, it would be neither possible nor
relevant to reconstruct the entire discourse on intersubjectivity and its various con-
ceptions of social understanding in this section. Instead, I will focus on a selection
of the most relevant classics—Husserl, Schütz, Sartre, and Habermas, with a gradu-
ally diminishing level of detail. After sketching the role of intersubjectivity in each
author’s philosophical approach, I will extract specific and decontextualizable in-
sights that I see as fruitful contributions to a sociological solution to the problem of
intersubjectivity.
The common denominator of these authors is a conception of intersubjectivity
as shared understanding (which Habermas interprets more normatively than do his
predecessors). Husserl, to begin with, does not believe that sharing knowledge, un-
derstanding each other, or even achieving consensus is a problem for everyday people.
He realizes that intersubjectivity is mainly a self-constructed puzzle of the philos-
ophy of the subject. This contrast follows Husserl’s general distinction between a
phenomenology of the “life-world” and a phenomenology of the transcendental ego
(Husserl 1976). The former, thought less highly of by Husserl than by Schütz, is an
essentially descriptive reconstruction of the “natural attitude” with which everyday
man faces the world. According to Husserl, the idealized belief that other people
are intelligible is part of this stance and paraphrased by him as a “general thesis of
the natural attitude” (Husserl 1976:51f/28–30, 1956:166/47). Since he believes that
rigid, scientifically satisfying explanations are not possible within a phenomenology
of the life-world, it was one of Husserl’s central aims from the early 1910s until his
death in 1938 to incorporate the notion of intersubjectivity into his phenomenology
of the transcendental ego.
However, tensions between the two modes of analysis are present all along, as
many of Husserl’s critics have noted (most stringently: Schütz 1966; see also Srubar
1983). In contrast to Kant, Husserl’s use of the adjective “transcendental” does not
refer to a certain type of (meta-)knowledge, but to a “motive to ask for the ultimate
sources of all cognitions [Erkenntnisbildungen], to let the apprehending person reflect
over her apprehensions and her apprehending life” (Husserl 1956:100f/26). This de-
sire to find the ultimate—that is, unquestionable—sources of one’s knowledge is to
be satisfied by means of the key “method” of Husserl’s phenomenology, the tran-
scendental reduction or epoché. This method is supposed to result in liberation from
all philosophical or unphilosophical prejudice and to reveal the essential characteris-
tics of the thereby divested “things themselves” (Husserl’s Sachen selbst, not Kant’s
46 SOCIOLOGICAL THEORY

unknowable Ding an sich). However, it does not consist of a procedure of steps one
could simply follow; rather, it is meant to be a single act of suspending the world’s
“web of social facts and values” (Geflecht der Geltungen; Husserl 1956:153/40). From
the countless publications and unpublications in which Husserl attempts to explicate
this method, no picture of a repeatable procedure or act that would satisfy method-
ologists emerges. On balance, the transcendental reduction comprises all of Husserl’s
attempts at distinguishing unessential from essential (unquestionable, transcendental)
characteristics. Among these, Husserl seems to have held Descartes’ priorization of
subjectivity to be the least dubitable. In consequence, the search 5 for intersubjectivity
within the jurisdiction of transcendental phenomenology violates the latter’s own
premises from the outset. Immanent reconstructions of Husserl’s phenomenology of
intersubjectivity are prone to the same fundamental flaw. 6
It requires a more eclectic reading of Husserl to extract insights useful for the so-
ciological purposes pursued here. By far the most systematic account of intersubjec-
tivity can be found in Husserl’s fifth Cartesian Meditation (Husserl 1992) especially
if supplemented with clarifications from his research manuscripts (Husserl 1973a,
1973b, 1973c). Critics often argue that the solution put forward in the fifth medita-
tion suffers from a confusion of the transcendental sphere with that of the life-world
(Downes 1965; Reynaert 2001; Scherbel 1994; Schütz 1966). Instead of seeing this ap-
parent confusion as a defect, I suggest taking it as a sociological plus. 7 Here, Husserl
addresses intersubjectivity as a practical problem that members of any given society
need to solve. He emphasizes with all due clarity that the “general thesis of the
natural attitude” is not sufficient to make another mind interpretable—real mental
events are not available to another person’s experience (see already Husserl 1901:34,
1973a:1–4, 189, 371–73). Accordingly, a person’s attempt at observing another mind
needs to begin with “empathy” (Einfühlung), by which term Husserl means imaginary
observation of the other mind in her own mind. Since this made-up observation is
a mental operation of a real individual, this individual as well as third parties may
assume, Husserl argues, that the observed mind carries out the same operation as
the observing mind. The result is a sociocognitive symmetry between the individual’s
and her alter ego’s attempts at observing each other (Husserl 1973b:9). Of course,
the only data available to each of them concern the outer world, including data
about each other’s bodies. Thus, a modern-day reader of Husserl will remark that
neither of the two can be sure that the other is actually attempting to observe his
or her respective other. However, this form of security is by no means necessary to
establish the viable assumption that another mind is present and observing in just
the same way as oneself. The assumption can be made valid by some other force
Husserl neither mentions nor excludes—for instance, an evolutionarily selected, uni-
versal tendency to observe other bodies and faces as minds, in the way suggested by
Levin.
At any rate, if this assumption is made by all participants in an encounter, an-
other body will “indicate” (Husserl 1992:120/53) another mind and external events

5 Husserl’s search for an adequate conception of intersubjectivity is documented in the almost 2,000
pages of “research manuscripts” published posthumously (Husserl 1973a, 1973b, 1973c).
6 For important examples of such immanent and relatively orthodox reconstructions and defenses, see
Kozlowski (1991), Römpp (1992), and Zahavi (1996). All of these authors remain fascinated—if not to
say dazzled—by the complexity and the paradoxes of Husserl’s theory; none attempts anything like a
clear and consistent summary of his conception of intersubjectivity.
7 The price of this selective reconstruction is that it will be necessary to ignore Husserl’s (in any case,
paradoxical) idea of a “double phenomenological reduction” (Husserl 1992:118/52). For a more favorable
interpretation, see Scherbel (1994:278f).
INTERSUBJECTIVITY 47

observed by one mind can be assumed equally observable by the other mind. Thus,
the individual and her alter ego are able to share both knowledge about the world
and knowledge about the fact that they both do so. This type of knowledge is not
shared because it is transmitted from one mind to another (which is impossible),
but because each individual produces such assumptions independently yet with ex-
pectably high likelihood. Although this idea of Husserl constitutes another important
step toward a sociological theory of intersubjectivity, I will try to show in the next
section that Husserl overlooks the role of social interaction in establishing shared
knowledge. Due to the influence of Bergson, Husserl and many of his followers
(e.g., Schütz, Sartre, and Merleau-Ponty) see simultaneity as a sufficient condition
for intersubjectivity. However, if co-observation of indexical events is crucial, more
complex forms of shared knowledge can only be generated if it is up to the observer
and her observee to willingly generate such events. This means that an adequate the-
ory of intersubjectivity has to account for the function of the temporal organization
of interaction. 8
With Husserl’s achievement in mind, it will be instructive to turn to one of his stu-
dents to whom the “enigma of how man can understand his fellow-man” (see again
Schütz 1962d:179) seemed most pressing. The main concern of Alfred Schütz’s social
phenomenology is to naturalize (i.e., detranscendentalize) the problem of intersubjec-
tivity and thereby to situate it in the life-world of everyday man (Schütz 1974:56). I
would like to suggest that Schütz carries out this task in two different and incompat-
ible directions. The first and certainly more prominent direction only adds confusion
to Husserl’s already naturalized solution. Within the life-world, Schütz conceptu-
alizes intersubjectivity as a gradual problem. Although “intended meanings” (Max
Weber’s gemeinter Sinn) are inaccessible to another person, the “intelligibility of the
other’s mental states” is not “denied in general” (Schütz 1974:139f/19; even more
radically: Schütz 1966:72). Schütz goes on to ensure us that “fragments” (Schütz
1974:147/20) of another person’s stream of experiences are available to an observer,
but he neither explains exactly what parts are available, nor how such fragments are
made available to the outside world, nor if and how an observer can distinguish
beforehand between expectably available and unavailable bits of experience. He does
emphasize that simultaneity of experience is necessary to “catch the other’s thought
in its vivid presence” (Schütz 1962d:173; also Schütz 1974:144f/20). However, all
his attempts at explaining how simultaneous thoughts become shared thoughts drive
him into highly questionable, pseudotopological classifications of different degrees
(e.g., typical vs. idiosyncratic, knowledge of vs. knowledge about) of shared knowl-
edge (Schütz 1962b, 1962c, 1964a:98ff, 1964b:124f). 9 Unfortunately, it was primarily
this rhetorical and ultimately paradoxical explanation of intersubjectivity that was
popularized in sociology by Berger and Luckmann (1966; for a similar reading of
Schütz, see Grathoff and Waldenfels 1983).
In a set of paragraphs in his early monograph as well as some of its later sum-
maries, Schütz takes a second, more fruitful direction toward such an explanation.
There, “understanding the other” (Fremdverstehen) is not conceived of as a prob-
lem of asymptotic approximation, but as an interactant’s hypothetical reconstruction
of the observee’s motives (Schütz 1974:155f/21). In this and only this version, the

8 A temporal perspective is alluded to, however, when Husserl speaks of a “testable accessibility of the
originally inaccessible” (bewährbarer Zugänglichkeit des original Unzugänglichen; Husserl 1992:117/52).
9 Interestingly, the early Schütz is explicitly critical of the usage of topological metaphors in philosophy;
see Schütz (1974:143/20).
48 SOCIOLOGICAL THEORY

problem of intersubjectivity is from the outset simplified—not the other’s inner in-
finity, but only her motives to carry out certain observable acts have to be made
transparent. Schütz argues that since real psychological motives are normally not
observable from the outside, the observer has to presume that a rational-teleological
“in-order-to motive” was underlying a certain act. This presumed motive does not
necessarily correspond to the causal chain of mental events that made the observee
carry out her act. It is a purely observational rationalization that serves to attribute
mental states to the observee. 10 In the extreme case where the only observation-
ally available datum is a certain external effect (e.g., a smashed window, and no
person standing near), the observer can rationalize the act by attributing an in-
tention to bring about this effect (Schütz 1974:243/35). Interestingly, Schütz takes
such attributions to be necessary for communicative interaction. To discover “ex-
pressive acts” behind “expressive movements” (or expressive sounds), the hearer
needs to observe the speaker by means of an attributed in-order-to motive (Schütz
1974:162f/23). Like Hollis, Schütz thereby suggests an unequivocally attributive ap-
proach to social observation, and the two authors can be seen as complementary.
To observe another mind at least hypothetically, an individual needs to assume that
this other relates behavior and inner states in a consistent manner. Each act is
then identifiable as consistently grounded in an in-order-to motive to bring about
this act, which means that observable information becomes a predictor of unob-
servable information. The role of motives or intentions in this observation is a
purely constructive one; they serve as virtual anchor points of real-world data.
These valuable theoretical ideas do not compensate for the lack of a truly inter-
actional perspective in Schütz, a deficit that was already noted in Husserl, but they
pave the way to the solution to the problem of intersubjectivity that I shall sketch
below.
Jean-Paul Sartre, like Schütz an indirect student of Husserl, is equally critical of
Husserl’s transcendental conception of intersubjectivity. 11 He demonstrates in detail
that neither the universality of human nature nor the universality of temporal ex-
perience is sufficient to render concrete contents of different minds communicable
(Sartre 1976:265). Similar to Schütz, he wants to situate intersubjectivity in the con-
crete reality of everyday life, although his argument is more sensitive to the logical
intricacies of this decision. Not unlike Mead (1934), Sartre’s concern for intersubjec-
tivity derives from his interest in the role of the anticipated other in the development
of reflexivity. 12 If an individual cannot register the presence of other individuals in
her world, this world will have the shape of a series of plain and unproblematic
experiences revolving around the self like planets around a central star. The indi-
vidual exists in this world, but she is not stimulated to reflect upon it. 13 As soon
as another self enters, this world becomes “decentered.” All the appearances she
perceives can now be interpreted as organized around a different hub. Furthermore,

10 According to Schütz, actions are carried out due to psychological “because motives”; “in-order-to
motives” have the sole purpose of making “actions” meaningful (to another person or even to the agent
himself) and turning them ex post facto (“You did X in order to . . . ”) or modo futuri exacti (“You will
have done Y in order to . . . ”) into “acts” (Schütz 1962a:71).
11 I will not consider Sartre’s early views as laid out in his Transcendence of the Ego (Sartre 1988), since
his major philosophical work, Being and Nothingness, claims to invalidate them (Sartre 1976:274).
12 Although Sartre’s reader should keep in mind that his existentialism regards the genesis of the self
as an ontological problem, thereby abstracting from fundamental social-psychological processes (e.g.,
socialization and language acquisition).
13 In Sartre’s beautiful French prose, the ego lives in a “pure manière de me perdre dans le monde, de
me faire boire par les choses comme l’encre par un buvard” (Sartre 1976:298).
INTERSUBJECTIVITY 49

in meeting the other’s eyes, the individual realizes that she, too, can become the
object of someone else’s experience. According to Sartre, the individual responds to
this archetypal social situation by controlling and adjusting her own appearances, a
mechanism that he interprets emphatically as a fundamental yet necessary restriction
of an individual’s freedom (Sartre 1976:304).
Sartre’s existentialism is mainly interested in the consequences of this situation for
the individual, not for the social situation or society in general. From a sociological
perspective, Sartre’s analysis of the relationship between observability and causality
seems to be his most original contribution to the discourse on intersubjectivity, since
it clarifies one of the key logical conditions under which attempts at observing ob-
servers operate. Like several of the authors discussed above, Sartre sees the existence
of the other as a purely observational construction. This construction should not be
assessed with respect to its epistemological adequacy but merely its practical utility:
“The other . . . is an a priori hypothesis, which has no other justification than the
unity, which it allows to operate in our experience” (Sartre 1976:265). However, by
transforming an observational object into an observing subject, the individual bur-
dens herself with self-generated contingency. This means that the laws of causality
are observationally inapplicable as long as the hypothesis is operative. Causal explana-
tions about the behavior of the other can be reintroduced at any time, but only at the
price of suspending the hypothesis that the other is an observer (Sartre 1976:264f,
341). In Luhmann’s terminology (see the next section), the difference between ob-
serving someone as a (causally determined) object and a (causally underdetermined)
subject corresponds to the distinction between first-order observation and second-
order observation. Unfortunately, the crucial question of how observation of others
as subjects can proceed without using causal explanations is already outside Sartre’s
scope.
Even Jürgen Habermas assigns the notion of intersubjectivity an important posi-
tion in his theory of communicative action (Habermas 1981a, 1981b). This theory
distinguishes between four “types” of action (teleological, normatively regulated,
dramaturgical, communicative) and claims that all but the last are unsuitable for
the purpose of reaching social understanding, as they create or presuppose a social
situation in which participants fail to acknowledge each other as equally valuable,
knowledgeable, and rational interaction partners. Habermas’s interest in the con-
cept of communicative action can be viewed in terms of his broader theoretical
project, which focuses on the communicative achievement of consensus. His vin-
dication of communicative action gives him the leverage to argue for the ethical
superiority of a procedure (i.e., “discourse”) that, he thinks, methodically guarantees
this achievement. Intersubjectivity is seen as both a requirement for, and a result
of, communicative action—Habermas claims that the intersubjectively acknowledged
validity of social rules (e.g., speech act rules) is a precondition to the communicative
achievement of shared meaning and consensus (Habermas 1982:411f). This poten-
tially circular reasoning makes it impossible to explain intersubjectivity within the
framework of the theory of communicative action (Schneider 1994b:195–210). In
consequence of these premises, and due to Habermas’s faithful application of the
speech-act-theoretical model of communication, 14 he never considers it necessary to
deliberate on the theoretical problem of how empirical understanding is possible.

14 Schegloff criticizes this recourse to speech act theory when he mentions that Habermas “has con-
tributed to the subversion of his own goals by relying on an analytic resource that in effect casts action
as atomistic, individualistic, atemporal, asequential, and asocial” (Schegloff 1992b:1339).
50 SOCIOLOGICAL THEORY

Despite Habermas’s sociological prominence, I do not see insights on which a solu-

tion to the problem of intersubjectivity could draw.
It will be useful to sum up the selectively extracted contributions of Husserl,
Schütz, and Sartre in their logical rather than historical order. The first explanatory
step is provided by Sartre, whose distinction between (deterministic) causality and
(indeterministic) selectivity reveals that observation of others as subjects creates a
problem of self-imposed contingency. If they are difficult to observe, this is not due
to their intrinsic capriciousness, but due to the observer’s own decision to observe
them as causally underdetermined observers. In the next logical step, Husserl and
Schütz provide us with complementary explanations of how this informational gap
can be refilled without falling back into observation of objects. In his fifth Cartesian
meditation, Husserl discusses worldly events and objects as indexes that may be used
for acquiring, in modernized terminology, first-order knowledge (i.e., knowledge of
events) and second-order knowledge (i.e., knowledge of co-interactant’s knowledge of
events). Each individual’s perceptual and mental organization makes the generation
of such knowledge likely. In consequence, first- and second-order knowledge becomes
expectably distributed over all participants in a social encounter. Schütz’s theory of
in-order-to motives explains how interactants can transform some indexical events
(especially “expressive movements” or expressive sounds) into acts of co-interactants.
They bluntly construct a direct correspondence between an act and their alter ego’s
intention to bring about this act, thereby refilling the aforementioned informational
gap attributively. Together with the contributions from analytical philosophy dis-
cussed above, something like a “big picture” of the achievement of intersubjectivity
in social interaction begins to emerge. The next section will move from philosophy
to sociology and inspect systems theorists’ thoughts on this issue.

THE (SOCIOLOGICAL) PROBLEM OF DOUBLE CONTINGENCY

According to the theories of social systems by Parsons and Luhmann, the mutual
intransparency of all participants in any social encounter and the consequent un-
predictability of each other’s behavior lead to an indeterminate situation that gives
room to the autonomous emergence of a third force—social systems. However, im-
portant differences between the two theorists can already be found in their respective
suggestions of social solutions to this primordial situation of “double contingency.”
Since both authors use the theorem of double contingency exclusively to justify their
theoretical starting points and hardly ever return to it at more advanced or more
applied points in their complex theoretical constructions, 15 some brief remarks on
these theories will be sufficient to characterize the role Parsons’s and Luhmann’s
respective solutions to the problem of double contingency play in them.
In Talcott Parsons’s work, social systems emerge as normative solutions to the
“Hobbesian problem of order” (Parsons 1951:36), the war of “every man against
every man.” 16 This unbearable situation is brought about by the double contingency
of interaction. Since there may never have been a human population without some
form of normative order, Parsons introduces double contingency as an analytical

15 Luhmann notes this about Parsons (Luhmann 1984:149, 174f). However, the same must be said of
Luhmann’s own theory, since Chapter 3 of the cited monograph, titled Double Contingency, features his
only detailed discussion of the concept.
16 The full quote is: “Hereby it is manifest that during the time men live without a common power to
keep them all in awe, they are in that condition which is called war; and such a war as is of every man
against every man” (Hobbes 1651:Ch. [XIII]).
INTERSUBJECTIVITY 51

problem—a problem that does not exist in the ideal typical shape given to it by
sociological theory. It is merely an implicit reference problem of social systems. In
its ideal typical form, double contingency is induced by “ego’s” 17 ability to select
meaningfully motivated and goal-directed behavior, in other words, to act (Parsons
1968:436). Since all of her acts, therefore, are tied to prior expectations regarding
the achievement of social goals, they are from the outset susceptible to the “polarity
of gratification and deprivation” (Parsons and Shils 1951:234). Any motivated and
goal-directed act can result in a future state that satisfies or disappoints ego, but
only because it is (by Parsons’s definition) a constituent of action to seek satisfaction
and to avoid disappointment.
In interaction, the fulfillment of ego’s expectations depends not only on her own
action (plus the contingent structure of the world), but also on the actions of
other “social objects or alters” (Parsons and Shils 1951:14f). Since these actions
are tied to their respective actor’s expectations, ego’s capacity to carry out a success-
ful act will depend on her ability to produce expectations about alter’s expectations
(Parsons and Shils 1951:65). Relative to such second-order expectations, the funda-
mental distinction between gratification and deprivation is instantiated by the more
specific distinction between conformity and nonconformity. Everything alter does
will automatically be either conforming or not conforming to ego’s expectations
(Parsons 1951:37). If we interchange the two roles, everything ego does will be either
conforming or not conforming to alter’s expectations. This leads to a dilemma. In
order to observe each other adequately, ego and alter would have to produce third-
order expectations, fourth-order expectations, and so on ad infinitum (cf. Goffman
1990:30). Interaction does not come with a built-in rule that specifies where to stop
this proliferation of expectations. The psychological truism that neither ego nor alter
is capable of developing infinitely complex expectations does not take the edge off
this problem, since neither of them can be sure that the other will stop this pro-
cess at any specific (say, at exactly the third) nesting level. It is for this reason that
Parsons sees ideal-typical interaction as characterized by the dilemma of “double
contingency” (Parsons 1951:10, 1968:436; Parsons and Shils 1951:16).
It is one of the central tenets of Parsons’s theory of social systems that the ensuing
“problem of order” cannot be solved locally, that is, from within the encounter. Any
party’s private decision to be honest or trustworthy is insufficient to reduce either
party’s uncertainty about the other party’s next step. Parsons believes that social
order has to enter from the outside, in the form of constraints on interaction that
are organized in social systems. He regards social norms as the most important
form of interactional constraints. In his version of systems theory, it is up to a
“shared basis of normative order” in the form of a “common culture” or a “symbolic
system” to guarantee the “stability” of social interaction (Parsons 1968:437). To use
March and Simon’s terminology, norms—preferably internalized ones—are expected
to “absorb uncertainty” by forcing interactants to act conformingly and by making
their conforming actions expectable (March and Simon 1958). Consequently—and
this consequence is vital—Parsons envisages norms as conditionals that have to be
simply followed.

17 Parsons uses “ego” and “alter” as purely formal reference points (Parsons and Shils 1951:15). Ego
is a symbol for the actor the theorist is chiefly observing, whereas alter refers to the one the theorist
observes insofar as he is observed by, or orienting to, ego. This terminology, which Luhmann uses as
well, converges nicely with the usage of “she/he” here, and I therefore merge the two conventions in the
remainder of the article.
52 SOCIOLOGICAL THEORY

I would like to suggest that Parsons’s exposition of the dilemma of double contin-
gency is a valid, pertinent, and genuinely sociological characterization of the problem
of intersubjectivity, whereas his normative solution to the problem is flawed. In line
with contributions to the discourses on other minds and phenomenological intersub-
jectivity, Parsons sees observational contingency as a self-induced problem. He goes
beyond these contributions when he characterizes the socially multiplied result of this
form of contingency, but he falls behind them as soon as he opts for a normative
solution. Naturally, if alter follows internalized norms, and if ego can expect that he
does so, she can also predict his behavior. However, Parsons’s solution has to operate
symmetrically on alter as well as ego, which means that ego would be constrained
to let her behavior depend only on such expectations about alter’s actions that alter
himself can expect. Independent of the nesting level of expectations allowed by the
“normative order,” ego and alter are transformed into simple norm-followers rather
than norm-interpreters. Norms are supposed to define human behavior as software
defines the behavior of a computer. Any actor’s attempt at covertly breaking norms
will threaten the normative order just as a computer processor’s failure to correctly
execute a program will result in a crash. Thus, Parsons’s solution must be rejected on
the ground of turning interactants into “judgmental dopes” (Garfinkel 1967:66ff). 18
It may be part of the Freudian heritage in Parsons’s works that they express much
greater faith in the constraining power of the unconscious than that of the conscious,
of internalized dispositions than of incentives for rational but prosocial utility maxi-
mization. Furthermore, Parsons’s solution must be (and has been) criticized for being
empirically implausible, since the “normative order” of present day Western society
clearly stimulates a great deal of individual rationality. 19 If most social encounters
in modern society still proceed without sudden outbreaks of violence or long awk-
ward silences, which seems to be the case, 20 Parsons’s solution has to be an at least
incomplete account of the social order operative in typical interactions.
Applying Sartre’s distinction between observing causality (i.e., objects) and ob-
serving selectivity (subjects/observers), Parsons’s solution to the problem of double
contingency implies that society makes people observable by never really letting them
achieve the interactional status of subjects. Parsons’s “normative order” is a world
where double contingency is abolished by way of double conditioning. However,
like a dictatorship, a closed mental hospital, or a concentration camp, this order
can only survive if it can prevent the emergence of creative, self-referential actors
(Burns and Flam 1987:11). A look at the reality of interaction in “total institutions”
demonstrates that causal conditioning is not, and has never been, possible (Goffman
1961:12ff; Sofsky 1997).
Niklas Luhmann is one of the most astute critics of Parsons’s inability to incor-
porate situational contingencies, individual creativity (in his words: the “operative
autonomy of mental systems”), and conflict into systems theory. His own theory of

18 Compare Wrong’s rejection of Parsons’ “oversocialized” conception of man (Wrong 1961). Wrong’s
criticism remains topical even though his alternative, Freudian concept of internalization is only slightly
less oversocialized (“for in psychoanalytic terms to say that a norm has been internalized, or introjected
to become part of the superego, is to say no more than that a person will suffer guilt-feelings if he fails
to live up to it, not that he will in fact live up to it in his behavior” (Wrong 1961:187)).
19 This has been demonstrated in detail in Goffman’s empirical studies of people’s situational enactment
of norms (especially, Goffman 1961, 1990; see also Garfinkel 1967:Ch. 5).
20 Note the efforts it took Garfinkel and his students to “breach” the normal course of interactions
and to produce embarrassment (Garfinkel 1963). Even under disconcerting conditions, some form of
communication went on (e.g., conflict) and neither physical violence nor silence was likely. Schneider
remarks that these experiments should be seen as proof of the robustness of communication, rather than
of its fragility (Schneider 1994a:228).
INTERSUBJECTIVITY 53

social systems (in the version developed after 1980 21 ) is assembled on premises that
are, among other things, intended to mend these defects. Interestingly enough, the
construction of his theory begins with the same reference problem—double contin-
gency. However, instead of using Parsons’s pragmatist exposition of the problem, he
adopts a more abstract construction that is grounded in the cybernetic theory of
black boxes. Because he imports concepts like “system,” “autonomy,” or “observa-
tion” from this theory as well, it deserves a close look.
The problem of “black” or impenetrable objects arose first in wartime engineering
when specialists had to make sense of entities like wreckage or sealed bombs (Ashby
1956:86). Instead of taking them apart immediately, the engineer would treat them
like autonomous subjects, stimulate their surface (e.g., attach a set of cables), and
observe reactions. Depending on these reactions, he or she could form hypotheses
about the unobservable internal mechanism of the black box. The cybernetic black-
box theory of the 1950s generalized this procedure to a methodology of dealing with
incomplete knowledge about causal mechanisms. The experimenter and the black box
were seen as forming a composite system in which “input” from the experimenter
to the box and “output” in the opposite direction are linked in a causally closed
feedback loop. Starting with random assumptions, the experimenter would discover
patterns in the output behavior and ultimately be able to hypothesize a model of the
mechanism that transforms input into output (Ashby 1956:89ff).
If early black-box theory did not lead to a breakthrough in engineering, it certainly
did contribute original ideas to epistemology. Newer developments in black-box
theory eliminate inconsistencies in the older approach that are due to a naı̈ve concept
of causality (see especially von Foerster’s essays, collected in Foerster 1999). 22 The
ignorance that gives rise to a black-box experiment is no longer interpreted as a
temporary white spot in the universal chain of causes and effects. Instead, the black
box is renamed a system and seen as autonomous relative to external causes. Such
a system is capable of selecting itself the environmental effects to which it reacts.
Similarly, it is able to produce own effects. Given this conception, it is theoretically
quite unimportant what else a system may be (e.g., whether it consists of elements
plus relationships or temporalized operations 23 )—its single most important property
is that it decides by itself what it does. The newer black-box theory (Foerster 1999;
Glasersfeld 1995) does not maintain that systems simply exist “out there,” but it does
make the following two claims. First, it argues that the constructivist assumption that
some things in the world (especially living organisms) are systems leads to far more
parsimonious explanations than does causal reductionism. In support of this view, it
deserves mentioning that according to Levin, the failure of biological behaviorism has
shown that explanations that treat organisms as objects and not as subjects/systems
are either simplistic or inefficient. Second, newer black-box theory asserts that the
assumption made in the first claim mirrors the practical constructivism that is built
into the ways in which higher organisms deal with each other in interaction—an
idea that was discussed earlier in this article.

21 Luhmann’s publication of Soziale Systeme in 1984 marks his turn from a theory of action systems to
communication systems (for reasons behind this turn, see Stichweh 2000).
22 Von Foerster and his colleagues are now known as representatives of a successor to classical cyber-
netics sometimes named second-order cybernetics. Wiener, the founding father of classical cybernetics, did
not problematize the notion of causality—perhaps because he was chiefly interested in control, thereby
taking causality for granted (Wiener 1961:39; see also Esposito 1996; Glanville 1979, 1982; Glasersfeld
1995; Maturana and Varela 1980).
23 Luhmann (whose concept of “autopoietic” system emergence derives from second-order cybernetics)
opts for the latter explanation (Luhmann 1978, 1984:76ff).
54 SOCIOLOGICAL THEORY

Luhmann conceives of double contingency as a real (i.e., practical) problem in-

duced by the interaction of two systems. He rejects Parsons’s pragmatist formulation
and normative resolution of the problem, arguing that Parsons is too concerned
with the achievement of consensus (Luhmann 1984:150). Thoroughly considered,
Luhmann claims, double contingency makes consensus—as a static correspondence
between beliefs or preferences held by separate minds—socially irrelevant because
agreement or shared understanding of these minds will be unobservable. 24 For the
same reason, norms for the attainment of consensus are not sufficient to make con-
sensus socially expectable. In consequence, Luhmann sees the social relevance of
norms as altogether reduced, and they enter his theory at a conceptually late and
comparatively marginal point.
Luhmann suggests that double contingency is solved dynamically instead of stat-
ically (Luhmann 1984:150). If double contingency is not an analytical problem but
one occurring in the dynamics of concrete social encounters, almost any random
“self-constrainment” (Selbsteinschränkung/Selbstfestlegung) of either participant will
contribute to its reduction. Any behaviorally consequential attempt by ego to form
an assumption about what alter is planning or intending, Luhmann argues, can be
treated by alter as an interactional fact to which he can react. In Luhmann’s words:
“The experience of contingency brings about the . . . transformation of random events
into probabilities for the formation of structure” (Luhmann 1984:170f, emphasis in
the original). A sequence of interactional moves absorbs uncertainty by stabilizing
ego’s and alter’s expectations, not their behavior (Luhmann 1984:158). In Luhmann’s
theoretical edifice, such sequences of moves are theorized more precisely, but also
more abstractly, under the heading of communication instead of interaction. Commu-
nication is conceived of as a “synthesis”—unfortunately, one of many unexplained
concepts in Luhmann’s theory—of three selections, namely, utterance, information,
and understanding (Luhmann 1984:Ch. 4; a more detailed analysis of understand-
ing is given in Luhmann 1986). It involves two systems (not necessarily co-present
ones, as in the case of written communication), ego and alter, with alter selecting an
utterance meant to convey a bit of information, and ego understanding what alter
is doing by way of distinguishing the utterance from the information conveyed by
it. Of these three selections, Luhmann holds, understanding is the decisive one, since
it “synthesizes” the other two, thereby producing one single, compact operation:
communication.
Puzzling about this link in the chain of reasoning, Luhmann’s reader can hardly
avoid the impression that the theoretical edifice of systems theory is not so robust
after all. Contrary to earlier promises (e.g., Luhmann 1984:158), ego’s and alter’s
expectations are never coordinated. Instead, a social system appears on stage and
takes over control. According to Luhmann’s oft-repeated definition, the operation
of communication is the fundamental and unit-like constituent of social systems.
It operates autonomously vis-à-vis the involved mental systems, decomposing “it-
self” into utterance, information, and understanding (Luhmann 1990:24). 25 Similarly,
double contingency is understood as an “autocatalytic” problem that contributes

24 Compare Luhmann’s critique of the notion of intersubjectivity: “The ‘inter’ contradicts the ‘subject’.
Or more precisely: every subject has its own intersubjectivity” (1995:170).
25 Luhmann never explains how communication can be a “unit” and at the same time be able to
“decompose itself.” Furthermore, some statements in Luhmann’s works suggest (in explicit contradiction
to the statements referred to above) that the “unity” of the selections is provided by the social system,
whereas the selections (utterance, etc.) are not (Luhmann 1984:658, 1990:38f). For a more favorable
reconstruction of Luhmann’s theory of communication by one of his students, see Fuchs (1993).
INTERSUBJECTIVITY 55

continuously to its own reduction by “consuming” selections of mental systems

without ever being “consumed” itself (Luhmann 1984:169f). It leads to the emer-
gence of a social system, which is to be seen as a new level of order, similar to the
idea that mental operations represent a higher level of order than neural events. The
subsequent parts of Luhmann’s argument as well as most of his applied analyses of
specific social systems (law, politics, science, among others) or types of social sys-
tems (society, interaction, organization) simply presuppose the autonomy of social
systems.
Regardless of the question of whether Luhmann’s theorem of emergence has to
be rejected as metaphysical or not, the crucial question seems to be whether double
contingency does or does not prevent the interactional production of observably
shared knowledge. Is intersubjectivity unattainable? Luhmann’s contempt for this
concept (Luhmann 1995) often leads him to an affirmative answer to this question
(Luhmann 1990:26, 1997:206). Yet his skepticism is usually targeted at the early
phenomenological conception of intersubjectivity as a static correspondence between
independent subjectivities. At the same time, Luhmann views communication as a
process that promotes the social “creation of redundancy”—in the sense of shared
knowledge (Luhmann 1995:178, referring to Bateson 2000:412). Although he does
not spell out the details, his theory can thus be interpreted as converging with
ethnomethodological/conversation-analytical perspectives on intersubjectivity as a
sequential achievement (Garfinkel 1967:30; Schegloff 1991:152, 1992b:1298).
Just like the discourses on other minds and phenomenological intersubjectivity,
theories of social systems bring us small but important steps closer to a solution
to the issue in question. The lesson to be learned from Parsons is that we need
to convert the psychological problem of mental intransparency into a sociological
problem of social coordination to be able to solve both problems. Ego’s and alter’s
problem of understanding the respective other have to be solved at the same time,
which means that their assumptions about each other need to be coordinated by an
interactional “meta-solution.” Accordingly, a theory of intersubjectivity must come
in the form of a theory of interaction. Luhmann, on the other hand, shows that
any such coordination (whether the one he envisages or any other one) has to be
a process, it must expend time in order to allow ego and alter to generate and
validate assumptions about each other in a sequence of moves. These suggestions
extend the field we have to survey for contributions to the problem(s) of intersubjec-
tivity. Sociological research on interaction and talk needs to be seen as potentially
relevant. Hence, it is no coincidence that it will be the founding father of conver-
sation analysis who leads us toward a first rough sketch of such a coordinating
meta-solution.

A SOLUTION TO THE PROBLEM(S) OF INTERSUBJECTIVITY

My problem-oriented, eclectic reading of three discourses on three different, largely
disconnected problems of intersubjectivity is likely to leave exegetes unsatisfied. The
advantage of this approach is that it allows us now to take stock and consider to
what extent the three discourses add up to, or approximate, a solution. A number
of properties of such a solution have certainly emerged. First, it has become clear
that its scope can and must remain modest, as social understanding can never result
in more than selected bits and pieces of shared knowledge—the inner infinity of
another person remains forever unavailable (Wittgenstein, Schütz). Second, intersub-
jectivity is not a fixed state but an interactional achievement (Husserl), depending on
56 SOCIOLOGICAL THEORY

the workings of a communicative infrastructure that coordinates the way in which

interactants acquire knowledge of each other (Husserl, Schütz, Parsons, Luhmann,
Schegloff). Third, intersubjectivity requires certain prior cognitive attitudes on the
part of ego and alter, the interactants (Sartre). Most importantly, it demands ego to
(1) interpret alter’s behavior as intentional, as grounded in in-order-to motives (Levin,
Schütz) and to (2) attribute alter a generalized disposition to organize his inner states
(beliefs, desires, etc.) in a socially expectable—that is, internally consistent—manner
(Hollis).
The review of the three discourses also provides us with a fourth conclusion, one
which none of the mentioned authors considers in a sufficiently radical way. Al-
though social understanding, as an empirical phenomenon, occurs frequently and
with remarkable reliability, it cannot be the result of an actual “observation” car-
ried out by ego when she interacts with alter. Schütz’s and Hollis’s attributive turn
forces us to acknowledge that a conception of the problem of intersubjectivity as
an issue of social observability is circular and therefore misguided. This means that
the underlying problem needs to be reframed. In this section, I shall argue that we
need to frame it as a problem of accountability: social understanding is achieved by
means of an interactional infrastructure, which holds agents accountable to make
themselves understandable. This basic idea comes from the works of Harvey Sacks;
it is an idea that I continue to find stunning not only in its intellectual audacity
but also in its sociological reach. It can be found in two rather isolated (and long
unpublished) parts of his intellectual legacy, 26 essentially unconnected to the more
narrowly focused enterprise of conversation analysis for which he became known
(Segerdahl 1998; Silverman 1998).
Whereas all other authors discussed in this article situate the problem of intersub-
jectivity in a symmetrical social situation, Sacks constructs his argument around an
initially asymmetrical encounter. Ego tries to observe alter and is in a position to
draw consequences from her observation, consequences that will affect alter’s posi-
tively or negatively. For instance, ego is a police officer, alter a stroller, and ego has
to decide whether to stop him and check his ID (Sacks 1972:282). Now, what if ego,
though notoriously unable to look inside alter’s head, just decides to observe him
arbitrarily? For instance, she may establish a rule that everybody wearing a grim
expression has criminal intentions and is worth checking. If ego were to use such a
rule, it would be vital for alter to be able to count on her using it, even though he
may find it unreasonable or unfair. As soon as he is able to anticipate ego’s policing
rule, it will be rational for him to adjust his behavior accordingly—for instance, by
smiling at her. Having cultivated an early interest in the Old Testament, Sacks seems
to take pleasure in finding the prototype of this situation in the tale of the Fall of
Man and in the Book of Job. 27 To recall the observational problem narrated in the
former tale, consider that God is (or behaves as if He is) confronted with the task
of finding out why Adam and his wife are hiding themselves behind the trees. Adam
answers to God’s inquiry by avowing that he is ashamed of his nakedness—thereby

26 The first of the two is a student paper Sacks wrote for a seminar with Goffman while he was a
graduate student in Berkeley in the early 1960s; it was published three years before Sacks’s tragic death
(Sacks 1972). The second is an early, unusually theoretical lecture Sacks gave to undergraduates at UCLA,
first published posthumously in an abridged version in Dijk (1985) and later in Sacks’s collected lectures
on conversation. I will refer to it according to this later and unabridged version (Sacks 1992).
27 For Sacks’s interesting intellectual biography, see Schegloff (1992a). A special interest in the Book of
Job is a quirk Sacks shares with Goffman, who makes numerous references to Job and the problem of
“vital tests” throughout his works (Creelan 1984).
INTERSUBJECTIVITY 57

revealing that he has acquired knowledge of good and bad by having eaten from the
forbidden tree. The consequences of God’s observation (and Adam’s failure to antic-
ipate it) are, of course, devastating—the discovery of “being an observable” (Sacks
1972:444, n1) comes at a high price.
The observational problem related in the Book of Job is equally dramatic. Job’s
friends find themselves in the predicament of having to make sense of the series of
bad blows by which his life has been struck. Since they, like Job, believe that a happy
and prosperous life is God’s reward for the good, they urge him to confess his guilt.
Job’s problem—Sacks calls it “one of the central dilemmas of Western civilization”
(Sacks 1992:118)—is precisely that the social observation of a fact that refers to the
internal state of another mind (Job’s alleged lack of faith) is actually carried out
independently of the internal state of this mind.
In Job’s case, this observation is based on wrong assumptions, but the fact that
a mistake has been made remains unrecognized until God reinstates Job’s social
position. In the case of Adam and Eve, God’s observation leads Him to correct
conclusions. In this tale, it is particularly instructive that God uses mere verbal cues
although He does not need them in order to look into the minds of men. God draws
a conclusion that any grown-up, socialized, and interaction-competent member of
society could have drawn in His place. Adam, Eve, and their children learn that their
observable behavior commits them to socially consequential hypotheses about their
unobservable self. They discover that they have to anticipate how social observations
are carried out in order to gain control over their application. The very next time
someone burdens himself with guilt, he will anticipate the consequences of his social
observability and act on this anticipation—when God asks Cain where his brother
Abel is, Cain tries to act the fool. His unavoidable punishment reflects the crime of
dissimulation rather than the crime of murder. In Sacks’s words: “He, we recall, is
condemned not to death, but to ‘observability for life’” (Sacks 1972:444, n2).
These considerations are more than just an original interpretation of ancient Jew-
ish mythology. Sacks uses the Old Testament as a quarry for sociologically instructive
tales. The problem he illustrates by the two tales is of general character. There exists
no social situation in which the other’s mental states or events are “observable” in
the traditional, correspondence-theoretical sense of the word. Observation of other
minds, however it may be attempted, cannot establish equivalence between the rep-
resented and the representing. In consequence, the impenetrability of the observee
cannot constitute, and can never have constituted, a “problem.” The society we all
live in, the encounters we all participate in must be of a kind that allows us to
do without such information. It must therefore be up to people themselves to aid
their observers in making sense of them. How can they be stimulated to do this?
Obviously, influenced by the dominant of Parsons during 1960s, Sacks suggests that
people are socialized to adapt themselves to the unsophisticated observational “ap-
paratus” by means of which society matches individuals with social categories like
gender, class, or status (Sacks 1992:117). Socialization, Sacks argues, does not have
to (and could never) convert children into norm-following “judgmental dopes,” but
it must be able to convert them into rational social beings who behave in crucial
situations like judgmental dopes—in other words, who behave sufficiently expectable
(including expectably creative, expectably eccentric, expectably role-distanced, and so
forth).
This final part of Sacks’s argument falters for essentially the same reasons that
were brought up against Parsons’s dissolution of the problem of double contingency.
However, we only have to connect Sacks’s key idea to the results from the review of
58 SOCIOLOGICAL THEORY

the discourses on intersubjectivity in order to arrive at a completed solution. It can

be expressed in three steps.

1. Understanding of (or knowledge sharing with) another person is possible in

two ways: observational understanding and communicative understanding. Ob-
servational understanding, a simplistic case, which I mention for the sake of
analytical exhaustiveness, works only for mental or affective states, which have
a hardwired link to bodily states or behaviors—for example, anger, blushing,
shame, or nervousness. In other words, it works for internal states, which are
not intentional or chosen (recall Sartre!). The understander looks at the un-
derstandee’s body or behavior, discerns a specific phenomenon (say, a red face
and a trembling voice), decides it is unintentional and links it to an internal
state (say, shame).
2. Communicative understanding is initiated when the understander attributes the
understandee both intentionality and internal consistency, the former being
necessary to reconstruct in-order-to motives and the latter to exclude “hid-
den” motives and causes (say, abnormal psychological dispositions). The un-
derstander then uses socially standardized rules in order to decide what the
understandee’s behavior is supposed to tell her about what he thinks, knows,
wants to tell her, and so on. The rules have to be socially standardized because
the understandee must be able to anticipate their use by the understander. For
example, the understandee says “Thanks for nothing!” and the understander
uses her knowledge of relevant social rules and of the current situation in
order to decide that he wants to tell her that he is not feeling grateful. As the
example shows, rules have to be standardized but may vary with culture.
3. Application of communicative understanding is a socially viable strategy be-
cause of the interlocking of two coordinative interactional mechanisms.
a. Pragmatic accountability. Understood persons can be held interactionally
accountable for the information they convey because they tend to depend
in some way on the cooperativeness and the responses of the understander.
For example, if alter behaves in a way that tells ego that he dislikes her,
she may choose to upset him by insulting him. Thus, alter finds himself
forced to anticipate just how (on the basis of what situational context,
imputable in-order-to motive, and standardized rules) ego will proceed in
reconstructing information that is technically unavailable to her because it
resides inside his mind. How he behaves, what he says, what he conceals
will or will fail to provide ego with information that can become relevant
to her next actions. In short, alter must deal with the fact that he is under
the obligation to make himself intelligible, to select and control his actions
in accordance with his anticipations of ego’s anticipations.
b. Sequential adjustability. In sequentially open forms of interaction (e.g., face-
to-face conversation), participants have the possibility to calibrate the per-
sonal information, which they have given or given off to others. The inter-
mittency of turns provides them with options for refinement, clarification,
or repair. In such contexts, therefore, communicative understanding is also
a robust strategy, as it does not commit ego to assumptions that cannot be
changed afterward.

This solution coheres with the properties and limitations of intersubjectivity that
were discussed in the article. It also provides a clear idea of what is culturally
INTERSUBJECTIVITY 59

universal (coordinative mechanisms) and culturally specific (interpretive rules) about

the achievement of social understanding in different societies and different social
contexts.

CONCLUSION
The aim of this article was to review several discourses on the problem(s) of in-
tersubjectivity and condense them into a theoretical answer that satisfies the usual
requirements of theory-design in the social sciences. Sociological appropriations of
extended philosophical debates can be tricky; the strategy pursued here was to keep
a constant eye on the basic puzzle underlying these debates (“People understand
each other although they have separate minds”) and investigate how each author
contributes to a sociological explanation of this puzzle. The proposed solution is
certainly interactional in character, although it may be called “situational,” “rela-
tional,” or “institutional” with equal justification. The key idea was provided by
Sacks, who allowed us to replace the—unanswerable—question of “How does ego
see through alter although she cannot?” with the question of “How does alter re-
spond to the fact that ego tries to understand him and acts on her results in ways
that may benefit or harm him?”
Intersubjectivity is a basic theoretical concept of the social sciences; it is men-
tioned in most textbooks on social theory and often seen as marking a fundamental
reference problem of social conduct. However, its relevance is not limited to theory
alone. Empirical fields like education, conflict resolution, or intercultural communi-
cation are dealing on a daily basis with the vagaries of social understanding and
the potentially disastrous effects of misunderstanding. To provide an idea of the
relevance and the limitations of the proposed theoretical solution to applied fields
such as these, I shall briefly revisit the questions that I posed in the introduction. I
will confine myself to preliminary answers to the first three questions, as I believe
that the remaining four should be addressed through empirical research.
Under what social conditions is intersubjectivity likely to be accomplished? Face-to-
face interaction is often identified as the primordial site for social conduct (Goffman
1983). Sociobiological evolution has equipped human beings with behavioral dispo-
sitions and physiognomic resources that are geared to observational understanding
by co-present others (e.g., recognition of emotions or attitudes; see only Ekman and
Friesen 1975). As it is obvious that communicative understanding is often facilitated
by observational understanding, it makes sense to hypothesize that intersubjectiv-
ity is (ceteris paribus) most easily achieved in face-to-face interaction—or at least
voice-to-voice interaction, where prosodic cues are still available. Apart from that,
the solution of the preceding section predicts that intersubjectivity is more likely
to be accomplished if the social situation allows for intermittent turns, not only to
enable repair but also to provide ego with opportunities of holding alter accountable
for the way in which he “makes himself understood.” Interestingly, the solution also
suggests that intersubjectivity does not depend on the cleverness of the interactants.
Achieving it is a coordination problem in the sense of Schelling (1980); a particu-
larly clever interactant may be better at understanding others but is more difficult
to understand.
How much—or how little—intersubjectivity can be achieved under optimal conditions?
By definition, observational understanding is only fit for reconstructing nonchosen
internal states. This renders it insufficient for application to even the simplest forms
of social conduct. Communicative understanding, on the other hand, is incapable of
60 SOCIOLOGICAL THEORY

granting ego any sort of privileged access to alter’s internal states. It only encom-
passes states, which ego must presume, alter wanted her to understand. This result
may be disheartening to proponents of a normative-emphatic conceptualization of
intersubjectivity, but it is a direct consequence of the idea that intersubjectivity is
first and foremost a device for coordinating interdependent social actions.
Are agents only concerned with establishing intersubjectivity or just as much with
simulating or exploiting it for their own purposes? Intersubjectivity is a coordination
device that depends to a significant extent on the cooperation of the interaction
partners. More precisely, communicative understanding rests on the viability of ego’s
assumption that alter wants to be understood by her. This structural design is not
only vulnerable to dishonesty, it is the very reason why dishonesty—as opposed to
dissimulation—is a part of the social world that human beings populate (dissimu-
lation can be found in extremely simple organisms, but dishonesty presupposes a
“theory of mind” and is therefore unknown even among the great apes; Tomasello
et al. 2005:684ff). Thus, the mystery is not that people sometimes manipulate the
ways in which they are being understood, but that they often allow others to under-
stand them in correct or adequate ways. In my view, this is a mystery that sociologists
should leave to evolutionary anthropologists, trusting instead to the happy intuition
that intersubjectivity is, after all, a core aspect of social life.

REFERENCES
Ashby, W. Ross. 1956. An Introduction to Cybernetics. London, UK: Chapman & Hall.
Aune, Bruce. 1961. “The Problem of Other Minds.” Philosophical Review 70:320–39.
Avramides, Anita. 2001. Other Minds. London, UK: Routledge.
Bateson, Gregory. 2000. “Cybernetic Explanation.” Pp. 405–16 in Steps to an Ecology of Mind, edited by
G. Bateson. Chicago, IL: University of Chicago Press.
Berger, Peter L. and Thomas Luckmann. 1966. The Social Construction of Reality: A Treatise in the
Sociology of Knowledge. Garden City, NY: Doubleday.
Boccaccio, Giovanni. 1982. Decameron, translated by J. Payne. Berkeley, CA: University of California
Press.
Botero, Giovanni. 1997. La Ragion di Stato. Roma: Donzelli.
Burns, Tom R. and Helena Flam. 1987. The Shaping of Social Organization: Social Rule System Theory
with Applications. London, UK: Sage.
Creelan, Paul. 1984. “Vicissitudes of the Sacred: Erving Goffman and the Book of Job.” Theory and
Society 13:663–95.
Davidson, Donald. 1980. Essays on Actions and Events. Oxford, UK: Clarendon.
Dennett, Daniel C. 1987. The Intentional Stance. Cambridge, MA: MIT Press.
Descartes, René. 1996. Discourse on the Method: And, Meditations on First Philosophy, translated by D.
Weissman. New Haven, CT: Yale University Press.
Dijk, Teun A. van. 1985. Handbook of Discourse-Analysis, Vol. 3. London, UK: Academic Press.
Downes, Chauncey. 1965. “Husserl and the Coherence of the Other Minds Problem.” Philosophy and
Phenomenological Research 26:253–59.
Dretske, Fred I. 1973. “Perception and Other Minds.” Noûs 7:34–44.
Ekman, Paul and Wallace V. Friesen. 1975. Unmasking the Face: A Guide to Recognizing Emotions from
Facial Clues. Englewood Cliffs, NJ: Prentice Hall.
Esposito, Elena. 1996. “Observing Interpretation: A Sociological View of Hermeneutics.” MLN 111.3:593–
619.
Foerster, Heinz von. 1999. Sicht und Einsicht: Versuche zu einer operativen Erkenntnistheorie, translated
by W. K. Köck. Heidelberg: Carl-Auer-Systeme Verlag.
Fuchs, Peter. 1993. Moderne Kommunikation: Zur Theorie des operativen Displacements. Frankfurt am
Main: Suhrkamp.
Garfinkel, Harold. 1963. “A Conception of, and Experiments with, ‘Trust’ as a Condition of Stable
Concerted Actions.” Pp. 187–238 in Motivation and Social Interaction: Cognitive Determinants, edited
by O. J. Harvey. New York: Ronald Press.
INTERSUBJECTIVITY 61

———. 1967. Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.

Gergely, György and Gergely Csibra. 2003. “Teleological Reasoning in Infancy: The Naı̈ve Theory of
Rational Action.” Trends in Cognitive Sciences 7:287–92.
Gilovich, Thomas, Kenneth Savitsky, and Victoria Husted Medvec. 1998. “The Illusion of Transparency:
Biased Assessments of Others’ Ability to Read One’s Emotional States.” Journal of Personality and
Social Psychology 75:332–46.
Glanville, Ranulph. 1979. “The Form of Cybernetics: Whitening the Black Box.” Pp. 35–42 in General
Systems Research: A Science, a Methodology, a Technology, edited by J. Miller. Louisville, KY: Society
for General Systems Research.
———. 1982. “Inside Every White Box There are Two Black Boxes Trying to Get Out.” Behavioral
Science 27:1–11.
Glasersfeld, Ernst von. 1995. Radical Constructivism: A Way of Knowing and Learning. London, UK:
Falmer.
Glendinning, Simon. 1998. On Being with Others: Heidegger, Derrida, Wittgenstein. London, UK: Rout-
ledge.
Goffman, Erving. 1961. Asylums: Essays on the Social Situation of Mental Patients and Other Inmates.
New York: Anchor Books.
———. 1983. “The Interaction Order.” American Sociological Review 48:1–17.
———. 1990. Stigma: Notes on the Management of Spoiled Identity. London, UK: Penguin.
Gordon, Robert M. 2004. “Folk Psychology as Mental Simulation.” in The Stanford Encyclopedia of
Philosophy, edited by E. N. Zalta. Stanford, CA: Stanford University.
Grathoff, Richard and Bernhard Waldenfels. 1983. “Sozialität und Intersubjektivität: Phänomenologische
Perspektiven der Sozialwissenschaften im Umkreis von Aron Gurwitsch und Alfred Schütz.” München:
Fink.
Guicciardini, Francesco. 1994. Dialogue on the Government of Florence, translated by A. Brown. Cam-
bridge, UK: Cambridge University Press.
Habermas, Jürgen. 1981a. Theorie des Kommunikativen Handelns: Band 1, Handlungsrationalität und
gesellschaftliche Rationalisierung. Frankfurt am Main: Suhrkamp.
———. 1981b. Theorie des Kommunikativen Handelns: Band 2, Zur Kritik der funktionalistischen Vernunft.
Frankfurt am Main: Suhrkamp.
———. 1982. Zur Logik der Sozialwissenschaften. Frankfurt am Main: Suhrkamp.
———. 1988. Nachmetaphysisches Denken: Philosophische Aufsätze. Frankfurt am Main: Suhrkamp.
Hegel, Georg W. F. 1970. Phänomenologie des Geistes (Werke Bd. 3). Frankfurt am Main: Suhrkamp.
Hirschman, Albert O. 1977. The Passions and the Interests. Princeton, NJ: Princeton University Press.
Hobbes, Thomas. 1651. Leviathan. London, UK: Crooke and Crooke.
Hollis, Martin. 1977. Models of Man: Philosophical Thoughts on Social Action. Cambridge, UK: Cam-
bridge University Press.
Husserl, Edmund. 1901. Logische Untersuchungen, Zweiter Theil: Untersuchungen zur Phänomenologie und
Theorie der Erkenntnis. Tübingen: Niemeyer.
———. 1956. Die Krisis der europäischen Wissenschaften und die transzendentale Phänomenologie: Eine
Einleitung in die phänomenologische Philosophie (Husserliana VI). Den Haag: Nijhoff.
———. 1973a. Zur Phänomenologie der Intersubjektivität: Texte aus dem Nachlass, 1. Teil: 1905–1920.
Den Haag: Nijhoff.
———. 1973b. Zur Phänomenologie der Intersubjektivität: Texte aus dem Nachlass, 2. Teil: 1921–1928.
Den Haag: Nijhoff.
———. 1973c. Zur Phänomenologie der Intersubjektivität: Texte aus dem Nachlass, 3. Teil: 1929–1935.
Den Haag: Nijhoff.
———. 1976. Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie, 1. Buch: Allge-
meine Einführung in die reine Phänomenologie (Husserliana III/1). Den Haag: Nijhoff.
———. 1992. Cartesianische Meditationen: Eine Einleitung in die Phänomenologie. Hamburg: Meiner.
Hyslop, Alec. 1995. Other Minds. Dordrecht, The Netherlands: Kluwer.
Kant, Immanuel. 1974. Kritik der reinen Vernunft. Frankfurt am Main: Suhrkamp.
Kirk, Carol A. van. 1986. “Kant and the Problem of Other Minds.” Kant-Studien 77:41–58.
Knorr-Cetina, Karin and Urs Bruegger. 2002. “Global Microstructures: The Virtual Societies of Financial
Markets.” American Journal of Sociology 107:905–50.
Kozlowski, Richard. 1991. Die Aporien der Intersubjektivität: Eine Auseinandersetzung mit Edmund Husserls
Intersubjektivitätstheorie. Würzburg: Königshausen & Neumann.
Levin, Michael E. 1984. “Why We Believe in Other Minds.” Philosophy and Phenomenological Research
44:343–59.
62 SOCIOLOGICAL THEORY

Luhmann, Niklas. 1978. “Temporalization of Complexity.” Pp. 95–111 in Sociocybernetics, Vol. 2, edited
by R. F. Geyer and J. V. D. Zouwen. Den Haag: Nijhoff.
———. 1984. Soziale Systeme: Grundriß einer allgemeinen Theorie. Frankfurt am Main: Suhrkamp.
———. 1986. “Systeme verstehen Systeme.” Pp. 72–117 in Zwischen Intransparenz und Verstehen: Fragen
an die Pädagogik, edited by N. Luhmann and K. E. Schorr. Frankfurt am Main: Suhrkamp.
———. 1989. “Staat und Staatsräson im Übergang von traditionaler Herrschaft zu moderner Politik.”
Pp. 65–148 in Gesellschaftsstruktur und Semantik: Studien zur Wissenssoziologie der modernen
Gesellschaft, Band 3, edited by N. Luhmann. Frankfurt am Main: Suhrkamp.
———. 1990. Die Wissenschaft der Gesellschaft. Frankfurt am Main: Suhrkamp.
———. 1995. “Intersubjektivität oder Kommunikation: Unterschiedliche Ausgangspunkte soziologischer
Theoriebildung.” Pp. 169–88 in Soziologische Aufklärung 6, edited by N. Luhmann. Opladen: West-
deutscher Verlag.
———. 1997. “Grenzwerte der ökologischen Politik: Eine Form von Risikomanagement.” Pp. 195–221 in
Risiko und Regulierung: Soziologische Beiträge zu Technikkontrolle und präventiver Umweltpolitik, edited
by G. Krücken and P. Hiller. Frankfurt am Main: Suhrkamp.
Machiavelli, Niccolò. 1988. The Prince, translated by Q. Skinner and R. Price. Cambridge, UK: Cambridge
University Press.
Malcolm, Norman. 1958. “Knowledge of Other Minds.” Journal of Philosophy 55:969–78.
March, James G. and Herbert A. Simon. 1958. Organizations. New York: Wiley.
Maturana, Humberto R. and Francisco J. Varela. 1980. Autopoiesis and Cognition: The Realization of the
Living. Dordrecht, The Netherlands: Reidel.
Mead, George Herbert. 1934. Mind, Self, and Society: From the Standpoint of a Social Behaviorist.
Chicago, IL: University of Chicago Press.
Mill, John Stuart. 1872. An Examination of Sir William Hamilton’s Philosophy and of the Principal
Philosophical Questions Discussed in His Writings. London, UK: Longmans, Green .
Papafragou, Anna. 2002. “Mindreading and Verbal Communication.” Mind and Language 17:55–67.
Parsons, Talcott. 1951. The Social System. Glencoe, IL: Free Press.
———. 1968. “Interaction: Social Interaction.” Pp. 429–41 in International Encyclopedia of Social Sciences,
Vol. 7, edited by D. L. Sills. New York: Macmillan.
Parsons, Talcott and Edward A. Shils. 1951. Toward a General Theory of Action. New York: Harper &
Row.
Putnam, Hilary. 1975. “Other Minds.” Pp. 342–61 in Philosophical Papers: Mind, Language and Reality,
Vol. 2, edited by H. Putnam. Cambridge, UK: Cambridge University Press.
Quine, Willard Van Orman. 1960. Word and Object. Cambridge, MA: MIT Press.
Reynaert, Peter. 2001. “Intersubjectivity and Naturalism – Husserl’s Fifth Cartesian Meditation Revisited.”
Husserl Studies 17:207–16.
Römpp, George. 1992. Husserls Phänomenologie der Intersubjektivität und ihre Bedeutung für eine Theorie
intersubjektiver Objektivität und die Konzeption einer phänomenologischen Philosophie. Dordrecht, The
Netherlands: Kluwer.
Russell, Bertrand. 1993. Our Knowledge of the External World. London, UK: Routledge.
Sacks, Harvey. 1972. “Notes on Police Assessment of Moral Character.” Pp. 280–93, 444–46 in Studies
in Social Interaction, edited by D. Sudnow. New York: Free Press.
———. 1992. “The Inference-Making Machine (Lecture Fall 1964 – Spring 1965).” Pp. 113–25 in Lectures
on Conversation I, edited by G. Jefferson. Oxford, UK: Blackwell.
Sartre, Jean-Paul. 1976. L’être et le néant: Essai d’ontologie phénoménologique. Paris: Gallimard.
———. 1988. La Transcendence de l’ego. Paris: J. Vrin.
Savigny, Eike von. 1988. Wittgensteins Philosophische Untersuchungen: Ein Kommentar für Leser, Bd. I:
§§ 1–315. Frankfurt am Main: Klostermann.
Schegloff, Emanuel. 1991. “Conversation Analysis and Socially Shared Cognition.” Pp. 150–71 in Perspec-
tives on Socially Shared Cognition, edited by S. D. Teasley, J. M. Levine, and L. B. Resnick. Washington,
DC: American Psychological Association.
———. 1992a. “Introduction (to Sacks, Lectures on Conversation, I).” Pp. ix–lxii in Lectures on Conver-
sation I, edited by G. Jefferson. Oxford, UK: Blackwell.
———. 1992b. “Repair After Next Turn: The Last Structurally Provided Defense of Intersubjectivity in
Conversation.” American Journal of Sociology 97:1295–1345.
Schelling, Thomas C. 1980. The Strategy of Conflict. Cambridge, MA: Harvard University Press.
Scherbel, Martina. 1994. “Deskription oder Postulat? Zur Intersubjektivitätstheorie in der V. Cartesianis-
chen Meditation Edmund Husserls.” Perspektiven der Philosophie 20:275–88.
Schneider, Wolfgang Ludwig. 1994a. Die Beobachtung von Kommunikation: Zur kommunikativen Konstruk-
tion sozialen Handelns. Opladen: Westdeutscher Verlag.
 INTERSUBJECTIVITY 63

———. 1994b. “Intersubjektivität als kommunikative Konstruktion.” Pp. 189–238 in Der Mensch – das
Medium der Gesellschaft? edited by P. Fuchs and A. Göbel. Frankfurt am Main: Suhrkamp.
Schütz, Alfred. 1962a. “Choosing Among Projects of Action.” Pp. 67–96 in Collected Papers I: The
Problem of Social Reality, edited by A. Schütz. Den Haag: Nijhoff.
———. 1962b. “Common-Sense and Scientific Interpretation of Human Action.” Pp. 3–47 in Collected
Papers I: The Problem of Social Reality, edited by A. Schütz. Den Haag: Nijhoff.
———. 1962c. “On Multiple Realities.” Pp. 207–59 in Collected Papers I: The Problem of Social Reality,
edited by A. Schütz. Den Haag: Nijhoff.
———. 1962d. “Scheler’s Theory of Intersubjectivity and the General Thesis of the Alter Ego.”
Pp. 150–79 in Collected Papers I: The Problem of Social Reality, edited by A. Schütz. Den Haag:
Nijhoff.
———. 1964a. “The Stranger: An Essay in Social Psychology.” Pp. 91–105 in Collected Papers II: Studies
in Social Theory, edited by A. Schütz. Den Haag: Nijhoff.
———. 1964b. “The Well-Informed Citizen: An Essay on the Social Distribution of Knowledge.”
Pp. 120–34 in Collected Papers II: Studies in Social Theory, edited by A. Schütz. Den Haag: Nijhoff.
———. 1966. “The Problem of Transcendental Intersubjectivity in Husserl.” Pp. 51–84 in Collected Papers
III: Studies in Phenomenological Philosophy, edited by A. Schütz. Den Haag: Nijhoff.
———. 1974. Der sinnhafte Aufbau der sozialen Welt: Eine Einleitung in die verstehende Soziologie. Frank-
furt am Main: Suhrkamp.
Segerdahl, Pär. 1998. “Scientific Studies of Aspects of Everyday Life: The Example of Conversation
Analysis.” Language and Communication 18:275–323.
Shoemaker, Sydney. 1965. “Ziff’s Other Minds.” Journal of Philosophy 62:587–89.
Silverman, David. 1998. Harvey Sacks: Social Science and Conversation Analysis. New York: Oxford
University Press.
Simon, Josef. 2001. “Intersubjektivität: Ein philosophisch problematischer Begriff.” Archivio di Filosofia
69:41–50.
Sober, Elliott. 2000. “Evolution and the Problem of Other Minds.” Journal of Philosophy 97:365–86.
Sofsky, Wolfgang. 1997. The Order of Terror: The Concentration Camp. Princeton, NJ: Princeton Univer-
sity Press.
Srubar, Ilja. 1983. “Abkehr von der transzendentalen Phänomenologie: Zur philosophischen Position
des späten Schütz.” Pp. 68–84 in Sozialität und Intersubjektivität: Phänomenologische Perspektiven der
Sozialwissenschaften im Umkreis von Aron Gurwitsch und Alfred Schütz, edited by R. Grathoff and B.
Waldenfels. München: Fink.
Stichweh, Rudolf. 2000. “Systems Theory as an Alternative to Action Theory? The Rise of ‘Communica-
tion’ as a Theoretical Option.” Acta Sociologica 43:5–13.
Strawson, Peter F. 1959. Individuals. London, UK: Methuen.
Tomasello, Michael. 2006. “Why Don’t Apes Point?” Pp. 506–24 in Roots of Human Sociality: Culture,
Cognition and Interaction, edited by N. J. Enfield and S. C. Levinson. Oxford, UK: Berg.
Tomasello, Michael, Malinda Carpenter, Josep Call, Tanya Behne, and Henrike Moll. 2005. “Under-
standing and Sharing Intentions: The Origins of Cultural Cognition.” Behavioral and Brain Sciences
28:675–91.
Wiener, Norbert. 1961. Cybernetics: Or Control and Communication in the Animal and the Machine.
Cambridge, MA: MIT Press.
Wittgenstein, Ludwig. 1953. Philosophische Untersuchungen/Philosophical Investigations, translated by G.
E. M. Anscombe. Oxford, UK: Blackwell.
Wrong, Dennis H. 1961. “The Oversocialized Conception of Man in Modern Sociology.” American
Sociological Review 26:183–93.
Zahavi, Dan. 1996. Husserl und die transzendentale Intersubjektivität: Eine Antwort auf die sprachpragma-
tische Kritik. Dordrecht, The Netherlands: Kluwer.
Ziff, Paul. 1965. “The Simplicity of Other Minds.” Journal of Philosophy 62:575–84.

View publication stats