Communication

Theory
Two:
Three
Poonam Pillai August
1992
Pages:
221-233

Rereading Stuart Hall’s

Encoding/Decoding Model

This article is a reading of Stuart Hall’s encoding/decoding model through his later, more
mature work on articulation theory. I t analyzes the underlying assumptions of the model,
accounts for the criticisms made against it, and points out ways in which the theory of
articulation is an advance over the earlier model. While the notion of articulation is
emergent in the encoding/decoding model, it is not adequately theorized. One o f t h e
consequences is the problematic equivalence between the preferred meaning of the text
and dominant ideology. Limitations such as these can be overcome through the present
reading. By situating the model within the movements of Hall’s intellectual history, the
article furthers the task of recasting the model f o r the analysis of encoding and decoding
practices.

Stuart Hall’s encoding/decoding model, which appeared in his article “Encod-

ing/Decoding”’ (Hall, 1980a), has been both appreciated and criticized.’ Al-
though it makes significant shifts (to be discussed later) from traditional models
of mass communication, such as effects research, selective perception theories,
uses and gratifications, and so on, it has several limitations and is in need of
reca~ting.~ I argue that Hall’s later articulation theory provides an incisive cri-
tique of the encoding/decoding model. Read through the notion of articulation,
the encoding/decoding model can be clarified and several of its limitations over-
come. Moreover, a revised version of the model provides an important frame-
work within which the theorization of encoding and decoding practices can be
developed. Although Hall’s recent work has focused more on questions of ethnic-
ity rather than encoding and decoding, the assumptions underlying the encoding/
decoding model and articulation theory reflect major trends within cultural stud-
ies. It is with these considerations that I reread the model, assessing both its
strengths and weaknesses.
As suggested above, the encoding/decoding model overcomes several prob-
lems that persist in traditional models of mass communication. First, Hall
(1980a) moves away from the linearity of these models and introduces the notion
of complexity into the circuit of communication. He criticizes traditional models
for overemphasizing the site of message exchange, thereby neglecting other rno-
ments of the circuit. Hall shifts ground by conceptualizing the process of commu-
nication as a “complex structure of relations.” The circuit is constituted through
practices of production, circulation, distribution/consumption, and reproduc-
tion. What is crucial is that although the moments of the circuit maintain their

221
 Communication
Theory

specificity, they are connected to the others through the process of articulation,
a concept to be discussed later in the article. This helps to distinguish discursive
production from nondiscursive production, and is another advance made by the
encoding/decoding model over previous models of communication. The influ-
ence of Hall’s (1974) reading of Marx’s “1857 Introduction” to the Grundrisse,
“Marx’s most substantial text on ‘method”’ (p. 132) upon this formulation of
the model is unmistakable. The specific affiliations are traced throughout the
present article.
Although the advances made by the encoding/ decoding model are consider-
able, several problems remain in the form in which it appears in the “Encoding/
Decoding” (Hall, 1980a) article. First, Hall uses the terms “preferred meaning”
and “preferred reading” interchangeably, assuming an equivalence between
them. Since these terms are central to his formulation of encoding and decoding
practices, the distinction between them needs to be clarified. The preferred mean-
ing of a televisual text refers to the dominant or preferred codes with which it is
inscribed by practices of encoding and media production located within the
institution of broadcasting. It is a property of the text. Practices of encoding,
according to Hall (1980a), attempt to prefer but cannot prescribe a correspon-
dence between the encoded and decoded meanings. A preferred reading occurs
when such a correspondence is achieved through practices of decoding. Unlike
the preferred meaning, it is a property of the reader. Secondly, terms such as
“prefer,” “represent,” “reproduce,” and “fix,” which are key to Hall’s formula-
tion of the circuit of communication, are left unclarified in the article. Due to an
indiscriminate use of these terms, Hall seems to assume unproblematically that
the preferred meaning and the preferred reading of a text are equivalent to the
dominant ideology. The process of “preferring” the meaning of a text, at the
moment of encoding, is simply read as repeating the dominant ideology. A
similar problem occurs at the moment of decoding. The notion that equivalences
among preferred meanings, preferred readings, and dominant ideologies have to
be achieved through practices of encoding and decoding, which Hall wants to
emphasize through the notion of hegemony, is neglected. The ambiguous use of
certain key terms also leads to the possibility of transparent communication
being read into the model (cf. Lewis, 1983).
Other criticisms of the model arise out of the fact that at the time of the
“Encoding/Decoding” model, Hall (1980a) was still working within an extended
version of the base-superstructure4 model of the social totality (cf. Hall, 1977,
1980b, 1983). This has allowed the possibility of direct determination by the
economic base being read into the model (cf. Fiske, 1987). Although Hall’s
theory of articulation is emergent in the “Encoding/Decoding” article, it is not
well developed. I will attempt to show the ways in which the articulation theory
is a theoretical advance over the encoding/decoding model and allows the model
to be recast more usefully.’
In analyzing the underlying assumptions of the “Encoding/Decoding” article,
it is important to begin by situating it in the movements of Hall’s intellectual
history. One way in which his work can be periodized is by distinguishing the
moment of the encoding/decoding model from his later theory of articulation,

222
 Rereading Hall’s Encoding/ Decoding Model

which was first elaborated in his work on subcultures (Hall & Jefferson, 1976).
I will focus primarily on the moment of the “Encoding/Decoding” article and
briefly discuss the reasons for the emergence of later positions. I will begin with
an analysis of the assumptions underlying the model as formulated in “Encoding/
Decoding.” This involves a discussion of articulation theory. Next, I will con-
sider the formulation of the circuit of communication and its affiliation to Marx’s
“1857 Introduction.” Following this, I will analyze Hall’s formulation of the
encoding and decoding moments of the circuit, their limitations, and the critique
provided by the theory of articulation. I will conclude by discussing the limita-
tions that still remain in the encoding/ decoding model and the implications of
this analysis.

Assumptions
The theoretical context Hall works against in writing “EncodinglDecoding” is
one in which the television message is assumed to have one predetermined, fixed
meaning that can be transparently read by audiences. This entails a notion of
communication as a simple reflection of reality. Hall attempts to move away
from this notion for several reasons. First, it does not emphasize the determinacy
of the message form itself. Second, it does not integrate the notion of meaning as
multiaccentual, a term borrowed from (Volosinov, 1973). It is the notion of the
multiaccentuality of the sign that allows Hall to conceptualize the message itself
as being “structured in dominance.” Hence, this is a key move for Hall.
From a marxist framework, the denial of the transparency of communication
has been accompanied by the assertion of the concept of ideology, traditionally
understood as false consciousness. However , the notion of ideology as false
consciousness, as argued by Hirst (1979),is empiricist since it assumes the possi-
bility of “true consciousness.” For an empiricist, objects are knowable through
the subject’s experience, there being in principle a correspondence possible be-
tween experience and the nature of the object in the case of recognition, and
noncorrespondence in the case of misrecognition. This implies that ideology as
false consciousness can in principle be removed by reforming the alienating na-
ture of the real conditions of existence, which are the source of distortion, the
false representation of the real, or ideology. Hall moves away from the notion of
ideology as false consciousness. His notion of ideology, which has been influ-
enced by both Althusser and Gramsci, is more extensively developed in his later
work (cf. Hall, 1982, 1985). Gramsci’s work has been particularly influential in
introducing the notion of struggle into the functioning of ideology. An elaborate
discussion of this term is not possible here.
The notion of ideology as false consciousness has developed from within a
classic base-superstructure model of the social totality that assumes the primacy
of the economic. In shifting away from the notion of communication as a simple
reflection or representation of reality, Hall simultaneously moves away from an
economically deterministic model of the social totality. He develops a nonreduc-
tionist theory of social practices and determinations through his theory of articu-
lation. A cogent analysis of the nature and structure of the social formation is

223
 Communication
Theory

found in his more mature work (Hall, 1977, 1980b, 1983). The works of Althus-
ser, Gramsci, and Laclau and Mouffe in different ways have been crucial for
Hall in this regard.
As suggested above, one way in which Hall (1980a) moves away from the no-
tion of communication as a simple reflection of reality is by emphasizing the deter-
minacy of the discursive. The discursive form in fact has a privileged position in
the process of exchange of messages from sender to receiver. This does not under-
mine the event itself or its social and political consequences. On the contrary, it
simply suggests that the discursivity of the message is both “necessary” and “deter-
minate.” For the process of communication to be fully effective, the message has to
be incorporated into social practices at a different stage of the process.
In emphasizing the determinacy of the discursive, Hall assumes a distinction
between the discursive and the nondiscursive. His point is to emphasize that dis-
courses do not exhaust the effects of real social relations and are a product of the
articulation of language with real social relations. Articulation, as Hall (1985) ar-
gues, is a way of understanding the central marxist notion of determinations. It
allows him to think of the relation between any two elements of the social totality
in terms of “no necessary correspondence.” Hall (1985) provides a succinct defini-
tion of articulation in a footnote in the essay “Significance, Representation, Ideol-
ogy: Althusser and the Post-Structuralist Debates.” He clarifies that

By the term “articulation,”I mean a connection or a link which is not necessarily given in
all cases, as a law or a fact of life, but which requires particular conditions of existence to
appear at all, which has to be positively sustained by specific processes, which is not
“eternal” but has constantly to be renewed, which can under some circumstances disappear
or be overthrown, leading to old linkages being dissolved and new connections- rearticula-
tions-being forged. (p. 113)

Hall (1985) argues that articulated elements do not lose their specificity or condi-
tions of existence but remain “distinctions within a unity.” It is precisely because
of “no necessary correspondence” between the elements that under certain condi-
tions the unity can be reconstituted. The unity is not formed randomly but has a
determinate configuration. Hence, it is not an expressive totality but a complex
structure sustained through relations of domination and subordination. Hall
(1980b) aptly describes the nature of the unity through Althusser’s phrase-a
“complex unity, structured in dominance” (Hall, 1980b, p. 325).
While reading Althusser’s (1969) For M a r x , Hall (1985) recognizes that there
are different social contradictions and origins and that they are not always lo-
cated at the same place nor have the same effect. Particular practices can be
determined by more than one set of relations, theorized by Althusser’s concept
of “overdetermination.”6 The task is to think about the articulations between
different contradictions. Hall (1985) agrees with Althusser that advances over a
“delayed teleology,” that is, the assurance of a final ground or origin, are already
there in Marx’s “1857 Introduction.” It is no longer possible to read off the
ideology of a class from its socioeconomic position. This allows the social forma-
tion to be conceptualized in terms of a much more contingent model than the

224
 Rereading Hall’s Encoding/Decoding Model

traditional marxist position where determinations can ultimately be reduced to

the economic base.
Articulation theory, according to Hall (1986), provides a way of understand-
ing how and under what social and historical conditions discourses are formed
and become associated with, or are appropriated by different historical subjects.
It is a way of theorizing the “contingent” but “non necessary” relation between
ideologies, practices, subjects, and subject positions. In an interview, Hall (1986)
suggests that the theory of articulation

enables us to think how an ideology empowers people, enabling them to begin to make
some sense or intelligibility of their historical situation, without reducing those forms of
intelligibility to their socio-economic or class location or class position. (p. 53)

The major advance made by the theory of articulation is that it rethinks the
question of determinacy in terms of the effects rather than the origins of social
forces and practices. It provides a way of understanding determinations in a
nonreductionist manner. In Lawrence Grossberg’s (1986) terms, it is an attempt
to rethink the “dialectic of determination as struggle,” a struggle against neces-
sary correspondence or noncorrespondence between practices and their effects.
The theory of articulation in many ways overcomes the problems of the encod-
ing/decoding model. This will be discussed in the following sections.
In contrast to his later work, Hall (1980a) makes a clear distinction between
discourses and practices in the “Encoding/Decoding” article. For instance, he
argues that

It is in this discursive form that the circulation of the ‘product’ takes place . . . as well as
its distribution to different audiences. Once accomplished, the discourse must then be
translated- transformed, again-into social practices if the circuit is to be both complete
and effective. If no ‘meaning’ is taken, there can be no ‘consumption.’If the meaning is not
articulated in practice, it has no effect. (p. 128)

Later, in an interview, Hall (1986) acknowledges that he has come far in rethink-
ing practices as functioning discursively, like languages. This has allowed the
reintegration of the notion of a decentered, nonunitary, and contradictory sub-
jectivity into marxism and structuralism. These had already shifted away from
the notion of a unified, empirical, and sociological subject to thinking of “a
history without subjects” and “a language with no speakers.” Hall acknowledges
the seminal work of Laclau and Mouffe (Laclau, 1977; Laclau & Mouffe, 1985)
in this regard.
Yet, Hall holds on to the analytic distinction between the discursive and the
nondiscursive and does not consider it a problem for his theory. He clarifies
(1986) the difference and specificity of his position from Laclau and Mouffe’s
later work in Hegemony and Socialist Strategy (Ladau & Mouffe, 1985). In
contrast to their position, Hall argues against conceptualizing practices as noth-
ing but discourses, against thinking of historical subjects and positions as simply
“discursively constituted subjectivities” and positionalities. While he stresses that

225
 Communication
Theory

the social operates like a language, he is not willing to go so far as to suggest

that the social is language. Hall argues that “the fully discursive position is a
reductionism upward rather than a reductionism downward, as economism was”
(1986, p. 57). Thus, even at the expense of being theoretically less accurate, he
holds on to the distinction between discourse and practice.

The Circuit of Communication

Hall (1980a) conceptualizes the process of communication in terms of a circuit, a
“structure produced and sustained through the articulation of linked but distinct
moments-production, circulation, distribution/consumption, reproduction”
(p. 128). The specificity of each of these moments is marked by the distinct
manner in which they operate and the conditions under which they exist. While
each moment is necessary to the circuit, none assumes the position of a final
determinant origin. Each of the moments exerts its determinacy upon the next,
but there is no necessary correspondence between any two moments. The mo-
ments are relatively autonomous to each other. This implies that to a certain
degree, each moment is autonomous, exerts its own determinacy, but none can
fully determine another. Although production initiates the circuit, it does not
constitute a closed system. The emphasis is on the difference and specificity of
each moment and the complexity and openendedness of the circuit as a whole.
Practices are required to bring about linkages that relatively constrain the next
moment.
This formulation marks a major shift from the classical marxist notion of
necessary correspondence between different elements of the circuit of production.
Hall’s reading of Marx’s “1857 Introduction” shows that the model of such an
overdetermined contingent circuit is already emergent in this text. He acknowl-
edges this in his reading of Althusser (Hall, 1985). It is important to recognize
the influence of this text on the encoding/decoding model.
One of the key methodological and theoretical premises of Marx’s “1857
Introduction” is the notion of a “differentiated unity.” This assumption is com-
mon to Hall as well. As he ( 1 9 8 0 ~ argues,
) the unity of a social formation is
seen as constructed and sustained through the difference rather than identity of
practices. There is no essence to which the historical forms of a totality can be
reduced. Rather, the totality is formed through the articulation of different mo-
ments whose determinations and differences are maintained. The influence of
the “1 857 Introduction” is also seminal in conceptualizing the relations between
the different phases of the circuit of production, exchange, distribution, con-
sumption, and reproduction, that is, in analyzing the relations between the parts
of a complexly structured whole. Marx’s dialectical method traces out the “inner
connection” rather than the “mere juxtaposition” of different moments.

The method which merely sets opposites together in an external way, which assumes
that, because things are neighbors, they must therefore be related, but which cannot move
from oppositions to contradictions, is “dialectical” only in its surface form. (Hall, 1974, p.
140)

226
 Rereading Hall’s Encoding/ Decoding Model

The articulations of the different moments d o not proceed in succession but

are related internally and dialectically through their differences. In theorizing a
nonlinear circuit of communication, Hall (1980a) can be seen as attempting to
integrate Marx’s notion of an internal, dialectical connection between different
moments of the circuit. A trait borrowed directly from Marx’s method is that
the identity of different moments remains distinct. Although the moments are
mutually dependent and indispensable to each other, they remain external to
each other. Each moment has its determinate conditions and “[nlothing except
the maintenance of these determinate conditions can guarantee the continuity of
this mode of production over time” (Hall, 1980a, p. 145). As seen earlier, this is
also a characteristic of the communication circuit as conceptualized by Hall.
In Marx’s method, the moment of production is privileged over the others in
two ways. It, rather than the moments of consumption or distribution, initiates
the circuit and is also the moment that “determines in the last instance.” Hall
argues that Marx gives an “overdeterminacy” to production (Hall, 1974). To
quote Hall (1974)

the distribution of its products, its results, in the form of wages or rent, cannot be its
starting point. Once this distribution of instruments and agents has been made, they form
the starting conditions for the realization of value within the mode; this realization process
generates its own distributive forms. This second type of distribution, however, is clearly
subordinate to production in this wider, mode-specific sense, and must be considered as
over-determined by it. (p. 146)

While a detailed exposition of the “1857 Introduction” is outside the scope of this
article, it may suffice to say that it lays the initial groundwork for a contingent
overdetermined circuit, later theorized by Hall.
As mentioned earlier, Hall conceptualizes the encoding /decoding circuit as a
“complex structure in dominance” (previously discussed with reference to the
concept of articulation). This is an Althusserian term and is significantly inte-
grated by Hall in his theory of articulation (Hall, 1980b, 1985). The notion of
the social formation as a complex structure can itself be traced back to Marx, as
is evident from Hall’s reading of the “1857 Introduction.” Yet, these terms, like
Hall’s theory of articulation, are not fully developed in the ‘‘Encoding1Decoding”
article. Although he states that the circuit is a complex structure, he has not at
this time theorized the reason for its complexity. This is one way in which the
theory of articulation makes an advance over the encoding/decoding model.

Encoding and theNotlon of Preferred Meaning

Practices of encoding, situated within structures of broadcasting, produce televi-
sion messages drawing upon various events, agendas, professional ideologies,
notions about the audience, and so on. They exist in a relationship of relative
autonomy and determinateness with relations of production and the technical
infrastructure. Certain codes are employed by the encoder to encode the program
and the decoder to decode the program as meaningful. The circulation of the

227
 Communication
Theory

product takes place in the message form, which exerts its own determinateness.
According to Hall (1980a), both structural differences of relation and position
between the encoder and decoder, and differences between the codes employed
by them, determine the degree of symmetry and asymmetry between encoding
and decoding.
Practices of encoding, according to Hall, attempt to prefer a certain reading
of the text. This inscribed meaning, called the preferred meaning, is a property
of the text. Codes, which primarily at the level of connotation help the viewer to

‘reference’ a wide domain of social meanings, relations and associations, are the means by
which the widely distributed forms of social knowledge, social practices, the taken for
granted knowledge which society’smemberspossess of its institutions, beliefs, ideas and legit-
imations are‘broughtwithin the horizon’of language and culture. (Hall, 1977, p. 330)

These codes link certain domains of meaning while excluding others in a certain
context. Hence they are preferred meanings, “structured in dominance.” How-
ever, the process of encoding is determinate but not determining, that is, it is
open-ended and cannot guarantee any correspondence between the encoded and
decoded moments. There is “no necessary correspondence” between encoding
and decoding, which have distinct conditions of existence.
Hall characterizes the television text as polysemic rather than plural. This im-
plies that the connotative readings of the text are not equally available to the reader.
They are structured in dominance, that is, hierarchically organized into dominant
or preferred meanings. The text therefore prefers certain ways of reading over oth-
ers. This notion of the message as a “structured polysemy” allows Hall to retain a
notion of power structuring inscribed within the text. The notion of power struc-
turing both within and outside the process of communication is also crucial for
Hall. One condition for this is the difference in codes available to broadcast jour-
nalists and viewers. The former have greater access and hence greater power to
determine what gets transmitted through the television screen. This imbalance in
power is theoretically manifest in the nature of the television text.
One possible reading of Hall’s formulation made by Lewis (1983)is that it con-
siders practices of encoding as a secondary signifying process acting upon an “al-
ready constituted system of signs”- the primary level of preferred meanings. The
television message is seen as simply reproducing rather than producing meanings,
this being the secondary level of preferred meanings. Lewis argues that

the problem with understanding television as a secondary sign producing agency, operat-
ing on the work done by primary agencies, is that it implies the existence of a set of privi-
leged agencies, with privileged access to the signifier, without identifying them. (Lewis,
1983, p. 153)

According to him, the relationship of the encoders to the defining elite and that
of encoding to other practices of signification remains unspecified. Lewis (1983)
raises the question whether dominant meanings should be understood in relation
to the socioeconomic position of the authors. Although Hall suggests that broad-

228
 Rereading Hall’s Encoding/Decoding Model

cast journalists are relatively autonomous to the dominant ideology, Lewis rejects
this way of theorizing the relationship. He argues that such a formulation neg-
lects the fact that practices of encoding and decoding have different conditions
of existence, and can be reduced to those models that consider the television
message to be simply a representation or a misrepresentation of reality rather
than constituting reality.
Problems also arise when the model is considered semiologically. In semiotics,
a “signifier” refers to a word, object, or any material entity, and is the unit prior
to meaning. The “signified” refers to the meaning given to the signifier, the
concept that exists in people’s minds. These combine to produce the “sign,”
reducible to neither, through the process of signification. This process requires
the presence of a constituting subject. According to Lewis, then, the problem in
considering the television message as inscribed with a preferred meaning is that
it denies that the sign has a material existence (since meaning is brought to
signifiers and not inscribed within them) and that it requires the presence of a
constituting subject. This rejects the activity of television as a sign producing
agency, which Lewis emphasizes. An alternative, he suggests, is to consider the
world prior to encoding as consisting of signifiers rather than significations.
I will now respond to the above criticisms, some of which can be circumvented
if read through Hall’s articulation theory. First, although in the “Encoding/
Decoding” article Hall emphasizes the openendedness of the television message
and the interpretive work required to prefer one reading of the text over another,
what makes Lewis’s reading possible is the problematic way in which the terms
“signification,” “production,” “reproduction,” “fix,” “prefer,” and so on, have
been used in the article. Second, with reference to the signification of complex
social relations, Hall (1977) clarifies that we must not think of such relations as
having a meaning, but must think of language, the principle mediator of ideologi-
cal practice, as enabling things to mean. This is in clear contrast to Lewis’s
reading of Hall. Third, in defense of Hall it can be argued that the semiological
problems in specifying the relationship between practices of encoding and other
signifying practices notwithstanding, the world prior to encoding cannot be as-
sumed to be a world of signifiers but of discourses and ideologies that practices
of encoding articulate.’
In theorizing the encoding/decoding model, Hall assumes that (a) the pre-
ferred meaning represents the dominant ideology in the sense of repeating it,
and (b) there is no necessary correspondence between the encoded and decoded
meanings. He also suggests that professional broadcasters are relatively autono-
mous to the dominant ideology. But Hall does not adequately theorize the pro-
cess of articulation at the level of production of the television text. As a result,
the process of fixing and preferring the meaning of the text is simply read as
repeating and reproducing the dominant ideology. In a later essay, Hall (1985)
clarifies that

without some arbitrary “fixing” or what I am calling “articulation,” there would be no

signification or meaning at all. What is ideology but precisely this work of fixing meaning
through establishing, by selection and combination, a chain of equivalences? (p. 9 3 )

129
 Communication
Theory

Preferred meanings and dominant ideologies need to be articulated through prac-

tices of encoding in order to achieve an equivalence between them. The theoriza-
tion of this moment of articulation is neglected in the “Encoding/Decoding”
article. Finally, in the context of Hall’s theory of articulation, dominant mean-
ings cannot be read off from the socioeconomic position of the authors but need
to be understood in terms of how they are articulated by the latter. There is no
necessary correspondence or noncorrespondence between dominant ideologies
and preferred meanings, a point not clarified in “Encoding/Decoding.”

Decoding and the Notion of Preferred Reading

Reading, according to Hall, is a “highly structured, asymmetrical and non-
equivalent process” (1980a, p. 135). It is never as individualized as selective
perception theory suggests (Hall, 1980a). Hall distinguishes among three hypo-
thetical decoding positions, namely, “dominant-hegemonic or preferred,” “nego-
tiated,” and “oppositional.” This is broadly understood as the preferred reading
model. David Morley (1980) points out that this formulation of the three ideal
typical positions is based on Parkin’s (1973) scheme of ideal decoding positions.*
A preferred reading occurs when a decoder operating within the dominant code
decodes a message in terms of the codes with which it has been encoded. This,
according to Hall, is the hypothetical instance of “perfectly transparent commu-
nication” (Hall, 1980a). Such a characterization has allowed the possibility of
transparent communication being read into the mode. In light of the theory of
articulation, this can be seen as the effect of an articulation between the codes
through which the text is constructed and the codes available to the decoder.
Neither the text nor the reader is situated in isolation but in relation to various
signifying agencies and practices that exert their determinations upon this articu-
lation. It is at the moment of the preferred reading that it has been possible to
exert power through discourse. This is the “moment of hegemony” in which
plausibility for the way in which events have been encoded has been won. This
formulation also suggests that in spite of the activity of producing significations,
the audience is passive.
To emphasize the notion of no necessary correspondence between encoding
and decoding, Hall (1980a) also distinguishes between “negotiated” and “opposi-
tional” readings. A negotiated reading occurs when in spite of recognizing the
authenticity of hegemonic definitions the viewer contests them through particular
or “situated” logics. An oppositional reading, according to Hall, occurs when
the viewer understands both the literal and implicative meanings of a discourse
but decodes the message in an entirely heterogenous way. It is here, he argues,
that the “politics of signification” meets the “struggle in discourse,” indicating
the failure of practices of encoding to achieve a hegemonic reading of the text.
From the point of view of decoding research, Hall’s preferred reading model
is less deterministic than the Screen position or the uses and gratifications posi-
tion in that it sees decoding as a product of both the reader and the text.’ As has
been pointed o u t (cf. Morley, 1980), the Screen position assumes a totally effec-
tive text producing an inscribed subject positionality, where the subject is simply
understood as a textual production. Here, the distinction between a textual

230
 Rereading Hall’s Encoding/Decoding Model

and a real subject is collapsed. Uses and gratifications, on the other hand, limits
all determinations on the side of the decoder, thereby ascribing to him or her
complete power to determine readings. Hall’s theorization of the three decoding
positions shows that readings are articulated both by the codes present in the
text and those available to the reader, and the asymmetrical structural positions
of the encoder and decoder.
To illustrate how the encodingldecoding model can be recast more usefully
in the context of articulation theory, I will briefly consider John Fiske’s reading of
Hall. Fiske (1987) interprets Hall’s analysis of decoding positions in ‘‘Encoding/
Decoding” as suggesting that “there is a necessary correlation between people’s
social situations and the meanings that they may generate from a television
program” (p. 260). Here, Fiske reads Hall as arguing that the codes available to
a reader are directly determined by his or her social situation, which determines
whether the reader prefers, opposes, or negotiates with the text. Although Hall
does not state anywhere in “Encoding/Decoding” that there is a necessary corre-
lation between the viewer’s social situation, the codes available to her or him,
and the meanings constituted, a possible reason for Fiske’s reading is that at the
time the article (Hall, 1980a) was published, Hall was still working within an
extended version of the base-superstructure model. However, Hall’s articulation
theory suggests that he no longer wants to hold on to even an expanded version
of the deterministic model. His later work foregrounds questions of culture and
ideology, abandoning the privilege accorded to the economic but not its effecti-
vity. In the context of the theory of articulation, Fiske’s reading is rendered
insignificant. Audience readings can no longer be assumed to be directly deter-
mined by their socioeconomic positions.

Limitationsand Conclusion
In terms of audience research, several problems remain in the encoding/ decoding
model. First, the use of the term “negotiated readings” as formulated by Hall in
“Encoding/Decoding” is limited. It refers to a single hypothetical decoding posi-
tion and fails to acknowledge the fact that all readings are in a sense negotiated.
Secondly, as Morley (1980) points out in his critique of Parkin’s (1983) model,
the three categories d o not reflect the diversity of discourses in society. Further,
Lewis (1983) criticizes the encoding/decoding model for distinguishing only
three hypothetical decoding positions in relation to a single text. This, he argues,
inadequately reflects the multireferentiality of the text. Thus, from the point of
view of audience research, the assumption that audience readings fall within
three hypothetical positions remains highly problematic. As both Morley (1981)
and Lewis (1983) argue, categories in decoding research should not be theoreti-
cally presupposed but should be established by empirical evidence. Moreover, in
distinguishing three hypothetical decoding positions in terms of the decoder’s
structural location and access to discourses, the preferred reading model cannot
account for the fact that the same decoder may produce contradictory readings.
This problem has also been pointed out by Lewis (1983).
As already stated, although the text has a preferred meaning, different read-
ings are possible, according to Hall. These differences are the effects of the

231
 Communication
Theory

structural location and discourses that the reader brings to the text. But, although
the preferred reading model is able to theorize the determinacy of the text on
audience readings, it can adequately explain the differences in audience readings
only in terms of factors external to the text, but not in terms of the text itself.
The assumption of the equivalence among preferred meaning, preferred reading,
and the dominant ideology, points to another serious problem. Such an equiva-
lence assumes that if, for instance, a decoder opposes the preferred reading of a
text, he or she necessarily opposes the dominant ideology as well. This, however,
need not be the case. It is possible for a decoder to oppose the preferred meaning
of a text and still remain within the terms set by the dominant ideology.
Further, the production of subjects and subject positions in the process of
encoding and decoding, and their relation to the nature of the text, is an area
mostly neglected in ‘‘Encoding/ Decoding.” Although there is no direct discussion
of the production of subjectivity, Hall’s analysis of the decoding moments sug-
gests a rather unified notion of the decentered subject. The attempt to move
away from the notion of the subject as a fixed constant entity can be seen if the
three decoding positions are not read simply as sociological descriptors but are
assumed to be possible even within the same decoder. However, it is only in his
later work that Hall develops the notion of a multiaccentual, processual, but
historically contingent subject.”
Although the encoding/decoding model has various limitations as discussed
above, it continues to provide an important framework within which the theori-
zation of specific cultural practices can be developed. As I have demonstrated in
this article, Hall’s more advanced articulation theory provides a critical lens
through which the encoding/decoding model can be read more usefully. This
clarifies various criticisms of the model. In addition, by reinterpreting terms such
as “prefer,” “fix,” and “represent,” that are central to the encoding/decoding
model, through the theory of articulation, several limitations are overcome. The
model no longer posits problematic equivalences among dominant ideologies,
preferred meanings, and readings of a text. Such equivalences are products
of articulations that are achieved through practices of encoding and decoding.
Recast in this manner, the encodingldecoding model provides an important
framework within which the theorization of specific cultural practices can be
developed.

Poonam Pillai is (as of Fall 1992) an assistant professor in the Department of Communica-
Author tion, The Ohio State University, Columbus, OH 43210. Her present research is on femi-
nist and postcolonial theory. This is a revised version of a paper presented at the Seventh
International Conference on Culture and Communication, Philadelphia, October 1989.
The author wishes to thank Professors Ian Angus, Jon Cruz, andlustin Lewis and the re-
viewers of Communication Theory for their valuable comments.
’ This is an edited version of an earlier article, “Encoding and Decoding in Media Dis-
Notes course,” published as an occasional paper for the Centre for Contemporary Cultural Stud-
ies, Birmingham, England, in 1973.
See David Morley’s (1980) The “Nationwide” Audience for the use of an extended

231
 Rereading Hall’s Encoding/Decoding Model

version of Hall’s model in audience research. For an incisive semiological critique of the
model see Lewis (1983).
This process is well under way; see Morley (1980, 1981). For a critique of Morley’s
revisions, see Lewis (1983).
“Base-superstructure” refers to the classic marxist conception of the social totality,
where primacy is given to the economic base over the superstructure constituted by political
and ideological practices.
The discussion of the theory of articulation is limited in this article but sufficiently aids
the reading.
For a detailed exposition of this concept, read “Contradiction and Overdetermination”
in Althusser’s (1969)For Marx.
’ A critique of Lewis’s (1983) assumptions is outside the scope of this article.
* For a critique of Parkin’s model, see Morley (1980).
See Morley (1980) for an excellent discussion of the three positions. For a critique of
the preferred reading model, see Lewis (1983, p. 156).
lo
For instance, see Hall (1985).
Althusser, L. (1969).For Marx (B. Brewster, Trans.). London: New Left Books.
Fiske, J. (1987). British cultural studies. In R. C. Allen (Ed.), Channels of discourse: References
Television and contemporary criticism (pp. 254-289). Chapel Hill and London: Uni-
versity of North Carolina Press.
Grossberg, L. (1986). History, politics and postmodernism: Stuart Hall and cultural stud-
ies. ]ournal of Communication Inquiry, 10(2),61-77.
Hall, S. (1973). Encoding and decoding in the media discourse. Stencilled paper no. 7,
Centre for Contemporary Cultural Studies, University of Birmingham.
Hall, S. (1974).Marx’s notes on method: A “reading” of the “1857 Introduction.” Working
Papers in Cultural Studies, 6,132-171.
Hall, S. (1977). Culture, the media, and the “ideological effect.” In J. Curran, M. Gure-
vitch, & J. Wollacott (Eds.), Mass communication and society (pp. 315-348). London:
Edward Arnold.
Hall, S. (1980a).Encoding/decoding. In S. Hall, D. Hobson, A. Lowe, & P. Willis (Eds.),
Culture, media, language (pp. 128-138). London: Hutchinson.
Hall, S. (1980b). Race, articulation and societies structured in dominance. In Sociological
theories: Race and colonialism (pp. 305-345). Paris: Unesco.
Hall, S. ( 1 9 8 0 ~ )Cultural
. studies: Two paradigms. Media, Culture and Society, 2,33-48.
Hall, S . (1982). The rediscovery of “ideology”: Return of the repressed in media studies.
In M. Gurevitch, T. Bennet, J. Curran, & S. WOllaCOtt (Eds.), Culture, society, and
the media (pp. 56-90). London: Metheun.
Hall, S. (1983).The problem of ideology-Marxism without guarantees. In B. Matthews
(Ed.), Marx 100 years on (pp. 28-60). London: Lawrence & Wishart.
Hall, S. (1985). Significance, representation, ideology: Althusser and the post-structuralist
debates. Critical Studies in Mass Communication, 2,91-114.
Hall, S . (1986).O n postmodernism and articulation: An interview with Stuart Hall. [Inter-
view with Lawrence Grossberg] Journal of Communication Inquiry, I0(2),45-60.
Hall, S., & Jefferson, T. (Eds.). (1976). Resistance through rituals: Youth subcultures in
post-war Britain. London: Hutchinson.
Hirst, P. (1979). Althusser and the theory of ideology. In On law and ideology (pp. 40-
74). Atlantic Heights, NJ: Humanities Press.
Laclau, E. (1977). Politics and ideology in Marxist theory: Capitalism, fascism, populism.
London: New Left Books.
Laclau, E., & Mouffe, C. (1985). Hegemony and socialist strategy. London: New Left
Books.
Lewis, J. (1983). The encoding/decoding model: Criticisms and redevelopments for re-
search on decoding. Media, Culture and Society, 5 , 151-169.
Morley, D. (1980). The “nationwide”audience. London: British Film Institute.
Morley, D. (1981). The “nationwide” audience: A critical postscript. Screen Education,
39,3-14.
Parkin, F. (1973). Class inequality and political order: Social stratification in capitalist
and communist societies. London: Paladin.
Volosinov, V. (1973). Marxism and the philosophy of language (L. Matejka & 1. R.
Titunik, Trans.). New York: Seminar Press.

233